An open, prospective evaluation of the effects of dichloroacetate on morbidity and survival time was done in 29 pediatric and adult patients with lactic acidosis. Dichloroacetate was administered intravenously over 30 minutes as two 50 mg/kg body weight doses separated by 2 hours. Five patients underwent retreatment with two additional drug doses and were considered new cases when analyzing for treatment response. Survival, however, was determined from the time of initial entry into the study. Patients were considered to respond to treatment if arterial lactate concentration decreased at least 20% from the pretreatment level within 6 hours of beginning the first dichloroacetate infusion. Using this criterion, 26 cases responded to therapy with dichloroacetate. For all cases, patients' mean arterial lactate concentration decreased 52% (P = 0.0009), arterial bicarbonate concentration increased 35% (P = 0.0003), and arterial pH increased (P = 0.024) to normal, defined as the range 7.35 to 7.45. Among responders, however, arterial lactate concentration decreased 74% (P = 0.0001), arterial bicarbonate level increased 47% (P = 0.0001), and arterial pH increased (P = 0.0004) to the normal range. Median survival time among responders was 60 hours, compared to 26 hours among nonresponders (P less than 0.001). There was no evidence of toxicity to dichloroacetate.