EHR

Tgflm(htrtEwanwoS

rmmldbsa

UA

2

xtracorporeal Life Support:istory and New Directions

obert H. Bartlett, MD

This review recounts the development of extracorporeal life support (ECLS, ECMO) fromthe laboratory and early clinical trials to routine clinical application. Lessons from neonatalECMO have led to better understanding of neonatal lung physiology, improved methods oftreatment, and application of ECLS to other patient populations.Semin Perinatol 29:2-7 © 2005 Elsevier Inc. All rights reserved.

KEYWORDS extracorporeal life support (ECLS), extracorporeal membrane oxygenation(ECMO), cardiopulmonary bypass (CPB), extracorporeal life support organization (ELSO),adult respiratory distress syndrome (ARDS)

ovctobac2ebtpisscagdpl

haftfcaw

he first neonatal extracorporeal membrane oxygenation(ECMO) survivor is 30 years old this year. In one medical

eneration, the use of prolonged extracorporeal circulationor life support in the intensive care unit has gone from aaboratory curiosity to clinical trials to routine practice. Every

ajor Children’s Hospital has an extracorporeal life supportECLS) program to sustain the life of patients with severeeart or lung failure when other treatments fail. Although theechnology was first developed for the care of neonatal respi-atory failure, growth and innovation are now primarily inhe pediatric and adult intensive care units. In fact, the use ofCLS for neonatal respiratory failure is steadily decreasing ase develop a better understanding of the pathophysiology

nd better methods of prevention and treatment in the neo-atal population. After three decades of experience, it isorthwhile to review the current status and future directionsf ECLS in neonatology, which is the subject of this issue ofeminars in Perinatology.

The heart/lung machine (a mechanical device to tempo-arily replace heart and lung function) was developed pri-arily by surgeon John Gibbon, beginning in 1939 and cul-inating in the first successful heart operation using a heart–

ung machine in 1954.1 Dr. Gibbon’s motivation was toevelop a technique to treat massive pulmonary embolism,ut what resulted instead was the entire field of intracardiacurgery. The artificial heart was simply a blood pump and thertificial lung was direct exposure of the flowing blood to

niversity of Michigan Medical Center, Ann Arbor, MI.ddress reprint requests to Robert H. Bartlett, MD, University of Michigan

Medical Center, 1500 E. Medical Center Drive, Taubman Center 2920,

bAnn Arbor, MI 48109-0331. E-mail: robbar@umich.edu

0146-0005/05/$-see front matter © 2005 Elsevier Inc. All rights reserved.doi:10.1053/j.semperi.2005.02.002

xygen gas. For cardiac surgery, all the venous return is di-erted into the machine and pumped into the systemic cir-ulation, leaving the heart empty long enough to repair in-racardiac defects or operate on the coronary circulation. Thepportunity to operate directly on the heart was miraculous,ut the heart–lung machine itself caused damage to the fluidnd solid elements of the blood. In fact, the heart–lung ma-hine caused fatal complications if it was used for more thanor 3 hours. The major cause of blood damage was the direct

xposure of blood to gas.2,3 Interposing a gas exchange mem-rane of plastic4 or cellulose5 between the flowing blood andhe gas solved most of the blood-damage problems, but ex-erimental devices required very large surface areas and were

mpractical for any clinical use.4-6 This changed when thinheets of dimethyl polysiloxane polymer (commonly calledilicone rubber) became available in the 1960s. Using a sili-one rubber membrane, artificial lungs with potential clinicalpplication were designed and studied.7-10 By eliminating theas interface it was possible to use the heart–lung machine forays at a time, and the physiology and pathophysiology ofrolonged extracorporeal circulation was worked out in the

aboratory.11-13

The first successful use of prolonged life support with aeart–lung machine was conducted by J. Donald Hill and hisssociates in 1971.14 The patient was a young man sufferingrom the Adult Respiratory Distress Syndrome. (ARDS washe name applied to a new syndrome of severe respiratoryailure complicating pneumonia, shock, trauma, or sepsis. Ofourse, ARDS had always existed, but was newly recognizednd named. The entire discipline we now call Critical Careas evolving at the same time. The concept of chronic intu-

ation and mechanical ventilation was only a decade old, and

wvrceppw1lpatmcavtictrHe

irhaypthwrmwawpnmittetyfBt

btfiard

etciamt1tlbpBEiyEcrcn

sbfwadiaboctpvputifiEFafifpwnhttrEl

Extracorporeal life support 3

as itself considered a radical intervention of questionablealue.) After Hill’s case, several other successful cases wereeported in children and adults with severe pulmonary andardiac failure.15 At the same time, there seemed to be anpidemic of “ARDS” and it looked like extracorporeal sup-ort would be the answer. A multi-center clinical trial ofrolonged extracorporeal circulation for adults with ARDSas commissioned by the National Institutes of Health in975. This was the first prospective randomized trial of a

ife-support technique in acute fatal illness in which the end-oint was death. There were many problems with the designnd execution of that clinical trial, but from it we learned thathe mortality for all patients with ARDS was 66%, and theortality for severe ARDS was 90%. We learned that extra-

orporeal support attempted by inexperienced teams in venorterial mode for 1 week did not improve the ultimate sur-ival in severe ARDS. We learned (the hard way) the mistakeso avoid when conducting a prospective trial in acute fatalllness. And we developed a name for the technology: extra-orporeal membrane oxygenation (ECMO). The results ofhat study were published in 197916; laboratory and clinicalesearch on ECLS in adults essentially stopped for a decade.owever, the results in neonatal respiratory failure were very

ncouraging.At the planning meeting for the adult NIH-ECMO trial

n 1975, we reported the first successful case of ECLS forespiratory failure in a newborn infant.17 Our laboratoryad been studying membrane oxygenator developmentnd prolonged extracorporeal circulation in animals for 10ears. We and others had used extracorporeal support forostoperative cardiopulmonary failure in children withhe first successful case in 1972.18 White19 and Dorson20

ad initiated clinical trials in neonatal respiratory failureithout success. Our patient was a full-term neonate with

espiratory failure secondary to meconium aspiration. Theajor cause of hypoxemia was pulmonary hypertensionith right to left shunting through the ductus arteriosus

nd foramen ovale. This was mysterious at the time, bute soon recognized that persistence of the fetal circulationhysiology (persistent pulmonary hypertension of theewborn, PPHN) is the underlying pathophysiology forost causes of respiratory failure in full-term newborn

nfants. Using techniques of vascular access, anticoagula-ion, ventilator management, and extracorporeal circula-ion which had been developed in the laboratory, we usedxtracorporeal support for infants with a variety of condi-ions. Forty newborn patients were treated over the next 5ears with 50% survival.21 Neonatologists and surgeonsrom other institutions joined us to learn the technology.y 1986, 18 neonatal centers had successful ECMOeams.22

As with intubation and mechanical ventilation for new-orn infants two decades before, the advent of this newechnology met with skepticism among the relatively neweld of neonatology. A prominent senior neonatologistnnounced at a national meeting that anyone pursuingesearch in ECMO was committing academic suicide. Pe-

iatric journals declined to publish reports of the growing E

xperience with ECMO, hence most of the reports fromhis early experience are in the artificial organs and surgi-al literature. We conducted the first prospective random-zed trial of ECMO in neonatal respiratory failure, using andaptive design to correct some of the mistakes we hadade in the earlier adult trial. After lengthy discussions,

he report of this trial was published in Pediatrics in985,23 along with invited commentaries criticizing theechnology and the trial.24 This criticism and controversyead to the second prospective randomized trial performedy O’Rourke and associates at the Boston Children’s Hos-ital.25 A similar adaptive trial design was used by theoston group with similar results (94% survival in theCMO group). The Michigan trial was criticized for expos-

ng critically ill infants to the high risks of ECMO. Twoears later, the Boston study was criticized for denyingCMO to the patients in the control group. Academicontroversies aside, neonatologists realized that ECMOegularly resulted in high survival rates of happy, healthyhildren and ECMO became standard treatment for neo-atal respiratory failure unresponsive to other measures.In 1990, the National Institutes of Health held a work-

hop on the diffusion of high-tech medicine from bench toedside using neonatal ECMO as an example.26 Severalactors were identified in the neonatal ECMO experiencehich facilitated the translation from concept to routine

pplication. All aspects of the technology were thoroughlyeveloped in the bench and animal laboratory before clin-

cal trials. Clinical trials progressed through Phases I, II,nd III with reporting and discussion at regular intervalsut without sensational reports in the lay press. A registryf all cases in all centers was kept from the beginning oflinical application, which proved to be very valuable ashe technology grew. The use of ECMO allowed study ofatients who would have been dead without it. This un-eiled many aspects of neonatal respiratory failure patho-hysiology and treatment, which in turn resulted in betternderstanding and the implementation of other simplerechniques. As the technology developed, it was standard-zed, disseminated, studied, and improved in an organizedashion by the actual and potential users. This group ofnvestigators and clinicians was formally organized as thextracorporeal Life Support Organization (ELSO) in 1989.or the last 15 years, that group has developed guidelinesnd practices, published the standard textbook in theeld,27 and maintained a registry of ECLS cases. ELSO hasollowed and documented the growth of ECLS in otheropulations and other diseases. As of January 2005, thereere over 30,000 patients in the registry, half of whom areewborn infants. Two other prospective randomized trialsave been reported.28,29 One of these is the large neonatalrial conducted in the United Kingdom.29 The design ofhis trial solved almost all of the problems of prospectiveandomized trials in acute fatal illness. Major advances inCLS technology and application have occurred over the

ast decade. This volume is focused only on the use of

CLS for neonatal respiratory failure.

CoEurpfioodpTctueswhtfoppsddfswiwaw

ctlcvo

LNBtoilmuWsb

tmrlpmecltmpwaofilcmfF

mdbfc

caedhtswmbi

wdfastmDutpctbp

E

4 R.H. Bartlett

urrent Statusf Neonatal ECMO

CMO is currently used for neonates with respiratory failurenresponsive to other treatment. Overall survival is 85%,anging from 98% in meconium aspiration to 55% in dia-hragmatic hernia. The use of ECLS in neonatal respiratoryailure has allowed the identification of aspects of pathophys-ology and treatment which should ultimately make ECMObsolete, or more accurately, unnecessary for the treatmentf neonatal respiratory failure. This is, in fact, happening asemonstrated by the number of neonatal ECMO cases re-orted by the Registry since the beginning of the technology.he use of ECLS peaked in 1992 and has been steadily de-reasing since that time. The primary reasons are identifica-ion and avoidance of ventilator-induced lung injury and these of inhaled nitric oxide as a pulmonary vaso dilator. In ourxperience, 40% of the patients who would have requiredupport with ECMO in 1992 are now successfully treatedith pressure-controlled (low stretch) ventilation and in-aled nitric oxide using standard or oscillation ventilatorechniques. Another factor is the successful use of surfactantor prematurity-related infant respiratory distress syndrome,bviating the need for extracorporeal support in most of theremature babies suffering from that condition. As other as-ects of treatment improve, those patients who do not re-pond (and are therefore treated with ECMO) are the mostifficult cases: bilateral hypoplasia secondary to congenitaliaphragmatic hernia, fulminant neonatal sepsis, pulmonaryailure complicating congenital heart disease, and the occa-ional patient who does not respond to gentle ventilationith nitric oxide inhalation. These problems are under study

n the laboratory and in clinical trials, and the need for ECLSill decrease even further in the future. However, there will

lways be some newborn infants with respiratory failure whoill not recover and survive without extracorporeal support.Today almost all of our patients are managed with a spe-

ially designed double lumen right atrial catheter placedhrough the internal jugular vein.30 This simplifies the vascu-ar access and avoids the complications of veno arterial ac-ess. Bleeding is rarely a problem, neurologic injury and de-elopmental delay are rare, and the technique is simply a partf routine management in the neonatal ICU.

essons from theeonatal Experience

asic research studies of prolonged extracorporeal circula-ion which began in the 1960s continue today in many lab-ratories and in industry. Elimination of the direct blood gasnterface solved most of the problems which limited pro-onged use of extracorporeal circulation. The advantages of

embrane oxygenators for prolonged use led to the routinese of membrane oxygenators for all of cardiac surgery.hen microporous membranes became available, laboratory

tudies showed that those surfaces appeared as solid mem-

ranes to flowing blood, even though there is direct exposure s

o gas through the micropores.31 Microporous materials areuch easier to handle and manufacture than solid silicone

ubber membrane, so that today almost all of the membraneungs used for cardiac surgery are made of microporousolypropylene, in hollow fiber configurations. This type ofembrane lung has greatly reduced the air and particulate

mbolism complications of open-heart surgery. Paradoxi-ally, microporous membrane lungs are not suitable for pro-onged extracorporeal support because plasma will leakhrough the micropores in an unpredictable fashion afterany hours of use. The cause of this plasma leakage is phos-ho lipid accumulation on the blood side of the membraneith subsequent “wetting” of the membrane and plasma leak-

ge.32 Basic research today has focused on the developmentf membrane lungs, which use solid membrane in hollowber configurations.33-35 These membrane lungs have very

ow blood flow resistance and very high gas exchange effi-iency. They are used for ECMO in Europe and in Japan withajor improvements in clotting and anticoagulation, platelet

unction, and ease of use. These devices have not yet receivedDA approval for use in the United States.Identification of the problems of direct gas interface led toore intense study of gas interfaces which still exist in car-iac surgery (open blood reservoirs and direct suction oflood from the operative field). Eliminating these gas inter-aces is another major step in improving the safety of extra-orporeal circulation for cardiac surgery.

Simple bench studies of the characteristics of vascular ac-ess catheters have led to the development and use of doublend single lumen catheters especially designed for prolongedxtracorporeal support. The management of anticoagulationuring clinical ECMO is based on laboratory studies inealthy animals. Heparin-bonded surfaces looked very at-ractive in the laboratory,36 but do not eliminate the need forystemic anticoagulation in patients. However, experienceith heparinized surfaces identified platelet adhesion as theajor problem in surface-induced thrombogenesis. Current

asic research is focused on plastic surfaces which are mod-fied to prevent platelet adhesion.37

All of these developments in basic bioengineering researchould be of minor importance if it were not for concomitantevelopment of devices in biomedical industry. To be uni-ormly useful, devices must be standardized, manufactured,nd generally available to practitioners at a reasonable cost. Aizable medical industry devoted to extracorporeal circula-ion has built up over the years, based on the much widerarket of devices for cardiac surgery. In 1976, the Food,rug, and Cosmetic Act was expanded to include FDA reg-lation of biomedical devices. Although very important forhe standardization and safety of devices for prolonged sup-ort, the market is relatively small and the expense of FDAertification has slowed the development of such devices inhe United States. The basic research continues everywhere,ut the transition from clinical implementation is now seenrimarily in Europe and Japan.We have learned a lot about clinical research from the

CMO experience. Principles and practices which apply to

imple interventions (such as a drug, an artificial joint, or a

cseoerccocbcwIwpaisrtr

bntsmhndoctdgwtwcpU

bFtltEtmos

aiaw

dmdi“ttvtcomsstshecwspnlocnpogmdpbpflC

TESfEHspwptagcfI

Extracorporeal life support 5

ardiac pace maker) do not apply to complex life supportystems like ECLS. Before beginning Phase I clinical trials, thentire hospital must learn and be prepared for the technol-gy. In the case of ECLS this includes the operating room, themergency room, the blood bank, the clinical laboratory,adiology, the patient transfer and referral system, the finan-ial administrative system, and all the ICUs and the intensiveare staff that would be involved with the trials. The technol-gy has to be thoroughly learned before beginning clinicalases. Protocols covering the possible complications have toe developed and practiced before starting clinical cases. Thelinical team may include perfusionists (who are very familiarith the technology but not with intensive care medicine),

CU nurses and respiratory therapists (who are very familiarith critical care but not with perfusion technology), andhysicians who may need to learn both perfusion technologynd ICU nursing regimens in addition to the complex phys-ology and pathophysiology of extracorporeal life support. Atudy of such a complex technology which includes inexpe-ienced centers will give poor results (like the 1970s ECMOrial). Such poor results can inhibit laboratory and clinicalesearch on a new technology for decades.

ECLS is the only life support technique which has evereen studied by prospective randomized trials. There hasever been, and never will be, a prospective trial of intuba-ion and mechanical ventilation, the use of antibiotics forepsis, the use of vasoactive drugs for shock, or the use ofechanical renal replacement for acute renal failure. ECLSas been studied in four prospective randomized trials ineonatal respiratory failure23,25,28,29 and two prospective ran-omized trials in adult respiratory failure.16,38 The problemsf designing a study of life support in acute fatal illness in-lude the ethics and the practicalities of assigning high-mor-ality risk patients to the control group, the logistics of twoifferent management protocols in a single hospital or a sin-le ICU, the logistics of standardizing patient entry criteriaithin diverse diseases, the logistics of patient transporta-

ion, and economics. Informed consent is a major problemhen the two arms of the study are almost certain death with

ontinuing conventional therapy versus temporary life sup-ort. Most of these problems were solved in the design of theK Neonatal ECMO Trial.29

As in any other clinical research project some of the sideenefits or “spin off” is as important as the primary project.or example, such common practices as bedside operationso ligate patent ductus arteriosis,39 avoiding stretch-inducedung injury,40 and the use of whole blood activated clottingime to measure heparin effect41 began with the neonatalCMO experience. Similarly, the use of oscillation ventila-

ion and inhaled nitric oxide for treatment of neonatal pul-onary hypertension grew out of the studies of pathophysi-

logy on ECMO patients42 and the availability of ECMO as aafety net when trials of the new therapy failed.

While using ECLS for life support, we all have learned a lotbout the biology of neonatal respiratory failure. When clin-cal trials of ECMO began in the 1970s and 1980s, respiratorylkalosis induced by high pressure, high volume ventilation

as standard treatment for neonatal respiratory failure. ECLS t

oes not provide any direct treatment but simply allowsodification of existing treatment. Turning the ventilatorown to low oxygen and low pressure settings usually results

n lung recovery. This observation, coupled with reports ofgentle ventilation with CO2 retention,”43 made it obvioushat ventilator techniques which were being used were part ofhe cause of ongoing lung injury. This changed the practice ofentilation in neonates, and a decade later led to a change inreatment of respiratory failure in children and adults. Unlikehildren and adults, however, the underlying pathophysiol-gy in most infants with neonatal respiratory failure is pul-onary vaso spasm with right to left shunting. This led to the

tudy of various inhaled pulmonary vaso dilators, the mostuccessful of which is nitric oxide gas. Similar studies iden-ified pulmonary hypertension as the primary culprit in re-piratory failure associated with congenital diaphragmaticernia. Two decades ago, standard practice was immediatemergency repair of diaphragmatic hernia, assuming theause of hypoxemia was the presence of the hernia itself. Nowe know that the presence of the hernia causes lung hypopla-

ia, which in turn leads to pulmonary hypertension withulmonary vasospasm. However, reduction of the hernia isow the last step in treatment rather than the first. This same

ine of investigation led to the observation that fetal trachealcclusion in animal44 and clinical45 diaphragmatic herniaauses lung growth and avoids the hypoplasia and pulmo-ary hypertension following delivery. Unfortunately, theractice of fetal operations still leads to significant incidencef premature labor and fetal loss, so pressure-induced lungrowth is no longer practiced clinically even in an experi-ental setting. However, pressure-induced lung growth iniaphragmatic hernia patients performed after birth whileatients are on ECMO support is not only feasible, is likely toe successful in solving this last remaining high-mortalityroblem in neonatal respiratory failure. Clinical trials ofuoro carbon-induced lung distention in ECMO-supportedDH patients are currently underway.46

he Future ofCLS in Neonatology

teps in prevention and treatment of neonatal respiratoryailure, discussed above, will decrease the need for and use ofCLS for respiratory failure in full-term newborn infants.owever, there will always be infants with severe neonatal

epsis, hypoplasia, diaphragmatic hernia, and pulmonary hy-ertension unresponsive to other methods of treatment, whoill survive and lead healthy lives after a period of extracor-oreal support. The use of fluoro carbon-induced lung dis-ention to cause lung growth in hypoplasia, as discussedbove, has the potential to result in routine survival in thisroup of patients. There are some full-term infants with in-urable lung disease, such as alveolo capillary dysplasia, pro-ound hypoplasia, and alveolar proteinosis in the newborn.n time these conditions will be treated with lung transplan-

ation. ECMO support will be needed from birth to the time

ou

titdadb

cpamiiosatdtctcgtvmtfpimcccttptpawt

tdwstproscv

utcrmHaarwItu

hhaitou

R

1

1

1

1

1

6 R.H. Bartlett

f transplantation, and in some cases, after transplantationntil the transplant is working successfully.ECMO will always be needed for severe cardiac failure in

he newborn. Whether bridging to recovery, bridging to anmplanted device, or bridging to cardiac repair or transplan-ation, extracorporeal support will be required for some chil-ren with severe cardiac malformations which are symptom-tic from birth. In addition, ECLS provides excellent supporturing shock caused by sepsis or hypovolemia in the new-orn.In the next decade, the most significant laboratory and

linical investigation will be in the area of ECLS for the veryremature, very low birth weight infant. Even with the avail-bility of surfactant as treatment, prolonged intubation andechanical ventilation are still required for many premature

nfants. The mortality and the incidence of intracranial bleed-ng and neurodevelopmental delay are significant. It is obvi-us that a fetus living in utero is much simpler and muchafer than a premature infant trying to live while breathing airnd subject to the complications of intensive care. Because ofhis, we go to great extremes to try to prevent prematureelivery. One approach being studied in the research labora-ory is modification of the ECMO system to an artificial pla-enta, permitting simulation of in utero growth rather thanrying to make a premature infant survive on extrauterineonditions. An artificial placenta is simply a membrane oxy-enator and hemofilter which is driven by arterial pressurehrough an arterial venous fistula, ideally via the umbilicalessels. Management of the extracorporeal gas exchange isodified to maintain fetal hypoxemia to encourage, rather

han eliminate, the physiology of pulmonary vaso spasm andetal circulation. Excretory and nutritional functions of thelacenta are fairly easily duplicated with a microporous filter

n the circuit. With such a system, growing fetuses would beaintained in a simulated artificial uterus (floating in artifi-

ial amniotic fluid and maintained in a small temperature-ontrolled, fluid-controlled, and electrolyte-controlledhamber). The goal of current research is to maintain prema-ure fetal animals in such a system until they reach maturity,hen “deliver” the fetus from the artificial uterus and thelacenta directly to normal newborn conditions. The impor-ance of this research related to neonatology includes theotential elimination of the complications of current man-gement of low birth weight premature infants. This systemould also greatly decrease the expense of managing prema-

ure infants who must survive outside the mother.These improvements in ECLS rely on improvements in the

echnology which are under study in many laboratories. Theevelopment of plastic which is inherently nonthrombogenicill permit the use of extracorporeal support without any

ystemic anticoagulation and without the complications ofhrombocytopenia, embolism, and bleeding. This would ex-and the potential applications of ECLS to newborn infants atisk for intracranial bleeding, newborns who require majorperations, etc. New techniques of vascular access undertudy are the use of tidal blood flow through a single lumenatheter, modified double lumen catheters to allow single

ein access, and techniques and devices to avoid spasm of the

mbilical vessels, allowing umbilical vessel access for prema-ure infants. Many of the potential complications of ECLS asurrently practiced relate to the use of the high-blood flowesistance, relatively thrombogenic, solid-silicone rubberembrane lung as standard practice in the United States.ollow fiber devices of various sizes using solid membranes

re available in Europe and Japan and will some day be avail-ble in the United States, greatly simplifying the extracorpo-eal circuit. This simplification will allow routine use of ECLSithout a specialized technical team as currently practiced.

n the future, ECLS will be managed as a routine part ofreatment, similar to the way that mechanical ventilators aresed today.In summary, the use of prolonged extracorporeal support

as prolonged the life of many critically ill infants intoealthy adulthood. In the process, we have identified manyspects of the technology of the devices and the pathophys-ology of neonatal respiratory failure which have improvedhe care of many other patients. Current research is focusedn simplifying the technology and applying it to other pop-lations, such as the very premature infants.

eferences1. Gibbon JH: Application of a mechanical heart and lung apparatus to

cardiac surgery. Minn Med 37:171, 19542. Lee WH Jr, Krumhar D, Fonkalsrud EW, et al: Denaturation of plasma

proteins as a cause of morbidity and death after intracardiac operations.Surgery 50:29-39, 1961

3. Dobell ARC, Mitri M, Galva R, et al: Biological evaluation of blood afterprolonged recirculation through film and membrane oxygenators. AnnSurg 161:617-622, 1965

4. Clowes GHA Jr, Hopkins AL, Neville WE: An artificial lung dependentupon diffusion of oxygen and carbon dioxide through plastic mem-branes. J Thorac Surg 32:630-637, 1956

5. Kolff WJ, Effler DB: Disposable membrane oxygenator (heart lung ma-chine) and it use in experimental and clinical surgery while the heart isarrested with potassium citrate according to the Melrose technique.ASAIO Trans 2:13-21, 1956

6. Pierce EC II: Modification of the Clowes membrane lung. J ThoracCardiovasc Surg 39:438, 1960

7. Kolobow T, Bowman RL: Construction and evaluation of an alveolarmembrane artificial heart lung. ASAIO Trans 9:238, 1963

8. Bramson ML, Osborn JJ, Main FB, et al: A new disposable membraneoxygenator with integral heat exchanger. J Thorac Cardiovasc Surg50:391, 1965

9. Landé AJ, Dos SJ, Carlson RG, et al: A new membrane oxygenator-dialyzer. Surg Clin North Am 47:1461, 1967

0. Bartlett RH, Isherwood J, Moss RA, et al: A toroidal flow membraneoxygenator: four day partial bypass in dogs. Surg Forum 20:152-153,1969

1. Kolobow T, Zapol W, Pierce J: High survival and minimal blood dam-age in lambs exposed to long term (1 week) veno-venous pumping witha polyurethane chamber roller pump with and without a membraneblood oxygenator. ASAIO Trans 15:172-177, 1969

2. Bartlett RH, Fong SW, Burns NE, et al: Prolonged partial venoarterialbypass: physiologic, biochemical and hematologic responses. Ann Surg180:850-856, 1974

3. Fong SW, Burns NE, Williams G, et al: Changes in coagulation andplatelet function during prolonged extracorporeal circulation (ECC) insheep and man. Trans Am Soc Artif Intern Organs 20:239-246, 1974

4. Hill JD, O’Brien TG, Murray JJ, et al: Extracorporeal oxygenation foracute post-traumatic respiratory failure (shock-lung syndrome): use of

the Bramson Membrane Lung. N Engl J Med 286:629-634, 1972

1

1

1

1

1

2

2

2

2

2

2

2

2

2

2

3

3

3

3

3

3

3

3

3

3

4

4

4

4

4

4

4

Extracorporeal life support 7

5. Bartlett RH: Extracorporeal life support for cardiopulmonary failure.Curr Probl Surg 27:621-705, 1990

6. Zapol WM, Snider MT, Hill JD, et al: Extracorporeal membrane oxy-genation in severe acute respiratory failure: a randomized prospectivestudy. J Am Med Assoc 242:2193-2196, 1979

7. Bartlett RH, Gazzaniga AB, Jefferies R, et al: Extracorporeal membraneoxygenation (ECMO) cardiopulmonary support in infancy. Trans AmSoc Artif Intern Organs 22:80-88, 1976

8. Bartlett RH, Gazzaniga AB, Fong SW, et al: Prolonged extracorporealcardiopulmonary support in man. J Thorac Cardiovasc Surg 68:918-932, 1974

9. White JJ, Andrews HG, Risemberg H, et al: Prolonged respiratory sup-port in newborn infants with a membrane oxygenator. Surgery 70:288-296, 1971

0. Dorson WJ, Baker E, Cohen ML, et al: A perfusion system for infants.ASAIO Trans 15:155, 1969

1. Bartlett RH, Andrews AF, Toomasian JM, et al: Extracorporeal mem-brane oxygenation (ECMO) for newborn respiratory failure: 45 cases.Surgery 92:425-433, 1982

2. Toomasian JM, Snedecor SM, Cornell R, et al: National experience withextracorporeal membrane oxygenation (ECMO) for newborn respira-tory failure: data from 715 cases. ASAIO Trans 34:140-147, 1988

3. Bartlett RH, Roloff DW, Cornell RG, et al: Extracorporeal circulation inneonatal respiratory failure: a prospective randomized study. Pediatrics4:479-487, 1985

4. Ware JH, Epstein MF: Extracorporeal circulation in respiratory failure(commentaries). Pediatrics 76:849-851, 1985

5. O’Rourke PP, Krone R, Vacanti J, et al: Extracorporeal membrane oxy-genation and conventional medical therapy in neonates with persistentpulmonary hypertension of the newborn: a prospective randomizedstudy. Pediatrics 84:957-963, 1989

6. Wright L: Report of the workshop on diffusion of ECMO technology.NIH Publication #93-3399, 1993

7. Zwischenberger JB, Stanborn RH, Bartlett RH: ECMO ExtracorporealCardiopulmonary Support in Critical Care(ed 2). Ann Arbor, MI, Ex-tracorporeal Life Support Organization, 2000

8. Schumacher RE, Roloff DW, Chapman R, et al: Extracorporeal mem-brane oxygenation in term newborns. A prospective cost-benefit anal-ysis. ASAIO J 39:873-879, 1993

9. UK Collaborative ECMO Trial Group: UK collaborative randomizedtrial of neonatal extracorporeal membrane oxygenation. Lancet 348:75-82, 1996

0. Zwischenberger JB, Toomasian JM, Drake K, et al: Total respiratorysupport with single cannula venovenous ECMO: double lumen contin-uous flow vs single lumen tidal flow. Trans ASAIO 31:610-615, 1985

1. Bartlett RH, Fong SW, Woldanski C, et al: Hematologic responses to

prolonged extracorporeal circulation (ECC) with microporousmembrane devices. Trans Am Soc Artif Intern Organs 21:250-257,1975

2. Montoya JP, Shanley CJ, Merz SI, et al: Plasma leakage through micro-porous membranes: role of phospholipids. ASAIO 38:M399-M405,1992

3. Peek GJ, Killer HM, Reeves R, et al: Early experience with a polymethylpentene oxygenator for adult extracorporeal life support. ASAIO J 48:480-482, 2002

4. Mueller XM, Marty B, Tevaearai HT, et al: A siliconized hollow fibermembrane oxygenator. ASAIO J 46:38-41, 2000

5. Tsukiya T, Tatsumi E, Nishinaka T, et al: Design progress of the ultra-compact integrated heart lung assist device. II. Optimization of thediffuser vane profile. Artif Organs 27:914-919, 2003

6. Toomasian JM, Hsu L-C, Hirschl RB, et al: Evaluation of Duraflo IIheparin coating in prolonged extracorporeal membrane oxygenation.Trans ASAIO 34:410-414, 1988

7. Annich GM, Meinhardt JP, Mowery KA, et al: Reduced platelet activa-tion and thrombosis in extracorporeal circuits coated with nitric oxiderelease polymers. Crit Care Med 28:915-920, 2000

8. Peek GJ: Conventional ventilation or ECMO for severe adult respiratoryfailure. Available at: http://www.cesar-trial.org. Established 2003.

9. Hubbard C, Rucker RW, Realyvasquez F, et al: Ligation of the patientductus arteriosus in newborn respiratory failure. J Pediatr Surg 21:3-5,1986

0. Kolobow T: On how to injure healthy lungs (and prevent sick lungsfrom recovering). ASAIO Trans 34:31-34, 1988

1. Bartlett RH, Isherwood J, Moss RA, et al: A toroidal flow membraneoxygenator: four day partial bypass in dogs. Surg Forum 20:152-153,1969

2. Inhaled nitric oxide in term and near-term infants: neurodevelopmen-tal follow-up of the neonatal inhaled nitric oxide study group (NINOS).J Pediatr 136:611-617, 2000

3. Wung JT, James LS, Kilchevsky E, et al: Management of infants withsevere respiratory failure and persistence of the fetal circulation withouthyperventilation. Pediatrics 76:488-494, 1985

4. Fauza DO, Hirschl RB, Wilson JM: Continuous intrapulmonary disten-sion with perfluorocarbon accelerates lung growth in infants with con-genital diaphragmatic hernia: initial experience. J Pediatr Surg35:1237, 2001

5. Harrison MR, Adzick NS, Bullard KM, et al: Correction of congenitaldiaphragmatic hernia in utero VII: a prospective trial. J Pediatr Surg32:1637-1642, 1997

6. Redmond CR, Heaton J, Calix J, et al: Correlation of pulmonary hyp-oplasia, mean airway pressure, and survival in congenital diaphrag-matic hernia treated with extracorporeal membrane oxygenation. J Pe-

diatr Surg 22:1143-1149, 1987