Abstract
Background Although silent myocardial infarction (SMI) is prognostically important, the risk of sudden cardiac death (SCD) among patients with incident SMI is not well established. Methods and Results We examined 2 community-based cohorts: the ARIC (Atherosclerosis Risk in Communities) study (n=13 725) and the CHS (Cardiovascular Health Study) (n=5207). Incident SMI was defined as electrocardiographic evidence of new myocardial infarction during follow-up visits that was not present at the baseline. The primary study end point was physician-adjudicated SCD. In the ARIC study, 513 SMIs, 441 clinically recognized myocardial infarctions (CMIs), and 527 SCD events occurred during a median follow-up of 25.4 years. The multivariable hazard ratios of SMI and CMI for SCD were 5.20 (95% CI, 3.81-7.10) and 3.80 (95% CI, 2.76-5.23), respectively. In the CHS, 1070 SMIs, 632 CMIs, and 526 SCD events occurred during a median follow-up of 12.1 years. The multivariable hazard ratios of SMI and CMI for SCD were 1.70 (95% CI, 1.32-2.19) and 4.08 (95% CI, 3.29-5.06), respectively. The pooled hazard ratios of SMI and CMI for SCD were 2.65 (2.18-3.23) and 3.99 (3.34-4.77), respectively. The risk of SCD associated with SMI is stronger with White individuals, men, and younger age. The population-attributable fraction of SCD was 11.1% for SMI, and SMI was associated with an absolute risk increase of 8.9 SCDs per 1000 person-years. Addition of SMI significantly improved the predictive power for both SCD and non-SCD. Conclusions Incident SMI is independently associated with an increased risk of SCD in the general population. Additional research should address screening for SMI and the role of standard post-myocardial infarction therapy.