Predictors of Left Ventricular Function Improvement in Patients Prescribed Sacubitril/Valsartan in a Tertiary-Care Community-Based Heart Failure Cohort

Abstract

Introduction Sacubitril/valsartan (ARNI) effects on outcome and quality of life in patients with HFrEF are well established. Limited data exist regarding which patient characteristics predict ARNI-mediated improvements in left ventricular function (EF). Methods Our cohort included 131 patients prescribed ARNI who had serial assessments of EF derived from 2D echocardiography at a community-based practice, Franciscan Health Indianapolis, from July 2016 to April 2018. Baseline patient characteristics predicting improvements in EF were assessed using univariate analysis. For continuous variables, logistic regression was used to determine predictors of EF improvement. Results Mean absolute EF improvement for the cohort was 6.7%. Presence of non-ischemic cardiomyopathy (NICM) vs ischemic cardiomyopathy (ICM) and absence of coronary artery disease (CAD) vs CAD were predictors of significant improvements in EF (3.5% vs 9.2%, p=0.003; and 4.6% vs 9.8%, p=0.009, respectively), while initiation of intermediate-dose vs low-dose ARNI and black vs caucasian trended toward significance (p=0.059 and 0.113, respectively) (see Table). No significant differences between NICM and ICM patients were observed with respect to baseline diastolic blood pressure (DBP), baseline EF, ARNI/beta-blocker dosing, duration of ARNI exposure, or presence of biventricular implantable defibrillator. Furthermore, higher baseline DBP (p=0.036) and lower baseline EF (p=0.0006) correlated with greater improvements in EF on ARNI. Conclusions In this retrospective cohort of HFrEF patients treated with ARNI, we observed greater improvements in EF in patients with a non-ischemic etiology as well as lower baseline EF and higher DBP. Interestingly, despite very low numbers of black patients, the largest absolute mean EF increase occurred in this group. Future studies exploring the mechanistic underpinnings for these findings and validation in a larger cohort are warranted.