Sustained low-efficiency dialysis for critically ill patients requiring renal replacement therapy

The replacement of renal function for critically ill patients is procedurally complex and expensive, and none of the available techniques have proven superiority in terms of benefit to patient mortality. In hemodynamically unstable or severely catabolic patients, however, the continuous therapies have practical and theoretical advantages when compared with conventional intermittent hemodialysis (IHD).

We present a single center experience accumulated over 18 months since July 1998 with a hybrid technique named sustained low-efficiency dialysis (SLED), in which standard IHD equipment was used with reduced dialysate and blood flow rates. Twelve-hour treatments were performed nocturnally, allowing unrestricted access to the patient for daytime procedures and tests.

One hundred forty-five SLED treatments were performed in 37 critically ill patients in whom IHD had failed or been withheld. The overall mean SLED treatment duration was 10.4 hours because 51 SLED treatments were prematurely discontinued. Of these discontinuations, 11 were for intractable hypotension, and the majority of the remainder was for extracorporeal blood circuit clotting. Hemodynamic stability was maintained during most SLED treatments, allowing the achievement of prescribed ultrafiltration goals in most cases with an overall mean shortfall of only 240 mL per treatment. Direct dialysis quantification in nine patients showed a mean delivered double-pool Kt/V of 1.36 per (completed) treatment. Mean phosphate removal was 1.5 g per treatment. Mild hypophosphatemia and/or hypokalemia requiring supplementation were observed in 25 treatments. Observed hospital mortality was 62.2%, which was not significantly different from the expected mortality as determined from the APACHE II illness severity scoring system.

SLED is a viable alternative to traditional continuous renal replacement therapies for critically ill patients in whom IHD has failed or been withheld, although prospective studies directly comparing two modalities are required to define the exact role for SLED in this setting.