Abstract 975: Left Ventricular Diastolic Dysfunction and Risk of Sudden Death in the General Population
Background: Left ventricular systolic dysfunction (LVSD) is an established predictor of sudden cardiac death (SCD) but recent studies show limited utility for risk stratification in the community. We therefore evaluated the potential effects of LV diastolic dysfunction (LVDD) on SCD in the general population.
Methods: SCD cases and controls ≥35 yrs were identified from an ongoing prospective evaluation of SCD in a population of one million in the northwest US (Feb 2002-Mar 2009). Analysis was limited to cases and controls with echocardiographic evaluation available; for cases the echocardiogram was prior to and unrelated to the arrest. LVDD was defined as combined LV hypertophy (LVH, >1.2 cm) and left atrial enlargement (LAE, >4cm). Univariate comparisons of cases and controls were evaluated using Pearson’s chi-square tests and independent samples t-tests. Logistic regression was used to evaluate the effect of LVDD on SCD, adjusting for age, gender, and LV function.
Results: Information on LA size and LVH was available for 131 cases and 141 controls. Cases were somewhat older (69.9 vs. 66.9, p=0.06), and the majority were male (68% and 64%, p=0.53). Mean LA size was greater in cases than controls (4.6 vs 4.2 cm, p=0.0005), while LA enlargement (>4 cm) was common in both cases and controls (68% vs 58%, p=0.09). Cases were more likely to have prevalent LVH (cases 35% vs controls 24%, p=0.047). Prevalence of severe LV dysfunction was similar in cases and controls (p=0.41). All controls had documented coronary artery disease (CAD); 94% of cases had documented or assumed CAD. In patients with normal LV systolic function, LVDD was more prevalent in cases (27%) than in controls (10%) (p=0.05). A similar association was observed in patients with mild-moderate LVSD (p=0.08), but the effect disappeared in patients with severe LVSD (p=0.43). In a multiple logistic regression model adjusting for age and gender, LVDD predicted SCA only in patients with normal LV function (OR 4.6, 95% CI 1.3–16.3).
Conclusions: LV diastolic dysfunction is a predictor of SCD risk particularly with normal LV systolic function. The potential for facilitating identification of the high risk patient with reversible coronary disease warrants further investigation.