A total of 101 patients (age 65 卤 11 years, 69 men, 18 New York Heart Association (NYHA) class IV, QRS 173 卤 23 ms) scheduled for CRT underwent clinical assessment, echocardiography, and brain-type natriuretic peptide (BNP) measurements before and 6.4 卤 2.3 months after CRT. Baseline SRSsept (all systolic stretch after initial shortening in the septum) was quantified by speckle tracking echocardiography. Primary composite end point was death, urgent cardiac transplantation, or left ventricular assist device implantation at the end of the study. Secondary end points were intermediate-term (6 months) response, quantified as decreases in left ventricular end-systolic volume (螖LVESV) and BNP (螖BNP). After a mean clinical follow-up of 15.6 卤 9.0 months; 23 patients had reached the primary end point. Baseline SRSsept (hazard ratio [HR] 0.742; 95%confidence intervals [CI] 0.601-0.916, P < .01]) was independently associated with a better outcome and NYHA class (HR 5.786: 95%CI 2.341-14.299, P < .001) with a worse outcome. Contrary to baseline NYHA class, baseline SRSsept was an independent predictor of both 螖LVESV (beta 0.53; P < .001) and 螖BNP (beta聽0.29; P < .01). Intermediate-term 螖LVESV and 螖BNP were associated with a favorable long-term outcome.
SRSsept at baseline is a strong, independent predictor of long-term prognosis after CRT and of improvements in left ventricular remodeling and neurohormonal activation at intermediate term.