Cisapride is a prokinetic agent that is useful to relieve gastrointestinal symptoms that often occur in dialysis patients. However, sudden death has been reported and alarmed the use of cisapride in dialysis patients. This study was performed to identify whether the use of cisapride increases the risk of death. We retrospectively analyzed all records of dialysis patients followed during the period November 1997 to March 2000. Cisapride dosage and risk factors associated with increased risk of sudden death such as cardiovascular disease, hypokalemia, and possible interacting drugs were recorded. Of 364 dialysis patients, 85 had been prescribed with cisapride (group A) whereas 279 had not (group B). Group A patients were older, with more female patients and longer duration of dialysis. There was no significant difference in mortality or causes of death between the two groups after adjusting for the baseline demographic differences. For group A, eight patients (9.41%) died while still on cisapride and 19 (22.4%) died after cisapride had been stopped. The causes of death were peritonitis (n = 2), infection (n = 2), ischemic heart disease (n = 1), malignancy (n = 1), sudden death (n = 1), and unknown (n = 1). Low serum albumin (p=0.013) and hypokalemia (p=0.066) were potential predictors of death while taking cisapride, but the presence of diabetes mellitus, maximum dosage of cisapride, and underlying cardiovascular disease were not. There was no drug interaction leading to cisapride toxicity. Patients who were given cisapride were older, more often women, and had a longer duration of dialysis. Low serum albumin and hypokalemia were significant predictors of death in patients given cisapride. Although no excessive risk of death was documented, the use of cisapride in dialysis patients should still be cautious and potential drug interactions and electrolyte disturbances should be avoided.