Applicability of a risk score for prediction of the long-term benefit of the implantable cardioverter defibrillator in patients receiving cardiac resynchronization therapy.

ABSTRACT

AIMS: The Goldenberg risk score, comprising five clinical risk factors (New York Heart Association class >2, atrial fibrillation, QRS duration >120 ms, age >70 years, and urea >26 mg/dL), may help identify patients in whom the survival benefit of the defibrillator may be limited. We aim at assessing whether this score can accurately predict the long-term all-cause mortality risk of patients receiving cardiac resynchronization therapy (CRT) and identify those who are more likely to benefit from the defibrillator. METHODS AND RESULTS: In this retrospective observational cohort study, 638 patients with ischaemic or non-ischaemic dilated cardiomyopathy who had CRT-defibrillator (CRT-D) (n = 224) vs. CRT-pacemaker (CRT-P) (n = 414) implantation were prospectively followed up for survival outcomes. The long-term outcome of patients with CRT-D vs. CRT-P was compared within risk score categories and in patients with severe renal dysfunction. Mean follow-up in surviving and deceased patients was 62.7 and 32.5 months, respectively. This score showed higher discriminative performance in all-cause mortality prediction in CRT-D vs. CRT-P patients (area under the curve 0.718 ± 0.041 vs. 0.650 ± 0.032, respectively, P = 0.001). In those with scores 0-2, a CRT-D device decreased mortality rates in the first 4 years of follow-up compared with CRT-P (11.3 vs. 24.7%, P = 0.041), but this effect attenuated with longer follow-up duration (21.2 vs. 32.7%, P = 0.078). In this group, the benefit of CRT-D during the follow-up was seen after adjusting for traditional mortality predictors (hazard ratio 0.339, P = 0.001). No significant differences in mortality rates were seen in patients with score ≥3 (57.9% with CRT-D vs. 56.9%, P = 0.8) and those with severe renal dysfunction (92.9% in CRT-D vs. 76.2%, P = 0.17). Similar results were seen following propensity score matching. CONCLUSION: A simple risk stratification score comprising five clinical risk factors may help identify CRT patients who are more likely to benefit from the presence of the defibrillator.