Echocardiographic Predictors of Ventricular Tachyarrhythmias in Patients With Cardioverter-Defibrillator Implanted for Primary Prevention of Sudden Cardiac Death. Results From a two-Year Prospective Follow-up Study

Abstract

Aim      To compare variables of transthoracic EchoCG for determining echocardiographic predictors and their prognostic role in the development of persistent paroxysmal ventricular tachyarrhythmias (VT) in patients with ischemic CHF who had been implanted with a cardioverter defibrillator (CD) for primary prevention of sudden cardiac death.Material and methods  This single-site prospective study included 176 patients with CHF of ischemic origin aged 58.7±7.4 years with a left ventricular ejection fraction (LV EF) of 30 % [25; 34] % who had been implanted with CD. The follow-up duration was 24 months. The primary endpoint was a newly developed persistent paroxysm of VT (duration ≥30 sec) detected in the “monitored” VT area or a VT paroxysm that required electric treatment. The echocardiographic picture was evaluated by 28 variables. Statistical analysis was performed with the c2, Fisher’s, and Mann—Whitney tests, and the one-factor logistic regression (LR). Prognostic models were developed with a multifactorial LR. The model accuracy was evaluated by 4 metrics: area under the ROC (AUC), sensitivity, specificity, and diagnostic efficacy.Results The primary endpoint was observed in 60 (34 %) patients. Mean time to a persistent VT episode was 19.2±0.8 months (95 % confident interval (CI): 17.5–20.8). Superior-inferior dimensions of the right and left atria (RA and LA, respectively) and the left atrial volume (LAv) were independent predictors for VT. The odds of VT development in patients of the study cohort increased with RAl ≥4.5 cm (odds ratio (OR), 1.6; 95 % CI: 1.4–1.9; р=0.03), LAl ≥5.5 cm (OR, 2.5; 95 % CI: 1.01–6.1; р=0.04), LAv ≥95 ml (OR, 3.2; 95 % CI: 1.3–17.5; р=0.01). A comprehensive analysis of echocardiographic variables proved the prognostic potential of LAv that was linearly associated with the development of VT. The metrics of the best prognostic model were AUC 0.7±0.07 with 95 % CI: 0.54–0.83; specificity, 20.9 %; sensitivity, 95.7 %; and diagnostic efficacy, 47 %.Conclusion      This study allowed evaluation of capabilities of transthoracic EchoCG for predicting the probability of VT in patients with CHF of ischemic origin and reduced LV EF. It was shown that linear and volumetric atrial dimensions could be used for stratification of risk of VT and for determining the tactics for primary prevention of sudden cardiac death in this patient category.