Abstract 031: Predictors and Cardiovascular Outcomes Associated with ECG-Evident Left Ventricular Hypertrophy in a Large Population

Abstract

Left ventricular hypertrophy that is evident on the electrocardiogram (ECG LVH) is a common condition affecting 1-5% of the general population and up to one third of patients with hypertension. However, there have been no studies examining the prevalence, correlates and associated cardiovascular disease (CVD) outcomes associated with ECG LVH in a very large and ethnically diverse population. In this study, we aimed to determine the prevalence of ECG LVH, its demographic and clinical correlates as well as its association with longitudinal CVD outcomes among members of Kaiser Permanente of Northern California. We performed a retrospective observational cohort study using an existing database of randomly selected, single ECGs taken on 1.4 million persons over age 30 as part of routine medical care between 1995 and 2009. We linked the index ECG data with other health plan databases, including hospitalizations and vital status through 8/31/12. The total accumulated follow-up time was 9.8 million person-years. ECG LVH was defined by sex-specific Cornell voltage Criteria (for men: S in V3 plus R in aVL >2.8 mV [28 mm]; for women: S in V3 + R in aVL >2.0 mV [20 mm] derived by processing raw ECG measurements from these two leads. The overall prevalence of ECG LVH was 5.1% (71,745 / 1,398,524), it was higher among women (7.4%; 55,296 / 743,466) than in men (2.5%; 16,449 / 655,058); in blacks (10%; 11,258 / 112,230) compared to whites (4.9%; 37,692 / 763,445), Asians (4.6%; 7,014 / 153,780) and Latinos (4.3%; 7,253 / 168,686); in persons over age 65 (9.3 %; 37,071 / 398,666) and in patients with a history of hypertension (9.2%; 46,415 / 502,462). In Cox regression, after excluding persons with prior cardiovascular events and after adjusting for age, gender, race/ethnicity, smoking status, obesity, diabetes, hypertension and hyperlipidemia, ECG LVH was independently associated with subsequent acute coronary syndrome (n=38,239; HR [hazard ratio]=1.66; 95% CI 1.60 to 1.73), cardiac arrest (n=12,144; HR=2.02; 95% CI 1.90 to 2.14), total stroke (n=49,518; HR=1.67; 95% CI 1.62 to 1.72), heart failure (n=47,294; HR=2.43; 95% CI, 2.37 to 2.50) and all cause-mortality (n=127,036; HR=1.78; 95% CI, 1.74 to 1.81). In conclusion ECG LVH is a relatively common ECG finding and a robust independent predictor of cardiovascular events and all-cause mortality in the largest population yet examined.