Risk stratifying asymptomatic left ventricular systolic dysfunction in the community: beyond left ventricular ejection fraction.

Abstract

Midwall fractional shortening (MWFS) is a measure of left ventricular (LV) systolic function that is more reliable in case of concentric LV geometry compared to LV ejection fraction (LVEF). We hypothesized that MWFS might predict heart failure (HF) and death in a high-risk asymptomatic population, beyond other echocardiographic parameters. Among 4047 subjects aged ≥55/≤80 years followed by 10 general practitioners in northern Italy, the DAVID-Berg study prospectively enrolled 623 asymptomatic outpatients at increased risk for HF. Baseline evaluation included clinical visit, electrocardiogram, N-terminal pro-brain natriuretic peptide (NT-proBNP), and echocardiogram. Mean age of the population was 69 ± 7 years, 56% were men, 88% had hypertension, mean LVEF was 61 ± 9%, and mean MWFS 16.2 ± 3.3. During a median follow-up of 5.7 years, 95 subjects experienced HF/death events. At Cox analysis, lower MWFS was the only echocardiographic parameter, among structural/functional ones, associated with higher risk of HF/death [hazard ratio (HR) 0.89, 95% confidence interval (CI) 0.84-0.95, Padjusted < 0.001]. The risk of HF/death related to clinical data and NT-proBNP (baseline model) was reclassified by echocardiography only when MWFS was included into the model (baseline C-statistics 0.761; adding conventional structural/functional echocardiographic data 0.776, P = 0.09; adding MWFS 0.791, P = 0.007). Compared to subjects with normal LVEF and MWFS, only subjects with combined systolic dysfunction (11% of the population) were at higher risk (P = 0.001 for both abnormal; P > 0.24 for either LVEF or MWFS abnormal). DAVID-Berg data suggest to include MWFS assessment in clinical practice, a simple and reliable echocardiographic parameter able to improve risk stratification in subjects at high risk for HF.