Comparison of pump-driven and spontaneous continuous haemofiltration in postoperative acute renal failure

In a comparison of spontaneous continuous arteriovenous haemofiltration (CAVH) and pump-driven haemofiltration (PDHF) for acute renal failure after surgery, 116 patients admitted to a surgical intensive care unit were assigned CAVH (48) or PDHF (68). The method of assignment was that a patient was treated by PDHF if he or she was the only patient requiring treatment at that time (only one pump was available); any other patient coming to the unit would be treated by CAVH. The groups were slightly unbalanced because there were fewer simultaneous cases than expected. The main endpoints were survival rate, control of uraemia, and additional application of haemodialysis. There were no differences between the patient groups in age, duration of treatment, severity of illness, serum creatinine concentration at the start of treatment, or cause of acute renal failure. Both treatments adequately controlled uraemia and fluid overload. However, the survival rate was significantly higher with PDHF than with CAVH (6 [12.5%] vs 20 [29.4%]; p less than 0.05). The daily ultrafiltrate volume was significantly higher with PDHF than with CAVH (15.7 [95% confidence interval 13.6-17.8] vs 7.0 [6.6-7.4] l/day; p less than 0.05). The volume of ultrafiltrate in patients with ischaemic or sepsis-induced acute renal failure was correlated with the survival rate. This finding suggests that the better survival rate in the PDHF group was due to faster elimination of toxic mediators (of molecular weight 800-1000 daltons) through the filter membrane by high-volume haemofiltration.