CDU Contributing Writer

NEW ORLEANS, Louisiana – Robert Bartlett, MD, pioneered the use of extracorporeal membrane oxygenation (ECMO) to treat meconium aspiration syndrome (MAS) in neonates at the University of Michigan (Ann Arbor, Michigan). Although ECMO initially proved to be of little value in treating respiratory distress syndrome in adults, new advances in the technique have now expanded its application beyond neonates to use in the cardiopulmonary bypass (CPB) sector. Although adult support rates still are much lower than those of neonates, ECMO is rapidly becoming a proving ground for advances in cardiopulmonary bypass. This was quite evident at this year's Extracorporeal Life Support Organization (ELSO; Ann Arbor, Michigan) conference, held here in September. Claude Mondiere, MD, of Doctors Hospital (Houston, Texas) showed how the control of the tension of pump tubing allows an extracorporeal circuit to provide self-regulating, pulsatile flow over a range of flow-volume needs, from neonatal ECMO to adult cardiopulmonary bypass. To accomplish this, silicone ovoid pump tubing of four different relaxed cross-sectional widths and two different lengths were used with rotors of three different sizes. The internal pressure of the circuit was set at 250 mmHg for all tubing sizes but one, where it was set at 120 mmHg. The rotor pump was run at 11 speeds ranging from 1 rpm to 50 rpm. Pump discharge volume was changed via one of two mechanisms. At a fixed pump speed, the axial tension of the pump tubing between the roller pins was changed, either by changing the inter-pin distance, or by increasing the stretch of the pump tubing over the same wheel at constant inter-pin distance. Alternatively, at a given axial tension of the pump tubing between the roller pins, the speed of the pump motor can be varied. In this way, the discharge of the system can range from 8 ml/min to as much as 6,350 ml/min.

Robert Remenapp, MD, of the University of Michigan, discussed the suitability of the Metaplus pump (currently sold by Jostra, formerly sold by Baxter Edwards, and originally sold by Avecor as the Affinity pump) for ECMO and prolonged extracorporeal life support. Metaplus is a variable-flow peristaltic pump that has a seamed polyurethane pump chamber stretched over a three-cammed rotor. The inlet of the pump fills passively by siphon and gravity. The discharge pressure of the pump is a function of the resistances of the membrane oxygenator and the arterial cannula, the bed height and the pressure within the vascular bed. This design blocks the transmission of negative pressure to the venous side of the circuit and prevents the over-pressurization of the arterial side of the circuit.

Although cerebral blood flow and changes in cerebral oxygenation during ECMO have been assessed using microspheres and ultrasonic Doppler, these techniques do not provide continuous measurement. Charles Hunter, MD, of Loma Linda University School of Medicine (Loma Linda, California), said he believes that continuous, real-time measurements of cerebral blood flow could detect previously undetected changes due to vessel ligation and the initiation of ECMO. In order to test his hypothesis, Hunter made laser Doppler flow measurements in lambs undergoing ECMO. He observed a transient (less than 90 second) fall in cerebral blood flow on both sides of the brain after ligation of the right common carotid artery, suggesting that rapid chemical and/or mechanical adaptations occur in response to abrupt changes in bilateral cerebral blood flow. Laser Doppler flow may prove to be a useful monitoring technique during ECMO in particular and during cardiopumonary bypass in general.

Brijesh Gill, MD, of the University of Texas Medical School (Houston, Texas) noted that trials of less-invasive respiratory support measures serve to delay the onset of ECMO in infants, resulting in fewer infants surviving to and from ECMO. A review of data from all patients with MAS in the national ECMO registry for the years 1989 to 1998 indicated that infants receiving ECMO more rapidly had both shorter ECMO runs and shorter hospital stays.

Jonathan Haft, MD, of the University of Michigan, detailed the performance of an artificial lung in patients with acute pulmonary hypertension, a condition associated with a variety of chronic lung and heart conditions. In particular, Haft demonstrated the effect of an artificial lung, applied in parallel with the native pulmonary circulation, on right ventricular load in the setting of pulmonary hypertension. In the study, sheep were connected to a low-resistance hollow fiber oxygenator made by MC3 (Ann Arbor, Michigan) via 16 mm tubing connect end-to-side with the pulmonary artery and the left atrium. It was found that more than 60% of total right ventricular output flows through the artificial lung, thereby meeting a large portion of total exchange requirements. Additionally, the artificial lung reduced mean pulmonary artery pressure, pulmonary vascular resistance, right ventricular power and the tension time index. All of this points to the artificial lung not only providing oxygenation but also improving right ventricular performance and efficiency.

Some institutions have used inhaled nitric oxide in neonates undergoing ECMO. Kenneth Bandy, BS, RPT, of the University of Michigan, evaluated the impact of and the cost associated with inhaled nitric oxide in a prospective, non-randomized cohort evaluated from 1996-1999. Of the 137 cases reviewed, 46% responded to nitrous oxide (NO), while 54% did not respond to NO; 39% received ECMO support, while 15% received other supportive therapies. Of the 46% that responded to NO, 94% survived with a sustained improvement in oxygenation. Non-responders to NO had only a 73% survival rate, with 50% mortality observed in those managed without the use of ECMO. It was concluded that neonates that do not respond to NO should receive ECMO support. In fact, it was recommended that NO therapy should not be initiated in any center that cannot provide ECMO support.

Extracorporeal support during cardiopulmonary resuscitation (ECPR) is a recent outgrowth of ECMO. The most effective application of ECPR remains to be defined. B.J. Taylor of Arkansas Children's Hospital (Little Rock, Arkansas) reviewed the experience with ECPR at his hospital, predominantly in neonates and children, to assess its efficacy as rescue therapy for respiratory and cardiac disease and to determine if outcomes could be correlated with any pre-ECPR variables. Overall, neonatal respiratory disease was found to have the best survival of ECPR, though long-term morbidity in this group was not assessed. Short-term survival was improved in all patients, but the effectiveness of ECPR seems to be age- and diagnosis-related. Outcomes did not seem to depend upon the degree of acidosis and hypoxemia, but there was so much variability among the 77 patients in the study that true assessment of this relationship requires the study of a much larger group of patients.

Craig Reikert, MD, of the University of Michigan, reviewed experience in ECPR at his institution, based on the charts of all patients treated from January 1996 to May 2000. Among the 75 patients identified (47 adult, 16 pediatric, and 12 neonatal), ECPR was attempted, but could not successfully be completed in five (four adults, one pediatric). Survival was 23% for adults, 25% for pediatric patients and 33% for neonates. Survival did not correlate with blood gases or with specific diagnosis. Survival was the same in patients in impending cardiac arrest and those with ongoing CPR. Within 24 hours, ECPR was discontinued in 34% of adult patients and 43% of pediatric patients. Reikert said he would like to have some way of predicting who is a successful candidate for ECPR and who isn't. In addition, according to Judy Jones, RT, of Children's Hospital-New Orleans (New Orleans, Louisiana), veno-arterial ECMO is increasingly becoming an option for patients who cannot be easily weaned off of conventional cardiac bypass.

The occurrence of TB pneumonia progressing to respiratory failure is rare and so are the cases which require ECMO support. Michael Herd, MD, of Emory University School of Medicine (Atlanta, Georgia) presented the case of a 15-year-old black female with a previous history of TB exposure, positive PPD and nine months of treatment with INH, admitted with severe respiratory failure, pleural effusion and pneumothorax of unknown cause. Because of continued difficulties with ventilation, venovenous ECMO was initiated on the patient with an initial maximum flow of 71 ml/kg/min while lung ventilation was continued for four days. The patient was treated with INH, rifampin, pyrazinamide and ethambutol and was decannulated after 152 hours of ECMO. The patient went on to have a candidal infection and developed a left pneumothorax. She was extubated eight days after ECMO was discontinued.

Gregor Alexander, MD, of the Arnold Palmer Hospital for Women and Children (Orlando, Florida), presented a case on the use of ECMO after neonatal myocardial infarction, demonstrating that ECMO can be used effectively for initial stabilization and management. ECMO allowed time for the recovery of ventricular function and for the post-natal decrease in pulmonary vascular resistance. Richard Powers, MD, of Children's Hospital (Oakland, California) reported the survival of two infants with Bordetella pertusis infections who underwent prolonged courses of ECMO to manage the complication of pulmonary hypertension. One infant was a six-week-old male who was intubated and ventilated for 72 hours prior to the start of veno-arterial ECMO for poor left ventricular function and was treated with erythromycin. This infant remained on ECMO for 19 days and was discharged 84 days after the end of ECMO. The other infant, a six-week-old female, was intubated and ventilated for less than 24 hours prior to the start of veno-arterial ECMO for refractory hypoxia. The patient remained on ECMO for 31 days and was discharged 73 days after the end of ECMO. From these two patients, who had remarkably similar courses, Powers concluded that severe pulmonary hypertension associated with pertusis infection is not necessarily fatal, and weaning from ECMO can be successfully accomplished even in the face of significant pulmonary hypertension.

Adnan Bhutta, MD, of Children's Hospital in Little Rock, Arkansas, reported the survival of an infant with Pneumocystis carinii who was treated with ECMO for respiratory distress syndrome. The patient was a 3.5-month-old female who was immunosuppressed on high-dose oral steroids after two laser treatments for the removal of large cutaneous hemangiomas on the left side of her face. The patient came to the emergency room after three days of coughing, fever and nasal congestion, and one day of difficulty breathing and grunting. She was admitted to the pediatric intensive care unit for respiratory monitoring and within a few hours, she was intubated for worsening respiratory distress. She was placed on veno-arterial ECMO on day 11 of her hospitalization because her oxygenation did not improve on high-frequency jet ventilation. The seven-day ECMO run led to improvement in her lung disease without complication. The patient was extubated eight days after coming off ECMO and then received pulmonary rehabilitation until discharge.

Until recently, ECMO was performed using veno-arterial cannulation. Now cannulas are available allowing sufficient flow for veno-venous ECMO. In cases where patients deteriorate while on veno-venous ECMO, conversion to veno-arterial ECMO is required. The University of Michigan's Reikert reviewed the experience with patients who required conversion from veno-arterial to veno-venous ECMO. He found that patients converted within the first 24 hours generally had inadequate circulatory support due to either underestimated cardiac pathology or inadequate blood flows. Those patients converted after 24 hours fell into two groups: those with acute decompensation due to a single, life-threatening event, converted due to cardiac arrest, super-ventricular tachycardia or pulmonary hemorrhage, and those with steady deterioration due to acidosis, hypoxemia or an increasing need for drugs to increase blood pressure. Conversion after 24 hours of ECMO was uniformly associated with an adverse outcome in adults. Survival was poor in adults (13%) requiring conversion, but survival was reasonable in neonates (64%) and pediatric patients (40%). It was thought that earlier conversion might improve outcomes for this steadily deteriorating group by offering earlier improvements in oxygenation and perfusion.

Venous cannula size may be a critical determinant of mortality in newborn ECMO patients, according to Tom Pullano, MD, of Presbyterian Hospital (Dallas, Texas). Pullano came to this conclusion by reviewing the National Registry Database of ELSO for the years 1985 to 2000 in order to determine the survival of patients in whom an 8 Fr venous cannula was placed. The data revealed that if a venous cannula greater than 8 Fr cannot be placed at the time of cannulation, mortality is substantially increased. Although the basis of this finding is not clear, Pullano postulated that there may be developmental restriction of the internal jugular vein or other anomalous vasculature structures in patients in whom a cannula that is at least 8 Fr cannot be placed.

Nevertheless, there is, as yet, no good way of predicting mortality and outcomes in ECMO patients. Cathy Ducato, RN, of Children's Hospital (Buffalo, New York) sought to assess the validity of the well-accepted PRISM score in patients undergoing ECMO. The charts of 42 patients (more than 75% of them non-cardiac) have been reviewed so far. A PRISM score was calculated from data available on the first day of ECMO support. Death occurred in 17 patients, compared to the 15 patients predicted by PRISM. Statistical analysis of the data revealed that the relationship between the PRISM score and mortality is random.

Several presentations concerned methods for increasing the efficiency with which patients undergoing ECMO are treated. Paul Scott, RN, of Vanderbilt Children's Hospital (Nashville, Tennessee) demonstrated that the use of the internal jugular artery, cannulated to provide venous access in ECMO, could be safely used after ECMO for insertion of a silastic (Broviac) catheter to provide central venous access for medications, blood drawing and parenteral nutrition.

ECMO is beginning to have applications outside the cardiac operating room. Wen-Je Ko, MD, of the National Taiwan University Hospital (Tapei, Taiwan), reviewed opportunities for the use of ECMO in organ transplantation. In a series of 24 lung transplants done at his institution, 15 required ECMO, of which 14 received ECMO. ECMO provided less bleeding, less transfusion and stable cardiopulmonary status throughout the transplantation procedure. ECMO could also be used to sustain non-heart beating organ donors, according to Ko. At his institution, six organ donors were under ECMO support for nearly an hour before abdominal organ procurement. Among these donors, 12 kidneys and one liver were harvested and transplanted. All of these grafts had immediate function after transplantation, except in three cases where the recovery of renal function required dialysis support. Finally, Ko noted that ECMO could be used during heart transplantation to maintain adequate perfusion, thus avoiding ischemic reperfusion injury and acute rejection. During the early transplant period, Ko successfully used IABP on eight heart recipients and ECMO on five heart recipients. Primary indications for ECMO in heart transplant recipients are primary graft failure and acute resistance. Ko said he believes that ECMO is a necessity for every center doing organ transplantation.

Samir Awad, MD, of the University of Michigan, reported on early work using ECMO to clear toxins such as ammonia, aromatic amino acids, free fatty acids and billirubin in patients with acute hepatic failure. Awad believes that continuous veno-venous hemodiafiltration with albumin dialysis should decrease elevated levels of hepatic toxins, reverse the ratio of unbranched chain to aromatic amino acids, reverse hepatic encephalopathy and maintain stable hemodynamics. In a Phase I clinical trial, nine patients with acute hepatic failure were placed on continuous veno-venous hemofiltration with continuous counter-current dialysis using a 10% albumin solution flowing through an activated charcoal cartridge. Serial blood samples were collected to look at toxin levels in the blood. Results of the trial indicated that a modified ECMO system can clear some of the toxins implicated in hepatic encephalopathy with minimal hemodynamic effects and no mechanical complications.