There has been considerable debate regarding the value of programmed electrical stimulation in patients who present with asymptomatic, or minimally symptomatic, nonsustained VT. Unfortunately, there has never been a sufficiently large study of an untreated group of patients to make any sense of the issue. We culled the literature for reports published between 1986 and 1990 that met certain minimum requirements, the most important of which were adequate patient profiling and outcome data. The survey identified 12 studies of 926 patients, mean age 61 years, with a 5:1 male preponderance. Underlying heart disease was coronary in 818 patients, including 665 who had experienced previous but not recent myocardial infarctions. Of these, 302 (33%) had inducible sustained ventricular arrhythmias (monomorphic VTs in 264). Eighty-three percent of these patients were treated with antiarrhythmic drugs compared to only 13% of the noninducible group (p < 0.0001). Sudden death or a sustained arrhythmic event occurred in 54 (18%) of the 302 patients in the inducible group compared with 46 (7%) of the 624 in the noninducible group (p < 0.001). The sensitivity, specificity, and positive and negative predictive accuracies of the test were 54%, 70%, 18%, and 93%, respectively. Thus a patient with an inducible sustained arrhythmia who manifests nonsustained VT is two and a half times as likely to have a major arrhythmic event, but a negative result bodes well for the patient. However, widespread application of the technique cannot be recommended until these results are confirmed in a large, prospective study in which antiarrhythmic therapy is controlled. © 1992.