Left ventricular global longitudinal strain is a long-term predictor of pacemaker-associated cardiomyopathy

Abstract

Abstract Background Implantation of a permanent pacemaker is an effective treatment for symptomatic bradycardia. However, chronic right ventricular pacing (RVP) may lead to compromised left ventricular (LV) function, atrial fibrillation, heart failure, and increased mortality via pacing induced electrical and mechanical dyssynchrony. A decrease in LV ejection fraction (LVEF) of ≥ 10% with a baseline LVEF > 50% prior to RVP and no alternative explanation for the decline in LVEF is termed pacing-induced cardiomyopathy (PICM). Predictive clinical and imaging-derived measures to identify patients at high risk for pacemaker induced cardiomyopathy (PMIC) are lacking, but essential. Objective The study is a prospective, single center, double-blinded trial evaluating the impact of right ventricular pacing on left ventricular function and clinical outcome. Methods In 105 patients (59 male / 46 female), who underwent a dual-chamber pacemaker implantation at the Charité Campus Virchow Klinikum, RVP burden, LVEF and left ventricular global longitudinal strain (LV-GLS) were obtained. Mean follow-up per patient was 29.6 ±13.9 months. We used a stepwise analytic approach including ROC analysis, univariate testing, binary regression as well as Cox-regression models and Kaplan-Meier analysis to compare low RVP burden (< 30%) and high RVP burden (≥ 30%) cohorts according to change in echocardiographic parameters and clinical outcome (all-cause mortality). Results Mean LVEF was 61 ± 6% and mean LV-GLS was 18 ± 4% at baseline. We identified 7 patients with a LVEF decline in the course of the study. In Pearson’s Chi-squared test, patients with a high RVP burden were more likely to develop LVEF and LV-GLS deterioration (6 patients versus 1 patient for LVEF, p<0.05; 28 versus 16 patients for LV-GLS, p<0.001). In multivariable binary logistic regression, LV-GLS (OR: 1.410, 95% CI: 1.201-1.610, p<0.001) deterioration was significantly associated with a LVEF decline and a high RVP burden ≥ 30% (OR: 1.358, 95% CI: 1.160-1.534, p<0.01). In a log-rank test, time to LVEF decline was significantly shorter in patients with a RVP burden ≥ 30% compared to a RVP burden < 30% (45.2±2.9 versus 55.7±1.0 months, p<0.05). In the Cox-regression analysis, LV-strain decline within 12 months (HR: 7.210, 95% CI: 4.239-9.516, p<0.05) was a significant predictor for a later LVEF decline. In a log-rank test, time to LVEF deterioration was significantly shorter in patients with LV-strain decline within 12 months compared to those without (34.7 ± 4.2 versus 53.7 ± 1.4 months, p<0.001). No significant difference in all-cause mortality between patients with a high RVP burden compared to those with low RVP burden (log-rank test, p=0.2). Conclusion A decline in LV-GLS is a sensitive long-term parameter of subsequent LVEF deterioration in patients with normal preimplant LVEF, who require RVP ≥ 30%. and is associated with a poor clinical outcome.