Neonates placed on ECMO frequently develop fluid accumulation resulting in the need for diuresis. Intravenous furosemide is the most commonly used diuretic, and doses of 2-4 mg/kg/dose every six to eight hours may be needed. To date, no in vitro or in vivo studies have been performed evaluating furosemide disposition in neonates during ECMO.

An in vitro analysis was performed using four complete ECMO circuits to determine if furosemide is adsorbed to the ECMO circuit and, if so, to quantify the extent of the adsorption. The ECMO circuits (0.8 m2 membrane oxygenator and 1/4" polyvinyl chloride (PVC) tubing) were primed with plasmalyte to which 10 cc of 25% albumin was added. Two units of“super-packed” PRBC's (Hct>70) containing 50 cc THAM, 50 cc fresh frozen plasma, 300 mg Calcium Gluconate, and 100 units Heparin were added and allowed to circulate for 24 hours. A flow rate of 360 ml/min., a temperature of 37° C, and a pH of 7.3-7.6 were maintained throughout the study. Two doses of furosemide, 5 mg and 10 mg, were analyzed in duplicate. Furosemide was injected into the ECMO circuit immediately after the bladder. Blood samples of 2 ml each were drawn at two, four, six, 10, 15, and 30 minutes and at one, two, three, and four hours following drug administration. Samples were stored in amber glass vials at -20° C and analyzed within three weeks using HPLC. An analysis of variance with Duncan's Multiple Range Test was performed to compare the expected and measured serum furosemide concentrations.

Data analysis showed that 63-87% of the furosemide dose was adsorbed to the ECMO circuit (p=0.0001). Saturation of the ECMO circuit with furosemide occurred 30 minutes following drug administration. The need for large doses of furosemide in neonates on ECMO can be explained by the rapid adsorption of drug by the circuit.