Bicarbonate dialysate for continuous renal replacement therapy in intensive care unit patients with acute renal failure

Lactate-buffered peritoneal solution traditionally has been used as dialysate for continuous renal replacement therapy (CRRT) in the United States because no bicarbonate solution is commercially available. Since 1994, the Cleveland Clinic Foundation Dialysis Unit has prepared a bicarbonate solution (sodium 144 +/- 3 mEq/L, HCO3 37 +/- 2 mEq/L, potassium 3 or 4 mEq/L, calcium 3.0 +/- 0.3 mEq/L, and magnesium 1.4 +/- 0.3 mg/dL) replicating the dialysate for chronic intermittent hemodialysis. No solute precipitation, as calcium or magnesium salts, were observed, and several cultures of the solution, performed at various time periods, remained negative. Fifty critically ill acute renal failure patients have been treated with bicarbonate-CRRT. All patients were in multiple organ failure and required mechanical ventilation; 37 were receiving vasopressors. Forty-four continuous venovenous hemodialysis sessions and eight continuous arteriovenous hemodialysis sessions were performed with a mean duration of 7.8 +/- 6.1 days. The mean inflow dialysate rate was 1,249 +/- 225 mL/hr and the mean outflow rate (dialysate plus ultrafiltration) was 1,399 +/- 237 mL/hr; the inflow rate was constantly kept lower or equal to the outflow rate to avoid an enhanced potential for backfiltration. No related fever spikes or sepsis episodes were noted. The metabolic control achieved during bicarbonate-CRRT was good, with the following mean (+/-SD) daily values: blood urea nitrogen 70.3 +/- 29.0 mg/dL, creatinine 3.6 +/- 1.3 mg/dL, sodium 135.7 +/- 3.7 mEq/L, potassium 4.6 +/- 0.5 mEq/L, chloride 99.9 +/- 4.6 mEq/L, carbon dioxide content 21.4 +/- 3.4 mEq/L, calculated anion gap 14.4 +/- 4.8 mEq/L, arterial pH 7.39 +/- 0.05, arterial PCO2 36.6 +/- 5.4 mm Hg, total calcium 8.7 +/- 0.9 mg/dL (corrected for albumin 9.6), phosphorus 4.2 +/- 1.4 mg/dL, and magnesium 2.06 +/- 0.26 mg/dL. A subgroup of 13 patients was treated with two dialysate types, lactate-based solution (Dianeal 1.5%; Baxter Healthcare Corporation, Deerfield, IL) for 3.2 +/- 1.5 days and bicarbonate solution for 7.4 +/- 1.6 days, and the obtained metabolic control under both types of dialysate was compared. Mean values +/- SD (with probability values) obtained with lactate dialysate versus bicarbonate dialysate were as follows: blood urea nitrogen 77.6 +/- 34.4 mg/dL versus 71.0 +/- 20.8 mg/dL (P = NS), creatinine 4.1 +/- 0.9 mg/dL versus 3.3 +/- 1.6 mg/dL (P = NS), sodium 132.8 +/- 4.8 mEq/L versus 135.6 +/- 2.9 mEq/L (P = 0.04), chloride 95.8 +/- 5.4 mEq/L versus 98.5 +/- 4.2 mEq/L (P = NS), carbon dioxide content 17.8 +/- 3.1 mEq/L versus 21.8 +/- 3.4 mEq/L (P = 0.002), calculated anion gap 19.3 +/- 4.4 mEq/L versus 15.2 +/- 3.8 mEq/L (P = 0.008), arterial pH 7.36 +/- 0.07 versus 7.40 +/- 0.06 (P = NS), arterial PCO2 32.1 +/- 5.3 mm Hg versus 37.8 +/- 3.8 mm Hg (P = 0.01), total calcium 8.3 +/- 1.1 mg/dL versus 8.8 +/- 1.0 mg/dL (P = NS), phosphorus 4.7 +/- 1.3 mg/dL versus 3.8 +/- 1.4 mg/dL (P = NS), magnesium 1.95 +/- 0.14 mg/dL versus 2.04 +/- 0.34 mg/dL (P = NS), and glucose 200.5 +/- 80.4 mg/dL versus 146.7 +/- 40.4 mg/dL (P = 0.04). The bicarbonate solution is simple to prepare and is cost-effective. In our experience, its use as dialysate for CRRT is safe, free of complications, and provides an excellent metabolic control. (C) 1995 by the National Kidney Foundation, Inc.