OBJECTIVE To compare acid-base balance, lactate concentration, and haemodynamic and O2 transport variables during haemofiltration with replacement fluid containing 44.5 mmol/l Na+ lactate or 40 mmol/l Na+ HCO3- and 3 mmol/l lactic acid. METHODS A prospective, randomized trial. METHODS A multidisciplinary, adult intensive care unit in a university hospital. METHODS Forty acidotic patients who required haemofiltration, were dependent on mechanical ventilation, and had PA catheters in situ. METHODS During haemofiltration patients received lactate or bicarbonate replacement fluid at a mean rate of 1.7 l/h (SD 0.3). Arterial blood gases, plasma lactate, and haemodynamic and O2 transport variables were measured before and after 12 and 24 h haemofiltration. Ultrafiltrate was collected for lactate estimation. RESULTS As means (SD). The net gain of lactate was 63 mmol/h (12 mmol) with Na+ lactate and 0 mmol/h (0.3 mmol) with Na+ HCO3-. There was a significant increase in pH and [lactate] in both groups, but [lactate] was higher in patients receiving lactate. Twenty-one patients survived to ICU discharge, these patients were significantly less acidotic after filtration (lactate group: 0 h: pH 7.23 (0.09), [lactate] 2.4 mmol/l (1.7); 12 h: pH 7.34 (0.09), [lactate] 4.7 mmol/l (2.4); 24 h: pH 7.36 (0.07), [lactate] 4.7 mmol (2.7). HCO3 group: 0 h: pH 7.23 (0.09), [lactate] 2.3 (1.3); 12 h: pH 7.32 (0.06), [lactate] 2.9 mmol/l (1.8); 24 h: pH 7.35 (0.08), [lactate] 2.8 mmol/l (2.0). Base deficit: survivors: 0 h: 9 mmol/l (4); 12 h: 2 mmol/l (3). Non-survivors: 0 h: 10 mmol/l (3); 12 h: 6 mmol/l (3)). Haemodynamic and O2 transport variables were not significantly affected by treatment group or outcome. CONCLUSIONS The degree of correction of acidosis during the first 24 h of haemofiltration was determined by patients outcome but was not affected by the substitution of bicarbonate- for lactate-containing replacement fluids.