In-treatment stroke volume predicts cardiovascular risk in hypertension

Abstract

To evaluate whether lower stroke volume during antihypertensive treatment is a predictor of cardiovascular events independent of left ventricular geometric pattern. The association between left ventricular stroke volume and combined cardiovascular death, stroke and myocardial infarction, the prespecified primary study endpoint, was assessed in Cox regression analysis using data from baseline and annual follow-up visits in 855 patients during 4.8 years of randomized losartan-based or atenolol-based treatment in the Losartan Intervention For Endpoint reduction in hypertension (LIFE) echocardiography substudy. During follow-up, a total of 91 primary endpoints occurred. At baseline, lower left ventricular stroke volume was associated with smaller body size, female sex, lower left ventricular mass and stress-corrected midwall shortening, higher relative wall thickness and total peripheral resistance, more concentric left ventricular geometry and impaired diastolic relaxation (all P < 0.01). Baseline stroke volume did not predict outcome. However, in time-varying multivariable Cox regression analysis, lower in-treatment left ventricular stroke volume indexed for height was associated with higher risk of cardiovascular events {hazard ratio 1.69 per 1 SD (6 ml/m) lower stroke volume [95% confidence interval (CI) 1.35-2.11], P < 0.001} independent of in-treatment left ventricular mass and concentric geometry and in a secondary model also independent of stress-corrected midwall shortening, impaired diastolic relaxation, heart rate, new-onset atrial fibrillation and study treatment [hazard ratio 1.46 per 1 SD (6 ml/m) lower stroke volume (95% CI 1.13-1.88)]. Assessment of in-treatment left ventricular stroke volume may reflect cardiac and vascular remodeling and impairment and, hence, adds information on cardiovascular risk in treated hypertensive patients beyond assessment of left ventricular structure alone. NCT 00338260.