Embed Size (px)
Citation preview
Equipped with his five senses, man explores the universe around him
and calls the adventure Science.
Edwin Powell Hubble, "The Nature of Science", 1954.
To my granny, my greatest fan ever
Acknowledgements
This thesis is respectfully presented to the Katholieke Universiteit Leuven through its rector
Prof. Dr. M. Waer and its vice-rector and chairman of Biomedical Science Prof. Dr. M. Casteels, to the
Faculty of Medicine through its dean Prof. Dr. B. Himpens, and to the Department Women and Child
through its chairman Prof. Dr. J. Deprest.
I am grateful to the members of the jury, Prof. Dr. F. Penninckx, Prof. Dr. P. Moerman, Prof.
Dr. V. Mais and Prof. Dr. S. Gordts for their most valuable remarks and critical reading of the
manuscript and to the chairman of the jury Prof. R. Pijnemborg. Thanks to Prof. Dr. D. De Ridder who
chaired the reading committee.
I would thank my promoter, Prof. Dr. Philippe R. Koninckx, who guided me during this great
adventure always with enthusiasm and who taught me the importance of perseverance and rigor as
well as imagination in science and in life.
Thanks to Prof. G. B. Melis from my University in Cagliari who, from the very beginning of my
career as medical doctor and scientist, believed in me and who has been always a great supporter of
this research project. I would also like to thank Prof. S. Angioni who has been for me a great tutor
and a good friend.
I thank all my colleagues from the Laboratory of Experimental Gynaecology. I am thankful to
Maria Merceds Binda for teaching me the mouse model and for her patience in reading carefully all
the manuscripts but, most of all, for being such a good friend. I’m grateful to Karina Mailova, Prof. R.
Pijnenborg, Rieta Van Bree, Lieve Verbist, Lisbeth Vercruysse, and Catherine Luyten for all their help.
Thank you to Ron Schonman and Carlo De Cicco for their help in the surgery and in the
experimental work and for their friendship: I spent with them the best moments in Leuven.
I want also to thank Diane Wolput, Marleen Craessaerts and Petra Stevens for all the logistic
support.
Thanks to Jasper Verguts for bringing me back to the clinic activity and for having me
encouraged to study Dutch, something that turned out to be fundamental for my career; thanks to
Prof. Dr. B. Vanacker, Sandra Meeus, the team from the A-cluster, Chris Verheyden and Mark
Declerck from the technical department, for their substantial help in the clinical trials. Thanks to all
my colleagues and all the nurses of the Gynaecological department for replying, always in good
mood, to all my practical questions.
Thanks to Thomas Koninckx, Bob Koninckx and Steven Declerck from eSaturnus®, Eugeen
Steurs from Storz®, Michael Blackhurst from Fisher and Paykel®, Philippe Billet and Christophe
Lauwerys for providing the equipments and the softwares that allowed the translation of this
research from the lab to the surgery.
Thanks to all my friends in Leuven, Brussels and Cagliari for their support and especially Inga
Sandaite, Elisabetta Groaz, Annamaria Sanna, Hans Aerts, Fabiana Cocco and Francesca Carta that
always stayed by my side and never let me feel lonely. A special thanks to Elisa Donè for being in
these years both my best friend, my stronger supporter and my putative sister.
I want especially to thank my parents and my sisters Francesca and Veronica who, with their
unconditional support, have always encouraged me during all the most important steps of my life: it
would not have been possible without them.
Finally, I want to thank Piergiorgio, the person who have filled my life with light, happiness
and love.
Bruxelles, August 2011
9
Table of Contents
Chapter 1. Introduction .............................................................................................................................. 11
1.1 Clinical significance of adhesions ........................................................................................................... 13
1.2 Pathophysiology of adhesion formation ................................................................................................ 14 1.2.1 The classic model ............................................................................................................................ 14 1.2.2 The mesothelial barrier and the role of the entire peritoneal cavity ............................................. 15
1.3 Adhesion prevention today .................................................................................................................... 18
1.4 The role of nitrous oxide in laparoscopy ................................................................................................. 22
1.6 Post-operative pain and inflammation in laparoscopy .......................................................................... 23
Chapter 2. Aims of this thesis ...................................................................................................................... 25
Chapter 3. General materials and methods ................................................................................................ 27
3.1 Animal models ........................................................................................................................................ 27 3.1.1 The laparoscopic mouse model ...................................................................................................... 27 3.1.2 The mouse model for open surgery ................................................................................................ 31
3.2 Humidification, desiccation and temperature during endoscopic surgery ................................................. 32 3.2.1 In Vitro experiments : a box mimicking the peritoneal cavity ......................................................... 32 3.2.2 In vivo set up .................................................................................................................................... 34
3.3 Clinical studies ........................................................................................................................................ 37 3.3.1 Adding 4% oxygen to the pneumoperitoneum ............................................................................... 37 3.3.2 Full conditioning of the peritoneal cavity ....................................................................................... 37
3.4 Statistics ................................................................................................................................................. 38
Chapter 4. Results ....................................................................................................................................... 39
4.1 Animal experiments ............................................................................................................................. 39 4.1.1 Results of aim 1: The key role of the entire peritoneal cavity and the importance of the quality of surgery on adhesion formation in mice. ...................................................................................................... 39
a. The effect of manipulation during surgery as a cofactor in adhesion formation ............................ 40 b. The importance of training upon adhesion formation .................................................................... 43 c. The effect of blood on adhesion formation .................................................................................... 47 d. The genetic effect on adhesion formation ...................................................................................... 51 e. The role of mesothelial cells on adhesion prevention .................................................................... 55 f. The effect of intercoat gel on adhesion prevention ......................................................................... 58 f. Post-operative acute inflammation of the peritoneal cavity: a common mechanism mediating the effect of the peritoneal fluid upon adhesion formation .......................................................................... 63
4.1.2 Results aim 2. To establish the optimal peritoneal environment to prevent adhesion formation . 70 a. Investigations on N2O and the effect of using mixtures of N2O and O2 in CO2 upon adhesion formation ................................................................................................................................................. 70 b. Translation of peritoneal cavity conditioning during laparoscopic surgery to conditioning in open surgery : equally important for adhesion formation. .............................................................................. 72
10
4.2 Investigations in humans ........................................................................................................................ 77 4.2.1 Results of aim 3. To translate the observations on adhesion prevention and peritoneal conditioning in mice into clinical practice ......................................................................................................................... 77
a. The desiccation during laparoscopic surgery at different settings of flow rate, temperature and humidity ................................................................................................................................................... 78 b. The effect of cooling the peritoneal cavity and the relation between intraperitoneal temperature and core body temperature ..................................................................................................................... 83
4.2.2 Clinical studies ................................................................................................................................. 84 a. The effect of using different gases for the pneumoperitoneum upon inflammatory parameters and upon post-operative pain ........................................................................................................................ 85 b. The effect of extensive abdominal lavage upon inflammatory parameters ................................... 90 c. A RCT comparing traditional laparoscopic surgery with full conditioning and the use of Hyalobarrier gel upon adhesion formation .............................................................................................. 92 d. The clearance rate of peritoneal fluid after surgery with full conditioning .................................... 96
Chapter 5. Discussion .................................................................................................................................. 99
5.1 Animal investigations ........................................................................................................................ 103 5.1.1 The role of the entire peritoneal cavity and the importance of training and quality of surgery on adhesion formation .................................................................................................................................... 103 5.1.2 The genetic effect on adhesion formation.................................................................................... 108 5.1.3 Inflammation and adhesions ....................................................................................................... 113 5.1.4 Effect of Nitrous oxide on adhesions ............................................................................................ 116 5.1.5 Peritoneal conditioning in open surgery ....................................................................................... 119 5.1.6 Mesothelial cells in prevention of adhesions ................................................................................ 120
5.2 Translation to clinical practice ................................................................................................................ 121 5.2.1 Intra-peritoneal temperature and desiccation during endoscopic surgery .................................. 121 5.2.2 Full conditioning of the peritoneal cavity ..................................................................................... 124
5.3 Conclusions ............................................................................................................................................ 127
5.4 Summary ................................................................................................................................................ 128
5.5 Samenvatting ......................................................................................................................................... 131
Chapter 6. Bibliography ........................................................................................................................... 133
Chapter 7. Career and bibliography of Roberta Corona ............................................................................ 149
Addenda ...................................................................................................................................................... 153
11
Chapter 1. Introduction
Since its beginning, surgery has been inevitably followed by the formation of adhesions
irrespective of its existence and importance realization by surgeons. The first descriptions of
adhesions date from as early as the XVI century when Vesalius (1) and Harvey described adhesions
between the bowels. In 1853 Virchow (2) described the important role of fibrin in adhesion
formation.
Adhesions are defined as abnormal fibrous connections joining mesothelial tissue surfaces in
abnormal locations (3) and more generally as pathologic formations between the mesothelial
surfaces within body cavities following tissue damage caused by surgical trauma, infection, ischemia,
exposure to foreign materials, etc.(4). Generally 3 types are distinguished: (i) adhesion formation are
adhesions forming at the operative sites, (ii) de novo adhesions are adhesions formed at non-
operative sites and (iii) adhesion reformation are adhesions forming after lyses of previous adhesions
(5).
In this thesis we will discuss the new concept about the importance of the peritoneal cavity
for adhesion formation in addition to the traditional concept of local trauma and opposing defects.
Although important for all postoperative adhesion formation, this concept might be even more
important for endoscopic surgery because of the pneumoperitoneum used and the possibilities of
conditioning of the gas used for the pneumoperitoneum. Surgery here being the major contributor to
adhesion formation since most frequently adhesions are formed postoperatively (6).
During the early years of surgery, adhesion formation received little attention, the focus
being on infection and survival. With reproductive medicine and infertility becoming a subspecialty in
the seventies, reproductive surgery has developed bringing into the front the importance of
minimizing postoperative adhesion formation. Reproductive surgery embraced microsurgery as a
magnification tool permitting tubal reanastomosis and as a principle of surgery emphasizing the
12
prevention of desiccation and gentle tissue handling (7-10;11). Prevention of adhesion formation was
mainly based upon careful observational medicine and common sense, and most of the principles
became only much later experimentally evaluated. Some mistakes, however, were also introduced
such as the free peritoneal graft to cover denuded peritoneal areas, a technique shown later to be
strongly adhesiogenic (12;13). Almost simultaneously, reproductive medicine developed diagnostic
laparoscopy with a special interest in endometriosis, and subsequently endoscopic surgery.
After its introduction in mid-eighties endoscopic surgery has had an explosive development
and replaced progressively microsurgery and also laparotomy, not only in gynaecology but also in
abdominal surgery and in urology. Though microsurgery was initially performed in highly specialized
centres (14;15) reproductive surgery when performed laparoscopically, became a mainstream
surgery losing the microsurgical focus on prevention of adhesion formation. It has been taken for
granted from the start that laparoscopy by its nature is a ‘minimally invasive’ technique and
adhesions would be a minor problem since then. This assumption has leaded to neglect the
principles of adhesion prevention (16;17). Endoscopic surgery was considered to potentially offer
reduced operative time and morbidity, less pain and less postoperative ileus compared with the
traditional open surgical procedures and therefore were expected to be associated with less
adhesion formation. The data remain controversial. Most studies found it unclear whether
laparoscopy has a less adhesiogenic effect than laparotomy (18-20). Less adhesion formation during
laparoscopy was found in animal models (21;22). In humans a decrease only in de novo adhesion
formation during laparoscopy compared to laparotomy has been seen (23;24). With the realization
that laparoscopic surgery was not the solution to prevent adhesion formation (25;26), laboratory
research and clinical interest in adhesion formation revived, recently supporting that classical
principles of the “gentle tissue handling” in preventing adhesion formation are still valid in
laparoscopy, and new products were developed, and an attention to ‘quality of surgery’ has been
renewed.
13
1.1 Clinical significance of adhesions
The last decade gave us evidence for the clinical significance of adhesions and adhesion-
related complications.
Approximately 30% to 40% of all adults in Western countries will have had some form of
abdominal surgery during their lifetime (27). Adhesions are shown to be the most frequent
complication of abdominal surgery (28) and represent one of the greatest unresolved problems in
gynaecologic surgery today, occurring in 70% to 95% of the patients (27;29;30).
Depending on localisation and structure adhesions might be asymptomatic or the cause of
serious complications such as bowel obstruction, chronic pelvic pain (31-33) and infertility. In
patients with bowel obstruction 40% of cases are adhesion-related (34). It has been reported that on
average, 4% of patients undergoing abdominal or pelvic surgery depending on the procedure will be
readmitted to the hospital with adhesion related complications. Furthermore, in those patients
undergoing repeat abdominal surgery (not due to adhesion related problems), 39.9% will be
complicated by the presence of adhesions (35).
The formation of adhesions following open gynaecological surgery has a considerable
epidemiological and clinical impact. It has been reported that intra-abdominal adhesions occur in 60–
90% of women who have undergone major gynaecological procedures (29;36;37). Further, the recent
SCAR study by Lower et al38 conducted in Scotland reported that women undergoing an initial open
surgery for gynaecological conditions had a 5% likelihood of being re-hospitalized because of
adhesions over the next 10 years. In a considerable proportion of cases adhesions cause infertility,
where adhesions account for up to 40% (39) and where treatment by adhesiolysis leads to an
increase in pregnancy rate as was already observed 30 years ago (40). Infertility (4;39;40-45)
together with pelvic pain (4;31;42;46-48) and intestinal obstruction (34;49-52) results in a reduced
14
quality of life (53) often requiring readmission to hospital and additional more complicated surgical
procedures (53-57) and indeed increased surgical costs (55;58;59).
1.2 Pathophysiology of adhesion formation
1.2.1 The classic model
The pathophysiology of adhesion formation is traditionally viewed as a local phenomenon,
resulting from the surgical trauma to the peritoneal surfaces involving besides the mesothelial cells
also the basal membrane and the sub-endothelial connective tissue. This leads to a local
inflammatory reaction and a cascade of events, exudation (60), fibrin deposition and capillary growth
at the site of injury (61). The extent of adhesions depends on the balance of natural healing
mechanisms requiring fibrin removal. This fibrin is normally rapidly removed by fibrinolysis (62) while
simultaneously the peritoneal repair process is started. Within hours after injury, the injured area is
covered by what is believed macrophages and ‘tissue repair cells, which within 3 to 4 days
differentiate into mesenchymal cells. Repair starts specifically from numerous small islands and the
repair of small or large areas therefore is similar. Given the concept of mesenchymal stem cells, the
discussion about the exact nature of macrophages and tissue repair cells has acquired a new
dimension, whereas the specific mechanism of repair starting from numerous small islands is easily
understood (63). If the normal rapid repair of peritoneal lesions fails or when repair is delayed other
processes which were activated become dominant. Within 4 to 6 days fibroblast proliferation
invading the fibrin scaffold and angiogenesis start, leading invariably to adhesion formation (64). The
importance of the fibrin scaffold between 2 injured surfaces was elegantly demonstrated since
separating these areas by silastic membranes for up to 30 hours abolished adhesion formation (65).
This type of experiments, reinforced the belief that adhesion formation is a local process, that
prevention should aim at separating the surfaces for at least a to 2 days, whereas medical treatment
15
given intravenously or intra-peritoneally has been considered less important since this type of
treatment would have difficulties to reach the injured zone because of local ischemia and since it is
shielded by the fibrin plug. The pathophysiology of this local process has been considered an
inflammatory reaction, with players and mechanisms as fibrinolysis, plasmin activation and PAI’s,
local macrophages and their secretion products and the overall oxygenation of the area or the
absence thereof, driving angiogenesis, fibroblast proliferation and mesothelial repair. The focus on
macrophages and tissue repair cells is changing rapidly given the actual concept to consider these
stem cells.
This concept of a local process is widely supported by reduced adhesions with
administration of antibodies against PAI-1 (66) or fibrinolitic agents (67), neutralizing antiserum
against VEGF, antibodies against PlGF and against VEGFR-1 (68).
1.2.2 The mesothelial barrier and the role of the entire peritoneal cavity
The present understanding of the processes involved in adhesion formation requires a closer
look at the structure and function of the mesothelial cells of the peritoneum, e.g. the morphologic
changes that occur during physiologic or pathologic events in the abdominal cavity.
The mesothelium lines the peritoneal cavity with visceral and parietal surfaces
covering the internal organs and body wall, respectively (69). In the era of stem cells the search for
understanding the origins of the mesothelial cells has been on its way and much evidence has
accumulated to consider that the cells are derived from a progenitor cell originating embryologically
in the splanchnic mesothelium (62). Moreover recently the descriptions of mesothelial stem cells
have shown their ability to differentiate into mesothelial cells, endothelial cells, smooth muscle cells,
myofibroblasts, neuronal cells, adipocytes, chondrocytes and osteoblasts.
The roles of mesothelial cells in maintaining normal serosal membrane integrity and function
is still only partially understood. They secrete glycosaminoglycans and surfactant to allow the parietal
and visceral serosa to slide over each other. They actively transport fluids, cells and particulates
16
across the serosal membrane. They actively modulate gas resorption as CO2 from the
pneumoperitoneum (70;71). Mesothelial cells also produce a multitude of cytokines and growth
factors which can regulate inflammatory responses and stimulate tissue repair as well as play a role
in the immune response. The secretion of pro-coagulants such as tissue factor and fibrin stabilizers
plasminogen activator inhibitor (PAI)-1 and -2, as well as fibrinolytic mediators including the
plasminogen activators (PA) urokinase PA (uPA) and tissue PA (tPA) by the mesothelium,
demonstrates an importance in regulating haemostasis and fibrin clearance (72). Following serosal
injury, there is a fine balance between these processes, which if disrupted may result in the
formation of adhesions, bands of fibrous tissue that occur in up to 95% of patients following surgery.
Thus the uncompromised function of the mesothelial cells does determine the normal sequences of
physiology events within the body.
These special functions of mesothelial cells take place within the abdominal cavity and more
specifically within the conditions provided by the peritoneal fluid. Peritoneal fluid is a specific micro-
environment with protein and hormone concentrations which are much different from plasma
(63;64). Mesothelial cells are highly specialized cells regulating the transport of fluid and proteins,
especially those larger than 20.000 Daltons, between the peritoneal cavity and the blood stream.
Peritoneal fluid contains high amounts of macrophages, which secrete, especially when activated,
such as in endometriosis, a large variety of cytokines and growth factors. The involvement and
mutual effect of the mesothelial cells and peritoneal cavity in this process is obvious but certain
questions like understanding how exactly these mesothelial cells communicate with each other and
with surrounding cells remain unclear.
Mesothelium is now considered to be a dynamic membrane that plays a pivotal role in the
homeostasis of the peritoneal cavity, contributing to the control of fluid and solute transport,
inflammation, and wound healing. The role of the progenitor cells and the relationship between
mesothelial stem cells and peritoneal repair following injury is still unclear. It remains debated
17
whether mesothelial cells originate from the peritoneal fluid, from the mesothelium, from the
submesothelial connective tissue, from the vascular endothelium or from blood cells. In any case, the
concept of mesothelial stem cells is expected to be important for our understanding of peritoneal
repair and of adhesion formation. The development of endoscopic surgery with its CO2
pneumoperitoneum has triggered research on the effect of peritoneal milieu and gas conditioning
upon adhesion formation. During the last decade the entire peritoneal cavity has been demonstrated
to be a cofactor enhancing adhesion formation after local injury. Identified so far are hypoxia of the
mesothelial cells due to CO2 pneumoperitoneum, hyperoxia, hypoxia and desiccation of cells and
increased temperatures. These factors damage mesothelial cells altering mesothelial cell
morphology (73;74) and general structure of the mesothelial layer: trauma to the large and flat
mesothelial cells will induce them to retract as a defence mechanism leading to up to now unknown
exact mechanisms through which adhesion formation is further modulated. We only can speculate
that macrophages and their secretion products, blood constituents or other inflammatory products
affect directly the repair process or the differentiation of stem cells at the injured area.
Pneumoperitoneum has been demonstrated to increase adhesions and this increasing is time and
pressure dependent (25). Peritoneal hypoxia was suggested as a driving mechanism (25;26;75;76)
since pneumoperitoneum-enhanced adhesions could be reduced by addition of 2-4% of O2 to CO2 in
various animal models (rabbits, mice), since the mesothelial layer stains hypoxic (77) and since the
effect is absent in mice partially deficient for hypoxemia inducible factor (HIF1a and in HIF2a). This
concept was supported by the observed effects in mice deficient for PAI-1, VEGF or PIGF (78). An
oxygen concentration over 10%, i.e. a higher partial oxygen pressure than 30-40mm Hg as is
physiologic for mesothelial cells, enhances adhesions probably through an increase in ROS since this
enhanced adhesion formation could be reduced by ROS scavengers (79). Also ischemia-reperfusion
during insufflation/deflation (80;81) could be involved since known to increase ROS. Lower
intraperitoneal temperatures seem attenuate the effect of these adhesion enhancing factors, indeed
that hypothermia decreases adhesion formation in mice was elegantly demonstrated by Binda (82),
18
and in rats, irrigation with saline at >37°C, i.e. between 37 and 60°C, increases adhesion formation
(83).
CO2 is resorbed from the CO2 pneumoperitoneum and has profound metabolic effects
leading to acidosis, later to metabolic acidosis and to metabolic hypoxemia (71;84). This can be
prevented by hyperventilation as is done normally during anaesthesia. The resorbtion rate of CO2
clearly increases over time, while this increase is prevented by adding 2-6 % of oxygen to the
pneumoperitoneum (67). Both observations lend support to mesothelial damage as the driving
mechanism.
1.3 Adhesion prevention today
Adhesion formation between opposing injured peritoneal surfaces are acknowledged to be
different from adhesion reformation following lyses of adhesions and from de novo adhesion
formation outside the areas of surgery. Since, only adhesion prevention for the former has been
investigated adequately, the following paragraphs will not discuss de novo adhesions nor adhesion
reformation.
Microsurgery developed the principles of good surgical practice, such as moistening tissues
by continuous irrigation (and with the knowledge of today cooling), and moistening of abdominal
packs, such as gentle tissue handling introducing glass or plastic rods for mobilization of tissues, and
precise micro-instruments. Only recently these factors were proven experimentally, underlining how
important and accurate these clinical observation were.
Adhesion formation between traumatized areas has traditionally been considered as a local
process and, based on this concept, the only clinically available products for adhesion prevention
today, are barriers in several formulations.
Flotation agents as Ringers lactate are marginally effective and its efficacy has not been
proven whereas Adept (Icodextrin) (53;85;86) a macromolecular sugar with a higher retention time
19
in the peritoneal cavity, was expected and shown to be efficacious in adhesion reduction. The
efficacy claimed to be 40 to 50%. A major advantage is the safety and absence of side effects, which
were well established since extensively used for peritoneal dialysis. The strength of the available
evidence demonstrating efficacy, was in a Cochrane review not considered very solid (87).
Mechanical barriers produced as sheets (Seprafilm, Interceed or Gore Tex) and gels
(Spraygel, hyalobarrier gel, Intercoat) also have been shown some 40 to 50% effectiveness. Thus
sheet barriers as Seprafilm (Hyaluronic acid-carboxymethylcellulose) (88-90), Interceed (Oxidized
regenerated cellulose) (53;86) and Gore-tex (Expanded polytetrafluoroethylene) (93) though proven
to be effective did not become very popular for various reasons. Seprafilm is difficult to use during
laparoscopy whereas to be efficacious any remaining bleeding of the traumatized area should be
avoided.
Since Intergel (0.5% ferric hyaluronate gel) has been withdrawn from the market, only
Hyalobarrier gel (Auto-cross linked hyaluronic acid gel)(94), Spraygel ( Polyethylenglycol) and
Intercoat/Oxiplex (95;96) remain available for clinical use. Overall efficacy appears to be similar for
all 3 products. A comparison between these 3 gels can unfortunately not be made since comparative
trials do not exist. Also the strength of the available evidence varies and a Cochrane review of
hyaluronic acid and spraygel concluded that only for hyaluronic acid the evidence was solid (87).
Most important is that for none of these products efficacy has been proven for endpoints
that really matter, i.e. pain, infertility, bowel obstruction or reoperation rate. We moreover should
realize that large RCT trials were needed because of the high intra-individual variability, and that in
these trials the surgical interventions were limited to rather simple and straightforward interventions
as cystectomy and myomectomy. In addition, these trials have been performed during interventions
performed by laparotomy or by laparoscopy under conditions of CO2 pneumoperitoneum enhanced
adhesion formation and slight desiccation. It therefore is still unclear to what extend the available
results of efficacy can be extrapolated to more severe or other types of surgery, whether highly
20
variable efficacy in adhesions and in prevention, is patient or intervention or surgeon dependent and
whether in the human the effect will be additive to good surgical practice and conditioning of the
peritoneal cavity (97).
A wide range of products has been shown to be effective in animal models. Efficacy of all
products described so far has been extensively investigated in our laparoscopic mouse model. It
should be realized that in this model all criteria of good surgical practice as described are fulfilled,
with standardized lesions, controlled duration of surgery, strict control of temperature and absence
of desiccation. It should moreover be realized that the laparoscopic mouse model, is a model for
three distinct pneumoperitoneum conditions, normoxia, hypoxia and hyperoxia. A first model
intends to minimize any peritoneal damage except for the lesions inflicted to induce adhesions. Thus,
adhesions will form according to the classic model, with little or no effect of the peritoneal cavity. In
this model 4% of oxygen was added to the CO2 pneumoperitoneum to prevent mesothelial hypoxia.
The second model is the ‘hypoxia model, since adhesions are enhanced by CO2 pneumoperitoneum
induced mesothelial hypoxia. In this model pure CO2 was used. In the third model, called hyperoxia
model, 12% of oxygen was added to the CO2 pneumoperitoneum, a concentration known to enhance
adhesions probably by cell damage by reactive oxygen species.
Dexamethasone decreased adhesions by some 30% in the hypoxia model (98), by 60% in the
hyperoxia model, and, especially, by some 76% in the normoxia model when it is combined with low
temperature. ROS scavengers decrease adhesions by 10 to 15% in both the hypoxia and hyperoxia
model, an effect too small to be demonstrated in the normoxia model, with much less adhesions to
start with. Calcium channel blockers decrease adhesion formation by some 35% of inhibition in both
hypoxia (98) and hyperoxia models, and around 58% in the normoxia model when is combined with
low temperature; recombinant Plasminogen Activator (rPA) decrease adhesion formation by 40% in
the hypoxia (99) and normoxia models whereas less inhibition, around 17%, was observed in the
hyperoxia model. Ringers lactate as a flotation agent was marginally but significantly effective (100).
21
The effects of other flotation agents as carboxymethylcellulose (CMC) and Hyskon were marginal (97)
and surfactants as phospholipids gave some 35% of inhibition in the hypoxia and hyperoxia models
and 58% in the normoxia model when is combined with low temperature. Icodextrin, (Adept, 4% α 1-
4 glucose polymer) unfortunately could not be evaluated since it degrades enzymatically in mice.
Barriers as hyalobarrier gel, spraygel and Intercoat were highly effective in all models with a
reduction of adhesions between (56) and 90%.
Prevention of angiogenesis also reduces adhesion formation, as demonstrated in PlGF
knockout mice and by the administration of anti VEGF and anti PlGF monoclonal antibodies (68;101-
105).
The transplantation of cultured mesothelial cells into the peritoneal cavity also is effective in
decreasing adhesion formation (106;107) and in remodelling the area of mesothelial denudation.
More recently, mesothelial cells were used as transplantable tissue-engineered artificial peritoneum
and research is focusing on the use of mesothelial progenitor cells (108).
Prevention of adhesion formation in humans therefore traditionally has focused upon good
surgical practice and has been limited to barriers preventing fibrinous attachments between injured
surfaces or flotation agents with a reduction of adhesion formation that ranges for all products
between 40% to 50%.
Today the key element would be assurance that believed trends coming from the animal
research data can be extrapolated to the human. The extrapolation can be done considering the
following data that the effect of CO2 pneumoperitoneum, the duration dependent increased CO2
resorbtion, observed in mice and in rabbits, also occurs in women.
Taking into account the findings in animal models, good surgical practice,
pneumoperitoneum conditioning, today should include the following. Firstly the insufflation gas
should be conditioned in order to minimize hypoxia and desiccation; this requires humidification of
22
the gas and the altering of the pneumoperitonem by addition of oxygen and or nitrous oxide to the
CO2. Moreover, cooling of the peritoneal cavity is important since cells become more resistant to
metabolic damage at lower temperatures. Secondly the duration of surgery should be kept to a
minimum, as well as the amount of bleeding and the extent of tissue manipulation as will be clearly
showed in this thesis.
Observation of strict sterility remains mandatory to prevent any kind of infection. This simple
statement should be balanced against the observation that it is difficult to completely disinfect the
umbilicus, and that each time the vagina is opened, at least some risk of infection occurs. This is even
more likely with entry into the bowel. Whether extensive lavage following surgery might reduce
adhesion formation or the risk of some minor infection is unknown.
Taken together, these measures of good surgical practice, conditioning of the
pneumoperitoneum, cooling, prevention of inflammation in combination with a barrier, should
reduce adhesion formation by more than 80%.
1.4 The role of nitrous oxide in laparoscopy
Considering the cardiovascular and other side effects of CO2, other gases have been explored
for pneumoperitoneum especially, helium and nitrous oxide (N2O). Helium is an inert gas but has a
low solubility in water. N2O has a solubility and lung exchange capacity similar to CO2.
N2O has been used clinically in laparoscopy for sterilizations or diagnostic procedures and
was shown to have beneficial effects on pain during and after surgery when compared to carbon
dioxide.
The induction of pneumoperitoneum under local anaesthesia, clearly is less or not painful in
comparison with CO2 as evidenced in prospective trials (109;110). The mechanism why exactly N2O
causes little peritoneal pain upon insufflation in contrast to CO2 remains largely unexplained. More
evidence accumulated over the next decades demonstrating that the use of N2O pneumoperitoneum
23
caused less pain after surgery, as evidenced in randomized controlled trials (111-113) and in
observational studies.
The solubility of N2O in blood and the exchange capacity in the lungs is comparable or better
than CO2, thus making it an extremely safe gas, thus avoiding the metabolic effects of CO2 resorbtion
(114-117). Nitrous oxide has besides its anaesthetic and analgesic properties, no cardiopulmonary
side effects of does CO2 (117). As investigated by Wong et al. carbon dioxide caused a slight
worsening of the parietal peritoneal acidosis at higher intra-abdominal pressure (12-15 mmHg),
whereas, nitrous oxide, helium and lifting of the abdomen did not cause parietal peritoneal acidosis
(118;119).
The clinical use of N2O for inducing the pneumoperitoneum during laparoscopy, never
became popular however, because of the explosion risk with concentrations higher than 30% of N2O
(120). It has shown that when N2O is used as part of the anaesthetic, it can reach concentrations in
the peritoneal cavity that can support combustion of bowel gas if there is bowel perforation (121).
The maximum reported concentrations of methane and hydrogen in bowel gas are 56% and 69%,
respectively. The concentration of N2O necessary to support combustion of 56% methane is
approximately 47%. By contrast, the concentration of nitrous oxide needed to support combustion of
69% hydrogen is approximately 29%. Therefore, it is recommended to use less than 30% in order to
avoid this risk of explosion.
1.6 Post-operative pain and inflammation in laparoscopy
Postoperative pain is believed to be caused by the surgical injury of nerve fibres and by the
subsequent inflammatory reaction, a necessary mediator for healing of tissues. Pain after
laparoscopy is believed to be caused for 50% by the surgical incisions of the wall , for some 30% by
the pneumoperitoneum (PP) and for another 20% by the operation wound itself (122). This concept
has lead to the relatively widespread use of local anaesthesia of the skin before making an incision
for laparoscopy. Visceral pain is known to be very different from somatic pain. Visceral pain is indeed
24
more mediated by stretching than by injury, while over 80% of the visceral nociceptors are sleeping.
Following inflammation, the dormant nociceptors are rapidly recruited and the threshold of the
nociceptors is decreased leading to a much increased firing activity to any stimulus including bowel
movements (123).
The exact pathophysiology of postoperative pain, however remains largely unknown. The
speed of recruitment and activation of dormant nociceptors has not been investigated in detail. It is
unclear what the exact time courses of pain are over the first 48 hours after surgery. This is not
surprising given the difficulty to measure pain and the fact that most estimations are indirect.
It is unclear whether, besides pain of the abdominal incision pain after laparoscopic surgery
is less than after laparotomy (124-126). Certainly, after laparoscopic surgery the systemic immune
response is less pronounced in comparison with open surgery, as estimated by levels and duration of
postoperative C-reactive protein (CRP) (127) and interleukin-6 (128), while the exact contribution of
each component to pain and how this fits with the known mechanisms of visceral pain mediation
remains unclear.
25
Chapter 2. Aims of this thesis
Aim 1. To provide additional information on the pathophysiology of adhesion formation and
the role of the peritoneal cavity in our laparoscopic mouse model
a. The effect of manipulation during surgery as a cofactor in adhesion formation
b. The importance of surgeon training upon adhesion formation
c. The effect of bleeding during surgery on adhesion formation
d. The genetic effect on adhesion formation
e. The role of mesothelial cells on adhesion formation/prevention
f. The effect of using barriers on adhesion formation: intercoat gel
g. Post-operative acute inflammation of the peritoneal cavity: a common
mechanism mediating the effect of the peritoneal fluid upon adhesion formation
Aim 2. To establish the optimal peritoneal environment to prevent adhesion formation
during laparoscopy and laparotomy
a. Investigations on N2O and the effect of using mixtures of N2O and O2 in CO2
upon adhesion formation
b. Translation of peritoneal cavity conditioning during laparoscopic surgery to
conditioning in open surgery: equally important for adhesion formation.
26
Aim 3. To translate the observations on adhesion prevention and peritoneal conditioning in
mice into clinical practice: safety and results
a. The desiccation during laparoscopic surgery at different settings of flow rate,
temperature and humidity in humans
b. The effect of cooling the peritoneal cavity and the relation between intra-
peritoneal temperature and core body temperature in humans
c. The effect of using different gases for the pneumoperitoneum on blood CRP
concentrations and upon post-operative pain (pneumoperitoneum with 100% CO2 versus the
use of an altered gas: 96% CO2 + 4% oxygen or 86% CO2 + 4% oxygen + 10% nitrous oxide)
d. The effect of extensive abdominal lavage upon inflammatory parameters
e. To confirm in a RCT that the combination of peritoneal cavity conditioning
and a local therapy results in an over 90% adhesion reduction.
f. The clearance rate of peritoneal fluid after surgery with full conditioning.
27
Chapter 3. General materials and methods
3.1 Animal models
All animal studies of this project were approved by the Institutional Review Animal Care
Committee (n°: P040/2006 and 2010). Animals are kept under standard laboratory conditions and
diet at the animal facilities of the Katholieke Universiteit Leuven (KUL).
3.1.1 The laparoscopic mouse model
Each animal model has advantages and limitations. The main disadvantage of mice is the size.
In addition, the thickness of the entire abdominal wall is equivalent to the first muscle layer in rats
and the same relationship exists between rats and rabbits. The ratio of peritoneal surface area/ body
weight is higher in small than in larger animals, being 0.195 m2/kg in mice, 0.115 m2/kg in rats, 0.084
m2/kg in rabbits and 0.026 m2/kg in humans. The volume of peritoneal fluid is disproportional higher
in small than in larger animals being 18.1 ml/kg in mice, 10.7 ml/kg in rats, 7.7 ml/kg in rabbit and 2.4
ml/kg in humans (129).
The mouse model has also several advantages: they are inexpensive and easy to handle.
Several well characterised types are available, such as inbred strains, genetically manipulated mice,
e.g. mice non-expressing or over-expressing specific genes, mice with altered immune system, e.g.
nude and SCID. Moreover, many specific assays and monoclonal antibodies exist. In addition, mice
and rats, in contrast with rabbit, do not require strict sterile conditions for surgery.
In our laboratory was developed a mouse model in which many parameters were modulated
and/or monitored: conditions of the pneumoperitoneum (duration, pressure, type of gas,
humidification and temperature), body temperature, mouse strain and ventilation (99;130). All these
parameters can have an influence in adhesion development.
28
Adhesion formation increases with the duration of the pneumoperitoneum (25), for this
reason it was decided to maintain it for 60 min and this model is called “pneumoperitoneum-
enhanced adhesion”. The pneumoperitoneum is induced using the Thermoflator (Karl Storz,
Tuttlingen, Germany) through a 2 mm endoscope with a 3.3 external sheath for insufflation (Karl
Storz, Tuttlingen Germany) introduced into the abdominal cavity through a midline incision caudal to
the xyphoid appendix. The incision is closed gas tight around the endoscope in order to avoid
leakage.
The insufflation pressure increases adhesions as demonstrated in rabbits (76) and in mice
(25;131). The insufflation pressure used to induce the pneumoperitoneum in our experiments is 15
mm of Hg, we decided to use this maximum pressure in order to observe maximum effect of
ischemia/hypoxia and since adhesions are pressure dependent. In addition, a pneumoperitoneum
with a pressure of 15 mm Hg could be relevant since it can mimic the conditions that happen in
surgery on humans.
The humidification of the insufflation gas has also to be controlled. When non-humidified gas
is used to induce the pneumoperitoneum, desiccation occurs and adhesions increase 86. In addition,
desiccation is associated with cooling. For humidification, the Storz Humidifier 204320 33 (Karl Storz,
Tuttlingen, Germany) is used.
The type of gas has also an influence in adhesion formation. Adhesion formation decreases
with the addition of 4% of oxygen to both CO2 (25), and increased when higher concentrations than
3% Oxygen were added to the pneumoperitoneum (131). For all these observations, mesothelial
hypoxia and inflammation of the peritoneal cavity was postulated as a cofactor in adhesions and
altered gas, with adding of oxygen and/ or nitrous oxide in different concentration to the CO2
pneumoperitoneum, are also used depending of the endpoint of the experiment.
Proper ventilation, as was clearly showed in previous experiments (132), has an
important role in adhesion formation, in fact, CO2 pneumoperitoneum causes acidosis and
29
hypercarbia and this can be corrected with assisted ventilation. Moreover, it was suggested that
pneumoperitoneum-induced acidosis plays a role in the mechanism of CO2 pneumoperitoneum-
enhanced adhesion formation. For this reason, after anaesthesia of the animals with i.p.
pentobarbital (Nembutal, Sanofi Sante Animale, Brussels, Belgium) with a dose of 0.08 mg/g, animals
are intubated with a 20-gauge catheter and ventilated with a mechanical ventilator (Mouse
Ventilator MiniVent, Type 845, Hugo Sachs Elektronik-Harvard Apparatus GmbH, March-Hugstetten,
Germany) using humidified air with a tidal volume of 250 µl at 160 strokes/min.
Mouse body temperature is critical for adhesion formation and is affected by several
factors, i.e. anaesthesia, ventilation with non-humidified gas, environmental temperature and use of
non-humidified gas to induce the pneumoperitoneum (82;133). Therefore, animals and equipment
are placed in a closed chamber at a preset temperature between 32 and 37 °C (heated air,
WarmTouch, Patient Warming System, model 5700, Mallinckrodt Medical, Hazelwood, MO). The
temperature of the environment in the chamber is measured with Testo 645 (Testo N.V./S.A.,
Lenzkirch, Germany), whereas the temperature of the animal is measured via the rectum with the
Hewlett Packard 78353A device (Hewlett Packard, Böblingen, Germany). All these factors must be
well controlled since it has been demonstrated that hypothermia prevents adhesions (133).
After the establishment of the pneumoperitoneum, two 14-gauge catheters (Insyte-W,
Vialon, Becton Dickinson, Madrid, Spain) were inserted under laparoscopic vision. The adhesions are
then induced by standardized 10 mm x 1.6 mm lesions performed in the antimesenteric border of
both right and left uterine horns and in right and left pelvic side walls with bipolar coagulation (20W),
standard coagulation mode, Autocon 350, Karl Storz, Tuttlingen, Germany). The surgical lesions are
necessary in order to induce adhesions, in fact, it was shown that pneumoperitoneum alone is not
sufficient and de novo adhesions were never presents when the surgical procedure was not
performed (104).
30
Finally, the genetic background has an influence in adhesion formation. Some mouse
strains developed more adhesions than others and, in addition, outbreed strains have more
variability in the results than inbred strains (134), therefore, to have the maximum of adhesions with
less variability 9-10 weeks-old female BALB/c OlaHsd mice inbred strain are generally used. Swiss,
BALB/cByJ and BALB/cJ@Rj mice were also used when we wanted to compare different mouse
strain.
Since scoring adhesions on day 7 or day 14 or day 28 (25), gave similar results scores we
systematically performed day 7 since more practical. Adhesions are scored quantitatively
(proportion) and qualitatively (extent, type, tenacity) blindly by laparotomy under a
stereomicroscope. The qualitative scoring system assessed: extent (0, no adhesions; 1, 1–25%; 2, 26–
50%; 3, 51– 75%; 4, 76–100% of the injured surface involved, respectively), type (0, no adhesions; 1,
filmy; 2, dense; 3, capillaries present), tenacity (0, no adhesions; 1, easily fall apart; 2, require
traction; 3, require sharp dissection). The quantitative scoring system assessed the proportion of the
lesions covered by adhesions (135) using the following formula: (sum of the length of the individual
attachments/length of the lesion) ×100. The results are presented as the average of the adhesions
formed at the four individual sites (right and left visceral and parietal peritoneum), which are scored
individually. The entire abdominal cavity is visualized using a xyphopubic midline and a bilateral
subcostal incision. After the evaluation of ports sites and viscera (omentum, large and small bowels)
for de novo adhesions, the fat tissue surrounding the uterus is carefully removed. The length of the
visceral and parietal lesions and adhesions are measured. Adhesions, when present, are lysed to
evaluate their type and tenacity. The terminology of Pouly is used (5), describing de novo adhesion
formation for the adhesions formed at non-surgical sites, adhesion formation for adhesions formed
at the surgical site and adhesion reformation for adhesions formed after the lysis of previous
adhesions.
31
Figure 1. The laparoscopic mouse model from Binda, Fertil & Steril (82)
3.1.2 The mouse model for open surgery
All the factors validated for laparoscopic surgery such as animals, anaesthesia, ventilation,
humidification and temperature control, and the general experiment set up were kept identical for
the mouse model in open surgery.
Figure 2. The mouse model for open surgery (Corona, modified from Binda).
32
Only the surgical approach and the conditioning of the peritoneal cavity were different. The
model was developed after several pilot experiments as explained in the chapter 3 of this thesis. In
summary, after anaesthesia and intubation, the mouse is placed in a transparent plexiglass box
measuring 22cm x10cm x30cm, with 3 ports: one for the ventilation tube, one for the inflowing gas
and one port for the out flowing gas. The box is closed with a sliding transparent cover that could be
removed to perform the surgery or other treatments (figure 2). A 5 mm diameter tubing is connected
to the box for the inflowing gas (Thermoflator (Karl Storz, Tuttlingen, Germany) in the lower part of a
lateral wall and since density of CO2 and N2O is higher than air, the box is filled progressively until the
gas escapes by overflow. All further procedures thus are performed in the specific gas environment
either air, or CO2 or a mixture of CO2 with oxygen of N2O. A midline xyphopubic incision is performed
and the abdomen is kept open and exposed to the environment with 2 pins. As performed in the
laparoscopic mouse model, standardized 10 mm x 1.6 mm lesions are performed in the
antimesenteric border of both right and left uterine horns and in both the right and left pelvic side
walls with bipolar coagulation (20W, standard coagulation mode, Autocon 350, Karl Storz, Tuttlingen,
Germany). After surgery the mouse is kept with the abdomen open inside the box under various
environmental conditioning for 30 minutes. After 7 days, the same scoring system is used as in the
laparoscopic mouse model.
3.2 Humidification, desiccation and temperature during endoscopic surgery
3.2.1 In Vitro experiments: a box mimicking the peritoneal cavity
Knowing the importance of temperature and desiccation for adhesion formation, and the
association of desiccation cooling, we investigated the quantitative relationship between flow rate,
temperature and RH of the insufflation gas upon intraperitoneal temperature, and RH together with
the effect of cooling of the peritoneal cavity with a third means.
33
For all the studies are used the Thermoflator (Karl Storz GmbH & Co.KG, Tuttlingen,
Germany) for insufflation and either the Storz humidifier, model 204320 33 (Karl Storz GmbH &
Co.KG, Tuttlingen, Germany), the Fisher & Paykel Healthcare (F&P) humidifier MR860 (Panmure,
Auckland, New Zealand), and a modified F&P humidifier for humidification of the gas. The Storz
humidifier blows gas over water warmed to 37°C and uses non insulated tubing, 2.7 m in length and
7 mm in diameter (Kendall, Covidien, Mansfield, MA). The F&P humidifier blows gas through a
heated water chamber. In order to avoid condensation, the tubing that delivers the gas was heated
to over 40°C with a heating wire. Heating in the heating chamber is adapted in order to deliver gas
with 100% RH at 37°C. The modified F&P humidifier, modified by eSaturnus, permits higher flow
rates than the original device by removing the luer lock. Moreover, the heating of the chamber and
tubing were continuously adapted by an electronic feedback loop to maintain 100% RH at the end of
the tubing but a preset temperature between 25 and 36°C at flow rates between 0.5 and 30 l/min.
A box was used to mimic the peritoneal cavity in the in vitro experiments: the bottom
and side walls of a closed polystyrene box were covered with plastic bags containing a total of 6 litres
of water at 37°C (Figure 3). Flow rate, RH and temperature of the gas at the end of the insufflation
tubing, i.e. entering the box, and the gas flowing out of the desiccation box were measured twice a
second; temperature and RH were captured with a digital sensor (SHT75, Sensirion AG, Zurich,
Switzerland). This permitted calculation of water loss, a marker for desiccation, in real time. The
quantitative relationship between the amount of water hold by a gas, the relative humidity and the
temperature are expressed by the formula: humidity (g/m³) = RH (%) / 100 x (3.1243 10E-4 x T³ +
8.1847 x 10E-3 x T² + 0.32321 x T + 5.018) where T is the temperature in °C; between 20 and 40°C.
Cooling by a third means of the peritoneal cavity was accomplished by nebulizing 3 ml/min of water
at room temperature or at 0°C with a nozzle set at 2 bar entry pressure. This
cooling/nebulization/humidification device had a diameter of less than 5 mm so that it could be used
through a standard 5 mm trocar.
34
In order to validate measurements, a “desiccation recipient” with protruding moistened
towels was placed inside the box in order to compare direct water loss, measured by measuring the
weight of the desiccation recipient, with the water loss calculated form the measurements of flow,
RH in, Temp in RH out and temp out. Because of the importance this was performed under different
conditions and flow rates. It was assumed that temperature and RH coming out of the box would
reflect temperature and RH inside the box. When the tubing was used for non-humidified CO2 or for
gas humidified with the Storz humidifier, the gas was permitted to cool to room temperature or was
heated to 37°C by putting the tubing inside a heated chamber.
3.2.2 In vivo set up
Before to start the RCT designed to demonstrate that, as in the mouse model, cooling and
humidification of the peritoneal cavity could have beneficial effects on adhesion formation and
postoperative pain in humans (Institutional Review Board approval has been granted for the RCT and
registration through clinicaltrial.gov was performed: NCT01344486), several investigations were
conducted as pilot experiments for a RCT.
Since the measurements of flow rate, RH, and temperature were accurate, as judged by the
linear relationship between the calculated and measured water loss obtained with the in vitro
experiments, the in vitro set up (figure 4) was translated to the clinical practice. Surgery involved no
unusual risks since all the instruments used were standard. The only difference was that the relevant
parameters were measured continuously to evaluate desiccation and temperature in the abdomen.
The cooling device posed no risk, since administering saline, 2-4 ml/min, was less intensive than the
normal practice of saline irrigation at room temperature when as much as 8 L might be used. In order
to permit higher flow rates, the luer lock limiting flow rates to 7.8 L/min at 15 mm Hg insufflation
pressure present in the original F&P humidifier. In order to avoid condensation, the tubing delivering
the gas is heated to over 40°C with a heating wire.
35
Figure 3. In vitro setup, « desiccation box », Corona, Am J Obstet Gynecol.
Figure 4: In vivo set up, Corona, Am J Obstet Gynecol.
36
In this F&P humidifier, modified by eSaturnus, the heating of the chamber and the heating of
the tubing was continuously adapted by an electronic feedback loop in order to maintain at the end
of the tubing 100% relative humidity (RH) at a preset temperature between 25°C and 36°C and this at
any flow rate between 0.5 and 30 L/min.
For our in vivo study, we used a previously-described endoscopic surgery setup, which
consisted of an umbilical metal trocar (Karl Storz GmbH & Co.KG, Tuttlingen, Germany) with a 7-mm
side-opening and 3 secondary 5-mm disposable trocars. When a straight laparoscope was used,
insufflation was performed through the umbilical trocar while aspiration was done from a secondary
trocar. When an operative laparoscope (Karl Storz GmbH & Co.KG, Tuttlingen, Germany) with a large
7-mm side-opening and a CO2 laser were used, insufflation was done through the laparoscope, which
prevented blooming of the CO2 laser beam, and aspiration was performed at the umbilical trocar. For
insufflation, the intraperitoneal pressure was maintained at the preset 15 mmHg for flow rates up to
30 L/min. The outflow of gas was regulated by opening and closing a 3-way valve.
All women included in the study underwent surgery for deep endometriosis using the
standard setup. After about 15 min of surgery, different flow rates (2.5, 5, 7.5, 10, and 15 L/min)were
evaluated. Subsequently, continuous flow rates between 7-10 L/ min were used to remove smoke, as
is standard for our CO2 laser operations. During the entire procedure, either flow rates, by a sensor
provided by eSaturnus especially for these studies, temperature-in, RH-in, temperature-out, and RH-
out, by a sensor for temperature and relative humidity (model SHT75, Sensirion, Switzerland), were
measured; water-in, water-out, and desiccation were calculated twice a second with MatLab
software. Non humidified CO2 was used in 3 procedures, humidified CO2 via the Storz humidifier was
used in 4 surgeries, and humidified CO2, provided by the modified F&P humidifier, was used in
another 4. At least 3 operations were recorded, and at the end of surgery, the effect of the cooling
device upon intraperitoneal temperature was assessed. Core body temperature, via oesophageal
temperature, was continuously monitored by the anaesthesiologist.
37
3.3 Clinical studies
All patients for the clinical trials were recruited at the University Hospital Leuven, campus
Gasthuisberg. Written and informed consent was obtained for all the patients.
Inclusion criteria always consisted of women undergoing gynaecologic laparoscopic surgery
for at least 60 minutes of planned laparoscopy time and signed informed consent.
Exclusion criteria always mentioned: pregnancy and women with a pre-existing condition
increasing the risk of surgery such as clotting disorders, heart or lung impairment. All conditions that
might interfere with inflammatory parameters such as immunodeficiency, chronic inflammatory
disease as Crohn’s disease were also exclusion criteria. Approval of the ethical committee from the
University Hospital Leuven was obtained for all clinical trials.
Clinical trials were registered through clinicaltrials.gov:
NCT00678366: Evaluation of Adding Small Amounts of Oxygen to the CO2 Pneumoperitoneum Upon
Pain and Inflammation
NCT01344499: Clearance Rate of Peritoneal Fluid after Full Conditioning Pneumoperitoneum
NCT01344486: Peritoneal Cavity Conditioning Decreases Pain, Inflammation and Adhesions
3.3.1 Adding 4% oxygen to the pneumoperitoneum
Initial clinical trials where performed with 4% of oxygen added to the carbon dioxide
pneumoperitoneum the Thermoflator plus® (Karl Storz AG, Tuttlingen, Germany) was used for
insufflation. This insufflator allowed a manual regulation of the amount of oxygen added to the
carbon dioxide pneumoperitoneum from 1 to 12%.
3.3.2 Full conditioning of the peritoneal cavity
38
After we had realised the importance of N2O and the marginally additive effect of O2, the
importance of avoiding desiccation and of cooling the peritoneal cavity, and the importance of fibrin
deposition for adhesion formation, we decided to define full conditioning as follows. As gas 86%
carbon dioxide +10% nitrous oxide + 4% oxygen was used (premixed bottles with a certified correct
mixture). Since the gas had the same composition at the beginning and at the end of the bottle, no
different composition of the gas occurred over time (Ijsfabriek Strombeek). In addition to the gas
mixture an the insufflation temperature at 32°C with 100% RH using the eSaturnus modified Fisher
and Paykel (F&P) humidifier (Type: MR860 AEA, SN: 070514000016, Fisher & Paykel Healthcare®,
Auckland, New Zealand) together with cooling (patent nr WO2005117779A1, US2008039911, LRD,
KUL) of the peritoneal cavity with a third mean using nebulization of 2-3 ml/min Hartman solution at
RT in order to ensure that upon entrance of gas in the peritoneal cavity some cooling and
condensation occurs. In addition the sprinkled cooling fluid contained 5000 IU of Heparin/litre in
order to avoid fibrin deposition. The use of humidified, mixed (86% CO2 + 4% O2 + 10% N2O) gas at
32°C with the use of the sprinkler is referred to as “full conditioning”.
3.4 Statistics
Statistical evaluation was done with Graph Pad Prism (Graph Pad Software Inc., San Diego,
CA) or with the SAS system (SAS Institute, Cary, NC). As appropriate paired and unpaired t-tests for
normally distributed populations or Wilcoxon-Kruskis-Wallis, (proc n-par1way) for non-normally
distributed populations, and Spearman or Pearson test for correlations.Two way analysis of variance
was done with two way ANOVA or proc GLM, for normally and not normally distributed populations
respectively. For all, animal and human experiments, if possible, a factorial design was used. Indeed a
strictly two x two factorial design (latin cross) generates almost similar significances and power for
two variables with in addition interaction, for almost half the number of mice as would have been
required when each factor would have been investigated separately (136)
39
Chapter 4. Results
4.1 Animal experiments
4.1.1 Results of aim 1: The key role of the entire peritoneal cavity and the importance of the quality of
surgery on adhesion formation in mice.
Introduction
We realized the importance of the surgical quality already in 1996, when Ordonez, with a key
experiment in rabbits 1996 (75), demonstrated that when an inexperienced surgeon started to do
laparoscopic nephrectomies in rabbits, the amount of adhesions decreased progressively and as did
the duration of surgery. Since the duration of surgery reflected both the duration of CO2
pneumoperitoneum and the surgical skills both factors were investigated separately. The duration of
CO2 pneumoperitoneum as a co-factor in adhesion formation was explored in detail by “experienced
surgeons” in subsequent experiments and this resulted in the discovery of the relative importance of
mesothelial hypoxia, hyperoxia, desiccation and temperature and dexamethasone. Since the
importance of surgical experience could only be investigated by inexperienced surgeons, all new
investigators had to perform a learning curve of 80 mice- (i.e. with a constant pneumoperitoneum of
60 min) prior to do any experiments. The results were recently published (4.1.1 b). When I and Ron
Schonman arrived at the lab we decided to do the learning curve as a prospective randomized trial
evaluating in addition the effect of adding 4% of oxygen and dexamethasone (4.1.1 a).
40
a. The effect of manipulation during surgery as a cofactor in adhesion formation
Effect of Upper Abdomen Tissue Manipulation on Adhesion Formation between Injured Areas in a Laparoscopic Mouse Model. Schonman R, Corona R, Bastidas A, De Cicco C, Koninckx PR. J Minim Invasive Gynecol. 2009 May-Jun;16(3):307-12.
Experimental design.
Experiment I: (n=80 mice) was performed by inexperienced surgeons to evaluate adhesions
formation during sequential experiments of adhesion induction. The design of the experiments was
factorial permitting a 2 way analysis of variance of adding 4% of oxygen to the pneumoperitoneum
and of dexamethasone. In each block of 4 animals, animals were randomised to either the induction
of intraperitoneal adhesions without any further treatment (Group I). In group II, two doses of 40
mcg dexamethasone, were given. In group III, 3% oxygen was added to the insufflating gas. Group IV
received both treatments of groups 2 and 3, i.e. the addition of 3% oxygen to the insufflating and two
doses of 40 mcg dexamethasone.
Results
The results showed an obvious learning curve in the control group with an exponential
decline in adhesion formation, whether evaluated as proportion or as total amount of adhesions (1
way analysis of variance, P<0.001 for both). Dexamethasone decreased adhesion formation whether
evaluated as adhesion proportion (P=0.0001), as total adhesion score (P=0.0009), extension
(P<0.0001), type (P=0.0063) or tenacity (P=0.007). Interestingly, during dexamethasone treatment
the learning curve almost disappears (Fig. 5) which led to the hypothesis that dexamethasone
eventually had a specific effect on the effect of manipulation. The decreasing of adhesions with the
addition of 3% of oxygen was not significant (P=0.0658) probably as a consequence of the high
variability during the learning curve.
41
0 5 10 15 200
10
20
30 ControlOxygenDexametasone + O2Dexametasone
Blocks
Prop
ortio
n of
adh
esio
ns (%
)
Figure 5 – Learning curve in adhesion formation model influenced by dexamethasone and intraperitoneal oxygen treatment.
Experiment II (n=32 mice) was planned to evaluate specifically the effect of manipulation
upon adhesions formation by letting experienced surgeons touch and grasp tissue in the upper
abdomen. Group I (n=8) was the control group were the standard model for induction of adhesions
was used i.e. 60 minutes of CO2 pneumoperitoneum plus surgical lesion with bipolar of lateral wall
and uterine horn bilaterally. In groups II(n=8) and IV(n=8) adhesions were further increased by
manipulation of the peri tubal fat and large and small bowel up that were moved up and down the
abdomen for 5 and 10 minutes respectively, with the shaft of a 1.5 mm grasper and the shaft of the
bipolar electrode. In group III (n=8 mice), the peritubal fat and small bowel were grasped and
manipulated repetitively for 5 min with 1.5mm non traumatic grasper.
Manipulation of omentum and bowels increased adhesions and the effect increased with
time. Groups II, III and IV, had higher adhesions proportion than the control group (t-test, P=0.0036,
P=0.0008 and P<0.0001 respectively). The effect of manipulation increased with duration, i.e. 5 min
versus 10 min, for adhesion proportion and adhesion extension (P=0.0301 and P=0.0241 respectively,
prog GLM). Also grasping increased adhesions but the effect was not more pronounced than
manipulation (Fig. 6).
42
0
20
40
60
80
Control Touching 5min
Grasping 5min
Touching 10min
19.15±5.1
42±14.7
46.66±8.6
62.77±6.04
Prop
ortio
n of
adh
esio
ns (%
)
Figure 6 – Adhesion proportion in mice model after manipulation of tissue.
Experiment III (n=32 mice) was planned to evaluate specifically the effect of dexamethasone
upon tissue manipulation. Group I was the control group, i.e. CO2 pneumoperitoneum enhanced
adhesions. Group II received two doses of 40 mcg dexamethasone, one immediately after the end of
pneumoperitoneum and another 24 hours later. In group III a similar manipulation as in experiment II
of the peri tubal fat and large and small bowel was performed for 5 minutes. Group IV received a
similar manipulation for 5 minutes as group III but in addition dexamethasone as in group 2.
This experiment confirmed that manipulation increased adhesions (2 ways analysis of
variance, P<0.0001) and that dexamethasone reduced adhesion formation (2 ways analysis of
variance, P=0.0001) in both the control group, i.e. CO2 pneumoperitoenum enhanced adhesions, and
in the group with an addition manipulation enhanced adhesion (Fig. 7). Qualitatively, the adhesion
scoring for extension, type, tenacity and total score were lower under dexamethasone treatment
(P<0.0001, <0.0001, P<0.0001and P<0.0001, respectively, prog GLM) and were significantly higher
with grasping manipulation for 5 min (P<0.0001, P<0.0001, P<0.0001 and P<0.0001, respectively,
prog GLM). While the effect of dexamethasone was proportional to the adhesions in the control
43
group, the manipulation enhanced adhesion formation remains more important than in the control
(P=0.008).
*P<0.0001 for manipulation effect (Two way ANOVA); **P=0.0001 for dexamethasone treatment (Two way ANOVA)
Figure 7 – Influence of dexamethasone and manipulation on adhesion formation
Conclusion. Manipulation in the upper abdomen increases adhesion formation between opposing
lesions in the lower abdomen. The effect of dexamethasone is not specific for manipulation
enhanced adhesions.
b. The importance of training upon adhesion formation
The impact of the learning curve upon adhesion formation in a laparoscopic mouse model.
R Corona, J Verguts, , MM Binda, R Schonman, CR Molinas, PR Koninckx. Fertil Steril. 2011 Jul;96(1):193-7.
Experimental design
The experiment was designed to evaluate the effect of training, assessed by the effect of
consecutive surgeries, upon operating time and upon the extension of post-operative adhesion
formation. The operating time was measured from T20, i.e. from the beginning of surgery, exactly 20
44
min after induction of anaesthesia, until the end of the procedure with the removal of catheters for
instrumentation and port sites closure. Adhesions were scored blindly and the codes of intervention
order were broken only at the end of the training experiment. Each trainee performed sequentially
80 interventions to induce adhesion formations. All trainees were inexperienced for the laparoscopic
procedure in a mouse model, but some of them were experienced gynaecologists after their training
in gynaecology (3 trainees) whereas others were inexperienced at the end of medical school (2
trainees).
Results
With training, assessed by consecutive surgeries, duration of surgery decreased exponentially
in both groups, demonstrating a clear learning curve that can be described as a two-phase
exponential decay model (non-experienced: R2=0.72; experienced: R2=0.73) (Fig.8).
Figure 8: Duration of surgery during training. Learning curve expressed as duration of the surgery vs. number of consecutive
procedures for experienced (n=3) and non-experienced surgeons (n=2). Dots represent the real operating times for each mouse. The lines
were calculated from the individual operating times following a two-phase exponential decay model.
45
In Balb/c mice the real and calculated operating times for non-experienced surgeons
decreased from 19 ± 6 min and 22 ± 4 min for the first ten procedures to 7 ± 1 min and 7 ± 1 min for
the last ten procedures (P=0.0001; P=0.0001, respectively). For experienced surgeons time decreased
from 8 ± 1 min and 10 ± 3 min for the first ten procedures to 4 ± 0.3 min and 4 ± 0.03 min for the last
ten procedures. In Swiss mice the real and calculated operating times from non-experienced
surgeons decreased from 17 ± 6 min and 22 ± 4 min for the first 10 procedures to 7 ± 1 min and 7 ± 1
min for the last 10 procedures (P=0.0001; P=0.0001, respectively). For experienced surgeons time
decreased from 8 ± 3 min and 10 ± 3 min for the first ten procedures to 4 ± 0.3 min and 5 ± 0.03 min
for the last ten procedures (P=0.0001; P=0.0001, respectively). As expected, the real and calculated
operating times both for Balb/c and Swiss mice were not statistically different (T test) both for the
first 10 procedures and for the last 10 procedures. When the effects of training upon time, the
experience of the surgeon and the mouse strain were evaluated simultaneously using multifactorial
analysis (proc GLM), we confirmed the importance of decreasing operating time (P= 0.0001) and the
effect of experience (P=0.0001), whereas the mouse strain was not important as could be anticipated
(P=NS) (Fig. 9).
Figure 9. Adhesion formation expressed as proportion of adhesions vs. number of consecutive procedures in both experienced
(n=3) and non-experienced surgeons (n=2). Number of procedures was grouped by groups of 10 blocks (1 block= 2 procedures: 1 Balb/c + 1
Swiss). Mean and SD are indicated.
46
Similarly, the effects of training (expressed by the consecutive number of surgeries),
operating time, experience and mouse strain upon adhesion formation were evaluated using multi-
factorial analysis (proc GLM) (Fig. 9 and 10). Adhesion formation was lower with the consecutive
number of surgeries (proportion: P= 0.01; total: P= 0.01; extension: P= 0.02; type: P= 0.01; tenacity:
P= 0.006), for surgeries of shorter operating time (proportion: P= 0.02; total: P= 0.04; extension: P=
0.02; type: P= 0.02; tenacity: P= 0.04), among experienced surgeons (proportion: P< 0.0001; total: P=
0.04; extension: P= 0.001; type: P= 0.03; tenacity: P= 0.04), and in Swiss mice (P= NS) (Fig. 10).
It was decided arbitrarily to make 4 groups of ten blocks (G1, block 1-10; G2, block 11-20; G3,
block 21-30; G4, block 31-40). With this grouping, the inter-animal variability for operating time and
for adhesion formation was assessed using the coefficient of variation (CV). The CV of operating time
among non-experienced surgeons were 41%, 35%, 25% and 20% for Balb/c mice and 39%, 40%, 22%
and 17% for Swiss mice for G1, G2, G3 and G4, respectively; among experienced surgeons were 45%,
24%, 20% and 22% for Balb/c mice and 37%, 24%, 20% and 21% for Swiss mice for G1, G2, G3 and
G4, respectively. The CV of the proportion of adhesions were 62%, 48%, 54% and 38% for Balb/c mice
and 106%, 92%, 96% and 77% for Swiss mice for G1, G2, G3 and G4, respectively. The effects of the
consecutive number of surgeries, measured by groups, and of mouse strain upon operating time and
its CV and upon the proportion of adhesions and its CV were evaluated with a multifactorial analysis
(proc GLM). Both operating time and its CV decreased with the number of surgeries (P=0.0001;
P=0.0001, respectively), whereas they were not affected by mouse strain (P=NS; P=NS, respectively).
Both the proportion of adhesions and its CV decreased with the number of surgeries (P=0.002; P=NS)
and were affected by mouse strain, i.e. for Balb/c mice the proportion of adhesions was higher and
the CV was lower.
47
Figure 10: Duration of the surgery and adhesion formation during learning curve in different mouse strain. Duration of the
surgery (black) and adhesion formation (red) during learning curves in Balb/c (n=200) and Swiss mice (n=200). Number of procedures was
grouped by blocks of 10. Mean and SD are indicated.
Conclusion. This study confirms that surgical training is extremely important, not only in a hospital
setting when dealing with patients but also when performing surgical studies with the aim of
analyzing the outcome. In this case, training experience has a marked effect on the development of
postoperative adhesion formation.
c. The effect of blood on adhesion formation
The addition of 10% N2O to the CO2 pneumoperitoneum strongly decreases the dose dependent adhesiogenic effect of blood in a laparoscopic mouse model.
Roberta Corona, Karina Mailova, Maria Mercedes Binda, Jasper Verguts and Philippe R. Koninckx. Submitted
Introduction
Based upon the role of fibrin and fibrinolysis in adhesion formation, good haemostasis during
surgery has become considered of prime importance to decrease adhesion formation. Direct
evidence, however, is to the best of our knowledge lacking, which is not surprising considering
48
that blood and fibrin deposition could be important not only at the lesion site but blood could
also induce acute inflammation of the entire peritoneal cavity.
Therefore, this study was designed to evaluate in our laparoscopic mouse model the effect of blood
or of its constituents upon adhesion formation when added to the peritoneal cavity. In addition, we
wanted to confirm that decreasing acute inflammation through peritoneal conditioning also
decreased adhesion formation induced by blood.
Materials and methods
Mice blood collection. Blood was obtained from the retro-orbital senus with a Pasteur
pipette or by cardiac puncture from anesthetized mice before each experiment. Blood was collected
into a heparinised tube and centrifuged for 10 min at 1000 RPM at 4°C to separate plasma and RBC.
The pellet containing the RBC was resuspended in 1 ml of a isotonic solution (145 mM NaCl, 1 mM
CaCl, 5 mM d-glucose, 10 mM MOPS; pH 7.4). Blood or the plasma or the RBCs obtained from 1 ml of
blood were injected intra-peritoneally at the end of the pneumoperitoneum. Blood, RBC and plasma
had been kept at 4°C until used.
Following some pilot observations that the addition of blood in the peritoneal cavity at the
end of the pneumoperitoneum increased adhesions (99) the I experiment was designed to confirm
that observation and to evaluate whether the adhesiogenic effect was mediated by plasma or by the
red blood cells. In all the groups, PP was induced for 60 min with pure CO2 and bipolar lesions were
performed. In the Group I, the control group, no blood was added. The other groups received at the
end of the PP, 1 ml of blood (group II), of plasma (group III) or of RBCs coming from 1 ml of blood
(group IV) (5 mice/group).
The second experiment was designed to evaluate the effect of the amount of blood during 60
min of pure CO2 pneumoperitoneum in order to mimic normal laparoscopic surgery. Besides the
control group without any blood added, 0.125, 0.25, 0.5 and 1 ml of blood were i.p. injected
49
respectively (n= 5/group). In order to confirm the effect of peritoneal conditioning upon adhesion
formation, CO2 with either with 10% of N2O, or 4% of O2 and with both 10% of N2O + 4% of O2 was
compared to pure CO2 PP. In addition, the effect of adding 0.5 ml of blood upon adhesion formation
was also evaluated using these three gas mixtures.
Results
The I experiment demonstrated that adhesions increased by the addition of blood, plasma or
RBCs in comparison to the control group (p < 0. 0001, p < 0. 0001 and p=0. 0103, respectively). Blood
is more adhesiogenic than plasma or RBC only (p < 0. 0001 for both comparisons) and plasma
increases adhesions more than RBCs (p < 0. 0001) (Fig. 11).
Control Blood Plasma RBC0
10
20
30
40
50
60
70
80
90
100
RBC
ControlBloodPlasma
***
***
*
*** p < 0. 0001, * p=0. 0103, t student test vs control
Prop
ortio
n of
adh
esio
ns (%
)
Figure 11: Effect of adding blood or its components to the pure CO2 pneumoperitoneum on adhesion formation. Bipolar lesions
were performed during 60 min of pneumoperitoneum with 100% CO2 and 1 ml of blood, plasma or RBCs were added respectively. Mean
and SEM are showed.
II Experiment. The dose response curve demonstrated that as little as 0.125 ml of blood
increased adhesions (p < 0. 0001), and that the adhesiogenic effect increased exponentially with the
amount of blood up to 0.5ml (p < 0. 0001 for every amount). Adding 1 ml of blood in comparison
with 0.5 ml had little additive effect (NS) (Fig. 12).
50
In the control group we confirmed the adhesion reducing effect of using for the
pneumoperitoneum, CO2 with 10% of N2O, with 4% of O2, or with both 10% of N2O +4% of O2 (p < 0.
0001, p = 0. 009, and p < 0. 0001 respectively). Clearly the addition of 10% of N2O was the most
effective, with little additional effect of adding also 4% of O2. Also, after the addition of 0.5ml of
blood, the use of these three gas mixtures decreased adhesions in comparison with the group with
100% CO2 (p < 0. 0001 for the three comparisons). The addition of 10% of N2O was more effective
than 4% of O2 with little additive effect of adding 4% of oxygen to the 10% of N2O (NS(Fig. 12)) .
Figure 12. Dose response curve and effect of using different PP compositions upon adhesion formation. The upper graphs
shows the dose response curve of adding blood after performing the bipolar lesions during 60 min of pneumoperitoneum with 100% CO2.
The lower graph shows the effect of replacing the 100% CO2 pneumoperitoneum with CO2 with 10% of N2O, 4% of O2 or both 10% of N2O
and 4% of O2, in the control groups (without blood) and after the addition of 0.5 ml of blood; mean and SD are shown.
51
Conclusion. Blood in the peritoneal cavity increases adhesions and the effect is mediated both by
plasma and RBC suggesting fibrin and inflammatory reaction to be involved. The adhesion reducing
effect of N2O is confirmed in blood enhanced adhesions.
d. The genetic effect on adhesion formation
Adhesion formation varies with the substrains of inbred BALB/c mice. Binda M.M, Corona R, Casorelli A and Koninckx PR.
Article in preparation.
Introduction
It is known that substrains can be spontaneously generated from a known strain and this can
give variability in the experiments. To the best of our knowledge, this has not been explored yet in
any adhesion formation model. Three companies providing BALB/c mice were found in Belgium and,
then, we raised the question “how identical” these three BALB/c substrains (BALB/cByJ,
BALB/cOlaHsd and BALB/cJRj) were. Therefore, the aim of this study was, firstly, to investigate
whether these small differences in the genetic background can have an influence upon postoperative
adhesion. Secondly, since we demonstrated that manipulation of intraperitoneal organs in the upper
abdomen enhances adhesion formation at trauma sites (137), we wanted to check if the
manipulation enhanced adhesions has the same effect in these three BALB/c substrains.
Materials and methods
For all groups, a CO2 pneumoperitoneum was induced for 60 min. BALB/cOlaHsd (group I and
II), BALB/cByJ (group III and IV) and BALB/cJRj (group V and VI) substrains were used (6 mice per
group, n=36). For the manipulation enhanced adhesions, extra five minutes of manipulation were
added in groups II, IV and VI. After performing the bipolar lesions, the grasper was removed from the
left side and the incision closed by suturing. To avoid a possible effect of the light, the light source
was decreased till the minimum which allow to see the abdominal cavity and 5 min of manipulation
52
was done with the plastic tip of the right catheter by touching the upper abdomen (the metal sharp
tip was kept at the interior of the catheter in order to avoid bleeding). After the manipulation, the
catheter was removed, the hole closed and the light turned off.
Other parameters were carefully registered during the experiment since they might have an
effect upon adhesion formation: body weight, body temperature and intestinal obstruction during
the second look. First, since animals were breeding in three different places, mouse body weight
might vary. Second, body temperature was measured every 20 min since we previously
demonstrated that hypothermia reduced adhesion formation (133).
Results
The effects of all the variables analyzed simultaneously (substrains, manipulation, mouse
body weight, mean of temperature during PP and intestinal obstruction) upon postoperative
adhesions were first evaluated using multifactorial analysis (proc Logistic). Adhesion formation is
influenced significantly by the substrain used (proportions: p<0.01; extent: p<0.03; type: NS;
tenacity: NS; total: p<0.04) and by the extra manipulation (proportions: p<0.003; extent: p<0.01;
type: p<0.01; tenacity: NS; total: p<0.02). Mouse body weight, mean of temperature during PP and
intestinal obstruction do not have a significant effect upon postoperative adhesions (NS all
comparisons).
Using also a multivariate analysis (proc Logistic), we compare specifically the effect of the
substrains, manipulation, body temperature during PP and intestinal obstruction upon adhesion
formation in two substrains. Comparing the substrains BALB/cOlaHsd and BALB/cByJ (groups I, II, III,
IV together), adhesion formation is influenced by the substrain itself (proportions: p<0.02; extent:
<0.05; type: NS; tenacity: NS; total: p<0.01) and by the manipulation (proportions: p<0.005; extent:
p<0.01; type: p<0.008; tenacity: NS; total: p<0.02). Intestinal obstruction and body temperature
during PP do not have a significant effect upon adhesion formation (NS all comparisons).
Comparing the substrains BALB/cOlaHsd and BALB/cJ@Rj (groups I, II, V and VI together),
adhesion formation is influenced by the substrain itself (proportions: p<0.01; extent: p<0.02; type:
53
p<0.05; tenacity: NS; total: p<0.01) and by the manipulation (proportions: p<0.002; extent: p<0.007;
type: p<0.002; tenacity: p<0.01; total: p<0.002). Intestinal obstruction and body temperature during
PP do not have a significant effect upon adhesion formation (NS all comparisons).
Group
BALB/c
Substrains
5 min
Manipulation
Qualitative Scoring (Mean±SE)
Extent Type Tenacity Total
I
BALB/cOlaHsd
no 0.45±0.12 0.55±0.16 0.75±0.19 1.75±0.46
II yes 1.33±0.25a 1.25±0.13
a 1.33±0.17 3.91±0.54a
III
BALB/cByJ
no 1.25±0.18a 1.21±0.16 1.42±0.19 3.88±0.51
IV yes 1.60±0.19a 1.30±0.12a 1.35±0.15 4.25±0.40a
V
BALB/cJ@Rj
no 1.25±0.11a 1.05±0.05a 1.10±0.10 3.40±0.23a
VI yes 1.25±0.11a 1.25±0.11a 1.35±0.10a 3.85±0.26a
Table 1: Postoperative adhesions after laparoscopic surgery in different BALB/c substrains. Postoperative adhesions were induced during
laparoscopy by performing four bipolar lesions during 60 min of CO2 pneumoperitoneum. After performing the lesions, five minutes
manipulation was done in groups II, IV, VI. Three BALB/c substrains were used: BALB/cOlaHsd, BALB/cByJ and Balb/cJ@Rj. Postoperative
adhesions were scored after one week during laparotomy using a qualitative scoring system (extend, type, tenacity and total). aComparison
vs. group I: p<0.05 (Wilcoxon)
Comparing the substrains BALB/cByJ and BALB/cJ@Rj (groups III, IV, V, VI), adhesion
formation is influenced only by the substrain itself (proportions: p<0.03; extent: p<0.03; type: NS;
tenacity: NS; total: p<0.05). Manipulation, intestinal obstruction and temperature during PP do not
have any effect upon adhesion formation.
The multivariate analysis was confirmed using a univariate analysis (Wilcoxon two sample
test) for adhesion formation. BALB/cOlaHsd substrain has significantly less postoperative adhesions
54
than BALB/cByJ (group I vs. III: proportions: p<0.024; extent: p<0.048; type: NS; tenacity: NS; total:
NS) and then BALB/cJ@Rj (group I vs. V: proportions: p<0.033; extent: p<0.031; type: p=0.05;
tenacity: NS; total: p<0.03) substrains. There are not differences in adhesion formation between
BALB/cByJ and BALB/cJ@Rj substrains (group III vs. V: all comparison: NS).
The multivariate analysis was also confirmed using a univariate analysis (Wilcoxon 2 sample
test) for manipulation. When five minutes manipulation is performed during pneumoperitoneum,
adhesion formation increased significantly only in BALB/c OlaHsd substrain (group I vs. II:
proportions: p<0.03; extent: p<0.03; type: p<0.035; tenacity: NS; total: p<0.025). For the BALB/cByJ
(group III vs. IV) and BALB/cJ@Rj (group V vs. VI) substrains, the comparisons were not significant.
When the effect of the substrain, manipulation and adhesion formation upon bowel
instruction was evaluated (proc Logistic), only the type of adhesions shows a significant effect upon
bowel instruction (p=0.05). We observed that mice without manipulation had less bowel obstruction
(6/18; 33.3%) than mice with manipulation (10/18; 55.5%), however this differences are NS probably
due to the small sample size (fig. 13).
0
10
20
30
40
+5' manipulationno manipulation
BALB/cOlaHsd BALB/cByJ BALB/cJ@Rj Substrain
a
a a
I II III IV V VI Group
a
a
Prop
ortio
n of
adh
esio
ns(%
) (m
ean ±
SEM
)
Figure 13: Postoperative adhesions after laparoscopic surgery in different BALB/c substrains
Postoperative adhesions were induced during laparoscopy by performing four bipolar lesions during 60 min of CO2 pneumoperitoneum.
After performing the lesions, five minutes manipulation was done in groups II, IV, VI. Three BALB/c substrains were used: BALB/cOlaHsd,
BALB/cByJ and Balb/cJ@Rj. Postoperative adhesions were scored after one week during laparotomy using a quantitative scoring system
(proportions). aComparison vs. group I: p<0.03 (Wilcoxon)
55
Conclusion. The magnitude of CO2 enhanced adhesions and of manipulation enhanced
adhesions varies not only with mice strains (134) but also with inbred mice BALBc substrains. This
could be used as a model to investigate individual susceptibility in humans.
e. The role of mesothelial cells on adhesion prevention
Intraperitoneal injection of cultured mesothelial cells decrease CO2 pneumoperitoneum enhanced adhesions in a laparoscopic mouse model. J. Verguts, A. Coosemans, R. Corona, M. Praet, K. Mailova, P. R. Koninckx. Gynecol Surg, Jan 2011 DOI 10.1007/s10397-011-0658-8.
Introduction
The concept that cultured mesothelial cells can decrease adhesion formation in an abrasion
model thus seems well established. The amount of cells required, however, the culture method and
the origin of the cells are still unclear (138). Since an abrasion model is a pure surgical lesion
requiring laparotomy, we therefore wanted to evaluate whether cultured mesothelial cells also
affected CO2 pneumoperitoneum-enhanced adhesion formation.
Material and methods
Collection and culture of mesothelial cells. Mesothelial cells were obtained from 9-16 week
old inbred female Balb/c mice. Following anaesthesia with intradermal pentobarbital (Nembutal,
Sanofi Sante Animale, Brussels, Belgium; 0.06 mg/g) a 14-gauge catheter (Insyte, Vialon, Becton
Dickinson) was inserted into the abdomen at the umbilicus and fixed leak-proof with a Prolene 5/0
purse suture.
The cells were collected via a 2-step procedure per mouse. Firstly, 10ml phosphate buffered
saline pH 7.3 (DPBS) was injected, of which at least 8 ml could be re-aspirated. This first set of ‘free
floating cells’ was centrifuged at 200G for ten minutes and the pellet was then suspended in culture
medium (RPMI 1640 (Lonza, BioWhittaker) supplemented with 15% fetal calf serum (FCS; CSL UK Ltd,
Andover, UK), 4 mM L-glutamine (Gibco, Paisley, UK), 0.4 mg/ml hydrocortisone (Sigma Aldrich,
Poole, UK) and antibiotics: penicillin, 120mg/l, amphothericine (2,5mg/l) and gentamicine (4mg/l) as
56
used by Foley-Comer et al. (139). Secondly, 10ml of DPBS with 0.25% Trypsin and 1 mM EDTA4Na at
37°C was injected through the same catheter and the mouse was softly shaken. After ten minutes
the instilled fluid was re-aspirated. Next, the abdominal cavity was washed with 5ml culture medium;
to finish, the abdominal cavity was opened and gently rinsed 8 times with 2 ml culture medium.
All mesothelial cells obtained via this second procedure were collected together, centrifuged
for 10 minutes at 200g and re-suspended in culture medium. Each time two mice had undergone the
procedure, cells of those mice were pooled, keeping cells of step 1 and 2 separately. Cells were then
put in an incubator at 37°C with 5% CO2. If cells were confluent, which was normally the case after 1
week, cells were split at a ratio 1:2, after having them detached with Trypsin-EDTA. After 3 weeks,
cells were detached and counted for further use. To ascertain the growth of mesothelial cells, cells
from 3 weeks old cultures were seeded on slides. After 2 days of culture, slides were fixed and
endogenous peroxidase activity was blocked by 0.5 % H2 O2 in methanol. Cells were evaluated for
intercellular connections and for cytokeratin staining.
All mice (n°20) received 4 standard bipolar lesions and 60 minutes of CO2
pneumoperitoneum as described above. After all laparoscopic ports had been closed, 1ml of DPBS
with or without mesothelial cells, was injected intraperitoneally.
Results
Experiment I. In the first experiment the injection of 400.000 cultured mesothelial cells
caused a reduction of the proportion of adhesion formation from 19% to 7% (p<0.046) and a
reduction of the total adhesion score from 3.1 (+/- 1.74) to 1.7 (+/-1.68) (NS, p<0.086).
Experiment II. The experiment was designed to obtain a dose response curve of adding
different amounts of cultured mouse mesothelial cells in adhesion reduction. Considering the effect
of 400.000 cells from the first experiment, and taking into account the number of cells available, it
57
was decided to inject per group of 5 mice 400.000 cells, 133.000 cells and 44.000 cells, i.e. a
logarithmic decrease. The control group which only received DPBS consisted of 5 mice.
This experiment confirmed that mesothelial cells can reduce CO2 pneumoperitoneum
enhanced adhesion formation. With 44.000, 133.000 and 400.000 cells the proportion of adhesions
dropped from 13.5% to 11.5% (NS), 8% (P<0.055) and 5% (P<0.007) respectively. The total adhesion
score showed a similar decline being for the 44.000, 133.000 and 400.000 cells not significant,
P=0.0093 and P=0.0025 respectively (Figure 1). When adhesions are plotted against the logarithm of
the concentration of cells, the amount of cells required to decrease adhesion formation by half was
estimated at 100.000 cells.
Figure 14. Total adhesion scores after post-operative intraperitonal injection of different concentration of mesothelial cells.
Conclusions. Cultured mesothelial cells can prevent adhesion formation in mice and the
effect increases with the number of cells. In Balb/c mice, the number of cells needed to reduce
adhesions by 50% seems to be around 100,000. Considering that 100,000 cells cover in culture 10
cm², this suggests that the effect is mediated by incorporation in the entire peritoneal cavity and not
only at the lesion site.
Dose Response Curve
0 100000 200000 300000 400000 5000000.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
Dose: number of cells
Tota
l Adh
aesio
n Sc
ore
(mea
n/SD
)
58
f. The effect of intercoat gel on adhesion prevention
Intercoat gel (oxiplex): efficacy, safety, and tissue response in a laparoscopic mouse model. Schonman R, Corona R, Bastidas A, De Cicco C, Mailova K, Koninckx PR. J Minim Invasive Gynecol. 2009 Mar-Apr;16(2):188-94.
Introduction
The mechanism of action of any product used for adhesion prevention could thus be local at
the trauma site, e.g., by keeping the surfaces separated until reepithelialization, or through the
peritoneal cavity by preventing or reducing mesothelial trauma or by preventing deleterious effects
from the peritoneal cavity to reach the traumatized area. This prompted us to evaluate the last
developed product, Intercoat, in our laparoscopic mouse model.
Materials and methods
Intercoat, a viscoelastic gel composed of polyethylene oxide and carbomethylcellulose was
received from Ethicon Inc (FzioMed, Ethicon, Somerville, NJ). Intercoat was administered with a
sterile syringe and a 14-gauge catheter.
Experiment I (n =24) was performed to evaluate the efficacy of Intercoat gel on adhesion
formation in model 1 (hypoxia-enhanced adhesion formation) and in model 2 (hypoxia-and
manipulation-enhanced model) and to evaluate whether efficacy was similar in both models. A
factorial design with 6 animals in each of the 4 groups was used, i.e., CO2 pneumoperitoneum-
enhanced adhesions without and with manipulation, each of them without and with Intercoat
application (0.7 ml of Intercoat) applied on the lesions immediately after the lesion was made for
model 1 or immediately after manipulation for model 2).
Experiment II (n = 49, 7 animals/group) was designed to answer 2 questions. First, is the
effect of Intercoat caused by the barrier effect between the opposing lesions only, or does Intercoat
in addition attenuate the effect of hypoxia by shielding the lesion from the hypoxic effect of CO2.
Second, we wanted to exclude that the inflammatory/edematosus reaction observed in the first
59
experiment might contribute to adhesion formation, thus Intercoat having possibly a dual effect, a
barrier effect reducing adhesions and an inflammatory reaction possibly enhancing adhesions. To
answer the first question, 0.7 ml of Intercoat was applied on the lesions immediately after the lesions
were made (group II) or at the end of pneumoperitoneum (group III) in comparison with a control
group (group I). To answer the second question, 0.2, 0.5, and 1 ml of Intercoat were administered in
the upper abdomen after induction of the lesions (group V, VI, and VII, respectively).
In addition, in group IV, 0.7 ml of Intercoat was administered on the lesions together with 1
ml in the upper abdomen.
Histology. To evaluate tissue reaction to Intercoat, biopsy were taken from the omentum in
group 7 (1 ml of Intercoat), i.e., the highest dose of Intercoat used. Because 85.7% of the mice in
group 4 (0.7 ml on the lesion and 1 ml at the upper abdomen of Intercoat) died, this group could not
be used. The biopsy specimens were embedded with JB solution (Canemco, Quebec, Canada) and
fixed in paraffin. Control samples were stained with haematoxylin- eosin. Samples from the treated
mice were stained with haematoxylin-eosin and immunohistochemistry staining for CD45 with
purified rat antimouse CD45, in a 3-step staining procedure with combination with biotin-conjugated
rabbit antirat as the secondary antibody and streptavidin together with diaminobenzidine as a
detection system. Because we looked for inflammatory reaction, we used anti-CD45, because CD45 is
found on all cells of haematopoietic origin except erythrocytes. Its presence distinguishes leukocytes
from non-haematopoietic cells.
60
Results
The I experiment confirmed the increase in adhesion formation by manipulation and the
decrease by Intercoat. Adhesion reduction was found whether evaluated as quantitative adhesion
proportions (p<.0001) (Fig. 15) or as a qualitative scoring of adhesion formation, i.e., total adhesion
score, extension, type, or tenacity (p ,.0001, p ,.0001, p 5.0002, and p 5.0002, respectively, proc
GLM). The reduction, moreover, was not specific for manipulation-enhanced adhesions, because the
percent reduction in adhesions was similar in both models, i.e., with and without manipulation.
Fig. 15. Reduction of adhesion formation (quantitative scoring) by Intercoat in 2 laparoscopic mouse models. In the first model,
adhesions following coagulation lesions were enhanced by 60 minutes of CO2 pneumoperitoneum. In second model, adhesions were
further enhanced by bowel manipulation. W/O = Without.
In this II experiment Intercoat similarly reduced adhesion formation whether applied
immediately after induction of the lesion or at the end of pneumoperitoneum (p 5.097 and p 5.0051,
respectively, Wilcoxon test) (Fig. 16). Intercoat applied in the upper abdomen did not increase
adhesion formation; to the contrary, a decrease in adhesion proportion was found with higher doses.
This reduction was observed for the quantitative scoring (0.2 ml vs. 0.5 ml and 1 ml; p 5.0177 and p
5.0024, respectively, Wilcoxon test) and for the qualitative adhesion score, i.e., total score,
extension, type, and tenacity (p 5.0113, p 5.0024, p 5.0338, and p 5.0278, respectively, proc GLM).
61
Intercoat in higher doses was associated with increased mortality in mice (p 5.0058, Spearman
correlation) being 85.7% when 1.7 ml was used (1 ml in the upper abdomen and 0.7 on the lesion)
versus 28.5%, 28.5%, and 0% when 1, 0.5, and 0.2 ml were used in the upper abdomen, respectively,
and 14.2% after 0.7 ml on the pelvic lesions. In total, 12 mice died between 4 and 7 days
postoperatively. Intercoat application in mice was associated on day 7 with ascites and hyperaemia.
Fig. 16. Effect of Intercoat on adhesion formation when applied on lesion at beginning or end of pneumoperitoneum (PP) and
when applied in upper abdomen.
The histology showed in the Intercoat-treated mice important capillary dilation, hyperaemia,
and cellular edema of the connective tissue cells, which were absent in the control samples (Fig. 17).
Immunohistochemistry staining with CD45 that stain leukocytes and lymphocytes did not show
marked inflammatory response in this tissue.
62
Figure 17. Histologic biopsy specimens from Intercoat-treated mouse showing omentum with severe capillary dilation (A) in comparison with control omentum biopsy specimens (B) and edema of connective tissue cells (C). CD45 staining does not show inflammatory cells (D).
Conclusions. Intercoat is confirmed to be an effective antiadhesion product in our
laparoscopic mouse model. It acts as a local barrier separating the lesions, but it cannot be excluded
that it also reduces the deleterious effect of the peritoneal cavity on adhesion formation. The
vascular congestion and intracellular edema, caused by high osmotic effect of its components, could
contribute to adhesion formation prevention.
Most important is that the effect of Intercoat is similar when applied immediately after the
lesion or after 1 hour of pneumoperitoneum. This is consistent with the concept that CO2
pneumoperitoneum enhancement of adhesions is mediated by the entire peritoneal cavity and not
by a direct effect on the lesion.
63
f. Post-operative acute inflammation of the peritoneal cavity: a common mechanism mediating the effect of the peritoneal fluid upon adhesion formation
Post-operative inflammation in the abdominal cavity increases adhesion formation in a laparoscopic mouse model. R Corona, J Verguts, R Schonman, MM Binda, K Mailova, PR Koninckx. Fertil Steril 2011 Mar 15;95(4):1224-8.
Introduction
The inflammatory reaction at the lesion site is widely believed to be a driving mechanism of
adhesion formation (140-150), Therefore it was surprising that neither nonsteroidal anti-
inflammatory drugs (NSAIDs) such as cyclooxygenase-1 (COX-1) or COX-2 inhibitors, nor antibodies
neutralizing anti-tumor necrosis factor-a (TNF-a) had any effect upon adhesion formation in hypoxia-
enhanced adhesion (pure CO2 pneumoperitoneum) (98) or hyperoxia-enhanced adhesion
(pneumoperitoneum with more than 12% O2) (79) models. In contrast, dexamethasone, a steroidal
anti-inflammatory drug, reduced adhesions by 30% and 62% in the hypoxia- and hyperoxia-enhanced
adhesions models, respectively. We therefore investigated the relationship between adhesion
formation and acute inflammation in the entire peritoneal cavity.
Materials and methods
Histological inflammation score. A scoring system of acute inflammation based upon the
known pathophysiology of acute inflammation (151-155). Acute inflammation has three major
components: (a) alterations in vascular calibre leads to an increase in blood flow, (b) neoangiogenesis
and structural changes in the microvasculature, permitting plasma proteins and leukocytes to leave
circulation, and (c) diapedesis of leukocytes from the microcirculation and their accumulation and
activation in the lesion (156). We scored, (a) the number of vessels, reflecting neoangiogenesis, (b)
the number of PMNs in diapedesis, reflecting the increased permeability of vessels; and (c) the PMNs
accumulation on the site of the injury. Scoring was as follows: neoangiogenesis (0 = < 3 vessels, 1 = 4-
64
8, 2 = 9-12, 3 = >12), vessel permeability (0 = 0 PMN’s in diapedesis, 1 = <2, 2 = 3-4, 3 = >4) and
leukocytes activation and accumulation (0 = <3 of PMN’s, 1 =4-8, 2 = 9-12, 3 = >12). Total
inflammation score was the sum of the neoangiogenesis, permeability and leukocyte activation
scores.
Histology and immunohistochemistry. Under stereomicroscopic vision a biopsies were taken
during laparotomy. After dissecting skin and muscles from the peritoneum, a tissue adherent
(Lyostipt, Braun) was applied over the whole length of the lesion plus 5 mm of peritoneum on each
side of the lesions. The specimens were then fixed with JB fix (157) for 24 hours, embedded in
paraffin, oriented and four 4-6µm sections were taken perpendicularly to the surface and
perpendicularly to the lesion. Each section thus permitted to evaluate changes of the lesion and of
the surrounding peritoneum from the surface to the depth of the biopsy. Sections were
immunohistochemically stained for CD45 to detect activated leukocytes (LCA, Ly-5, T200) (BD
Pharmigen) in citrate bluffer 80°C, pH 6, dilution 1/400.
Inflammation parameters were blindly scored under a microscope with a camera for imaging
system (Axio scope, Axio cam MRc5, KS 400 imaging system, Zeiss, Germany). For each slide, 4 high
power fields were randomly chosen, 2 of the lesion and 2 of the surroundings (1 for each side) and
counted the number of vessels, of PMN’s in diapedesis and of leukocyte stained by CD45.
Experiment I was designed to determine which day after surgery acute inflammation and
adhesion formation should be scored. After 60 min of CO2 pneumoperitoneum (n=8) inflammation
and adhesion scores were evaluated after 1, 2, 4 and 7 days (2 mice per day) (Fig. 1). Based upon
these results we decided to evaluate inflammation and adhesions 48 hours after surgery in all further
experiments.
Experiment II (16 mice: 4 mice per group) was designed to evaluate the inflammation score
and adhesions in control animals after 60 min of CO2 pneumoperitoneum with 4% O2 (Group I), after
60 min CO2 pneumoperitoneum (group II), after 60 min CO2 pneumoperitoneum plus manipulation
65
(group III) and after manipulation-enhanced adhesion formation as group III but adding
dexamethasone (Group IV) known to decrease adhesions. Manipulation consisted of manipulating fat
and bowels in the upper abdomen, i.e. the omentum and large and small bowels were moved gently
up and down across the abdomen for 5 minutes with a 1.5 mm non-traumatic grasper for 5 min as
described (137). Dexamethasone (Aacidaxim 5 mg for injection; Organon, Brussels, Belgium) was
given intraperitoneally, by an intraperitoneal injection of the medication through the midline of the
lower abdominal wall with a 31-gauge, 8mm (5/16") BD Ultra-Fine™ Short Needle holding the mouse
in upside down position to avoid bowel injuries, at the dose of 40 µg at the end of
pneumoperitoneum and 40 µg after 24 hours.
Experiment III (16 mice: 4 mice per group) was similar to experiment II, but inflammation was
scored at the parietal peritoneum in the upper abdomen at distance from the surgical site.
Results
In the I experiment quantitative adhesion scores after 24, 48, 72 hrs and 7 days were 2.32
±0.29, 3.01±0.18, 3.35±0.31 and 3.74±0.32, respectively. Total adhesion scores and the total
inflammation, the former increasing and the latter decreasing progressively are shown in fig 18. Also
the scores of neoangiogenesis, diapedesis and leukocytes accumulation decreased progressively. We
therefore decided to evaluate acute inflammation and adhesion formation on day 2 in all subsequent
experiments.
66
Figure 18. Total adhesion and inflammation score during the first 7 days after a surgical bipolar lesion and 60 minutes of CO2
pneumoperitoneum.
In the II experiment the adhesion scores on day 2 confirmed previous observations on day 7
after surgery (Fig. 19 and 20). In comparison with the control group (group I), both total and
proportion of adhesions increased when a pure CO2 pneumoperitoneum was used (group II; P=0.007
and P = 0.0053, respectively). Adhesions further increased in group III (P<0.0001 for both total and
proportion). The addition of dexamethasone (group IV) decreased adhesion formation (P<0.0001 for
both adhesion scores).The total inflammation score at the central part of the surgical lesion was
slightly higher in comparison with the inflammation score at 0.5 cm from the lesion, being 1.5±0.40
and 2.625 ±0.25 (p=0.0203), 0.625±0.25 and 2.125±0.25 (p= 0.0154), 4.125±0.25 and 6.375±0.25
(p=0.0321), 2.375±0.25 and 3.875±0.25 (p= 0.0591) for groups I, II, III and IV, respectively.
Depth of the inflammatory reaction spanned 2 mm till a maximum of 4 mm in group III which
had the highest inflammation score.
67
Figure 19. Correlation between total adhesion and inflammation scores (Statistics: p < 0.0001, Spearman Test (for details see
table 1 grey highlighted).
Changes in total inflammation scores at the lesion (mean of the inflammation scores in
the central part and in the periphery) and in adhesions scores were strikingly similar (Fig. 20).
At the level of the surgical lesion pure CO2 pneumoperitoneum in comparison with group I
slightly increased the total inflammation score (p=0.07), neoangiogenesis (p=0.0577) and
lymphocytes accumulation (p=0.0796). In group III, the inflammation score further increased
i.e. the total inflammation (group II versus group III P<0.0001), neoangiogenesis (P=0.0007),
vasodilatation and permeability (P=0.0052) and lymphohistiocytic activation and accumulation
(p=0.0022). When dexamethasone was added after surgery total mean inflammation score
decreased (group III vs. group IV: P<0.0001), an effect observed for all parameters i.e.
neoangiogenesis (p=0.0154), diapedesis (P=0.0016) and lymphocytes activation-accumulation
(P=0.001). Most strikingly, however, was the strong correlation between adhesion scores and
inflammatory parameters (table 2). Total acute inflammation score (Fig. 19), neoangiogenesis,
diapedesis and leukocytes accumulation (table 2) strongly correlated with total adhesion
scores.
68
Inflammatory parameters
Biopsy
Location
Adhesion score (P Value)
Total Extent Type Tenacity Proportion (%)
Neoangiogenesis
Central
Surrounding
0.0002
0.0002
NS
NS
<0.0001
<0.0001
0.0032
0.0021
NS
NS
Permeability Central
Surrounding
0.0391
0.0286
NS
NS
0.0411
0.0382
NS
NS
NS
NS
Leucocytes accumulation
Central
Surrounding
0.0003
0.0002
0.0043
0.0032
0.0029
0.0021
<0.0001
<0.0001
0.0031
0.0028
Total score Central
Surrounding
<0.0001
<0.0001
0.0017
0.0005
<0.0001
<0.0001
<0.0001
<0.0001
0.0005
0.0005
Table 2. The table shows the p values for the correlations between the quantitative and qualitative adhesions scoring
system and the inflammation scoring. (For grey fields see Fig. 19). Statistics: Spearman correlation.
This experiment III confirmed the results of experiment II. In addition, the inflammation
scores of the parietal peritoneum were found to be comparable with those in at the surgical lesion
(Fig 20). Indeed the use of pure CO2 pneumoperitoneum in comparison with CO2 + 4% O2 increased
slightly the total inflammation score (NS, P= 0.06); manipulation further increase the total
inflammation score (group II versus group III: P<0.0001) and dexamethasone decreased the total
inflammation score (group III vs. group IV: P<0.0001).
69
Figure 20. Total adhesion and inflammation score at the surgical lesion (experiment II) and at the parietal peritoneum far from
the surgical lesion (experiment III).
Conclusion: Acute inflammation of the entire peritoneal cavity plays a key role in the
enhancement of adhesions between opposing surgical lesions. The degree of acute inflammation
results from the cumulative effect of all known bad and good factors enhancing and decreasing
adhesion formation respectively.
70
4.1.2 Results aim 2. To establish the optimal peritoneal environment to prevent adhesion
formation
a. Investigations on N2O and the effect of using mixtures of N2O and O2 in CO2 upon adhesion formation
The addition of 5% N2O to the CO2 pneumoperitoneum is enough to reduce adhesion formation in a laparoscopic mouse model. Karina MAILOVA, Roberta CORONA, Jasper VERGUTS, Leila ADAMYAN and Philippe R. KONINCKX MD. Submitted
Introduction
Nitrous oxide (N2O), well known in anaesthesiology has been used instead of CO2, mainly in
the 1970s (158). N2O is a safe gas with solubility in water (1.5 mg/l) and diffusion in the lungs similar
to CO2, but its use was abandoned due to its explosion risks at least at concentrations of more than
29%. N2O nevertheless has some advantages over CO2. It is less irritating with less postoperative pain
in comparison with CO2 (110). N2O has anaesthetic and analgesic properties, without the
cardiopulmonary side effects of CO2 (117;158).
We therefore investigated the effects of N2O during laparoscopy upon adhesion formation in
our laparoscopic mouse model. Indeed, if concentrations lower than 29% would be effective in
reducing adhesion formation, its use might be reconsidered, given its solubility and safety is
comparable to CO2 (121).
Materials and methods
Experiment I (n=18) was a designed as an exploratory pilot experiment evaluating pure
carbon dioxide (group I; n=9) with pure nitrous oxide (group II; n=9) upon adhesion formation when
used for the pneumoperitoneum during 1 hour. In group II, two animals died during the procedure
because of intubation problems.
71
Experiment II (n=30) was designed as a dose finding experiment aimed to find the efficacy of
lower concentrations of N2O upon adhesion formation. The study comprised six groups (n = 5 in each
group) comparing pure CO2 (group I) with 5%, 10%, 25%, 50% of N2O in CO2 (groups II, III, IV and V)
and with pure N2O ( group VI).
Results
Using 100% N2O for the pneumoperitoneum instead of CO2 decreased adhesion formation.
Besides the quantitative score (P= 0,012, Fig. 21), also the qualitative score was decreased. Total
adhesion score, type, tenacity and extent decreased from 3,0±0,3 to 1,6±0,4 (P=0,0091 ), from 10,8±
to 5,5±0.6 (P=0,0228), from 10,8±1.2 to 5,6±0.4 (P=0,0253) and from 11,3±1.3 to 4,9±0.8 (P=0,0048)
respectively.
Fig. 21 Effect of the type of the insufflation gas (CO2 vs. N2O) upon adhesion formation. Proportion of adhesions ± SD is
indicated. P = 0.012
Comparing in the second experiment 100% CO2 pneumoperitoneum with 5, 10, 25, 50 and
100% of N2O the proportion score of adhesion formation decreased (Fig. 22) already with addition of
5% of N2O to the CO2 pneumoperitoneum and there was an high significant difference between all
the groups that received in addition N2O and the group with pure CO2 (P=0,0001 for all groups).
72
0 5 10 25 50 1000
5
10
15
20
25
30
Proportion of N2O (%)
Prop
ortio
n of
adh
esio
ns (%
)
Figure 22. Effect of the addition of different proportions of nitrous oxide to CO2 pneumoperitoneum on adhesion formation,
scored 7 days after surgery.
Conclusions. N2O, from concentrations of 5% onwards, is the single most important adhesion
reducing factor.
b. Translation of peritoneal cavity conditioning during laparoscopic surgery to conditioning in open surgery : equally important for adhesion formation.
Is the conditioning of the operative field possible in open surgery? A mouse model for open surgery and the effect of conditioning upon adhesion formation. Roberta Corona, Karina Mailova, Maria Mercedes Binda and Philippe R. Koninckx Article in preparation
Introduction
Adhesion formation after open surgery is partly due to the perioperative exposure of the
operative field to the ambient air leading to various local processes that cause inflammation and
damage the mesothelial peritoneal layer (159-161). These adhesiogenic processes include superficial
desiccation and exposure to the atmospheric partial oxygen (O2) pressure with ensuing hyperoxia
and oxidative stress (79;159;162).
73
We believe that the peritoneal factors identified in laparoscopy are the same occurring in an
open surgical field therefore the aim of this study was, first, to develop an animal model for open
surgery to study the feasibility of a peritoneal conditioning of the open surgical field. Secondly, to
compare adhesion formation in laparoscopy versus laparotomy and third, to evaluate the effect of
peritoneal conditioning in open surgery upon adhesion formation.
Materials and methods
Development of the mouse model for open surgery. In a first pilot experiment we planned
to evaluate the effect of humidified 60 min of CO2, humidified CO2+ N2O (50/50) and N2O in 6 mice (2
mice per group, randomized). They unfortunately all died, which was explained by excessive
desiccation (visually obvious), due to mixing of the gas with the ambient air as described by Persson
(163). Therefore in all subsequent experiment the box was covered as explained in details in the
chapter of material and methods.
In order to confirm the effect of desiccation (exp 2), non-humidified CO2 and humidified CO2
were compared for 30 and 60 min of “open exposure” (n=12 mice). This confirmed the importance of
covering the small box, and the importance of desiccation since after 60 min without humidification
all mice died. This experiment was repeated (exp 3) with N2O instead of CO2, confirming the
laparoscopic findings that N2O was less harmful than CO2. This experiment was therefore repeated
(pilot 4) with adding 4% of Oxygen to the CO2 instead of pure CO2, equally confirming the
laparoscopic findings that adding 4% of oxygen is beneficial.
After these pilot experiments we decided to do all further experiments with humidification, a
covered box and operating times of 30 or 60 minutes.
Experiment I. In a first experiment the dose response of the effect of N2O upon adhesion
formation was evaluated in the open model. Except for a control group, where mice were kept
74
exposed only to air outside the box, in the all the others group’s mice were inside the cover box and
kept exposed to humidified gas for 30 minutes.
Experiment II. Being the addition of 10% of N2O to CO2 much more important than the
addition of 4% of O2, the second experiment was designed to evaluate, whether adding 4% of oxygen
still had an additive effect when 10% of N2O in CO2 was used. Since adhesion formation was already
so low, this was evaluated in both the open model and in the laparoscopic model using a factorial
design with 10 mice in each cell. In the open model, mice were inside the cover box and kept
exposed to 100% humidified gas for 60 minutes. In the laparoscopic model, the pneumoperitoneum
was maintained for 60 minutes with 100% humidified gas. The same mixing of gas, CO2 + 10% N2O or
CO2+ 10% N2O + 4% O2, were used in the open and laparoscopic model.
Results
In the pilot experiments adhesions increase significantly with the increasing of the exposure
time at the gas for all the different settings used, with or without humidifier (p=0.002 CO2 30 min vs.
CO2 60 min, p< 0.0001 for all the other groups). With 60 minutes of 100% CO2 dry gas exposure, all
the mice dead. With 30 minutes we had 1 dead out of 3. In the groups without humidifier with CO2
and with CO2+ O2 were always found “de novo” adhesions in the upper abdomen. In all the groups
with N2O were never found “de novo” adhesions. Using N2O gas with 100% of relative humidity,
there were no adhesions after 30 minutes of gas exposure and very few after 60 minutes, significant
less than in all the other groups (p < 0.0001). With the adding of O2 at the CO2 gas there is a
significant decreasing in adhesions formation in comparison with the CO2 gas only ( p= 0.0011 after
30 min of humidified gas, p= 0.003 after 30 min of humidified gas, p<0.0001 after 30 minutes of dry
gas exposure) (fig.23).
For every different settings there is a high statistical difference between the groups with or
without humidifier, both if used 30 minutes or 60 minutes for the different gas (p<0.0001).
75
CO2/30 CO2/60 N20/30 N20/60 CO2+O2/30 CO2+O2/600
5
10
15
20
Humidifier NO Humidifier
X
X Mortality 100%
255075
100
Gas/minutes
Prop
ortio
n of
adh
esio
ns (%
)
Figure 23. Cumulative view of pilot experiments results after establishing of the model with a covered box.
In the dose response (exp I) in open surgery (figure 24), there was not a significant reduction
of adhesions in comparison with CO2 only when 0. 3% or 1% was added to the CO2
pneumoperitoneum; it was instead significant with the addition of 3% or 10% (p=0. 0061 and p= 0.
0006). It’ enough the addition of 10% N2O to reach almost the same effect of 100% N2O with
complete prevention of adhesions (p=0. 1551). There is still a significant difference between 3% N2O
and 10% (p=0. 0291).
Figure 24. Proportion of adhesions when floading the open operative field with CO2 plus different concentration of N2O.
76
In the results of the II experiment not significant difference was noted between laparoscopy
and open surgery. It was showed instead an addictive effect of O2 in both laparoscopy and open
surgery (p=0. 0028 and p=0. 0175) (figure 25).
CO2 + N2O 10% CO2 + N2O 10% + 4% O20.00.10.20.30.40.50.60.70.80.91.0
open surgery laparoscopy
Pro
po
rtio
n o
f ad
hes
ion
s (%
)
Figure 25. Comparison between open surgery and laparoscopy for conditioning with CO2 plus 10 %N2O only or plus 10% N2O and
4% O2; a little addictive effect of O2 is showed.
Conclusions. Conditioning seems as important, if not more for open surgery, as
demonstrated for laparoscopy. The dose response effect of N2O as established in mice is confirmed
and extended for lower concentrations suggesting that around 5% a maximal effect if obtained.
When 10% of N2O is used the effect of oxygen is marginal. We therefore suggest to use CO2 + 10% of
N2O and eventually 4% of oxygen.
77
4.2 Investigations in humans
4.2.1 Results of aim 3. To translate the observations on adhesion prevention and peritoneal conditioning in mice into clinical practice
Introduction
The quantitative relationship between desiccation, peritoneal temperature and mesothelial
damage during endoscopic surgery were not yet investigated in detail. Desiccation will obviously vary
with flow rate, relative humidity and temperature of the gas used for the pneumoperitoneum. This
will affect intra-peritoneal and mesothelial temperature, mesothelial damage, peritoneal acute
inflammation and ultimately postoperative adhesion formation and pain. Ideally, according to mouse
experiments, the absence of desiccation should be combined with a slightly lower intra-peritoneal
temperature, requiring a gas with a close to 100%RH entering a peritoneal cavity cooled by a third
means.
The translation of the mouse experiments on adhesion formation, cooling and desiccation, to
the human thus suggested no desiccation together with an intra-peritoneal temperature of 31-32°C,
a temperature achieving more than 80% of the cooling effect upon adhesion. We therefore planned
to investigate in the human the relationship between flow rate, temperature and relative humidity of
the gas used for insufflation, desiccation, cooling with a third means and the intra-peritoneal
temperature.
78
a. The desiccation during laparoscopic surgery at different settings of flow rate, temperature and humidity
Intra-peritoneal temperature and desiccation during endoscopic surgery. R Corona, J Verguts, R Koninckx, K Mailova, MM Binda, PR Koninckx. Am J Obstet Gynaecol. In press
In vitro experiments. Experiment I: First, the accuracy of the measurements of flow rate, RH, and
temperature of the gas were validated. The exact water loss was measured by the difference in
weight of the “desiccation recipient” protruding moistened towels, before and after the experiment,
and compared with the water loss calculated from the flow rate, RH, and temperature of the in-
flowing and out-flowing gas with the following formula: humidity (g/m³) = RH (%) / 100 x (3.1243
10E-4 x T³ + 8.1847 x 10E-3 x T² + 0.32321 x T + 5.018). Since the accuracy of desiccation calculation
was crucial for the in vivo experiments, the measurements were validated over a wide range of
conditions and performed in triplicate (table 3).
Flow rate (l/min) Measured water loss (g) Calculated water loss (g)
Storz humidifier
2.5 1.91 + 0.12 2.22 + 0.14
5 7.20 + 0.16 8.11 + 0.09
10 9.73 + 0.13 11.21 + 0.12
20 11.84 + 0.2 12.92 + 0.22
F & P humidifier
2.5 1.72 + 0.08 2,01 + 0.07
5 5.31 + 0.10 6.14 + 0.15
10 7.60 + 0.25 9.32 + 0.23
20 10.43 + 0.26 12.61 + 0.30
No humidifier
2.5 3.11 + 0.11 4.22 + 0.14
5 9.31 + 0.15 10.10 + 2.11
10 11.93 + 0.27 13.41 + 0.26
20 14.72 + 0.31 16.20 + 0.33
Table 3. Measured and calculated water loss for different humidifiers and flow rates.
79
Experiment II was designed to measure the impact of inflowing gas temperature, RH and of
desiccation upon the temperature and relative humidity in the box, confirmed as expected that the
effect of desiccation was most important. Using a non-isolated insufflating tubing of more than 1.5 m
the gas temperature rapidly equilibrated with the ambient temperature and this for flow rates up to
20L/min. Thus, when the tubing was in a chamber at 37°C the temperature was as expected stable at
37°C; otherwise the temperatures were similar to room temperatures.
When dry CO2 at room temperature was used for insufflation desiccation and a drop in
temperature occurred as shown in figure 26. It should be realized that thus the water loss/min and
the heat loss/min increased almost linearly with flow rate. With the Storz humidifier, (Figure 1) the
temp “in” at the end of the tubing was as expected at room temperature while the RH decreased
with flow rates. The near 100% RH at low flow rates is explained by condensation in the tubing due
to cooling (Fig.26). With the F&P humidifier, with an isolated tubing and with the gas heated to
avoiding condensation, the RH at the end of the tubing were adequate for any flow rate. The
temperature increased however with the flow rate being 37.2 + 0.4 °C, 38.8 + 0.6 °C, 40.2 + 1.2 °C
and 41.7 + 1.8 °C for flow rates of 2.5, 5, 10 and 20 L/min respectively. It should be noted that in
these experiments we removed the luer lock of the original tubing, which prevents flow rates higher
than 7.8 L/min at 15 mm insufflation pressure.
The third experiment confirmed that with the modified F&P humidifier the temperature was
constant within 0.2°C for any flow rate for preset temperatures between 28 and 34 °C with a RH over
90% as shown in fig 27 insert for a preset temperature of 32°C. When flow rates were suddenly
increased or decreased, the new equilibrium was reached within 2-3 minutes. The temp “out” and
RH ‘out’ were as expected with a slight decrease in temperature due to desiccation at higher flow
rates.
80
Figure 26. Temperature and relative humidity for different flow rates with dry gas, Storz humidifier and modified Fisher and
Paykel humidifier in vitro.
Cooling. Spraying 3-4 ml of water at 0° decreased the temp ‘out’ by an average of 2,0°C + 0.5
within seconds, and this for flow rates up to 10 L/min. This cooling decreased, as expected the
calculated desiccation and at lower flow rates even some condensation occurred in the box.
In vivo investigations
The results of temp and RH, in and out were strikingly similar in vivo (figure 2) and in vitro
(figure 26).
With dry CO2 entering the peritoneum cavity at room temperature (Fig.27), resulted in a
calculated water loss of 0.151 g/min + 0.002, 0.120 g/min + 0.005, 0.171 g/min + 0.005, 0.201 g/min
+ 0.006 and 0.303 g/min + 0.006 for flow rates 2.5, 5, 7.5, 10 and 15 L/min respectively. With the
Storz humidifier desiccation was as expected much less. the water loss being 0.011 g/min + 0.003,
0.023 g/min + 0.004, 0.031 g/min + 0.006, 0.053 g/min + 0.008 and 0.070 g/min + 0.009 respectively.
With the modified F&P humidifier with a preset temperature of 32°C, desiccation was 0.00021 g/min
81
+ 0.00001, 0.0009 g/min + 0.0002, 0.0010 g/min + 0.0008, 0.0041 g/min + 0.0012 and 0.0090 g/min +
0.0021 respectively.
The differences in desiccation caused by dry CO2 and the Storz humidifier and by the Storz
humidifier and the modified F&P humidifier were significant (P<0.01) at any flow rate.
Temperature changes and heat loss
Both for the in vitro and the in vivo experiments all changes in temperature between “in”
and “out” could be attributed entirely to desiccation and to the temperature change of the water in
the gas. assuming that to heat 1 ml of gas by 1°C required 0.00003 cal), and 1 ml of water 1 cal, while
desiccation of 1 ml of water required 577cal/g of water. These data also permitted to calculate the
heat exchange between the water bags or the body and the gas used for insufflation. This heat loss
was found never to exceed 50 to 100 cal/min.
Figure 27. Temperature and relative humidity for different flow rates with dry gas, Storz humidifier and modified Fisher and Paykel humidifier during laparoscopy in vivo. (In the insert corrispective in vitro results)
82
Cooling and electrosurgery
With the cooling device we achieved as in the vitro experiments, a decrease of the
temperature of some 2°C (Fig. 28). When the spray was directed to the upper abdomen the
endoscopic image was never affected to a degree impairing vision and surgery. A similar fast increase
and decrease in temperature of 1 °C is observed during surgical procedures such as electrosurgery
(Fig. 28) and rinsing of the abdomen.
Figure 28. Changes in temperature with cooling and coagulation during laparoscopy.
Compartmentalization of the peritoneal cavity.
When insufflation was done through the operative laparoscope with outflow of gas through
the central trocar, the differences between temperature ‘in’ and ‘out’ and RH ‘in’ and ‘out’ were
much less than when the outflow occurred through a secondary port. This can be explained by the
fact that temperature and relative humidity in the abdominal cavity are not homogeneously
distributed. Especially when using a CO2 laser setup with insufflation through the laparoscope and
aspiration through the central trocar results in a small compartment with relative high flow rates
with gas flowing from the tip of the laparoscope to the central trocar. We should realize however,
that when the compartment is small, even low flow rates can cause important local desiccation as
83
could be visually observed. Therefore also any medication administered together with the
insufflating gas or the sprinkled water will not affect homogeneously the entire peritoneal cavity.
b. The effect of cooling the peritoneal cavity and the relation between intraperitoneal
temperature and core body temperature
With high flow rates of 10 L/min, as used for smoke evacuation during CO2 Laser surgery, dry
gas and the Storz humidifier caused constant desiccation of 0.3 and 0.05 mg/min during the first hour
of surgery. After 70 to 90 minutes of surgery the temperature “out”, reflecting the intra-peritoneal
temperature, suddenly decreased to 28.6 + 0.4° C and 29.4+ 0.6° C causing desiccation to drop in all
women within 5 minutes. Simultaneously with decreasing of temperature, with dry gas desiccation
drop to zero while with the Storz humidifier the same phenomenon occurred but desiccation
“inverted” to condensation of almost all inflowing humidity (Fig. 29). With the modified F&P
humidifier desiccation was minimal and this phenomenon was not observed (Fig. 29).
-60 -50 -40 -30 -20 -10 0 10 20 30 40 50 60
-0.2
-0.1
0.0
0.1
0.2
Dry Gas Storz humidifier F&P modified humidifier
Desiccation and condensation during surgery
Time (min)
Wat
er lo
ss (g
/min
)
Figure 29. Desiccation and condensation during surgery of more than 1.5 hour with flow rates between 8 and 10 L/min as used
during CO2 Laser surgery. After 70 to 80 minutes of surgery outflowing temperatures, reflecting bowel temperature, suddenly decreased
causing desiccation suddenly dropping to zero when dry gas was used (n=3). With the Storz humidifier (n=3) the same phenomenon
occurred and the desiccation “inverted” to condensation of almost all inflowing humidity. With the modified Fisher and Paykel humidifier
desiccation was minimal and this phenomenon was not observed. The moment that desiccation and temperature started to drop is
indicated as zero ; 60 min before and after that moment and Mean and standard deviations over a 10 min period are indicated.
84
The core body temperature of the patients, as measured by esophageal temperature, was
36.5 + 0.3°C (mean + SD of all the patients included in the study); in none of these women the core
body temperature was affected by Intra-peritoneal temperatures being always below 32°C.
Conclusions. In the absence of humidification the intra-abdominal temperatures are
surprisingly low mainly caused by desiccation. Even with a Storz humidifier temperatures hardly
exceed 28-30°C. Desiccation could be prevented completely by a modified F&P humidifier
maintaining the peritoneal cavity temperature at 31-32°C together with a little cooling. Surprisingly, a
little cooling by a third means was sufficient, as could be explained by the fact that the temperature
regulation of bowel and abdominal wall is different from the core body temperature and rather
similar to the limbs, reacting rapidly with vasoconstriction in order to prevent heat loss during
surgery. Clinically important is that a decreasing in the core body temperature during surgery, is
almost exclusively caused by desiccation. In the absence of desiccation we anticipate that intra-
abdominal temperatures up to 25°C will hardly affect core body temperature.
4.2.2 Clinical studies
Preliminary clinical trials were started in 2008 in order to investigate the effect of adding 4%
of oxygen to the CO2 pneumoperitoneum upon postoperative pain.
In order to initiate clinical trials on the effect of full conditioning, we first had to translate
cooling of the peritoneal cavity with a third means in order to avoid any desiccation into a clinical
device, which was developed by eSaturnus NV, as explained in the precedent paragraphs.
Initiation of these clinical trials was further delayed by the discovery that adding N2O even as
low as 5 to 10% had such strong effects on adhesion formation. This observation – N2O has emerged
as the single most important factor- obviously had to be explored in detail before being translated to
the human.
85
a. The effect of using different gases for the pneumoperitoneum upon inflammatory parameters and upon post-operative pain
Investigations with Oxygen only
Adding 4% of oxygen to the CO2 pneumoperitoneum decreases pain and inflammation in the post-operative phase after laparoscopic sacrocolpopexia. Verguts J, Corona R, Vanacker B, Binda M, Deprest J, Koninckx P. Manuscript submitted
Materials and methods
All patients undergoing a laparoscopic colpopexia from May 2008 till November 2009 at least
18 years of age were eligible for this trial. In order not to have any interference with other
inflammatory conditions or pre-existing pain, women would be excluded as stated above or in case
of chronic pain (i.e. peripheral neuropathy, pathology of the vertebral column, osteo-articular
disease) or acute pain (i.e. trauma).
Randomized, double-blind, controlled study 1:1 into one of the two groups (CO2 or CO2 + 4%
oxygen) was based on block randomization using a sealed envelope system of 6 envelopes per block.
The envelopes were kept sealed until induction of anaesthesia. The details of the study were kept
blinded to the investigator that collected the postoperative data.
All operations were carried out by the same team of surgeons. Laparoscopy was standardized
with an insufflation pressure of 15 mm of Hg. Promontofixation started by dissecting the peritoneum
of the promontory, paracolic gutter up to the vaginal vault and the posterior vaginal vault up to the
level of the levator ani muscle. A polypropylene mesh was sutured anterior and posterior to the vault
by PDS 0 sutures (Ethicon, Inc) and fixed at the level of the promontory by stapling. The peritoneum
was closed using a monocryl 3/0 (Ethicon, Inc.) running suture.
The local analgesic protocol as designed by the department of anaesthesiology (J. De Coster
MD and E. Vandermeulen MD, 2006) was used for all patients.
For each patient, age, body mass index (BMI) and duration of surgery (from incision to
closure of the laparoscopic ports) were recorded. To measure pain intensity, the Visual Analogue
86
Score (0-100 mm scale) pain score was measured by a blinded investigator at 4, 24, 48 and 72 hours
after surgery to asses pain intensity in the shoulder, trocar wound, subcostal and visceral pain. VAS
pain scores of 30 mm or less are categorized as mild, scores from 31-69 mm as moderate and scores
of 70 mm or more as severe pain. A blood sample was taken the day before surgery and on day 1, 2,
3 and 4 after surgery as is standard in our centre for patients undergoing a laparoscopic colpopexia.
C-reactive protein (CRP), white blood cell count (WBC) always expressed as 109/L and CA-125 always
expressed as IU/L was evaluated.
Results
24 patients were included in the study. Groups were comparable for age, BMI, laparoscopic
operating time and pre-operative scores for pain, CRP, WBC’s and CA 125.
Pain-score in the groups where oxygen was added to the CO2 pneumoperitoneum was
significantly lower (p<0.03) compared to the control group on the day of surgery (day 0) (Figure 1).
The first day after surgery the difference was not statistically significant anymore.
Inflammatory reaction as defined by CRP (Figure 30) and WBC showed no significant
decrease in the group where oxygen was added to the CO2 pneumoperitoneum. Levels of CA-125
were not different between the two groups. The means for CRP were however twice as high in the
control group.
The use of pain killers after surgery was not significantly different between the two groups
although less pain medication was used by the study group receiving oxygen on all postoperative
days. The difference was the greatest for the use of paracetamol at the day of the surgery (p<0.09)
and at the second day after surgery for ketorolac (p<0.16).
87
Figure 30: General VAS scores and C-reactive protein (CRP) before surgery (day -1/pre-op) and the first 4 days for group 1 (oxygen) and group 2 (carbon dioxide). Mean and SEM.
Conclusions. The beneficial effect of adding O2 to the pneumoperitoneum, as shown in the
animal models, is confirmed in humans with a significant decreasing in post-operative pain and a
reduction in CRP.
Investigations with full conditioning
Full conditioning of the pneumoperitoneum will prevent mesothelial damage and decrease post-operative inflammation and pain. Verguts J, Corona R, Vanacker B, Binda M and Koninckx P. Manuscript in preparation
Materials and methods
Postoperative pain was assessed by Visual Analogue Scales (VAS). Patients were asked to
locate their pain-symptoms (shoulder, subcostal, trocar wound and visceral pain) and score the
severity. VAS was collected 2-4 hours after surgery and every post-operative day until discharge.
Patients were free to use pain medication after surgery but preferentially ibuprofen was used in
order to permit easier comparison of pain killer intake. Postoperative inflammatory reaction was
assessed by inflammatory parameters as CRP, WBC, and temperature and was recorded as done
88
routinely. CRP and WBC were measured preoperative and every post-operative day until day 3 or
until discharge.
Patients were randomized 1:1 for standard treatment (CO2 pneumoperitoneum) or full
conditioning of the peritoneum (trial 2).
In trial 2 women undergoing laparoscopy for total or subtotal hysterectomy, myomectomy,
colpopexia or adhesiolysis were included. Women in trial 1 undergoing laparoscopy for deep
endometriosis were also included, but separately analyzed as they also received dexamethasone 5mg
at the induction of pneumoperitoneum and application by the end of surgery of Hyalobarrier®gel at
the surgical wound. Surgery and postoperative follow-up were performed as standard. Antibiotics
were given as specific for the surgery performed and immediate post-operative analgesia as is
standard (see local analgesic protocol, cfr. supra)
A prospective series of 20 patients in each arm (i.e. 40 in total) should be sufficient to reach
statistical significance per stratum. Statistical analysis was performed as an intention to treat analysis
in a 2 way ANOVA. This will allow analysis per stratum and per type of gas used. If in one procedure
the effect of full conditioning is not obtained, this will be visible in the analysis.
Preliminary results.
As expected, VAS was significant lower in the group receiving full conditioning. (Fig.31).
Shoulder tip pain 2-4 hours after surgery was significantly lower in the group receiving full
conditioning: 1/22 vs. 11/27 (p<0.0001).
89
Figure 31: General VAS scores (mean and SD) for women having laparoscopy with CO2 PP (square) or with full conditioning (dot)
for Trial 1 (endometriosis) and Trial 2 (rest)
Difference in CRP has a trend of decreasing but was not significant in trial 1 while was high
significant in trial 2 on day 1, 2 and 3. Leucocytosis was unaffected by full conditioning of the
peritoneal cavity. There was a trend of decreased pain killer intake (ibuprofen) in the group with full
conditioning (figure 32).
Figure 32: Pain killers’ intake (mean and SD) for women having laparoscopy with CO2 PP (square) or with full conditioning (dot)
for Trial 1 and Trial 2.
Conclusions. These data confirm the important reduction in pain, pain killers’ intake and a
decrease in CRP.
90
b. The effect of extensive abdominal lavage upon inflammatory parameters
Extensive abdominal lavage during laparoscopic discoid excision of deep endometriosis decreases bowel complications rate. C De Cicco, R Corona, R Schonman, A Ussia, PR Koninckx. Accepted for publication in Surgical Endoscopy
Introduction
Extensive abdominal lavage decreases mortality, morbidity and post-operative adhesion
formation in animal models and in patients with peritonitis (156;157). Historically the lavage of the
peritoneal cavity was introduced some 100 years ago (158) to decrease and dilute the abdominal
contamination and has become standard treatment to manage patients with severe peritonitis
(156;157;159). This was demonstrated in the human (160) and in animal models 3 demonstrating
that a decrease in bacterial concentration decreases mortality from peritonitis. Moreover, lavage is
claimed to remove material that may promote bacterial proliferation and pro-inflammatory
cytokines recruitment that may enhance local inflammation (161;162). The addition to the saline of
antiseptics, antibiotics and substances affecting osmolality and pH, remain controversial
(159;163;164).
Considering important the inflammatory reaction, as judged by the increased C-reactive
protein (CRP), following laparoscopic full thickness deep endometriosis resection, and the known
effects of peritoneal rinsing, we decided to evaluate, in these women, the effect of extensive
abdominal lavage on women undergoing this kind of surgery.
Materials and methods
Randomized controlled trial. Lavage of the peritoneal cavity after surgery was performed in
20 women with randomization to standard lavage or extensive lavage with 8 litres of saline. Women
in the lavage group and in the control group were comparable for age, weight and duration of
surgery.
91
Comparison with the historical series. After implementation of extensive lavage of the
peritoneal cavity, complications following full thickness resection of bowel endometriosis such as late
bowel perforations disappeared. We prospectively compared all cases thereafter to historic controls.
Results
In the trial postoperative CRP was significantly lower in the lavage group from day 1 onwards
to day 7 (P=0.01). Moreover, in the lavage group, CRP declined faster, being less than 13 mg/l on day
5, whereas in the control group CRP still was markedly elevated on day 6 with a mean of 22 mg/l.
(Figure 33) WBC counts did not show any difference.
Figure 33. Mean and standard error of CRP concentrations following full thickness resection of deep endometriosis of the rectum in case of extensive lavage with extensive lavage (8 L) or standard (no) lavage. Overall significance: P=0.01 (repeated measurement ANOVA)
Comparison with historical series. In 65 patients who received 8 litres of extensive
abdominal lavage, no bowel complications occurred (0/65). In the historical series of 84 patients who
received the standard abdominal rinsing, 8 bowel
92
complications occurred (8/84). (P<0.02 Chi-Square test). With extensive lavage, CRP is significantly
decreased and late bowel perforation did no longer occur.
Conclusion. Extensive abdominal lavage decreases inflammatory reaction following excision of deep
endometriosis and is suggested to decrease the risk of late bowel complications.
c. A RCT comparing traditional laparoscopic surgery with full conditioning and the use of Hyalobarrier
gel upon adhesion formation
Full conditioning of the peritoneal cavity prevents mesothelial damage and decrease adhesion formation. Koninckx P, Verguts J, Binda M, Vanacker B, Craessaerts M, and Corona R. Manuscript in preparation
Aim of the study: Full conditioning of the pneumoperitoneum will prevent mesothelial damage and
decrease adhesion formation.
Materials and methods
Women undergoing laparoscopic excision of deep endometriosis were randomized in this
study since this surgery is associated with severe postoperative adhesions and infertility and
therefore a second laparoscopic look can be beneficial in these patients.
40 women should be randomized 1:1. 20 women receiving standard treatment with a CO2
pneumoperitoneum. 20 women being randomized to full conditioning of the peritoneum plus rinsing
with heparinised saline (5000 IU/l) plus injection of 5mg dexamethasone at creation of the
pneumoperitoneum plus application of an anti-adhesions local barrier (Hyalobarrier Gel Endo®) at
the site of surgical wound by the end of surgery.
A second laparoscopy to assess adhesion formation was performed some two weeks after
surgery under general anaesthesia using a 5 mm laparoscope. Early repeat laparoscopy for
93
atraumatic lysis of pelvic adhesions is used routinely in many centres for infertility patients. During
this early repeat laparoscopy any existing adhesions were lysed and adhesions were scored by the
surgeon.
Scoring of adhesions. A new scoring system for adhesion formation was developed based
upon the lessons learned from previous classification systems as used in the human and in animal
models, where adhesions are scored for type, tenacity and extend..
We scored in 7 areas type and tenacity (0 to 3) together with extend, which r was scored
as the diameter of a theoretical circle comprising all adhesions (table 4). Type and tenacity were
scored as 0 to 3 when adhesions were absent, filmy with easy lysis and little or no coagulation,
mixed filmy and dense, and dense adhesions requiring sharp dissection and coagulation respectively.
The area involved was scored quantitatively as the diameter of a theoretical circle comprising all
adhesions. This area in fact describes the sum of all denuded areas caused by adhesiolysis. Filmy
adhesions thus almost invariably will cover a small area, i.e. often long but very thin. The areas
scored separately were kept to a minimum and were based upon clinical importance and surgical
difficulty. Were scored separately: (1) vescico-uterine fold, (2) adhesions between abdominal wall
and bowels or omentum (almost invariably a consequence of previous surgery) and (3) adhesions
between the posterior wall of the uterus and the pouch of Douglas or bowels (as a consequence of
previous surgery, PID of deep endometriosis). Areas 3 and 4 comprised adhesions around left and
right ovaries and oviducts together and areas and scored adhesions between bowels and lateral side-
walls in the pelvis. Finally group 8, in order to be complete, grouped all other adhesions outside the
pelvis without further specification. For these trials we calculated the adhesion load at each site by
multiplying scoring of type and tenacity (severity) with the area of adhesions in mm of diameter
(extension).
94
Areas Type/Tenacity Extension Partial Total x area
Vescico-uterine fold
0 to 3 Diameter in mm Type/Tenacity x Extension
Anterior pelvic wall
0 to 3 Diameter in mm Type/Tenacity x Extension
Posterior uterine side/Douglas
0 to 3 Diameter in mm Type/Tenacity x Extension
Left adnex 0 to 3 Diameter in mm Type/Tenacity x Extension
Right adnex 0 to 3 Diameter in mm Type/Tenacity x Extension
Left pelvic wall 0 to 3 Diameter in mm Type/Tenacity x Extension
Right pelvic wall 0 to 3 Diameter in mm Type/Tenacity x Extension
TOTAL SCORE ∑ partial total
Outside pelvis Yes or Not; short description for location and severity
Table 4. Clinical scoring system. 7 pelvic areas are scored for severity (type + tenacity) and extension (diameter of a theoretical circle
comprising all adhesions); the total score is obtained by multiplying severity score and extension score.
95
Preliminary results
20 women were randomized 1:1. 9 women received standard treatment with a CO2
pneumoperitoneum, 11 women were randomized to full conditioning of the peritoneum plus rinsing
with heparinised saline plus Hyalobarrier gel.
Statistically significant differences between the two groups were observed (fig.34). In the
control group severe adhesions were systematically found, whereas in the full conditioning group,
adhesions were either minimal (in 2 women) or not existent (in 9 women).
Figure 34. Adhesions scored at the initial laparoscopy (beginning and end of surgery) and at repeat laparoscopy two weeks later
in the control group (square) and in the full conditioning group (dot).
Conclusion. The adhesion formation reduction is almost complete exceeding the most
optimistic expectations.
96
d. The clearance rate of peritoneal fluid after surgery with full conditioning
Absorption of Ringer lactate and Adept after full conditioning of the peritoneal cavity during laparoscopy. Verguts J, Corona R, Timmerman D, Koninckx P. Manuscript in preparation
Aim of the study. In order to test the efficacy of two fluids, the clearance rate of Ringer-
lactate versus Adept® in case of classic pneumoperitoneum with 100% CO2 and a gas mixture ( 86%
CO2 + 4% oxygen + 10% nitrous oxide) was studied in humans.
Materials and methods
Prospective randomised trial using pure CO2 or full conditioning with 1000 ml of fluid: Ringer-
Lactate or Adept® left in the abdomen following laparoscopic hysterectomy.
20 women were randomized 1:1. N=10 for each subgroup. The volume was estimated by ultrasound
at times 0, 24h, 48h and 72h as described later. All women undergoing a laparoscopic subtotal
hysterectomy were eligible for this trial. No limitation in body mass index or uterine size was made.
Laparoscopy was standardized by adjusting the insufflation pressure exactly to 15 mm of Hg
in all patients. Rinsing of the abdomen was performed with the randomized product (Ringer-Lactate
or Adept®). One litre of the fluid was left in place after surgery.
The first measurements were performed immediately after surgery in order to account for
any differences in instilled fluid. We estimated the residual volume of fluid in the abdominal cavity in
a 30° anti-Trendelenburg position after 1 minute in this position. The volume was estimated by
ultrasound immediately after surgery (time 0) and after 24, 48 and 72 hours.
Results
As expected, fluid absorption was fastest in the carbon dioxide group when Ringer lactate
was used: no detectable fluid after 48 hours and the slope of the curve was the steepest.
97
The slowest resorbtion occurred however in the carbon dioxide group where Adept® was
used where a pocket of fluid was still detected after 72 hours. Resorbtion in the full conditioning
group when Adept® was used was slower than when Ringer lactate was used (Fig.35).
Figure 35. Fluid absorption as calculated by consecutive ultrasound measurements, mean and SEM
Conclusion. The clearance of Ringer and Adept shows a clear exponential decline in contrast
with the data in the literature. Both the fluids have a better clearance with slower resorption after
surgery with conditioning.
98
99
Chapter 5. Discussion
The new concept of adhesion formation is that the entire peritoneal cavity is a cofactor in
adhesion formation. This concept and that conditioning of the peritoneal cavity is a key element in
the prevention of postoperative adhesion formation has originated almost exclusively from our
group derived from 15 years of experiments performed in rabbit and mouse model and fit with all
experiments performed so far.
Especially the development of a laparoscopic mouse model, which has been fully
characterized by now, has been important in demonstrating that factors from the peritoneal cavity
can enhance 4 to 10 times adhesion formation following surgery.
The mechanism involved is understood as follows. The peritoneum is a surface lining the
abdominal cavity composed by mesothelial cells. These cells are normally large, flat cells resting upon
the basal membrane and covering the entire peritoneal cavity (fig. 36).
It was demonstrated that when ‘traumatized’ these large flat cells retract and bulge, thus
exposing between them directly the extracellular matrix (73). Following injury and the exposure of
the basal membrane an acute inflammatory reaction, with release of inflammatory mediators and
other unidentified factors, immediately starts (fig. 37).
100
__________________________________________________________________________________
Figure 36. Electro- microscopical view of the intact peritoneal layer (magnification x710) by Volz, Surg Endosc 1999 (left) and draw of a schematic section (right), with permission of Fisher & Paykel.
__________________________________________________________________________________
Figure 37. Electro- microscopical view of the peritoneal layer after 2 hours of pneumoperotoneum (magnification x710) by Volz, Surg Endosc 1999 (left) and draw of a schematic section (right), Corona modified with permission of Fisher & Paykel.
______________________________________________________________________________
Figure 38. Electro- microscopical view of the peritoneal layer after 12 hours of pneumoperotoneum (magnification x1,200) by Volz, Surg Endosc 1999 (left) and draw of a schematic section (right), Corona modified with permission of Fisher & Paykel.
101
It is classically accepted that, the inflammatory reaction is associated with exudation and
fibrin deposition. This fibrin, unless removed within a few days will constitute a scaffold for fibroblast
and vessel growth thus attaching the 2 opposing surfaces with adhesions (figure 38).
This process of fibrin removal and healing or adhesion formation occurs within a few days
after surgery. A key experiment has been the observation in animal models that keeping the surfaces
separated by a membrane for 3 days is sufficient to prevent adhesion formation (65). In this thesis
(4.1.1 a, 4.1.1 g) we clearly demonstrated for the first time that this acute inflammatory process
strongly correlated with adhesion formation playing as a driving mechanism. Most important, we
showed that the inflammation involve equally the entire peritoneal cavity and that “traumatizing”
the abdomen, even far from the surgical lesion, inflammation and post-operative adhesions between
opposing injuries increase some 5 to 10 fold largely extending and exceeding the classic concept of a
local process.
We identified as “bad factors” of the peritoneum, besides direct surgical trauma, mesothelial
hypoxia, e.g. the CO2 pneumoperitoneum during endoscopic surgery, mesothelial hyperoxia e.g. the
20% oxygen in the air during open surgery, ROS, desiccation and blood. Surgical manipulation of the
bowels in the upper abdomen increases adhesions in the lower abdomen and this increase is dose
dependent with duration of manipulation. CO2, generally used for the pneumoperitoneum for safety
reasons because of its high solubility in water and its high exchange capacity in the lungs, causes
hypoxia of the upper mesothelial layers Air contains 20% of oxygen, i.e. a supraphysiological partial
pressure of some 150mm of Hg resulting in reactive oxygen species (ROS) and an increase in
adhesion formation which as expected is duration and pressure dependent. Desiccation and blood
also increases adhesion formation and the effect is dose dependent. “Good factors” have been also
identified such as cooling, N2O, new gas mixture and dexamethasone (figure 39).
102
______________________________________________________________________________
Figure 39. Bad and good factors from the peritoneal cavity involved in adhesion formation/prevention.
The experiments presented in this thesis will be discussed separately in the following paragraphs.
103
5.1 Animal investigations
5.1.1 The role of the entire peritoneal cavity and the importance of training and quality of surgery on adhesion formation
That training and quality of surgery are directly related to adhesion formation is a widely
accepted concept, but while the correlation with decreasing in operation time, post-operative
complication and costs are demonstrate in several experiments (75;165;166) and clinical trials (167-
169), the effect on adhesion formation it was only based on clinical observations and common sense.
The experiments include in this thesis demonstrate that tissue manipulation enhances
adhesion formation and thus supports the concept of gentle tissue handling. The effect is evident
with different types of tissue manipulation, i.e. touching and grasping. Also the higher adhesion
formation at the beginning of the learning curve is consistent with more manipulation by the
inexperienced surgeon and decreased adhesions during the learning curve (Figure 1). The effect
clearly is independent of the duration of surgery since in these experiments the duration of the
pneumoperitoneum and of the anaesthesia was kept constant at 60 min.
The tissue manipulation increased the adhesion formation and this effect is evident with
different types of tissue manipulation, i.e. touching and grasping. Also the higher adhesion formation
at the beginning of the learning curve is consistent with more manipulation by the inexperienced
surgeon and decreased adhesions during the learning curve. Clinically this has far reaching
importance, since during surgery mobilization cannot be completely avoided during exposition and
dissection. It confirms that gentle tissue handling is important and extends the observations of
learning curves to tissue manipulation trauma which clearly is much more important when the
surgeon has less experience.
The experiments on tissue manipulation during surgery confirm and extend the model that
the peritoneal cavity is a cofactor in adhesion formation between traumatised peritoneal areas.
104
Although the adhesions occur only at the lesions' sites, i.e. the uterus and the abdominal wall, the
amount of adhesion formation depends on factors in the peritoneal cavity or fluid. These
experiments demonstrating that manipulation outside the lesions, will effect adhesion formation
unequivocally demonstrate that the effect has to be mediated through the peritoneal cavity or fluid.
The mechanism and the factors involved are still unknown, although they seem to be a consequence
of the trauma caused by manipulation, or by desiccation, or by hypoxia, or by reactive oxygen
species. Obviously this is not the peritoneal macroscopic injury, since no adhesion formed between
the organs manipulated, but only at the area of injury. Adhesion formation increased with the
duration of manipulation, i.e. also the amount of manipulation, thus supporting the concept of
gentle tissue handling.
Clinical adhesion prevention today is limited to barriers and flotation agents, both of which
have been demonstrated to be effective. The mechanism of their action, however, could be
interpreted differently. We speculate that the positive effect of these agents, particularly for barriers,
can be due to their capability to play as a protective layer avoiding the bad peritoneal factors to
reach the site of the surgical lesions. This concept is supported by the finding that the histological
specimens taken after application of Intercoat gel in our experiment, did not presented an
inflammatory reaction.
The concept of the peritoneal cavity as a cofactor in adhesion formation has other far
reaching implications for adhesion prevention, since therapy should also be oriented to minimizing or
preventing these peritoneal factors. Besides quality of surgery, i.e. gentle tissue handling and
duration of surgery, and minimising the mesothelial trauma of ROS, hypoxia and desiccation by
humidification, adding small amounts of oxygen and cooling other factors should be considered e.g.
dexamethasone. In these experiments we specifically used dexamethasone since proven efficacious
in previous experiments (98;179), and since we anticipated some inflammatory reaction following
manipulation. Dexamethasone confirms to be highly effective to attenuate the effect of manipulation
105
during surgery, as shown by reducing manipulation enhanced adhesion formation. This effect
however, is not specific for manipulation enhanced adhesions, since the reduction was only
proportional to the control group, i.e. adhesions after manipulation and dexamethasone treatment
remaining more important than after dexamethasone only.
The model that the peritoneal cavity is a cofactor in adhesion formation at the level of
peritoneal trauma is bound to influence our clinical concept of adhesion prevention. Firstly the
clinical wisdom of gentle tissue handling and careful and blood less surgery becomes experimentally
supported. The importance of experience and training gets another dimension, since besides
reducing complication rates and operation time it also reduced tissue manipulation. We thus begin
to understand why training and experience are the first key factors in adhesion prevention.
The impact of surgical training upon adhesion formation is clearly proved in our experiment
where experienced and non-experienced surgeons are compared in a learning curve. Experienced
surgeons not only started with lower duration of surgery but also achieved the plateau earlier (after
10 procedures) whereas non-experienced surgeons started with longer duration of surgery and,
although the duration decreased, it didn’t achieve the level of experienced surgeons even after 80
consecutive procedures. A less traumatic, more precise and gentle surgical technique gained with
experience appears to be important since post-operative adhesion formation was lower already in
the ten first procedures within the group of experienced surgeons in comparison with non-
experienced surgeons confirming the data of the effect of manipulation-enhanced adhesions. This
difference among groups of surgeons decreased with number of surgical procedure underlining the
importance of training. Gentle tissue handling, however, obviously is more complex than is reflected
in the duration of surgery, since for similar operating times; experienced surgeons had always fewer
adhesions than inexperienced surgeons during the entire learning curve.
Postoperative adhesion formation is influenced not only by surgical training but also by the
duration of the pneumoperitoneum which in these studies was kept constant at 60 min. Less
106
adhesion formation with shorter procedures may happen due to less exposure to CO2
pneumoperitoneum and thus to less pneumoperitoneum-enhanced hypoxia, CO2 and pH changes. In
the study performed in rabbits, duration of surgery also seemed to be important in the outcome of
postoperative adhesion formation although it is difficult to completely separate both effects since
duration of surgery and training of surgeon are intimately related.
Other reports studying only the effect of CO2 or helium pneumoperitoneum upon post-
operative adhesion formation have shown an important effect of the time of exposure to both CO2
and helium pneumoperitoneum upon adhesion formation, those studies show the same situation,
the longer the exposure, the higher the adhesion formation (25;26).
Based on the classic process of adhesion formation, fibrin related, it’s commonly believed
that, during surgery, bleeding should be avoided and a good haemostasis should be adopted as one
of the general adhesion reduction measures (180;181) Despite this concept is widely accepted, it’s
mainly based only upon careful observational medicine and common sense and upon indirect data as
the effectiveness of some anticoagulants (182-184) in adhesion prevention but it was not yet directly
experimentally evaluated.
To the best of our knowledge, this is the first experimental study detailing the effect of blood
in the peritoneal cavity upon adhesion formation. The effect was clearly shown to be dose
dependent with a sigmoid relationship between adhesion formation and amount of blood. As little as
0.125 ml already significantly increased adhesions, the effect increased further exponentially up to
0.5 of blood, with little higher effect of 1 ml. It should be realized, however, that 0.5 and 1 ml of
blood are high volumes of blood considering that the total amount of blood that can be retrieved
from 1 mouse is around 2 ml. The adhesiogenic effect of blood corresponded strikingly to the sum of
the adhesiogenic effect of plasma and red blood cells separately. This suggests that besides the likely
fibrin deposition, acute inflammation of the entire peritoneal cavity plays an important role. These
data thus are consistent with the observations that acute inflammation in the entire peritoneal cavity
107
is an important mechanism enhancing adhesion formation between injured areas. These data
moreover lend further support to the concept that he adhesion enhancing mechanism through acute
inflammation of the peritoneal cavity is quantitatively much more important than the adhesions
resulting from the peritoneal trauma only, which however remain a prerequisite for adhesion
formation since in none of the animals de novo adhesions were found, also not after the addition of
large amounts of blood. We therefore suggest two separate roles of fibrin deposition in adhesion
formation. After a peritoneal injury, exudation and local fibrin deposition occurs and if this fibrin is
not removed by fibrinolysis within a few days, adhesions start to form locally. Blood and/or fibrin,
reaching the peritoneal cavity in addition, cause an acute inflammation of the entire peritoneal cavity
as evidenced by the pain and the rise in CRP following an intra-abdominal bleeding (185;186). It is
unclear to what extend fibrin deposition in the peritoneal cavity plays a specific role in adhesion
formation besides the acute inflammation. Indeed, an overload of fibrin could decrease the
availability of plasmin necessary for the local fibrinolysis between injured areas. These concepts also
shed new light on the unclear and conflicting results of the effect of tPA upon adhesion formation.
In conclusion, these data confirm and extend the concept that the acute inflammation of the
entire peritoneal cavity is quantitatively the most important driving mechanism in adhesion
formation, and that good surgical practice and peritoneal cavity conditioning are the cornerstones of
adhesion prevention. Identified good factors today are the replacement of pure CO2 for the
pneumoperitoneum with CO2 with at least 5% of N2O eventually with 10% N2O with some 3-4% of O2,
the reduction of the pneumoperitoneum temperature to below 32°C , the prevention of desiccation
using humidified gas86, which requires cooling with a third means (paragraph 4.2.1 a,b), the
decreasing of the mechanical surgical trauma and, finally the avoiding of bleeding in the peritoneal
cavity during surgery.
108
5.1.2 The genetic effect on adhesion formation
Our studies indicate that genetic background could also influence adhesion formation, at
least after laparoscopic surgery (134). Balb/c mice, an inbred strain, showed lower inter animal
variability and higher adhesion formation (the later not significant) in comparison to Swiss mice an
outbreed strain, observation to take into account when developing a standardized animal model for
the study of adhesion formation. Strain differences have been reported for other processes involving
fibrosis and healing responses such as hepatic, lung, and colorectal fibrosis (187-189), myocardial and
ear wound healing (190;191) and bone regeneration (192). This is not surprising because inbred
strains, maintained by sibling (brother × sister) mating for 20 or more generations, are genetically
almost identical, homozygous at virtually all loci, and with high phenotypic uniformity (193). This less
interanimal variability in inbred strains has been reported for many processes such as sleeping time
under anaesthesia (194). Among the strains evaluated we found that Swiss, NMRI, and BALB/c mice
developed more adhesions compared to FVB and C57BL/6J mice, in which adhesion formation was
minimal. About the mechanisms causing these interstrain differences, at present, we only can
speculate.
Our study demonstrated that adhesion formation varies also depending on the BALB/c
substrain used. Mice after going through the same laparoscopic procedure performed by the same
experienced and well trained surgeon, have shown different quantity of postoperative adhesions, i.e.
BALB/c OlaHsd substrain developed less adhesions (proportions around 10%) than substrains
BALB/cByJ and BALB/cJ@Rj; and substrains BALB/cByJ and BALB/cJ@Rj have developed the same
quantity (proportions both around 25%). Differences were also observed for manipulation, i.e.
substrain BALB/c OlaHsd was significant more sensitive to the manipulation than the other two
strains, developing twice adhesions that the control without manipulation (10% vs. 20%). BALB/cByJ
substrain was a bit sensitive to the manipulation (25% vs. 35%) and BALB/cJ@Rj strain was not
109
sensitive to the effect of manipulation developing the same proportion than the control group
without manipulation (both 25%).
About how to explain these differences in adhesion formation and in manipulation enhanced
adhesions, we can only hypothesize. Inbreed strains offers several advantages over outbreed mice,
i.e. genetic and phenotypic uniformity. Mouse strains can be termed inbred if they have been mated
brother x sister for 20 or more consecutive generations, and individuals of the strain can be traced to
a single ancestral pair at the 20th or subsequent generation. At this point the individuals' genomes
will on average have only 0.01 residual heterozygosis (excluding any genetic drift) and can be
regarded for most purposes as genetically identical. Established inbred strains may genetically
diverge with time into substrains, due to a number of circumstances. For instances, if two branches
are separated after 20 but before 40 generations of inbreeding there still will be enough residual
heterozygosis that two genetically different substrains will result (195); if branches are separated for
more than 20 generations from a common ancestor, it is likely that genetic variation between the
branches will have occurred by mutation and genetic drift; or if genetic differences are proven by
genetic analysis to have occurred between branches. This might have happened with these three
substrains of BALB/c.
If we have a look to the nomenclature, inbred strains nomenclature is a combination of the
parent strain and substrain designation. Substrains are given the root symbol of the original strain,
followed by a forward slash (/) and a substrain designation. The designation is usually the laboratory
code of the individual or laboratory originating the strain. Substrains may give rise to further
substrains by continued maintenance by a different investigator or through establishment of a new
colony. In addition, substrains arise if demonstrable genetic differences from the original substrain
are discovered. In either case, further substrain designations are added, without the addition of
another slash (196). For instances, in this study, three BALB/c substrains were used, i.e. BALB/cByJ,
BALB/cOlaHsd and BALB/cJ@Rj. Taking into account the nomenclature of The Jackson Laboratory
110
(197) and the articles of Dr Staats (198) and of Dr Davisson (199), “By” comes from Dr Donald W.
Bailey (The Jackson Laboratory, Bar Harbor, USA), “J” is the laboratory code for The Jackson
Laboratory (Bar Harbor, USA) in the first substrain . In the second substrain, “Ola” is the laboratory
code for Oxford shire Laboratory Animal Colonies (Shaws Farm, Bicester, Oxon, England) and “Hsd”
for Harlan Sprague-Dawley, Indianapolis). In the third substrain, “J” is the laboratory code for “The
Jackson Laboratory” and “Rj” for Centre D'Elevage R. Janvier, meaning the strain BALB/c substrain J is
bred at (@) the Centre D'Elevage R. Janvier (Rj). Therefore, it is clear we are talking about in three
substrains defined.
The history and characteristics of the BALB/c family have been carefully reviewed by M.
Potter and it begins in 1913 when Dr. Halsey J. Bagg obtained an albino breeding outbred stock from
a pet dealer in Ohio (200). From the data sheet information provided by the companies where mice
were bought, BALB/cByJ substrain (Charles River Laboratories) is coming from The Jackson
Laboratory when pedigree inbred pairs from were introduced into Charles River Laboratory France in
May 1982 at F156 (http://jaxmice.jax.org/strain/001026.html). The BALB/c OlaHsd substrain (Harlan
Laboratories) is derived from the Laboratory Animal Centre, Carshalton UK from the Jackson
Laboratory, Bar Harbor, Maine in 1955 and in 1976 they were bred by OLAC (now Harlan UK)
(http://www.harlan.com/products_and_services/research_models_and_services/research_models_
by_product_type/inbred_mice/balbc.hl). The source of BALB/cJ@Rj strain (Janvier, France), which is
bred now in Janvier France, is Zentralinstitut Für Versuchstierzucht- Hanover 1988 at F172 (data
sheet provided by Bio Service, nl). Therefore, it is clear that three sublines or substrains were
generated that can explained the differences in postoperative adhesions in this laparoscopic mouse
model.
Every mammalian species studied to date possess a tightly linked cluster of genes, the major
histocompatibility complex (MHC), whose products are associated with intercellular recognition and
with self/nonself discrimination. The MHC complex is a region of multiple loci that play major roles in
111
determining whether transplanted tissue will be accepted as self (histocompatible) or rejected as
foreign (non-histoincompatible). The MHC plays a central role in the development of both humoral
and cell-mediated immune responses. For this reason, the MHC has been implicated in the
susceptibility to disease and in the development of autoimmunity. The MHC is referred to as the HLA
complex in humans and as the H-2 complex in mice. The loci constituting the MHC are highly
polymorphic, that many alternate forms of the gene, or alleles, exist at each locus. Each set of alleles
is referred to as a haplotype. An individual inherits one haplotype from the mother and one form the
father. In outbred populations, the offspring are generally heterozygotes. In inbred mice, however,
each H-2 locus is homozygous because maternal and paternal haplotypes are identical (201).
Therefore, these three substrains of BALB/c have the same MHC H2 haplotype, the H2d.
Although it is often difficult to decipher the exact genetic causes, substrains from different
sources can lead to phenotypic differences among mice. For instance behavioural differences among
substrains (202-203); tissue rejection among related substrains e.g. 129 substrains (204), tumor
susceptibility differences e.g. C3H substrains (205); body and organs weight (206) and sperm
abnormality differences (207).
Genetic differences among substrains can occur by several mechanisms, i.e. residual
heterozygosis at the time of separation; undetected spontaneous mutations that become fixed in the
colony; or undetected genetic contamination or deliberate outcrossing of strains for specific
experimental purposes (208). Examples of undetected spontaneous mutations that become fixed in
the colony are a deletion of the alpha-synuclein locus (a presynaptic involved in a range of
neurodegenerative diseases) has been observed in C57BL/6JOlaHsd7 (209), mutations at the Raf-1
locus (controlling the expression of alpha-fetoprotein) only in BALB/cJ, at the Gdc-1 locus (governing
cell surface antigens) in BALB/cHeA only, the Qa-2 locus (governing L-glycerol 3-phosphate
dehydrogenase activity in the liver) in BALB/cByA (206). However, if these phenotype differences are
due to these genetic mutations was not clear.
112
These observations (i.e., strain and substrain differences in adhesion formation potential and
in interanimal variability) point to genetics effects on adhesion formation, which is not surprising and
confirms clinical observations. The importance of these observations is twofold. First, to study the
genetic involvement in detail, the use of two strains with high and low adhesion formation potential
can be considered as an experiment of nature. Second, to study adhesion formation and prevention,
it is preferable to use a strain with high adhesion formation potential and low interanimal variability,
such as BALB/c mice. Furthermore, fewer inbred animals will be needed to achieve a given level of
statistical precision than if outbred animals had been used (134). We, however, want to point out
that inbred strains in general weigh less than outbred strains (average of 20 g vs. 32 g), which
increases the technical skills required to do the experiments, especially those involving laparoscopic
surgery.
In conclusion, our data demonstrate that some mouse strains develop more postoperative
adhesions than others and that the interanimal variability in inbred strains is less. These data should
not be underestimated for adhesion formation studies in animal models and, although very
preliminary, can open new insights in the pathogenesis and treatment of adhesion formation in
humans.
It is well known that therapies doesn’t work for all patients all of the time and clinical
research has provided insights into various factors that impact how drugs work for different patients,
including disease state, comorbidities, concomitant drugs, nutritional state, and pharmacogenomics
(PGx). PGx is the study of how gene variations and the different ways genes are expressed can impact
how an individual responds to drugs. Significant developments in the field of genetics are finally
enabling health care providers, pharmacists and patients to think about new ways to optimize drug
therapy based on the very new concept of personalized medicine.
113
5.1.3 Inflammation and adhesions
The observed parallelism between changes in adhesion formation and in acute inflammation
scores is striking with a linear correlation between adhesion and inflammation scores. This suggests
that acute inflammation is an important common and driving mechanism modulating adhesion
formation. The similarity of acute inflammation at the lesion and, in the peritoneum 5 mm from the
lesion and in the entire peritoneal cavity strongly supports the concept that the entire peritoneal
cavity is a cofactor affecting adhesion formation at the lesion site. Especially the effect of bowel
manipulation in the upper abdomen i.e. at a distance from the bipolar lesion strongly supports the
concept that some peritoneal cavity factors stimulated by the mesothelial trauma with exposure of
the basal membrane enhance adhesion formation and also inflammation, at the lesion site. The
absence of de novo adhesions confirms that adhesion formation requires a peritoneal lesion, but
that quantitatively the inflammatory reaction in the entire peritoneal cavity is the most important
factor.
However, the exact mechanism by which the inflammatory reaction of the entire peritoneal
cavity affects adhesion formation at the lesion site is still unclear. Since mesothelial cells are known
to retract and to bulge during CO2 pneumoperitoneum without affecting the basal membrane, and
since bowel manipulation was done very gently, a very superficial mesothelial cells trauma is
suggested as causing an inflammatory reaction of the entire peritoneal cavity. Subsequently some
substances or cells could be released, activated or attracted into the peritoneal fluid affecting
adhesion formation at the lesion site. We speculate that chemokines could be involved since they are
known as inflammatory mediators involved in the activation and migration of leukocytes into the
tissue (210;211). Indeed according to the position of the first two cysteine residues (212), some being
chemoattractants and activators of non-PMNs leukocytes while others attract neutrophiles. Also
macrophages and leucocytes attracted into the peritoneal cavity and their secretion products as
cytokines are likely to be involved.
114
Whereas non-steroidal anti-inflammatory drugs (NSAIDs), as ibuprofen, tenoxicam,
nimesulide, parecoxib, and TNF alpha neutralizing antibodies did not affect adhesion formation (98),
dexamethasone was confirmed to decrease adhesion formation (98;161)while decreasing the acute
inflammatory reaction. Also the mechanism by which dexamethasone decreases the inflammatory
reaction in the peritoneal cavity can only be speculated upon.
In this study, dexamethasone did decrease the acute inflammatory reaction and
proportionally adhesion formation as previously demonstrated (98;213). This suggest that
dexamethasone can be acting through other postulated mechanisms such as inhibition of fibroblast
proliferation, depression of procollagen gene expression through a decreased transforming growth
factor secretion (214), or immunosuppressive effects the production and release of cytokines (215) .
Our data are tempting to consider the entire process of adhesion formation to be
determinated by acute inflammation. Since it was demonstrated that dexamethasone produces its
anti-inflammatory effect by inducing the expression of mitogen-activated protein kinase (MAPK)
phosphatase-1 (MKP-1), we postulate that this pathway could be involved in the adhesion formation
process. The MAP kinase phosphatase (MKP)-1 is a negative regulator of cytokines production in
innate immune cells (216) and it has a negative regulation effect on the mitogen-activated protein
kinases (MAPKs) production (217). MAPKs include the extracellular signal-regulating kinase (ERK),
p38 MAPK and c-Jun N-Terminal protein Kinase (JNK) and they all play an important role in cell
proliferation, apoptosis and many other nuclear events. MKP-1 has been shown to inhibit a number
of cellular responses mediated by ERK and p38 MAPK.
MAPKs also regulate the metalloproteinases (MMPs) and MMPs have a role in fibrinolysis
and in adhesion formation. A key factor in adhesion prevention is fibrinolysis, which is regulated by
the plasminogen system. The inactive proenzyme plasminogen is converted into plasmin by tissue-
type plasminogen activator (tPA) and/or urokinase type plasminogen activator (uPA). The fibrin
matrix serves as a scaffold for fibroblasts and capillary ingrowth and for extracelullar matrix (ECM)
115
deposition. During normal healing, the fibrin matrix is rapidly removed and the ECM will be degraded
by MMPs. The traditional concept is that when the fibrin matrix persists too long, or when the ECM
degradation is inhibited, peritoneal adhesions will be formed. In addition, MMP-2 was demonstrated
to be expressed in mature human peritoneal adhesions (218).
Moreover, MMPs have also been implicated as important factor in the control of the tumor
implantation. MMPs play roles in pathological conditions involving untimely and accelerated
turnover of extracellular matrix, including inflammation, angiogenesis and metastasis (219-221).
Among MMPs, we focus our attention on matrix metalloproteinase 2, a secreted endopeptidase
homologous with interstitial collagenase but which possesses an additional fibronectin-like domain.
Specific cell surface receptors bind to fibronectines. These receptors include the traditional
fibronectin receptor, also called integrin alpha5 beta1, the major fibronectin receptor on most cells,
and several other integrines. Several studies have shown that the adhesive extracellular matrix
protein fibronectin and its integrin receptors function in certain types of adhesive contact as well as
playing a major role in matrix assembly (222). Integrin ligands, such as fibronectin, are not passive
adhesive molecules but are active participants in the cell adhesive process that leads to signal
transduction (223).
MMPs secretion is stimulated also by nitric oxide (NO), induced by the expression of the gene
nitric oxide synthase (i-NOS), and also associated to the angiogenesis process (219).
This enzyme is associated to the angiogenesis process (220) and, in addition its expression,
like the MPK-1 expression, is also inhibited by dexamethasone (222). These two important effects of
dexamethasone, not observed using NSAIDs, would explain why dexamethasone is the only anti-
inflammatory drug strongly effective on adhesion prevention.
In summary, the inflammation at the peritoneal cavity level, due to the mesothelial trauma,
causes an activation of a signal transduction pathway like MAPKs that induces an expression of
MMPs; at the same time, due to a local wound following the surgery, there is a fibrin matrix and
116
extracelullar matrix (ECM) deposition. We speculate that this two simultaneous events bring to an
over production of MMP- 2 that, miming the effect of the fibronectin, increase the fibrin and the
extracellular matrix assembling that lead to adhesion formation, confirming the traditional concept
of fibrin matrix persisting and insufficient ECM degradation.
In conclusion, these data strongly suggest that acute inflammation in the entire peritoneum
cavity is the driving mechanism of adhesion formation at the lesion in our laparoscopic mouse model.
In addition, MMPs may play an important role being a link between the two processes. Of course,
new experiments should be done to confirm our hypothesis.
5.1.4 Effect of Nitrous oxide on adhesions
For the first time in this thesis is showed that the use of N2O instead of CO2 can reduce
postoperative adhesion formation in our laparoscopic mouse model, in the open model and in
humans. Moreover, surprisingly this effect was already obtained with as little as 5% of N2O in CO2.
The effect of N2O, especially at concentrations of 5% was unexpected and cannot be
explained by current knowledge. The effect of lower temperatures (133) could be explained by
making cells more resistant to damage e.g. by hypoxia, the addition of a physiologic concentration of
oxygen (4%) could be explained by correction the mesothelial hypoxia during pure CO2
pneumoperitoneum 25 or mesothelial hyperoxia when more than 10% of oxygen was used (131).
The effect of humidification was easily explained by preventing desiccation, whereas gentle tissue
handling should decrease mechanical trauma (82;137). The mechanism of action of N2O, however
has to be different from the effect of oxygen, since the effect of even 100% of N2O is comparable to
the effect of 5%, whereas oxygen concentrations above 10% clearly increase adhesions. This explains
the additive effects of N2O and low concentrations of oxygen observed in the mouse model for open
surgery. This together with the investigation of the lowest effective concentration, and the effect
117
upon acute inflammation of the peritoneal cavity, which is almost linearly correlated with the extend
of adhesion formation in our models of hypoxia enhanced adhesions, hypoxia and manipulation
enhanced adhesions, and hypoxia and trauma enhanced adhesions and dexamethasone (137;223). In
addition we demonstrated that both gases, especially N2O can prevent the enhancing effect upon
adhesion formation of blood. We believe that these effects upon adhesion formation are mediated
through a decreased acute inflammation in the entire peritoneal cavity. That 4% of O2 is effective if it
is considered to be a consequence of preventing the mesothelial hypoxia. The effect of N2O,
especially at low concentrations is surprising, N2O could have anti-inflammatory effect on acute
inflammation since N2O is effective in decreasing the post-operative pain and blocking directly one of
the most precocious and important phase of the inflammatory process as the chemotaxis (224).
Chemotaxis of leucocytes also clearly decreases for more than 50% with N2O administration.
Suppression of chemotaxis to corneal inflammation by nitrous oxide (225). It remains surprising that
as little as 5 or 10% of N2O has these strong effects.
Nitrous oxide, commonly known as "laughing gas", is a colourless non-flammable gas used in
surgery and dentistry for its anaesthetic and analgesic effects. Nitrous oxide is a weak general
anaesthetic, so it is used as a carrier gas in a 2:1 ratio with oxygen for more powerful general
anaesthetic agents such as sevoflurane or desflurane. It has a MAC (minimum alveolar concentration)
of 105% and a blood: gas partition coefficient of 0.46. Less than 0.004% is (minimally) metabolized in
humans. Nitrous oxide is relatively non-polar, with a molecular weight of 44.013 g/mol and high lipid
solubility. As a result, it diffuses quickly into the phospholipid cell membranes. Though nitrous oxide's
exact mechanism of action is still open to some conjecture it is known that it acts as an NMDA
antagonist at partial pressures similar to those used in general anaesthesia. N2O affects the GABA
receptor (226) but this is still controversial. It has a lower potency by acting as a positive allosteric
modulator. N2O, like other volatile anaesthetics, activates twin-pore potassium channels, albeit
weakly and these channels are largely responsible for keeping neurons at the resting (unexcited)
potential (227). Thus the studies comparing N2O and CO2 pneumoperitoneum during laparoscopic
118
cholecystectomy (228) and demonstrating that N2O pneumoperitoneum produced less postoperative
pain and required a decreased quantity of anaesthetic for the surgical procedure than did CO2
pneumoperitoneum could be connected with nitrous oxides general anaesthetic’s properties.
N2O at concentrations less than 29% is safe as a gas to be used for the pneumoperitoneum
(127). Indeed below 29% N2O does not have an explosion risk. Above 29% N2O might maintain
combustion (229). Thus if gases as methane would escape from the intestine and if these gases got
ignited by electrosurgery some explosion risk could exist. This risk exists theoretically whereas it
seems supported by some reported accidents (230;231). Used at low concentrations of 10% or 5 %
therefore does not constitute an explosion risk even if used together with 3% of oxygen, and its use
can be considered absolutely safe. In addition the solubility in water is similar for N2O and for CO2
being 1.5 and 1.45 mg/l respectively. This contrasts sharply with the low solubility of O2 and N2 in
water. This high solubility with a high exchange capacity in the lungs makes N2O a safe gas to be used
during pneumoperitoneum.
It was reported that postoperative pain is less with N2O pneumoperitoneum than with CO2
pneumoperitoneum, and that intraoperative ventilation and metabolic management is easier by the
use of N2O pneumoperitoneum instead of CO2 pneumoperitoneum. It therefore was recommended
to change to N2O in case of refractory hypercarbia and to use it as the primary insufflation gas in
patients with little ventilatory reserve (121).
Resorption of N2O from the peritoneal cavity certainly when used as 10% N2O in CO2 is so low
that it is not detected in the expired gas (77).
The clinical implications of the effect of low concentrations of N2O are far reaching. First
given the direct effects upon the mesothelial cells between laparoscopic surgery and laparotomy, all
beneficial effects observed at laparoscopy, such as reduction in temperature, humidification, 4% of
oxygen and e.g. 10% of N2O are also applicable to open surgery. In addition, we expect that as
119
demonstrated for the addition of 4% of oxygen to the CO2 pneumoperitoneum (77), N2O, will also
reduce tumour cell implantation.
In conclusion, the addition of low concentrations of N2O to the CO2 pneumoperitoneum has
important advantages in addition to adding 4% of oxygen. Besides decreasing adhesion formation it
is expected to decrease postoperative pain. Since N2O has a similar density as CO2, it might
beneficially be used also for flooding the operative cavity during open surgery. If confirmed the
mixture CO2 + 10% of N2O + 4% of oxygen would become the gas of choice to be used for
pneumoperitoneum during laparoscopy and for flooding the cavity during open surgery.
5.1.5 Peritoneal conditioning in open surgery
How it has been proven in this thesis, adhesions can be prevented with conditioning of the
peritoneal cavity with prevention of desiccation, cooling and prevention of hypoxia during
laparoscopy (25;82;133). Our model for open surgery clearly showed that the same conditioning of
the surgical field is feasible and efficient also during laparotomy. In air the concentration of O2 is
much higher than the intracellular O2 partial pressure this inevitably leading to local hyperoxia with
occurring of ROS and oxidative stress, deleterious for adhesion formation (131;98). Since the gas
used to flood the surgical field (i.e. CO2 86% +N2O 10% + O2 4%) has an higher density of air it will act
as a protective field separating the operative field from the air environment and from the
consequent oxidative stress. The addition of a physiologic concentration of oxygen (4%) to this
mixture of gases corrects the mesothelial hypoxia during flooding with pure CO2 (25). The effect of
lower temperatures could be explained by making cells more resistant to damage e.g. by hypoxia in
laparoscopy (133) as well in open surgery. Moreover, during open surgery the organs are also in
contact with dried air that it is not a physiologic condition for the wet tissues. The effect of
desiccation was prevented by using of humidified gas and a cover box to avoid high velocity jet and
turbulent mixing with the ambient air that could bring to increasing in O2 concentration (232) and
120
desiccation rate (233). Since the translation of a cover box to the clinic could be difficult, this last
effect could be avoided in clinical practice by using for insufflation a gas diffuser (155).
In conclusion, the conditioning of the surgical field in open surgery looks feasible. The
mixture of gas with CO2 + 10% of N2O + 4% of oxygen is confirmed to be the gas of choice for flooding
the cavity during open surgery as well for the pneumoperitoneum during laparoscopy. When using
this gas mixture there is no difference in adhesion formation between laparoscopy or laparotomy.
5.1.6 Mesothelial cells in prevention of adhesions
These experiments confirm and extend the observations that cultured mesothelial cells
obtained after trypsinization of the peritoneal cavity can reduce adhesion formation as described in
rabbits and rats with mesothelial cells obtained from the omentum (116;117;138).
Since we know that following a 60 minute CO2 pneumoperitoneum mesothelial cells retract
thus exposing directly the basal membrane (73), we speculate that injected mesothelial cells may fill
the gaps between the retracted mesothelial cells thus attenuating the adhesion enhancing effect of
the CO2 pneumoperitoneum by preventing the deleterious inflammatory reaction of the peritoneal
cavity. Considering that 100.000 cells are needed for a half maximal effect, the hypothesis that the
cells affect the entire peritoneal cavity seems attractive, as 100.000 cells exceed what is necessary to
cover the injury which only measures 1cm by 1.6mm. Whether this observation on CO2
pneumoperitoneum enhanced adhesions can be extended to hyperoxia, desiccation, or manipulation
enhanced adhesions is likely but remains to be established.
The mechanisms of adhesion reduction by cultured mesothelial cells can only be speculated
upon. Whereas in pure surgical models, incorporation in the lesion as demonstrated in rats (150)
seems logic, in the peritoneal cavity enhanced models, incorporation in between retracted
121
mesothelial cells of the entire peritoneal cavity cannot be excluded. Moreover, it is unclear whether
the origin of the mesothelial cells and the culture conditions could be important or whether these
cells actively secrete substances in the peritoneal fluid. We even cannot ascertain that the effect is
specific for mesothelial cells since Alpay et al. demonstrated that also fibroblasts display different
immunologic characteristics with an effect on adhesion formation (103).
The clinical implications of these experiments are far reaching whereas the use of
mesothelial cells for adhesion prevention becomes attractive. Extensive lavage during and after
surgery might remove besides debris and fibrin also mesothelial cells and macrophages, and the
latter might be detrimental. Whereas collecting and culturing mesothelial cells for adhesion
prevention in the same patient is not realistic today, this method could become important if
fibroblast or other cell lines with similar effects could be developed, i.e. stem cells and if these cells
could be preserved from the peritoneal lavage during surgery as is the case with peri-operative cell
salvage for blood, this could be a reliable method to prevent adhesions and the detrimental effects it
may cause.
5.2 Translation to clinical practice
5.2.1 Intra-peritoneal temperature and desiccation during endoscopic surgery
From the experiments comparing the real water loss as measured by the weight loss and the
calculated water loss, we can conclude that the measurements of flow rate, RH en Temp were
accurate, at least for flow rates up to 20L/min.
With the Storz humidifier RH decreases rapidly with flow rate. The main problem is the
cooling of the gas to room temperature when non isolated tubing is used. Since at 25°C CO2 cannot
hold more than 25mg of water/litre (=100% RH), this causes at low flow rates condensation in the
tubing and apparently adequate humidification up to 5L/min; entering a peritoneal cavity which is
122
initially at 37°C, the gas temperature will increase (100% RH at 37°C = 44mg of water/l) resulting
unavoidably in peritoneal desiccation. This combination of insufflation at room temperature and
desiccation resulted in unexpectedly low peritoneal cavity temperatures often below 30°C. As
expected, with non-humidified gas, peritoneal temperatures were even lower. The F&P humidifier
prevents cooling in the tubing using a heating wire. At higher flow rates however, the gas
temperature can be higher than 37°C. The luer lock at the end of the tubing limits flow rates to 7.8
L/min at 15mm Hg pressure. Therefore when used according to operating instructions, this
humidifier provides adequate humidification while the slight increase in temperature is compensated
by cooling in the trocar. When the luer lock is removed in order to permit higher flow rates outside
the control limits of the humidifier, to be used with the Thermoflator for CO2 laser surgery,
humidification still is adequate but the intra-peritoneal temperature could be slightly increased.
Considering the compartmentalization of the peritoneal cavity during surgery it should be stressed
however, that especially at higher flow rates and a CO2 laser set up, desiccation and temperature
changes mainly occur within a small compartment.
For adhesion formation it was demonstrated that desiccation is harmful while a slightly lower
mesothelial temperature is beneficial (82;133). Adhesions indeed decrease exponentially with
temperature at least up to 25°C, with over 80% of this beneficial effect being achieved at 31-32 °C.
With higher temperature adhesions increase rapidly and above 37°C the increase dramatically (133).
Clearly neither the Storz nor the F&P humidifiers are fully adequate to prevent adhesion formation,
the former resulting in desiccation the latter in intra-peritoneal temperatures up to 37°C. For both,
the effects upon desiccation and temperature respectively, increase with flow rates. We therefore
modified the F&P humidifier in order to deliver fully humidified gas at a preset value of e.g. 31-32°C.
This temperature was chosen since most of the beneficial effect was reached at that temperature,
whereas clinically and biologically we wanted to avoid temperatures below 28°C. In order to avoid
heating of the gas upon entrance, and thus desiccation we anticipated that the peritoneal cavity had
to be cooled by a third means such as a sprinkler with water at room temperature, and these
123
experiments were designed to estimate in the human how much cooling would be necessary to
prevent desiccation.
To our surprise, when fully humidified gas at 32°C was used, the desiccation and the heating
of the gas was within a few minutes much less than anticipated. This can be explained only by rapid
vasoconstriction of the peritoneal surface, especially in the compartment with the higher gas flow.
Therefore cooling with a third means to maintain 31-32°C in the peritoneal cavity and thus avoid
desiccation, turned out to be much easier than anticipated. Indeed little cooling by sprinkling
intermittently with 2-3 ml of saline at room temperature, together with the irrigation normally used,
turned out to prevent within minutes heating of the gas in the peritoneal cavity and desiccation.
The hypothesis that the bowel/peritoneal cavity temperature is regulated differently than
the core body temperature is supported by the observation that after some 70-80 min of surgery
with gas at room temperature and desiccation (dry gas or Storz humidifier) the temperature of the
bowel and wall suddenly dropped even below RT, and this without affecting core body temperature
as measured by the oesophageal body temperature. These observations make us conclude that the
temperature regulation of the bowel/peritoneal content is different from the core body temperature
while being more similar to the arms and legs which can react by vasoconstriction in order to
decrease heat loss and thus preserve core body temperature. In addition a constant Intra-peritoneal
temperature below 32°C did not affect the core body temperature, probably as a consequence of
important vasoconstriction of mesenteric vessels. This vasoconstriction also explains why in none of
the experiments, the calculated heat loss from the body exceeded 100 cal/min and why
condensation in any surgery caused fogging of the optics.
In conclusion in the absence of humidification the intra-abdominal temperatures are
surprisingly low mainly caused by desiccation. Even with a Storz humidifier temperatures hardly
exceed 28-30°C. Desiccation could be prevented completely by a modified F&P humidifier
maintaining the peritoneal cavity temperature at 31-32°C together with a little cooling. Surprisingly, a
124
little cooling by a third means was sufficient, as could be explained by the fact that the temperature
regulation of bowel and abdominal wall is different from the core body temperature and rather
similar to the limbs, reacting rapidly with vasoconstriction in order to prevent heat loss during
surgery.
5.2.2 Full conditioning of the peritoneal cavity
All findings from previous trials where only 4% of oxygen was added to the CO2
pneumoperitoneum were confirmed.
End tidal CO2 as a measurement for CO2 absorption further decreased to an almost steady
state after some 20 minutes. From this we may confirm that the mesothelial layer maintains its
integrity. This finding also adds up to the safety of the full conditioning where hypercarbia is almost
unthinkable under these conditions.
Pain was improved by full conditioning with decreased VAS especially during the first
postoperative days. This also resulted in a decreased use of pain killers (i.e. ibuprofen).
Looking at fluid absorption rates in the different groups, we saw the slowest absorption in
the women who received Adept®, and the fastest absorption when Ringer Lactate was used. There
was however an inversed correlation with the use of full conditioning, where Adept had an increased
absorption when full conditioning was used and a decreased absorption of Ringer lactate when full
conditioning was used. We can hypothesize that the full conditioning keeps the mesothelial layer
intact and thus providing more active transport of Adept® whereas Ringer is retained with an intact
mesothelial layer, explaining the somehow inversed relation. With the use of carbon dioxide and the
thus damaged mesothelial layer the Ringer lactate is easily absorbed through third space where on
the other hand there is no more active transport for Adept® possible and thus retained for a longer
period.
125
The preliminary results of the trial 1, confirmed all the finding in the animal models and the
strong effect of full conditioning on adhesion prevention. In this trial adhesion formation was scored
in a second laparoscopic look with a new clinical scoring system. In animals adhesions can be scored
quantitatively or qualitatively. A quantitative scoring, i.e. the area of the surgical lesions covered by
adhesions can only be done in animal models with a well standardized lesion. A qualitative scoring
scores separately extent, tenacity and type at a given localization and subsequently these scores are
added. Past experience in clinical trials on adhesion prevention made us aware of the limitations of
this scoring system. Firstly extent and tenacity and type of adhesions are strongly correlated, filmy
adhesions generally having low tenacity and covering a small area only. Secondly clinically important
adhesions such as thin, long, dense and vascularized adhesions are too rare to be picked up. Finally,
the number of sites to be scored remains debated (234). Moreover none of these scoring systems
have been clinically validated while statistical evaluation of adhesion data left us with the
conclusion that scoring adhesions at too many locations generally is rather counterproductive (54). In
order to be clinically useful, scoring systems should permit immediate scoring after surgery by the
surgeon. Otherwise subsequent blind review and scoring of video-recorded operations will be
necessary. Although this may at first glance be considered the most objective method, reviewing
videos balances between reviewing and scoring short video clips, and reviewing and scoring the
complete videotape of an intervention. The former has the drawback that not all adherences can be
grasped in short video clips, since the extent of a frozen pelvis can only be judged during and after
surgery. Reviewing complete registrations of the entire intervention, requiring advanced video-
compression and accelerated replay only recently has become technically realistic, while the time
required to review still might be prohibitively long to become practically realistic.
To the best of our knowledge all available adhesion scoring systems are based upon common
sense, surgical difficulty and anticipated impairment of fertility but have not been validated. It indeed
is unclear whether the scoring of extent, type, tenacity can be simply added or whether some should
get more weight in order to predict or measure risk of occurrence after surgery, or to predict
126
recurrence after lysis, or to predict fertility, pain or reoperation. The concept underlying this new
scoring system is similar as the proposed endometriosis classification (235), i.e. a design which will
permit validation afterwards. Firstly tenacity and type are so strongly intercorrelated that probably
one scoring will be sufficient. Before this will be confirmed and validated we today simply add the
separate scorings. Secondly for extend we prefer not to make classes when not necessary or which
can easily be calculated afterwards. Clinically the areas involved in different sites are variable: a
frozen pelvis indeed will have and extend at the backside of the uterus of more than 100mm
diameter, whereas adhesions of the adnexa will rarely cover an area of more than 50 mm in
diameter. We therefore use the diameter of each area involved in order to have a total score which
reflect the total area of adhesions involved. If extend would be scored as 0 to 3 in each site, smaller
sites will be overestimated. Anyway a scoring of extend in diameter can easily be converted to a 0 to
3 scoring if afterwards would be demonstrated that this is sufficient. With collection of more data it
is the aim to model the relative importance of area involved and type of adhesions in order to obtain
a more predictive total score of adhesions at any of the clinically important areas.
127
5.3 Conclusions
In conclusion, these data confirm the concept that the acute inflammation of the entire
peritoneal cavity is quantitatively the most important driving mechanism in adhesion formation, and
that good surgical practice and peritoneal cavity conditionings are the cornerstones of adhesion
prevention. Identified good factors today are the replacement of pure CO2 for the
pneumoperitoneum with CO2 with at least 5% of N2O eventually with 10% N2O with some 3-4% of O2,
the reduction of the pneumoperitoneum temperature to below 32°C , the prevention of desiccation
using humidified gas, which requires cooling with a third means, the decreasing of the mechanical
surgical trauma and, finally, the avoiding of bleeding in the peritoneal cavity during surgery.
Adhesion formation between traumatized areas has traditionally been considered as a local
process. However in laparoscopic surgery the pneumoperitoneum conditioning has lately shown to
be of importance as well as awareness that secretions and/or cells from the entire peritoneal cavity
strongly influence this local phenomenon has grown. Desiccation, mesothelial hypoxia occurring
during CO2 pneumoperitoneum, ROS occurring during open surgery and mesothelial trauma resulting
from grasping and manipulation have been also identified. If judged from animal models this
peritoneal effect is quantitatively much more important than the local phenomenon.
The concept emphasizing the importance of the peritoneal cavity has opened new
approaches to prevention. Gentle tissue handling is getting a new dimension while during surgery the
peritoneal cavity should be conditioned by preventing inflammation caused my mesothelial damage.
This damage can be avoided with the addition of N2O to the pneumoperitoneum, by preventing
hyperoxia and ROS, by preventing hypoxia by adding 3-4% of oxygen, by preventing desiccation and
by cooling. If used together with products as dexamethasone, and barriers an overall efficacy over
95% is obtained.
128
5.4 Summary
The new concept of adhesion formation is that the entire peritoneal cavity is a cofactor in
adhesion formation. This concept and that conditioning of the peritoneal cavity is a key element in
the prevention of postoperative adhesion formation has originated almost exclusively from our
group derived from 15 years of experiments performed in rabbit and mouse model and fit with all
experiments performed so far.
Especially the development of a laparoscopic mouse model, which has been fully
characterized by now, has been important in demonstrating that factors from the peritoneal cavity
can enhance 4 to 10 times adhesion formation following surgery.
The mechanism involved is understood as follows. The peritoneum is a surface lining the
abdominal cavity composed by mesothelial cells. These cells are normally large, flat cells resting upon
the basal membrane and covering the entire peritoneal cavity.
It was demonstrated that when ‘traumatized’ these large flat cells retract and bulge, thus
exposing between them directly the extracellular matrix. Following injury and the exposure of the
basal membrane an acute inflammatory reaction, with release of inflammatory mediators and other
unidentified factors, immediately starts.
It is classically accepted that, the inflammatory reaction is associated with exudation and
fibrin deposition. This fibrin, unless removed within a few days will constitute a scaffold for fibroblast
and vessel growth thus attaching the 2 opposing surfaces with adhesions.
This process of fibrin removal and healing or adhesion formation occurs within a few days
after surgery. A key experiment has been the observation in animal models that keeping the surfaces
separated by a membrane for 3 days is sufficient to prevent adhesion formation to a large extent. In
this thesis we clearly demonstrated for the first time that this acute inflammatory process strongly
correlated with adhesion formation playing as a driving mechanism. Most important, we showed that
129
the inflammation involve equally the entire peritoneal cavity and that “traumatizing” the abdomen,
even far from the surgical lesion, inflammation and post-operative adhesions between opposing
injuries increase some 5 to 10 fold largely extending and exceeding the classic concept of a local
process.
The acute inflammation of the abdominal cavity is shown to be caused, besides direct
surgical trauma, by duration and severity of mesothelial hypoxia during endoscopic surgery with CO2
pneumoperitoneum, mesothelial hyperoxia e.g. the 20% oxygen in the air during open surgery, ROS,
desiccation and blood. Surgical manipulation of the bowels in the upper abdomen increases
adhesions in the lower abdomen and this increase is dose dependent with duration of manipulation.
CO2, generally used for the pneumoperitoneum for safety reasons because of its high solubility in
water and its high exchange capacity in the lungs, causes hypoxia of the upper mesothelial layers Air
contains 20% of oxygen, i.e. a supraphysiological partial pressure of some 150mm of Hg resulting in
reactive oxygen species (ROS) and an increase in adhesion formation which as expected is duration
and pressure dependent. Desiccation and blood also increases adhesion formation and the effect is
dose dependent.
We demonstrated that in order to achieve a maximal adhesion reduction, the local barriers
should be combined with full conditioning of the pneumoperitoneum. The latter should aim at a
physiologic oxygen tension through the addition of 4% of oxygen and the prevention of desiccation
and the cooling of the abdominal cavity. The effect of each of these deleterious factors is greatly
attenuated when the temperature is lower, when the desiccation is avoided using humidified gas and
when oxygen and nitrous oxide are added to the pneumoperitoneum. At lower temperature cells are
more resistant to the trauma of hypoxia and hyperoxia as expected. They also are more resistant to
mechanical trauma and to the trauma of desiccation. The addition of 4% of oxygen recreate a
physiological cellular partial oxygen pressure, and that the deleterious effect of CO2
pneumeperitoneum is absent when the consequences of the hypoxia cascade are abolished as in HIF
130
knockout mice, in mice lacking angiogenetic factors as VEGF or PLGF, or when the effect of these
factors is blocked. N2O act most probably decreasing the acute inflammatory reaction but the exact
mechanism it remains unclear.
Finally we demonstrated that with peritoneal cavity conditioning (cooling, adding 4% of
oxygen and preventing desiccation) adhesions can be decreased by some 75%. Combining this
conditioning during surgery with a local barrier after surgery adhesions decrease by over 90%.
131
5.5 Samenvatting
Het nieuwe concept is dat postoperatieve vergroeiingen wel geïnitieerd wordt door een
lokaal trauma maar de ernst ervan vooral wordt bepaald door een inflammatoir proces in de ganse
buikholte, veroorzaakt door een combinatie van, chirurgische techniek, te weinig of teveel zuurstof,
uitdroging, fibrine of bloed. Deze stimulering van adhesies wordt tegengegaan door normoxie, dwz
de toevoeging van O2, door een lagere temperatuur, en het voorkomen van uitdroging. Het meest
belangrijke effect wordt bekomen door toevoeging van meer dan 5% N2O. Dit concept en zijn
toepassing om adhesies te voorkomen leidde tot het concept van conditionering van de peritoneale
holte als het sleutelelement in de preventie van postoperatieve verklevingen.
Dit concept is bijna uitsluitend afkomstig van Leuven na 15 jaar van experimenten in
diermodellen. Opvallend is dat alle experimenten tot nog toe volledig passen in dit model.
Het beland van de ganse buikholte is dan ook kwantitatief tot 20 keer belangrijker dan het
chirurgisch trauma zelf. De mechanisme werkt als volgt. Het buikvlies is een groot oppervlakte
bekleed met mesotheelcellen, grote, platte cellen die rusten op de basale membraan met als fuctie
gljden van darmen en vocht en transport. Bij het minste ‘trauma’ zullen deze grote platte cellen
samentrekken en uitpuilen, zodat de extracellulaire matrix niet meer bedekt is.
Dit heeft tot gevolg, dat water en gas sneller zullen diffunderen terwijl een acute
ontstekingsreaktie optreedt met secretie van stoffen in de buikholte.
Het klassieke concept is dat het chirurgisch trauma leidt tot een lokale ontstekingsreaktie met lokale
fibrinedepositie. Indien deze fibrine niet wordt verwijderd, vormt dit binnen een paar dagen een
steiger voor fibroblast ingroei en dus adhaesies. Het lokaal trauma is een noodzakelijke voorwaarde
voor vergroeiingen, maar kwantitatief veel minder belangrijk dan de graad van acute
ontstekingsreaktie in het ganse abdomen.
132
Dit concept werd vertaald naar de kliniek in translationeel onderzoek van de veiligheid en de
efficaiteit van de gekende gunstige faktoren. Dat het voorkomen van uitdroging geen probleem stelt
is duidelijk, maar dat men in afwezigheid van uitdroging de temperatuur van de buikholte kan laten
dalen tot onder de 30°C zonder de core body temperatuur te beïnvloeden is niet alleen nieuw, maar
ook belangrijk voor veiligeid. Het meest belangrijke effect, dit van toevoeging van 5-10% N2O is
eveneens volkomen veilig gezien de oplosbaarheid van N2O in water beter is dan van CO2. Toevoegen
van 3-4% zuurstof is onvoldoende voor een harttamponade. Het toevoegen van lage dosissen
heparine is reeds uitvoerig gebruikt in het verleden net zoals het geven van dexamethasone.
De klinische trials werden mogelijk door samenwerking met eSaturnus limited die de
koeling/bevochtiger bouwde. De resultaten van de klinische trials overtreffen de stoutste
verwachtingen. Indien gebruikt met een barrier verdwijnen adhaesies zo goed als volledig: dit is des
te meer waar omdat intentioneel lange en meoilijke chirurgie – diepe endometriose- als model werd
gekozen. Bovendien is er een duidelijke vermindering van postoperatieve pijn na chirurgie. Ook de
veel mindere resorptie ven CO2 gedurende chirurgie opent nieuwe perspectieven voor langdurige
ingrepen , zeker bij sterke Trendelenburg bij patiënten met een verminderde longfunctie.
Men mag bovendien verwachten dat de conklusies van laparoscopische chirurgie onverkort
zullen gelden voor open chirurgie. Bovendien, mag men verwachten dat conditionering eenzelfde
gunstig effect zal hebben op het voorkomen van tumorimplantatie bij kan chirurgie.
133
Chapter 6. Bibliography
1. Vesalius A. De Corpori Humani Fabrica. 1543.
2. Virchow. Peritonitis. In: Semashko NA, editor. 1928.
3. Baakdah H TT. Adhesion in gynecology complication, cost, and prevention: a review. Surg.Technol.Int 2005;14:185-90.
4. Diamond MP, Freeman ML. Clinical implications of postsurgical adhesions. Hum.Reprod.Update. 2001;7:567-76.
5. Pouly JL, Seak-San S. Adhesions: laparoscopy versus laparotomy. In: diZerega GS, editor. Peritoneal Surgery. New York: Springer; 2000. p. 183-92.
6. Weibel MA, Majno G. Peritoneal adhesions and their relation to abdominal surgery. A postmortem study. Am.J.Surg. 1973;126:345-53.
7. Brosens IA, Boeckx W, Page G. Microsurgery of Ovarian Endometriosis. Hum Reprod. 1988;3:365-66.
8. Boeckx W, Gruft L, Brosens I. Training in Tubal Microsurgery. Br J Obstet Gynaecol. 1985;92:266-69.
9. Gordts S, Boeckx W, Brosens I. Microsurgery of Endometriosis in Infertile Patients. Fertil Steril 1984;42:520-25.
10. Boeckx W, Gordts S, Brosens I. Microsurgery of Ovarian and Peritoneal Endometriosis. Contracept Fertil Sex (Paris) 1983;11:475-79.
11. Gomel V. Clin.Obstet.Gynecol. 1980;23:1301-10.
12. Karabulut B, Sonmez K, Turkyilmaz Z, Demirogullari B, Karabulut R, Sezer C et al. Omentum prevents intestinal adhesions to mesh graft in abdominal infections and serosal defects. Surg Endosc. 2006;20:978-82.
13. Ellis H. The Causes and Prevention of Intestinal Adhesions. Br J Surg. 1982;69:241-43.
14. Gomel V. Operative laparoscopy: time for acceptance. Fertil.Steril. 1989;52:11.
15. Gomel V. Infertility surgery: microsurgery. Curr.Opin.Obstet Gynecol 1992;4:390-99.
16. Gomel V. From microsurgery to laparoscopic surgery: a progress. Fertil Steril 1995;63:464-68.
17. Gomel V, Urman B, Gurgan T. Pathophysiology of adhesion formation and strategies for prevention. J Reprod Med.1996;41:35-41.
18. Filmar S, Gomel V, McComb PF. Operative laparoscopy versus open abdominal surgery: a comparative study on postoperative adhesion formation in the rat model. Fertil Steril. 1987;48:486-89.
134
19. Jorgensen JO, Lalak NJ, Hunt DR. Is laparoscopy associated with a lower rate of postoperative adhesions than laparotomy? A comparative study in the rabbit. Aust.N.Z.J.Surg. 1995;65:342-44.
20. Marana R, Luciano AA, Muzii L, Marendino VE, Mancuso S. Laparoscopy versus laparotomy for ovarian conservative surgery: a randomized trial in the rabbit model. Am.J.Obstet.Gynecol. 1994;171:861-64.
21. Luciano AA, Maier DB, Koch EI, Nulsen JC, Whitman GF. A comparative study of postoperative adhesions following laser surgery by laparoscopy versus laparotomy in the rabbit model. Obstet.Gynecol. 1989;74:220-24.
22. Schippers E, Tittel A, Ottinger A, Schumpelick V. Laparoscopy versus laparotomy: comparison of adhesion-formation after bowel resection in a canine model. Dig.Surg. 1998;15:145-47.
23. Lundorff P, Hahlin M, Kallfelt B, Thorburn J, Lindblom B. Adhesion formation after laparoscopic surgery in tubal pregnancy: a randomized trial versus laparotomy. Fertil Steril. 1991;55:911-15.
24. Bulletti CPVNVGEaFC. Adhesion formation after laparoscopic myomectomy. J.Am.Assoc.Gynecol.Laparosc. 1996;3:533-36.
25. Molinas CR, Mynbaev O, Pauwels A, Novak P, Koninckx PR. Peritoneal mesothelial hypoxia during pneumoperitoneum is a cofactor in adhesion formation in a laparoscopic mouse model. Fertil Steril. 2001;76:560-67.
26. Molinas CR, Koninckx PR. Hypoxaemia induced by CO(2) or helium pneumoperitoneum is a co-factor in adhesion formation in rabbits. Hum.Reprod. 2000;15:1758-63.
27. Ellis H. The magnitude of adhesion-related problems. In: diZerega GS, editor. Peritoneal surgery. New York: Springer; 2000.
28. Parker MC WMvGHMBJJDJMDWSEH. Adhesions and colorectal surgery - call for action. Colorectal Dis. 2007;9:66-72.
29. Liakakos T, Thomakos N, Fine PM, Dervenis C, Young RL. Peritoneal adhesions: Etiology, pathophysiology, and clinical significance - Recent advances in prevention and management. Dig Surg. 2001;18:260-73.
30. Jeekel H. Cost implications of adhesions as highlighted in a European study. Eur.J.Surg.Suppl 1997;43-45.
31. Duffy DM, diZerega GS. Adhesion controversies: pelvic pain as a cause of adhesions, crystalloids in preventing them. J.Reprod.Med. 1996;41:19-26.
32. diZerega GS. Use of adhesion prevention barriers in ovarian surgery, tubalplasty, ectopic pregnancy, endometriosis, adhesiolysis, and myomectomy. Curr.Opin.Obstet.Gynecol. 1996;8:230-37.
33. Steege JF. Chronic pelvic pain: what constitutes evidence? J.Am.Assoc.Gynecol.Laparosc. 2000;7:443-46.
34. Menzies D, Ellis H. Intestinal obstruction from adhesions--how big is the problem? Ann.R.Coll.Surg.Engl. 1990;72:60-63.
135
35. Davey AK MPJ. Surgical adhesions: a timely update, a great challenge for the future. J Minim Invasive Gynecol 2007;14:15-22.
36. Monk BJ BMMFJ. Adhesions after extensive gynecologic surgery: Clinical significance, etiology, and prevention. Am.J.Obstet.Gynecol. 1994;170:1396-403.
37. Metwally M WALRVP. Fluid and pharmacological agents for adhesion prevention after gynaecological surgery. Cochrane Database Syst Rev 2006;CD001298.
38. Lower AM HREHOFBSCAM. The impact of adhesions on hospital readmissions over ten years after 8849 open gynaecological operations: an assessment from the Surgical and Clinical Adhesions Research Study. Br J Obstet Gynaecol. 2000;107:855-62.
39. Milingos S, Kallipolitis G, Loutradis D, Liapi A, Mavrommatis K, Drakakis P et al. Adhesions: laparoscopic surgery versus laparotomy. Ann N Y Acad Sci. 2000;900:272-85.
40. Gomel V.Salpingo-ovariolysis by laparoscopy in infertility.Fertil Steril.1983;40:607-11.
41. Becker JM; et al. Prevention of postoperative abdominal adhesions by a sodium hyaluronate-based bioresorbable membrane: a prospective, randomized, double-blind multicenter study. J.Am.Coll.Surg. 1996;183:297-306.
42. Risberg B. Adhesions: preventive strategies. Eur.J.Surg.Suppl 1997;32-39.
43. Nagata Y. et al. Peri-ovarian adhesions interfere with the diffusion of gonadotrophin into the follicular fluid. Hum.Reprod. 1998;13:2072-76.
44. Milingos SD. Et al. Laparoscopic treatment of hydrosalpinx: factors affecting pregnancy rate. J.Am.Assoc.Gynecol.Laparosc. 2000;7:355-61.
45. Vrijland WW JJVGHSDBHJ. Abdominal adhesions: intestinal obstruction, pain, and infertility. Surg.Endosc. 2003;17:1017-22.
46. Howard FM. Et al. Conscious pain mapping by laparoscopy in women with chronic pelvic pain. Obstet.Gynecol. 2000;96:934-39.
47. Swank DJ, et al. A prospective analysis of predictive factors on the results of laparoscopic adhesiolysis in patients with chronic abdominal pain. Surg.Laparosc.Endosc.Percutan.Tech. 2003;13:88-94.
48. Hammoud A, et al. Adhesions in patients with chronic pelvic pain: a role for adhesiolysis? Fertil.Steril. 2004;82:1483-91.
49. Ellis H, Moran BJ, Thompson JN, Parker MC, Wilson MS, Menzies D et al. Adhesion-related hospital readmissions after abdominal and pelvic surgery: a retrospective cohort study. Lancet 1999;353:1476-80.
50. Duron JJ, et al. Prevalence and mechanisms of small intestinal obstruction following laparoscopic abdominal surgery: a retrospective multicenter study. French Association for Surgical Research. Arch.Surg. 2000;135:208-12.
51. Tulandi T. Introduction--prevention of adhesion formation: the journey continues. Hum.Reprod.Update. 2001;7:545-46.
52. Schölin J, et al. Bowel obstruction after laparoscopic and open colon resection for cancer: Results of 5 years of follow-up in a randomized trial. Surg.Endosc. 2011; In Press.
136
53. Menzies D, Pascual MH, Walz MK, Duron JJ, Tonelli F, Crowe A et al. Use of icodextrin 4% solution in the prevention of adhesion formation following general surgery: from the multicentre ARIEL Registry. Ann R.Coll Surg Engl. 2006;88:375-82.
54. Diamond MP et al. Pelvic Adhesions. Surg.Technol.Int 1998;VII:273-83.
55. Beck DE, et al. Effect of previous surgery on abdominal opening time. Dis.Colon Rectum 2000;43:1749-53.
56. Coleman MG, et al. Impact of previous surgery on time taken for incision and division of adhesions during laparotomy. Dis.Colon Rectum 2000;43:1297-99.
57. Van Der Krabben AA, et al. Morbidity and mortality of inadvertent enterotomy during adhesiotomy. Br.J.Surg. 2000;87:467-71.
58. Ivarsson ML, et al. Cost of bowel obstruction resulting from adhesions. Eur.J.Surg. 1997;163:679-84.
59. Menzies D, et al. Small bowel obstruction due to postoperative adhesions: treatment patterns and associated costs in 110 hospital admissions. Ann.R.Coll.Surg.Engl. 2001;83:40-46.
60. diZerega GS. Biochemical events in peritoneal tissue repair. Eur.J.Surg.Suppl 1997;10-16.
61. diZerega GS. Peritoneum, peritoneal healing and adhesion formation. In: diZerega GS, editor. Peritoneal surgery. New York: Springer; 2000. p. 3-38.
62. Holmdahl L. The role of fibrinolysis in adhesion formation. Eur J Surg Suppl 1997;24-31.
63. Mutsaers SE. Mesothelial cells: their structure, function and role in serosal repair. Respirology. 2002;7:171-91.
64. Herrick SE, Mutsaers SE. The potential of mesothelial cells in tissue engineering and regenerative medicine applications. Int.J.Artif.Organs 2007;30:527-40.
65. diZerega GS, Campeau JD. Peritoneal repair and post-surgical adhesion formation. Hum.Reprod.Update. 2001;7:547-55.
66. Falk K, Bjorquist P, Stromqvist M, Holmdahl L. Reduction of experimental adhesion formation by inhibition of plasminogen activator inhibitor type 1. Br.J.Surg. 2001;88:286-89.
67. Hellebrekers BW, Trimbos-Kemper TC, Trimbos JB, Emeis JJ, Kooistra T. Use of fibrinolytic agents in the prevention of postoperative adhesion formation. Fertil.Steril. 2000;74:203-12.
68. Molinas CR, Binda MM, Carmeliet P, Koninckx PR. Role of vascular endothelial growth factor receptor 1 in basal adhesion formation and in carbon dioxide pneumoperitoneum-enhanced adhesion formation after laparoscopic surgery in mice. Fertil Steril 2004;82 Suppl 3:1149-53.
69. Herrick SE, Mutsaers SE. Mesothelial progenitor cells and their potential in tissue engineering. Int.J.Biochem.Cell Biol. 2004;36:621-42.
70. Mynbaev OA, Molinas CR, Adamyan LV, Vanacker B, Koninckx PR. Pathogenesis of CO2 pneumoperitoneum-induced metabolic hypoxemia in a rabbit model. J.Am.Assoc.Gynecol.Laparosc. 2002;9:306-14.
137
71. Mynbaev OA, Molinas CR, Adamyan LV, Vanacker B, Koninckx PR. Reduction of CO(2)-pneumoperitoneum-induced metabolic hypoxaemia by the addition of small amounts of O(2) to the CO(2) in a rabbit ventilated model. A preliminary study. Hum.Reprod. 2002;17:1623-29.
72. Sitter T, et al. Imbalance between intraperitoneal coagulation and fibrinolysis during peritonitis of CAPD patients: the role of mesothelial cells. Nephrol Dial Transplant. 1995;10:677-83.
73. Volz J, Koster S, Spacek Z, Paweletz N. Characteristic alterations of the peritoneum after carbon dioxide pneumoperitoneum. Surg.Endosc. 1999;13:611-14.
74. Hazebroek EJ, Schreve MA, Visser P, De Bruin RW, Marquet RL, Bonjer HJ. Impact of temperature and humidity of carbon dioxide pneumoperitoneum on body temperature and peritoneal morphology. J.Laparoendosc.Adv.Surg.Tech.A 2002;12:355-64.
75. Ordonez JL, Dominguez J, Evrard V, Koninckx PR. The effect of training and duration of surgery on adhesion formation in the rabbit model. Hum.Reprod. 1997;12:2654-57.
76. Yesildaglar N, Koninckx PR. Adhesion formation in intubated rabbits increases with high insufflation pressure during endoscopic surgery. Hum.Reprod. 2000;15:687-91.
77. Verguts, J. Can prevention of mesothelial damage during endoscopic surgery decrease adhesion formation and tumor implantation? Katholieke Universiteit Leuven. Thesis 2011.
78. Molinas CR, Campo R, Elkelani OA, Binda MM, Carmeliet P, Koninckx PR. Role of hypoxia inducible factors 1alpha and 2alpha in basal adhesion formation and in carbon dioxide pneumoperitoneum-enhanced adhesion formation after laparoscopic surgery in transgenic mice. Fertil.Steril. 2003;80 Suppl 2:795-802.
79. Binda MM, Molinas CR, Koninckx PR. Reactive oxygen species and adhesion formation: clinical implications in adhesion prevention. Hum.Reprod. 2003;18:2503-07.
80. Glantzounis GK, Tselepis AD, Tambaki AP, Trikalinos TA, Manataki AD, Galaris DA et al. Laparoscopic surgery-induced changes in oxidative stress markers in human plasma. Surg.Endosc. 2001;15:1315-19.
81. Souza AMB, Rogers MS, Wang CC, Yuen PM, Ng PS. Comparison of peritoneal oxidative stress during laparoscopy and laparotomy. J.Am.Assoc.Gynecol.Laparosc. 2003;10:65-74.
82. Binda MM, Molinas CR, Hansen P, Koninckx PR. Effect of desiccation and temperature during laparoscopy on adhesion formation in mice. Fertil.Steril. 2006;86:166-75.
83. Kappas AM, Fatouros M, Papadimitriou K, Katsouyannopoulos V, Cassioumis D. Effect of intraperitoneal saline irrigation at different temperatures on adhesion formation. Br.J.Surg. 1988;75:854-56.
84. Mynbaev OA, Adamyan LV, Mailova K, Vanacker B, Koninckx PR. Effects of adding small amounts of oxygen to a carbon dioxide-pneumoperitoneum of increasing pressure in rabbit ventilation models. Fertil.Steril. 2008.
85. Brown CB, Luciano AA, Martin D, Peers E, Scrimgeour A, DiZerega GS. Adept (icodextrin 4% solution) reduces adhesions after laparoscopic surgery for adhesiolysis: a double-blind, randomized, controlled study. Fertil Steril 2007;88:1413-26.
138
86. DiZerega GS, Verco SJ, Young P, Kettel M, Kobak W, Martin D et al. A randomized, controlled pilot study of the safety and efficacy of 4% icodextrin solution in the reduction of adhesions following laparoscopic gynaecological surgery. Hum Reprod 2002;17:1031-38.
87. Metwally M, Watson A, Lilford R, Vandekerckhove P. Fluid and pharmacological agents for adhesion prevention after gynaecological surgery. Cochrane.Database.Syst.Rev. 2006;CD001298.
88. Zeng Q, Yu Z, You J, Zhang Q. Efficacy and safety of Seprafilm for preventing postoperative abdominal adhesion: systematic review and meta-analysis. World J Surg 2007;31:2125-31.
89. Beck DE. Seprafilm Review and Meta-analysis. World J Surg 2008.
90. Mohri Y, Kusunoki M. Efficacy and Safety of Seprafilm: Systematic Review and Meta-Analysis. World J Surg 2008.
91. DiZerega GS. Use of adhesion prevention barriers in ovarian surgery, tubalplasty, ectopic pregnancy, endometriosis, adhesiolysis, and myomectomy. Curr.Opin.Obstet Gynecol 1996;8:230-37.
92. Larsson B. Efficacy of Interceed in adhesion prevention in gynecologic surgery: a review of 13 clinical studies. J Reprod Med. 1996;41:27-34.
93. Baakdah H, Tulandi T. Adhesion in gynecology complication, cost, and prevention: a review. Surg Technol Int. 2005;14:185-90.
94. Mais V, Bracco GL, Litta P, Gargiulo T, Melis GB. Reduction of postoperative adhesions with an auto-crosslinked hyaluronan gel in gynaecological laparoscopic surgery: a blinded, controlled, randomized, multicentre study. Hum Reprod 2006; 21:1248-54
95. DiZerega GS, Coad J, Donnez J. Clinical evaluation of endometriosis and differential response to surgical therapy with and without application of Oxiplex/AP* adhesion barrier gel. Fertil Steril 2007;87:485-89.
96. Lundorff P, Donnez J, Korell M, Audebert AJ, Block K, DiZerega GS. Clinical evaluation of a viscoelastic gel for reduction of adhesions following gynaecological surgery by laparoscopy in Europe. Hum Reprod 2005;20:514-20.
97. Binda MM, Molinas CR, Bastidas A, Jansen M, Koninckx PR. Efficacy of barriers and hypoxia-inducible factor inhibitors to prevent CO(2) pneumoperitoneum-enhanced adhesions in a laparoscopic mouse model. J Minim Invasive Gynecol 2007;14:591-99.
98. Binda MM, Molinas CR, Bastidas A, Koninckx PR. Effect of reactive oxygen species scavengers, antiinflammatory drugs, and calcium-channel blockers on carbon dioxide pneumoperitoneum-enhanced adhesions in a laparoscopic mouse model. Surg Endosc 2007;21:1826-34.
99. Binda, M. M. Pathophysiology and prevention of adhesion formation in a laparoscopic mouse model. Leuven University press, thesis 2008.
100. Elkelani OA, Molinas CR, Mynbaev O, Koninckx PR. Prevention of adhesions with crystalloids during laparoscopic surgery in mice. J.Am.Assoc.Gynecol.Laparosc. 2002;9:447-52.
139
101. Saltzman AK, Olson TA, Mohanraj D, Carson LF, Ramakrishnan S. Prevention of postoperative adhesions by an antibody to vascular permeability factor/vascular endothelial growth factor in a murine model. Am J Obstet Gynecol. 1996;174:1502-06.
102. Saed GM, Jiang Z, Fletcher NM, Diamond MP. Modulation of the BCL-2/BAX ratio by interferon-gamma and hypoxia in human peritoneal and adhesion fibroblasts. Fertil Steril 2007.
103. Alpay Z, Ozgonenel MS, Savasan S, Buck S, Saed GM, Diamond MP. Possible role of natural immune response against altered fibroblasts in the development of post-operative adhesions. Am J Reprod Immunol. 2006;55:420-27.
104. Saed GM, Zhao M, Diamond MP, Abu-Soud HM. Regulation of inducible nitric oxide synthase in post-operative adhesions. Hum Reprod 2006;21:1605-11.
105. Chiang SC, Cheng CH, Moulton KS, Kasznica JM, Moulton SL. TNP-470 inhibits intraabdominal adhesion formation. J Pediatr.Surg 2000;35:189-96.
106. Di Paolo N, Sacchi G, Vanni L, Corazzi S, Terrana B, Rossi P et al. Autologous peritoneal mesothelial cell implant in rabbits and peritoneal dialysis patients. Nephron 1991;57:323-31.
107. Di Paolo N, Vanni L, Sacchi G. Autologous implant of peritoneal mesothelium in rabbits and man. Clin Nephrol. 1990;34:179-84.
108. Di Paolo N, Sacchi G, Del Vecchio MT, Nicolai GA, Brardi S, Garosi G. State of the art on autologous mesothelial transplant in animals and humans. Int.J Artif.Organs 2007;30:456-76.
109. Crabtree JH, Fishman A, Huen IT. Videolaparoscopic peritoneal dialysis catheter implant and rescue procedures under local anesthesia with nitrous oxide pneumoperitoneum. Adv.Perit.Dial. 1998;14:83-86.
110. Minoli G, Terruzzi V, Spinzi GC, Benvenuti C, Rossini A. The Influence of Carbon-Dioxide and Nitrous-Oxide on Pain During Laparoscopy - A Double-Blind, Controlled Trial. Gastrointest Endosc. 1982;28:173-75.
111. Tsereteli Z, Terry ML, Bowers SP, Spivak H, Archer SB, Galloway KD et al. Prospective randomized clinical trial comparing nitrous oxide and carbon dioxide pneumoperitoneum for laparoscopic surgery. J Am Coll Surg. 2002;195:173-79.
112. Aitola P, Airo I, Matikainen M. Laparoscopic versus open preperitoneal inguinal hernia repair: A prospective randomised trial. Ann Chir Gynaecol. 1998;87:22-25.
113. Rammohan A, Manimaran AB, Manohar RR, Naidu RM. Nitrous oxide for pneumoperitoneum: no laughing matter this! A prospective single blind case controlled study. Int.J.Surg 2011;9:173-76.
114. el Minawi MF, Wahbi O, El Bagouri IS, Sharawi M, El Mallah SY. Physiologic changes during CO2 and N2O pneumoperitoneum in diagnostic laparoscopy. A comparative study. J.Reprod.Med 1981;26:338-46.
115. Ooka T, Kawano Y, Kosaka Y, Tanaka A. [Blood gas changes during laparoscopic cholecystectomy--comparative study of N2O pneumoperitoneum and CO2 pneumoperitoneum]. Masui 1993;42:398-401.
140
116. Rademaker BM, Odoom JA, de Wit LT, Kalkman CJ, ten Brink SA, Ringers J. Haemodynamic effects of pneumoperitoneum for laparoscopic surgery: a comparison of CO2 with N2O insufflation. Eur.J.Anaesthesiol. 1994;11:301-06.
117. Hunter JG, Staheli J, Oddsdottir M, Trus T. Nitrous-Oxide Pneumoperitoneum Revisited - Is There A Risk of Combustion. Surg Endosc 1995;9:501-04.
118. Wong YT, Shah PC, Birkett DH, Brams DM. Peritoneal pH during laparoscopy is dependent on ambient gas environment - Helium and nitrous oxide do not cause peritoneal acidosis. Surg Endosc 2005;19:60-64.
119. Wong YT, Shah PC, Birkett DH, Brams DM. Carbon dioxide pneumoperitoneum causes severe peritoneal acidosis, unaltered by heating, humidification, or bicarbonate in a porcine model. Surg Endosc 2004;18:1498-503.
120. El Kady AA, Abd-El-Razek M. Intraperitoneal explosion during female sterilization by laparoscopic electrocoagulation. A case report. Int.J.Gynaecol.Obstet. 1976;14:487-88.
121. Gunatilake DE. Case report: fatal intraperitoneal explosion during electrocoagulation via laparoscopy. Int.J.Gynaecol.Obstet. 1978;15:353-57.
122. Mouton WG, Bessell JR, Otten KT, Maddern GJ. Pain after laparoscopy. Surg.Endosc. 1999;13:445-48.
123. Cervero F, Laird JMA. Visceral pain. Lancet 1999;353:2145-48.
124. Muzii L, Basile S, Zupi E, Marconi D, Zullo MA, Manci N et al. Laparoscopic-assisted vaginal hysterectomy versus minilaparotomy hysterectomy: A prospective, randomized, multicenter study. J Minim Invasive Gynecol.2007;14:610-15.
125. Mcmahon AJ, Russell IT, Ramsay G, Sunderland G, Baxter JN, Anderson JR et al. Laparoscopic and Minilaparotomy Cholecystectomy - A Randomized Trial Comparing Postoperative Pain and Pulmonary-Function. Surgery 1994;115:533-39.
126. McCloy R, Randall D, Schug SA, Kehlet H, Simanski C, Bonnet F et al. Is smaller necessarily better? A systematic review comparing the effects of minilaparoscopic and conventional laparoscopic cholecystectomy on patient outcomes. Surg Endosc. 2008;22:2541-53.
127. Tolhurst Cleaver CL, Hopkins AD, Kee Kwong KC, Raftery AT. The effect of postoperative peritoneal lavage on survival, peritoneal wound healing and adhesion formation following fecal peritonitis: an experimental study in the rat. Br.J.Surg 1974;61:601-04.
128. Vittimberga FJ, Foley DP, Meyers WC, Callery MP. Laparoscopic surgery and the systemic immune response. Ann Surg. 1998;227:326-34.
129. Wiseman DM. Animal adhesion models: design, variables, and relevance. diZerega G. Peritoneal Surgery. New York: Springer-Verlag; 2000. p. 459-76.
130. Mynbaev OA, Corona R. Possible mechanisms of peritoneal tissue-oxygen tension changes during CO2-pneumoperitoneum: the role of design, methodology and animal models. Hum.Reprod. 2009;24:1242-46.
131. Elkelani OA, Binda MM, Molinas CR, Koninckx PR. Effect of adding more than 3% oxygen to carbon dioxide pneumoperitoneum on adhesion formation in a laparoscopic mouse model. Fertil.Steril. 2004;82:1616-22.
141
132. Molinas CR, Tjwa M, Vanacker B, Binda MM, Elkelani O, Koninckx PR. Role of CO(2) pneumoperitoneum-induced acidosis in CO(2) pneumoperitoneum-enhanced adhesion formation in mice. Fertil.Steril. 2004;81:708-11.
133. Binda MM, Molinas CR, Mailova K, Koninckx PR. Effect of temperature upon adhesion formation in a laparoscopic mouse model. Hum.Reprod. 2004.
134. Molinas CR, Binda MM, Campo R, Koninckx PR. Adhesion formation and interanimal variability in a laparoscopic mouse model varies with strains. Fertil.Steril. 2005;83:1871-74.
135. Holmdahl L, al Jabreen M, Risberg B. Experimental models for quantitative studies on adhesion formation in rats and rabbits. Eur.Surg.Res. 1994;26:248-56.
136. Shaw R, Festing MF, Peers I, Furlong L. Use of factorial designs to optimize animal experiments and reduce animal use. ILAR.J. 2002;43:223-32.
137. Schonman R, Corona R, Bastidas A, De Cicco C, Koninckx PR. Effect of Upper Abdomen Tissue Manipulation on Adhesion Formation between Injured Areas in a Laparoscopic Mouse Model. J Minim Invasive Gynecol. 2009;16:307-12.
138. Betram P, Tietze L, Hoopmann M, Treutner KH, Mittermayer C,Schumpelick V (1999) Intraperitoneal transplantation of isologous mesothelial cells for prevention of adhesions. Eur J Surg 165:705–709
139. Foley-Comer AJ, Herrick SE, Al Mishlab T, Prele CM, Laurent GJ, Mutsaers SE. Evidence for incorporation of free-floating mesothelial cells as a mechanism of serosal healing. 2002;115:1383-89.
140. Guvenal T, Cetin A, Ozdemir H, Yanar O, Kaya T. Prevention of postoperative adhesion formation in rat uterine horn model by nimesulide: a selective COX-2 inhibitor. Hum Reprod 2001;16:1732–5.
141. Siegler AM, Kontopoulos V, Wang CF. Prevention of postoperative adhesions in rabbits with ibuprofen, a nonsteroidal anti-inflammatory agent. Fertil Steril 1980;34:46–9.
142. Luciano AA, Hauser KS, Benda J. Evaluation of commonly used adjuvants in the prevention of postoperative adhesions. Am J Obstet Gynecol 1983;146:88–92.
143. Aldemir M, Ozturk H, Erten C, Buyukbayram H. The preventive effect of Rofecoxib in postoperative intraperitoneal adhesions. Acta Chir Belg 2004;104:97–100.
144. Celebioglu B, Eslambouli NR, Olcay E, Atakan S. The effect of tenoxicam on intraperitoneal adhesions and prostaglandin E2 levels in mice. Anesth Analg 1999;88:939–42.
145. Golan A, Bernstein T, Wexler S, Neuman M, Bukovsky I, David MP. The effect of prostaglandins and aspirin—an inhibitor of prostaglandin synthesis— on adhesion formation in rats. Hum Reprod 1991;6:251–4.
146. Tayyar M, Basbug M. The effects of intraperitoneal piroxicam and low molecular weight heparin in prevention of adhesion reformation in rat uterine horn. Res Exp Med (Berl) 1999;198:269–75.
142
147. Nishimura K, Nakamura RM, diZerega GS. Biochemical evaluation of postsurgical wound repair: prevention of intraperitoneal adhesion formation with ibuprofen. J Surg Res 1983;34:219–26.
148. Nishimura K, Nakamura RM, diZerega GS. Ibuprofen inhibition of postsurgical adhesion formation: a time and dose response biochemical evaluation in rabbits. J Surg Res 1984;36:115–24.
149. Rodgers K, Girgis W, diZerega GS, Bracken K, Richer L. Inhibition of postsurgical adhesions by liposomes containing nonsteroidal anti-inflammatory drugs. Int J Fertil 1990;35:315–20.
150. Greene AK, Alwayn IP, Nose V, Flynn E, Sampson D, Zurakowski D, et al. Prevention of intra-abdominal adhesions using the antiangiogenic COX-2 inhibitor celecoxib. Ann Surg 2005;242:140–6.
151. Krenn V, Morawietz L, Burmester GR, Kinne RW, Mueller-Ladner U, Muller B, et al. Synovitis score: discrimination between chronic low-grade and highgrade synovitis. Histopath 2006;49:358–64.
152. Krenn V, Morawietz L, Haupl T, Neidel J, Petersen I, Konig A. Grading of chronic synovitis—a histopathological grading system for molecular and diagnostic pathology. Pathol Res Pract 2002;198:317–25.
153. Krenn V, Morawietz L. Synovitis score: proposal for a histopathological grading system of chronic synovitis. Ann Rheum Dis 2004;63:9.
154. Haywood L, McWilliams DF, Pearson CI, Gill SE, GanesanA,WilsonD, et al. Inflammation and angiogenesis in osteoarthritis. Arthritis Rheum 2003;48:2173–7.
155. Pessler F, Dai L, Diaz-Torne C, Gomez-Vaquero C, Paessler ME, Zheng DH, et al. The synovitis of ‘‘non-inflammatory’’ orthopaedic arthropathies: a quantitative histological and immunohistochemical analysis. Ann Rheum Dis 2008;67:1184–7.
156. Kumar V, Abbas A, Fausto N. Acute and chronic inflammation. In: Kumar V, Abbas A, Fausto N, editors. Robbins and Cotran’s pathologic basis of disease. 7th ed. Philadelphia: Saunders; 2005 p. 82–3.
157. Beckstead JH. A simple technique for preservation of fixation-sensitive antigens in paraffin-embedded tissues: addendum. J Histochem Cytochem 1995;43:345.
158. Robinson JS, Thompson JM, Wood AW. Letter: Laparoscopy explosion hazards with nitrous oxide. Br.Med J. 1975;3:764-65.
159. Menes T, Spivak H. Laparoscopy: searching for the proper insufflation gas. Surg.Endosc. 2000;14:1050-56.
160. The Practice Committee of the American Society for Reproductive Medicine. Pathogenesis, consequences, and control of peritoneal adhesions in gynecologic surgery. Fertil.Steril. 2008;90:44-49.
161. Ryan GB, Grobety J, Majno G. Mesothelial injury and recovery. Am.J.Pathol. 1973;71:93-112.
143
162. Nkere UU, Whawell SA, Sarraf CE, Schofield JB, Thompson JN, Taylor KM. Perioperative histologic and ultrastructural changes in the pericardium and adhesions. Ann.Thorac.Surg 1994;58:437-44.
163. Svenarud P, Persson M, van der LJ. Effect of CO2 insufflation on the number and behavior of air microemboli in open-heart surgery: a randomized clinical trial. Circulation 2004;109:1127-32.
164. Persson M, van der LJ. Intraoperative field flooding with warm humidified CO2 may help to prevent adhesion formation after open surgery. Med Hypotheses 2009;73:521-23.
165. Arnesjo B, Breland U, Petersson BG. The effect of peritoneal lavage on the postoperative course after colonic anastomosis and perforation in the rat. Acta Chir Scand. 1975;141:433-436.
166. Sortini D, Feo CV, Maravegias K et al. Role of peritoneal lavage in adhesion formation and survival rate in rats: an experimental study. J Invest Surg. 2006;19:291-297.
167. Torek F. The treatment of of diffuse suppurative peritonitis following appendicitis. Med Rec. 1906;70:849-858.
168. Platell C, Papadimitriou JM, Hall JC. The influence of lavage on peritonitis. J Am Coll Surg. 2000;191:672-680.
169. Ohno Y, Furui J, Kanematsu T. Treatment strategy when using intraoperative peritoneal lavage for perforated appendicitis in children: a preliminary report. Pediatr Surg Int. 2004;20:534-537.
170. Adam U, Ledwon D, Hopt UT. [Programmed lavage as a basic principle in therapy of diffuse peritonitis]. Langenbecks Arch Chir. 1997;382:S18-S21.
171. Hall JC, Heel KA, Papadimitriou JM et al. The pathobiology of peritonitis. Gastroenterology. 1998;114:185-196.
172. Lally KP, Trettin JC, Torma MJ. Adjunctive antibiotic lavage in experimental peritonitis. Surg Gynecol Obstet. 1983;156:605-608.
173. Polubinska A, Winckiewicz M, Staniszewski R et al. Time to reconsider saline as the ideal rinsing solution during abdominal surgery. Am J Surg. 2006;192:281-285.
174. Molinas CR, Binda MM, Mailova K, Koninckx PR. The rabbit nephrectomy model for training in laparoscopic surgery. Hum.Reprod. 2004;19:185-90.
175. Molinas CR, Cabral CR, Koninckx PR. Effect of the diameter of the endoscope and of surgeon training on the duration and quality of laparoscopic surgery in a rabbit model. J.Am.Assoc.Gynecol.Laparosc. 1999;6:447-52.
176. Aggarwal R, Tully A, Grantcharov T, Larsen CR, Miskry T, Farthing A et al. Virtual reality simulation training can improve technical skills during laparoscopic salpingectomy for ectopic pregnancy. Br J Obstet Gynecol. 2006;113:1382-87.
177. McClusky DA, Gallagher AG, Ritter EM, Lederman AB, Van Sickle KR, Baghai M et al. Virtual reality training improves junior residents' operating room performance: Results of a prospective, randomized, double-blinded study of the complete laparoscopic cholecystectomy. J Am Coll Surg. 2004;199:S73.
144
178. Van Sickle KR, Ritter EM, Baghai M, Goldenberg AE, Huang IP, Gallagher AG et al. Prospective, Randomized, Double-Blind Trial of Curriculum-Based Training for Intracorporeal Suturing and Knot Tying. J Am Coll Surg.2008;207:560-68.
179. Binda MM, Koninckx PR. Prevention of adhesion formation in a laparoscopic mouse model should combine local treatment with peritoneal cavity conditioning. Hum Reprod 2009;24:1473-79.
180. Gadallah MF, Torres-Rivera C, Ramdeen G, Myrick S, Habashi S, Andrews G. Relationship between intraperitoneal bleeding, adhesions, and peritoneal dialysis catheter failure: a method of prevention. Adv.Perit.Dial. 2001;17:127-29.
181. Gamal EM, Metzger P, Szabo GY, Brath E, Peto K, Olah A et al. The influence of intraoperative complications on adhesion formation during laparoscopic and conventional cholecystectomy in an animal model. Surg Endosc 2001;15:873-77.
182. Cömert M et al. Does intraabdominal use of Ankaferd Blood Stopper cause increased intraperitoneal adhesions? Ulus Travma Acil Cerrahi Derg. 2010;16:383-89.
183. Wiseman DM, Kamp L, Linsky CB, Jochen RF, Pang RHL, Scholz PM. Fibrinolytic Drugs Prevent Pericardial Adhesions in the Rabbit. J Surg Res. 1992;53:362-68.
184. Wiseman DM, Gottlick LE, Diamond MP. Effect of Thrombin-Induced Hemostasis on the Efficacy of An Absorbable Adhesion Barrier. J Reprod Med.1992;37:766-70.
185. Gupta N, Dadhwal V, Deka D, Jain SK, Mittal S. Corpus luteum hemorrhage: Rare complication of congenital and acquired coagulation abnormalities. J Obstet Gynaecol Res 2007;33:376-80.
186. Teng SW, Tseng JY, Chang CK, Li CT, Chen YJ, Wang PH. Comparison of laparoscopy and laparotomy in managing hemodynamically stable patients with ruptured corpus luteum with hemoperitoneum. J.Am.Assoc.Gynecol Laparosc. 2003;10:474-77.
187. Baba Y, Uetsuka K, Nakayama H, Doi K. Rat strain differences in the early stage of porcine-serum-induced hepatic fibrosis. 2004;55:325-30.
188. Ma B, Blackburn MR, Lee CG, Homer RJ, Liu W, Flavell RA et al. Adenosine metabolism and murine strain-specific IL-4-induced inflammation, emphysema, and fibrosis. J Clinic Invest. 2006;116:1274-83.
189. Skwarchuk MW, Travis EL. Murine strain differences in the volume effect and incidence of radiation-induced colorectal obstruction. Int J Radiat Oncol Biol Phys. 1998;41:889-95 .
190. Caldwell RL, Opalenik SR, Davidson JM, Caprioli RM, Nanney LB. Tissue profiling MALDI mass spectrometry reveals prominent calcium-binding proteins in the proteome of regenerative MRL mouse wounds. Wound Repair Regen. 2008;16:442-49.
191. Osadchii O, Norton G, Deftereos D, Woodiwiss A. Rat strain-related differences in myocardial adrenergic tone and the impact on cardiac fibrosis, adrenergic responsiveness and myocardial structure and function. Pharmacol Res 2007;55:287-94.
192. Mitchell CA, Grounds MD, Papadimitriou JM. The Genotype of Bone-Marrow-Derived Inflammatory Cells Does Not Account for Differences in Skeletal-Muscle Regeneration Between Sjl/J and Balb/C Mice. Cell Tissue Res. 1995;280:407-13.
145
193. McClearn GE, Hofer SM. Genes as gerontological variables: uniform genotypes. Neurobiol Aging 1999;20:95-104.
194. Homanics GE, Le NQ, Kist F, Mihalek R, Hart AR, Quinlan JJ. Ethanol tolerance and withdrawal responses in GABA(A) receptor alpha 6 subunit null allele mice and in inbred C57BL/6J and strain 129/SvJ mice. Alcoholism. Clinic Exp Res 1998;22:259-65.
195. Green V. Adrenal Hormones and Ethanol Ingestion in C57Bl-Crgl and C3H-Crgl-2 Mice. Behav Neural Biol. 1981;31:56-66.
196. Janan T.Eppig Goran Levan. Guidelines for Nomenclature of Mouse and Rat Strains . 1-9-2010. International Committee on Standardized Genetic Nomenclature for Mice. Ref Type: Data File
197. Staats J. Standardized Nomenclature for Inbred Strains of Mice: eighth listing. Cancer Res. 1985;45:945-77.
198. Davisson MT. Rules and guidelines for genetic nomenclature in mice: excerpted version. Committee on Standardized Genetic Nomenclature for Mice. Transgenic Res. 1997;6:309-19.
199. Potter M. History of the BALB/c family. Curr.Top.Microbiol.Immunol. 1985;122:1-5.
200. Janis Kuby. In: WH Freeman and company, editor. Immunology. USA: 1997. p. 223-46.
201. Crabbe JC, Wahlsten D, Dudek BC. Genetics of mouse behavior: interactions with laboratory environment. Science 1999;284:1670-72.
202. Crawley JN, Belknap JK, Collins A, Crabbe JC, Frankel W, Henderson N et al. Behavioral phenotypes of inbred mouse strains: implications and recommendations for molecular studies. Psychopharmacol (Berl) 1997;132:107-24.
203. Stiedl O, Radulovic J, Lohmann R, Birkenfeld K, Palve M, Kammermeier J et al. Strain and substrain differences in context- and tone-dependent fear conditioning of inbred mice. Behav.Brain Res. 1999;104:1-12.
204. Simpson EM, Linder CC, Sargent EE, Davisson MT, Mobraaten LE, Sharp JJ. Genetic variation among 129 substrains and its importance for targeted mutagenesis in mice. Nat.Genet. 1997;16:19-27.
205. Glant TT, Bardos T, Vermes C, Chandrasekaran R, Valdez JC, Otto JM et al. Variations in susceptibility to proteoglycan-induced arthritis and spondylitis among C3H substrains of mice: evidence of genetically acquired resistance to autoimmune disease. Arthritis Rheum. 2001;44:682-92.
206. Hilgers J, Arends J. A series of recombinant inbred strains between the BALB/cHeA and STS/A mouse strains. Curr.Top.Microbiol.Immunol. 1985;122:31-37.
207. Ohta H, Sakaide Y, Wakayama T. Age- and substrain-dependent sperm abnormalities in BALB/c mice and functional assessment of abnormal sperm by ICSI. Hum.Reprod. 2009;24:775-81.
208. Fall. The Importance of Understanding Substrains in the Genomic Age JAX® . 2003.
209. Specht CG, Schoepfer R. Deletion of the alpha-synuclein locus in a subpopulation of C57BL/6J inbred mice. BMC Neurosci. 2001;2:11.
146
210. Binnebosel M, Klink CD, Serno J, Jansen PL, von Trotha KT, Neumann UP et al. Chronological evaluation of inflammatory mediators during peritoneal adhesion formation using a rat model. Langenbecks Arch.Surg 2011;396:371-78.
211. Adams DH, Lloyd AR. Chemokines: Leucocyte recruitment and activation cytokines. Lancet 1997;349:490-95.
212. Murphy PM. The Molecular-Biology of Leukocyte Chemoattractant Receptors. Annu Rev Immunol 1994;12:593-633.
213. Bladh LG, Johansson-Haque K, Rafter I, Nilsson S, Okret S. Inhibition of extracellular signal-regulated kinase (ERK) signaling participates in repression of nuclear factor (NF)-kappa B activity by glucocorticoids. Biochim Biophys Acta 2009;1793:439-46.
214. Brunton, L. L., Lazo, J. S., and Parker, K. L. Goodman & Gilman's The pharmacological basic of therapeutics. 2006. United States of America, Mac Graw-Hill. Ref Type: Generic
215. Wang XX, Nelin LD, Kuhlman JR, Meng XM, Welty SE, Liu YS. The role of MAP kinase phosphatase-1 in the protective mechanism of dexamethasone against endotoxemia. Life Sci. 2008;83:671-80.
216. Binnebosel M, Klinge U, Rosch R, Junge K, Lynen-Jansen P, Schumpelick V. Morphology, quality, and composition in mature human peritoneal adhesions. Langenbecks Arch.Surg. 2008;393:59-66.
217. Nakano A, Tani E, Miyazaki K, Yamamoto Y, Furuyama JI. Matrix Metalloproteinases and Tissue Inhibitors of Metalloproteinases in Human Gliomas. J Neurosurg 1995;83:298-307.
218. Kripke BJ, Kupferman A, Luu KC. Suppression of chemotaxis to corneal inflammation by nitrous oxide. Zhonghua Min Guo Wei Sheng Wu Ji Mian Yi Xue Za Zhi. 1987 Nov;20(4):302-10.
219. Price NM, Farrar J, Chau TTH, Mai NTH, Hien TT, Friedland JS. Identification of a matrix-degrading phenotype in human tuberculosis in vitro and in vivo. J Immunol 2001;166:4223-30.
220. Rao JS. Molecular mechanisms of glioma invasiveness: The role of proteases. Nat Rev Cancer 2003;3:489-501.
221. Yubero S, Ramudo L, Manso MA, De Dios I. Mechanisms of dexamethasone-mediated chemokine down-regulation in mild and severe acute pancreatitis. Biochim Biophys Acta 2009;1792:1205-11.
222. Ruoslahti E. Fibronectin and its integrin receptors in cancer. Adv Cancer Res. 76 1999;76:1-20.
223. Lin YM, Jan HJ, Lee CC, Tao HY, Shih YL, Wei HW et al. Dexamethasone reduced invasiveness of human malignant glioblastoma cells through a MAPK phosphatase-1 (MKP-1) dependent mechanism. Eur.J.Pharmacol. 2008;593:1-9.
224. Corona R, Verguts J, Schonman R, Binda MM, Mailova K, Koninckx PR. Post-operative inflammation at the abdominal cavity increases adhesion formation in a laproscopic mouse model. Fertil Steril. 2011 Mar 15;95(4):1224-8.
225. Nyerges A. Pain Mechanisms in Laparoscopic Surgery. Semin Laparosc Surg 1994; Dec;1:215-18.
147
226. Jevtovic-Todorovic V, Todorovic SM, Mennerick S, Powell S, Dikranian K, Benshoff N et al. Nitrous oxide (laughing gas) is an NMDA antagonist, neuroprotectant and neurotoxin. Nat.Med 1998;4:460-63.
227. Gruss M, Bushell TJ, Bright DP, Lieb WR, Mathie A, Franks NP. Two-pore-domain K+ channels are a novel target for the anesthetic gases xenon, nitrous oxide, and cyclopropane. Mol.Pharmacol. 2004;65:443-52.
228. Aitola P, Airo I, Kaukinen S, Ylitalo P. Comparison of N2O and CO2 pneumoperitoneums during laparoscopic cholecystectomy with special reference to postoperative pain. Surg Laparosc Endosc. 1998;8:140-44.
229. Neuman GG, Sidebotham G, Negoianu E, Bernstein J, Kopman AF, Hicks RG et al. Laparoscopy explosion hazards with nitrous oxide. Anesthesiol1993;78:875-79.
230. Corall IM, Elias JA, Strunin L. Letter: Laparoscopy explosion hazards with nitrous oxide. Br.Med J. 1975;4:288.
231. Persson M, van der LJ. De-airing of a cardiothoracic wound cavity model with carbon dioxide: theory and comparison of a gas diffuser with conventional tubes. J.Cardiothorac.Vasc.Anesth. 2003;17:329-35.
232. Persson M, van der LJ. Can wound desiccation be averted during cardiac surgery? An experimental study. Anesth.Analg. 2005;100:315-20.
233. Bertram P, Tietze L, Hoopmann M, Treutner KH, Mittermayer C, Schumpelick V. Intraperitoneal transplantation of isologous mesothelial cells for prevention of adhesions. Eur J Surg 1999;165:705-09.
234. Trew G, Pistofidis G, Pados G, Lower A, Mettler L, Wallwiener D et al. Gynaecological endoscopic evaluation of 4% icodextrin solution: a European, multicentre, double-blind, randomized study of the efficacy and safety in the reduction of de novo adhesions after laparoscopic gynaecological surgery. Hum Reprod 2011;26:2015-27.
235. Koninckx PR, Ussia A, Adamyan L, Wattiez A. An endometriosis classification, designed to be validated. Gynecol Surg 2011;8:1-6.
148
149
Chapter 7. Career and bibliography of Roberta Corona
EDUCATION:
- 1-11-2009: Specialist in Obstetrics and Gynaecology (50/50 cum laude)
- 2004- 2009: Resident in Obstetrics and Gynaecology, University of Cagliari, Italy
- 1-10-2004: Graduated from University Medical School of Cagliari ( 110/110 cum laude)
- 1-07-1998: Maturità scientifica “Liceo scientifico statale “Pitagora”, Cagliari, Italy.
CLINICAL AND RESEARCH EXPERIENCE:
Principal interests in clinical and scientific activity:
• Minimally invasive surgery • Endometriosis • Adhesions • Reproductive medicine • Urogynaecology
- From January 2011 fellow in urogynaecology at the Dep. Obstetrics and Gynaecology UZ Leuven Gasthuisberg with Prof. J. Deprest (clinical and surgical approach and pelvic floor ultrasound).
- From July 2007 Clinical fellow in benign gynaecology with specific interest on pelvic surgery for endometriosis with Prof. P.R. Koninckx at Dep. Obstetrics and Gynaecology UZ Leuven Gasthuisberg.
INTERNATIONAL PUBLICATION:
1. R Corona, J Verguts, R Koninckx, K Mailova, PR Koninckx. Intra-peritoneal temperature and desiccation during endoscopic surgery. Am J Obstet Gynaecol. In press
2. Verguts J, Corona R, Binda MM, Koninckx PR. Concerns: Carbon dioxide pneumoperitoneum
prevents postoperative adhesion formation in a rat cecal abrasion model. J Laparoendosc Adv Surg Tech A. 2011 Jul-Aug;21(6):539-40.
3. R Corona, J Verguts, , MM Binda, R Schonman, CR Molinas, PR Koninckx. The impact of the learning curve upon adhesion formation in a laparoscopic mouse model. Fertil Steril 2011 Jul;96(1):193-7
150
4. C De Cicco, R Corona, R Schonman, A Ussia, PR Koninckx. Extensive abdominal lavage during laparoscopic discoid excision of deep endometriosis decreases bowel complications rate. Accepted for pubblication in Surgical Endoscopy
5. Manodoro S, Werbrouck E, Veldman J, Verguts J, Corona R, Van der Aa F, Coremans G, De Ridder D, Spelzini F, Deprest J. The laparoscopic approach to pelvic floor surger. Obst Gynec Forum May 2011;21:13-19.
6. R Corona, J Verguts, R Schonman, MM Binda, K Mailova, PR Koninckx. Post-operative
inflammation at the abdominal cavity increases adhesion formation in a laparoscopic mouse model. Fertil Steril. 2011 Mar 15;95(4):1224-8.
7. Mais V, Angioli R, Coccia E, Fagotti A, Landi S, Melis GB, Pellicano M, Scambia G, Zupi E, Angioni S, Arena S, Corona R, Fanfani F, Nappi C. Minerva Ginecol. 2011 Feb;63(1):47-70. Italian.
8. De Cicco C, Corona R, Schonman R, Mailova K, Ussia A, Koninckx P. Bowel resection for deep
endometriosis: a systematic review. BJOG. 2011 Feb;118(3):285-91
9. J. Verguts, A. Coosemans, R. Corona, M. Praet, K. Mailova, P. R. Koninckx. Intraperitoneal injection of cultured mesothelial cells decrease CO2 pneumoperitoneum enhanced adhesions in a laparoscopic mouse model. (DOI 10.1007/s10397-011-0658-8 early online) Gynecological Surgery, January 2010.
10. Binda MM, Corona R, Verguts J, Koninckx PR. Peritoneal Infusion with Cold Saline Decreased Postoperative Intra-abdominal Adhesion. World J Surg. 2010 Aug 31.
11. Koninckx PR, Corona R, de Cicco C, Verguts J, Ussia A. Pudendal neuralgia. J Minim Invasive Gynecol. 2010 Sep-Oct;17(5):666.
12. Mynbaev OA, Corona R. Possible mechanisms of peritoneal tissue-oxygen tension changes during CO2-pneumoperitoneum: the role of design, methodology and animal models.
Hum Reprod. 2009 Jun;24(6):1242-6.
13. Schonman R, Corona R, Bastidas A, De Cicco C, Koninckx PR. Effect of Upper Abdomen Tissue Manipulation on Adhesion Formation between Injured Areas in a Laparoscopic Mouse Model. J Minim Invasive Gynecol. 2009 May-Jun;16(3):307-12.
14. Schonman R, Corona R, Bastidas A, De Cicco C, Mailova K, Koninckx PR. Intercoat gel (oxiplex): efficacy, safety, and tissue response in a laparoscopic mouse model. J Minim Invasive Gynecol. 2009 Mar-Apr;16(2):188-94.
151
15. Schonman R, De Cicco C, Corona R, Soriano D, Koninckx PR. Accident analysis: factors contributing to a ureteric injury during deep endometriosis surgery. BJOG 2008 Dec;115(13):1611-5.
16. Koninckx PR, De Cicco C, Schonman R, Corona R, Betsas G, Ussia A. Endometriosis lesions that compromise the rectum deeper than the inner muscularis layer have more than 40% of the circumference of the rectum affected by the disease. J Minim Invasive Gynecol. 2008 Nov-Dec;15(6):774-5; author reply 775-6.
17. Ussia A, Betsas G, Corona R, De Cicco C, Koninckx PR. Pathophysiology of cyclic hemorrhagic ascites and endometriosis. J Minim Invasive Gynecol. 2008 Nov-Dec;15(6):677-81.
18. Corona R, De Cicco C, Schonman R, Verguts J, Ussia A, Koninckx PR. Tension-free Vaginal Tapes and Pelvic Nerve Neuropathy. Minim Invasive Gynecol. 2008 May-Jun;15(3):262-7.
BOOK CHAPTER:
• Co-author of the chapter 2: “Postoperative adhesions and their prevention” P.R. Koninckx, M.M. Binda, R. Corona, C.R. Molinas in the book: “Reconstructive and Reproductive Surgery in Gynecology” Victor Gomel, Andrew Brill (September 2010). Informa Healthcare, London, UK. ISBN-13:9780415419550.
Other publications (abstracts and national journals)
1. Stefano Manodoro, Jasper Verguts, Roberta Corona, Frank Van der Aa, Georges Coremans, Dirk De Ridder, Federico Spelzini, Jan Deprest. Laparoscopic sacrocolpopexy. SAJOG 2011; 17 (2):45.
2. R. Corona, R. Schonman, M.M. Binda, K. Maylova, J. Verguts, P.R. Koninckx. The role of acute inflammation at the peritoneal cavity-enhanced adhesion formation. Gynecol Surg (2010) 7 (Suppl 1):S1–S30
3. Angioni S, Corona R, Arena I, Melis GB, Diagnosi di endometriosi profonda. Il Ginecologo vol. 4 No. 1, Marzo 2009.
4. C. De Cicco, R. Corona, R. Schonman, K. Mailova, A Ussia, PR Koninckx “Bowel resection for endometriosis: outcome and complications”, Gynaecological Surgery, Supplement 1, October 2009
152
5. C. De Cicco, G. Betsas, A. Ussia, R. Corona, R. Schonman, P.R. Koninckx. “Prophylactic suturing of the ureter following extensive ureterolysis during deep endometriosis laparoscopic excision”, Gynaecological Surgery, Supplement 1, October 2009
6. R. Corona, C. De Cicco, R. Schonman, J. Verguts, A.Ussia, Gb. Melis, PR. Koninckx. Tension-free vaginal tapes and pelvic nerve neuropathy. Gynaecological Surgery, Supplement 1, October 2009
7. R. Schonman, C. De Cicco, R. Corona, D. Soriano, P.R. Koninckx “Iatrogenic ureter injury during deep endometriosis surgery: analysis of factors influencing judgment and mistakes” Gynaecological Surgery, Supplement 1, October 2008
8. C. De Cicco, R. Schonman, R. Corona, B. Van Kleynenbreugel, P.R. Koninckx “Ureteral lesions following laparoscopic excision of deep endometriosis” Gynaecological Surgery, Supplement 1, October 2008
9. C. De Cicco, R. Schonman, R. Corona, B. Van Kleynenbreugel, P.R. Koninckx “Extensive abdominal lavage decreases the risk of bowel perforation following discoid excision of deep endometriosis” Gynaecological Surgery, Supplement 1, October 2008
10. R. Schonman, A. Bastidas, R. Corona, C. De Cicco, P.R. Koninckx “Evaluation of Oxiplex\SP Gel (Intercoat) in our laparoscopic mouse model”. Gynaecological Surgery, Supplement 1, October 2008
11. R. Schonman, R. Corona, A. Bastidas, C. De Cicco, P.R. Koninckx “Tissue manipulation increases adhesion formation in a laparoscopic mouse model” Gynaecological Surgery, Supplement 1, October 2008.
153
Addenda
154
Surgeon’s Corner www.AJOG.org
AQ: 1
Intraperitoneal temperature and desiccationduring endoscopic surgeryIntraoperative humidification and cooling of theperitoneal cavity can reduce adhesions
Roberta Corona, MD; Jasper Verguts, MD, PhD; Robert Koninckx, PhD; Karina Mailova, MD;Maria Mercedes Binda, PhD; Philippe R. Koninckx, MD, PhDmorkdiga
1.45acticde
onsesolat
erath
wedandheenan
n-d-ais-s-t,-s-dls-
is-s
hs--
e,i-
vennd.
nd
, a
h.B.,
AQ: 3
AQ: 4AQ: 5
Problem: adhesion formationDuring endoscopic surgery a pneuperitoneum is needed to create a woing space.1 For safety reasons, carbonoxide (CO2) is used as an insufflationbecause it is highly soluble in water (mg/L), and it has a high exchange capity in the lungs.2,3 This routine prachas been identified as a culprit in thevelopment of postoperative adhesias a consequence of the associated mthelial hypoxia and desiccation. Theter results from the flow rate, tempture, and relative humidity (RH) ofgas.
Trauma to the peritoneum folloby a local inflammatory reactionmesothelial healing can lead to adsions. The acute inflammation of thetire peritoneal cavity, however, is qu
From the Department of Obstetrics andGynecology, Katholieke Universiteit Leu(Drs Corona, Verguts, Mailova, Binda, aP.R. Koninckx), and eSaturnus NV (Dr RKoninckx), Leuven, Belgium, and theDepartment of Reproductive Medicine aSurgery, Moscow State University ofMedicine and Dentistry, Moscow, Russia(Dr Mailova).
Karl Storz GmbH & Co.KG (Tuttlingen,Germany) provided a humidifier andendoscopic equipment; Fisher and PaykelHealthcare Ltd (Auckland, New Zealand)provided a humidifier; and eSaturnus NV(Leuven, Belgium) provided equipment tomeasure relative humidity and temperaturemodified humidifier, and an instrument forcooling.
R.K. is an employee of eSaturnus NV, whicprovided equipment. R.C., J.V., K.M., M.Mand P.R.K. report no conflict of interest.
0002-9378/$36.00© 2011 Mosby, Inc. All rights reserved.doi: 10.1016/j.ajog.2011.06.091
In press
---s
-e-,---e
---
titatively the most important factor iadhesion formation.4 This acute inflammation results from the balance of bafactors such as mesothelial hypoxia associated with CO2 pneumoperitoneum,mesothelial hyperoxia when the pO2
�40 mm Hg, as occurs during open surgery (if air is roughly 20% oxygen, it haa pO2 of about 150 mm Hg), and desiccation of the mesothelium.5-7 The lasdesiccation, has 2 opposing effects on adhesion formation–it directly damagecells by dehydrating them, but it also decreases the temperature of the affectearea, and this is advantageous, since celare more resistant to injury, such as hypoxia, at lower temperatures.7,8 For threason, the damaging effect of desiccation has been difficult to isolate–and habeen underestimated.8
The degree of desiccation varies witflow rate, RH, and temperature of the gaused for the pneumoperitoneum, subsequently altering intraperitoneal and mesothelial temperature, mesothelial damagperitoneal acute inflammation, and ult
This study was conducted to document quadesiccation during laparoscopic surgery. Thwere measured in vitro and during laparoscthe abdomen. This permitted us to calculatedry CO2 or CO2 humidified with 100% relativ25 and 37°C. The study showed that desexpected with the flow rates and relative humainly with desiccation. Temperature reguintraperitoneal temperature without affecthumidifier, desiccation could be eliminatedature between 31 to 32°C.
Key words: core body temperature, desicclaparoscopy, relative humidity
Cite this article as: Corona R, Verguts J, Koninckxduring endoscopic surgery. Am J Obstet Gyneco
MONTH 2011 Ame
mately postoperative adhesion formationand pain. Over the last decade, humidifi-cation and temperature of the gas usedduring endoscopic surgery has received in-creasing attention.3,9 For example, it isclear that in a peritoneal cavity at 98.6°F(37°C) with a 100% RH, nonhumidifiedgas at 37°C will cause desiccation and somerelated cooling. Desiccation also obviouslyincreases with the flow rate of the deliveredgas.10 The relationship among desiccation,flow rate, and cooling, however, is morecomplex. First, the maximum amount ofwater a gas can hold increases linearlywith temperature, and these maximumamounts are equivalent to 100% RH. At25°C, 100% RH equals 25 mg/L whereas at37°C this is 44 mg/L. Second, when flowrate increases desiccation increases, butwhen flow rate is too high, equilibration nolonger occurs and the outgoing RH drops.Thirdly, when desiccation causes coolingthis results in a cooling of the gas (thus thegas can hold less water) while simultane-ously the cooling of tissues will slow downthe rate of desiccation. Use of warm hu-
atively the intraperitoneal temperature andmperature, relative humidity, and flow rate
c surgery, at the entrance and at the exit ofsiccation for various flow rates using eitherumidity at any preset temperature betweenation, both in vitro and in vivo, varies asity while intraperitoneal temperature variesion of bowels is specific and drops to thecore body temperature. With a modifiedile maintaining the intraperitoneal temper-
n, intraperitoneal temperature,
t al. Intraperitoneal temperature and desiccation11;205:x.ex-x.ex.
ntite teopidee hicc
midlat
ingwh
atio
R, el 20
rican Journal of Obstetrics & Gynecology 1.e1
ingh
n-ra-gyterf aofn-n-calto
nlyst,a-dyre,m-easmeu-d.si-ndn,ofreo-ti-
rmbere-e-
easd-
-
.-
Surgeon’s Corner www.AJOG.org
AQ: 6
AQ: 7
AQ: 17
AQ: 8
AQ: 9
AQ: 10
AQ: 11
AQ: 12
F1
midified gas has also been said to resultdecreased postoperative pain,9 althouthis remains controversial.11,12
To understand cooling, the relatioship among humidification, tempeture, enthalpy of a gas, and the enerrequirements for evaporation of wamust be considered. While heating ohumidified gas requires the amountenergy needed to heat the water cotained in that gas– by definition, the eergy to heat 1 mL of water by 1°C is 1(4.1858 joules)–the energy requiredheat 1 mL of dry gas by 1°C is o0.00003 cal (0.0001 joules). In contra577 cal (2415.2 joules) are needed to vporize 1 mL of water at normal botemperature of 98.6°F (37°C). Therefopractically, desiccation is the only iportant factor causing cooling, whercold humidified gas could cause socooling, and the temperature of nonhmidified gas can almost be disregarde
Desiccation and cooling are intrincally related and have detrimental afavorable effects on adhesion formatiorespectively. The quantitative effectdesiccation and peritoneal temperatuon mesothelial damage during endscopic surgery has not yet been invesgated in detail. Indeed, the use of wahumidified gas has been suggested tosuperior to dry and cold gas but itmains controversial11,12 whether it dcreases postoperative pain, wherthere still is no evidence of decreased a
FIGURE 1
In vitro system was used to validate accusurements: heated chamber, maintained atubing, while box with water bags at 98Protruding moistened towels, acting as “dment and at its conclusion.Corona. Intraperitoneal temperature and desiccation du
hesion formation. The relationship be-
1.e2 American Journal of Obstetrics & Gynecology
tween peritoneal damage caused bywarm (37°C) and humidified gas and ahigh temperature and by cold and drygas causing desiccation and coolingcould be biphasic, with an optimumachieved with a little desiccation to-gether with a little cooling. Experimentsin mice, moreover, indicate that ideally,to reduce adhesions, prevention of des-iccation should be combined with aslightly lower intraperitoneal tempera-ture of 87.8-89.6°F (31-32°C), a temper-ature that achieves �80% of the coolingeffect upon adhesion formation com-pared to gas at 25°C.7 So when an insuf-flation gas with close to 100% RH is used,a third means of cooling is required toachieve the total effect on adhesion for-mation of cooling.
Our solutionWe performed several studies in prepa-ration for a randomized controlled trial(RCT) that will investigate the effects offlow rate, temperature (T), and RH ofthe insufflation gas combined with ex-ternal cooling and thus desiccation andintraperitoneal temperature upon painand adhesions in human beings. For allthe studies we used the Thermoflator(Karl Storz GmbH & Co.KG, Tuttlingen,Germany) for insufflation. For humidi-fication of the insufflation gas, the model204320 33 humidifier (Karl Storz GmbH& Co.KG), the MR860 humidifier(Fisher and Paykel Healthcare Ltd,
of temperature and relative humidity (RH) mea.6°F (37°C), contained insufflator, humidifier, andF (37°C) served as model of peritoneal cavityccation recipient,” were weighted before experi
endoscopic surgery. Am J Obstet Gynecol 2011.
Auckland, New Zealand), and a modi-
MONTH 2011
fied humidifier (Fisher and PaykelHealthcare Ltd) were used.
The Storz humidifier blows gas overwater warmed to 98.6°F (37°C) and usesa noninsulated tubing, 2.7 m in lengthand 7 mm in diameter (Kendall; Covi-dien, Mansfield, MA). Adequate humid-ification is provided at low flow rates butthe tubing causes rapid cooling of the gasto ambient temperature. The standardF&P humidifier blows gas through aheated water chamber. To avoid conden-sation, the tubing that delivers the gaswas heated to �104°F (40°C) with aheating wire. To permit higher flow ratesthan the device usually delivers, the luerlock was removed. In the modified F&Phumidifier, modified by eSaturnus NV(Leuven, Belgium), the heating of thechamber and tubing was continuouslyadapted by an electronic feedback loopto maintain 100% RH at the end of thetubing, a preset temperature between 77-96.8°F (25-36°C) at flow rates between0.5-30 L/min.
The flow rate, RH, and temperature ofthe gas at the end of the insufflation tub-ing and the gas flowing from the perito-neal cavity– or a box used to mimic theperitoneal cavity in the in vitro experi-ments–were measured twice a second;temperature and RH were captured witha digital sensor (SHT75; Sensirion AG,Zurich, Switzerland). This permittedcalculation of water loss, a marker fordesiccation, in real time. Cooling by athird means of the peritoneal cavity wasaccomplished by nebulizing 3 mL/min ofwater at room temperature or at 32°F(0°C) with a nozzle set at 2 bar entrypressure. This cooling/nebulization/hu-midification device had a diameter of �5mm so that it could be used through astandard 5-mm trocar.
To validate the setup in vitro, the bot-tom and side walls of a closed polysty-rene box were covered with plastic bagscontaining a total of 6 L of water at 98.6°F(37°C) (Figure 1). It was assumed thattemperature and RH coming out of thebox would reflect temperature and RHinside the box. When the tubing wasused for nonhumidified CO2 or for gashumidified with the Storz humidifier,the gas was permitted to cool to room
racyt 98.6°esi
ring
temperature or was heated to 98.6°F
a
-er
”
--/-
e0e-
s-
y-
-s-,,,,r
t-e-
t0t---0ae-
e
eyzf
e-
-
t,
rs-
t
e
f
e-e
-
-tt
-e
-t
r
eg
TABL
E1
Valid
atio
nof
mea
sure
men
tsof
flow
rate
,rel
ativ
ehu
mid
ity,a
ndte
mpe
ratu
rea
Varia
ble
Stor
zhu
mid
ifier
bF&
Phu
mid
ifier
cNo
hum
idifi
er
Flow
rate
,L/m
in2.
55
1020
2.5
510
202.
55
1020
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
Mea
sure
dw
ater
loss
,g1.
91�
0.12
7.20
�0.
169.
73�
0.13
11.8
4�
0.2
1.72
�0.
085.
31�
0.10
7.60
�0.
2510
.43
�0.
263.
11�
0.11
9.31
�0.
1511
.93
�0.
2714
.72
�0.
31
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
Calc
ulat
edw
ater
loss
,g2.
22�
0.14
8.11
�0.
0911
.21
�0.
1212
.92
�0.
222.
01�
0.07
6.14
�0.
159.
32�
0.23
12.6
1�
0.30
4.22
�0.
1410
.10
�2.
1113
.41
�0.
2616
.20
�0.
33
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
Data
are
expr
esse
das
mea
nan
dSD
.Diff
eren
ces
wer
eca
lcul
ated
with
Wilc
oxon
/Kru
skal
-Wal
lisun
paire
dsa
mpl
este
stus
ing
Grap
hPad
Pris
m5
softw
are.
aM
easu
rem
ents
wer
em
ade
atea
chflo
wra
teov
er30
-min
perio
d;b
Mod
el20
4320
33hu
mid
ifier
(Kar
lSto
rzGm
bH&
Co.K
G);c
MR8
60hu
mid
ifier
(Fis
hera
ndPa
ykel
Heal
thca
reLt
d,Au
ckla
nd,N
ewZe
alan
d).
Cor
ona.
Intr
aper
iton
ealt
empe
ratu
rean
dde
sicc
atio
ndu
ring
endo
scop
icsu
rger
y.A
mJO
bste
tGyn
ecol
2011
.
www.AJOG.org Surgeon’s Corner
AQ: 13
T1
AQ: 14
F2
(37°C) by putting the tubing insideheated chamber.
First, the accuracy of the measurements of flow rate, RH, and temperaturof the gas were validated. The exact wateloss was measured by the difference inweight of the “desiccation recipient,protruding moistened towels, before andafter the experiment, and compared withthe water loss calculated from the flowrate, RH, and temperature of the inflowing and outflowing gas with the following formula: humidity (g/m3) � RH (%)100 � (3.1243 10e-4 � T3 � 8.1847 � 10e3 � T2 � 0.32321 � T � 5.018). RHs andtemperatures at inflow and outflowequilibrated within 2 minutes, so wused the mean values during the last 1minutes of the experiment to calculatwater loss. Since the accuracy of desiccation calculation was crucial for the invivo experiments, the measurementwere validated over a wide range of conditions and performed in triplicate.
The measurements of flow rate, RH, andtemperature were accurate, as judged bthe linear relationship between the calculated and measured water loss (MatLabsoftware, MathWorks ™, Natick, Massachusetts, U.S.A.) (Table 1). Measurementof temperature and RH, made twice a second, had a coefficient of variation of 0.8%0.9%, 0.5%, and 0.5%, and of 2.2%, 3%12.4%, and 22.9% at flow rates of 2.5, 5, 10and 20 L/min, respectively. Thereforemean values over 5 minutes were used foall further calculations.
In the second experiment, the impacof desiccation, temperature of the inflowing gas, and humidification on thtemperature and RH in the box was measured. To do this, dry gas or humidifiedgas was used at room temperature or a98.6°F (37°C) at flow rates from 2.5-2L/min. We confirmed, as expected, thathe effect of desiccation was the most important factor for cooling. The gas temperature rapidly equilibrated with the ambient temperature at flow rates up to 2L/min. Thus, when the tubing was inchamber at 98.6°F (37°C), the temperaturremained stable; otherwise the temperatures were similar to room temperatures.
When dry CO2 at room temperaturwas used for insufflation, desiccation
and a drop in temperature occurred, in-dicating that the water loss/min and thheat loss/min increased almost linearlwith flow rate (Figure 2). With the Storhumidifier, temperature-in at the end othe tubing was, as predicted, at roomtemperature, while the RH decreasedwith flow rates. The near 100% RH at lowflow rates is explained by condensationin the tubing due to cooling. With thFisher and Paykel Healthcare Ltd humidifier, when the gas was heated toavoid condensation, the RHs at the endof the tubing were adequate for any flowrate. However, the temperature increased with the flow rate: it was 99 �0.68°F (37.2 � 0.4°C), 101.8 � 1.1°F(38.8 � 0.6°C), 104.4 � 2.1°F (40.2 �1.2°C), and 107 � 3.3°F (41.7 � 1.8°C) aflow rates of 2.5, 5, 10, and 20 L/minrespectively. It should be noted that inthese experiments, we removed the luelock on the original tubing, which preventflow rates �7.8 L/min at 15 mm Hg insufflation pressure. Because of the increase intemperature, the modified Fisher andPaykel Healthcare Ltd humidifier was nofurther evaluated in vivo.
The third experiment evaluated thmodified Fisher and Paykel Healthcare Ltdhumidifier and its ability to deliver 100%RH at the end of the tubing regardless othe preset temperature for flow rates upto 30 L/min. Temperature and RH at thend of the tubing were measured for different preset temperatures to determinaccuracy and fluctuations over time andthe rapidity of response when the flowrates were changed. The third experiment confirmed that with the modifiedFisher and Paykel Healthcare Ltd humidifier, the temperature was constanwithin 32.4°F (0.2°C) for any flow rate apreset temperatures between 82.4-93.2°F(28-34°C) with an RH �90% for any preset temperature. When flow rates wersuddenly increased or decreased, the newequilibrium was reached within 2-3 minutes. The temperature-out and RH-ouwere as expected, with a slight decrease intemperature due to desiccation at higheflow rates.
We also examined the value of thcooling device and the effect of coolinon desiccation with a constant deliveredgas temperature of 89.6°F (32°C) and an
RH of 100%. Spraying 3-4 mL of nebu-MONTH 2011 Americ
an Journal of Obstetrics & Gynecology 1.e3
stcfaCaww
Surgeon’s Corner www.AJOG.org
T2
F3
lized water at 32°F (0°C) decreased thetemperature-out by an average of 35.6 �0.9°F (2 � 0.5°C) within seconds for flowrates up to 10 L/min. This cooling de-creased, as expected, the calculated des-iccation, and at lower flow rates, somecondensation occurred in the box.
For our in vivo study, we used a previ-ously described endoscopic surgery setup,which consisted of an umbilical metal tro-car (Karl Storz GmbH & Co.KG) with a7-mm side opening and 3 secondary5-mm disposable trocars.13 When atraight laparoscope was used, insuffla-ion was performed through the umbili-al trocar while aspiration was donerom a secondary trocar. When an oper-tive laparoscope (Karl Storz GmbH &o.KG) with a large 7-mm side openingnd a CO2 laser were used, insufflationas done through the laparoscope,hich prevented blooming of the CO2
laser beam, and aspiration was per-formed at the umbilical trocar. For in-sufflation, the intraperitoneal pressurewas maintained at the preset 15 mm Hgfor flow rates up to 30 L/min.14 The out-flow of gas was regulated by opening and
FIGURE 2
Temperature and relative humidity (RH) in and2.5-15 L/min. These were measured while usinroom temperature, and modified Fisher and Payk(32°C). Similar observations were made in vitroCorona. Intraperitoneal temperature and desiccation during
closing a 3-way valve.
1.e4 American Journal of Obstetrics & Gynecology
All women included in this study un-derwent surgery for deep endometriosisusing the standard setup. After about 15minutes of surgery, different flow rates(2.5, 5, 7.5, 10, and 15 L/min) were eval-uated. Subsequently, continuous flowrates between 7-10 L/min were used toremove smoke, as is standard for ourCO2 laser operations. During the entireprocedure, flow rates, temperature-in,RH-in, temperature-out, and RH-outwere measured; water-in, water-out, anddesiccation were calculated twice a sec-ond. Nonhumidified CO2 was used in 3procedures, humidified CO2 via theStorz humidifier was used in 4 surgeries,and humidified CO2 provided by themodified F&P humidifier was used inanother 4. At least 3 operations were re-corded, and at the end of surgery, the ef-fect of the cooling device on intraperito-neal temperature was assessed. Corebody temperature, via esophageal tem-perature, was continuously monitoredby the anesthesiologist.
These investigations were conductedas pilot experiments for a RCT designedto demonstrate that, as in the mouse
during laparoscopic surgery with flow rates ofry gas at room temperature, Storz humidifier atealthcare Ltd (F&P) humidifier preset at 89.6°F
ert).scopic surgery. Am J Obstet Gynecol 2011.
model, cooling and humidification of
MONTH 2011
the peritoneal cavity could have benefi-cial effects on adhesion formation andpostoperative pain in human beings (in-stitutional review board approval hasbeen granted for the RCT, and registra-tion through clinicaltrial.gov was per-formed: NCT01344486) Surgery in-volved no unusual risks since all theinstruments used were standard. Theonly difference was that the relevant pa-rameters were measured continuously toevaluate desiccation and temperature inthe abdomen. The cooling device posedno risk, since administering saline, 2-4mL/min, was less intensive than the nor-mal practice of saline irrigation at roomtemperature when as much as 8 L mightbe used. The modified F&P humidi-fier maintained temperature at 89.6°F(32°C), an advantage over devices thatallow temperatures to fluctuate between98.6°F (37°C) (in the absence of flow)and 77°F (25°C) (when high flow with-out humidification was used).
The results for temperature-in and– out, and RH-in and – out, were strik-ingly similar in vivo and in vitro (Figure2 and Table 2). In both, all changes be-tween temperature-in and -out could beattributed entirely to desiccation and tothe temperature change of the water inthe gas. These data permitted calculationof the heat exchange between the waterbags or the body and the gas used forinsufflation. Heat loss never exceeded50-100 cal/min.
As in the in vitro experiments, weachieved a temperature decrease of about35.6°F (2°C) (Figure 3). When the spraywas directed to the upper abdomen, theendoscopic image was never distorted.Similarly, swift temperature changes ofabout 33.8°F (1°C) are observed duringprocedures such as electrosurgery andrinsing of the abdomen.
Insufflation through the operative lap-aroscope with outflow of gas through thecentral trocar resulted in smaller differ-ences between temperature-in and -outand RH-in and -out than when the out-flow occurred through a secondary port.Although differences in temperature atdifferent locations in the abdominal cav-ity were not measured directly, it seemsplausible that gas mainly flows directly
outg del H(insendo
from the insufflation port to the outflow
In press
sacmfltawleawwe
fscmotpc01h
mtiwtchc
tp0wt
rlcrrImpuc
cpr
C endo
ABLE
2at
erlo
ssw
hen
dry
carb
ondi
oxid
ew
arm
edto
room
tem
pera
ture
ente
rspe
riton
ealc
avity
a
aria
ble
Dry
CO2
Stor
zhu
mid
ifier
bM
odifi
edF&
Phu
mid
ifier
c
low
rate
,/m
in2.
55
7.5
1015
2.5
57.
510
152.
55
7.5
1015
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
...
alcu
late
dat
erlo
ss,
/min
0.15
1�
0.00
20.
120
�0.
005
0.17
1�
0.00
50.
201
�0.
006
0.30
3�
0.00
60.
011
�0.
003
0.02
3�
0.00
40.
031
�0.
006
0.05
3�
0.00
80.
070
�0.
009
0.00
021
�0.
0000
10.
0009
�0.
0002
0.00
10�
0.00
080.
0041
�0.
0012
0.00
90�
0.00
21
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
...
O 2,c
arbo
ndi
oxid
e.at
aar
eex
pres
sed
asm
ean
and
SD.
Diffe
renc
esin
desi
ccat
ion
caus
edby
dry
CO2
and
Stor
zhum
idifi
eran
dby
Stor
zhum
idifi
eran
dm
odifi
edFi
sher
and
Payk
elHe
alth
care
Ltd
(Auc
klan
d,Ne
wZe
alan
d)hu
mid
ifier
wer
esi
gnifi
cant
(P�
.01)
atan
yflo
wra
te;b
Mod
el20
4320
33hu
mid
ifier
(Kar
lSto
rzGm
bH&
Co.K
G,(T
uttli
ngen
,Ger
man
y));
cM
R860
hum
idifi
er(F
ishe
rand
Payk
elHe
alth
care
Ltd)
mod
ified
byeS
atur
nus
NV.
oron
a.In
trap
erit
onea
ltem
pera
ture
and
desi
ccat
ion
duri
ngen
dosc
opic
surg
ery.
Am
JObs
tetG
ynec
ol20
11.
www.AJOG.org Surgeon’s Corner
AQ: 15
F4
port. Temperature and RH in the ab-dominal cavity thus will not be homoge-neously distributed. Using a CO2 laseretup with insufflation through the lap-roscope and aspiration through theentral trocar produces a small compart-ent with relatively high flow rates; gas
ows from the tip of the laparoscope tohe central trocar. While these consider-tions do not affect the overall results,hen the compartment is small, even
ow flow rates can cause important localffects, including desiccation. Therefore,ny medication administered togetherith the insufflating gas or the nebulizedater will not homogeneously affect the
ntire peritoneal cavity.With flow rates of 10 L/min, as is used
or smoke evacuation during CO2 laserurgery, dry gas and the Storz humidifieraused constant desiccation of 0.3 and 0.05g/min, respectively, during the first hour
f surgery. After 70-90 minutes of surgeryhe temperature-out, reflecting the intra-eritoneal temperature, suddenly de-reased to a mean of 83.5 � 0.7°F (28.6 �.4°C) in patients given dry gas and 84.9 �°F (29.4 � 0.6°C) for patients givenumidified gas with Storz humidifier, re-
FIGURE 3
bservations during surgery: decrease in tempeerature increased with electrosurgical coagulatorona. Intraperitoneal temperature and desiccation during
inutes. Simultaneously, with decreasedemperature, dry gas desiccation was elim-nated. With the Storz humidifier, not onlyas desiccation abolished but condensa-
ion of almost all inflowing humidity oc-urred (Figure 4). With the modified F&Pumidifier, desiccation was minimal andondensation never occurred.
The core body temperature of the pa-ients, as measured by esophageal tem-erature, averaged 97.7 � 0.5°F (36.5 �.3°C), and remained unaffected in allomen despite constant intraperitoneal
emperatures �89.6°F (32°C).From the experiments comparing the
eal water loss, as measured by weightoss, and calculated water loss, we canonclude that the measurements of flowate, RH, and temperature were accu-ate, at least for flow rates up to 20 L/min.t should be stressed that these experi-
ents were designed to evaluate intra-eritoneal temperature and desiccationnder different conditions of humidifi-ation, flow rate, and external cooling.
With the Storz humidifier, RH de-reased rapidly with flow rate. The mainroblem was the cooling of the gas tooom temperature when noninsulated
re was noted with use of cooling device; tem-
scopic surgery. Am J Obstet Gynecol 2011.
O ratup ion.
ducing desiccation in all women within 5 tubing was used. Since at 77°F (25°C),
T W V F L ... C w g ... C D a C
MONTH 2011 American Journal of Obstetrics & Gynecology 1.e5In press
1ttctt4dtcn(n
cAplrTiaccWh
mctitstammsbe7e3s(
aatawmmv
fidwnptcTttcItrng
tbboursfiwpttdpchatabitp
Surgeon’s Corner www.AJOG.org
CO2 cannot hold �25 mg of water/L at00% RH, condensation occurred in theubing at low flow rates and humidifica-ion up to 5 L/min; entering a peritonealavity that is initially at 98.6°F (37°C),he gas temperature will increase (at thisemperature and 100% RH, the gas holds4 mg of water/L), resulting in peritonealesiccation. This combination of insuffla-
ion of gas at room temperature and desic-ation resulted in unexpectedly low perito-eal cavity temperatures, often �86°F30°C). With nonhumidified gas, perito-eal temperatures were even lower.The standard F&P humidifier prevented
ooling in the tubing using a heating wire.t higher flow rates however, the gas tem-erature can be �98.6°F (37°C). The luer
ock at the end of the tubing limits flowates to 7.8 L/min at 15 mm Hg pressure.herefore, when used according to operat-
ng instructions, this humidifier provideddequate humidification and a slight in-rease in temperature, which is probablyompensated by cooling in the trocar.
hen the luer lock is removed to permit
FIGURE 4
Desiccation and condensation is noted during suwere used as is usual during carbon-dioxide lasperature-out, reflecting bowel temperature, suddcaused desiccation to drop to 0 (3 patients). Wioccurred, and desiccation “inverted” to condenshumidifier (Fisher and Paykel Healthcare Ltd [F&mal. Moment that desiccation and temperature sand after that moment and mean and SD over 1Corona. Intraperitoneal temperature and desiccation during
igher flow rates to be used with the Ther-
1.e6 American Journal of Obstetrics & Gynecology
oflator for CO2 laser surgery, humidifi-ation remains adequate but the intraperi-oneal temperature could be slightlyncreased. However, it should be stressedhat with use of a CO2 laser setup and as-ociated higher flow rates, desiccation andemperature changes mainly occur within
small compartment. For adhesion for-ation, it was demonstrated in the mouseodel that desiccation is harmful while a
lightly lower mesothelial temperature iseneficial.7,8 Adhesions indeed decreasexponentially with temperatures of at least7°F (25°C), with �80% of this beneficialffect being achieved at 87.8-89.6°F (31-2°C). With higher temperatures, adhe-ions increase rapidly, and at �98.6°F37°C), the increase is dramatic.7,13
Clearly, neither the Storz nor the Fisherand Paykel Healthcare Ltd humidifier cancompletely prevent adhesion formation.The former resulted in desiccation, and thelatter caused rather high intraperitonealtemperatures of 98.6°F (37°C). For both,effects on desiccation and temperature in-creased with flow rates. We, therefore,
ry lasting �1.5 hours; flow rates of 8-10 L/minsurgery. When dry gas was administered, tem-
decreased after 70-80 minutes of surgery; thistorz humidifier (3 patients), same phenomenonn of almost all inflowing humidity. With modifiedAuckland, New Zealand), desiccation was mini-ted to drop is indicated as 0; 60 minutes beforeinute period are indicated.scopic surgery. Am J Obstet Gynecol 2011.
modified the Fisher and Paykel Healthcare t
MONTH 2011
Ltd humidifier to deliver fully humidifiedgas at a preset value of 87.8-89.6°F (31-32°C), since most of the beneficial effectwas reached at that temperature. For clin-ical and biological reasons, we wanted toavoid temperatures �82.4°F (28°C). Tovoid heating of the gas upon entrance,nd thus desiccation, we anticipated thathe peritoneal cavity had to be cooled bynother means, such as a nozzle deliveringater at room temperature. These experi-ents were designed to estimate howuch cooling would be necessary to pre-
ent desiccation in human beings.To our surprise, when fully humidi-
ed gas at 89.6°F (32°C) was used, theesiccation and the heating of the gasas, within minutes, much less pro-ounced than expected. This can be ex-lained only by rapid vasoconstriction ofhe peritoneal surface, especially in theompartment with the higher gas flow.herefore, additional cooling to maintain
emperatures of 87.8-89.6°F (31-32°C) inhe peritoneal cavity–and avoiding desic-ation–was much easier than anticipated.ndeed, a little extra cooling with intermit-ent application of 2-3 mL of saline atoom temperature, combined with ordi-ary irrigation, prevented heating of theas in the peritoneal cavity and desiccation.
Our finding that the temperatures ofhe bowel and peritoneal cavity seem toe regulated differently from the coreody temperature was supported by thebservation that, after some 70-80 min-tes of surgery with insufflating gas atoom temperature and desiccation (aseen with dry gas or the Storz humidi-er), the temperature of the bowel andall suddenly dropped below room tem-erature without affecting core bodyemperature. We conclude, then, thatemperature regulation of this region isifferent from that of the core body tem-erature. Rather, it is similar to what oc-urs in the arms and legs, which diminisheat loss and preserve core body temper-ture through vasoconstriction. In addi-ion, a constant intraperitoneal temper-ture �89.6°F (32°C) did not affect coreody temperature, probably because of
mportant vasoconstriction of mesen-eric vessels. Vasoconstriction also ex-lains why the calculated heat loss from
rgeerenlyth SatioP],tar
0-mendo
he body never exceeded 100 cal/min and
In press
www.AJOG.org Surgeon’s Corner
why condensation never caused foggingof the optical instruments.
In conclusion, in the absence of humid-ification, the intraabdominal tempera-tures were surprisingly low, which wasmainly the result of desiccation. Even witha Storz humidifier, temperatures hardlyexceeded 82.4-86°F (28-30°C). Desicca-tion could be prevented completely when amodified Fisher and Paykel Healthcare Ltdhumidifier that maintains the peritonealcavity temperature at 87.8-89.6°F (31-32°C) is combined with additional cool-ing. Surprisingly, minimal cooling was suf-ficient, and this could be explained by theunexpected finding that, as with the limbs,temperature regulation of the bowel andabdominal wall appears to be differentfrom that of core body temperature. Evi-dently, rapid vasoconstriction preventsheat loss during surgery. f
ACKNOWLEDGMENTSWe would like to thank Bernard Van Acker, De-partment of Anesthesiology; Andre D’Hoore,Department of Surgery; and the nurses of theoperating room at Katholieke Universiteit Leu-ven for their cooperation. Philippe Billet, Chris-
tophe Lauwerys, and Steven Declerck (eSatur-nus NV), were of great help, as were MichaelBlackhurst (Fisher and Paykel Healthcare Ltd)and Thomas Koninckx (eSaturnus NV). Wewould also like to acknowledge the assistanceof Marleen Craessaerts and Diana Wolput.
REFERENCES1. Lau WY, Leow CK, Li AK. History of endo-scopic and laparoscopic surgery. World J Surg1997;21:444-53.2. Menes T, Spivak H. Laparoscopy: searchingfor the proper insufflation gas. Surg Endosc2000;14:1050-6.3. Neudecker J, Sauerland S, Neugebauer E, etal. The European Association for EndoscopicSurgery clinical practice guideline on the pneu-moperitoneum for laparoscopic surgery. SurgEndosc 2002;16:1121-43.4. Corona R, Verguts J, Schonman R, BindaMM, Mailova K, Koninckx PR. Postoperative in-flammation in the abdominal cavity increasesadhesion formation in a laparoscopic mousemodel. Fertil Steril 2011;95:1224-8.5. diZerega GS. Biochemical events in peritonealtissue repair. Eur J Surg Suppl 1997;577:10-6.6. diZerega GS, Campeau JD. Peritoneal repairand post-surgical adhesion formation. Hum Re-prod Update 2001;7:547-55.7. Binda MM, Molinas CR, Mailova K, KoninckxPR. Effect of temperature upon adhesion for-mation in a laparoscopic mouse model. Hum
Reprod 2004;19:2626-32. gMONTH 2011 Ame
In press
8. Binda MM, Molinas CR, Hansen P, KoninckxPR. Effect of desiccation and temperature dur-ing laparoscopy on adhesion formation in mice.Fertil Steril 2006;86:166-75.9. Sammour T, Kahokehr A, Hill AG. Meta-analy-sis of the effect of warm humidified insufflation onpain after laparoscopy. Br J Surg 2008;95:950-6.10. Yesildaglar N, Koninckx PR. Adhesion for-mation in intubated rabbits increases with highinsufflation pressure during endoscopic sur-gery. Hum Reprod 2000;15:687-91.11. Manwaring JM, Readman E, Maher PJ. Theeffect of heated humidified carbon dioxide onpostoperative pain, core temperature, and re-covery times in patients having laparoscopicsurgery: a randomized controlled trial. J MinimInvasive Gynecol 2008;15:161-5.12. Hamza MA, Schneider BE, White PF, et al.Heated and humidified insufflation during lapa-roscopic gastric bypass surgery: effect on tem-perature, postoperative pain, and recovery out-comes. J Laparoendosc Adv Surg Tech A2005;15:6-12.13. Koninckx PR, Meuleman C, Demeyere S,Lesaffre E, Cornillie FJ. Suggestive evidencethat pelvic endometriosis is a progressive dis-ease, whereas deeply infiltrating endometriosisis associated with pelvic pain. Fertil Steril1991;55:759-65.14. Koninckx PR, Vandermeersch E. The per-sufflator: an insufflation device for laparoscopyand especially for CO -laser-endoscopic sur-
2ery. Hum Reprod 1991;6:1288-90.
rican Journal of Obstetrics & Gynecology 1.e7
Received
publishe
R.C. has n
nothing
to disclo
Supported
care, N
Reprint req
of Obste
(E-mail:
0015-028doi:10.10
The impact of the learning curve on adhesionformation in a laparoscopic mouse modelRoberta Corona, M.D.,a Jasper Verguts, M.D.,a Maria Mercedes Binda, Ph.D.,a
Carlos Roger Molinas, M.D., Ph.D.,b Ron Schonman, M.D.,c and Philippe R. Koninckx, M.D., Ph.D.a
a Department of Obstetrics and Gynaecology, University Hospital Gasthuisberg, Katholieke Universiteit Leuven, Leuven,
Belgium; b Neolife-Medicina y Cirug�ıa Reproductiva, Centro M�edico La Costa, Asuncion, Paraguay; and c Department of
Obstetrics and Gynaecology, Sheba Medical Center, affiliated to Tel Aviv University, Tel Hashomer, Israel
Objective: To evaluate the impact of surgeon training on adhesion formation in a laparoscopic mouse model.Laparoscopic surgery and bowel manipulation was demonstrated to enhance postoperative adhesion formation.Design: Prospective randomized, controlled trial.Setting: University laboratory research center.Animal(s): 200 BALB/c and 200 Swiss female mice.Intervention(s): Adhesions were induced by opposing bipolar lesions and 60 minutes of pneumoperitoneum. Eachsurgeon operated on 80 mice (40 Swiss and 40 BALB/c), the only variable thus being his/her increasing experience.Some surgeons were already experienced gynecologists, others were starting their training.Main Outcome Measure(s): End points were the duration of surgery while performing the lesions. The adhesionformation was scored quantitatively (proportion and total) and qualitatively (extent, type, and tenacity) after 7 days.Result(s): With training, duration of surgery and adhesion formation decreased exponentially for all surgeons,whether experienced or not. Experienced surgeons had initially a shorter duration of surgery, less adhesion forma-tion, and less de novo adhesions than inexperienced surgeons.Conclusion(s): These data suggest that laparoscopic skills improve with training, leading to a decrease in the du-ration of surgery and formation of adhesions. Therefore completion of a standardized learning curve should beman-datory when initiating adhesion formation studies both in laboratory or clinical setting. (Fertil Steril� 2011;96:193–7. �2011 by American Society for Reproductive Medicine.)
Key Words: Learning curve, laparoscopy, adhesions, mouse model, training, surgery
Surgical training has always been an essential part in the residency ofthe young surgeon. Since the introduction of laparoscopic surgery,learning curves are frequently used to analyze the effect of trainingand to assess the competency of surgeons. This can be done, withoutany risk for the patient (1), thanks to the development of differentmethods of laparoscopic surgical training, such as live and cadaveranimal training, human cadaver training, box trainer, video trainer,and virtual reality training by computer simulation (2). Learningcurves are different for different type of surgery and surgeons canbe experienced in one procedure and not in another (3–5).
Learning curves in endoscopic surgery, as reported in the litera-ture, showed that they always improve outcomes in different surgi-cal procedures (2, 6–11) and after a rapid improvement in skills atthe beginning, which varies with the skills investigated (tying,suturing, cutting, dissecting, lifting, grasping, and transferringobjects with both hands) (12), a plateau is reached (i.e., knot tyingis learned much quicker than stitching) (13). Learning curves are re-flected in a progressively decreasing operating time, an enhanced
January 24, 2011; revised and accepted April 15, 2011;
d online May 24, 2011.
othing to disclose. J.V. has nothing to disclose. M.M.B. has
to disclose. C.R.M. has nothing to disclose. R.S. has nothing
se. P.R.K. has nothing to disclose.
by the Leuven Quality Surgery Fund (Fisher and Paykel Health-
ordic Pharma, and eSaturnus NV).
uests: Roberta Corona, M.D., U.Z. Gasthuisberg, Department
trics & Gynaecology, Herestraat 49, B3000 Leuven, Belgium
2/$36.0016/j.fertnstert.2011.04.057 Copyright ª2011 American S
precision, and a decrease in complications (6, 7, 14). Specifically,the duration of surgery (14–19) and complication rate (18, 19) issignificantly lower for experienced surgeons compared with juniorones (14), with an important impact on costs (20, 21).
The impact of learning curves on experiments involving animalmodels has rarely been investigated (14). This might be especiallyimportant in experiments investigating postoperative adhesion for-mation, as adhesion formation has been shown to decrease witha shorter duration of surgery (22) and with gentle tissue handling(23, 24). Good surgical techniques is widely believed to decreasepostoperative adhesions.
During the past years our group developed and validated a strictlystandardized laparoscopic mouse model for the study of postopera-tive adhesion formation and tumor implantation. To investigate theminimum experience required to perform experiments we prospec-tively investigated the learning curves and the impact of training onthe extent and the variability of postoperative adhesion formation, aswell as the effect on operating time.
MATERIALS AND METHODSThe Laparoscopic Mouse Model for Adhesion FormationThe experimental setup (i.e., animals, anesthesia and ventilation,laparoscopic surgery and induction, and scoring of peritoneal adhe-sions) has been described in detail previously (22–31). Briefly, themodel consisted of performing bipolar lesions during laparoscopyfollowed by 60 minutes of pure CO2 pneumoperitoneum. Thepneumoperitoneum was induced using the Thermoflator (Karl
Fertility and Sterility� Vol. 96, No. 1, July 2011 193ociety for Reproductive Medicine, Published by Elsevier Inc.
FIGURE 1
Duration of surgery during training. Learning curve expressed as
duration of the surgery versus number of consecutive proceduresfor experienced (n¼ 3) and nonexperienced surgeons (n¼ 2). Dots
represent the real operating times for each mouse. The lines were
calculated from the individual operating times after a two-phase
exponential decay model.
Corona. Techniques and instrumentation. Fertil Steril 2011.
Storz) through a 2-mm endoscope with a 3.3 external sheath forinsufflation (Karl Storz) introduced into the abdominal cavitythrough a midline incision caudal to the xyphoid appendix. Theincision was closed gas tight around the endoscope to avoidleakage. The insufflation pressure was 15 mm Hg and humidifiedgas (Humidifier 204320 33; Karl Storz) was used. After theestablishment of the pneumoperitoneum, two 14-gauge catheters(Insyte-W, Vialon; Becton Dickinson) were inserted underlaparoscopic vision. Standardized 10- � 1.6-mm lesions wereperformed in the antimesenteric border of both right and leftuterine horns and in both the right and left pelvic side walls withbipolar coagulation (20 W, standard coagulation mode, Autocon350; Karl Storz). Because anesthesia and ventilation influencebody temperature (28), the timing between anesthesia (T0), intuba-tion (at 10 minutes, T10), and the onset of the experiment (at 20min-utes, T20) was strictly controlled. After 7 days, adhesions werescored quantitatively (proportion and total) and qualitatively (extent,type, tenacity) blindly during laparotomy under a stereomicroscope.The entire abdominal cavity was visualized using a xyphopubicmidline and a bilateral subcostal incision. After the evaluation ofports sites and viscera (omentum, large and small bowels) for denovo adhesions, the fat tissue surrounding the uterus was carefullyremoved. The length of the visceral and parietal lesions and adhe-sions were measured. Adhesions, when present, were carefully lysedto evaluate their type and tenacity. The terminology of Pouly andSeak-San (32) was used, describing de novo adhesion formationfor the adhesions formed at nonsurgical sites and adhesion forma-tion for adhesions formed at the surgical site.
AnimalsThe present study was performed in 400, 12–13-week-old femalemice (i.e., 200 BALB/cJ@Rj mice; inbred strain) weighting 20–30g and 200 Swiss mice (outbred strain) weighting 30–40 g. Animalswere kept under standard laboratory conditions and diet at the ani-mal facilities of the Katholieke Universiteit Leuven. The studywas approved by the Institutional Review Animal Care Committeeof the Katholieke Universiteit Leuven.
Experimental Design and Surgical ProceduresThe experiment was designed to evaluate the effect of training, as-sessed by the consecutive surgeries, on operating time and on the ex-tent and variability of postoperative adhesion formation. Theoperating time was measured from T20 (i.e., from the beginningof surgery, exactly 20 minutes after induction of anesthesia) untilthe end of the procedure with the removal of catheters for instrumen-tation and port sites closure. In addition, the codes of interventionorder were broken only at the end of the training experiment.Each trainee performed sequentially 80 interventions to induce ad-hesion formations. All trainees were inexperienced for the laparo-scopic procedure in a mouse model, but some of them wereexperienced gynecologists after their training in gynecology (3trainees), whereas other surgeons were inexperienced at the end ofmedical school (2 trainees).
All experiments were performed using block randomization bydays. Therefore, a block of animals comprising one animal ofeach strain, one BALB/c mouse and one Swiss mouse were alwaysoperated in a single session on the same day to avoid day-to-day var-iability. In addition, within a block, experiments were performed inrandom order. Before surgical procedure initial body temperatureand weight were measured.
194 Corona et al. Techniques and instrumentation
StatisticsStatistical analyses were performed using GraphPad Prism version 4(GraphPad Software Inc.). Multifactorial analyses were achievedwith GLM and Logistic procedures using the SAS System (SAS In-stitute). The coefficient of variation (CV), a normalized measure ofdispersion, was calculated as the ratio of the standard deviation tothe mean multiplied by 100.
RESULTSWith training, assessed by consecutive surgeries, duration of surgerydecreased exponentially in both groups, demonstrating a clear learn-ing curve that can be described as a two-phase exponential decaymodel (nonexperienced: R2 ¼ 0.72; experienced: R2 ¼ 0.73)(Fig. 1). In BALB/c mice the real and calculated operating timesfor nonexperienced surgeons decreased from 19 � 6 minutes and22 � 4 minutes for the first 10 procedures to 7 � 1 minutes and7 � 1 minutes for the last 10 procedures (P¼.0001; P¼.0001,respectively). For experienced surgeons time decreased from 8 �1 minutes and 10 � 3 minutes for the first 10 procedures to 4 �0.3 minutes and 4 � 0.03 minutes for the last 10 procedures. InSwiss mice the real and calculated operating times from nonexper-ienced surgeons decreased from 17� 6 minutes and 22� 4 minutesfor the first 10 procedures to 7� 1 minutes and 7� 1 minutes for thelast 10 procedures (P¼.0001; P¼.0001, respectively). For experi-enced surgeons time decreased from 8� 3 minutes and 10� 3 min-utes for the first 10 procedures to 4 � 0.3 minutes and 5 � 0.03minutes for the last 10 procedures (P¼.0001; P¼.0001, respec-tively). As expected, the real and calculated operating times bothfor BALB/c and Swiss mice were not statistically different (t-test)both for the first 10 procedures and for the last 10 procedures.When the effects of training with time, the experience of the surgeonand the mouse strain were evaluated simultaneously using multifac-torial analysis (proc GLM), we confirmed the importance of de-creasing operating time (P¼.0001) and the effect of experience
Vol. 96, No. 1, July 2011
FIGURE 2
Duration of the surgery and adhesion formation during learning
curve in different mouse strain. Duration of the surgery andadhesion formation during learning curves in BALB/c (n¼ 200) and
Swiss mice (n ¼ 200). Number of procedures were grouped by
groups of 10 blocks (1 block¼ 2 procedures: 1 BALB/cþ 1 Swiss).
Mean and SD are indicated.
Corona. Techniques and instrumentation. Fertil Steril 2011.
(P¼.0001), whereas the mouse strain was not important as could beanticipated (P¼not significant [NS]) (Fig. 2).
Similarly, the effects of training (expressed by the consecutivenumber of surgeries), operating time, experience, and mouse strainon adhesion formation were evaluated using multifactorial analysis(proc GLM) (Figs. 2 and 3). Adhesion formation was lower with theconsecutive number of surgeries (proportion: P¼.01; total: P¼.01;extension: P¼.02; type: P¼.01; tenacity: P¼006), for surgeries ofshorter operating time (proportion: P¼.02; total: P¼.04; extension:P¼.02; type: P¼.02; tenacity: P¼.04), among experienced surgeons
FIGURE 3
Adhesion formation during learning curve. Adhesion formation
expressed as proportion of adhesions versus number ofconsecutive procedures in both experienced (n ¼ 3) and
nonexperienced surgeons (n ¼ 2). Mean and SD are indicated.
Corona. Techniques and instrumentation. Fertil Steril 2011.
Fertility and Sterility�
(proportion: P<.0001; total: P¼.04; extension: P¼.001; type:P¼.03; tenacity: P¼.04), and in Swiss mice (P¼NS) (Fig. 2).
It was decided arbitrarily to make four groups of 10 blocks (G1,block 1–10; G2, block 11–20; G3, block 21–30; G4, block 31–40).With this grouping, the interanimal variability for operating timeand for adhesion formation was assessed using the CV. The CV ofoperating time among nonexperienced surgeons were 41%, 35%,25%, and 20% for BALB/c mice and 39%, 40%, 22%, and 17%for Swiss mice for G1, G2, G3, and G4, respectively; among expe-rienced surgeons were 45%, 24%, 20%, and 22% for BALB/c miceand 37%, 24%, 20%, and 21% for Swiss mice for G1, G2, G3, andG4, respectively. The CVof the proportion of adhesions were 62%,48%, 54%, and 38% for BALB/c mice and 106%, 92%, 96%, and77% for Swiss mice for G1, G2, G3, and G4, respectively. The ef-fects of the consecutive number of surgeries, measured by groups,and of mouse strain on operating time and its CVand on the propor-tion of adhesions and its CV were evaluated with a multifactorialanalysis (proc GLM). Both operating time and its CV decreasedwith the number of surgeries (P¼.0001; P¼.0001, respectively),whereas they were not affected by mouse strain (P¼NS; P¼NS, re-spectively). Both the proportion of adhesions and its CV decreasedwith the number of surgeries (P¼.002; P¼NS) and were affected bymouse strain (i.e., for BALB/c mice the proportion of adhesions washigher and the CV was lower).
DISCUSSIONTo the best of our knowledge this is the first article aiming tostudy the effect of surgical training on postoperative adhesion for-mation in a laparoscopic mouse model. A previous study, performedin rabbits, showed that postoperative adhesion formation, durationof surgery, and complication rate decreased with surgeon training,expressed by the consecutive number of procedures performed(14, 33, 34).
In the present study, we compared two groups of surgeons (i.e.,experienced and nonexperienced surgeons). Experienced surgeonsnot only started with lower duration of surgery but also achievedthe plateau earlier (after 10 procedures), whereas nonexperiencedsurgeons started with longer duration of surgery and, although theduration decreased, it did not achieve the level of experienced sur-geons even after 80 consecutive procedures. A less traumatic,more precise and gentle surgical technique gained with experienceappears to be important as postoperative adhesion formation waslower already after the first 10 procedures within the group of expe-rienced surgeons in comparison with nonexperienced surgeons, con-firming the data of the effect of manipulation-enhanced adhesions(23). This difference among groups of surgeons decreased withthe number of surgical procedures, underlining the importance oftraining. Our data confirm the well-known effect of training on thelearning curve (6, 12, 13, 22, 25–28, 30, 35, 36). Gentle tissuehandling, however, is more complex than is reflected in theduration of surgery, as for similar operating times, experiencedsurgeons had always less adhesions than inexperienced surgeonsduring the entire learning curve.
Postoperative adhesion formation is influenced not only by surgi-cal training but also by the duration of the pneumoperitoneum,whichin the present study was kept constant at 60 minutes. Less adhesionformation with shorter procedures may happen due to less exposureto CO2 pneumoperitoneum and thus to less pneumoperitoneum-enhanced hypoxia, CO2, and pH changes. In the study performedin rabbits, duration of surgery also seemed to be important in the out-come of postoperative adhesion formation, although it is difficult to
195
completely separate both effects as duration of surgery and trainingof surgeon are intimately related.
Other reports studying only the effect of CO2 or helium pneumo-peritoneum on postoperative adhesion formation have shown an im-portant effect of the time of exposure. Those studies show the samesituation—the longer the exposure, the higher the adhesion forma-tion (22, 37).
Our study indicates that genetic background could also influenceadhesion formation, at least after laparoscopic surgery (38). BALB/cmice, an inbred strain, showed lower interanimal variability andhigher adhesion formation (the latter not significant) in comparisonwith Swiss mice, an outbred strain, observation to take into accountwhen developing a standardized animal model for the study of adhe-sion formation. Strain differences have been reported for other pro-cesses involving fibrosis and healing responses such as hepatic,lung, and colorectal fibrosis (39–41), myocardial and ear woundhealing (42, 43), and bone regeneration (44). This is not surprisingbecause inbred strains, maintained by sibling (brother� sister) mat-ing for 20 or more generations, are genetically almost identical,homozygous at virtually all loci, and with high phenotypic unifor-mity (45). This less interanimal variability in inbred strainshas been reported for many processes such as sleeping time underanesthesia (46).
The mouse model has many advantages compared with other an-imal models because it is relatively cheap, easy to handle, and does
196 Corona et al. Techniques and instrumentation
not require strict sterile conditions for surgery. Furthermore, it isparticularly useful for diverse studies because of the availability ofanimals with low genetic variability (i.e., inbred mice), underex-pressing or overexpressing specific genes (i.e., transgenic mice),and immunodeficient by spontaneous mutation (i.e., nude mice[T-cell deficient] and SCID mice [T&B cell deficient]). In addition,many specific mouse assays and monoclonal antibodies areavailable.
In conclusion, first, this study confirms that surgical training is ex-tremely important, not only in a hospital setting when dealing withpatients but also when performing surgical studies with the aim ofanalyzing the outcome. In this case, training experience has amarkedeffect on the development of postoperative adhesion formation. Sec-ond, it is preferable to use a strain with high adhesion formation po-tential and low interanimal variability such as BALB/c mice. Webelieve fewer inbred animals will be needed to achieve a given levelof statistical precision than if outbred animals are used (44). It is,however, necessary to point out that inbred strains in general weighless than outbred strains (average of 20 g vs. 32 g), which increasesthe technical skills required to do the experiments, especially thoseinvolving laparoscopic surgery.
Acknowledgments: The authors thank Mads Riiskjaer for participating in the
learning curve. We do thank Storz AG for their generous supply of
equipment.
REFERENCES
1. Moorthy K, Munz Y, Jiwanji M, Bann S, Chang A,
Darzi A. Validity and reliability of a virtual reality up-
per gastrointestinal simulator and cross validation us-
ing structured assessment of individual performance
with video playback. Surg Endoscop Other Intervent
Techn 2004;18:328–33.
2. Munz Y, Moorthy K, Bann S, Shah J, Ivanova S,
Darzi SA. Ceiling effect in technical skills of surgical
residents. Am J Surg 2004;188:294–300.
3. Herrell SD, Smith JA. Laparoscopic and robotic rad-
ical prostatectomy: what are the real advantages?
BJU Intern 2005;95:3–4.
4. Tekkis PP, Fazio VW, Lavery IC, Remzi FH,
Senagore AJ, Wu JS, et al. Evaluation of the learning
curve in heal pouch-anal anastomosis surgery. Ann
Surg 2005;241:262–8.
5. Tekkis PP, Senagore AJ, Delaney CP, Fazio VW.
Evaluation of the learning curve in laparoscopic
colorectal surgery—comparison of right-sided
and left-sided resections. Ann Surg
2005;242:83–91.
6. Aggarwal R, Tully A, Grantcharov T, Larsen CR,
Miskry T, Farthing A, et al. Virtual reality simulation
training can improve technical skills during laparo-
scopic salpingectomy for ectopic pregnancy. BJOG
2006;113:1382–7.
7. Aggarwal R, Black SA, Hance JR, Darzi A,
Cheshire NJ. Virtual reality simulation training can
improve inexperienced surgeons’ endovascular skills.
Eur J Vasc Endovasc Surg 2006;31:588–93.
8. Hyltander A, Liljegren E, Rhodin PH, Lonroth H. The
transfer of basic skills learned in a laparoscopic sim-
ulator to the operating room. Surg Endoscop Other
Intervent Techn 2002;16:1324–8.
9. Seymour NE, Gallagher AG, Roman SA,
O’Brien MK, Bansal VK, Andersen DK, et al. Virtual
reality training improves operating room perfor-
mance—results of a randomized, double-blinded
study. Ann Surg 2002;236:458–64.
10. Watterson JD, Beiko DT, Kuan JK, Denstedt JD. A
randomized, prospective blinded study validating
the acquisition of ureteroscopy skills using a computer
based virtual reality endourological simulator. J Urol
2002;168:1928–32.
11. Grantcharov TP, Kristiansen VB, Bendix J,
Bardram L, Rosenberg J, Funch-Jensen P. Random-
ized clinical trial of virtual reality simulation for
laparoscopic skills training. Br J Surg
2004;91:146–50.
12. Van Sickle KR, Ritter EM, Baghai M,
Goldenberg AE, Huang IP, Gallagher AG, et al.
Prospective, randomized, double-blind trial of
curriculum-based training for intracorporeal sutur-
ing and knot tying. J Am Coll Surg
2008;207:560–8.
13. Vossen C, Van Ballaer P, Shaw RW, Koninckx PR. Ef-
fect of training on endoscopic intracorporeal knot ty-
ing. Hum Reprod 1997;12:2658–63.
14. Molinas CR, Binda MM, Mailova K,
Koninckx PR. The rabbit nephrectomy model for
training in laparoscopic surgery. Hum Reprod
2004;19:185–90.
15. Shabtai M, Rosin D, Zmora O, Munz Y, Scarlat A,
Shabtai EL, et al. The impact of a resident’s se-
niority on operative time and length of hospital
stay for laparoscopic appendectomy—outcomes
used to measure the resident’s laparoscopic skills.
Surg Endoscop Other Intervent Techn
2004;18:1328–30.
16. Farnworth LR. A comparison of operative times in ar-
throscopic ACL reconstruction between orthopaedic
faculty and residents: the financial impact of ortho-
paedic surgical training in the operating room. Iowa
Orthoped J 2001;21:31–5.
17. Babineau TJ, Becker J, Gibbons G, Sentovich S,
Hess D, Robertson S, et al. The ‘‘cost’’ of operative
training for surgical residents. Arch Surg 2004;
139:366–9.
18. Kauvar DS, Braswell A, Brown BD, Harnisch M. In-
fluence of resident and attending surgeon seniority on
operative performance in laparoscopic cholecystec-
tomy. J Surg Res 2006;132:159–63.
19. Wilkiemeyer M, Pappas TN, Giobbie-Hurder A,
Itani KMF, Jonasson O, Neumayer LA. Does resident
post graduate year influence the outcomes of inguinal
hernia repair? Ann Surg 2005;241:879–82.
20. Barth RJ, Babineau, Hiebert C, Deckers PJ,
Russell PS, Goldfarb WB, et al. The ‘‘cost’’ of opera-
tive training for surgical residents—discussion. Arch
Surg 2004;139:369–70.
21. Bridges M, Diamond DL. The financial impact of
teaching surgical residents in the operating room.
Am J Surg 1999;177:28–32.
22. Molinas CR, Mynbaev O, Pauwels A, Novak P,
Koninckx PR. Peritoneal mesothelial hypoxia during
pneumoperitoneum is a cofactor in adhesion forma-
tion in a laparoscopic mouse model. Fertil Steril
2001;76:560–7.
23. Schonman R, Corona R, Bastidas A, De Cicco C,
Koninckx PR. Effect of upper abdomen tissue manip-
ulation on adhesion formation between injured areas
in a laparoscopic mouse model. J Min Invas Gynecol
2009;16:307–12.
24. Corona R, Verguts J, Schonman R, Binda MM,
Mailova K, Koninckx PR. Post-operative inflamma-
tion at the abdominal cavity increases adhesion for-
mation in a laparoscopic mouse model. Fertil Steril
2011;95:1224–8.
25. Molinas CR, Campo R, Elkelani OA, Binda MM,
Carmeliet P, Koninckx PR. Role of hypoxia inducible
factors 1alpha and 2alpha in basal adhesion formation
and in carbon dioxide pneumoperitoneum-enhanced
adhesion formation after laparoscopic surgery in trans-
genic mice. Fertil Steril 2003;80(Suppl 2):795–802.
26. Molinas CR, Elkelani O, Campo R, Luttun A,
Carmeliet P, Koninckx PR. Role of the plasminogen
system in basal adhesion formation and carbon diox-
ide pneumoperitoneum-enhanced adhesion formation
after laparoscopic surgery in transgenic mice. Fertil
Steril 2003;80:184–92.
27. Molinas CR, Campo R, Dewerchin M, Eriksson U,
Carmeliet P, Koninckx PR. Role of vascular endothe-
lial growth factor and placental growth factor in basal
Vol. 96, No. 1, July 2011
adhesion formation and in carbon dioxide
pneumoperitoneum-enhanced adhesion formation af-
ter laparoscopic surgery in transgenic mice. Fertil
Steril 2003;80(Suppl 2):803–11.
28. Binda MM, Molinas CR, Mailova K, Koninckx PR.
Effect of temperature upon adhesion formation in
a laparoscopic mouse model. Hum Reprod 2004.
29. ElkelaniOA,BindaMM,MolinasCR,KoninckxPR. Ef-
fect of adding more than 3% oxygen to carbon dioxide
pneumoperitoneum on adhesion formation in a laparo-
scopic mouse model. Fertil Steril 2004;82:1616–22.
30. Elkelani OA,Molinas CR,Mynbaev O, Koninckx PR.
Prevention of adhesions with crystalloids during lap-
aroscopic surgery in mice. J Am Assoc Gynecol Lap-
arosc 2002;9:447–52.
31. Binda MM,Molinas CR, Hansen P, Koninckx PR. Ef-
fect of desiccation and temperature during laparos-
copy on adhesion formation in mice. Fertil Steril
2006;86:166–75.
32. Pouly JL, Seak-San S. Adhesions: laparoscopy versus
laparotomy. In: diZerega GS, editor. Peritoneal sur-
gery. New York: Springer; 2000. p. 183–92.
33. Molinas CR, Cabral CR, Koninckx PR. Effect of the
diameter of the endoscope and of surgeon training
on the duration and quality of laparoscopic surgery
in a rabbit model. J Am Assoc Gynecol Laparosc
1999;6:447–52.
Fertility and Sterility�
34. Ordonez JL, Dominguez J, Evrard V, Koninckx PR.
The effect of training and duration of surgery on ad-
hesion formation in the rabbit model. Hum Reprod
1997;12:2654–7.
35. Aggarwal R, Darzi A. Technical-skills training in the
21st century. N Engl J Med 2006;355:2695–6.
36. Aggarwal R, Mytton OT, Derbrew M, Hananel D,
Heydenburg M, Issenberg B, et al. Training and sim-
ulation for patient safety. Qual Saf Health Care
2010;19(Suppl 2):i34–43.
37. Baba Y, Uetsuka K, Nakayama H, Dot K. Rat strain
differences in the early stage of porcine-serum-
induced hepatic fibrosis. Exp Toxicol Pathol
2004;55:325–30.
38. Ma B, Blackburn MR, Lee CG, Homer RJ, Liu W,
Flavell RA, et al. Adenosine metabolism and murine-
strain-specific IL-4-induced inflammation, emphy-
sema, and fibrosis. J Clin Invest 2006;116:1274–83.
39. SkwarchukMW, Travis EL. Murine strain differences
in the volume effect and incidence of radiation-
induced colorectal obstruction. Intern J Radiation On-
col Biol Physics 1998;41:889–95.
40. Osadchii O, Norton G, Deftereos D, Woodiwiss A.
Rat strain-related differences in myocardial adrener-
gic tone and the impact on cardiac fibrosis, adrenergic
responsiveness and myocardial structure and func-
tion. Pharmacol Res 2007;55:287–94.
41. Caldwell RL, Opalenik SR, Davidson JM, Caprioli RM,
Nanney LB. Tissue profiling MALDI mass spectrometry
reveals prominent calcium-binding proteins in the pro-
teome of regenerative MRL mouse wounds. Wound Re-
pair Regen 2008;16:442–9.
42. Mitchell CA, Grounds MD, Papadimitriou JM. The
genotype of bone-marrow-derived inflammatory cells
does not account for differences in skeletal-muscle re-
generation between Sjl/J and Balb/C mice. Cell Tis-
sue Res 1995;280:407–13.
43. McClearn GE, Hofer SM. Genes as gerontological
variables: uniform genotypes. Neurobiol Aging
1999;20:95–104.
44. Homanics GE, Le NQ, Kist F, Mihalek R, Hart AR,
Quinlan JJ. Ethanol tolerance and withdrawal re-
sponses in GABA(A) receptor alpha 6 subunit
null allele mice and in inbred C57BL/6J and strain
129/SvJ mice. Alcohol Clin Exper Res 1998;
22:259–65.
45. Molinas CR, Koninckx PR. Hypoxaemia induced by
CO(2) or helium pneumoperitoneum is a co-factor
in adhesion formation in rabbits. Hum Reprod
2000;15:1758–63.
46. Molinas CR, Binda MM, Campo R, Koninckx PR.
Adhesion formation and interanimal variability in
a laparoscopic mouse model varies with strains. Fertil
Steril 2005;83:1871–4.
197
Received
January
R.C. has n
to discl
disclose
Reprint re
Gynaec
berg, He
gmail.co
FC
1224
Postoperative inflammation in the abdominal cavityincreases adhesion formation in a laparoscopicmouse model
Roberta Corona, M.D.,a,b Jasper Verguts, M.D.,a Ron Schonman, M.D.,a Maria Mercedes Binda, Ph.D.,a
Karina Mailova, M.D.,a and Philippe Robert Koninckx, M.D., Ph.D.a
a Department of Obstetrics and Gynaecology, University Hospital Gasthuisberg, Leuven, Belgium; and b Department of
Obstetrics and Gynaecology, University Hospital S. Giovanni di Dio, Cagliari, Italy
Objective: To investigate acute inflammation in the peritoneal cavity in adhesion formation.Design: Prospective randomized, controlled trial.Setting: University laboratory research center.Animal(s): 9- to 10-week-old BALB/c female mice.Intervention(s): In a laparoscopic mouse model, acute inflammation in the peritoneal cavity evaluated in CO2
pneumoperitoneum enhaced adhesions, by CO2 pneumoperitoneum plus manipulation, and in the latter groupplus dexamethasone.Main Outcome Measure(s): Qualitative and quantitative adhesion scores and an acute inflammation score (neoan-giogenesis, diapedesis, and leukocyte accumulation).Result(s): Adhesions at the lesion site were enhanced by the CO2 pneumoperitoneum, further enhanced by manip-ulation, and decreased by the administration of dexamethasone. The acute inflammation scores (total, neoangiogen-esis, diapedesis, and leukocyte accumulation) strongly correlated with the total adhesion score. Inflammationscores were similar at both the surgical lesion and the parietal peritoneum.Conclusion(s): Acute inflammation of the entire peritoneum cavity is an important mechanism involved inadhesion formation and enhances adhesion formation at the lesion site. (Fertil Steril� 2011;95:1224–8. �2011by American Society for Reproductive Medicine.)
Key Words: Adhesions, acute inflammation, laparoscopy, dexamethasone, inflammation score
Postoperative adhesion formation results from a series of local factor-a (TNF-a) had any effect upon adhesion formation in hyp-
events at the trauma site. Peritoneal injury by surgery, infection,or irritation initiates a local inflammatory reaction, exudation,and fibrin deposition into which white blood cells, macrophages,fibroblasts, and mesothelial cells can migrate, proliferate, and/ordifferentiate. Within a few hours the lesion is covered by macro-phages and other tissue repair cells whose exact precursors are stillunclear (1–3). The local interplay between inflammatory cells,macrophages, and cytokines is not well understood.These events at the trauma site are modulated by factors derivedfrom the peritoneal cavity. Adhesions at the lesion site are enhancedin a dose-dependent way by pure CO2 pneumoperitoneum (4), pneu-moperitoneumwith more than 10%O2 (5, 6), by desiccation (7), andby bowel manipulation at a remote site (8) believed to act throughmesothelial hypoxia, mesothelial hyperoxia and reactive oxygenspecies (ROS), desiccation, and trauma, respectively.
Because the inflammatory reaction at the lesion site is widelybelieved to be a driving mechanism of adhesion formation (9–19),it was surprising that neither nonsteroidal anti-inflammatorydrugs (NSAIDs) such as cyclooxygenase-1 (COX-1) or COX-2inhibitors, nor antibodies neutralizing anti-tumor necrosis
September 2, 2010; revised November 25, 2010; accepted
4, 2011; published online February 4, 2011.
othing to disclose. J.V. has nothing to disclose. R.S. has nothing
ose. M.M.B. has nothing to disclose. K.M. has nothing to
. P.R.K. has nothing to disclose.
quests: Roberta Corona, M.D., Department of Obstetrics and
ology, University Hospital Leuven, University Hospital Gasthuis-
restraat 49, B-3000, Leuven, Belgium (E-mail: coronaroberta@
m).
ertility and Sterility� Vol. 95, No. 4, March 15, 2011opyright ª2011 American Society for Reproductive Medicine, P
oxia-enhanced adhesion (pure CO2 pneumoperitoneum) (20) orhyperoxia-enhanced adhesion (pneumoperitoneum with morethan 12% O2) (6) models. In contrast, dexamethasone, a steroidalanti-inflammatory drug, reduced adhesions by 30% and 62% inthe hypoxia- and hyperoxia-enhanced adhesions models, respec-tively. We therefore investigated the relationship between adhe-sion formation and acute inflammation in the entire peritonealcavity.
MATERIALS AND METHODSLaparoscopic Mouse Model for Adhesion FormationThe experimental setup (i.e., animals, anesthesia, ventilation, laparoscopic
surgery, and induction and adhesion scoring) has been described in detail else-
where (4, 5, 21–25). The model consisted of a bipolar lesion made at the
begining of a 60 minutes pneumoperitoneum during laparoscopy. The
pneumoperitoneum was induced using the Thermoflator (Karl Storz,
Tuttlingen, Germany) through a 2 mm endoscope with a 3.3 external sheath
for insufflation (Karl Storz) introduced into the abdominal cavity through
a midline incision caudal to the xiphoid appendix. The incision was closed
gas tight around the endoscope to avoid leakage. The insufflation pressure
was 15 mm Hg. For humidification, the Storz Humidifier 204320 33 (Karl
Storz) was used.
After the establishment of the pneumoperitoneum, two 14-gauge catheters
(Insyte-W, Vialon; Becton Dickinson, Madrid, Spain) were inserted under
laparoscopic vision. Standardized 10 mm � 1.6 mm lesions were performed
in the antimesenteric border of both right and left uterine horns and in both
the right and left pelvic side walls with bipolar coagulation (20W, standard
coagulation mode, Autocon 350; Karl Storz). Because temperature is critical
for adhesion formation, animals and equipment were placed in a closed
chamber at 37�C (heated air, WarmTouch, Patient Warming System, model
0015-0282/$36.00ublished by Elsevier Inc. doi:10.1016/j.fertnstert.2011.01.004
FIGURE 1
Total adhesion and inflammation score during the first 7 days after
a surgical bipolar lesion and 60 minutes of CO2pneumoperitoneum.
Corona. Inflammation and adhesion formation. Fertil Steril 2011.
5700; Mallinckrodt Medical, Hazelwood, MO). Because anesthesia and ven-
tilation may influence body temperature (7), the timing from anesthesia (T0),
to intubation (at 10 minutes, T10), to the onset of the experiment (at 20
minutes, T20) was strictly controlled.
After 7 days, adhesions were scored quantitatively (proportion and total)
and qualitatively (extent, type, tenacity) blindly by laparotomy under a ste-
reomicroscope. The entire abdominal cavity was visualized using a xiphopu-
bic midline and bilateral subcostal incision. After the evaluation of port sites
and viscera (omentum, large and small bowels) for de novo adhesions, the fat
tissue surrounding the uterus was carefully removed. The length of the
visceral and parietal lesions and adhesions were measured. Adhesions,
when present, were lysed to evaluate their type and tenacity. Pouly and
Seak-San (44) terminology was used to describe the de novo adhesion
formation for the adhesions formed at nonsurgical sites and adhesion
formation for adhesions formed at the surgical site.
AnimalsThe study used 9- to 10-week-old female BALB/c mice of 20 to 24 g.
Animals were kept under standard laboratory conditions and diet at the ani-
mal facilities of the Katholieke Universiteit Leuven (KUL). The study was
approved by the Institutional Review Animal Care Committee.
Histologic Inflammation ScoreA scoring system of acute inflammation was based upon the known patho-
physiology of acute inflammation (26–30). Acute inflammation has three
major components: [1] alterations in vascular caliber leading to an increase
in blood flow, [2] neoangiogenesis and structural changes in the
microvasculature permitting plasma proteins and leukocytes to leave
circulation, and [3] diapedesis of leukocytes from the microcirculation
leading to their accumulation and activation in the lesion (31). We scored
[1] the number of vessels, reflecting neoangiogenesis; [2] the number of poly-
morphonuclear cells (PMNs) in diapedesis, reflecting the increased perme-
ability of vessels; and [3] the PMN accumulation at the site of the injury.
Scoring was as follows: neoangiogenesis (0 when %3 vessels, 1 when 4–8
vessels, 2 when 9–12 vessels, 3 when R12 vessels), vessel permeability
(0 when 0 PMNs in diapedesis, 1 when %2 PMNs, 2 when 3–4 PMNs,
3 when R4 PMNs), and leukocyte activation and accumulation (0 when
%3 activated PMNs are present, 1 when 4–8 PMNs, 2 when 9–12 PMNs,
3 whenR12 PMNs). Total inflammation score was the sum of the neoangio-
genesis, permeability, and leukocyte activation scores.
Histology and ImmunohistochemistryUnder stereomicroscopic vision, biopsy samples were obtained during lapa-
rotomy. After dissecting skin and muscles from the peritoneum, a tissue ad-
herent (Lyostypt; B. Braun Melsungen AG, Mesungen, Germany) was
applied over the whole length of the lesion plus 5 mm of peritoneum on
each side of the lesions. The specimens were then fixed with JB fix (32)
for 24 hours, embedded in paraffin, oriented, and four 4 to 6 mm sections
were taken perpendicularly to the surface and perpendicularly to the lesion.
Each section permitted evaluation of the changes in the lesion and the sur-
rounding peritoneum from the surface to the depth of the biopsy sampling.
Sections were immunohistochemically stained for CD45 to detect activated
leukocytes (LCA, Ly-5, T200) (BD Biosciences Pharmingen, San Diego,
CA) in citrate bluffer 80�C, pH 6.0, dilution 1/400.
Inflammation parameters were blindly scored under a microscope with
a camera (Axio scope, AxioCamMRc 5, KS 400 imaging system; Carl Zeiss
MicroImaging, Jena, Germany). For each slide, four high-power fields were
randomly chosen, two of the lesion and two of the surroundings (one for each
side), and we counted the number of vessels, PMNs in diapedesis, and
leukocyte stained by CD45.
Experiment DesignThe experiments were designed to investigate the effect on the inflammation
score of factors known to enhance adhesions (CO2 pneumoperitoneum and
manipulation) or to decrease adhesions (dexamethasone). Experiment 1
was designed to determine which day after surgery acute inflammation and
Fertility and Sterility�
adhesion formation should be scored. After 60 minutes of CO2 pneumoper-
itoneum (n ¼ 8), the inflammation and adhesion scores were evaluated after
1, 2, 4, and 7 days (two mice per day) (Fig. 1). Based on these results, we
decided to evaluate inflammation and adhesions 48 hours after surgery in
all further experiments.
Experiment 2 (16 mice: 4 mice per group) was designed to evaluate the in-
flammation score and adhesions in control animals after 60 minutes of CO2
pneumoperitoneum with 4%O2 (group 1), after 60 minutes CO2 pneumoper-
itoneum (group 2), and after 60 minutes CO2 pneumoperitoneum plus
manipulation (group 3); group 4 underwent the same manipulation-
enhanced adhesion formation as group 3 but with added dexamethasone,
which is known to decrease adhesions (8). Manipulation consisted of manip-
ulating fat and bowels in the upper abdomen: the omentum and large and
small bowels were moved gently up and down across the abdomen for 5 min-
utes with a 1.5-mm nontraumatic grasper for 5 minutes as described
elsewhere (8). Dexamethasone (Aacidexam, 5 mg for injection; Organon,
Brussels, Belgium) was administered by intraperitoneal injection of the med-
ication through the midline of the lower abdominal wall with a 31-gauge, 8
mm (5/16-in) BD Ultra-Fine Short Needle (BD, Franklin Lakes, NJ) holding
the mouse in an upside down position to avoid bowel injuries, at the dose of
40 mg at the end of pneumoperitoneum and 40 mg after 24 hours.
Experiment 3 (16 mice: 4 mice per group) was similar to experiment 2; in-
flammation was scored at the parietal peritoneum in the upper abdomen at
distance from the surgical site. Experiments 2 and 3 were block randomized
by day, meaning that one mouse from each group was treated randomly the
same day.
StatisticsStatistical analyses were performed with the SAS System (SAS Institute,
Cary, NC) using a Wilcoxon test for differences in adhesion formation,
and Spearman correlation between adhesion and total inflammation scores.
All the data are presented as mean � standard deviation.
RESULTSExperiment 1The time courses of both adhesion scores (total and proportion) weresimilar. Total adhesion scores after 24, 48, and 72 hours and 7 days
1225
FIGURE 2
Correlation between total adhesion and total inflammation scores
at the lesion (P< .0001, Spearman test; see Table 1).
Corona. Inflammation and adhesion formation. Fertil Steril 2011.
FIGURE 3
Total adhesion and total inflammation scores at the surgical lesion
(mean of scores of the lesion and 5mm from the lesion; experiment2) and at the parietal peritoneum (experiment 3).
Corona. Inflammation and adhesion formation. Fertil Steril 2011.
were 2.32� 0.29, 3.01� 0.18, 3.35� 0.31, and 3.74� 0.32, respec-tively. Total adhesion scores and the total inflammation, the formerincreasing and the latter decreasing progressively, are shown inFigure 1. Also the scores of neoangiogenesis, diapedesis, and leuko-cytes accumulation decreased progressively. We therefore decidedto evaluate acute inflammation and adhesion formation on day 2in all subsequent experiments.
Experiment 2Adhesion scores on day 2 confirmed the observations from day 7 af-ter surgery (8) (Figure 2 and Table 1). In comparison with the controlgroup (group 1), both total and proportion of adhesions increasedwhen a pure CO2 pneumoperitoneum was used (group 2; P¼.007and P¼.0053, respectively). Adhesions further increased in group3 (P<.0001 for both total and proportion). The addition of dexa-methasone (group 4) decreased adhesion formation (P<.0001 forboth adhesion scores).The total inflammation score at the centralpart of the surgical lesion was slightly higher in comparison withthe inflammation score at 0.5 cm from the lesion: 1.5 � 0.40 and2.625 � 0.25 (P¼.0203), 0.625 � 0.25 and 2.125 � 0.25(P¼.0154), 4.125 � 0.25 and 6.375 � 0.25 (P¼.0321), 2.375 �0.25 and 3.875 � 0.25 (P¼.0591) for groups 1, 2, 3, and 4,respectively.
Depth of the inflammatory reaction spanned 2 mm to a maximumof 4 mm in group 3 which had the highest inflammation score.Changes in total inflammation scores at the lesion (mean of theinflammation scores in the central part and in the periphery) andin adhesions scores were strikingly similar (Figs. 3 and 4). At thelevel of the surgical lesion, pure CO2 pneumoperitoneum in compar-ison with group 1 slightly increased the total inflammation score(P¼.07), neoangiogenesis (P¼.0577), and lymphocyte accumula-tion (P¼.0796). In group 3, the inflammation score furtherincreased: total inflammation (group 2 vs. group 3, P<.0001),neoangiogenesis (P¼.0007), vasodilatation and permeability(P¼.0052), and lymphohistiocytic activation and accumulation(P¼.0022). When dexamethasone was added after surgery, the totalmean inflammation score decreased (group 3 vs. group 4, P<.0001),an effect observed for all parameters: neoangiogenesis (P¼.0154),
1226 Corona et al. Inflammation and adhesion formation
diapedesis (P¼.0016), and lymphocytes activation-accumulation(P¼.001). Most strikingly, however, was the strong correlation be-tween adhesion scores and inflammatory parameters (Table 1). Totalacute inflammation score (Fig. 2), neoangiogenesis, diapedesis, andleukocyte accumulation (see Table 1) strongly correlated with totaladhesion scores.
Experiment 3Experiment 3 confirmed the results of experiment 2. In addition, theinflammation scores of the parietal peritoneum were found to becomparable with those at the surgical lesion (see Fig. 4). Indeed,the use of pure CO2 pneumoperitoneum in comparison with CO2
þ 4% O2 increased slightly the total inflammation score (not statis-tically significant, P¼.06); manipulation further increase the totalinflammation score (group 2 vs. group 3, P<.0001), and dexameth-asone decreased the total inflammation score (group 3 vs. group 4,P<.0001).
DISCUSSIONThese data confirm and extend previous observations on adhesionformation in our laparoscopic mouse model. Indeed, as observedon day 7, pure CO2 pneumoperitoneum increased adhesions on
Vol. 95, No. 4, March 15, 2011
TABLE 1Correlation between adhesion and inflammation scores.
Inflammatory parameters Biopsy location
Adhesion score (P value)
Total Extent Type Tenacity Proportion (%)
Neoangiogenesis Central .0002 NS < .0001 .0032 NS
Surrounding .0002 NS < .0001 .0021 NS
Permeability Central .0391 NS .0411 NS NS
Surrounding .0286 NS .0382 NS NSLeukocytes accumulation Central .0003 .0043 .0029 < .0001 .0031
Surrounding .0002 .0032 .0021 < .0001 .0028
Total score Central < .0001a .0017 < .0001 < .0001 .0005Surrounding < .0001a .0005 < .0001 < .0001 .0005
Note: Spearman correlation was performed. NS ¼ not statistically significant.a Data shown in Figure 2.
Corona. Inflammation and adhesion formation. Fertil Steril 2011.
day 2 in comparison with a pneumoperitoneumwith 96%CO2þ 4%O2, (4, 5, 24). When, in addition, mechanical trauma was added, theadhesions further increased at the lesion site (8), whereas dexameth-asone decreased the adhesions (8). Also the absence of de novo ad-hesions was confirmed. The parallelism between changes inadhesion formation and in acute inflammation scores was striking,with a linear correlation between adhesion and inflammation scores.This suggests that acute inflammation is an important common driv-ing mechanism modulating adhesion formation. The similarity ofacute inflammation at the lesion, in the peritoneum 5 mm from thelesion, and in the entire peritoneal cavity strongly supports the con-cept that the entire peritoneal cavity is a cofactor affecting adhesionformation at the lesion site. In particular, the effect of bowel manip-ulation in the upper abdomen (i.e., at a distance from the bipolar le-sion) strongly supported the concept that some peritoneal cavityfactors stimulated by mesothelial trauma with exposure of the basalmembrane may enhance adhesion formation and also inflammationat the lesion site. The absence of de novo adhesions confirmed thatadhesion formation requires a peritoneal lesion, but that quantita-tively the inflammatory reaction in the entire peritoneal cavity isthe most important factor.
The exact mechanism by which the inflammatory reaction of theentire peritoneal cavity affects adhesion formation at the lesion siteis still unclear. Because mesothelial cells are known to retract and tobulge during CO2 pneumoperitoneum without affecting the basalmembrane, and because bowel manipulation was done very gently,a very superficial mesothelial cell trauma is suggested as causing aninflammatory reaction of the entire peritoneal cavity. Subsequently,some substances or cells could be released, activated, or attractedinto the peritoneal fluid affecting adhesion formation at the lesionsite. We speculate that chemokines could be involved as they areknown to be inflammatory mediators involved in the activationand migration of leukocytes into the tissue (33). Indeed, according
Fertility and Sterility�
to the position of the first two cysteine residues (34), some arechemoattractants and activators of non-PMNs leukocytes, andothers attract neutrophils. Also, macrophages and leukocytesattracted into the peritoneal cavity and their secretion products ascytokines are likely to be involved.
Nonsteroidal anti-inflammatory drugs (NSAIDs) such as ibupro-fen, tenoxicam, nimesulide, parecoxib, and TNF-a-neutralizing an-tibodies do not affect adhesion formation (20), but dexamethasonehas been confirmed to decrease adhesion formation (20, 35) whiledecreasing the acute inflammatory reaction. The mechanism bywhich dexamethasone decreases the inflammatory reaction in theperitoneal cavity can only be speculated upon. Dexamethasone hasa wide range of effects such as inhibiting fibroblast proliferationand procollagen gene expression through a decreased transforminggrowth factor secretion (36) or cytokines (37).
Moreover, dexamethasone has an anti-inflammatory effect by in-ducing mitogen-activated protein kinase (MAPK) phosphatase-1(MKP-1), decreasing cytokines in innate immune cells (38) andMAPKs (38). The MAPKs include the extracellular signal-regulating kinase (ERK), p38 MAPK, and c-Jun N-Terminal proteinKinase (JNK), and they all play an important role in cell proliferation,apoptosis, and many other nuclear events. In particular, MKP-1 hasbeen shown to inhibit a number of cellular responses mediated byERK and p38 MAPK. The MAPKs regulate metalloproteinases(MMPs), thus affecting fibrinolysis and nitric oxide (NO) productionand angiogenesis (39–43). Our data strongly suggest that acuteinflammation in the entire peritoneum cavity is an importantmechanism involved in adhesion formation and enhancingadhesion formation at the lesion site.
Acknowledgments: The authors thank Rita VanBree, Robert Pijnenborg, and
Lisbeth Vercruysse for their essential help in the laboratory, and Gian
Benedetto Melis and Stefano Angioni for revising the manuscript.
REFERENCES
1. diZerega GS. Biochemical events in peritoneal tissue
repair. Eur J Surg Suppl 1997;577:10–6.
2. diZerega GS, Campeau JD. Peritoneal repair and
post-surgical adhesion formation. Hum Reprod
Update 2001;7:547–55.
3. diZerega GS. Peritoneum, peritoneal healing and ad-
hesion formation. In: diZerega GS, editor. Peritoneal
surgery. New York: Springer; 2000. p. 3–38.
4. Molinas CR, Mynbaev O, Pauwels A, Novak P,
Koninckx PR. Peritoneal mesothelial hypoxia during
pneumoperitoneum is a cofactor in adhesion forma-
tion in a laparoscopic mouse model. Fertil Steril
2001;76:560–7.
5. Elkelani OA, Binda MM, Molinas CR, Koninckx PR.
Effect of adding more than 3% oxygen to carbon
1227
dioxide pneumoperitoneum on adhesion formation
in a laparoscopic mouse model. Fertil Steril
2004;82:1616–22.
6. Binda MM, Molinas CR, Koninckx PR. Reactive ox-
ygen species and adhesion formation: clinical impli-
cations in adhesion prevention. Hum Reprod
2003;18:2503–7.
7. Binda MM, Molinas CR, Hansen P, Koninckx PR.
Effect of desiccation and temperature during laparos-
copy on adhesion formation in mice. Fertil Steril
2006;86:166–75.
8. Schonman R, Corona R, Bastidas A, De Cicco C,
Koninckx PR. Effect of upper abdomen tissue manip-
ulation on adhesion formation between injured areas
in a laparoscopic mouse model. J Minim Invasive Gy-
necol 2009;16:307–12.
9. Guvenal T, Cetin A, Ozdemir H, Yanar O, Kaya T.
Prevention of postoperative adhesion formation in
rat uterine horn model by nimesulide: a selective
COX-2 inhibitor. Hum Reprod 2001;16:1732–5.
10. Siegler AM, Kontopoulos V, Wang CF. Prevention of
postoperative adhesions in rabbits with ibuprofen,
a nonsteroidal anti-inflammatory agent. Fertil Steril
1980;34:46–9.
11. Luciano AA, Hauser KS, Benda J. Evaluation of com-
monly used adjuvants in the prevention of postopera-
tive adhesions. Am J Obstet Gynecol 1983;146:88–92.
12. Aldemir M, Ozturk H, Erten C, Buyukbayram H. The
preventive effect of Rofecoxib in postoperative
intraperitoneal adhesions. Acta Chir Belg 2004;104:
97–100.
13. Celebioglu B, Eslambouli NR, Olcay E, Atakan S.
The effect of tenoxicam on intraperitoneal adhesions
and prostaglandin E2 levels in mice. Anesth Analg
1999;88:939–42.
14. Golan A, Bernstein T, Wexler S, Neuman M,
Bukovsky I, David MP. The effect of prostaglandins
and aspirin—an inhibitor of prostaglandin synthe-
sis—on adhesion formation in rats. Hum Reprod
1991;6:251–4.
15. Tayyar M, Basbug M. The effects of intraperitoneal
piroxicam and low molecular weight heparin in pre-
vention of adhesion reformation in rat uterine horn.
Res Exp Med (Berl) 1999;198:269–75.
16. Nishimura K, Nakamura RM, diZerega GS. Bio-
chemical evaluation of postsurgical wound repair:
prevention of intraperitoneal adhesion formation
with ibuprofen. J Surg Res 1983;34:219–26.
17. Nishimura K, Nakamura RM, diZerega GS. Ibupro-
fen inhibition of postsurgical adhesion formation:
a time and dose response biochemical evaluation in
rabbits. J Surg Res 1984;36:115–24.
18. Rodgers K, Girgis W, diZerega GS, Bracken K,
Richer L. Inhibition of postsurgical adhesions by li-
posomes containing nonsteroidal antiinflammatory
drugs. Int J Fertil 1990;35:315–20.
1228 Corona et al. Inflammation an
19. Greene AK, Alwayn IP, Nose V, Flynn E, Sampson D,
Zurakowski D, et al. Prevention of intra-abdominal
adhesions using the antiangiogenic COX-2 inhibitor
celecoxib. Ann Surg 2005;242:140–6.
20. Binda MM, Molinas CR, Bastidas A, Koninckx PR.
Effect of reactive oxygen species scavengers, antiin-
flammatory drugs, and calcium-channel blockers on
carbon dioxide pneumo-peritoneum-enhanced adhe-
sions in a laparoscopic mouse model. Surg Endosc
2007;21:1826–34.
21. Molinas CR, Campo R, Elkelani OA, Binda MM,
Carmeliet P, Koninckx PR. Role of hypoxia inducible
factors 1a and 2a in basal adhesion formation and in
carbon dioxide pneumoperitoneum-enhanced adhe-
sion formation after laparoscopic surgery in trans-
genic mice. Fertil Steril 2003;80(Suppl 2):795–802.
22. Molinas CR, Elkelani O, Campo R, Luttun A,
Carmeliet P, Koninckx PR. Role of the plasminogen
system in basal adhesion formation and carbon diox-
ide pneumoperitoneum-enhanced adhesion formation
after laparoscopic surgery in transgenic mice. Fertil
Steril 2003;80:184–92.
23. Molinas CR, Campo R, Dewerchin M, Eriksson U,
Carmeliet P, Koninckx PR. Role of vascular endothe-
lial growth factor and placental growth factor in basal
adhesion formation and in carbon dioxide
pneumoperitoneum-enhanced adhesion formation af-
ter laparoscopic surgery in transgenic mice. Fertil
Steril 2003;80(Suppl 2):803–11.
24. Binda MM, Molinas CR, Mailova K, Koninckx PR.
Effect of temperature upon adhesion formation in
a laparoscopic mouse model. Hum Reprod
2004;19:2626–32.
25. Elkelani OA,Molinas CR,Mynbaev O, Koninckx PR.
Prevention of adhesions with crystalloids during lap-
aroscopic surgery in mice. J Am Assoc Gynecol Lap-
arosc 2002;9:447–52.
26. Krenn V, Morawietz L, Burmester GR, Kinne RW,
Mueller-Ladner U, Muller B, et al. Synovitis score:
discrimination between chronic low-grade and high-
grade synovitis. Histopathology 2006;49:358–64.
27. Krenn V, Morawietz L, Haupl T, Neidel J, Petersen I,
Konig A. Grading of chronic synovitis—a histopatho-
logical grading system for molecular and diagnostic
pathology. Pathol Res Pract 2002;198:317–25.
28. Krenn V, Morawietz L. Synovitis score: proposal for
a histopathological grading system of chronic synovi-
tis. Ann Rheum Dis 2004;63:9.
29. Haywood L, McWilliams DF, Pearson CI, Gill SE,
GanesanA,WilsonD,et al. Inflammationand angiogen-
esis in osteoarthritis. Arthritis Rheum 2003;48:2173–7.
30. Pessler F, Dai L, Diaz-Torne C, Gomez-Vaquero C,
Paessler ME, Zheng DH, et al. The synovitis of
‘‘non-inflammatory’’ orthopaedic arthropathies:
a quantitative histological and immunohistochemical
analysis. Ann Rheum Dis 2008;67:1184–7.
d adhesion formation
31. Kumar V, Abbas A, Fausto N. Acute and chronic
inflammation. In: Kumar V, Abbas A, Fausto N,
editors. Robbins and Cotran’s pathologic basis
of disease. 7th ed. Philadelphia: Saunders; 2005.
p. 82–3.
32. Beckstead JH. A simple technique for preservation of
fixation-sensitive antigens in paraffin-embedded tis-
sues: addendum. J Histochem Cytochem 1995;43:
345.
33. Adams DH, Lloyd AR. Chemokines: leucocyte re-
cruitment and activation cytokines. Lancet 1997;349:
490–5.
34. Murphy PM. The molecular-biology of leukocyte
chemoattractant receptors. Ann Rev Immunol
1994;12:593–633.
35. Binda MM, Koninckx PR. Hyperoxia and prevention
of adhesion formation: a laparoscopic mouse model
for open surgery. BJOG 2010;117:331–9.
36. Bladh LG, Johansson-Haque K, Rafter I, Nilsson S,
Okret S. Inhibition of extracellular signal-regulated
kinase (ERK) signaling participates in repres-
sion of nuclear factor (NF)-kB activity by glu-
cocorticoids. Biochim Biophys Acta 2009;1793:
439–46.
37. Brunton LL, Lazo JS, Parker KL, editors. Goodman
and Gilman’s the pharmacological basis of therapeu-
tics. 11th ed. New York: McGraw-Hill; 2006.
38. Wang XX, Nelin LD, Kuhlman JR, Meng XM,
Welty SE, Liu YS. The role of MAP kinase
phosphatase-1 in the protective mechanism of dexa-
methasone against endotoxemia. Life Sci 2008;83:
671–80.
39. Binneb€osel M, Klinge U, Rosch R, Junge K, Lynen-
Jansen P, Schumpelick V. Morphology, quality, and
composition in mature human peritoneal adhesions.
Langenbecks Arch Surg 2008;393:59–66.
40. Nakano A, Tani E, Miyazaki K, Yamamoto Y,
Furuyama JI. Matrix metalloproteinases and tissue
inhibitors of metalloproteinases in human gliomas.
J Neurosurg 1995;83:298–307.
41. Price NM, Farrar J, Chau TTH, Mai NTH, Hien TT,
Friedland JS. Identification of a matrix-degrading
phenotype in human tuberculosis in vitro and in
vivo. J Immunol 2001;166:4223–30.
42. Rao JS. Molecular mechanisms of glioma invasive-
ness: the role of proteases. Nat Rev Cancer
2003;3:489–501.
43. Yubero S, Ramudo L, Manso MA, De Dios I. Mech-
anisms of dexamethasone-mediated chemokine down-
regulation in mild and severe acute pancreatitis.
Biochim Biophys Acta 2009;1792:1205–11.
44. Pouly JL, Seak-San S. Adhesions: laparoscopy versus
laparotomy. In: diZerega GS, DeCherney AH,
Diamond MP, Ellis H, Gomel V, Haney AF, editors.
Peritoneal surgery. New York: Springer-Verlag;
2000. p. 183–92.
Vol. 95, No. 4, March 15, 2011
EDITORIAL COMMENTARY
Possible mechanisms of peritonealtissue-oxygen tension changes duringCO2-pneumoperitoneum: the role ofdesign, methodology and animalmodels
Editorial commentary on the articles ‘Peritonealtissue-oxygen tension during a carbon dioxidepneumoperitoneum in a mouse laparoscopic model withcontrolled respiratory support’ by Bourdel et al. (2007)and ‘Effects of supplemental perioperative oxygen onpost-operative abdominal wound adhesions in a mouselaparotomy model with controlled respiratory support’by Matsuzaki et al. (2007)
Ospan A. Mynbaev1,4 and Roberta Corona2,3
1Moscow State University of Medicine and Dentistry, Delegatskaya str. 20/1 127473, Moscow, Russian Federation 2Katholieke UniversiteitLeuven (KULeuven) Leuven, Belgium 3University of Cagliari, Cagliari, Italy
4Correspondence address. Tel: þ7-7252470137; Fax: þ7-7252530906; E-mail: [email protected]
Animal models and design ofexperimental studiesRecently, Bourdel et al. (2007) and Matsuzaki et al. (2007) publishedwell-documented articles covering the fundamental aspects of laparo-scopy. In order to comment on the study design and on the method-ology of studies using mouse ventilation models (VMs), we havecarefully studied the review by Bide et al. (1997) concerning minutevolume (MV) and body weight (BW) relationships since analogous sug-gestions with reference to respiratory parameters (RPs) and theirimpact on study results were pointed out in the article of Bourdelet al. (2007).
There is a very large variability in ventilation parameters (VPs)between animal models and between laboratories as well as between
anesthetized and non-anesthetized animals. Therefore, Bide et al.(1997) analyzed large numbers of studies to provide sufficient statisticaldata to reduce the effects of biological variability on VPs. Subsequently,we compared these incorporated values of RPs from physiologicalstudies (Bide et al., 1997; Zuurbier et al., 2002) with those in severalstudies by teams from Leuven (Molinas and Koninckx, 2000; Molinaset al., 2001, 2003a, b, 2004; Binda et al., 2004, 2006) and Clermont-Ferrand (Bourdel et al., 2007; Matsuzaki et al., 2007) using mice VMs.According to our comparison, very important RPs from most of theseexperimental studies with mice VMs did not correspond with RPsyielded by a great number of physiological studies (Table I).
We fully agree with Bourdel et al. (2007), who noted that ‘technicaldifficulties associated with endotracheal intubation are the mainreason why only a limited number of laboratories use mechanical
& The Author 2009. Published by Oxford University Press on behalf of the European Society of Human Reproduction and Embryology. All rights reserved.For Permissions, please email: [email protected]
Human Reproduction, Vol.24, No.6 pp. 1242–1246, 2009
Advanced Access publication on March 3, 2009 doi:10.1093/humrep/dep025
at KU
Leuven on July 6, 2011hum
rep.oxfordjournals.orgD
ownloaded from
ventilation in rodent models. Several methods require a great deal ofskill to perform . . . ’ and pointed out that difficult methodologicalaspects concerning manipulations and the choice of VPs such astidal volume and strokes.
We believe that researchers should study the physiology of theanimals used in their experiments very well since poorly performedexperiments may misrepresent results. In future, studies with VMsshould take into account all of the RPs such as tidal volume, respirat-ory strokes, the MV and BW ratio, as well as the fraction of inspiredoxygen (FiO2), in order to avoid flawed results and to allow standar-dized data comparison.
Based on our experience with rabbit models (Mynbaev et al.,2002a, 2007), we conclude that it is not surprising that CO2-pneumoperitoneum in itself, without perioperative controlled respirat-ory support, is enough to cause acidosis in mice, as shown by Bourdelet al. (2007), since absorption of CO2 through intact and injuredtissue (Kazama et al., 1998) as well as during intra- and extra-peritonealCO2 insufflations (Mullett et al., 1993) has been clearly demonstrated inboth spontaneously breathing and ventilated models (Neto et al., 2000;Mynbaev et al., 2002a, 2007; Bergstrom et al., 2008).
The impact of CO2-pneumoperitoneum on blood gases (BGs), acid–base balance (ABB) and metabolic parameters has been well studied insmaller (mice, rats) and larger (pigs, dogs, etc.) animals, but the mostintelligible mechanism of changes in these parameters was clearlyshown in rabbit models. We have also demonstrated that excess
amounts of insufflated gas induce changes in BGs and ABB homeostasis(Mynbaev et al., 2002a, b, 2007, in press): during CO2-pneumoperitoneum, the mesothelial surface CO2 tension in theabdominal cavity is considerably higher than in both venous and arterialblood. This tension gap makes CO2 pass into the blood through theparietal peritoneum tissue and the capillaries. Subsequently, partialpressure of venous and arterial CO2 increases to establish blood gasand acid–base equilibrium. Therefore, carboxemia and acidosis arethe key events, reflected by a pCO2 and HCO3
2 increase and by apH decrease. These changes lead to metabolic acidosis with an increasein lactate concentrations and a decrease in pH, acid–base equilibrium(ABE), standard base excess and standard bicarbonate. Finally,because of the Bohr effect, the changes result in metabolic hypoxemia.The latter changes are indeed associated with a desaturation, as a con-sequence of decreased oxygen availability in tissue, caused by increasedreduced or deoxyhemoglobin (RHb) and decreased hemoglobinoxygen affinity and, hence, decreased O2Hb.
In our opinion, experimental models must ‘quantitatively and qualitat-ively’ come as closely as possible to the physiologic conditions of theanimals used. Hence, to study the impact of CO2-pneumoperitoneumon BGs, ABB and metabolic parameters (Mynbaev et al., 2002a,2007), a rabbit model is the best one, since in smaller animals thereare repeated blood sample collection difficulties and with largeranimals there are technical and financial difficulties (which are veryexpensive and are therefore unavailable to many laboratories).
.............................................................................................................................................................................................
Table I Body weight and respiratory parameters of studies using mice ventilation models
Authors Mice strainsand conditions
BW (g) TV (mL) Strokes (min) MV (ml/min) MV/BW ratio
Adapted from Bideet al. (1997)
NAM 21 — — 31 1.491
AM 25 — — 26 1.044
CVSLA 25 — — 27 1.076
Binda et al. (2004) NMRI 30–40 (35) 250 160 40 0.875
Binda et al. (2006) NMRI 25–35 (30) 250 160 40 0.750BALB/c 19–21 (20) 0.500
250 LS 50 0.380
Bourdel et al. (2006) C57BL6 18–20 (20) 200 220 LT and AAC 44 0.432
Matsuzaki et al. (2007) C57BL6 18–20 (19) 200 220 44 0.432
Molinas et al. (2001) NMRI 45–55 (40) 1500 85 127.5 0.392
Molinas et al. (2003a, b) NMRI, WT andTM
30–40 (35) 500 85 42.5 0.824
Molinas et al. (2004) NMRI 30–35 (32.5) 500 a) 40, b) 60, c) 80, d)100, e) 120
a) 20, b) 30, c) 40,d) 50, e) 60
a) 1.625, b) 1.083, c)0.813, d) 0.650, e) 0.542
250 a) 80, b) 120, c) 160,d) 200, e) 240
Zuurbier et al. (2002) Swiss 33 300 100 30 1.100CD-1 32 1.067BalbC 27 0.900C57Bl6 27 0.900
AAC, anesthesia alone/control; AM, anesthetized mice; BW, body weight; CVSLA, calculated values for the ‘Standard’ laboratory animals; LS, laparoscopy; LT, laparotomy; MV, minutevolume; MW/BW ratio, minute volume and body weight ratio; NAM, non-anesthetized mice; TV, tidal volume.Mice strains: WT, wild type; TM, transgenic mice; NMRI, BALB/c, C57BL6, Swiss and CD-1 mice.
Role of design, methodology and animal models 1243
at KU
Leuven on July 6, 2011hum
rep.oxfordjournals.orgD
ownloaded from
Debates concerningCO2-pneumoperitoneum-inducedadhesion formationThe hypothesis that laparoscopy decreases post-operative adhesionswas predicted and subsequently confirmed by several clinical obser-vations, although this was in the absence of long-term randomizedstudies comparing adhesion formation (AF) after laparoscopy withopen surgery (Gutt et al., 2004; Sauerland and Neugebauer, 2005;Majewski, 2006).
Surprisingly, over the last decade, the impact of CO2-pneumoperitoneum on AF has been extensively discussed in a largebody of literature. The hypothesis that mesothelial hypoxia inducedby CO2-pneumoperitoneum can be a co-factor of post-operativeAF, was initially suggested by Professor Koninckx and has been exten-sively studied by his team (Molinas and Koninckx, 2000; Yesildaglarand Koninckx, 2000; Molinas et al., 2001, 2003a, b, 2004; Bindaet al., 2004, 2006; Elkelani et al., 2004).
Recently, a few reports have, either directly or indirectly, supportedthis phenomenon (Nagelschmidt et al., 2001; Bergstrom et al., 2003;de Souza et al., 2003).
Nevertheless, some other publications have presented oppositefindings, supporting the absence of any impact of CO2-pneumoperitoneum on either AF or tissue plasminogen activatorchanges during laparoscopic procedures (Ziprin et al., 2003; Miyanoet al., 2006).
Unfortunately, the direct subject of our study (Mynbaev et al., 2002b)was not correctly referenced in the paper (Matsuzaki et al., 2007): ‘onegroup demonstrated a reduction in AF (Mynbaev et al., 2002b; Elkelaniet al., 2004) and another showed no benefit (Demirbag et al., 2005)when 3% O2 was added to the CO2-pneumoperitoneum’.
We presented systemic BGs, ABB and metabolic changes andshowed the prevention of changes by adding small amounts of O2
to the CO2-pneumoperitoneum (Mynbaev et al., 2002b) and didnot study AF as closely as did the other team (Demirbag et al., 2005).
Misleading lapses by our colleagues (Demirbag et al., 2005), such asintraperitoneal insufflation of 100% O2 and its acidotic consequenceswith increased pCO2 and decreased pH in arterial blood, were clearlydemonstrated in our reply (Mynbaev, 2006). In their answer, theauthors fully agreed with our claims (Demirbag et al., 2006). Sub-sequently, data presented by this team (Demirbag et al., 2005),suggesting that intraperitoneally insufflated O2 can produce acidoticchanges, are misleading and contrary to the results of both thesestudies (Bourdel et al., 2007; Matsuzaki et al., 2007).
Mechanisms of a tissue-oxygentension on AFTissue tension and O2 blood partial pressure depend on the FiO2,humidity and temperature of air as well as on atmospheric pressure.Ragheb and Buggy (2004) wrote that ‘the tissue oxygen tension(tpO2) may be thought of as the local expression of global oxygendelivery in a particular tissue. It is the balance between oxygen per-fusion and oxygen consumption in the tissue at a given time. Intissues with stable oxygen consumption, tpO2 reflects tissue perfusionmore precisely . . . ’. FiO2 is a very important parameter during
mechanical ventilation, particularly in laparoscopy and other surgicalprocedures in patients with high cardiovascular risk.
It is well known that inflammation is characterized by significantchanges in metabolic activities since alterations in energy supply anddemand can result in diminished oxygen delivery and/or availability,leading to inflammation-associated tissue hypoxia and metabolicacidosis (Karhausen et al., 2005). Also, hypoxic conditions promoteinflammation while diminishing the ability of leukocytes to useoxygen (Kokura et al., 2002), while low tpO2 is associated withwound healing failure (Greif et al., 2000; Gottrup, 2004). This relation-ship between hypoxia, inflammation, tpO2 and wound healing mayexplain the adhesion prevention impact of perioperative supplementaloxygen observed in the second study of this team (Matsuzaki et al.,2007).
Bourdel et al. (2007) measured tpO2 in the retroperitoneum space.In our opinion, their observation was probably based on deep tpO2
measuring, at �1000 m or more. With histochemistry, we observedthat hypoxia during CO2-pneumoperitoneum is restricted to 50–100 m and only to the superficial mesothelial layers (,100 m) whichare hypoxic (Corona et al., unpublished data). Consequently, wesuggest that this tpO2 increase, measured at lower pneumoperito-neum pressure (2 mmHg), is due to reactive hyperemia.
Another mechanism of tpO2 increases during CO2-pneumoperitoneum can be explained by increased blood flow (BF)in the abdominal cavity, particularly in the intestinal muscular andmucosal layers. Recently, it was presented that overall, BF increasesat the beginning (at the 40 min) of both (CO2 or He) insufflations inmucosal and muscular intestine layers (cecum and sigmoid), thenreturns to near baseline levels at 80 and 120 min (Goitein et al.,2005). Increased BF was found in peritoneal layers (Brundell et al.,2002) and in tissues surrounding the abdominal cavity (Yavuz et al.,2003). Thus, changes in tpO2 observed in the study by Bourdelet al. (2007) during CO2-pneumoperitoneum could be related withelevated BF, since increased blood supply positively correlates withraised tpO2 (Schugart et al., 2008).
We would like to underline that this commentary does not intendto criticize our colleagues’ well-documented articles and their efforts,but simply aims to clarify a complex matter.
ConclusionsAppropriate designs, correct methodology and suitable animal modelsare very important factors in experimental studies dealing with funda-mental physiological processes aimed to broaden the insights in patho-logical human conditions.
AcknowledgementsWe acknowledge Ms Veronique Berkein and Professor PhilippeR. Koninckx for their help during the preparation of this manuscript.
ReferencesBergstrom M, Falk P, Holmdahl L. CO2 promotes plasminogen activator
inhibitor type 1 expression in human mesothelial cells. Surg Endosc2003;17:1818–1822.
1244 Mynbaev and Corona
at KU
Leuven on July 6, 2011hum
rep.oxfordjournals.orgD
ownloaded from
Bergstrom M, Falk P, Park P-O, Holmdahl L. Peritoneal and systemic pHduring pneumoperitoneum with CO2 and helium in a pig model. SurgEndosc 2008;22:359–364.
Bide RW, Armour SJ, Yee E. (1997) Estimation of Human Toxicity FromAnimal Inhalation Toxicity Data: 1. Minute Volume-Body WeightRelationships Between Animals And Man. Medicine Hat, Alberta, Canada:Defence Research Establishment Suffield, Box 4000, 1997. No.19980129 033, http://handle.dtic.mil/100.2/ADA336351.
Binda MM, Molinas CR, Mailova K, Koninckx PR. Effect of temperatureupon adhesion formation in a laparoscopic mouse model. Hum Reprod2004;19:2626–2632.
Binda MM, Molinas CR, Hansen P, Koninckx PR. Effect of desiccation andtemperature during laparoscopy on adhesion formation in mice. FertilSteril 2006;86:166–175.
Bourdel N, Matsuzaki S, Bazin JE, Pouly JL, Mage G, Canis M. Peritonealtissue-oxygen tension during a carbon dioxide pneumoperitoneum ina mouse laparoscopic model with controlled respiratory support.Hum Reprod 2007;22:1149–1155.
Brundell SM, Tsopelas C, Chatterton B, Touloumtzoglou J, Hewett PJ.Experimental study of peritoneal blood flow and insufflation pressureduring laparoscopy. Br J Surg 2002;89:617–622.
Demirbag S, Yesildaglar N, Gulec B, Erogul O, Pekcan M. The addition ofO2 to the CO2 does not prevent the systemic effects of the CO2pneumoperitoneum in a rabbit model. Surg Laparosc Endosc PercutanTech 2005;15:271–274
Demirbag S, Yesildaglar N, Gulec B, Erogul O, Pekcan M. Author reply.The addition of O2 to the CO2 does not prevent the systemiceffects of the CO2 pneumoperitoneum in a rabbit model. SurgLaparosc Endosc Percutan Tech 2006;16:125–126.
de Souza AM, Wang CC, Chu CY, Lam PM, Rogers MS. The effect ofintra-abdominal pressure on the generation of 8-iso prostaglandinF2alpha during laparoscopy in rabbits. Hum Reprod 2003;18:2181–2188.
Elkelani OA, Binda MM, Molinas CR, Koninckx PR. Effect of adding more than3% oxygen to carbon dioxide pneumoperitoneum on adhesion formation ina laparoscopic mouse model. Fertil Steril 2004;82:1616–1622.
Goitein D, Papasavas P, Yeaney W, Gagne D, Hayetian F, Caushaj P,Keenan R, Landreneau R. Microsphere intestinal blood flow analysisduring pneumoperitoneum using carbon dioxide and helium. SurgEndosc 2005;19:541–545.
Gottrup F. Oxygen in wound healing and infection. World J Surg 2004;28:312–315.
Greif R, Akca O, Horn EP, Kurz A, Sessler DI. Supplemental perioperativeoxygen to reduce the incidence of surgical-wound infection. OutcomesResearch Group. N Engl J Med 2000;342:161–167.
Gutt CN, Oniu T, Schemmer P, Mehrabi A, Buehler MW. Feweradhesions. Surg Endosc 2004;18:898–906.
Karhausen J, Haase VH, Colgan SP. Inflammatory hypoxia: role ofhypoxia-inducible factor. Cell Cycle 2005;4:256–258.
Kazama T, Ikeda K, Sanjo Y. Comparative carbon dioxide output throughinjured and noninjured peritoneum during laparoscopic procedures.J Clin Monit Comput 1998;14:171–176.
Kokura S, Yoshida N, Yoshikawa T. Anoxia/reoxygenation-inducedleukocyte-endothelial cell interactions. Free Radic Biol Med 2002;33:427–432.
Majewski WD. Fewer adhesions after laparoscopic surgery? Surg Endosc2006;20:999.
Matsuzaki S, Canis M, Bazin J-E, Darcha C, Pouly J-L, Mage G. Effects ofsupplemental perioperative oxygen on post-operative abdominalwound adhesions in a mouse laparotomy model with controlledrespiratory support. Hum Reprod 2007;22:2702–2706.
Miyano G, Yamataka A, Doi T, Okawada M, Takano Y, Kobayashi H,Lane GJ, Miyano T. Carbon dioxide pneumoperitoneum preventsintraperitoneal adhesions after laparotomy in rats. J Pediatr Surg 2006;41:1025–1028.
Molinas CR, Koninckx PR. Hypoxaemia induced by CO(2) or heliumpneumoperitoneum is a co-factor in adhesion formation in rabbits.Hum Reprod 2000;15:1758–1763.
Molinas CR, Mynbaev OA, Pauwels A, Novak P, Koninckx PR. Peritonealmesothelial hypoxia during pneumoperitoneum is a cofactor inadhesion formation in a laparoscopic mouse model. Fertil Steril 2001;76:560–567.
Molinas CR, Campo R, Dewerchin M, Eriksson U, Carmeliet P,Koninckx PR. Role of vascular endothelial growth factor and placentalgrowth factor in basal adhesion formation and in carbon dioxidepneumoperitoneum-enhanced adhesion formation after laparoscopicsurgery in transgenic mice. Fertil Steril 2003a;80:803–811.
Molinas CR, Elkelani O, Campo R, Luttun A, Carmeliet P, Koninckx PR.Role of the plasminogen system in basal adhesion formation andcarbon dioxide pneumoperitoneum-enhanced adhesion formationafter laparoscopic surgery in transgenic mice. Fertil Steril 2003b;80:184–192.
Molinas CR, Tjwa M, Vanacker B, Binda MM, Elkelani O, Koninckx PR.Role of CO2 pneumoperitoneum-induced acidosis in CO2pneumoperitoneum-enhanced adhesion formation in mice. Fertil Steril2004;81:708–711.
Mullett CE, Viale JP, Sagnard PE, Miellet CC, Ruynat LG, Counioux HC,Motin JP, Boulez JP, Dargent DM, Annat GJ. Pulmonary CO2elimination during surgical procedures using intra- or extraperitonealCO2 insufflation. Anesth Analg 1993;76:622–626.
Mynbaev OA. Re: the addition of O2 to the CO2 does not prevent thesystemic effects of the CO2 pneumoperitoneum in a rabbit model.Surg Laparosc Endosc Percutan Tech 2006;16:123–125.
Mynbaev OA, Molinas CR, Adamyan LV, Vanacker B, Koninckx PR.Pathogenesis of CO(2) pneumoperitoneum-induced metabolichypoxemia in a rabbit model. J Am Assoc Gynecol Laparosc 2002a;9:306–314.
Mynbaev OA, Molinas CR, Adamyan LV, Vanacker B, Koninckx PR.Reduction of CO(2)-pneumoperitoneum-induced metabolichypoxaemia by the addition of small amounts of O(2) to the CO(2)in a rabbit ventilated model. A preliminary study. Hum Reprod 2002b;17:1623–1629.
Mynbaev OA, Koninckx PR, Bracke M. A possible mechanism ofperitoneal pH changes during carbon dioxide pneumoperitoneum(letter). Surg Endosc 2007;21:489–491.
Mynbaev OA, Adamyan LV, Mailova K, Vanacker B, Koninckx PR. Effectsof adding small amounts of oxygen to a CO2-pneumoperitoneum ofincreasing pressure in rabbit ventilation models. Fertil Steril 2008;September 26 (Epub ahead of print).
Nagelschmidt M, Gerbecks D, Minor T. The impact of gas laparoscopy onabdominal plasminogen activator activity. Surg Endosc 2001;15:585–588.
Neto FJ, Luna SP, Massone F, Thomassian A, Vargas JL, Junior JR, D’UtraVaz BB, Crocci AJ. The effect of changing the mode of ventilation on thearterial-to-end-tidal CO2 difference and physiological dead space inlaterally and dorsally recumbent horses during halothane anesthesia.Vet Surg 2000;29:200–205.
Ragheb J, Buggy DJ. Editorial III: tissue oxygen tension (PTO2)in anaesthesia and perioperative medicine. Br J Anaesth 2004;92:464–468.
Sauerland S, Neugebauer E. Fewer adhesions after laparoscopic surgery?Surg Endosc 2005;19:862.
Role of design, methodology and animal models 1245
at KU
Leuven on July 6, 2011hum
rep.oxfordjournals.orgD
ownloaded from
Schugart RC, Friedman A, Zhao R, Sen CK. Wound angiogenesis as afunction of tissue oxygen tension: a mathematical model. Proc NatlAcad Sci USA 2008;105:2628–2633.
Yavuz Y, Ronning K, Lyng O, Gronbech JE, Marvik R. Effect of carbondioxide pneumoperitoneum on tissue blood flow in the peritoneum,rectus abdominis, and diaphragm muscles. Surg Endosc 2003;17:1632–1635.
Yesildaglar N, Koninckx PR. Adhesion formation in intubated rabbitsincreases with high insufflation pressure during endoscopic surgery.Hum Reprod 2000;15:687–691.
Ziprin P, Ridgway PF, Peck DH, Darzi AW. Laparoscopic-typeenvironment enhances mesothelial cell fibrinolytic activity in vitro via adown-regulation of plasminogen activator inhibitor-1 activity. Surgery2003;134:758–765.
Zuurbier CJ, Emons VM, Ince C. Hemodynamics of anesthetized ventilatedmouse models: aspects of anesthetics, fluid support, and strain. Am JPhysiol Heart Circ Physiol 2002;282:H2099–H2105.
Submitted on May 4, 2008; resubmitted on August 27, 2008; accepted onNovember 18, 2008
1246 Mynbaev and Corona
at KU
Leuven on July 6, 2011hum
rep.oxfordjournals.orgD
ownloaded from
Original Article
Effect of Upper Abdomen Tissue Manipulation on AdhesionFormation between Injured Areas in a Laparoscopic Mouse Model
Ron Schonman, MD*, Roberta Corona, MD, Adriana Bastidas, MD, MSc, Carlo De Cicco, MD,and Philippe Robert Koninckx, MD, PhDFrom the Departments of Obstetrics and Gynecology at the University Hospital Gasthuisberg, Leuven, Belgium (all authors) and the Chaim Sheba Medical
Center, Tel-Hashomer, Israel (Dr. Schonman).
ABSTRACT Study Objective: These experiments were designed to examine the effect of manipulation during surgery as a cofactor in
The authors have
products or comp
Corresponding au
Gynecology, Un
Belgium.
E-mail: ronschon
Submitted Novem
Available at www
1553-4650/$ - see
doi:10.1016/j.jmi
adhesion formation at trauma sites.
Design: Randomized, controlled trial. Canadian Task Force Classification–class 1.
Setting: University laboratory research center.
Subjects: A standardized laparoscopic mouse model (Balb\c mice 9–10 weeks old) for adhesion formation after opposing
bipolar lesions and 60 minutes of carbon-dioxide pneumoperitoneum. In this model adhesions are known to decrease after
the addition of 3% of oxygen, dexamethasone, or both. In addition, adhesions decrease with experience (i.e., with a decreasing
amount of manipulation during the learning curve).
Interventions: A factorial design was used to evaluate the effects of dexamethasone and of adding 3% of oxygen on manip-
ulation-enhanced adhesion formation during a learning curve. Blocks of 4 animals were thus randomized as controls (carbon-
dioxide pneumoperitoneum only) or received an additional 3% of oxygen, dexamethasone, or both. In a second experiment, the
effects of manipulation on adhesion formation were quantified. In a third experiment we evaluated whether dexamethasone had
a specific effect on manipulation-enhanced adhesion formation.
Measurements and Main Results: Qualitative and quantitative adhesion scoring 7 days after the intervention. The first ex-
periment confirmed that adhesion formation decreased during the learning curve (p ,.0001) and after the addition of dexameth-
asone whether assessed as the total adhesion score (p ,.0001 and p 5.0009, respectively) or a quantitative score (p ,.0001 and
p ,.0001, respectively). The second experiment showed that adhesion formation increased by standardized touching and
grasping of omentum and bowels (proportion score p 5.0059 and p 5.0003, respectively) and this effect increased with
duration of touching (p 5.0301). In the third experiment, dexamethasone was confirmed to decreased adhesion formation
(p 5.0001) but this effect was not specific for manipulation-enhanced adhesion formation.
Conclusion: Manipulation of intraperitoneal organs in the upper abdomen enhances adhesion formation at trauma sites,
confirming that the peritoneal cavity is a cofactor in adhesion formation. Dexamethasone decreases adhesion formation but
the effect is not specific for manipulation-enhanced adhesion formation. Journal of Minimally Invasive Gynecology (2009)
16, 307–12 � 2009 AAGL. All rights reserved.
Keywords: Adhesion formation; Tissue manipulation; Learning curve; Dexamethasone; Laparoscopy; Animal model
Postoperative adhesion formation remains an important
clinical problem causing small bowel obstructions, chronic
pain, and decreased fertility [1]. Previous abdominal surger-
ies were reported to be the cause of 33% to 50% and 79% of
no commercial, proprietary, or financial interest in the
anies described in this article.
thor: Ron Schonman, MD, Department of Obstetrics and
iversity Gasthuisberg, Herestraat 49 B-3000 Leuven,
ber 21, 2008. Accepted for publication January 8, 2009.
.sciencedirect.com and www.jmig.org
front matter � 2009 AAGL. All rights reserved.
g.2009.01.005
large and small bowel obstructions, respectively [2,3]. Adhe-
sions are believed to be an important cause of pelvic pain but
the efficacy of adhesiolysis [4] remains unclear [5]. Intraper-
itoneal (IP) adhesions are considered to be a cause of female
infertility [6] and are in 15% of the cases the only factor found
[7]. At repeated surgeries, adhesions increase difficulty and
duration of surgery and complication rates.
The widely accepted pathophysiology of adhesion forma-
tion after surgical injury to the peritoneal surfaces involves ex-
udation [8], inflammation, and fibrin deposition [9]. Fibrin is
removed by fibrinolysis and simultaneously the mesothelium
is repaired within a few days by proliferation of mesothelial
308 Journal of Minimally Invasive Gynecology, Vol 16, No 3, May/June 2009
cell islands on the injured surface [10]. If the repair process
fails, adhesions can be formed. Adhesions increase with
prolonged inflammation (i.e., infection or suture resorption)
whereas it is believed that the amount of necrotic tissue
increases inflammation and adhesions. Adhesion formation
thus was considered as a local phenomenon. It was shown
that adhesions increase with the duration of carbon-dioxide
(CO2) pneumoperitoneum, with the insufflation pressure,
and with desiccation whereas adhesions decrease with cool-
ing or after adding 3% to 4% of oxygen to the CO2 pneumo-
peritoneum [11–13]. These data show that the peritoneal
cavity is a cofactor in adhesion formation at the injured site.
The exact molecular or cellular mechanisms involved are still
unknown.
Good surgical practice is widely believed to reduce adhe-
sion formation [14] and microsurgery has claimed a reduction
of adhesions through minimizing surgical trauma and ische-
mia and by avoiding desiccation [15]. The effect of desiccation
was shown unequivocally in our mouse model [13]. The re-
duction of adhesion formation by gentle tissue handling was
not yet investigated directly. Indirect evidence is the decrease
of adhesion formation by surgical expertise, as revealed in the
rabbit nephrectomy model. Surgical expertise, however, al-
ways associates a decrease in duration of surgery with an in-
crease in quality as shown in numerous animal and human
training curves such as the pig model for paraaortic lymph
node dissection [16], pig model for nephrectomy [17], rabbit
nephrectomy model [18], and human training curves [19].
In our mouse model of adhesion formation it is well
known that the amount and the variability of adhesion forma-
tion decreases exponentially with experience during the first
50 to 80 experiments and a training curve was standard
practice for each new researcher before starting experiments.
Because our mouse model duration of CO2 pneumoperito-
neum was standardized to 60 minutes, the decrease in adhe-
sion formation with expertise should be the consequence of
a decrease in tissue manipulation. These experiments were
designed to specifically investigate the effect of tissue manip-
ulation on adhesion formation. Because the effect of tissue
manipulation might be mediated by inflammation, we
evaluated whether dexamethasone, known to have antiin-
flammatory effects, specifically could reduce manipulation-
enhanced adhesion formation.
Materials and Methods
The Laparoscopic Mouse Model for Adhesion Formation
The experimental setup (i.e., animals, anesthesia and ven-
tilation, laparoscopic surgery, and induction and scoring of IP
adhesions) was described in detail previously [11,12,20–24].
Briefly, the model consisted of pneumoperitoneum-enhanced
adhesions induced during laparoscopy by creation of oppos-
ing mechanical lesions. The pneumoperitoneum was main-
tained for 60 minutes using humidified CO2 at an
insufflation pressure of 15 mm Hg. Gas and body temperature
were kept strictly at 37�C using a heated chamber. Female
Balb\c mice, 9 to 10 weeks old and weighing 19 to 24 g,
were used because in this inbred strain adhesion formation
was high with low interanimal variability [25]. Animals
were kept under standard laboratory conditions and they
were fed with a standard laboratory diet with free access to
food and water. The study was approved by the institutional
review animal care committee.
Mice were anesthetized with IP 0.08 mg/g of pentobarbi-
tal, intubated with a 20-gauge catheter and mechanically ven-
tilated (Mouse Ventilator MiniVent, type 845, Hugo Sachs
Elektronik-Harvard Apparatus GmbH, March-Hugstetten,
Germany) using humidified room air with a tidal volume of
250 mL at 160 strokes/min to prevent cooling. To avoid var-
iability introduced by any other procedures as anesthetizing
and intubating the Balb\c mice, which are very small, the in-
vestigator was trained in this before the start of the study. The
investigator, moreover, was an experienced endoscopist.
A midline incision was performed caudal to the xyphoids,
a 2-mm endoscope with a 3.3-mm external sheath for insuf-
flation (Karl Storz, Tuttlingen, Germany) was introduced into
the abdominal cavity, and the incision was closed gas tight
around the endoscope to avoid leakage. The pneumoperito-
neum was created (Thermoflator Plus, Karl Storz) using hu-
midified insufflation gas [13,26]. After the establishment of
the pneumoperitoneum, 2 14-gauge catheters were inserted
under laparoscopic vision. Standardized 10- by 1.6-mm le-
sions were made in the antimesenteric border of both right
and left uterine horns and pelvic sidewalls with bipolar coag-
ulation, BICAP bipolar hemostasis probe BP-5200A 5F 200
cm (IMMED Benelux, Linkebeek, Belgium) at 20 W (stan-
dard coagulation mode, Autocon 200, Karl Storz, Tuttlingen,
Germany).
Adhesions were scored blindly (the investigator was not
informed of the group being evaluated) both qualitatively
and quantitatively under stereomicroscopic vision during lap-
arotomy 7 days later. The qualitative scoring system assessed
extent (0 5 no adhesions; 1 5 1%–25%; 2 5 26%–50%; 3 5
51%–75%; 4 5 76%–100% of the injured surface involved),
type (0 5 no adhesions; 1 5 filmy; 2 5 dense; 3 5 capillaries
present), and tenacity (0 5 no adhesions; 1 5 easily fall
apart; 2 5 require traction; 3 5 require sharp dissection) of
adhesions, from which a total score was calculated (extent
1 type 1 tenacity). The quantitative scoring system assessed
the proportion of the lesions covered by adhesions using the
following formula: adhesion (%) 5 (sum of the length of the
individual attachments/length of the lesion) ! 100. The re-
sults are presented as the average of the adhesions formed
at the 4 sites (right and left visceral and parietal peritoneum),
which were individually scored.
Setup and Design of the Experiments
To control temperature, animals and equipment (i.e., in-
sufflator, humidifier, water valve, ventilator, and tubing)
were placed in a closed chamber maintained at 37�C with
a heated air patient warming system (WarmTouch, model
Schonman et al. Tissue Manipulation Increases Adhesions 309
5700, Mallinckrodt Medical, Hazelwood, Mo). The insuffla-
tion gas temperature was determined by the environmental
temperature (i.e., at 37�C). Because anesthesia and ventila-
tion can influence body temperature and body temperature
can influence adhesion formation [13], the timing and tem-
perature were strictly controlled. Mice temperature was mea-
sured by rectal probe before anesthesia and was between
35�C and 37.7�C for all mice. The time of the anesthesia in-
jection was considered time 0. The animal preparation and
ventilation started after exactly 10 minutes. The pneumoper-
itoneum started at 20 minutes and was maintained for 60 min-
utes for a total of 80 minutes. In all experiments, animals
were block randomized by day (i.e., 1 animal of each group
was operated on the same day in random order) to eliminate
eventual variability caused by the day or surgery.
Experiment I (n 5 80) was performed to confirm the de-
crease in adhesion formation during sequential experiments
even when performed by an experienced endoscopist. Be-
cause duration of surgery was strictly standardized at 60 min-
utes, the only variable was decreasing manipulation with
experience. Using a factorial design, we specifically evalu-
ated in each sequential block of 4 animals the effect of dexa-
methasone and the effect of adding 3% of oxygen, because
both were proved previously to reduce adhesion formation
in this model [12,27]. In each block, mice thus were random-
ized to no additional treatment (group I); treatment with 40
mg of dexamethasone twice, the first after surgery and the
second after 24 hours (group II); the addition of 3% oxygen
to the CO2 pneumoperitoneum (group III); or both dexameth-
asone and oxygen treatment (group IV).
Experiment II (n 5 32) was designed to evaluate the effect
of touching and grasping in the upper abdomen on adhesion
formation in the lower abdomen (i.e., at the site of the injury).
Group I (n 5 8) was the control group. In groups II (n 5 8)
and IV (n 5 8) the omentum and large and small bowels were
moved gently up and down across the abdomen for 5 and 10
minutes, respectively, with the shafts of 2 graspers (i.e., tis-
sues were moved around without grasping). In group III (n
5 8), to check whether grasping was more traumatic thus
more adhesiogenic, the omentum and small and large bowels
were grasped and manipulated repetitively for 5 minutes with
a 1.5-mm atraumatic grasper.
Experiment III (n 5 32) was designed specifically to eval-
uate whether dexamethasone, known to reduce CO2 pneumo-
peritoneum-enhanced adhesions, had a specific effect on
manipulation-enhanced adhesions. A factorial design (n 5
8 animals/group) was used in which group I consisted of
the control group (i.e., CO2 pneumoperitoneum-enhanced
adhesions without any treatment). Group II received 2 doses
of 40 mg of dexamethasone, one immediately after the end of
pneumoperitoneum and another 24 hours later. In group III
a similar manipulation as in experiment II of the omentum
and large and small bowels was performed for 5 minutes
with a 1.5-mm nontraumatic grasper. Group IV received sim-
ilar manipulation for 5 minutes as group III but also received
dexamethasone in the same fashion as in group II.
Statistics
Software (Graph Pad Prism, Graph Pad Software Inc, San
Diego, Calif), was used for curve fitting. Statistical analyses
were performed with software (SAS System, SAS Institute,
Cary, NC) using nonparametric tests, because the numbers
were too small to confirm reliably a normal distribution.
For the first experiment a 3-way analysis of variance
(ANOVA) with the block number, dexamethasone, and oxy-
gen as independent variables (Proc GLM [general linear
methods]) was used to evaluate simultaneously the effects
of experience/manipulation, dexamethasone, and oxygen ad-
dition. This 3-way ANOVA was used as a simplification of
a 2-way ANOVA (factorial design of oxygen and dexameth-
asone treatment) within a repetitive measurement (sequential
blocks). In the second experiment differences in adhesions
between 2 specific groups were evaluated by Wilcoxon
rank sum test. For the third experiment a 2-way ANOVA
(Proc GLM) evaluated the effects of manipulation and of
dexamethasone. Although nonparametric tests were used
for statistical analysis, means and SD (instead of medians,
10th–90th percentiles) are given in text and figures because
they are easier to read and interpret.
Power analysis was based on the low interanimal variabil-
ity in inbred Balb\c mice (,5%) resulting in a power of 90%
to detect a difference of 4%, 7%, and 9% when comparing 2
groups of 40, 12, or 8 mice, respectively. Indeed a factorial
design, permitting a 2- or 3-way ANOVA, has almost the
same power for each variable as if the experiment were
conducted sequentially for each variable with the same
number of animals. The factorial design, moreover, permits
the exclusion of interaction between the factors investigated
[28].
In the first experiment 3 mice died immediately after sur-
gery in the first 3 blocks. In these blocks, adhesion formation
was, moreover, erratically high. Because this obviously was
a confounding factor caused by the surgeon’s limited labora-
tory experience and because statistical results did not vary
with or without these 3 blocks, the first 3 blocks were not
included in the graph or statistical analysis in this article.
Results
In experiment I, an obvious learning curve was observed
in the control group with a logarithmic decline in adhesion
formation, whether evaluated as proportion or as total amount
of adhesions (Proc GLM, p ,.0001 for both). Dexametha-
sone decreased adhesion formation whether evaluated as ad-
hesion proportion, total adhesion score, extension, type, or
tenacity (Proc GLM, p 5.0001, p 5.0009, p ,.0001, p
5.0063, and p 5.007, respectively). Of interest, the effect
of the learning curve almost disappeared with dexametha-
sone treatment (Fig. 1). The decrease in adhesion formation
with the addition of 3% of oxygen was not significant
(Proc GLM, p 5.0658) probably as a consequence of the
high variability during the learning curve.
0 5 10 15 200
10
20
30
Control
OxygenDexametasone + O2
Dexametasone
Blocks
Pro
po
rtio
n o
f ad
hesio
ns (%
)
Fig. 1. Adhesion formation during learning curve (i.e., first 80 animals). Us-
ing block, randomization of 4 sequential mice to 60 minutes of CO2 pneumo-
peritoneum only, addition of 3% of oxygen, addition of dexamethasone, or
both show that adhesions decrease with experience (p ,.0001), with addition
of dexamethasone (p ,.0001 for adhesion proportion), and with addition of
oxygen (not significant, p 5.06 for adhesion proportion).
0
10
20
30
40
50
60
70
Control Touching5min
Grasping5min
Touching10min
19.15±5.1
42±14.746.66±8.6
62.77±6.04
Pro
po
rtio
n o
f ad
hesio
ns (%
)
Fig. 2. Increase in adhesion formation in laparoscopic mouse model by ma-
nipulation of tissue either touching (p 5.0059 for adhesion proportion) or
grasping (p 5.0003 for adhesion proportion) for 5 minutes. Effect of touch-
ing is dose dependent (5 vs 10 minutes, p 5.0301 for adhesion proportion).
Bars present means and SD of each group.
40
60
80With out DexametasoneWith Dexametasone 52.7±9.3
33.4±15.2
ad
hesio
ns (%
)
310 Journal of Minimally Invasive Gynecology, Vol 16, No 3, May/June 2009
In experiment II, manipulation of omentum and bowels in-
creased adhesion formation and this effect increased with
time. Groups II, III, and IV, had higher adhesion proportion
in comparison with the control group (Wilcoxon test, p
5.0059, p 5.0003, and p 5.0003, respectively). The effect
of manipulation increased with duration (i.e., 5 vs 10 min-
utes) for adhesion proportion and adhesion extension (Proc
GLM, p 5.0301 and p 5.0241, respectively). In addition,
grasping increased adhesions but the effect was not more pro-
nounced than manipulation (Fig. 2). In none of the mice was
adhesion found in the upper abdomen.
Experiment III confirmed that manipulation increased ad-
hesion formation and that dexamethasone reduced adhesions
whether evaluated quantitatively (2-way ANOVA, p 5.0001
for both) (Fig. 3) or qualitatively. The adhesion scores for ex-
tension, type, tenacity, and total score were lower under dexa-
methasone treatment (Proc GLM, p ,.0001, p ,.0001,
p ,.0001, and p ,.0001, respectively) and were significantly
higher with grasping manipulation for 5 minutes (Proc GLM,
p ,.0001, p ,.0001, p ,.0001, and p ,.0001, respectively).
Dexamethasone decreased adhesions similarly in the control
and manipulation-enhanced adhesion groups. Manipulation
also enhanced quantitative adhesion scores when comparing
the 2 groups receiving dexamethasone (Wilcoxon, p 5
.008).
No Manipulation Manipulation0
20
24.09±6.5
12.3±4.5
Pro
po
rtio
n o
f
Fig. 3. Manipulation increases (p 5.0001 for adhesion proportion) and
dexamethasone decreases (p 5.0001 for adhesion proportion) adhesion for-
mation in laparoscopic mouse model (2-way ANOVA; proportion). Bars
present means and SD of each group.
Discussion
These results revealed that tissue manipulation in the
upper abdomen by itself enhances adhesion formation at
the site of lesions. The effect is seen with different types of
tissue manipulation such as touching and grasping, and is
dose dependent. In addition, the higher adhesion formation
at the beginning of the learning curve is consistent with
more manipulation by the inexperienced surgeon (Fig. 1)
because this was the only variable, as in these experiments
the duration of the pneumoperitoneum and of the anesthesia
was kept constant at 60 minutes. This effect of trauma
and manipulation strongly supports the concept of gentle
tissue handling as one of the important factors in adhesion
prevention.
That manipulation alone has such a profound effect on ad-
hesion formation (Fig. 2) could have far reaching clinical im-
portance. Indeed, although during surgery mobilization
cannot be avoided for exposition of structures and during dis-
section, the amount of manipulation will decrease with expe-
rience of the surgeon, as reflected also in shorter operating
times. It, moreover, suggests that during learning curves
not only is the duration and quality of surgery affected, but
Schonman et al. Tissue Manipulation Increases Adhesions 311
also the overall tissue trauma and manipulation and possibly
also human adhesion formation.
These experiments confirm and extend the model that the
peritoneal cavity is a cofactor in adhesion formation between
traumatized peritoneal areas. Although the adhesions occur
only at the lesion sites (i.e., the uterus and the abdominal
wall), the amount of adhesion formation depends on factors
in the peritoneal cavity or fluid. These experiments showing
that manipulation outside the lesions will affect adhesion for-
mation unequivocally reveal that the effect has to be medi-
ated through the peritoneal cavity or fluid. The mechanism
and the factors involved are still unknown, although they
seem to be a consequence of the trauma caused by manipula-
tion, desiccation, hypoxia, or reactive oxygen species (ROS).
Obviously this is not the peritoneal macroscopic injury, be-
cause no adhesion formed between the organs manipulated,
but only at the area of injury.
The model that the peritoneal cavity is a cofactor in adhe-
sion formation at the level of peritoneal trauma is bound to
influence our clinical concept of adhesion prevention. First,
the clinical wisdom of gentle tissue handling and careful
and bloodless surgery becomes experimentally supported.
The importance of experience and training gains another di-
mension, because in addition to reducing complication rates
and operation time it also reduced tissue manipulation. We
thus begin to understand why training and experience might
be key factors in adhesion prevention.
Clinical adhesion prevention today is limited to barriers
and flotation agents, both of which are effective. The mech-
anism of their action, however, could be interpreted differ-
ently, because barriers could also prevent peritoneal factors
from reaching the trauma sites whereas flotation agents could
attenuate these factors by massive dilution. The concept of
the peritoneal cavity as a cofactor in adhesion formation
has other far-reaching implications for adhesion prevention,
because therapy should also be oriented to minimizing or pre-
venting these peritoneal factors. In addition to quality of sur-
gery (i.e., gentle tissue handling and duration of surgery);
minimizing the mesothelial trauma of ROS, hypoxia, and
desiccation by humidification; and adding small amounts of
oxygen and cooling, other factors should be considered
(e.g., dexamethasone). In these experiments we specifically
used dexamethasone because it proved efficacious in previ-
ous experiments using this model [27], and because we antic-
ipated some inflammatory reaction after manipulation.
Moreover, in the first experiment the effect of dexamethasone
was so pronounced that we wanted to test whether the effect
might be specific for manipulation-enhanced adhesion for-
mation. Dexamethasone was confirmed to attenuate effec-
tively the adhesiogenic effect of CO2 pneumoperitoneum
plus manipulation although not more than the effect of CO2
pneumoperitoneum only. This effect thus seems not specific
for manipulation-enhanced adhesions, because the reduction
was only proportional to the control group (i.e., adhesions af-
ter manipulation and dexamethasone treatment remaining
more important than after dexamethasone only).
To estimate the clinical implications of these findings the
model used should be understood. The mouse model com-
bines minimal surgical trauma (enough to induce reliably al-
beit minimal adhesions) together with factors affecting the
entire peritoneal cavity and enhancing this posttraumatic ad-
hesion formation. In human surgery the surgical trauma is
generally much more pronounced whereas the intensity of
the other factors enhancing adhesions is variable. Today we
do not have clinical evidence of adhesion reduction by reduc-
ing these adhesion-enhancing factors (i.e., CO2 pneumoper-
itoneum, ROS, dessication, and tissue manipulation). The
importance of the latter 2 factors, however, has been postu-
lated clinically for decades. In addition, the adhesion-reduc-
ing effect of dexamethasone should be considered carefully.
Although effective in rabbits [29], efficacy was not consis-
tently confirmed in human beings [30] or monkey models
[31]. In our mouse model dexamethasone was shown to be
highly effective in reducing adhesions in a pure surgical
trauma model. When used in the adhesion enhancement
model (trauma or CO2 pneumoperitoneum) the adhesions re-
duction remains statistically significant but the effect of dexa-
methasone clearly is not specific for this adhesion
enhancement. Absence of significant adhesion reduction in
human and primate models thus is not surprising because
the enhancement factors involved in adhesion formation
were poorly controlled, whereas variability in non-inbred
strains is much higher. This interindividual variability, to-
gether with ethical constraints limiting standardization of hu-
man surgery, hampers clinical investigation. Understanding
the underlying mechanisms, however, could be important
in designing effective adhesion reduction regimens.
In conclusion, adhesion formation between traumatized
areas is enhanced by manipulation of pelvic organs. This fur-
ther supports the concept that the entire peritoneal cavity is
a cofactor in adhesion formation. Indeed, because hypoxia,
addition of oxygen, and desiccation affect both traumatized
areas and the peritoneal cavity it has been difficult to con-
clude unequivocally that the effect was through the peritoneal
cavity. These experiments, to the contrary, by revealing the
effect of manipulation in the upper abdomen at distance
from the trauma sites, show that the peritoneal cavity pro-
duces cofactor or cofactors modulating adhesion formation.
Clinically the concept of gentle tissue handling is supported,
emphasizing the importance of training and skill develop-
ment not only to reduce operating time and complications
but also to minimize unnecessary tissue manipulation.
We thank Maria Mercedes Binda (Department of Gyne-
cology and Obstetrics, Catholic University of Leuven,
Leuven, Belgium) for her guidance. We thank George
Betsas, (University of Thessaloniki, Thessaloniki, Greece)
for reviewing the manuscript.
References
1. Liakakos T, Thomakos N, Fine PM, Dervenis C, Young RL. Peritoneal
adhesions: etiology, pathophysiology, and clinical significance; recent
advances in prevention and management. Dig Surg. 2001;18:260–273.
312 Journal of Minimally Invasive Gynecology, Vol 16, No 3, May/June 2009
2. Attard JA, MacLean AR. Adhesive small bowel obstruction: epidemiol-
ogy, biology and prevention. Can J Surg. 2007;50:291–300.
3. Stricker B, Blanco J, Fox HE. The gynecologic contribution to intestinal
obstruction in females. J Am Coll Surg. 1994;178:617–620.
4. Duffy DM, diZerega GS. Adhesion controversies: pelvic pain as a cause
of adhesions, crystalloids in preventing them. J Reprod Med. 1996;41:
19–26.
5. Swank DJ, Jeekel H. Laparoscopic adhesiolysis in patients with chronic
abdominal pain. Curr Opin Obstet Gynecol. 2004;16:313–318.
6. Drake TS, Grunert GM. The unsuspected pelvic factor in the infertility
investigation. Fertil Steril. 1980;34:27–31.
7. Milingos S, Kallipolitis G, Loutradis D, et al. Adhesions: laparoscopic
surgery versus laparotomy. Ann N Y Acad Sci. 2000;900:272–285.
8. diZerega GS. Biochemical events in peritoneal tissue repair. Eur J Surg
Suppl. 1997;577:10–16.
9. diZerega GS. Peritoneum, peritoneal healing and adhesion formation.
In: diZerega GS, editor. Peritoneal Surgery. New York: Springer;
2000. p. 3–38.
10. diZerega GS, Campeau JD. Peritoneal repair and post-surgical adhesion
formation. Hum Reprod Update. 2001;7:547–555.
11. Molinas CR, Mynbaev O, Pauwels A, Novak P, Koninckx PR. Perito-
neal mesothelial hypoxia during pneumoperitoneum is a cofactor in ad-
hesion formation in a laparoscopic mouse model. Fertil Steril. 2001;76:
560–567.
12. Binda MM, Molinas CR, Mailova K, Koninckx PR. Effect of tempera-
ture upon adhesion formation in a laparoscopic mouse model. Hum
Reprod. 2004;19:2626–2632.
13. Binda MM, Molinas CR, Hansen P, Koninckx PR. Effect of desiccation
and temperature during laparoscopy on adhesion formation in mice.
Fertil Steril. 2006;86:166–175.
14. Holmdahl L, Risberg B, Beck DE, et al. Adhesions: pathogenesis and
prevention–panel discussion and summary. Eur J Surg Suppl. 1997;
577:56–62.
15. Taylor PJ, Leader A. Surgery of the fallopian tube. Int J Fertil. 1984;29:
65–72.
16. Occelli B, Narducci F, Lanvin D, LeBlanc E, Querleu D. Learning
curves for transperitoneal laparoscopic and extraperitoneal endoscopic
paraaortic lymphadenectomy. J Am Assoc Gynecol Laparosc. 2000;7:
51–53.
17. Traxer O, Gettman MT, Napper CA, et al. The impact of intense laparo-
scopic skills training on the operative performance of urology residents.
J Urol. 2001;166:1658–1661.
18. Molinas CR, Binda MM, Mailova K, Koninckx PR. The rabbit nephrec-
tomy model for training in laparoscopic surgery. Hum Reprod. 2004;19:
185–190.
19. Ghomi A, Littman P, Prasad A, Einarsson JI. Assessing the learning
curve for laparoscopic supracervical hysterectomy. JSLS. 2007;11:
190–194.
20. Elkelani OA, Binda MM, Molinas CR, Koninckx PR. Effect of adding
more than 3% of oxygen to carbon dioxide pneumoperitoneum upon ad-
hesion formation in a laparoscopic mouse model. Fertil Steril. 2004;82:
1616–1622.
21. Molinas CR, Campo R, Elkelani OA, Binda MM, Carmeliet P,
Koninckx PR. Role of hypoxia inducible factors 1alpha and 2alpha in
basal adhesion formation and in carbon dioxide pneumoperitoneum-en-
hanced adhesion formation after laparoscopic surgery in transgenic
mice. Fertil Steril. 2003;80(Suppl2):795–802.
22. Molinas CR, Campo R, Dewerchin M, Eriksson U, Carmeliet P,
Koninckx PR. Role of vascular endothelial growth factor and placental
growth factor in basal adhesion formation and in carbon dioxide pneu-
moperitoneum-enhanced adhesion formation after laparoscopic surgery
in transgenic mice. Fertil Steril. 2003;80(Suppl2):803–811.
23. Molinas CR, Elkelani O, Campo R, Luttun A, Carmeliet P,
Koninckx PR. Role of the plasminogen system in basal adhesion forma-
tion and carbon dioxide pneumoperitoneum-enhanced adhesion forma-
tion after laparoscopic surgery in transgenic mice. Fertil Steril. 2003;80:
184–192.
24. Elkelani OA, Molinas CR, Mynbaev O, Koninckx PR. Prevention of ad-
hesions with crystalloids during laparoscopic surgery in mice. J Am As-
soc Gynecol Laparosc. 2002;9:447–452.
25. Molinas CR, Binda MM, Campo R, Koninckx PR. Adhesion formation
and interanimal variability in a laparoscopic mouse model varies with
strains. Fertil Steril. 2005;83:1871–1874.
26. Glew PA, Campher MJ, Pearson K, Schofield JC, Davey AK. The effect
of warm humidified CO2 on the dissipation of residual gas following lap-
aroscopy in piglets. J Am Assoc Gynecol Laparosc. 2004;11:204–210.
27. Binda MM, Molinas CR, Bastidas A, Koninckx PR. Effect of reactive
oxygen species scavengers, antiinflammatory drugs, and calcium-chan-
nel blockers on carbon dioxide pneumoperitoneum-enhanced adhesions
in a laparoscopic mouse model. Surg Endosc. 2007;21:1826–1834.
28. Armitage P, Berry G. Factorial Designs, Statistical Methods in Medical
Research. 2nd ed. Oxford: Blackwell; 1987. p. 227–239.
29. Siegler AM, Kontopoulos V, Wang CF. Prevention of postoperative ad-
hesions in rabbits with ibuprofen, a nonsteroidal anti-inflammatory
agent. Fertil Steril. 1980;34:46–49.
30. Fayez JA, Schneider PJ. Prevention of pelvic adhesion formation by dif-
ferent modalities of treatment. Am J Obstet Gynecol. 1987;157:
1184–1188.
31. Seitz HM Jr, Schenker JG, Epstein S, Garcıa CR. Postoperative intraper-
itoneal adhesions: a double-blind assessment of their prevention in the
monkey. Fertil Steril. 1973;24:935–940.
Original Article
Intercoat Gel (Oxiplex): Efficacy, Safety, and Tissue Response ina Laparoscopic Mouse Model
Ron Schonman, MD*, Roberta Corona, MD, Adriana Bastidas, MD, MSc, Carlo De Cicco, MD,Karina Mailova, MD, and Philippe Robert Koninckx, MD, PhDFrom the Departments of Obstetrics and Gynecology at University Hospital Leuven, Campus Gasthuisberg, Leuven, Belgium (all authors) and The Chaim Sheba
Medical Center, Tel-Hashomer, Israel (Dr. Schonman).
ABSTRACT Study Objective: To study the efficacy and safety of Intercoat gel in a laparoscopic mouse model with pneumoperitoneum-
The authors have
products or comp
is a product of Fz
Corresponding au
Gynecology, The
Israel.
E-mail: ronschon
Submitted Octobe
Available at www
1553-4650/$ - see
doi:10.1016/j.jmig
enhanced adhesion formation.
Design: Randomized controlled trial. Evidence obtained from a properly designed, randomized, controlled trial (Canadian
Task Force classification I).
Setting: University laboratory research center.
Subjects: Balb\c female mice 9 to 10 weeks old.
Interventions: Two laparoscopic mouse models for adhesion formation were used. In the first model, adhesions following
bipolar opposing lesions in the pelvis were enhanced by 60 minutes of carbon-dioxide pneumoperitoneum. In the second
model, adhesions were further enhanced by bowel manipulation. The first experiment evaluated the efficacy of Intercoat in
both models. The second experiment evaluated the efficacy of Intercoat in the first model, when applied immediately on the
lesion, when applied at the end of the pneumoperitoneum, and when applied in the upper abdomen. Biopsy specimens were
taken after 7 days and were evaluated after hematoxylin-eosin and CD45 staining.
Measurements and Main Results: Qualitative and quantitative adhesion scoring. Morphology was evaluated by standard
light microscopy. In both models, Intercoat decreased adhesion formation whether applied immediately on the lesion or at
the end of the pneumoperitoneum (qualitative and quantitative scoring p ,.0001 and p ,.0001, respectively). Intercoat appli-
cation is associated with tissue redness, vascular congestion, and cellular edema but without an inflammatory reaction. Applied
in the upper abdomen, Intercoat does not increase adhesions, but decreases adhesions at higher doses (p 5.0024). Intercoat in
high doses had a toxic effect (p 5.0058).
Conclusion: Intercoat is an effective antiadhesion product. It is associated with tissue edema and vasodilatation as observed
after 7 days both macroscopically and by histology. Journal of Minimally Invasive Gynecology (2009) 16, 188–194 � 2009
AAGL. All rights reserved.
Keywords: Adhesion formation; Intercoat; Oxiplex; Laparoscopy; Animal Model
Barriers are widely used for adhesion prevention, because
adhesion formation has been viewed as a local process of
slow or incomplete fibrin degradation at the trauma site per-
mitting fibroblast proliferation instead of mesothelial repair.
Separation of peritoneal surfaces until mesothelization has
no commercial, proprietary, or financial interest in the
anies described in this article. Intercoat/Oxiplex/SP Gel
ioMed Inc, San Luis Obispo, California.
thor: Ron Schonman, MD, Department of Obstetrics and
Chaim Sheba Medical Center, Tel-Hashomer, 52621,
r 11, 2008. Accepted for publication December 18, 2008.
.sciencedirect.com and www.jmig.org
front matter � 2009 AAGL. All rights reserved.
.2008.12.014
occurred may prevent or lessen adhesion formation. The bar-
riers can be nonresorbable solid membranes, resorbable solid
membranes, or resorbable semisolid gels. Although efficacy
of nonresorbable solid membranes was elegantly proved
[1], they never became popular because a second intervention
was necessary to remove them. All resorbable barriers today
are based on carbohydrate polymers such as hyaluronic acid,
carboxymethylcellulose (CMC), polyethylene glycol (PEO),
oxidized regenerated cellulose, and polytetrafluoroethylene.
Combination of products and degree of chemical cross-
linking will result in solid membranes or gels.
Solid barriers such as Seprafilm (Genzyme, Cambridge,
MA), containing hyaluronic acid-CMC, and Interceed (John-
son & Johnson, Gynecare Unit, Somerville, NJ), oxidized
Schonman et al. Intercoat Effect in a Laparoscopic Model 189
regenerated cellulose reduced adhesion formation by some
50% in animal models and in human beings [2–8]. They
can be used adequately only in laparotomy. Semisolid gels
such as SprayGel (Confluent Surgical, Waltham, MA), using
PEO, and Hyalobarrier Gel (Baxter, Pisa, Italy), using auto-
cross-linked hyaluronic acid, seem to be equally effective
and safe. Moreover, they can be used easily during laparo-
scopic surgery [9–21]. Developed as a local adhesion preven-
tion barrier applicable at laparoscopy, Intercoat/Oxiplex/SP
Gel is composed of PEO and CMC. Manufactured as a thin
sheet, it was proved to be effective in a rabbit laparoscopy
model [22], and in the human being it reduced epidural fibro-
sis and radiculopathy after lumbar surgery [23,24]. A gel
preparation reduced adhesion formation after laparoscopic
adnexal surgery [25] in nonendometriosis and in endometri-
osis surgery [26]. These resorbable solid or semisolid barriers
appear to be safe, with few side effects, which is not surpris-
ing because they are based on natural products. Efficacy is
comparable, which also is not surprising because they are
chemically related. The pathophysiology of adhesion forma-
tion, however, is surprisingly poorly documented.
During the last decade, injury to the mesothelial cells of
the peritoneal cavity was revealed as a cofactor in adhesion
formation at the injury site [27]. This was shown for car-
bon-dioxide (CO2) pneumoperitoneum, which increases
adhesion formation as a time- and pressure-related effect
[28]. Scanning electron microscopy showed that mesothelial
cells retract with direct exposure of the extracellular mem-
brane. It was postulated to be mediated through mesothelial
hypoxia because it was prevented by adding low doses of
oxygen to the pneumoperitoneum [29] and absent in HIF1a
and HIF2a knockout mice and because it was decreased by
drugs preventing hypoxia-inducible factor induction [30].
Also, desiccation enhances adhesion formation [31] as does
mechanical manipulation of the bowel in the upper abdomen.
The mechanism of action of any product used for adhesion
prevention could thus be local at the trauma site, e.g., by
keeping the surfaces separated until reepithelialization, or
through the peritoneal cavity by preventing or reducing
mesothelial trauma or by preventing deleterious effects
from the peritoneal cavity to reach the traumatized area.
This prompted us to evaluate the last developed product,
Intercoat in our laparoscopic mouse model.
Materials and Methods
The Laparoscopic Mouse Model for Adhesion Formation
The experimental setup (i.e., animals, anesthesia and ven-
tilation, laparoscopic surgery, and induction and scoring of
intraperitoneal [IP] adhesions) was described in detail
previously [27–30,32–34]. Briefly, the model consisted of
pneumoperitoneum-enhanced adhesions induced by a me-
chanical lesion. The pneumoperitoneum was maintained for
60 minutes using pure and humidified CO2 at 15 mm Hg of
insufflation pressure. Gas and body temperatures were kept
strictly at 37�C using a heated chamber. Female Balb/c
mice, 9 to 10 weeks old and weighing 19 to 24 g, were
used because adhesion formation is high whereas the
interanimal variability is low in this inbred strain [35]. Ani-
mals were kept under standard laboratory conditions and
they were fed with a standard laboratory diet with free access
to food and water. The study was approved by the institu-
tional review animal care committee.
Mice were anesthetized with IP 0.08 mg/g of pentobarbi-
tal, intubated with a 20-gauge catheter, and mechanically
ventilated (Mouse Ventilator MiniVent, type 845, Hugo
Sachs Elektronik-Harvard Apparatus GmbH, March-
Hugstetten, Germany) using humidified room air with a tidal
volume of 250 mL at 160 strokes/min to prevent cooling.
A midline incision was performed caudal to the xyphoid,
a 2-mm endoscope with a 3.3-mm external sheath for insuf-
flation (Karl Storz, Tuttlingen, Germany) was introduced into
the abdominal cavity, and the incision was closed gas tight
around the endoscope to avoid leakage. The pneumoperito-
neum was created with the Thermoflator Plus (Karl Storz)
using humidified insufflation gas [31,36]. After the establish-
ment of the pneumoperitoneum, 2 14-gauge catheters were
inserted under laparoscopic vision. Standardized 10- by
1.6-mm lesions were performed in the antimesenteric border
of both right and left uterine horns and pelvic sidewalls with
bipolar coagulation (BICAP, bipolar hemostasis probe,
BP-5200A, 5 Fr, 200 cm; IMMED Benelux, Linkebeek,
Belgium) at 20 W (standard coagulation mode, Autocon
200, Karl Storz).
Adhesions were scored qualitatively and quantitatively
under microscopic vision by a blinded investigator during
laparotomy 7 days later. Indeed, scoring after 7 and 28
days was proved previously not to be different [27], whereas
scoring after 7 days is much more convenient for planning of
experiments. The qualitative scoring system assessed extent
(0: no adhesions; 1: 1%–25%; 2: 26%–50%; 3: 51%–75%;
4: 76%–100% of the injured surface involved), type (0: no
adhesions; 1: filmy; 2: dense; 3: capillaries present), and te-
nacity (0: no adhesions; 1: easily fall apart; 2: require trac-
tion; 3: require sharp dissection) of adhesions, from which
a total score was calculated (extent, type, tenacity). The quan-
titative scoring system assessed the proportion of the lesions
covered by adhesions using the following formula: adhesion
(%) 5 (sum of the length of the individual attachments/length
of the lesion) ! 100. The results are presented as the average
of the adhesions formed at the 4 sites (right and left visceral
and parietal peritoneum), which were individually scored.
To control temperature, animals and equipment (i.e.,
insufflator, humidifier, water valve, ventilator, and tubing)
were placed in a closed chamber maintained at 37�C (heated
air, WarmTouch, Patient Warming System, model 5700,
Mallinckrodt Medical, Hazelwood, MO). The insufflation
gas temperature was determined by the environmental tem-
perature, i.e., at 37�C. Because anesthesia and ventilation
can influence body temperature and body temperature can
influence adhesion formation [31], the timing and
190 Journal of Minimally Invasive Gynecology, Vol 16, No 2, March/April 2009
temperature were strictly controlled. Mouse temperature was
measured by rectal probe before anesthesia and was between
35�C and 37.7�C for all mice. The time of the anesthesia
injection was considered time 0. The animal preparation
and ventilation started after exactly 10 minutes. The pneumo-
peritoneum started at 20 minutes and was maintained for 60
minutes for a total time of 80 minutes.
Two models of adhesion formation were used to mimic
the clinical situation. The first model consisted of adhesion
formation following opposing bipolar lesions and 60 minutes
of pure CO2 pneumoperitoneum (CO2 pneumoperitoneum/
hypoxia-enhanced adhesions). In the second model, these
CO2 pneumoperitoneum-enhanced adhesions were further
enhanced by 5 minutes of manipulation of omentum and
bowels in the upper abdomen (hypoxia- and manipulation-
enhanced model). These models mimic the clinical situation
of a peritoneal lesion together with the effect of CO2 pneumo-
peritoneum (model 1), further enhanced by surgical manipu-
lation (model 2).
Design of the Experiments
Intercoat, a viscoelastic gel composed of polyethylene ox-
ide and carbomethylcellulose was received from Ethicon Inc
(FzioMed, Ethicon, Somerville, NJ). Intercoat was adminis-
tered with a sterile syringe and a 14-gauge catheter.
Experiment I (n 5 24) was performed to evaluate the ef-
ficacy of Intercoat gel on adhesion formation in model 1
(hypoxia-enhanced adhesion formation) and in model 2 (hyp-
oxia- and manipulation-enhanced model) and to evaluate
whether efficacy was similar in both models. A factorial
design with 6 animals in each of the 4 groups was used,
i.e., CO2 pneumoperitoneum-enhanced adhesions without
and with manipulation, each of them without and with Inter-
coat application (0.7 mL of Intercoat applied on the lesions
immediately after the lesion was made for model 1 or imme-
diately after manipulation for model 2). Animals were block
randomized by day, i.e., 1 animal of each group was operated
on the same day in random order.
Experiment II (n 5 49, 7 animals/group) was designed to
answer 2 questions. First, is the effect of Intercoat caused by
the barrier effect between the opposing lesions only, or does
Intercoat in addition attenuate the effect of hypoxia by shield-
ing the lesion from the hypoxic effect of CO2? Second, we
wanted to exclude that the inflammatory/edematous reaction
observed in the first experiment might contribute to adhesion
formation, thus Intercoat having possibly a dual effect, a bar-
rier effect reducing adhesions and an inflammatory reaction
possibly enhancing adhesions. To answer the first question,
0.7 mL of Intercoat was applied on the lesions immediately
after the lesions were made (group II) or at the end of pneu-
moperitoneum (group III) in comparison with a control group
(group I). To answer the second question, 0.2, 0.5, and 1 mL
of Intercoat were administered in the upper abdomen after
induction of the lesions (group V, VI, and VII, respectively).
In addition, in group IV, 0.7 mL of Intercoat was adminis-
tered on the lesions together with 1 mL in the upper abdomen.
Animals were block randomized by day, i.e., 1 animal of each
group was performed in 1 day, but in random order.
To evaluate tissue reaction to Intercoat, biopsy specimens
were taken from the omentum. The biopsy specimens were
embedded with JB solution (Canemco, Quebec, Canada)
and fixed in paraffin. Control samples were stained with he-
matoxylin-eosin. Samples from the treated mice were stained
with hematoxylin-eosin and immunohistochemistry staining
for CD45 with purified rat antimouse CD45, in a 3-step stain-
ing procedure with combination with biotin-conjugated rab-
bit antirat as the secondary antibody and streptavidin together
with diaminobenzidine as a detection system. Because we
looked for inflammatory reaction, we used anti-CD45, be-
cause CD45 is found on all cells of hematopoietic origin ex-
cept erythrocytes. Its presence distinguishes leukocytes from
nonhematopoietic cells.
Statistics
Statistical analyses were performed with the software
(SAS System, SAS Institute, Cary, NC) using for the first
experiment a 2-way analysis of variance (general linear
methods, proc GLM) with manipulation and Intercoat as vari-
ables. In the second experiment, differences in adhesion pro-
portions between 2 groups were evaluated by Wilcoxon test
(Graph Pad Software Inc, San Diego, CA). Correlation
between postoperative mortality and total amount of Inter-
coat administration was checked with Spearman correlation.
Power analysis was based on the low interanimal variabil-
ity in inbred Balb\c mice (,5%). The first experiment had
a power of 90% to detect a difference of 7% for each factor
because the factorial design with a 2 analysis of variance
has almost the same power for each variable as if the exper-
iment were conducted sequentially for each variable with 12
animals in each group [28]. The second experiment had
a power of 90% to detect a difference of 10% between
groups.
Results
The first experiment confirmed the increase in adhesion
formation by manipulation and the decrease by Intercoat.
Adhesion reduction was found whether evaluated as quantita-
tive adhesion proportions (p ,.0001) (Fig. 1) or as a qualita-
tive scoring of adhesion formation, i.e., total adhesion score,
extension, type, or tenacity (p ,.0001, p ,.0001, p 5.0002,
and p 5.0002, respectively, proc GLM). The reduction, more-
over, was not specific for manipulation-enhanced adhesions,
because the percent reduction in adhesions was similar in both
models, i.e., with and without manipulation.
In experiment II Intercoat similarly reduced adhesion for-
mation whether applied immediately after induction of the
lesion or at the end of pneumoperitoneum (p 5.097 and
p 5.0051, respectively, Wilcoxon test) (Fig. 2). Intercoat
applied in the upper abdomen did not increase adhesion
Fig. 1. Reduction of adhesion formation (quantitative scoring) by Intercoat
in 2 laparoscopic mouse models. In the first model, adhesions following co-
agulation lesions were enhanced by 60 minutes of CO2 pneumoperitoneum.
In second model, adhesions were further enhanced by bowel manipulation.
W/O 5 Without.
Schonman et al. Intercoat Effect in a Laparoscopic Model 191
formation; to the contrary, a decrease in adhesion proportion
was found with higher doses. This reduction was observed
for the quantitative scoring (0.2 mL vs 0.5 mL and 1 mL;
p 5.0177 and p 5.0024, respectively, Wilcoxon test) and
for the qualitative adhesion score, i.e., total score, extension,
type, and tenacity (p 5.0113, p 5.0024, p 5.0338, and
p 5.0278, respectively, proc GLM).
Intercoat in higher doses was associated with increased
mortality in mice (p 5.0058, Spearman correlation) being
85.7% when 1.7 mL was used (1 mL in the upper abdomen
and 0.7 on the lesion) versus 28.5%, 28.5%, and 0% when
1, 0.5, and 0.2 mL were used in the upper abdomen, respec-
tively, and 14.2% after 0.7 mL on the pelvic lesions. In total,
12 mice died between 4 and 7 days postoperatively.
Intercoat application in mice was associated on day 7 with
ascites and hyperemia. To evaluate whether this macroscopic
observation of hyperemia was caused by inflammation,
Fig. 2. Effect of Intercoat on adhesion formation when applied on lesion at beginn
biopsy specimens were taken in group 7 (1 mL of Intercoat),
i.e., the highest dose of Intercoat used. Because 85.7% of the
mice in group 4 (0.7 mL on the lesion and 1 mL at the upper
abdomen of Intercoat) died, this group could not be used. The
histology showed in the Intercoat-treated mice important cap-
illary dilation, hyperemia, and cellular edema of the connec-
tive tissue cells, which were absent in the control samples
(Fig. 3). Immunohistochemistry staining with CD45 that
stain leukocytes and lymphocytes did not show marked in-
flammatory response in this tissue.
Discussion
Intercoat was confirmed to effectively decrease adhesion
formation in both the laparoscopic mouse models, i.e., with
CO2 pneumoperitoneum-enhanced adhesions and in the
model of CO2- and manipulation-enhanced adhesions
(Fig. 1). The effect is proportional to the adhesions and is
not specific for either CO2-enhanced adhesions or manipula-
tion-enhanced adhesions. These results confirm and extend
the 91% reduction in adhesions by a similar CMC and PEO
film [22] in rabbits and rats. Our results also confirm and ex-
tend the previous indication of efficacy of barriers such as
SprayGel (Confluent Surgical) and Hyalobarrier Gel (Baxter)
in the pneumoperitoneum-enhanced laparoscopic adhesion
model.
Because the effect of Intercoat application on the lesion is
similar whether applied immediately after injury or at the end
of the pneumoperitoneum, we suggest that the direct effect of
the CO2 pneumoperitoneum on the lesion area is limited.
This is compatible with the hypothesis that the effect is trans-
mitted by factors released from the entire peritoneal cavity af-
ter hypoxic injury. The pathophysiology of the antiadhesive
effect of Intercoat is believed to be a consequence of its bar-
rier effect keeping the injured surfaces separated for suffi-
cient time [25]. It remains possible, however, that in
addition these barriers are also effective by preventing the
deleterious effects from the peritoneal cavity to reach the
injured areas, e.g., those induced by hypoxia, reactive oxygen
ing or end of pneumoperitoneum (PP) and when applied in upper abdomen.
Fig. 3. Histologic biopsy specimens from Intercoat-treated mouse showing omentum with severe capillary dilation (A) in comparison with control omentum
biopsy specimens (B) and edema of connective tissue cells (C). CD45 staining does not show inflammatory cells (D).
192 Journal of Minimally Invasive Gynecology, Vol 16, No 2, March/April 2009
species, desiccation, and manipulation. In addition, any semi-
liquid product could spread over the entire peritoneal cavity
and thus affect any deleterious influence from the peritoneal
cavity on adhesion formation at the injured areas. The exper-
iments with application in the upper abdomen clearly show
a dose-dependent antiadhesiogenic effect. Because 0.2 mL
in the upper abdomen is less effective, we suggest that the
effectiveness of the upper abdomen application of 0.5 and
1 mL is a consequence of spreading of the gel after surgery
by bowel and body movements.
Intercoat is a gel containing CMC and PEO (i.e., a polysac-
charide and a polyether, respectively). The added calcium
chloride creates an ion complex with PEO, determining rhe-
ology, tissue adherence, and resistance time [37]. Carboxy-
methylcellulose is tissue adhesive, thus acting as a local
barrier. The PEO has a high viscosity thus contributing to
the barrier effect. However, PEO also has a high osmotic
pressure thus increasing peritoneal fluid and tissue edema
with hyperemia. Leukocyte concentrations are reduced in
the peritoneal fluid after IP PEO treatment, but it is unclear
whether this is a mere dilution effect [38]. This osmotic effect
might also affect adhesion formation. Indeed, the osmotic
effect is associated at least with a dilution of all factors in
the peritoneal fluid whether increasing or decreasing adhe-
sion formation. The tissue edema and hyperemia could affect
tissue healing. This effect might also be responsible for the
high mortality when used in high doses in small animals
such as mice. In any case, none of the studies on Intercoat
revealed toxicity effect when given IP [25,26]. In this
experiment the doses used in mice in comparison with the
animals’ weight were much higher compared with the use
in human beings but were necessary to have sufficient cover-
age of the lesion.
The overall efficacy of Intercoat might thus be a conse-
quence of the local barrier effect of preventing deleterious
substances from the peritoneal cavity to reach the injured
area and by the osmotic effect. Our experiments unfortu-
nately do not permit quantitative differentiation between
these effects. First, mice are very small animals with obvious
spreading of the Intercoat over the peritoneal cavity. Second,
with Intercoat in our model being more than 80% effective, it
becomes difficult to show additive effects from the 3 different
potential mechanisms.
During both experiments we observed, during adhesion
scoring after 1 week, that mice treated with Intercoat had
marked redness of the peritoneal cavity and increased IP
fluid. This, to our knowledge, was not reported before. This
observation was confirmed by histology showing edema
and capillary dilatation (Fig. 3). Fortunately, we did not
find signs of inflammation whereas application of Intercoat
in the upper abdomen did not increase adhesion formation.
We, therefore, conclude that Intercoat does not cause inflam-
mation of the peritoneal cavity. We consider this very impor-
tant because Intercoat could have a double effect, i.e.,
decreasing the adhesions through a local barrier effect while
increasing adhesions through peritoneal cavity irritation.
Redness and increased peritoneal fluid 1 week after
Intercoat application was not reported before, nor was
Schonman et al. Intercoat Effect in a Laparoscopic Model 193
dose-dependent mortality in mice. Both could be considered
a reason for concern. This redness was shown to be a conse-
quence of local hyperemia and cellular edema but without
inflammatory reaction (Fig. 3). The increased IP fluid proba-
bly is also related to this osmotic effect. It is unknown,
although likely, whether a similar effect exists for other bar-
riers. The high mortality in mice after the application of
higher doses is suggested to be a consequence of this osmotic
effect and dehydration. In any case, none of the studies on
Intercoat indicated toxicity effect when given IP in human
beings [25,26]. The doses necessary in small animals such
as mice to cover the lesion indeed are relatively much higher
than in larger animals and in human beings. Whether this
local edema should be a cause of concern when applied
over bowel lesions or bowel sutures remains unknown.
In conclusion, Intercoat is proved to be an effective anti-
adhesion product in our model. It probably acts as a local bar-
rier at the level of the lesions, but it cannot be excluded that it
also reduces the deleterious effect of the peritoneal cavity on
adhesion formation. Moreover, Intercoat causes vascular
congestion and intracellular edema, probably because of
high osmotic effect of its components, an effect that also
might influence adhesion formation. Mice unfortunately are
animals too small to differentiate between these effects.
The mortality with high doses is believed to be a consequence
of the osmotic effect and should not be a concern in the
human being. Because all antiadhesive barriers today are
based on carbohydrate polymers, similar tissue effects can
be expected.
We thank Gere diZerega, Los Angeles, for reviewing the
article and for his important remarks. We thank Kathleen
Rogers, Los Angeles, and Maria Mercedes Binda, Leuven,
for reviewing the manuscript.
References
1. DeCherney AH, diZerega GS. Clinical problem of intraperitoneal post
surgical adhesion formation following general surgery and the use of
adhesion prevention barriers. Surg Clin North Am. 1997;77:671–688.
2. Kutlay J, Ozer Y, Isik B, Kargici H. Comparative effectiveness of sev-
eral agents for preventing postoperative adhesions. World J Surg. 2004;
28:662–665.
3. Ustun C, Kocak I, Akpolat I. Effects of Seprafilm (sodium hyaluranate-
based bioresorbable), Sepracoat (0.4% hyaluronic acid), and Ringer’s
lactate on the prevention of postsurgical adhesion formation in rat
models. J Obstet Gynaecol. 2000;20:78–80.
4. Eroglu A, Demirci S, Kurtman C, Akbay A, Eroglu N. Prevention of
intra-abdominal adhesions by using Seprafilm in rats undergoing bowel
resection and radiation therapy. Colorectal Dis. 2000;3:33–37.
5. Vrijland WW, Tseng LN, Eijkman HJ, et al. Fewer intraperitoneal adhe-
sions with use of hyaluronic acid-carboxymethylcellulose membrane:
a randomized clinical trial. Ann Surg. 2002;235:193–199.
6. Wallwiener M, Brucker S, Hierlemann H, Brochhausen C, Solomayer E,
Wallwiener C. Innovative barriers for peritoneal adhesion prevention:
liquid or solid? A rat uterine horn model. Fertil Steril. 2006;86:
1266–1276.
7. Nordic Adhesion Prevention Study Group. The efficacy of Interceed
(TC7)* for prevention of reformation of postoperative adhesions on ova-
ries, fallopian tubes, and fimbriae in microsurgical operations for
fertility: a multicenter study. Fertil Steril. 1995;63:709–714.
8. INTERCEED(TC7) Adhesion Barrier Study Group. Prevention of post-
surgical adhesions by INTERCEED(TC7), an absorbable adhesion bar-
rier: a prospective randomized multicenter clinical study. Fertil Steril.
1989;51:933–938.
9. Dunn R, Lyman MD, Edelman PG, Campbell PK. Evaluation of the
SprayGel adhesion barrier in the rat cecum abrasion and rabbit uterine
horn adhesion models. Fertil Steril. 2001;75:411–416.
10. Ferland R, Mulani D, Campbell PK. Evaluation of a sprayable polyeth-
ylene glycol adhesion barrier in a porcine efficacy model. Hum Reprod.
2001;16:2718–2723.
11. Johns DA, Ferland R, Dunn R. Initial feasibility study of a sprayable
hydrogel adhesion barrier system in patients undergoing laparoscopic
ovarian surgery. J Am Assoc Gynecol Laparosc. 2003;10:334–338.
12. Mettler L, Audebert A, Lehmann-Willenbrock E, Schive K, Jacobs VR.
Prospective clinical trial of SprayGel as a barrier to adhesion formation:
an interim analysis. J Am Assoc Gynecol Laparosc. 2003;10:339–344.
13. Mettler L, Audebert A, Lehmann-Willenbrock E, Schive-Peterhansl K,
Jacobs VR. A randomized, prospective, controlled, multicenter clinical
trial of a sprayable, site-specific adhesion barrier system in patients
undergoing myomectomy. Fertil Steril. 2004;82:398–404.
14. De Iaco PA, Stefanetti M, Pressato D, et al. A novel hyaluronan-based
gel in laparoscopic adhesion prevention: preclinical evaluation in an
animal model. Fertil Steril. 1998;69:318–323.
15. De Iaco PA, Muzzupapa G, Bigon E, et al. Efficacy of a hyaluronan
derivative gel in postsurgical adhesion prevention in the presence of
inadequate hemostasis. Surgery. 2001;130:60–64.
16. Belluco C, Meggiolaro F, Pressato D, et al. Prevention of postsurgical
adhesions with an autocrosslinked hyaluronan derivative gel. J Surg
Res. 2001;100:217–221.
17. Kocak I, Unlu C, Akcan Y, Yakin K. Reduction of adhesion formation
with cross-linked hyaluronic acid after peritoneal surgery in rats. Fertil
Steril. 1999;72:873–878.
18. Pellicano M, Bramante S, Cirillo D, et al. Effectiveness of autocros-
slinked hyaluronic acid gel after laparoscopic myomectomy in infertile
patients: a prospective, randomized, controlled study. Fertil Steril.
2003;80:441–444.
19. Pellicano M, Guida M, Bramante S, et al. Reproductive outcome after
autocrosslinked hyaluronic acid gel application in infertile patients
who underwent laparoscopic myomectomy. Fertil Steril. 2005;83:
498–500.
20. Acunzo G, Guida M, Pellicano M, et al. Effectiveness of auto-cross-
linked hyaluronic acid gel in the prevention of intrauterine adhesions af-
ter hysteroscopic adhesiolysis: a prospective, randomized, controlled
study. Hum Reprod. 2003;18:1918–1921.
21. Guida M, Acunzo G, Di Spiezio SA, et al. Effectiveness of auto-
crosslinked hyaluronic acid gel in the prevention of intrauterine adhe-
sions after hysteroscopic surgery: a prospective, randomized, controlled
study. Hum Reprod. 2004;19:1461–1464.
22. Rodgers KE, Schwartz HE, Roda N, Thornton M, Kobak W,
diZerega GS. Effect of Oxiplex* films (PEO/CMC) on adhesion forma-
tion and reformation in rabbit models and on peritoneal infection in a rat
model. Fertil Steril. 2000;73:831–838.
23. Rodgers KE, Robertson JT, Espinoza T, et al. Reduction of epidural
fibrosis in lumbar surgery with Oxiplex adhesion barriers of carboxy-
methylcellulose and polyethylene oxide. Spine J. 2003;3:277–283.
24. Kim KD, Wang JC, Robertson DP, et al. Reduction of radiculopathy and
pain with Oxiplex/SP gel after laminectomy, laminotomy, and discec-
tomy: a pilot clinical study. Spine. 2003;28:1080–1087.
25. Lundorff P, Donnez J, Korell M, Audebert AJ, Block K, diZerega GS.
Clinical evaluation of a viscoelastic gel for reduction of adhesions fol-
lowing gynecological surgery by laparoscopy in Europe. Hum Reprod.
2005;20:514–520.
26. diZerega GS, Coad J, Donnez J. Clinical evaluation of endometriosis
and differential response to surgical therapy with and without applica-
tion of Oxiplex/AP* adhesion barrier gel. Fertil Steril. 2007;87:
485–489.
194 Journal of Minimally Invasive Gynecology, Vol 16, No 2, March/April 2009
27. Molinas CR, Mynbaev O, Pauwels A, Novak P, Koninckx PR. Perito-
neal mesothelial hypoxia during pneumoperitoneum is a cofactor in
adhesion formation in a laparoscopic mouse model. Fertil Steril.2001;76:560–567.
28. Binda MM, Molinas CR, Mailova K, Koninckx PR. Effect of tempera-
ture upon adhesion formation in a laparoscopic mouse model. Hum
Reprod. 2004;19:2626–2632.
29. Elkelani OA, Binda MM, Molinas CR, Koninckx PR. Effect of adding
more than 3% of oxygen to carbon dioxide pneumoperitoneum upon
adhesion formation in a laparoscopic mouse model. Fertil Steril.
2004;82:1616–1622.
30. Molinas CR, Campo R, Elkelani OA, Binda MM, Carmeliet P,
Koninckx PR. Role of hypoxia inducible factors 1alpha and 2alpha in
basal adhesion formation and in carbon dioxide pneumoperitoneum-
enhanced adhesion formation after laparoscopic surgery in transgenic
mice. Fertil Steril. 2003;80:795–802.
31. Binda MM, Molinas CR, Hansen P, Koninckx PR. Effect of desiccation
and temperature during laparoscopy on adhesion formation in mice.
Fertil Steril. 2006;86:166–175.
32. Molinas CR, Campo R, Dewerchin M, Eriksson U, Carmeliet P,
Koninckx PR. Role of vascular endothelial growth factor and placental
growth factor in basal adhesion formation and in carbon dioxide
pneumoperitoneum-enhanced adhesion formation after laparoscopic
surgery in transgenic mice. Fertil Steril. 2003;80:803–811.
33. Molinas CR, Elkelani O, Campo R, Luttun A, Carmeliet P,
Koninckx PR. Role of the plasminogen system in basal adhesion forma-
tion and carbon dioxide pneumoperitoneum-enhanced adhesion forma-
tion after laparoscopic surgery in transgenic mice. Fertil Steril. 2003;80:
184–192.
34. Elkelani OA, Molinas CR, Mynbaev O, Koninckx PR. Prevention of
adhesions with crystalloids during laparoscopic surgery in mice. J Am
Assoc Gynecol Laparosc. 2002;9:447–452.
35. Molinas CR, Binda MM, Campo R, Koninckx PR. Adhesion formation
and interanimal variability in a laparoscopic mouse model varies with
strains. Fertil Steril. 2005;83:1871–1874.
36. Glew PA, Campher MJ, Pearson K, Schofield JC, Davey AK. The effect
of warm humidified CO2 on the dissipation of residual gas following lap-
aroscopy in piglets. J Am Assoc Gynecol Laparosc. 2004;11:204–210.
37. diZerega GS, Cortese S, Rodgers KE, et al. A modern biomaterial for
adhesion prevention. J Biomed Mater Res B Appl Biomater. 2007;81:
239–250.
38. Nagelschmidt M, Minor T, Saad S. Polyethylene glycol 4000 attenuates
adhesion formation in rats by suppression of peritoneal inflammation
and collagen incorporation. Am J Surg. 1998;176:76–80.
Peritoneal Infusion with Cold Saline Decreased PostoperativeIntra-abdominal Adhesion Formation: Letter to the Editor
M. M. Binda • R. Corona • J. Verguts •
P. R. Koninckx
Published online: 31 August 2010
� Societe Internationale de Chirurgie 2010
We read with interest the article by Fang et al. [1] dem-
onstrating that peritoneal infusion with cold saline for
30 min after surgery decreased adhesion formation in a
mouse model. These results indeed confirm our data
demonstrating that, in a laparoscopic mouse model, adhe-
sion formation decreases with hypothermia [2]. The data
are, moreover, consistent with our observation in mice that
intraperitoneal temperatures higher than 37�C increase
adhesion formation [3], as had already been observed after
irrigation with saline warmer than 37�C [4].
We would be most interested to have some additional
information. From the article by Fang et al., it is unclear
what the impact of cold saline infusion was upon body
temperature of the mice; i.e., to what extent the infusion
with cold saline affects the superficial layers, the abdomi-
nal cavity, or the entire body. We also were intrigued by
the results of cytokines which are nice, but difficult to
judge without standard deviations. Moreover we were
surprised that some effects lasted for more than 7 days,
although it is generally believed that peritoneal repair is
completed by postoperative day 7. Also group III is
intriguing. Because we know that desiccation occurs when
the abdomen is left open, this can be assumed to be the
driving mechanism of the increase of adhesions and of the
changes in cytokines observed. Therefore, we wonder
whether the authors observed any changes in the meso-
thelium indicating desiccation. Their data do, however,
confirm that the peritoneal cavity is a cofactor in adhesion
formation at the level of the surgical trauma [5].
Surprisingly, also in open surgery, Fang et al. did not see de
novo adhesions, thus confirming our observations in a
laparoscopic mouse model. The time courses of the cyto-
kine concentrations in group IV (desiccation only; no
surgical trauma) and group III (desiccation ? surgical
trauma) are intriguing in that opening the abdomen (and
desiccation) induced such an increase in adhesions, yet
affected cytokines only slightly.
In conclusion, we appreciate the promising results that
expand to open surgery our observations that in laparos-
copy temperature affects adhesion formation.
References
1. Fang CC, Chou TH, Lin GS et al (2010) Peritoneal infusion with
cold saline decreased postoperative intra-abdominal adhesion
formation. World J Surg 34:721–727
2. Binda MM, Molinas CR, Mailova K et al (2004) Effect of
temperature upon adhesion formation in a laparoscopic mouse
model. Hum Reprod 19:2626–2632
3. Binda MM, Molinas CR, Hansen P et al (2006) Effect of
desiccation and temperature during laparoscopy on adhesion
formation in mice. Fertil Steril 86:166–175
4. Kappas AM, Fatouros M, Papadimitriou K et al (1988) Effect of
intraperitoneal saline irrigation at different temperatures on
adhesion formation. Br J Surg 75:854–856
5. Schonman R, Corona R, Bastidas A et al (2009) Effect of upper
abdomen tissue manipulation on adhesion formation between
injured areas in a laparoscopic mouse model. J Minim Invasive
Gynecol 16:307–312
M. M. Binda (&) � R. Corona � J. Verguts � P. R. Koninckx
Department of Obstetrics and Gynaecology, University Hospital
Gasthuisberg, Katholieke Universiteit Leuven, Herestraat 49,
3000 Leuven, Belgium
e-mail: [email protected]
123
World J Surg (2011) 35:242
DOI 10.1007/s00268-010-0772-1
Letter to the Editor
Regarding: Carbon Dioxide Pneumoperitoneum PreventsPostoperative Adhesion Formation in a Rat Cecal
Abrasion Model
Jasper Verguts, MD, Roberta Corona, MD, Mercedes M. Binda, MSc, and Philippe R. Koninckx, MD, PhD
Dear Editor:
It is with great interest that we read the article on theprevention of postoperative adhesions by induction of a
carbon dioxide pneumoperitoneum prior to a laparotomy.1
The mechanism by which adhesions were prevented in thismodel was difficult to explain. Moreover, there seems to be acontradiction with the observation that CO2 pneumoper-itoneum enhances adhesion formation and that this en-hancement is pressure and duration dependent.
Traditionally, postoperative adhesion formation has beenconsidered a local process between two injured surfaces. Fol-lowing injury, exudation and fibrin deposition occur; simulta-neously, repair mechanisms are started. If this fibrin is notremoved within 2–3 days, the fibrin forms a scaffold for fibro-blasts, activated by the repair mechanisms, to form adhesions.
We demonstrated over the last decade that this local phe-nomenon is essential for adhesion formation, but quantita-tively not so important. Factors from the entire peritonealcavity can stimulate this local process of adhesion formationup to some 20-fold. It was demonstrated that factors thatdamage the mesothelial cells lining the peritoneal cavity havethis stimulating effect upon adhesion formation.2 Identified sofar as stimulating factors are mesothelial hypoxia (e.g., byusing CO2 for pneumoperitoneum), mesothelial hyperoxia(e.g., by exposing the mesothelial cells to air with a pO2 of150 mm Hg), desiccation, and direct trauma.3–5 These factors,alone or combined, result in an acute inflammation of theentire peritoneal cavity, as demonstrated by biopsies, and theseverity of this acute inflammation determines the degree ofadhesion formation enhancement (Corona et al. Submitted).At a lower peritoneal/mesothelial temperature than 37�C,this inflammatory reaction is decreased, that is, the meso-thelial layer is more resistant to damage.
As manipulation of the bowels in the upper abdomen af-fects in a dose-dependent way, the acute inflammatory reac-tion in the entire cavity and simultaneously the adhesionformation between injured areas in the lower abdomen, it isclear that the effect is one involving the entire peritonealcavity and not only a local effect on the injured areas. As allother factors obviously affect the entire peritoneal cavity and
thus also the traumatized area, it could have been theoreti-cally argued that the effect is localized only at the injured areaand that the acute inflammation is an epiphenomenon. Themanipulation experiments in the upper abdomen, enhancingadhesions at distance in the lower abdomen, and the acuteinflammatory reaction as a common mechanism for all factorsunequivocally demonstrate that the detrimental end benefi-cial effects upon adhesion formation are mediated through aneffect upon the entire peritoneal cavity lining.
A possible mechanism by which it might be explained thatCO2 pneumoperitoneum decreases adhesion formation in-duced by laparotomy is the ischemic preconditioning. Thishas been studied over the last two decades in coronary oc-clusion models, wherein ischemia and reperfusion may acti-vate a cascade of events leading to the death of myocardialcells. There is agreement that reactive oxygen species pro-duction by the mitochondria is an essential part in the pro-tective mechanism of ischemic preconditioning.6 Even briefischemic preconditioning puts the cardiac cells in a protectivephenotype for several hours, but this mechanism is not wellunderstood today. This mechanism of protection is only ex-erted in the reperfusion phase, not in the ischemic phase. In-deed, we noted that reactive oxygen species (ROS) scavengersprotected against adhesion formation in a laparoscopic mousemodel exposed to a carbon dioxide pneumoperitoneum.7
In conclusion, we suggest that the ischemic condition of acarbon dioxide pneumoperitoneum stimulates ROS pro-duction, which will have a protective effect in the reperfusionphase, which obviously occurs during laparotomy afterlaparoscopy.
References
1. Kece C, Ulas M, Ozer I, Ozel U, Bilgehan A, Aydog G, Dalgic T,Oymaci E, Bostanci B. Carbon dioxide pneumoperitoneumprevents postoperative adhesion formation in a rat cecal abra-sion model. J Laparoendosc Adv Surg Tech A 2010;20:25–30.
2. Schonman R, Corona R, Bastidas JA, Koninckx PR. Effect oftissue manipulation upon adhesion formation in a laparoscopicmouse model. J Minim Invasive Gynecol 2009;16:307–312.
3. Molinas CR, Mynbaev O, Pauwels A, Novak P, Koninckx PR.Peritoneal mesothelial hypoxia during pneumoperitoneum is
Department of Obstetrics and Gynecology, UZ Leuven, Leuven, Belgium.
JOURNAL OF LAPAROENDOSCOPIC & ADVANCED SURGICAL TECHNIQUESVolume --, Number -, 2011ª Mary Ann Liebert, Inc.DOI: 10.1089/lap.2011.0010
1
a cofactor in adhesion formation in a laparoscopic mousemodel. Fertil Steril 2001;76:560–567.
4. Elkelani OA, Binda MM, Molinas CR, Koninckx PR. Effect ofadding more than 3% oxygen to carbon dioxide pneumo-peritoneum on adhesion formation in a laparoscopic mousemodel. Fertil Steril 2004;82:1616–1622.
5. Binda MM, Molinas CR, Hansen P, Koninckx PR. Effect ofdesiccation and temperature during laparoscopy on adhesionformation in mice. Fertil Steril 2006;86:166–175.
6. Yang X, Cohen MV, Downey JM. Mechanism of cardiopro-tection by early ischemic preconditioning. Cardiovasc DrugsTher 2010;24:225–234.
7. Binda MM, Molinas CR, Bastidas A, Koninckx PR. Effect ofreactive oxygen species scavengers, antiinflammatory drugs,
and calcium-channel blockers on carbon dioxide pneumo-peritoneum-enhanced adhesions in a laparoscopic mousemodel. Surg Endosc 2007;21:1826–1834.
Address correspondence to:Jasper Verguts, MD
Department of Obstetrics and GynecologyUZ Leuven
Herestraat 493000 Leuven
Belgium
E-mail: [email protected]
2 LETTER TO THE EDITOR
Title: The effect of blood on adhesion formation and its prevention with the addition of 10%
N2O to the CO2 pneumoperitoneum in a laparoscopic mouse model
(Submitted to Human Reproduction)
1,3 Roberta Corona M.D., 2Karina Mailova M.D., 1Maria Mercedes Binda Ph.D., 1Jasper Verguts
M.D. and 1Philippe R. Koninckx M.D., Ph.D.
Affiliation
1Department of Obstetrics and Gynaecology, University Hospital Gasthuisberg, Katholieke
Universiteit Leuven, Leuven, Belgium.
2 Department of Reproductive Medicine and Surgery, Moscow State University of Medicine and
Dentistry, Moscow, Russia
3 Corresponding author:
Roberta Corona, U.Z. Gasthuisberg, Dept. Of Obstetrics & Gynaecology
Herestraat 49, B3000 Leuven, Belgium,
Tel. +32 16.34.08.28, Secretary: Tel. +32 16.34.46.34 – Fax +32 16.34.42.05.
email: [email protected], [email protected]
Abstract Objective: To evaluate the effect of blood, plasma and red blood cells on adhesion formation and prevention by peritoneal cavity conditioning through the addition of N2O or O2 to the CO2 pneumoperitoneum (PP). Interventions: Prospective RCT in Balb/c mice. In all experiments, the effect of blood, red blood cells (RBC) or plasma was standardized by intraperitoneal injection after adhesion induction. A 100% CO2 PP was used and adhesions were enhanced by injection of 1ml of blood, plasma or RBCs. A second experiment evaluated the dose response by adding different amounts of blood. Simutaneously, the adhesion reducing effect of using CO2 with 4%O2, 10%N2O, or both gases was evaluated. Results: Adhesions increased significantly after adding 1ml of blood, plasma or RBCs (p<0.0001, p<0.0001, p=0.0103). Blood was more adhesiogenic than plasma or RBC (p<0.0001) and plasma increased adhesions more than RBCs (p<0.0001). Adhesions increased by adding only 0.125ml of blood (p<0.0001) and this increased exponentially up to 0.5ml. In the control group, adhesions were lower when instead of CO2 PP, CO2 + 4%O2, 10%N2O or both were used (p=0.009, p<0.0001, p<0.0001). Conclusions: These data demonstrate the adhesiogenic effect of blood and the anti-adhesiogenic effect of using CO2 PP + 10%N20 or + 4%O2. These data extend the concept of the role of acute inflammation and support the importance of good surgical practice with little bleeding and peritoneal cavity conditioning in adhesion prevention. Keywords: blood, laparoscopy, adhesions, mouse model, pneumoperitoneum, surgery. Running title: Blood increases adhesion formation Capsule: Blood in the abdominal cavity has a strong adhesiogenic effect that can be prevented by adding 10% of N20 and/ or 4% of O2 to the CO2 pneumoperitoneum for laparoscopic surgery. The data of this study support the importance of good surgical practice with little bleeding and peritoneal cavity conditioning as the cornerstones of adhesion prevention.
Introduction Abdominal surgery has always been associated with postoperative adhesions in 60-90% of patients (1). Although laparoscopy has been believed to be less adhesiogenic compared to laparotomy this remains unclear and the risk of adhesion-related complications seems similar (2). The pathophysiology of adhesion formation is traditionally viewed as a local phenomenon, resulting from the surgical trauma to the peritoneal surfaces involving besides the mesothelial cells also the basal membrane and the subendothelial connective tissue. This leads to a local inflammatory reaction and a cascade of events, such as exudation (3), fibrin deposition and capillary growth at the site of injury (4). The extent of adhesions depends on the balance between the rate of natural healing and fibrinolysis while simultaneously the peritoneal repair process is started (5). These cascade of events at the trauma site leading to adhesions or repair are modulated by the degree of acute inflammation in the entire peritoneal cavity (6). Identified factors increasing this acute inflammation are desiccation and mechanical trauma as evidenced by manipulation of bowels in the upper abdomen (7) or by the manipulation effect during the learning curve (8). In addition, acute inflammation increases with the duration and pressure of CO2 pneumoperitoneum (PP) through mesothelial hypoxia or when CO2 PP with more than 10% of oxygen is used, through hyperoxia and reactive oxygen species (ROS). Identified factors which decrease acute inflammation are the prevention of desiccation by using humidified gas, gentle tissue handling as evidenced by the decreasing adhesions during the learning curve, and a physiologic mesothelial partial oxygen pressure around 30 mm of Hg by adding 4% of O2 to the CO2 PP. In addition, acute inflammation and adhesion formation are lower when the mesothelial temperature is lower (9), after the administration of dexamethasone and most importantly by the addition of as little as 5% of N2O to the CO2 pneumoperitoneum (submitted). These factors minimizing acute inflammation can be summarized as peritoneal conditioning. Blood and fibrin deposition are believed to enhance adhesion formation, based upon the observations that a bleeding enhances adhesion formation (10,11) and that adhesions decrease with the use of a blood stopper (12). Consistent
with this, , heparin has been added to the rinsing solution without evidence, however, that this effectively reduces adhesion formation in the human. In animal models, trauma with bleeding has been used widely as a method to induce postoperative adhesions (13). Based upon the role of fibrin and fibrinolysis in adhesion formation, good haemostasis during surgery has become considered of prime importance to decrease adhesion formation. Direct evidence, however, is to the best of our knowledge lacking, which is not surprised considering that blood and fibrin deposition could be important not only at the lesion site but blood could also induce acute inflammation of the entire peritoneal cavity Therefore, this study was designed to evaluate in our laparoscopic mouse model the effect of blood, or of its constituents upon adhesion formation when added to the peritoneal cavity. In addition, we wanted to confirm that decreasing acute inflammation through peritoneal conditioning also decreased adhesion formation induced by blood. Materials and methods The laparoscopic mouse model for adhesion formation. The model has been validated extensively and the experimental setup (animals, anesthesia, ventilation, laparoscopic surgery, induction and adhesion scoring) was described in detail previously (9,14-18). Briefly, the model consisted of performing 2 bipolar lesions during 60 min of laparoscopy using pure CO2 pneumoperitoneum. The pneumoperitoneum was induced using the Thermoflator (Karl Storz, Tuttlingen, Germany) through a 2 mm endoscope with a 3.3 external sheath for insufflation (Karl Storz, Tuttlingen Germany) introduced into the abdominal cavity through a midline incision caudal to the xyphoid appendix. The incision was closed gas tight around the endoscope in order to avoid leakage. The insufflation pressure was 15 mmHg and humidified gas (Humidifier 204320 33, Karl Storz, Tuttlingen, Germany) was used. After the establishment of the pneumoperitoneum, two 14-gauge catheters (Insyte-W, Vialon, Becton Dickinson, Madrid, Spain) were inserted under laparoscopic vision. Standardized 10 mm x 1.6 mm lesions were performed in the antimesenteric border of both right and left uterine horns and in both the right and left pelvic
side walls with bipolar coagulation (20W, standard coagulation mode, Autocon 350, Karl Storz, Tuttlingen, Germany). Since temperature is critical for adhesion formation (Binda 2004), animals and equipment were placed in a closed chamber at 37°C (heated air, WarmTouch, Patient Warming System, model 5700, Mallinckrodt Medical, Hazelwood, MO). Since anaesthesia and ventilation may influence body temperature (Binda 2004), the timing between anaesthesia (T0), intubation (at 10 min, T10) and the onset of the experiment (at 20 min, T20) was strictly controlled. After 7 days, adhesions were quantitatively (proportion) and qualitatively (extent, type, tenacity) scored, blindly during laparotomy using a stereomicroscope. The entire abdominal cavity was visualized using a xyphopubic midline and a bilateral subcostal incision. After the evaluation of ports sites and viscera (omentum, large and small bowels) for de novo adhesions, the fat tissue surrounding the uterus was carefully removed. The length of the visceral and parietal lesions and adhesions were measured. Adhesions, when present, were carefully lysed to evaluate their type and tenacity. The terminology of Pouly et al was used (19), describing de novo adhesion formation for the adhesions formed at non-surgical sites, adhesion formation for adhesions formed at the surgical site. Animals. The present study was performed in (number of mice) 9-10 weeks-old female BALB/c OlaHsd mice of 18g to 20g (Harlan Laboratories B.V., Venray, The Netherlands). Animals were kept under standard laboratory conditions and diet at the animal facilities of the Katholieke Universiteit Leuven (KUL). The study was approved by the Institutional Review Animal Care Committee (n°: P040/2010). Mice blood collection. Blood was obtained from the retro-orbital senus with a Pasteur pipette or by cardiac puncture from anesthetized mice before each experiment. Blood was collected into a heparinised tube and centrifuged for 10 min at 1000 RPM at 4°C to separate plasma and RBC. The pellet containing the RBC was resuspended in 1 ml of a isotonic solution (145 mM NaCl, 1 mM CaCl, 5 mM d-glucose, 10 mM MOPS; pH 7.4). Blood or the plasma or the RBC obtained from 1 ml of blood were injected intra-peritoneally at the end of the pneumoperitoneum. Blood, RBC and plasm had been kept at 4° C untill used.
Experiment design. All experiments were block randomised by day so that at least one animals of each experimental group was done on the same day. Following some pilot observations that the addition of blood in the peritoneal cavity at the end of the pneumoperitoneum increased adhesions (Doctoral Thesis Binda MM. Catholic University of Leuven, Belgium; http://hdl.handle.net/1979/1728) the first experiment was designed to confirm that observation and to evaluate whether the adhesiogenic effect was mediated by plasma or by the red blood cells. In all the groups, PP was induced for 60 min with pure CO2 and bipolar lesions were performed. In the Group I, the control group, no blood was added. The other groups received at the end of the PP, 1 ml of blood (group II), of plasm (group III) or of RBCs coming from 1 ml of blood (group IV) (5 mice/group). The second experiment was designed to evaluate the effect of the amount of blood during 60 min of pure CO2 pneumoperitoneum in order to mimic normal laparoscopic surgery. Besides the control group without any blood added, 0.125, 0.25, 0.5 and 1 ml of blood were i.p. injected respectively ( n= 5/group). In order to confirm the effect of peritoneal conditioning upon adhesion formation, CO2 with either with 10% of N2O, or 4% of O2 and with both 10% of N2O+ 4% of O2 was compared to pure CO2 PP . In addition, the effect of adding 0.5 ml of blood upon adhesion formation was also evaluated using these three gas mixtures. Statistics. Differences were calculated with Wilcoxon/Kruskal Wallis unpaired test with the SAS System (SAS Institute, Cary, NC). Results are expressed as a mean and standard deviations. Results Experiment I demonstrated that adhesions increased by the addition of blood, plasma or RBCs in comparison to the control group (p < 0. 0001, p < 0. 0001 and p=0. 0103, respectively). Blood is more adhesiogenic than plasma or RBC only (p < 0. 0001 for both comparisons) and plasma increases adhesions more then RBCs (p < 0. 0001). The dose response curve demonstrated that as little as 0.125 ml of blood increased adhesions (p < 0. 0001), and that the adhesiogenic effect
increased exponentially with the amount of blood up to 0.5ml (p < 0. 0001 for every amount). Adding 1 ml of blood in comparison with 0.5 ml had little additive effect (NS). In the control group we confirmed the adhesion reducing effect of using for the pneumoperitoneum, CO2 with 10% of N2O, with 4% of O2, or with both 10% of N2O +4% of O2 (p < 0. 0001, p = 0. 009, and p < 0. 0001 respectively). Clearly the addition of 10% of N2O was the most effective, with little additional effect of adding also 4% of O2. Also, after the addition of 0.5ml of blood,the use of these three gas mixtures decreased adhesions in comparison with the group with 100% CO2 (p < 0. 0001 for the three comparisons). The addition of 10% of N2O was more effective than 4% of O2 with little additive effect of adding 4% of oxygen to the 10% of N2O (NS). Discussion To the best of our knowledge, this is the first experimental study detailing the effect of blood in the peritoneal cavity upon adhesion formation. The effect was clearly shown to be dose dependent with a sigmoid relationship between adhesion formation and amount of blood. As little as 0.125 ml already significantly increased adhesions, the effect increased further exponentially up to 0.5 of blood, with little higher effect of 1ml. It should be realized, however, that 0.5 and 1 ml of blood are high volumes of blood considering that the total amount of blood that can be retrieved from 1mouse is around 2 ml. The adhesiogenic effect of blood corresponded strikingly to the sum of the adhesiogenic effect of plasma and red blood cells separately. This suggests that besides the likely fibrin deposition, acute inflammation of the entire peritoneal cavity plays an important role. These data thus are consistent with previous observations that acute inflammation in the entire peritoneal cavity is an important mechanism enhancing adhesion formation between injured areas. These data moreover lend further support to the concept that he adhesion enhancing mechanism through acute inflammation of the peritoneal cavity is quantitatively much more important than the adhesions resulting from the peritoneal trauma only, which however remain a prerequisite for adhesion formation since in none of the animals de novo adhesions were found, also not after the addition of large amounts of blood. We therefore
suggest two separate roles of fibrin deposition in adhesion formation. After a peritoneal injury, exudation and local fibrin deposition occurs and if this fibrin is not removed by fibrinolysis within a few days, adhesions start to form locally. Blood and/or fibrin, reaching the peritoneal cavity in addition, cause an acute inflammation of the entire peritoneal cavity as evidenced by the pain and the rise in CRP following an intra-abdominal bleeding (20, 21). It is unclear to what extend fibrin deposition in the peritoneal cavity plays a specific role in adhesion formation besides the acute inflammation. Indeed, an overload of fibrin could decrease the availability of plasmin necessary for the local fibrinolysis between injured areas. These concepts also shed new light on the unclear and conflicting results of the effect of tPA upon adhesion formation. We previously demonstrated that replacing the pure CO2 peritoneum with CO2 with 4% of O2 (14), with 10% of N2O or with both gases had important adhesion reducing effects (submitted), that the effect of adding 10% of N2O was quantitatively much more important than adding 4% of oxygen, and that adding 4% of oxygen in addition to 10% of oxygen had a marginal (NS) or no additive effect in comparison with 10% N2O only. The effects upon adhesion formation were confirmed in this study. In addition we demonstrated that both gases, especially N2O decreased also blood enhanced adhesion formation. We believe that these effects upon adhesion formation were mediated through a decreased acute inflammation in the entire peritoneal cavity, in fact was clearly showed by our group the correlation between post-operative adhesion formation and acute inflammation6. That 4% of O2 is effective if it is considered to be a consequence of preventing the mesothelial hypoxia. The effect of N2O, especially at low concentrations is surprising (submitted), N2O could have anti-inflammatory effect on acute inflammation since N2O is effective in decreasing the post-operative pain and blocking directly one of the most precocius and important phase of the inflammatory process as the chemotaxis (22). Chemotaxis of leucocytes also clearly decreases for more than 50% with N2O administration. Suppression of chemotaxis to corneal inflammation by nitrous oxide22. It remains surprising that as little as 5 or 10% of N2O has these strong effects.
In conclusion, these data confirm and extend the concept that the acute inflammation of the entire peritoneal cavity is quantitatively the most important driving mechanism in adhesion formation, and that good surgical practice and peritoneal cavity conditioning are the cornerstones of adhesion prevention. Identified good factors today are the replacement of pure CO2 for the pneumoperitoneum with CO2 with at least 5% of N2O eventually with 10% N2O with some 3-4% of O2, the reduction of the pneumoperitoneum temperature to below 32°C 9, the prevention of desiccation using humidified gas (24), which requires cooling with a third means (25), the decreasing of the mechanical surgical trauma and, finally the avoiding of bleeding in the peritoneal cavity during surgery.
Control Blood Plasma RBC0
10
20
30
40
50
60
70
80
90
100
RBC
ControlBloodPlasma
***
***
*
*** p < 0. 0001, * p=0. 0103, t student test vs control
Prop
ortio
n of
adh
esio
ns (%
)
Figure 1: Effect of adding blood or its components to the pure CO2 pneumoperitoneum on adhesion formation. Bipolar lesions were performed during 60 min of pneumoperitoneum with 100% CO2 and 1 ml of blood, plasm or RBC were added respectively.Mean and SEM are showed.
Figure 2 : Dose response curve and effect of using different PP compositions upon adhesion formation The upper graphs shows the dose response curve of adding blood after perfoming the bipolar lesions during 60 min of pneumoperitoneum with 100% CO2. The lower graph shows the effect of replacing the 100% CO2 pneumoperitoneum with CO2 with 10% of N2O, 4% of O2 or both 10% of N2O and 4% of O2, in the control groups (without blood) and after the addition of 0.5 ml of blood; mean and SD are shown. Acknowledgments We thank Ms. Rieta Van Bree for technical help. Steven Declerck (eSaturnus NV Belgium), Stefano Angioni and Gian Benedetto Melis (University of Cagliari, Italy) are acknowledged for reviewing the manuscript. This study was supported by the Agency for Innovation, Science and Technology (Innovatie door Wetenschap en Technologie) grant number 100969.
Reference List 1. Monk,B.J., Berman,M.L. & Montz,F.J. Adhesions After Extensive Gynecologic Surgery - Clinical-Significance, Etiology, and Prevention. American Journal of Obstetrics and Gynecology 170, 1396-1403 (1994). 2. Lower,A.M. et al. Adhesion-related readmissions following gynaecological laparoscopy or laparotomy in Scotland: an epidemiological study of 24 046 patients. Hum. Reprod. 19, 1877-1885 (2004). 1. diZerega,G.S. Biochemical events in peritoneal tissue repair. Eur. J. Surg. Suppl 10-16 (1997). 2. diZerega,G.S. Peritoneal surgery. diZerega,G.S. (ed.), pp. 3-38 (Springer, New York,2000). 3. diZerega,G.S. & Campeau,J.D. Peritoneal repair and post-surgical adhesion formation. Hum. Reprod. Update. 7, 547-555 (2001). 4. Corona R., Verguts J, Schonman R, Binda MM, Mailova K, Koninckx PR. Post-operative inflammation at the abdominal cavity increases adhesion formation in a laproscopic mouse model. Fertil Steril. 2011 Mar 15;95(4):1224-8. 5. Schonman,R., Corona,R., Bastidas,A., De Cicco,C. & Koninckx,P.R. Effect of Upper Abdomen Tissue Manipulation on Adhesion Formation between Injured Areas in a Laparoscopic Mouse Model. Journal of Minimally Invasive Gynecology 16, 307-312 (2009). 6. Corona R, Verguts J, Binda MM, Molinas CR, Schonman R, Koninckx PR. The impact of the learning curve on adhesion formation in a laparoscopic mouse model. Fertil. Steril. 2011 May 21, early online. In press. 7. Binda,M.M., Molinas,C.R., Mailova,K. & Koninckx,P.R. Effect of temperature upon adhesion formation in a laparoscopic mouse model. Hum. Reprod. (2004). 8. Gadallah,M.F. et al. Relationship between intraperitoneal bleeding, adhesions, and peritoneal dialysis catheter failure: a method of prevention. Adv. Perit. Dial. 17, 127-129 (2001). 9. Gamal,E.M. et al. The influence of intraoperative complications on adhesion formation during laparoscopic and conventional cholecystectomy in an animal model. Surgical Endoscopy and Other Interventional Techniques 15, 873-877 (2001). 10. Cömert M,et al. Does intraabdominal use of Ankaferd Blood Stopper cause increased
intraperitoneal adhesions? Ulus Travma Acil Cerrahi Derg. 16, 383-389 (2010). 11. Wiseman DM, Gottlick LE, Diamond MP. Effect of thrombin-induced hemostasis on the efficacy of an absorbable adhesion barrier. J Reprod Med. 1992 Sep;37(9):766-70. 12. Elkelani,O.A., Binda,M.M., Molinas,C.R. & Koninckx,P.R. Effect of adding more than 3% oxygen to carbon dioxide pneumoperitoneum on adhesion formation in a laparoscopic mouse model. Fertil. Steril. 82, 1616-1622 (2004). 13. Molinas,C.R. et al. Role of the plasminogen system in basal adhesion formation and carbon dioxide pneumoperitoneum-enhanced adhesion formation after laparoscopic surgery in transgenic mice. Fertil. Steril. 80, 184-192 (2003). 14. Molinas,C.R. et al. Role of vascular endothelial growth factor and placental growth factor in basal adhesion formation and in carbon dioxide pneumoperitoneum-enhanced adhesion formation after laparoscopic surgery in transgenic mice. Fertil. Steril. 80 Suppl 2, 803-811 (2003). 15. Molinas,C.R. et al. Role of hypoxia inducible factors 1alpha and 2alpha in basal adhesion formation and in carbon dioxide pneumoperitoneum-enhanced adhesion formation after laparoscopic surgery in transgenic mice. Fertil. Steril. 80 Suppl 2, 795-802 (2003). 16. Molinas,C.R., Mynbaev,O., Pauwels,A., Novak,P. & Koninckx,P.R. Peritoneal mesothelial hypoxia during pneumoperitoneum is a cofactor in adhesion formation in a laparoscopic mouse model. Fertil. Steril. 76, 560-567 (2001). 17. Pouly,J.L. & Seak-San S. Peritoneal Surgery. diZerega,G.S. (ed.), pp. 183-192 (Springer, New York,2000). 18. Gupta,N., Dadhwal,V., Deka,D., Jain,S.K. & Mittal,S. Corpus luteum hemorrhage: Rare complication of congenital and acquired coagulation abnormalities. Journal of Obstetrics and Gynaecology Research 33, 376-380 (2007). 19. Teng,S.W. et al. Comparison of laparoscopy and laparotomy in managing hemodynamically stable patients with ruptured corpus luteum with hemoperitoneum. J. Am. Assoc. Gynecol Laparosc. 10, 474-477 (2003). 20. Nyerges A. Pain Mechanisms in Laparoscopic Surgery. Semin Laparosc Surg Dec;1, 215-218 (1994). 21. Kripke,B.J., Kupferman,A. & Luu,K.C. Suppression of chemotaxis to corneal inflammation by nitrous oxide. Zhonghua Min
Guo. Wei Sheng Wu Ji. Mian. Yi. Xue. Za Zhi. 20, 302-310 (1987). 22. Binda,M.M., Molinas,C.R., Hansen,P. & Koninckx,P.R. Effect of desiccation and temperature during laparoscopy on adhesion formation in mice. Fertil. Steril. 86, 166-175 (2006). 23. Corona R, Mailova K, Koninckx R, Verguts J, koninckx PR. Intra-peritoneal temperature and desiccation during endoscopic surgery. Am. J. Obstet. Gynecol. In press, (2011).
Title: The addition of 5% N2O to the CO2 pneumoperitoneum is enough to reduce adhesion formation in a laparoscopic mouse model.
(Submitted) Authors: Karina MAILOVA MD1 2, Roberta CORONA MD1 , Jasper VERGUTS MD 1, Leila ADAMYAN MD, PhD 2 3, and Philippe R. KONINCKX MD, PhD 1. Affiliations: 1 Department of Obstetrics and Gynaecology, KULeuven, University Hospital Gasthuisberg, B3000 Leuven, Belgium; 2 Department of Reproductive Medicine and Surgery, Moscow State University of Medicine and Dentistry, Delegatskaya str. 20/1 127473, Moscow, Russia 3 Department of Operative Gynaecology, Scientific Center for Obstetrics, Gynecology and Perinatology, Oparin str 4 117997, Moscow, Russia Correspondence : Prof Philippe R. Koninckx UZ Gasthuisberg, Herestraat 49, 3000 Leuven, Belgium tel : +32 16 344202 - [email protected] Financial support: This study was supported by the Agency for Innovation by Science and Technology (Innovatie door Wetenschap en Technologie) grant number 100969 Disclosure: None of the authors have a conflict of interest. Abstract word count: 150. Text word count: 2743. Abstract Objective(s): To examine the effect of adding nitrous oxide (N2O) to the carbon dioxide (CO2) pneumoperitoneum upon postoperative adhesions. Study Design: Randomized controlled trial. University laboratory research center. A standardized laparoscopic mouse model (BALB/c) for adhesion formation. Adhesions were assessed 7 days later. First 100% N2O pneumoperitoneum was compared to 100% CO2 pneumoperitoneum. In a second experiment, different proportions of N2O in CO2 were evaluated. Results: The first experiment showed that adhesions decreased when pure nitrous oxide was used for the pneumoperitoneum (p = 0,009). The second experiment confirmed this effect of N2O upon adhesion formation, and demonstrated that as little as 5% of N2O in CO2 was sufficient to dramatically decrease adhesion formation (P = 0,0138). Conclusion: The addition of nitrous oxide to the CO2 pneumoperitoneum reduces adhesion formation and the effect is reached with as little as 5% of N2O. Keywords: Adhesions; Nitrous oxide; Laparoscopy; Animal model; Pneumoperitoneum. Running title: N2O reduces adhesion formation. Capsule: Adding as little as 5% of N2O to the CO2 pneumoperitoneum effectively decreases adhesion formation in a laparoscopic mouse model.
Introduction. Adhesion formation remains an important clinical problem causing small bowel obstructions, chronic pain, and decreased fertility (1-4). The pathophysiology of adhesion formation, is traditionally viewed as a local phenomenon, resulting from the surgical trauma to opposing peritoneal surfaces leading to a local inflammatory reaction, exudation (5) and fibrin deposition. If fibrin is not removed rapidly this is a scaffold for capillary and fibroblast growth. Repair of the mesothelium is rapid, within a few days, by proliferation of mesothelial cell islands on the injured surface (6). If the repair mechanisms fails adhesions will form. The exact role of inflammation, fibrinolysis, plasmin activation and PAI’s , local macrophages and their secretion products and the overall oxygenation of the area or the absence thereof, driving angiogenesis, fibroblast proliferation and mesothelial repair are unknown. During the last decade the peritoneal cavity has been shown to be a cofactor in adhesion formation. So far identified as adhesion enhancing factors are CO2 pneumoperitoneum, dessication , mechanical trauma , reactive oxygen species (ROS) and an oxygen concentration over 10% (7-12). All these factors are temperature dependent , higher temperatures increasing adhesions, slightly lower temperatures decreasing adhesions. As driving mechanisms were identified mesothelial trauma through hypoxia (pure CO2) hyperoxia (more than 10% oxygen) or directly by mechanical trauma whereas cells are more resistant to trauma at lower temperatures. Quantitatively, judged in a standardised mouse model the effect of the peritoneal cavity is the most important since factors from the peritoneal cavity can increase adhesions at least 20 times. A trauma however, remains essential since the deleterious effects caused by dessicaion, hypoxia, hyperoxia, or slight mechanical trauma never caused de novo adhesions in these models (13). Adhesion prevention in the human has been based upon the classical model considering
adhesion formation as a local phenomenon and aims at keeping injured areas separated for at least 36 hours either by barriers or by flotation agents (14). In animal models adhesion can be reduced by preventing the deleterious effects of the peritoneal cavity or mesothelial cells by using some 4% of oxygen in CO2 for the pneumoperitoneum, adequate humidification, and lowering the temperature to some 30 ° C. In this model, if used together with products as dexamethasone, and barriers an overall efficacy over 95% is obtained. In addition a series of factors as calcium channel blockers and phospholipids were described as reducing adhesion formation (15;16). CO2 is traditionally used for the pneumoperitoneum for safety reasons since CO2 has a high solubility in water (1.45 mg/l) and a high exchange capacity in the lungs (diffusion rates across membranes 20 times greater than air) thus minimizing the risk of gas embolism (17;18). Nitrous oxide (N2O), well known in anesthesiology has been used instead of CO2, mainly in the 1970s (19). N2O is a safe gas with solubility in water (1.5 mg/l) and diffusion in the lungs similar to CO2, but its use was abandoned due to its explosion risks at least at concentrations of more than 29%. N2O nevertheless has some advantages over CO2. It is less irritant with less postoperative pain in comparison with CO2 (20). N2O has anesthetic and analgesic properties, without the cardiopulmonary side effects of CO2 (21;22). We therefore investigated the effects of N2O during laparoscopy upon adhesion formation in our laparoscopic mouse model. Indeed, if concentrations lower than 29% would be effective in reducing adhesion formation, its use might be reconsidered, given its solubility and safety is comparable to CO2 (23). Materials and Methods. The laparoscopic mouse model for adhesion formation. The experimental setup (i.e., animals, anesthesia and ventilation, laparoscopic surgery, and induction and
scoring of IP adhesions) was described in detail previously (24-29). The model consisted of CO2 pneumoperitoneum-enhanced adhesions following a mechanical bipolar lesion during laparoscopy. The pneumoperitoneum was maintained for 60 minutes using humidified CO2 or N2O or a mixture at an insufflation pressure of 16 mm Hg. Gas and body temperature were kept strictly at 37°C using a heated chamber. Animals. Female Balb/c mice, 9 to 10 weeks old and weighing 18 to 28 g, were used because in this inbred strain adhesion formation was high with low interanimal variability (30). Animals were kept under standard laboratory conditions (temperature 20°C–22°C, relative humidity 50%–60%, 14 hours light and 10 hours dark) at the Center for Laboratory Animal Care of Katholieke Universiteit Leuven (KUL). They were fed with a standard laboratory diet (MuraconG, Carsil Quality, Turnhout, Belgium) with free access to food and water. The study was approved by the Institutional Review Animal Care Committee (n°: P040/2010). Anesthesia and Endotracheal Intubation. Mice were anesthetized with IP 0.08 mg/g of pentobarbital (Nembutal, Sanofi Sante Animale, Brussels, Belgium), then abdomen was shaved and animal secured to the table in supine position. The laparoscopic surgery required intubation with a 20-gauge catheter and ventilation with a mechanical ventilator (Mouse Ventilator MiniVent, Type 845, Hugo Sachs Elektronik-Harvard Apparatus GmbH, March-Hugstetten, Germany) using humidified air with a tidal volume of 250 µl at 160 strokes/min. Laparoscopic surgery for induction of adhesions. A midline incision was performed caudal to the xyphoid, a 2-mm endoscope with a 3.3-mm external sheath for insufflation (Karl Storz, Tüttlingen, Germany) was introduced into the abdominal cavity, and the incision was closed gas tight around the endoscope to avoid leakage. Laparoscopic surgery was performed by establishing pneumoperitoneum created by insufflators (Thermoflator Plus, Karl Storz) using humidified 100% CO2 or nitrous oxide or a mixture of carbon dioxide with 50, 25, 10 or 5% of N2O (Storz Humidifier 204320 33
(Karl Storz, Tüttlingen, Germany). This was achieved using two insufflators one for CO2, one for nitrous oxide or both together. To obtain a homogenous mixture, the output of both insufflators was mixed in a mixing chamber which connected to a water valve to with free escape of gas for a continuous flow and insufflation pressure. The insufflation pressure for both insufflators was 16 mm Hg. Mixture with 5% of nitrous oxide in CO2 was achieved by 9,5 l/min of CO2 and 0,5 l/min N2O, similarly 10% of N2O by 4,5 l/min CO2 and 0,5 l/min N2O, 25% of N2O by 3,0 l/min CO2 and 1,0 l/min N2O, 50% of N2O by 1,0 l/min CO2 and 1,0 l/min N2O. After the establishment of the pneumoperitoneum, two 14-gauge catheters (Insyte-W, Vialon, Becton Dickinson, Madrid, Spain) were inserted in both right and left flanks for the working instruments under laparoscopic vision. Standardized 10- by 1.6- mm lesions were performed in the antimesenteric border of both right and left uterine horns and pelvic side walls with bipolar coagulation by bipolar haemostasis probe (20W, standard coagulation mode, Autocon 350, Karl Storz, Tüttlingen, Germany). Since adhesion formation can be influenced by body temperature (9;10), factors affecting it had to be controlled. In order to maintain constant temperature, animals and equipment (insufflator, humidifier, water valve, ventilator and tubing) were placed in a chamber at 37°C (heated air patient warming system, WarmTouch, model 5700, Mallinckrodt Medical, Hazelwood, MO). The insufflation gas temperature was determined by the environmental temperature, i.e. 37°C. Since such factors as anaesthesia and ventilation modify body temperature the timing of all interventions was strictly determined. The time of the anaesthesia injection was considered time 0. The animal preparation and ventilation started after exactly 10 minutes. The pneumoperitoneum started at 20 minutes (T20) with subsequent induction of adhesions and was maintained for 60 minutes for all experiments. Evaluation of adhesions. Adhesions were qualitatively and quantitatively blindly
scored, under microscopic vision during laparotomy 7 days after laparoscopic surgery. The qualitative scoring system assessed extent (0: no adhesions; 1: 1%–25%; 2: 26%–50%; 3: 51%–75%; 4: 76%–100% of the injured surface involved), type (0: no adhesions; 1: filmy; 2: dense; 3: capillaries present) and tenacity (0: no adhesions; 1: easily fall apart; 2: require traction; 3: require sharp dissection) of adhesions, from which a total score was calculated(extent + type + tenacity). The quantitative scoring system assessed the proportion of the lesions covered by adhesions using the following formula: adhesion (%) = (sum of the length of the individual attachments/length of the lesion) × 100. The results are presented as the average of the adhesions formed at the four sites (right and left visceral and parietal peritoneum), which were individually scored. Experimental Design. All experiments were performed using block randomization by days. Therefore, a block of animals comprising one animal of each group was always operated the same day, avoiding day-to-day variability. In addition, within a block, experiments were performed in random order. All animals had a pneumoperitoneum for 60 min at 16 mm of Hg. Experiment I (n=18) was a designed as an exploratory pilot experiment evaluating pure carbon dioxide (group I; n=9) with pure nitrous oxide (group II; n=9) upon adhesion formation when used for the pneumoperitoneum during 1 hour. In group II, two animals died during the procedure because of intubation problems. Experiment II (n=30) was designed as a dose finding experiment aimed to find the efficacy of lower concentrations of N2O upon adhesion formation. The study comprised six groups (n = 5 in each group) comparing pure CO2 (group VI) with 5%, 10%, 25%, 50% of N2O in CO2 (groups V, IV, III and II) and with pure N2O ( group I ). Statistics. Statistical analyses were performed with the SAS software (SAS
System, SAS Institute, Cary, NC) using the non-parametric Kruskal-Wallis. Results Using 100% N2O for the pneumoperitoneum instead of CO2 decreased adhesion formation. Besides the quantitative score (P= 0,012, Fig. I), also the qualitative score was decreased. Total adhesion score, type, tenacity and extent decreased from 3,0±0,3 to 1,6±0,4 (P=0,0091 ), from 10,8± to 5,5±0.6 (P=0,0228), from 10,8±1.2 to 5,6±0.4 (P=0,0253) and from 11,3±1.3 to 4,9±0.8 (P=0,0048) respectively. Experiment II (n=30) was designed as a dose finding experiment aimed to find the efficacy of lower concentrations of N2O upon adhesion formation. The study comprised six groups (n = 5 in each group) comparing pure CO2 (group I) with 5%, 10%, 25%, 50% of N2O in CO2 (groups II, III, IV and V) and with pure N2O ( group VI). Comparing in the second experiment 100% CO2 pneumoperitoneum with 5, 10, 25, 50 and 100% of N2O the proportion score of adhesion formation decreased (Fig. II) already with addition of 5% of N2O to the CO2 pneumoperitoneum and there was an high significant differense between all the groups that received in addition N2O and the group with pure CO2 (P=0,0001 for all groups). Discussion These data demonstrate for the first time that the use of N2O instead of CO2 can reduce postoperative adhesion formation in our laparoscopic mouse model. Moreover, surprisingly this effect was already obtained with as little as 5% of N2O in CO2. The effect of N2O, especially at concentrations of 5% was unexpected and cannot be explained by current knowledge. At this moment we do not have an explanation for the effect of N2O in contrast with all previous observations in our model. The effect of lower temperatures (31) could be explained by making cells more resistant to damage e.g. by hypoxia, the addition of a physiologic concentration of oxygen (4%) could be explained by correction the
mesothelial hypoxia during pure CO2 pneumoperitoneum (32) or mesothelial hyperoxia when more than 10% of oxygen was used (33). The effect of humidification was easily explained by preventing dessication, whereas gentle tissue handling should decrease mechanical trauma (34;35). The mechanism of action of N2O, however has to be different from the effect of oxygen, since the effect of even 100% of N2O is comparable to the effect of 5% , whereas oxygen concentrations above 10% clearly increase adhesions. We therefore anticipate that N2O and low concentrations of oxygen will have additive effects. This together with the investigation of the lowest effective concentration, and the effect upon acute inflammation of the peritoneal cavity, which is almost linearly correlated with the extend of adhesion formation in our models of hypoxia enhanced adhesions, hypoxia and manipulation enhanced adhesions, and hypoxia and trauma enhanced adhesions and dexamethasone (36). Nitrous oxide, commonly known as "laughing gas", is a colorless non-flammable gas used in surgery and dentistry for its anesthetic and analgesic effects. Nitrous oxide is a weak general anesthetic, so it is used as a carrier gas in a 2:1 ratio with oxygen for more powerful general anesthetic agents such as sevoflurane or desflurane. It has a MAC (minimum alveolar concentration) of 105% and a blood: gas partition coefficient of 0.46. Less than 0.004% is (minimally) metabolized in humans. Nitrous oxide is relatively non-polar, with a molecular weight of 44.013 g/mol and a high lipid solubility. As a result, it diffuses quickly into the phospholipid cell membranes. Though nitrous oxide's exact mechanism of action is still open to some conjecture it is known that it acts as an NMDA antagonist at partial pressures similar to those used in general anaesthesia. N2O affects the GABA receptor (37) but this is still controversial. It has a lower potency by acting as a positive allosteric modulator. N2O, like other volatile anesthetics, activates twin-pore potassium channels, albeit weakly and these channels are largely responsible for keeping neurons at the resting (unexcited) potential (38). Thus the studies
comparing N2O and CO2 pneumoperitoneum during laparoscopic cholecystectomy (39) and demonstrating that N2O pneumoperitoneum produced less postoperative pain and required a decreased quantity of anesthetic for the surgical procedure than did CO2 pneumoperitoneum could be connected with nitrous oxides general anesthetic’s properties. N2O at concentrations less than 29% is safe as a gas to be used for the pneumoperitoneum (40). Indeed below 29% N2O does not have an explosion risk. Above 29% N2O might maintain combustion (41). Thus if gasses as methane would escape from the intestine and if these gases got ignited by electrosurgery some explosion risk could exist. This risk exists theoretically whereas it seems supported by some reported accidents (42;43). Used at low concentrations of 10% or 5 % therefore does not constitute an explosion risk even if used together with 3% of oxygen, and its use can be considered absolutely safe. In addition the solubility in water is similar for N2O and for CO2 being 1.5 and 1.45 mg/l respectively. This contrasts sharply with the low solubility of O2 and N2 in water. This high solubility with a high exchange capacity in the lungs makes N2O a safe gas to be used during pneumoperitoneum. It was reported that postoperative pain is less with N2O pneumoperitoneum than with CO2 pneumoperitoneum, and that intraoperative ventilation and metabolic management is easier by the use of N2O pneumoperitoneum instead of CO2 pneumoperitoneum. It therefore was recommended to change to N2O in case of refractory hypercarbia and to use it as the primary insufflation gas in patients with little ventilatory reserve (44). We recently demonstrated that in women undergoing promontofixation, postoperative pain was reduced by adding 4% of oxygen to the CO2. Since in the mouse model acute inflammation of the peritoneal cavity decreased by the addition of 4% of oxygen, it is suggested that postoperative pain is at least partially mediated by acute inflammation of the entire peritoneal cavity. Since N2O was reported to decrease pain
and postoperative discomfort (45-49) we anticipate that also N2O even at concentrations of 5% will decrease the acute inflammatory reaction and postoperative pain. Resorption of N2O from the peritoneal cavity certainly when used as 10% N2O in CO2 is so low that it is not detected in the expired gas (non published observation). These observations of even low concentrations of N2O decreasing adhesion formation obviously has to be confirmed in the human. If confirmed however, the clinical implications are far reaching. First given the direct effects upon the mesothelial cells between laparoscopic surgery and laparotomy, all beneficial effects observed at laparoscopy, such as reduction in temperature, humidification, 4% of oxygen and e.g. 10% of N2O probably are applicable
to open surgery as well (paper in preparation). In addition, we expect that as demonstrated for the addition of 4% of oxygen to the CO2 pneumoperitoneum, N2O,will also reduce tumor cell implantation. In conclusion, the addition of low concentrations of N2O to the CO2 pneumoperitoneum do have important advantages in addition to adding 4% of oxygen. Besides decreasing adhesion formation it is expected to decrease postoperative pain. Since N2O has a similar density as CO2, it might beneficially be used also for flooding the operative cavity during open surgery. If confirmed the mixture CO2 + 10% of N2O + 4% of oxygen would become the gas of choice to be used for pneumoperitoneum during laparoscopy and for flooding the cavity during open surgery.
CO2 versus N2O
Fig. I. Effect of the type of the insufflation gas (CO2 vs. N2O) upon adhesion formation. Proportion of adhesions ± SD are indicated. P = 0.012.
Effect of different proportions of N2O
0 5 10 25 50 1000
5
10
15
20
25
30
Proportion of N2O (%)
Prop
ortio
n of
adh
esio
ns (%
)
F I G U R E 3 Fig. II Effect of the addition of different proportions of nitrous oxide to CO2 pneumoperitoneum on adhesion formation, scored 7 days after surgery.
Acknowledgements We do thank the long standing collaboration in endoscopic surgery between the KULeuven, Leuven Belgium and Moscow State University of Medicine and Dentistry, Scientific Center for Obstetrics, Gynecology and Perinatology, Moscow, Russia. Reference List 1. Diamond MP, Freeman ML. Clinical implications of postsurgical adhesions. Hum.Reprod.Update. 2001;7:567-76. 2. Duffy DM, diZerega GS. Adhesion controversies: pelvic pain as a cause of adhesions, crystalloids in preventing them. J.Reprod.Med. 1996;41:19-26. 3. Ellis H. The magnitude of adhesion related problems. Ann.Chir Gynaecol. 1998;87:9-11. 4. Liakakos T, Thomakos N, Fine PM, Dervenis C, Young RL. Peritoneal adhesions: etiology, pathophysiology, and clinical significance. Recent advances in prevention and management. Dig.Surg. 2001;18:260-73. 5. diZerega GS. Biochemical events in peritoneal tissue repair. Eur.J.Surg.Suppl 1997;10-16. 6. diZerega GS, Campeau JD. Peritoneal repair and post-surgical adhesion formation. Hum.Reprod.Update. 2001;7:547-55. 7. Binda MM, Molinas CR, Hansen P, Koninckx PR. Effect of desiccation and temperature during laparoscopy on adhesion formation in mice. Fertil.Steril. 2006;86:166-75. 8. Elkelani OA, Binda MM, Molinas CR, Koninckx PR. Effect of adding more than 3% oxygen to carbon dioxide pneumoperitoneum on adhesion formation in a laparoscopic mouse model. Fertil.Steril. 2004;82:1616-22. 9. Binda MM, Molinas CR, Mailova K, Koninckx PR. Effect of temperature upon adhesion formation in a laparoscopic mouse model. Hum.Reprod. 2004;19:2626-32. 10. Binda MM, Molinas CR, Koninckx PR. Reactive oxygen species and adhesion formation: clinical implications in adhesion prevention. Hum.Reprod. 2003;18:2503-07. 11. Molinas CR, Mynbaev O, Pauwels A, Novak P, Koninckx PR. Peritoneal mesothelial
hypoxia during pneumoperitoneum is a cofactor in adhesion formation in a laparoscopic mouse model. Fertil.Steril. 2001;76:560-67. 12. Schonman R, Corona R, Bastidas A, De CC, Koninckx PR. Effect of upper abdomen tissue manipulation on adhesion formation between injured areas in a laparoscopic mouse model. J.Minim.Invasive.Gynecol. 2009;16:307-12. 13. diZerega GS, Campeau JD. Peritoneal repair and post-surgical adhesion formation. Hum.Reprod.Update. 2001;7:547-55. 14. Binda MM, Koninckx PR. Prevention of adhesion formation in a laparoscopic mouse model should combine local treatment with peritoneal cavity conditioning. Hum.Reprod. 2009;24:1473-79. 15. Ho HS, Gunther RA, Wolfe BM. Intraperitoneal carbon dioxide insufflation and cardiopulmonary functions. Laparoscopic cholecystectomy in pigs. Arch.Surg. 1992;127:928-32. 16. Uhlich GA. Laparoscopy: the question of the proper gas. Gastrointest.Endosc. 1982;28:212-13. 17. Minoli G, Terruzzi V, Tadeo G. Laparoscopy: the question of the proper gas. Gastrointest.Endosc. 1983;29:325. 18. Robinson JS, Thompson JM, Wood AW. Letter: Laparoscopy explosion hazards with nitrous oxide. Br.Med.J. 1975;4:760-61. 19. Robinson JS, Thompson JM, Wood AW. Letter: Laparoscopy explosion hazards with nitrous oxide. Br.Med.J. 1975;4:760-61. 20. Minoli G, Terruzzi V, Spinzi GC, Benvenuti C, Rossini A. The influence of carbon dioxide and nitrous oxide on pain during laparoscopy: a double-blind, controlled trial. Gastrointest.Endosc. 1982;28:173-75. 21. Hunter JG, Staheli J, Oddsdottir M, Trus T. Nitrous oxide pneumoperitoneum revisited. Is there a risk of combustion? Surg.Endosc. 1995;9:501-04. 22. Menes T, Spivak H. Laparoscopy: searching for the proper insufflation gas. Surg.Endosc. 2000;14:1050-56. 23. Tsereteli Z, Terry ML, Bowers SP, Spivak H, Archer SB, Galloway KD et al. Prospective randomized clinical trial comparing nitrous oxide and carbon dioxide
pneumoperitoneum for laparoscopic surgery. J.Am.Coll.Surg. 2002;195:173-79. 24. Elkelani OA, Binda MM, Molinas CR, Koninckx PR. Effect of adding more than 3% oxygen to carbon dioxide pneumoperitoneum on adhesion formation in a laparoscopic mouse model. Fertil.Steril. 2004;82:1616-22. 25. Binda MM, Molinas CR, Mailova K, Koninckx PR. Effect of temperature upon adhesion formation in a laparoscopic mouse model. Hum.Reprod. 2004;19:2626-32. 26. Molinas CR, Campo R, Dewerchin M, Eriksson U, Carmeliet P, Koninckx PR. Role of vascular endothelial growth factor and placental growth factor in basal adhesion formation and in carbon dioxide pneumoperitoneum-enhanced adhesion formation after laparoscopic surgery in transgenic mice. Fertil.Steril. 2003;80 Suppl 2:803-11. 27. Molinas CR, Campo R, Elkelani OA, Binda MM, Carmeliet P, Koninckx PR. Role of hypoxia inducible factors 1alpha and 2alpha in basal adhesion formation and in carbon dioxide pneumoperitoneum-enhanced adhesion formation after laparoscopic surgery in transgenic mice. Fertil.Steril. 2003;80 Suppl 2:795-802. 28. Molinas CR, Elkelani O, Campo R, Luttun A, Carmeliet P, Koninckx PR. Role of the plasminogen system in basal adhesion formation and carbon dioxide pneumoperitoneum-enhanced adhesion formation after laparoscopic surgery in transgenic mice. Fertil.Steril. 2003;80:184-92. 29. Molinas CR, Mynbaev O, Pauwels A, Novak P, Koninckx PR. Peritoneal mesothelial hypoxia during pneumoperitoneum is a cofactor in adhesion formation in a laparoscopic mouse model. Fertil.Steril. 2001;76:560-67. 30. Molinas CR, Binda MM, Campo R, Koninckx PR. Adhesion formation and interanimal variability in a laparoscopic mouse model varies with strains. Fertil.Steril. 2005;83:1871-74. 31. Binda MM, Molinas CR, Mailova K, Koninckx PR. Effect of temperature upon adhesion formation in a laparoscopic mouse model. Hum.Reprod. 2004;19:2626-32.
32. Molinas CR, Mynbaev O, Pauwels A, Novak P, Koninckx PR. Peritoneal mesothelial hypoxia during pneumoperitoneum is a cofactor in adhesion formation in a laparoscopic mouse model. Fertil.Steril. 2001;76:560-67. 33. Elkelani OA, Binda MM, Molinas CR, Koninckx PR. Effect of adding more than 3% oxygen to carbon dioxide pneumoperitoneum on adhesion formation in a laparoscopic mouse model. Fertil.Steril. 2004;82:1616-22. 34. Binda MM, Molinas CR, Hansen P, Koninckx PR. Effect of desiccation and temperature during laparoscopy on adhesion formation in mice. Fertil.Steril. 2006;86:166-75. 35. Schonman R, Corona R, Bastidas A, De CC, Koninckx PR. Effect of upper abdomen tissue manipulation on adhesion formation between injured areas in a laparoscopic mouse model. J.Minim.Invasive.Gynecol. 2009;16:307-12. 36. Corona R, Verguts J, Schonman R, Binda MM, Mailova K, Koninckx PR. Post-operative inflammation at the abdominal cavity increases adhesion formation in a laproscopic mouse model. Fertil.Steril. 2010;in press. 37. Jevtovic-Todorovic V, Todorovic SM, Mennerick S, Powell S, Dikranian K, Benshoff N et al. Nitrous oxide (laughing gas) is an NMDA antagonist, neuroprotectant and neurotoxin. Nat.Med. 1998;4:460-63. 38. Gruss M, Mathie A, Lieb WR, Franks NP. The two-pore-domain K(+) channels TREK-1 and TASK-3 are differentially modulated by copper and zinc. Mol.Pharmacol. 2004;66:530-37. 39. Aitola P, Airo I, Kaukinen S, Ylitalo P. Comparison of N2O and CO2 pneumoperitoneums during laparoscopic cholecystectomy with special reference to postoperative pain. Surg.Laparosc.Endosc. 1998;8:140-44. 40. Hunter JG, Staheli J, Oddsdottir M, Trus T. Nitrous oxide pneumoperitoneum revisited. Is there a risk of combustion? Surg.Endosc. 1995;9:501-04. 41. Neuman GG, Sidebotham G, Negoianu E, Bernstein J, Kopman AF, Hicks RG et al.
Laparoscopy explosion hazards with nitrous oxide. Anesthesiology 1993;78:875-79. 42. Corall IM, Elias JA, Strunin L. Letter: Laparoscopy explosion hazards with nitrous oxide. Br.Med.J. 1976;1:397. 43. Robinson JS, Thompson JM, Wood AW. Letter: Laparoscopy explosion hazards with nitrous oxide. Br.Med.J. 1975;3:764-65. 44. Tsereteli Z, Terry ML, Bowers SP, Spivak H, Archer SB, Galloway KD et al. Prospective randomized clinical trial comparing nitrous oxide and carbon dioxide pneumoperitoneum for laparoscopic surgery. J.Am.Coll.Surg. 2002;195:173-79. 45. Aitola P, Airo I, Kaukinen S, Ylitalo P. Comparison of N2O and CO2 pneumoperitoneums during laparoscopic cholecystectomy with special reference to postoperative pain. Surg.Laparosc.Endosc. 1998;8:140-44. 46. El-Minawi MF, Wahbi O, El-Bagouri IS, Sharawi M, El-Mallah SY. Physiologic changes during CO2 and N2O pneumoperitoneum in diagnostic laparoscopy. A comparative study. J.Reprod.Med. 1981;26:338-46. 47. Holzman M, Sharp K, Richards W. Hypercarbia during carbon dioxide gas insufflation for therapeutic laparoscopy: a note of caution. Surg.Laparosc.Endosc. 1992;2:11-14. 48. Minoli G, Terruzzi V, Spinzi GC, Benvenuti C, Rossini A. The influence of carbon dioxide and nitrous oxide on pain during laparoscopy: a double-blind, controlled trial. Gastrointest.Endosc. 1982;28:173-75. 49. Sharp JR, Pierson WP, Brady CE, III. Comparison of CO2- and N2O-induced discomfort during peritoneoscopy under local anesthesia. Gastroenterology 1982;82:453-56.
Is the conditioning of the operative field possible in open surgery? A mouse model for
open surgery and the effect of conditioning upon adhesion formation.
(Submitted)
1,3 Roberta Corona M.D., 2Karina Mailova M.D., 1Maria Mercedes Binda Ph.D., and 1Philippe
R. Koninckx M.D., Ph.D.
Affiliation 1Department of Obstetrics and Gynaecology, University Hospital Gasthuisberg, Katholieke
Universiteit Leuven, Leuven, Belgium. 2 Department of Reproductive Medicine and Surgery, Moscow State University of Medicine
and Dentistry, Moscow, Russia
3 Corresponding author: Roberta Corona, U.Z. Gasthuisberg, Dept. Of Obstetrics & Gynaecology
Herestraat 49, B3000 Leuven, Belgium, Tel. +32 16.34.08.28, email: [email protected]
Secretary: Tel. +32 16.34.46.34 – Fax +32 16.34.42.05
Introduction
Adhesion formation has a great clinical impact, infact between 60-90% of patients who have undergone major gynaecological surgery develop adhesions 1. Although adhesions may not appear to cause problems for many patients, in a considerable proportion there are serious short- and long-term consequences including fertility issues and a life-long risk of small bowel obstruction (SBO), with important associated morbidity, expense and a considerable risk of mortality. Adhesions also complicate future surgery with serious risks to the patient.
Although laparoscopy has been commonly believed to be less adhesiogenic and cause fewer de novo adhesions to form compared to laparotomy, the comparative risk of adhesion-related complications following open and laparoscopic gynaecological surgery is similar for most procedures 2.
The pathophysiology of adhesion formation is traditionally viewed as a local phenomenon, resulting in a local inflammatory reaction with exudation 3, fibrin deposition and capillary growth at the site of injury 4. The extent of adhesions depends on the balance of natural healing mechanisms requiring fibrin removal. This fibrin is normally rapidly removed by fibrinolysis 5 while simultaneously the peritoneal repair process is started 6.
In the last decade our group identified in a laparoscopic mouse model several factors derived from the peritoneal cavity that modulated this classical phenomenon. Adhesions at the lesion site are enhanced in a dose dependent way by a pneumoperitoneum (PP) with pure CO2 7, or a PP with more than 4% O2 8;9, by desiccation 10 and by the trauma produced during bowel manipulation at remote site 11;12. We believe that the driving mechanins are mesothelial hypoxia, mesothelial hyperoxia and reactive oxygen species (ROS), desiccation and surgical trauma, respectively. The trauma caused by the surgical lesion and by the peritoneal factors lead to an acute inflammatory reaction of the entire peritoneal
cavity suggested to be the driving mechanism of adhesion formation13.
Adhesion formation after open surgery is partly due to the perioperative exposure of the operative field to the ambient air leading to various local processes that cause inflammation and damage the mesothelial peritoneal layer 14-16. These adhesiogenic processes include superficial desiccation and exposure to the atmospheric partial oxygen (O2) pressure with ensuing hyperoxia and oxidative stress 9;14;17.
We believe that the peritoneal factors identified in laparoscopy are the same occurring in an open surgical field therefore the aim of this study was, first, to develop an animal model for open surgery to study the feasibility of a peritoneal conditioning of the open surgical field. Secondly, to compare adhesion formation in laparoscopy versus laparotomy and third, to evaluate the effect of peritoneal conditioning in open surgery upon adhesion formation.
Materials and methods.
The laparoscopic mouse model for adhesion formation. The model has been validated extensively and the experimental setup ( animals, anesthesia, ventilation, laparoscopic surgery and induction and adhesion scoring) was described in detail previously 7;8;10;18;18;19. Briefly, the model consisted of performing 2 bipolar lesions during 60 min of laparoscopy using pure CO2 pneumoperitoneum. The pneumoperitoneum was induced using the Thermoflator (Karl Storz, Tuttlingen, Germany) through a 2 mm endoscope with a 3.3 external sheath for insufflation (Karl Storz, Tuttlingen Germany) introduced into the abdominal cavity through a midline incision caudal to the xyphoid appendix. The incision was closed gas tight around the endoscope in order to avoid leakage. The insufflation pressure was 15 mmHg and humidified gas (Humidifier 204320 33, Karl Storz, Tuttlingen, Germany) was used. After the establishment of the pneumoperitoneum, two 14-gauge catheters
(Insyte-W, Vialon, Becton Dickinson, Madrid, Spain) were inserted under laparoscopic vision. Standardized 10 mm x 1.6 mm lesions were performed in the antimesenteric border of both right and left uterine horns and in both the right and left pelvic side walls with bipolar coagulation (20W, standard coagulation mode, Autocon 350, Karl Storz, Tuttlingen, Germany). Since temperature is critical for adhesion formation18, animals and equipment were placed in a closed chamber at 37°C (heated air, WarmTouch, Patient Warming System, model 5700, Mallinckrodt Medical, Hazelwood, MO). Since anaesthesia and ventilation may influence body temperature (Binda 2004), the timing between anaesthesia (T0), intubation (at 10 min, T10) and the onset of the experiment (at 20 min, T20) was strictly controlled.
Animals. The present study was performed in 9-10 weeks-old female BALB/c OlaHsd mice of 18g to 20g (Harlan Laboratories B.V., Venray, The Netherlands). Animals were kept under standard laboratory conditions and diet at the animal facilities of the Katholieke Universiteit Leuven (KUL). The study was approved by the Institutional Review Animal Care Committee (n°: P040/2010).
The mouse model for open surgery. All the factors validated for laparoscopic surgery such as animals, anaesthesia, ventilation, humidification and temperature control, and the general experiment set up were kept identical for the mouse model in open surgery.
Only the surgical approach and the conditioning of the peritoneal cavity were different. The model was developed after several pilot experiments as explained in the chapter 3 of this thesis. In summary, after anesthesia and intubation, the mouse is placed in a transparent plexiglass box measuring 22cm x10cm x30cm, with 3 ports: one for the ventilation tube, one for the inflowing gas and one port for the outflowing gas. The box is closed with a sliding transparent cover that could be removed to perform the surgery or other treatments (figure 1). A 5 mm diameter tubing is
connected to the box for the inflowing gas (Thermoflator (Karl Storz, Tüttlingen, Germany ) in the lower part of a lateral wall and since density of CO2 and N2O is higher than air, the box is filled progressively until the gas escapes by overflow. All further procedures thus are performed in the specific gas environment either air, or CO2 or a mixture of CO2 with oxygen of N2O. A midline xyphopubic incision is performed and the abdomen is kept open and exposed to the environment with 2 pins. As performed in the laparoscopic mouse model, standardized 10 mm x 1.6 mm lesions are performed in the antimesenteric border of both right and left uterine horns and in both the right and left pelvic side walls with bipolar coagulation (20W, standard coagulation mode, Autocon 350, Karl Storz, Tüttlingen, Germany). After surgery the mouse is kept with the abdomen open inside the box under various environmental conditioning for 30 minutes. After 7 days, the same scoring system is used as in the laparoscopic mouse model .
Figure 1. The mouse model for open surgery. Image modified from Binda Fertil Steril 2006
Scoring of adhesions. After 7 days, adhesions were quantitatively (proportion) and qualitatively (extent, type, tenacity) scored, blindly during laparotomy using a stereomicroscope. The entire abdominal cavity was visualized using a xyphopubic midline and a bilateral subcostal incision. After the evaluation of ports sites and viscera (omentum, large and small bowels) for de novo adhesions, the fat tissue surrounding the uterus was carefully removed. The length of
the visceral and parietal lesions and adhesions were measured. Adhesions, when present, were carefully lysed to evaluate their type and tenacity. The terminology of Pouly et al was used 20, describing de novo adhesion formation for the adhesions formed at non-surgical sites, adhesion formation for adhesions formed at the surgical site.
Results
Development of the mouse model for open surgery
Pilot experiments.
In a first pilot experiment we planned to evaluate the effect of humidified 60 min of CO2, humidified CO2+N2O (50/50) and N2O in 6 mice (2 mice per group, randomised). They unfortunately all died, which was explained by excessive desiccation (visually obvious), due to mixing of the gas with the ambient air as described by Person. Therefore in all subsequent experiment the box was covered as explained in details in the chapter of material and methods.
In order to confirm the effect of desiccation (exp 2), non humidified CO2 and humidified CO2 were compared for 30 and 60 min of “open exposure” (n=12 mice). This confirmed the importance of covering the small box , and the importance of desiccation since after 60 min without humidification all mice died. This experiment was repeated (exp 3) with N2O instead of CO2, confirming the laparoscopic findings that N2O was less harmful than CO2. This experiment was therefore repeated (pilot 4) with adding 4% of Oxygen to the CO2 instead of pure CO2, equally confirming the laparoscopic findings that adding 4% of oxygen is beneficial.
After these pilot experiments we decided to do all further experiments with humidification, a covered box and operating times of 30 or 60 minutes.
Adhesions increase significantly with the increasing of the exposure time at the gas for all the different settings used, with or without
humidifier (p=0.002 CO2 30 min vs CO2 60 min, p< 0.0001 for all the other groups). With 60 minutes of 100% CO2 dry gas exposure, all the mice dead. With 30 minutes we had 1 dead out of 3. In the groups without humidifier with CO2 and with CO2+O2 were always found “de novo” adhesions in the upper abdomen. In all the groups with N2O were never found “de novo” adhesions. Using N2O gas with 100% of relative humidity, there were no adhesions after 30 minutes of gas exposure and very few after 60 minutes, significant less than in all the other groups (p < 0.0001). With the adding of O2 at the CO2 gas there is a significant decreasing in adhesions formation in comparison with the CO2 gas only ( p= 0.0011 after 30 min of humidified gas, p= 0.003 after 30 min of humidified gas, p<0.0001 after 30 minutes of dry gas exposure) (figure 2).
For every different settings there is a high statistical difference between the groups with or without humidifier, both if used 30 minutes or 60 minutes for the different gas (p<0.0001).
CO2/30 CO2/60 N20/30 N20/60 CO2+O2/30 CO2+O2/600
5
10
15
20
Humidifier NO Humidifier
X
X Mortality 100%
255075
100
Gas/minutes
Pro
port
ion
of a
dhes
ions
(%
)
Figure 2. Cumulative view of pilot experiments results after establishing of the model with a covered box.
I experiment. In a first experiment the dose response of the effect of N2O upon adhesion formation was evaluated in the open model. Except for a control group, where mice were kept exposed only to air outside the box, in the all the others groups mice were inside the cover box and kept exposed to humidified gas for 30 minutes.
In the dose response in open surgery (figure 3), there was not a significant reduction of adhesions in comparison with CO2 only when 0. 3% or 1% was added to the CO2 pneumoperitoneum; it was instead significant with the addition of 3% or 10% (p=0. 0061 and p= 0. 0006). It’ enough the addition of 10% N2O to reach almost the same effect of 100% N2O with complete prevention of adhesions (p=0. 1551). There is still a significant difference between 3% N2O and 10% (p=0. 0291).
Figure 3. Proportion of adhesions when floading the open operative field with CO2 plus different concentration of N2O.
Experiment II. Being the addition of 10% of N2O to CO2 much more important than the addition of 4% of O2, the second experiment was designed to evaluate, whether adding 4% of oxygen still had an additive effect when 10% of N2O in CO2 was used. Since adhesion formation was already so low, this was evaluated in both the open model and in the laparoscopic model using a factorial design with 10 mice in each cell.
In the open model, mice were inside the cover box and kept exposed to 100% humidified gas for 60 minutes. In the laparoscopic model, the pneumoperitoneum was manteined for 60 minutes with 100% humidified gas. The same mixing of gas, CO2 + 10% N2O or CO2+ 10% N2O + 4% O2, were used in the open and laparoscopic model.
In the results not significant difference was noted between laparoscopy and open surgery.
It was showed instead an addictive effect of O2 in both laparoscopy and open surgery (p=0. 0028 and p=0. 0175) (figure 4).
CO2 + N2O 10% CO2 + N2O 10% + 4% O20.00.10.20.30.40.50.60.70.80.91.0
open surgery laparoscopy
Pro
po
rtio
n o
f ad
hes
ion
s (%
)
Figure 4. Comparison between open surgery and laparoscopy for conditioning with CO2 plus 10 %N2O only or plus 10% N2O and 4% O2; a little addictive effect of O2 is showed.
Discussion
Laparotomy has been commonly believed to be more adhesiogenic that laparoscopy and cause more de novo adhesions compared to open surgery, although the comparative risk of adhesion-related complications following open and laparoscopic gynaecological surgery is similar for most procedures 21.
How it has been proven in several experiments, adhesions can be prevented with conditioning of the peritoneal cavity with prevention of desiccation, cooling and prevention of hypoxia during laparoscopy7;10;18. Our model for open surgery clearly showed that the same conditioning of the surgical field is feasible and efficient also during laparotomy. In air the concentration of O2 is much higher than the intracellular O2 artial pressure this inevitably leading to lacal hyeroxia with occurring of ROS and oxidative stress, deleterious for adhesion formation8;22. Since the gas used to fload the surgical field ( i.e. CO2 86% +N2O 10% + O2 4%) has an higher density of air it will act as a protective field separating the operative field from the air environment and from the consequent
oxidative stress. The addition of a physiologic concentration of oxygen (4%) to this mixture of gases corrects the mesothelial hypoxia during floading with pure CO2 7. The effect of lower temperatures 18 could be explained by making cells more resistant to damage e.g. by hypoxia. The effect of desiccation was prevented by using of humidified gas and a cover box to avoid high velocity jet and turbulent mixing with the ambient air that could bring to increasing in O2 concentration23 and desiccation rate24. Since the translation of a cover box to the clinic could be difficult, this last effect could be avoided in clinical practice by using for insufflation of a gas diffuser25.
That the use of N2O instead of CO2 can reduce postoperative adhesion formation is clearly demonstrates in our laparoscopic mouse model (PhD thesis Corona), and preliminary data of an on going clinical trial in humans, with laparoscopy and full conditioning of the peritoneal cavity with addition of O2 and N2O, confirme this finding26.
The effect of N2O, especially at low concentrations was unexpected and cannot be explained by current knowledge. The mechanism of action of N2O, however has to be different from the effect of oxygen, since the effect of even 100% of N2O is comparable to the effect of 10% of N2O , whereas oxygen concentrations above 10% clearly increase adhesions. In addition we demonstrated that both gases, especially N2O can prevent the enhancing effect upon adhesion formation of blood (Corona thesis). We believe that these effects upon adhesion formation is mediated through a decreased acute inflammation in the entire peritoneal cavity. The effect of N2O, especially at low concentrations is surprising, N2O could have anti-inflammatory effect on acute inflammation since N2O is effective in decreasing the post-operative pain and blocking directly one of the most precocius and important phase of the inflammatory process as the chemotaxis27. Chemotaxis of leucocytes also clearly decreases for more than 50% with N2O administration. Suppression of chemotaxis to corneal
inflammation by nitrous oxide27. Nitrous oxide, commonly known as "laughing gas", is a colorless non-flammable gas used in surgery and dentistry for its anesthetic and analgesic effects. Nitrous oxide is a weak general anesthetic, so it is used as a carrier gas in a 2:1 ratio with oxygen for more powerful general anesthetic agents such as sevoflurane or desflurane. It has a MAC (minimum alveolar concentration) of 105% and a blood: gas partition coefficient of 0.46. Less than 0.004% is (minimally) metabolized in humans. Nitrous oxide is relatively non-polar, with a molecular weight of 44.013 g/mol and a high lipid solubility. As a result, it diffuses quickly into the phospholipid cell membranes. Though nitrous oxide's exact mechanism of action is still open to some conjecture it is known that it acts as an NMDA antagonist at partial pressures similar to those used in general anaesthesia. N2O affects the GABA receptor 28 but this is still controversial. It has a lower potency by acting as a positive allosteric modulator. N2O, like other volatile anesthetics, activates twin-pore potassium channels, albeit weakly and these channels are largely responsible for keeping neurons at the resting (unexcited) potential 29. Thus the studies comparing N2O and CO2 pneumoperitoneum during laparoscopic cholecystectomy 30 and demonstrating that N2O pneumoperitoneum produced less postoperative pain and required a decreased quantity of anesthetic for the surgical procedure than did CO2 pneumoperitoneum could be connected with nitrous oxides general anesthetic’s properties.
N2O at concentrations less than 29% is safe as a gas to be used for the pneumoperitoneum 31. Indeed below 29% N2O does not have an explosion risk. Above 29% N2O might maintain combustion32. Thus if gasses as methane would escape from the intestine and if these gases got ignited by electrosurgery some explosion risk could exist. This risk exists theoretically whereas it seems supported by some reported accidents 33;34. Used at low concentrations of 10% or 5 % therefore does not constitute an explosion risk even if used together with 3% of oxygen, and its use can be considered absolutely safe. In addition the
solubility in water is similar for N2O and for CO2 being 1.5 and 1.45 mg/l respectively. This contrasts sharply with the low solubility of O2 and N2 in water. This high solubility with a high exchange capacity in the lungs makes N2O a safe gas to be used during pneumoperitoneum. It was reported that postoperative pain is less with N2O pneumoperitoneum than with CO2 pneumoperitoneum, and that intraoperative ventilation and metabolic management is easier by the use of N2O pneumoperitoneum instead of CO2 pneumoperitoneum. It therefore was recommended to change to N2O in case of refractory hypercarbia and to use it as the primary insufflation gas in patients with little ventilatory reserve35.
Resorption of N2O from the peritoneal cavity certainly when used as 10% N2O in CO2 is so low that it is not detected in the expired gas26.
The clinical implications of the effect of low concentrations of N2O are far reaching. First given the direct effects upon the mesothelial cells between laparoscopic surgery and laparotomy, all beneficial effects observed at laparoscopy, such as reduction in temperature, humidification, 4% of oxygen and e.g. 10% of N2O are also applicable to open surgery. In addition, we expect that as demonstrated for the addition of 4% of oxygen to the CO2 pneumoperitoneum (ref), N2O, will also reduce tumor cell implantation and postoperative pain.
In conclusion, the conditioning of the surgical field in open surgery looks feasible. The mixture of gas with CO2 + 10% of N2O + 4% of oxygen is confirmed to be the gas of choice for flooding the cavity during open surgery as well for the pneumoperitoneum during laparoscopy. When using this gas mixture there are no difference in adhesion formation between laparoscopy or laparotomy.
Reference List
1. Monk BJ, Berman ML, Montz FJ. Adhesions After Extensive Gynecologic Surgery - Clinical-Significance, Etiology, and Prevention. American Journal of Obstetrics and Gynecology 1994;170:1396-403.
2. Lower AM, Hawthorn RJ, Ellis H, O'Brien F, Buchan S, Crowe AM. The impact of adhesions on hospital readmissions over ten years after 8849 open gynaecological operations: an assessment from the Surgical and Clinical Adhesions Research Study. BJOG 2000;107:855-62.
3. diZerega GS. Biochemical events in peritoneal tissue repair. Eur.J.Surg.Suppl 1997;10-16.
4. diZerega GS. Peritoneum, peritoneal healing and adhesion formation. In: diZerega GS, editor. Peritoneal surgery. New York: Springer; 2000. p. 3-38.
5. Holmdahl L. The role of fibrinolysis in adhesion formation. Eur J Surg Suppl 1997;24-31.
6. DiZerega GS, Campeau JD. Peritoneal repair and post-surgical adhesion formation. Hum.Reprod.Update. 2001;7:547-55.
7. Molinas CR, Mynbaev O, Pauwels A, Novak P, Koninckx PR. Peritoneal mesothelial hypoxia during pneumoperitoneum is a cofactor in adhesion formation in a laparoscopic mouse model. Fertil.Steril. 2001;76:560-67.
8. Elkelani OA, Binda MM, Molinas CR, Koninckx PR. Effect of adding more than 3% oxygen to carbon dioxide pneumoperitoneum on adhesion formation in a laparoscopic mouse model. Fertil.Steril. 2004;82:1616-22.
9. Binda MM, Molinas CR, Koninckx PR. Reactive oxygen species and adhesion formation: clinical implications in adhesion prevention. Hum.Reprod. 2003;18:2503-07.
10. Binda MM, Molinas CR, Hansen P, Koninckx PR. Effect of desiccation and temperature during laparoscopy on adhesion
formation in mice. Fertil.Steril. 2006;86:166-75.
11. Corona R, Verguts J, Binda MM, Molinas CR, Schonman R, Koninckx PR. The impact of the learning curve on adhesion formation in a laparoscopic mouse model. Fertil.Steril. 2011.
12. Schonman R, Corona R, Bastidas A, De Cicco C, Koninckx PR. Effect of Upper Abdomen Tissue Manipulation on Adhesion Formation between Injured Areas in a Laparoscopic Mouse Model. Journal of Minimally Invasive Gynecology 2009;16:307-12.
13. Corona R, Verguts J, Schonman R, Binda MM, Mailova K, Koninckx PR. Post-operative inflammation at the abdominal cavity increases adhesion formation in a laproscopic mouse model. Fertil.Steril. 2010;in press.
14. The Practice Committee of the American Society for Reproductive Medicine. Pathogenesis, consequences, and control of peritoneal adhesions ingynecologic surgery. Fertil.Steril. 2008;90:44-49.
15. Ryan GB, Grobety J, Majno G. Mesothelial injury and recovery. Am.J.Pathol. 1973;71:93-112.
16. Nkere UU, Whawell SA, Sarraf CE, Schofield JB, Thompson JN, Taylor KM. Perioperative histologic and ultrastructural changes in the pericardium and adhesions. Ann.Thorac.Surg 1994;58:437-44.
17. Svenarud P, Persson M, van der LJ. Effect of CO2 insufflation on the number and behavior of air microemboli in open-heart surgery: a randomized clinical trial. Circulation 2004;109:1127-32.
18. Binda MM, Molinas CR, Mailova K, Koninckx PR. Effect of temperature upon adhesion formation in a laparoscopic mouse model. Hum.Reprod. 2004.
19. Mynbaev OA, Molinas CR, Adamyan LV, Vanacker B, Koninckx PR. Reduction of CO(2)-pneumoperitoneum-induced metabolic
hypoxaemia by the addition of small amounts of O(2) to the CO(2) in a rabbit ventilated model. A preliminary study. Hum.Reprod. 2002;17:1623-29.
20. Pouly JL, Seak-San S. Adhesions: laparoscopy versus laparotomy. In: diZerega GS, editor. Peritoneal Surgery. New York: Springer; 2000. p. 183-92.
21. Lower AM, Hawthorn RJ, Clark D, Boyd JH, Finlayson AR, Knight AD et al. Adhesion-related readmissions following gynaecological laparoscopy or laparotomy in Scotland: an epidemiological study of 24 046 patients. Hum.Reprod. 2004;19:1877-85.
22. Binda MM, Molinas CR, Bastidas A, Koninckx PR. Effect of reactive oxygen species scavengers, antiinflammatory drugs, and calcium-channel blockers on carbon dioxide pneumo-peritoneum-enhanced adhesions in a laparoscopic mouse model. Surgical Endoscopy and Other Interventional Techniques 2007;21:1826-34.
23. Persson M, van der LJ. De-airing of a cardiothoracic wound cavity model with carbon dioxide: theory and comparison of a gas diffuser with conventional tubes. J.Cardiothorac.Vasc.Anesth. 2003;17:329-35.
24. Persson M, van der LJ. Can wound desiccation be averted during cardiac surgery? An experimental study. Anesth.Analg. 2005;100:315-20.
25. Persson M, van der LJ. Intraoperative field flooding with warm humidified CO2 may help to prevent adhesion formation after open surgery. Med Hypotheses 2009;73:521-23.
26. Verguts, J. Can prevention of mesothelial damage during endoscopic surgery decrease adhesion formation and tumor implantation? 2011. Katholieke Universiteit Leuven. Ref Type: Thesis/Dissertation
27. Nyerges A. Pain Mechanisms in Laparoscopic Surgery. Semin Laparosc Surg 1994; Dec;1:215-18.
28. Jevtovic-Todorovic V, Todorovic SM, Mennerick S, Powell S, Dikranian K, Benshoff N et al. Nitrous oxide (laughing gas) is an NMDA antagonist, neuroprotectant and neurotoxin. Nat.Med 1998;4:460-63.
29. Gruss M, Bushell TJ, Bright DP, Lieb WR, Mathie A, Franks NP. Two-pore-domain K+ channels are a novel target for the anesthetic gases xenon, nitrous oxide, and cyclopropane. Mol.Pharmacol. 2004;65:443-52.
30. Aitola P, Airo I, Kaukinen S, Ylitalo P. Comparison of N2O and CO2 pneumoperitoneums during laparoscopic cholecystectomy with special reference to postoperative pain. Surg Laparosc Endosc. 1998;8:140-44.
31. Hunter JG, Staheli J, Oddsdottir M, Trus T. Nitrous-Oxide Pneumoperitoneum Revisited - Is There A Risk of Combustion. Surgical Endoscopy-Ultrasound and Interventional Techniques 1995;9:501-04.
32. Neuman GG, Sidebotham G, Negoianu E, Bernstein J, Kopman AF, Hicks RG et al. Laparoscopy explosion hazards with nitrous oxide. Anesthesiology 1993;78:875-79.
33. Corall IM, Elias JA, Strunin L. Letter: Laparoscopy explosion hazards with nitrous oxide. Br.Med J. 1975;4:288.
34. Robinson JS, Thompson JM, Wood AW. Letter: Laparoscopy explosion hazards with nitrous oxide. Br.Med J. 1975;3:764-65.
35. Tsereteli Z, Terry ML, Bowers SP, Spivak H, Archer SB, Galloway KD et al. Prospective randomized clinical trial comparing nitrous oxide and carbon dioxide pneumoperitoneum for laparoscopic surgery. Journal of the American College of Surgeons 2002;195:173-79.
