LAUSR

Cardiac-resynchronization therapy (CRT) reduces morbidity and mortality in selected symptomatic patients with reduced left ventricular ejection fraction (LVEF) and wide QRS complex. However, some patients fail to benefit from CRT. Data on the differential role of baseline and follow-up left ventricular ejection fraction (LVEF) on outcome in patients with ischemic compared to non-ischemic cardiomyopathy (ICM, N-ICM) is controversial. To test, whether ICM and N-ICM patients differ in outcome after CRT during long-term follow-up and whether predictors for survival after CRT differ between the two groups. All patients undergoing CRT implantation at our institution between November 2000 and January 2015 were evaluated (n=418). All ICM/N-ICM patients with follow-up echocardiography within 1 year after CRT implantation (FU1) and a second echocardiography >1 year after FU1 (FU2) were included in the analysis (n=253). Primary post-hoc defined study endpoint was the composite of all-cause death, heart transplantation or implantation of a ventricular assist device. Compared to patients with N-ICM (n=160, median age 64 years [IQR 54–71], 71% male), ICM patients (n=93, median age 70 years [IQR 61–75], 84% male) were significantly older and had a higher prevalence of male gender, concomitant diabetes mellitus and arterial hypertension. There were no significant differences in pre-implantation echocardiographic features (LVEF, LVEDV, RV-FAC, severity of mitral regurgitation), QRS width and NT-proBNP levels between the groups. However, the hazard for reaching the primary endpoint was significantly higher in patients with ICM compared to N-ICM both on univariate analysis (HR 1.62 [95% CI 1.09–2.42], p=0.018) and after multivariate correction (aHR 2.13 [1.24–3.66], p=0.006). While higher NT-proBNP levels and greater right ventricular fractional area change were positively correlated with the hazard of death in both ICM and N-ICM (see Figure), lower LVEF at baseline was associated with an increased risk of death only in ICM but not in N-ICM (HR 0.95 [0.91–0.99], p=0.029 vs. HR 1.00 [0.96–1.04], p=0.945). Male gender, lower BMI and NYHA class ≥ III were positively correlated with the endpoint in N-ICM, but not in ICM. Importantly, LVEF at FU1 (median 4.7 months after implantation) and FU2 (median 47.1 months after implantation) were found to correlate signficantly with the endpoint in both ICM and N-ICM. Our findings highlight important differences in ischemic and non-ischemic patient populations undergoing CRT. While overall survival of patients with N-ICM exceeds survival in ICM, several other factors (including LVEF) have differential effects on response to CRT in these two patient groups.