A prospective study of 137 patients with premature coronary disease was begun in 1965 to evaluate the influence of phenformin on mortality from cardiovascular disease. The subjects were treated with phenformin plus diet or diet alone. Twenty-five of the 137 patients (18%) died during the study. There were 9 deaths among 54 patients on diet (17%) and 16 deaths among 83 taking phenformin (19%). Cumulative survival showed no significant differences in any year between the two groups, despite higher mean base-line lipid levels in the phenformin group. In a subgroup of 104 patients entering after the first year and randomized to a phenformin or a diet group, 9 deaths among 50 patients (18%) on diet and 6 deaths among 54 (11%) taking phenformin occurred. There was no significant difference in cumulative survival between the two groups. In this study phenformin did not adversely influence mortality from cardiovascular disease.