Abstract 974: Sudden Cardiac Death in San Francisco County, 2007: Coronary Artery Disease Accounts for Only 40% of SCDs but is Associated With a 3-Fold Higher Risk
Background: Sudden cardiac death (SCD) is the most common lethal manifestation of heart disease. Previous studies demonstrated that CAD underlies the majority of SCDs (~80%) with highly variable frequencies of active coronary lesions – from 20% to 80%. We hypothesized that the contemporary epidemiology of SCD in a diverse community, San Francisco (SF) County, differs substantially from historical data derived from homogenous populations, in particular reflecting a decreased contribution of CAD.
Methods: 3 physicians reviewed all deaths reported to the SF Medical Examiner (ME) in 2007 for SCD presentation: within 1 h of symptom onset (witnessed) or within 24 h of last having been symptom free (unwitnessed). Of these, we performed a comprehensive review of investigations and autopsy data to exclude non-arrhythmic SCDs (e.g. CHF deaths, drug overdose, PE). To evaluate the contribution of CAD, only SCDs undergoing autopsy were included in the analysis. To estimate prevalence in the general population, we identified a control group of demographically similar non-SCDs, accidental trauma deaths, the majority of which underwent autopsy. CAD was determined at autopsy using standard methods employed by the ME. Significant CAD was defined as > 75% stenosis in at least one coronary artery or an active coronary lesion (disrupted coronary plaque, luminal thrombus, or both); mild CAD was defined as < 75% stenosis in all coronary arteries.
Results: We identified 37 of 147 SCDs (mean 61.3 years, 70% male) and 192 of 238 accidental trauma controls (mean 56.7 years, 68% male) that underwent autopsy evaluation for CAD. Significant CAD was found in 15 SCDs (40.5%) and 38 controls (19.8%), p=0.007. After adjustment for age, sex, race, and BMI, the OR for presence of significant CAD was 3.05 (95% CI 1.31–7.08, p=0.01) for SCD. Only 1 of 37 (2.7%) SCDs and none of 192 controls had an active coronary lesion, p=0.17.
Conclusions: Active coronary lesion as the cause of SCD is far lower than previous estimates. Significant CAD accounted for only 40% of SCDs, half of recent estimates. Both results reflect the contemporary epidemiology of SCD and CAD in this diverse community. Nevertheless, significant CAD still imparted a three-fold higher risk of SCD as compared to accidental traumatic death.