A REVIEW OF VENOVENOUS HEMOFILTRATION IN SERIOUSLY ILL INFANTS

Between April 1989 and October 1991, 13 severely ill infants, median a ge 13 days (range 1-180 days), median weight 3.5 kg (range 2.5-4.8 kg) , received continuous venovenous haemofiltration (CVVH) for a median d uration of 39 h (range 5-234 h). Filtration was performed through a do uble lumen catheter inserted into a central vein. The indications for filtration included acute renal failure (8), fluid overload (5), inbor n errors of metabolism (3) and sepsis (1). Some infants had more than one indication. The median Paediatric Risk of Mortality (PRISM) score on the day of admission to the intensive care unit was 27 (range 8-42) . No change in the level of respiratory support was required following the commencement of CVVH. Serum electrolyte concentrations and plasma osmolality remained normal throughout. Serum creatinine fell from a m ean of 0.11 mmol/L (95% Cl 0.058-0.168) to 0.07 mmol/L (Cl 0.034-0.112 ). Urea fell from a mean of 9.5 mmol/L (Cl 4.4-14.6) to 6.5 mmol/L (Cl 2.7-10.3). Platelet counts fell by 40-50% from a mean of 126 X 10(6)/ mm3 (Cl 72-180) to 69 X 10(6)/mm3 (Cl 36-103) 18 h following commencem ent of filtration but no bleeding was encountered. The main complicati on was a thrombosis of the superior and inferior vena cava in one infa nt. Four infants survived to be discharged from intensive care. Contin uous venovenous haemofiltration, with its inherent advantages over art eriovenous haemofiltration, is feasible in small infants using standar d paediatric equipment.