OBJECTIVE To test the safety and efficacy of fluoxetine in patients with renal failure on dialysis. METHODS Fourteen patients with major depression and end stage renal disease on hemodialysis were randomized into two groups for an eight-week study. Subjects as well as investigators were blinded as to which subject received fluoxetine and which placebo. Patients were carefully monitored concerning adverse events, serum fluoxetine and norfluoxetine levels, and psychological measurements of degree of depression. RESULTS No patients discontinued treatment because of adverse events, all of which were minor. All psychological tests showed improvement in depression at the four-week and eight-weeks point, although statistical significance could only be demonstrated at the fourth week of this study. All patients in the active group had serum plasma concentrations of fluoxetine and norfluoxetine less than 250 ng/ml at eight weeks, similar to levels in patients with normal renal function in a previous open label study. CONCLUSIONS This study confirms the relative safety of fluoxetine in depressed patients in renal failure on hemodialysis. It also suggests that fluoxetine may be efficacious in depressed patients on dialysis.