The association between ventricular premature complexes (VPCs) detected on a rest 2-minute lead I electrocardiographic rhythm strip and sudden cardiac death (SCD), occurring within 1 hour of onset of symptoms, was evaluated in a prospective study of 15,637 apparently healthy white men, aged 35 to 57 years, at the first screening examination (1973 to 1975) to determine eligibility for the Multiple Risk Factor Intervention Trial in Minneapolis/St. Paul, Minnesota. The prevalence of any VPC was 4.4% (681 of 15,637). Over an average follow-up period of 7.5 years, a total of 381 deaths occurred. Of these, 34% (131 of 381) were ascribed to coronary artery disease (CAD) and 31% of the CAD deaths (41 of 131) occurred suddenly. The presence of any VPC was associated with a significantly higher risk for SCD (adjusted relative risk = 3.0; p < 0.025). On the other hand, the presence of any VPC was not associated with any significant increase in the risk of non-SCD or of total deaths from CAD (adjusted relative risks = 1.0 and 1.6, respectively). When VPC characteristics such as frequency (2 or more uniform VPCs every 2 minutes) and complexity (multiforms, pairs, runs, R-on-T) were examined, those with frequent or complex VPCs were at a significantly increased risk of SCD (adjusted relative risk = 4.2; p < 0.005), whereas for non-SCD no significant increase in risk was found (adjusted relative risk = 1.6; p = 0.28). The data also suggest that the presence of frequent/ complex VPCs carries a greater risk for SCD in men younger than 50 years of age than in those 50 years or older. The specificity of VPCs to SCD may bear an etiologic association and merits a careful evaluation of otherwise healthy men who demonstrate VPCs, particularly frequent/complex, on a short-term electrocardiographic recording.