Comparing Continuous Venovenous Hemodiafiltration and Peritoneal Dialysis in Critically Ill Patients with Acute Kidney Injury: A Pilot Study

Abstract

There are few reports on the role of peritoneal dialysis in critically ill patients requiring continuous renal replacement therapies. Patients with acute kidney injury and multi-organ involvement were randomly allotted to continuous venovenous hemodiafiltration(CVVHDF, group A) or to continuous peritoneal dialysis (CPD, group B). Cause and severity of renal failure were assessed at the time of initiating dialysis. Primary outcome was the composite correction of uremia, acidosis, fluid overload, and hyperkalemia. Secondary outcomes were improvement of sensorium and hemodynamic instability, survival, and cost. Groups A and B comprised 25 patients each with mean ages of 45.32 ± 17.53 and 48.44 ± 17.64 respectively. They received 21.68 ± 13.46 hours and 66.02 ± 69.77 hours of dialysis respectively (p = 0.01). Composite correction was achieved in 12 patients of group A (48%) and in 14 patients of group B (56%). Urea and creatinine clearances were significantly higher in group A (21.72 ± 10.41 mL/min and 9.36 ± 4.93 mL/min respectively vs. 22.13 ± 9.61 mL/min and 10.5 ± 6.07 mL/min, p < 0.001). Acidosis was present in 21 patients of group A (84%) and in 16 of group B (64%); correction was better in group B (p < 0.001). Correction of fluid overload was faster and the amount of ultrafiltrate was significantly higher in group A (20.31 ± 21.86 L vs. 5.31 ± 5.75 L, p<0.001). No significant differences were seen in correction of hyperkalemia, altered sensorium, or hemodynamic disturbance. Mortality was 84% in group A and 72% in group B. Factors that influenced outcome were the APACHE (Acute Physiology and Chronic Health Evaluation) II score (p=0.02) and need for ventilatory support (p < 0.01). Cost of disposables was higher in group A than in group B [INR 7184 ± 1436 vs. INR 3009 ± 1643, p < 0.001 (US$1=INR 47)]. Based on this pilot study, CPD may be a cost-conscious alternative to CVVHDF; differences in metabolic and clinical outcomes are minimal.