Abstract 17989: Association of Serial Changes of Left Ventricular Diastolic Function with Cardiovascular and Non-cardiovascular Mortality Over Time

Abstract

Background: While the presence & severity of changes in diastolic function over time are related to heart failure, little is known about age-dependent longitudinal changes & their relationship to mortality. We assessed the hypothesis that serial changes in diastolic function correlate with mortality.

Methods: Analysis was based on a prospective community-based cohort of participants enrolled in the Olmsted County Heart Function Study. Randomly selected participants ≥45 yrs (n = 2042) underwent clinical evaluation, medical record abstraction, & comprehensive Doppler echocardiography (1997-2000). After 4 years, participants were invited to return (2001-2004) with n = 1402 of 1960 surviving, 72%) returning. All cause & disease specific mortality was assessed over a further 8±2 yrs. LV diastolic function was graded as normal, mild, moderate/severe by validated Doppler techniques.

Results: During the 4±0.3 years between exams, the prevalence of diastolic dysfunction increased from 23.8% (95% CI 21.2%-26.4%) to 39.2% (95% CI, 36.3%-42.2%; P < 0.0001). During follow-up, 9.2% died (3.1% related to cardiovascular disease (CV) and 6.1% from non-CV causes). All-cause mortality occurred in 4.6% (95% CI, 3.3%-6.5%) of persons whose diastolic function normalized or remained normal & in 14% (95% CI, 11.0%-17.7%) of those with progression to diastolic dysfunction (P < 0.0001). Those with persistent or progression to diastolic dysfunction had a higher rate of both CV (5.3% (95% CI, 3.5%-7.9%) and non-CV (8.7% (95% CI, 6.3%-11.8%) death compared to those with normal/normalized diastolic function (0.5% (95% CI, 0.2%-1.4%) and (4.2% (95% CI, 2.9%-6.0%) respectively. When adjusted for age, sex, hypertension and diabetes, diastolic dysfunction was associated with CV (hazard ratio, 3.27 [95% CI, 1.07-14.27]) but not non-CV death (hazard ratio, 0.90 [95% CI, 0.50-1.66]).

Conclusion: Abnormalities on serial assessment by Doppler echocardiography of diastolic function are associated with mortality. While diastolic dysfunction is related to non-CV death, this association appears to be a reflection of age & comorbidities rather than a direct association. However diastolic dysfunction is associated with CV mortality independent of age and comorbidities.