Survival and renal function in pediatric patients following extracorporeal life support with hemofiltration Robyn J. Meyer, MD, MS; Patrick D. Brophy, MD; Timothy E. Bunchman, MD; Gail M. Annich, MD, MS; Norma J. Maxvold, MD; Theresa A. Mottes, RN; Joseph R. Custer, MD E xtracorporeal life support (ex- tracorporeal membrane oxy- genation) is used to provide more comprehensive cardiac and pulmonary support to critically ill patients when less invasive means of sup- port are inadequate. For several reasons, patients receiving extracorporeal life sup- port may also require hemofiltration. As a result of their underlying critical illness, patients may experience multiple organ system failure, including renal failure (1,2,3), or may receive large volumes of fluid causing edema (4,5). Extracorporeal life support itself often necessitates the administration of large amounts of blood products, leading to fluid overload (4,5). Finally, patients on extracorporeal life support may have primary renal disorders (6), electrolyte imbalances (2), or meta- bolic disorders requiring hemofiltration. For patients who experience these diffi- culties, continuous hemofiltration with or without countercurrent dialysis can be provided through the existing extracor- poreal circuit. There is evidence that the addition of hemofiltration may be beneficial for some patients on extracorporeal life support. In studies of neonatal extracorporeal life support, decreases in body weight were paralleled by improvements in oxygen- ation and lung compliance. Hypervol- emia on extracorporeal life support was significantly improved by hemofiltration, and the clearance of edema was related to ability to wean off extracorporeal support (2,4). In a study of pediatric patients on extracorporeal life support, hemofiltra- tion resulted in improved oxygenation and cardiac output in addition to allow- ing for increased nutrition (2). Variables associated with survival have been reported for pediatric patients re- ceiving either extracorporeal life support (7,8,9,10) or hemofiltration (11). Vari- ables affecting survival may be different for patients receiving both hemofiltration and extracorporeal life support concur- rently. Variables associated with survival in pediatric patients receiving extracor- poreal life support with hemofiltration have not been clearly identified. In addition, data concerning recovery of renal function in survivors after hemo- filtration on extracorporeal life support is limited (2). There has been concern that hemofiltration may contribute to perma- nent impairment of renal function. Infor- mation concerning outcomes would as- From the Department of Pediatrics, Section of Pediatric Critical Care, University of Arizona, Tucson, AZ (Dr. Meyer) and the Department of Pediatrics, Di- visions of Pediatric Critical Care and Nephrology, Uni- versity of Michigan, Ann Arbor, MI (Drs. Brophy, Bunchman, Annich, Maxvold, and Custer and Ms. Mottes) Presented, in part, at the Pediatric Critical Care Colloquium, Portland, OR, September 1999. Copyright © 2001 by the Society of Critical Care Medicine and the World Federation of Pediatric Inten- sive and Critical Care Societies Objective: To determine variables associated with survival in pediatric patients treated with hemofiltration while receiving ex- tracorporeal life support and to determine the probability for recovery of renal function among survivors. Design: Retrospective database analysis. Setting: University of Michigan pediatric nephrology database. Patients: All pediatric patients treated with continuous hemo- filtration while on extracorporeal life support at the University of Michigan between January 1990 and May 1999. A pediatric pa- tient was defined as any child between birth and 18 yrs of age, including children treated in both the pediatric intensive care unit and neonatal intensive care unit. Indications for extracorporeal life support included both cardiac and pulmonary failure. Interventions: Data analysis of patients who were treated with hemofiltration while on extracorporeal life support. Hemofiltration includes both ultrafiltration and hemofiltration with countercur- rent dialysis. Measurements and Main Results: Thirty-five patients with a mean age of 39 6 65 months (median, 3 months) underwent hemofiltration while on extracorporeal life support. Forty-three percent survived to hospital discharge (95% CI, 26%– 60%). All deaths occurred in the intensive care unit. Recovery of renal function occurred in 93% of survivors (95% CI, 79%–108%). Mean duration of hemofiltration in survivors, including time during and after extracorporeal life support, was 9 6 6 days. All nonsurvivors were on renal replacement therapy at the time of death. In this analysis, decreased survival was significantly associated with the use of vasopressor infusions (p 5 .01) and the presence of complications (p 5 .006). Vasopressor infusions were required in 89% of patients, and 37% of patients experienced complications. Conclusions: In patients receiving hemofiltration while on ex- tracorporeal life support, survival is comparable to that reported in other extracorporeal life support or hemofiltration populations. Decreased survival in these patients may be associated with the use of vasopressor infusions and the occurrence of complica- tions. Recovery of renal function occurs in most survivors. (Pediatr Crit Care Med 2001; 2:238 –242) KEY WORDS: life-support system; extracorporeal membrane oxygenation; cardiopulmonary bypass; extracorporeal circulation; patient outcome assessment; mortality prediction; pediatrics; acute renal failure; hemofiltration; oliguria; hemodialysis; ultrafil- tration 238 Pediatr Crit Care Med 2001 Vol. 2, No. 3