Since 1944, 91 patients (50 men and 41 women, mean age 68 years [range 39 to 86]) with ventricular septal rupture after acute myocardial infarction were seen at the Mayo Clinic. Patients were divided into 4 groups according to therapy and timing of surgical intervention. Fourteen patients seen before 1965, when surgery was not performed for such a complication or not readily available, were excluded from the analysis. Group 1 (n = 22) had surgery within 48 hours of septal rupture, group 2 (n = 6) underwent operation between 2 and 14 days, group 3 (n = 24) had surgery after 14 days, and group 4 (n = 25) only received medical treatment. Short-term (30 days) survivors (45%, 35 of 77 patients) were compared with nonsurvivors. Using logistic regression, by univariate analysis, 3 variables were significantly associated with outcome: age (p less than 0.01), cardiogenic shock (p less than 0.00001), and long delay between ventricular septal rupture and surgical intervention (p less than 0.004). By multivariate analysis, however, only cardiogenic shock (p less than 0.00001) and age (p less than 0.007) correlated with an adverse outcome. In patients with cardiogenic shock after septal rupture, the prognosis was uniformly fatal unless patients undergo early surgery. None of the 23 patients in groups 2, 3 or 4 survived, whereas 5 of 13 patients (38%) who had surgery within 48 hours of septal rupture survived.(ABSTRACT TRUNCATED AT 250 WORDS)