Comparison of left ventricular lead upgrade vs continued medical care among patients eligible for cardiac resynchronization therapy at the time of defibrillator generator replacement: Predictors of left ventricular lead upgrade and associations with long-term outcomes

Randomized trials evaluating cardiac resynchronization therapy (CRT) have excluded patients with a pre-existing implantable cardioverter–defibrillator (ICD). The association of CRT upgrade with clinical outcomes in patients with a pre-existing ICD is unclear. The purpose of this study was to examine a CRT-eligible population to evaluate clinical outcomes associated with CRT upgrade compared to patients who did not undergo CRT. Using the National Cardiovascular Data Registry (NCDR) ICD Registry between April 2010 and December 2014, we created a hierarchical logistic regression model to identify predictors of CRT upgrade in a CRT-eligible ICD population. In the subpopulation of patients with Medicare-linked claims data, differential outcomes were determined with censoring at 3 years. The primary endpoint of this study was all-cause mortality, with secondary endpoints of rates of hospitalization and procedural complications. CRT upgrade was performed in 75.5% of CRT-eligible patients with pre-existing ICD (n = 15,803). Presence of left bundle branch block conduction was the strongest predictor of CRT upgrade (odds ratio [OR] 4.56; 95% confidence interval [CI] 4.08–5.11; P