Deep sedation in natural orifice transluminal endoscopic surgery(NOTES): a comparative study with dogs

Mohammad Al-Haddad • Daniel McKenna • Jeff Ko • Stuart Sherman •

Don J. Selzer • Samer G. Mattar • Thomas F. Imperiale • Douglas K. Rex •

Attila Nakeeb • Seong Mok Jeong • Cynthia S. Johnson • Lynetta J. Freeman

Received: 19 December 2011 / Accepted: 2 April 2012

� Springer Science+Business Media, LLC 2012

Abstract

Background Natural orifice transluminal endoscopic sur-

gery (NOTES) has been mostly performed with the animal

under general and inhalational anesthesia (IA-NOTES). To

date, NOTES using propofol sedation (PS-NOTES) has not

been investigated. This study aimed to assess the feasibility

and safety of PS-NOTES for transgastric oophorectomy

with carbon dioxide insufflation and to compare its success

rates with those of conventional IA-NOTES.

Methods In this prospective randomized study, NOTES

oophorectomy was performed for 19 female dogs random-

ized to two conditions: PS (study group) and IA (control

group). Sedation success rates (ability to visualize and resect

ovaries without converting to IA), operative success rates

(ability to resect and retrieve both ovaries in full using only

NOTES), and vital parameters including hemodynamic and

respiratory changes were documented.

Results In the PS-NOTES group (n = 9), the sedation

success rate was 100 %. The operative success rate was

67 % (6 of 9 animals) compared with 80 % (8 of 10 animals)

in the IA-NOTES group. No purposeful movement occurred

during surgical manipulation and no respiratory or cardio-

vascular complications in occured the PS group. Heart rate

(HR) and end-tidal carbon dioxide (ETCO2) were signifi-

cantly higher in the PS group than in the IA group. Blood

pressure (BP) was significantly higher in the PS group only

during the middle part of the procedure. Only mild respi-

ratory depression was noted in the PS group, as indicated by

elevated but acceptable ETCO2. Elevations in BP and HR

are thought to be related to elevated CO2 but did not appear

to have an adverse impact on the course of the procedure.

Recovery was uneventful for all the animals.

Conclusion The use of PS-NOTES appears to be feasible,

resulting in outcomes comparable with those for IA in

dogs. Further studies are needed to determine the appli-

cability of this concept in human NOTES.

Keywords Inhalational anesthesia � NOTES �Oophorectomy � Propofol sedation

Natural orifice transluminal endoscopic surgery (NOTES)

has emerged as an ‘‘incisionless’’ transvisceral approach to

the peritoneal cavity [1, 2]. The advantages that NOTES

has over conventional surgery include less postoperative

pain, fewer complications, decreased anesthesia require-

ments, accelerated patient recovery and return to normal

function, and elimination of the risk for incisional hernia-

tion. Despite the identification of barriers to human uses in

the first white paper [3], human NOTES applications

continue to expand rapidly under investigational settings at

a limited number of centers [4–15].

Presented as a poster at the Digestive Disease Week held in Chicago,

IL, USA between 7–10 May 2011.

M. Al-Haddad (&) � S. Sherman � T. F. Imperiale � D. K. Rex

Department of Medicine, Division of Gastroenterology, Indiana

University School of Medicine, 550 N University Boulevard,

UH 4100, Indianapolis, IN 46202, USA

e-mail: moalhadd@iupui.edu

D. McKenna � D. J. Selzer � S. G. Mattar � A. Nakeeb

Department of Surgery, Indiana University School of Medicine,

Indianapolis, IN, USA

J. Ko � S. M. Jeong � L. J. Freeman

Purdue University School of Veterinary Medicine,

West Lafayette, IN, USA

C. S. Johnson

Division of Biostatistics, Indiana University School of Medicine,

Indianapolis, IN, USA

123

Surg Endosc

DOI 10.1007/s00464-012-2309-1

and Other Interventional Techniques

Similar to laparoscopy, to date, human NOTES has been

performed with the patient under general anesthesia for

hemodynamic and ventilating reasons. The increased

intraabdominal pressure from carbon dioxide (CO2)

insufflation during laparoscopy adversely affects the

hemodynamic stability of the patients by preventing

venous return and reducing cardiac output [16]. Further-

more, the increased abdominal pressure displaces the dia-

phragm cephalad, reducing lung’s vital capacity during

surgery and the early postoperative period [17–19]. Post-

operative pain after laparoscopic procedures also could be

at least partially attributed to insufflation of CO2.

A proposed benefit of NOTES is that it requires minimal

intraabdominal CO2 insufflation for visualization of struc-

tures and performance of procedures, and this combined with

less abdominal trauma could partly explain the reduced need

for postoperative analgesics in the clinical cases of NOTES

cholecystectomy performed recently [20]. Our group has

demonstrated that NOTES oophorectomy can be performed

easily with the subject under a CO2 insufflation pressure of 10

to 12 mmHg, which is lower than the 14 mmHg of pressure

required for the same procedure performed via laparoscopy

[21]. Reduced insufflation pressures with NOTES procedures

may preserve total lung capacity and maintain hemodynamic

stability of the anesthetized patient without the need for

assisted ventilation during the procedure [22].

We recently have demonstrated that NOTES oopho-

rectomy allows earlier recovery than open and laparoscopic

approaches [23, 24]. This has led to the assumption that

sedation rather than general anesthesia may be feasible for

NOTES due to the lower insufflation pressures compared

with laparoscopy.

Conscious sedation with or without local anesthesia has

been used successfully on a limited scale in gynecologic

and general surgical procedures [25–27]. A few NOTES

case reports have introduced this concept for a percutane-

ous endoscopic gastrostomy (PEG) rescue in one case [28]

and for peritoneoscopy in another case [29]. Propofol is a

short-acting hypnotic agent with the advantage of rapid

onset and offset of sedation as well as faster recovery of

neuropsychiatric function [30–33]. Its safety and effec-

tiveness for gastrointestinal (GI) procedures have been well

described in large studies and endorsed by the various GI

societies [30–32, 34–36]. Moreover, propofol use has

facilitated the performance of complex endoscopic inter-

ventions in which traditional conscious sedation could be

inadequate [37] and for high-risk patients [38].

Our hypothesis that NOTES can be performed with the

patient under deep sedation was derived from previous

observations of NOTES oophorectomy performed in dogs.

This study aimed to assess the feasibility and safety of

NOTES oophorectomy using propofol sedation (PS) com-

pared with general inhalational anesthesia (IA).

Methods

Animal model

The study was approved by Indiana University and Purdue

University Institutional Animal Care and Use Committees

(IACUC). A local animal shelter provided all the animals

for the study. The study used 20 healthy female dogs

weighing 11.6–26.4 kg. The animals were randomized into

blocks with procedures consecutively performed on all the

animals in the same group (PS or IA). One dog initially

assigned to the PS group was excluded when the preop-

erative exam showed a spaying scar. At completion of the

monitoring period, the dogs were returned to the same

shelter for adoption.

Study definitions

Sedation success with PS-NOTES was defined as the

ability to visualize and complete ovarian resection without

respiratory or hemodynamic compromise necessitating

conversion to general IA for completion of the procedure.

Operative success was defined as complete resection and

retrieval of both ovaries and gastric closure using the

NOTES technique.

Sedation, anesthesia, and monitoring

All the animals were fasted for 24 h, and baseline values for

heart rate, respiratory rate, and temperature were obtained.

In the PS-NOTES group, a loading dose of propofol (3 mg/

kg) was administered intravenously (IV) followed by inter-

mittent boluses (*1 mg/kg given every 1–2 min) for

maintenance titered to a desired depth of sedation and vital

parameters. The trachea was not intubated in this group, and

a sampling catheter was placed in the proximal trachea to

obtain a sample of expired air for end-tidal CO2. Supple-

mental oxygen also was provided, and the animals breathed

spontaneously without ventilation. The a priori criteria for

conversion from PS to IA included any sustained hemody-

namic or respiratory compromise or excessive animal

movement impairing the progress of the procedure.

The dogs in the IA group were premedicated with ace-

promazine 0.02 mg/kg and hydromorphone 0.1 mg/kg

followed by propofol induction (6 mg/kg, IV), endotra-

cheal intubation, isoflurane anesthesia, and periodic man-

ual ventilation. Mechanical ventilation then was used, and

end-tidal CO2 was maintained between 35 and 40 mmHg.

Overall and segment-specific procedure durations were

tabulated, and vital signs including heart rate (HR), mean

blood pressure (BP), respiratory rate (RR), oxygen satu-

ration (SpO2) levels, and end-tidal partial pressure of CO2

(ETCO2) levels were recorded every 5 min in each group.

Surg Endosc

123

The animals were positioned in dorsal recumbency on a

circulating water blanket to prevent anesthesia-induced

hypothermia. Intravenous fluids were given during the

procedure. Perioperative antibiotics (cefazolin 22 mg/kg

IV every 2 h intraoperatively) were given, and aseptic

procedures for clipping, preparing, and draping of the

abdomen were followed. The endoscopes and other

equipment underwent high-level disinfection after every

use. An overtube (U.S. Endoscopy, Mentor, OH, USA) was

used to reduce oral contamination. Both groups received

carprofen (4 mg/kg, subcutaneously) preoperatively and

24 h postoperatively for analgesia. The dogs were given

hydromorphone 0.05 mg/kg intramuscularly (IM) at the

end of the surgical procedure for postoperative analgesia. A

second dose of hydromorphone 0.05 mg/kg IM was

administered 6 h after surgery.

Postoperatively, each animal was monitored in recovery

until its body temperature was higher than 37.2 �C. Post-

operatively, the heart rate, respiratory rate, temperature, and

indirect blood pressure were recorded every 6 h. The ani-

mals were monitored for 48 h for postoperative pain and

complications before they were returned to the local shelter.

NOTES procedure

Starting with the IA group, NOTES oophorectomy was

performed as we have previously described (Fig. 1) [24].

After a flexible therapeutic endoscope (Olympus GIF 2T-

160; Olympus America Inc., Center-Valley, PA, USA) had

been passed into the stomach, cefazolin (1 g in 200 ml of

normal saline) was instilled for 10 min and then aspirated.

The gastrotomy was performed using an endoscopic bal-

loon dilator (CRE Esophageal/Colonic Wire-Guided Bal-

loon; Boston Scientific Corporation, Natick, MA, USA)

advanced over a percutaneously inserted guidewire to

create a 20 mm gastrotomy (Fig. 2) [24]. The endoscope

was passed into the abdominal cavity, and air insufflation

via the endoscope was provided (Fig. 3). Another 18-gauge

catheter was placed in the peritoneal cavity, and CO2

insufflation was provided by a standard laparoscopic

insufflator (Karl Storz Veterinary 183 Endoscopy America,

Goletta, CA, USA) with the pressure set to 10 mmHg in

addition to air from the endoscope. An alarm sounded

when the intraabdominal pressure exceeded 10 mmHg or

decreased to \8 mmHg.

The animal’s position was adjusted to expose each

ovary. A 3.0 9 4.5 cm hexagonal snare (AcuSnare; Cook

Medical Inc., Bloomington, IN, USA) was passed through

one of the working channels of the endoscope, and endo-

scopic grasping forceps (Polygrab Tripod; Olympus

Endoscopy, Center-Valley, PA, USA) were passed into the

second channel. Together, these instruments were used to

elevate and loop the ovary (Fig. 4).

The ovary was suspended to the abdominal wall using a

surgical suture and inspected before initiation of cautery

(Fig. 5). If needed, the ovary was released and resuspended

to achieve the optimum resection level. The blood supply

then was coagulated and transected using monopolar

electrocautery at 20–40 W of blended current (Endostat II

electrosurgical generator; Boston Scientific Corp.). The site

was examined for hemorrhage. When encountered, hem-

orrhage was treated by attempted endoscopic cautery. Each

Fig. 1 Cartoon demonstrating the path of the endoscope as it exits

the stomach and the consecutive steps for identifying, resecting, and

retrieving each of the ovaries

Fig. 2 Freshly created gastrostomy pictured after deflation of the

CRE balloon

Surg Endosc

123

ovary then was removed and examined to ensure complete

resection (Fig. 6). If the ovary was not present in the tissue

removed, another excision was performed.

After removal of both ovaries, the gastrotomy was

closed with prototype T fasteners using 2–0 nylon sutures

and a knotting element (developed by Cook Medical,

Bloomington, IN, USA; Fig. 7). Four T fasteners were

placed around the incision, and opposite sutures were

opposed and secured with the knotting element as previ-

ously described (Figs. 8, 9).

Statistical analysis

For the study, 10 animals were block-randomized to each

group. With 10 animals per group, a two-sided, two-sample

t test has 80 % power at the 0.05 level of significance to

detect an effect size of 1.3 standard deviations.

Due to variation in procedure times, only vital param-

eters taken during the first 145 min were used in the

analyses. Two-sample t tests were used to compare base-

line characteristics and procedure durations between the PS

and IA groups. Repeated measures analysis of variance

(ANOVA) was used to examine differences in mean blood

pressure, heart and respiratory rates, SpO2 levels, and

ETCO2 between the groups over time (Fig. 10, 11, 12, 13,

and 14).

Fig. 3 Abdominal wall as viewed by the endoscope after inflation of

the peritoneal cavity with air. The endoscope is subsequently oriented

caudad, and the left ovary is identified

Fig. 4 Endoscopic grasping forceps passed through a widely opened

snare to elevate the ovary

Fig. 5 Left ovary suspended to the abdominal wall using an external

surgical suture to optimize visualization of the ovary and facilitate

adjustment of the forceps grasp without dropping the ovary

Fig. 6 Resected ovary (white part, upper half of the specimen) as

seen after incision of the surrounding bursa

Surg Endosc

123

Group-by-time interaction was included in each model

to test for varying group differences over time. If the

interaction was not significant, it was removed, and the

final model included only the main effects. All p values

less than 0.05 were considered significant. If the interaction

term was significant, repeated measures ANOVA was

performed using data from each 20-min interval of the

procedure. If the interaction term was significant within a

20 min interval, two-sample t tests were performed at each

time. All analyses were done using SAS version 9.3 (SAS,

Cary, NC, USA).

Results

The study was conducted between July and August 2010.

The average total propofol dose in the PS-NOTES group

was 1,178.6 mg, and the average infusion rate was

0.49 mg/kg/min.

Sedation and operative outcomes

The animals in the PS group weighed significantly less

(p = 0.035) and had a significantly shorter distance to the

gastroesophageal (GE) junction (p = 0.029) than the ani-

mals in the IA group. The procedure durations did not

differ significantly between the groups except that signifi-

cantly less time was required to close the gastric incision in

the PS group than in the IA group (p = 0.021). In the PS

group (n = 9), the sedation success rate was 100 %, and

Fig. 7 Plastic cap on the end of the endoscope providing a safety

margin for deployment of the T-fasteners used to close the

gastrostomy. A suture-knotting device (in purple) is used to unite

each pair of opposing sutures by delivery of a small metallic binder

Fig. 8 Cartoon illustrating the mechanism of gastric closure using

the metallic binder to unite the sutures before they are cut with

endoscopic scissors

Fig. 9 Gastrostomy appearance after complete closure

Fig. 10 Mean blood pressure (mmHg) over time (min)

Surg Endosc

123

the operative success rate was 67 % (6 of 9 animals)

compared with 80 % in the IA group (8 of 10 animals).

The five operative failures were related to inadequate

hemostasis of the ovarian pedicle in four animals (2 in each

group) and inability to mobilize the ovary due to uterine size in

one animal (PS group). These failures were resolved by con-

version to open laparotomy with the animal under general IA.

There was one episode of desaturation lasting less than

1 min that required removal and repositioning of the

esophageal overtube in the first PS-NOTES animal. There

was no purposeful movement during surgical manipulation,

and no respiratory or cardiovascular adverse events were

encountered in either group. Recovery was uneventful,

with no need for rescue analgesia or postoperative com-

plications in any animal.

Vital parameters

Heart rate and ETCO2 were significantly higher in the PS

group (p = 0.002) than in the IA group (p = 0.036). There

were significant group-by-time interactions for mean blood

pressure (p \ 0.001), respiratory rate (p \ 0.001), and

SpO2 (p = 0.017) (Table 1). The mean blood pressure did

not differ significantly between the groups during the first

20 min of the procedure or after 70 min (Table 2). The

animals in the PS group had a significantly higher mean

blood pressure than the animals receiving IA for

25–70 min (Table 3). The animals receiving PS had a

significantly higher respiratory rate than the animals

receiving IA during the first 20 min of the procedure

(Table 1), but the respiratory rates did not differ signifi-

cantly between the groups after 20 min (Tables 2, 3).

During the first 20 min, and again between 50 and 70 min,

the animals receiving PS had a significantly greater SpO2

than the animals receiving IA. At other times during the

procedure, the SpO2 did not differ significantly between the

groups

Fig. 11 Mean heart rate over time (min)

Fig. 12 Mean respiratory rate over time (min)

Fig. 13 Mean oxygen saturation (SpO2) (mmHg) over time (min)

Fig. 14 Mean end-tidal carbon dioxide (CO2) (mmHg) over time

(min)

Surg Endosc

123

Discussion

This is the first animal study investigating the propofol-

only approach in NOTES. Propofol appears to be feasible

for performing NOTES oophorectomy in dogs without the

need for airway intubation and with outcomes similar to

those for inhalant general anesthesia. Further studies,

however, are needed to determine the applicability of this

concept for human NOTES.

We have defined procedural success using two criteria

(sedation and operative technique) to separate the failures

due to operative techniques from those due to sedation

inadequacy. The sedation success rate for NOTES was

100 %, and the operative success rates between the two

groups were comparable (67 vs. 80 %). The operative

failures were independent of the sedation or anesthesia

used and related to post-resection hemorrhage from the

ovarian pedicle in four animals and due to a large post-

partum uterus in one animal whose the ovaries could not be

mobilized. The time spent attempting to secure hemostasis

in those four animals resulted in variability in total pro-

cedure time between cases (Table 4). It can be argued that

operative failures can lead to ‘‘sedation’’ failures because

conversion to the open approach requires IA, although this

failure would not be directly related to the mode of seda-

tion or anesthesia. Such operative failures could be mini-

mized in future human trials if dedicated NOTES platforms

become available.

We observed no body movements or significant com-

promise in vital functions that required rescue intubation or

positive pressure ventilation for any animal in the PS

group. Despite mildly elevated partial pressure of CO2

Table 1 Comparison of vital parameters over time

Vital parameter Overall adjusted mean (SE) p value

PS-NOTES IA-NOTES Interaction Group Time

BP (combined) (mmHg) – – \0.001 – –

HR (per min) 132.4 (5.0) 106.9 (4.7) 0.693 0.002 0.001

RR (per min) – – \0.001 – –

SpO2 (%) – – 0.017

ETCO2 (mmHg) 58.6 (3.1) 48.9 (2.9) 0.621 0.036 0.011

SE standard error, PS propofol sedation, NOTES natural orifice transluminal endoscopic surgery, IA inhalational anesthesia, BP blood pressure,

HR heart rate, RR respiratory rate, SpO2 oxygen saturation, ETCO2 end-tidal carbon dioxide

Table 2 Comparison of vital

parameters over time when a

significant interaction term was

present

SE standard error, PS propofol

sedation, NOTES natural orifice

transluminal endoscopic

surgery, IA inhalational

anesthesia, BP blood pressure,

RR respiratory rate, SpO2

oxygen saturation

Vital parameter Overall adjusted mean (SE) p value

Time (min) PS-NOTES IA-NOTES Interaction Group Time

BP (combined) (mmHg) 0–20 98.1 (6.7) 79.1 (6.3) 0.206 0.054 0.003

25–45 – – 0.007 – –

50–70 – – 0.024 – –

75–95 127.0 (5.1) 71.0 (4.8) 0.756 \0.001 0.198

100–120 120.1 (7.8) 77.4 (8.4) 0.398 0.003 0.042

125–145 110.8 (6.7) 77.3 (6.4) 0.142 0.022 0.336

RR (per min) 0–20 16.8 (1.4) 12.0 (1.4) 0.248 0.026 0.752

25–45 15.3 (1.9) 14.7 (1.8) 0.069 0.815 0.132

50–70 – – \0.001 – –

75–95 18.6 (1.8) 15.4 (1.7) 0.330 0.222 0.801

100–120 16.2 (2.4) 16.2 (2.6) 0.267 0.988 0.307

125–145 12.0 (3.0) 14.3 (3.4) 0.788 0.632 0.703

SpO2 (%) 0–20 95.6 (0.5) 98.1 (0.5) 0.753 0.002 0.864

25–45 96.3 (0.4) 97.1 (0.3) 0.135 0.119 0.805

50–70 95.0 (0.5) 98.2 (0.4) 0.250 \0.001 0.912

75–95 95.9 (0.6) 97.5 (0.6) 0.433 0.076 0.316

100–120 96.7 (0.4) 97.6 (0.5) 0.358 0.237 0.001

125–145 96.4 (1.3) 96.3 (1.4) 0.240 0.951 0.490

Surg Endosc

123

(PCO2) indicating underlying hypoventilation, this has not

resulted in any significant hypoxemia. In the same group, a

moderate increase in intraperitoneal pressure (up to

10 mmHg) did not significantly compromise spontaneous

ventilation in our experience. No procedure was interrupted

due to respiratory or hemodynamic compromise.

Elevations in HR (throughout) and BP (25–70 min)

were significant in the PS group, and we believe these are

secondary to increased CO2 (in turn related to mild respi-

ratory depression) rather than inadequate sedation. The

elevated CO2 likely is related to permitting the animals to

breathe spontaneously with no active ventilation. However,

this has not resulted in any postprocedural delay of

recovery or resumption of feeding and activity in any

animal. It is possible that CO2 elevation could be a

reflection of inadequate sedation, but we believe this is less

likely because no inadvertent animal movement was

noticed during the procedure. Despite the use of propofol

doses adequate for performing oophorectomy (average,

0.49 mg/kg), we encountered no apnea, significant hyp-

oxemia, or hypotension in the PS group.

The use of propofol for GI sedation has been well

characterized in the literature during the last decade.

Findings have shown propofol administration by endos-

copists or supervised experienced nurses to be feasible

without a dramatic increase in the costs associated with

using anesthesia services, resulting in better patient satis-

faction than with conventional sedation [30–32].

In addition, a single intravenous agent for induction and

maintenance of anesthesia [total intravenous anesthesia

(TIVA)] without inhalational agents was introduced to

clinical practice more than a decade ago [39–46]. Propofol-

based TIVA provides rapid induction and recovery of

anesthesia, facilitates rapid turnover of patients, and pro-

vides access to anesthesia services outside the conventional

operating room setting where IA is unavailable. Never-

theless, TIVA usually is combined with analgesics and/or

Table 3 Two-sample t test comparisons of vital parameters at each

time

Vital parameter Time PS-

NOTES

IA-

NOTES

p value

(min) Mean (SE) Mean (SE)

BP (combined) (mmHg) 25 121.8 (9.0) 84.8 (7.8) 0.008

30 117.0 (5.5) 85.0 (6.3) 0.002

35 120.3 (5.8) 83.1 (8.0) 0.002

40 123.4 (6.0) 78.4 (7.5) \0.001

45 129.1 (7.3) 78.5 (5.7) \0.001

50 123.4 (5.3) 78.7 (6.7) \0.001

55 127.1 (5.5) 79.5 (6.7) \0.001

60 124.3 (6.1) 77.5 (5.7) \0.001

65 125.1 (5.8) 75.3 (5.6) \0.001

70 128.0 (4.7) 74.8 (6.5) \0.001

RR (/min) 50 13.6 (1.8) 15.8 (2.0) 0.417

55 13.8 (1.6) 14.7 (1.6) 0.688

60 14.9 (2.1) 14.5 (1.7) 0.886

65 16.9 (1.8) 13.1 (1.4) 0.119

70 16.4 (1.8) 14.6 (1.5) 0.436

PS propofol sedation, NOTES natural orifice transluminal endoscopic

surgery, IA inhalational anesthesia, SE standard error, BP blood

pressure, RR respiratory rate

Table 4 Baseline procedure-related data for the inhalational anesthesia and propofol groups

Inhalational anesthesia group Propofol group p value (t test)

n Mean SD Min Max n Mean SD Min Max

Weight (kg) 10 20.1 4.6 12.4 26.4 9 15.8 3.4 11.6 20 0.035

Distance to gastroesophageal junction (cm) 10 55.6 5.0 49 67 9 50.2 4.7 45 58 0.029

Induction to starting surgery (min) 10 39.4 15.2 23 60 9 47.7 7.7 37 60 0.162

Step 1a (min) 10 12.8 3.4 10 21 9 12.9 2.6 10 18 0.950

Step 2b (min) 10 8.3 5.2 3 20 9 6.6 5.1 2 16 0.470

Step 3b (min) 10 10.5 7.8 2 27 8 7.4 5.9 1 20 0.362

Step 4c (min) 9 4.1 2.8 1 10 8 9.4 7.6 1 22 0.099

Step 5c (min) 8 5.6 5.8 2 16 7 7.9 7.8 1 22 0.537

Step 6d (min) 8 23.4 5.0 17 31 6 17.0 3.6 13 22 0.021

Overall procedure time (min) 10 132.6 45.0 88 227 9 126.9 40.5 75 192 0.776

Min minimum, Max maximum, SD standard deviationa Create access to the peritoneal cavityb Access and resect the left ovaryc Access and resect the right ovaryd Close the gastric incision

Surg Endosc

123

benzodiazepines and muscle relaxants and typically

requires endotracheal intubation with positive ventilation

due to the risk of apnea associated with higher doses of

anesthetics.

None of these adjunct agents were used in the current

study, so we labeled the intervention as ‘‘sedation’’ rather

than anesthesia. Our data suggest that visceral resection can

be performed with the subject under deep sedation alone,

without the need for airway control, or combined with sys-

temic analgesics or muscle relaxants. In our opinion, the lack

of abdominal wall trauma (compared with open surgery) and

reduced intraperitoneal pressure (compared with laparos-

copy) is behind this observation in NOTES. Additionally,

there is evidence that propofol-based anesthesia is associated

with less increase in catecholamines, adrenocorticotropic

hormone, and cortisol intra- and postoperatively than inha-

lational agents [47, 48] which could have reduced the car-

diovascular burden of anesthesia in our study.

The study had some limitations. First, the limited

number of animals in each group may not have powered

the study sufficiently for comparison of the particular

outcomes we reported between the two groups, but we

believe that such results would hold true in a larger sample

due to the strength of the statistical significance.

Second, the study used block randomization of 10 ani-

mals each, starting with the IA group. The impact of the

timing in this case probably was minimal because our team

achieved the ‘‘plateau’’ of the learning curve earlier, as we

have demonstrated in a previous study [49]. For this study to

be applicable to humans, we are assuming that abdominal

wall compliance in humans and canines is similar. Never-

theless, few data exist to support this assumption. Moreover,

factors that could potentially alter abdominal wall compli-

ance in the dogs, such as previous pregnancy or weight loss,

were largely unknown for the animals selected for this study.

Additionally, we discovered after completion of the

study that the two groups were not well matched by weight

and that the animals in the IA group were significantly

heavier. Because all the animals were provided by the local

shelter, the investigators had limited control over the breed

or the weight of the animals sent for oophorectomy.

Finally, we did not closely track the recovery time in the

IA group to allow comparison with the PS group because

this study was designed mainly to assess sedation and

outcomes. We have previously reported on earlier post-

operative recovery and resumption of bowel function in

NOTES oophorectomy [21].

If further studies demonstrate the feasibility of the

described approach for humans, several advantages can be

achieved. The PS-NOTES approach could provide a way to

perform intraabdominal surgical interventions on an out-

patient basis or in an environment such as a battlefield,

in which a fully equipped hospital is not immediately

available. Additionally, this approach will provide further

advantages by circumventing the need for bulky IA

equipment, reducing costs due to a shorter hospital stay and

facilitating outpatient procedures.

Acknowledgments This study was funded by the NOSCAR (2010)

and the Glen A. Lehman Endowed Chair Fund in Gastroenterology.

Disclosures Mohammad Al-Haddad, Daniel McKenna, Jeff Ko,

Stuart Sherman, Don J Selzer, Samer G. Mattar, Thomas F. Imperiale,

Douglas K. Rex, Attila Nakeeb, Seong Mok Jeong, Cynthia S.

Johnson, and Lynetta J Freeman have no conflicts of interest or

financial ties to disclose.

References

1. Flora ED, Wilson TG, Martin IJ, O’Rourke NA, Maddern GJ

(2008) A review of natural orifice translumenal endoscopic sur-

gery (NOTES) for intra-abdominal surgery: experimental models,

techniques, and applicability to the clinical setting. Ann Surg

247:583–602

2. McGee MF, Rosen MJ, Marks J, Onders RP, Chak A, Faulx A

et al (2006) A primer on natural orifice transluminal endoscopic

surgery: building a new paradigm. Surg Innov 13:86–93

3. Rattner D, Kalloo A (2006) ASGE/SAGES Working Group on

Natural Orifice Translumenal Endoscopic Surgery, October 2005.

Surg Endosc 20:329–333

4. Horgan S, Thompson K, Talamini M, Ferreres A, Jacobsen G,

Spaun G et al. Clinical experience with a multifunctional, flexible

surgery system for endolumenal, single-port, and NOTES pro-

cedures. Surg Endosc 25:586–592

5. Inoue H, Minami H, Kobayashi Y, Sato Y, Kaga M, Suzuki M

et al. Peroral endoscopic myotomy (POEM) for esophageal

achalasia. Endoscopy 42:265–271

6. Lehmann KS, Ritz JP, Wibmer A, Gellert K, Zornig C, Burghardt

J et al. The German registry for natural orifice translumenal

endoscopic surgery: report of the first 551 patients. Ann Surg

252:263–270

7. Nau P, Anderson J, Yuh B, Muscarella P Jr., Christopher Ellison

E, Happel L et al. Diagnostic transgastric endoscopic periton-

eoscopy: extension of the initial human trial for staging of pan-

creatic head masses. Surg Endosc 24:1440–1446

8. Nikfarjam M, McGee MF, Trunzo JA, Onders RP, Pearl JP,

Poulose BK et al. Transgastric natural-orifice transluminal

endoscopic surgery peritoneoscopy in humans: a pilot study in

efficacy and gastrotomy site selection by using a hybrid tech-

nique. Gastrointest Endosc 72:279–283

9. Park PO, Bergstrom M. Transgastric peritoneoscopy and appen-

dectomy: thoughts on our first experience in humans. Endoscopy

42:81–84

10. Park PO, Bergstrom M, Rothstein R, Swain P, Ahmed I, Gomez

G et al. Endoscopic sutured closure of a gastric natural orifice

transluminal endoscopic surgery access gastrotomy compared

with open surgical closure in a porcine model: a randomized,

multicenter controlled trial. Endoscopy 42:311–317

11. Saad S, Schmischke D, Martin C, Schieren T. Hybrid laparo-

scopic colectomy with transluminal colonoscopic specimen

extraction: a step toward natural orifice surgery. Endoscopy

42(Suppl 2):E346–E347

12. Salinas G, Saavedra L, Agurto H, Quispe R, Ramirez E, Grande J

et al. Early experience in human hybrid transgastric and trans-

vaginal endoscopic cholecystectomy. Surg Endosc 24:1092–1098

Surg Endosc

123

13. Sylla P, Rattner DW, Delgado S, Lacy AM. NOTES transanal

rectal cancer resection using transanal endoscopic microsurgery

and laparoscopic assistance. Surg Endosc 24:1205–1210

14. Zornig C, Siemssen L, Emmermann A, Alm M, von Waldenfels

HA, Felixmuller C et al. NOTES cholecystectomy: matched-pair

analysis comparing the transvaginal hybrid and conventional

laparoscopic techniques in a series of 216 patients. Surg Endosc.

Epub 25 December 2010

15. Zorron R, Palanivelu C, Galvao Neto MP, Ramos A, Salinas G,

Burghardt J et al. International multicenter trial on clinical natural

orifice surgery–NOTES IMTN study: preliminary results of 362

patients. Surg Innov 17:142–158

16. Galizia G, Prizio G, Lieto E, Castellano P, Pelosio L, Imperatore

V et al (2001) Hemodynamic and pulmonary changes during

open, carbon dioxide pneumoperitoneum, and abdominal wall-

lifting cholecystectomy: a prospective, randomized study. Surg

Endosc 15:477–483

17. Chumillas MS, Ponce JL, Delgado F, Viciano V (1998) Pul-

monary function and complications after laparoscopic cholecys-

tectomy. Eur J Surg 164:433–437

18. Wallace DH, Serpell MG, Baxter JN, O’Dwyer PJ (1997) Ran-

domized trial of different insufflation pressures for laparoscopic

cholecystectomy. Br J Surg 84:455–458

19. Bures E, Fusciardi J, Lanquetot H, Dhoste K, Richer JP, Lacoste

L (1996) Ventilatory effects of laparoscopic cholecystectomy.

Acta Anaesthesiol Scand 40:566–573

20. Zorron R, Filgueiras M, Maggioni LC, Pombo L, Lopes Carvalho

G, Lacerda Oliveira A (2007) NOTES transvaginal cholecystec-

tomy: report of the first case. Surg Innov 14:279–283

21. Freeman LJ, Rahmani EY, Al-Haddad M, Sherman S, Chiorean

MV, Selzer DJ et al. Comparison of pain and postoperative stress

in dogs undergoing natural orifice transluminal endoscopic sur-

gery, laparoscopic, and open oophorectomy. Gastrointest Endosc

72:373–380

22. McGee MF, Rosen MJ, Marks J, Chak A, Onders R, Faulx A et al

(2007) A reliable method for monitoring intraabdominal pressure

during natural orifice translumenal endoscopic surgery. Surg

Endosc 21:672–676

23. Freeman L, Rahmani EY, Burgess RC, Al-Haddad M, Selzer DJ,

Sherman S, Constable P (2011) Evaluation of the learning curve

for natural orifice transluminal endoscopic surgery: bilateral

ovariectomy in dogs. Vet Surg 40(2):140–150

24. Freeman LJ, Rahmani EY, Sherman S, Chiorean MV, Selzer DJ,

Constable PD et al (2009) Oophorectomy by natural orifice

transluminal endoscopic surgery: feasibility study in dogs. Gas-

trointest Endosc 69:1321–1332

25. Pellicano M, Zullo F, Cappiello F, Di Carlo C, Cirillo D, Nappi C

(2000) Minilaparoscopic ovarian biopsy performed under con-

scious sedation in women with premature ovarian failure. J Re-

prod Med 45:817–822

26. Pellicano M, Zullo F, Fiorentino A, Tommaselli GA, Palomba S,

Nappi C (2001) Conscious sedation versus general anaesthesia

for minilaparoscopic gamete intra-Fallopian transfer: a prospec-

tive randomized study. Hum Reprod 16:2295–2297

27. Tytherleigh MG, Fell R, Gordon A (2004) Diagnostic conscious

pain mapping using laparoscopy under local anaesthetic and

sedation in general surgical patients. Surgeon 2:157–160

28. Marks JM, Ponsky JL, Pearl JP, McGee MF (2007) PEG ‘‘res-

cue’’: a practical NOTES technique. Surg Endosc 21:816–819

29. Lee CK, Lee SH, Chung IK, Lee TH, Kim HS, Park SH et al.

Human diagnostic transgastric peritoneoscopy with the submu-

cosal tunnel technique performed with the patient under con-

scious sedation (with video). Gastrointest Endosc 72:889–891

30. Rex DK (2006) Review article: moderate sedation for endoscopy:

sedation regimens for nonanaesthesiologists. Aliment Pharmacol

Ther 24:163–171

31. Rex DK, Heuss LT, Walker JA, Qi R (2005) Trained registered

nurses/endoscopy teams can administer propofol safely for

endoscopy. Gastroenterology 129:1384–1391

32. VanNatta ME, Rex DK (2006) Propofol alone titrated to deep

sedation versus propofol in combination with opioids and/or

benzodiazepines and titrated to moderate sedation for colonos-

copy. Am J Gastroenterol 101:2209–2217

33. Lichtenstein DR, Jagannath S, Baron TH, Anderson MA,

Banerjee S, Dominitz JA et al (2008) Sedation and anesthesia in

GI endoscopy. Gastrointest Endosc 68:815–826

34. Rex DK, Deenadayalu VP, Eid E, Imperiale TF, Walker JA,

Sandhu K et al (2009) Endoscopist-directed administration of

propofol: a worldwide safety experience. Gastroenterology

137:1229–1237 quiz 518–519

35. Dumonceau JM, Riphaus A, Aparicio JR, Beilenhoff U, Knape

JT, Ortmann M et al. European Society of Gastrointestinal

Endoscopy, European Society of Gastroenterology and Endos-

copy Nurses and Associates, and the European Society of Ana-

esthesiology Guideline: Nonanesthesiologist administration of

propofol for GI endoscopy. Endoscopy 42:960–974

36. Vargo JJ, Cohen LB, Rex DK, Kwo PY (2009) Position state-

ment: nonanesthesiologist administration of propofol for GI

endoscopy. Gastrointest Endosc 70:1053–1059

37. Imagawa A, Fujiki S, Kawahara Y, Matsushita H, Ota S, To-

moda T et al (2008) Satisfaction with bispectral index moni-

toring of propofol-mediated sedation during endoscopic

submucosal dissection: a prospective, randomized study. Endos

copy 40:905–909

38. Riphaus A, Stergiou N, Wehrmann T (2005) Sedation with pro-

pofol for routine ERCP in high-risk octogenarians: a randomized,

controlled study. Am J Gastroenterol 100:1957–1963

39. Collins SJ, Robinson AL, Holland HF (1996) A comparison

between total intravenous anaesthesia using a propofol/alfentanil

mixture and an inhalational technique for laparoscopic gynae-

cological sterilization. Eur J Anaesthesiol 13:33–37

40. Hogue CW Jr, Bowdle TA, O’Leary C, Duncalf D, Miguel R,

Pitts M et al (1996) A multicenter evaluation of total intravenous

anesthesia with remifentanil and propofol for elective inpatient

surgery. Anesth Analg 83:279–285

41. Matsumoto H, Shingu K, Numata K, Ogura S, Hanaoka K, Ito H

et al (1998) Total intravenous anesthesia with propofol is

advantageous than thiopental-sevoflurane anesthesia in the

recovery phase. Masui 47:1046–1058

42. Trinder TJ, Johnston JR, Lowry KG, Phillips AS, Cosgrove J

(1998) Propofol and alfentanil total intravenous anaesthesia: a

comparison of techniques for major thoracic surgery. Acta Ana-

esthesiol Scand 42:452–459

43. Juckenhofel S, Feisel C, Schmitt HJ, Biedler A (1999) TIVA mit

propofol/remifentanil oder balancierte anasthesie mit sevofluran/

fentanyl bei laparoskopischen operationen. Hamodynamik, au-

fwachverhalten und nebenwirkungen [TIVA with propofol-rem-

ifentanil or balanced anesthesia with sevoflurane-fentanyl in

laparoscopic operations: hemodynamics, awakening and adverse

effects]. Anaesthesist 48(11):807–812

44. Visser K, Hassink EA, Bonsel GJ, Moen J, Kalkman CJ (2001)

Randomized controlled trial of total intravenous anesthesia with

propofol versus inhalation anesthesia with isoflurane-nitrous

oxide: postoperative nausea with vomiting and economic analy-

sis. Anesthesiology 95:616–626

45. Rohm KD, Piper SN, Suttner S, Schuler S, Boldt J (2006) Early

recovery, cognitive function, and costs of a desflurane inhala-

tional vs a total intravenous anaesthesia regimen in long-term

surgery. Acta Anaesthesiol Scand 50:14–18

46. Lerman J, Johr M (2009) Inhalational anesthesia vs total intra-

venous anesthesia (TIVA) for pediatric anesthesia. Paediatr

Anaesth 19:521–534

Surg Endosc

123

47. Ledowski T, Bein B, Hanss R, Paris A, Fudickar W, Scholz J et al

(2005) Neuroendocrine stress response and heart rate variability:

a comparison of total intravenous versus balanced anesthesia.

Anesth Analg 101:1700–1705

48. Satani M, Hamada T, Nakada K, Umemoto Y, Fujii T, Takaki O

(2005) Comparison of total intravenous anesthesia and inhalation

anesthesia regarding hormonal responses during lung lobectomy.

Masui 54:1109–1115

49. Freeman L, Rahmani EY, Burgess RC, Al-Haddad M, Selzer DJ,

Sherman S et al. Evaluation of the learning curve for natural

orifice transluminal endoscopic surgery: bilateral ovariectomy in

dogs. Vet Surg 40:140–150

Surg Endosc

123