Venovenous extracorporeal membrane oxygenation affects renal function

Abstract

Objective: We evaluated the effect of venovenous extracorporeal membrane oxygenation (ECMO) on renal function and fluid balance in neonates with severe respiratory failure.

Design: We retrospectively reviewed the charts of 30 consecutive patients who met criteria for treatment with ECMO. Twelve were managed without ECMO (comparison group) and 18 were treated with venovenous ECMO (treatment group).

Setting: The study was conducted in a single level III neonatal intensive care unit in a regional children's hospital accepting medical and surgical neonatal transfers. Our hospital does not have an inborn service.

Patients: Neonates were included if their gestational age was more than 34 weeks, they weighed more than 2 kg, and their respiratory failure was severe enough to warrant consideration of ECMO as a mode of support. All the neonates in this study were treated with high-frequency ventilation before being considered for ECMO; none were treated with nitric oxide. Criteria used to determine whether a neonate was a candidate for ECMO included: (1) alveolar-arterial oxygen difference greater than 60 kPa (610 torr) for 8 hours; (2) alveolar-arterial oxygen difference greater than 59 kPa (605 torr) and a peak airway pressure greater than 3.7 kPa (38 cm H2O) for 4 hours; (3) oxygenation index greater than 40 on three of five postductal blood gases obtained at least 30 minutes apart and unstable patient condition; or (4) refractory, severe respiratory failure with sudden decompensation (partial pressure of arterial oxygen 3.4 kPa or lower, 35 torr) despite maximal medical management for 2 hours. We did not include patients with congenital diaphragmatic hernia.

Main results: There were no differences between the groups in gestational age, birth weight, age at admission, gender, or diagnoses. Over the course of the 108 hours reviewed for each case, neonates treated with ECMO had higher positive fluid balance (P < .001), lower urine flow rates (P < .01), and higher blood urea nitrogen (P < .01) and creatinine (P < .01) levels than neonates managed without ECMO. There were no differences in mean blood pressure, protein intake, serum albumin, or use of diuretic therapy that might explain the differences between the groups.

Conclusion: We conclude that venovenous ECMO is associated with transient impairment in renal function and marked fluid retention.