AUTHOR=Merkel Eperke D. , Schwertner Walter R. , Behon Anett , Kuthi Luca , Veres Boglárka , Osztheimer István , Papp Roland , Molnár Levente , Zima Endre , Gellér László , Kosztin Annamária , Merkely Béla TITLE=Predicting the survival benefit of cardiac resynchronization therapy with defibrillator function for non-ischemic heart failure—Role of the Goldenberg risk score JOURNAL=Frontiers in Cardiovascular Medicine VOLUME=9 YEAR=2023 URL=https://www.frontiersin.org/journals/cardiovascular-medicine/articles/10.3389/fcvm.2022.1062094 DOI=10.3389/fcvm.2022.1062094 ISSN=2297-055X ABSTRACT=Aims

Primary prevention of sudden cardiac death (SCD) in non-ischemic heart failure (HF) patients remains a topic of debate at cardiac resynchronization therapy (CRT) implantation requiring individual risk assessment. Using the Goldenberg SCD risk score, we aimed to predict, which non-ischemic HF patients will benefit from the addition of an implantable cardioverter defibrillator (ICD) to CRT at long-term.

Methods

Between 2000 and 2018 non-ischemic HF patients undergoing CRT implantation were collected into our retrospective registry. The Goldenberg risk score (GRS) was calculated by the presence of atrial fibrillation, New York Heat Association (NYHA) class > 2, age > 70 years, blood urea nitrogen > 26 mg/dl and QRS > 120 ms. The primary endpoint was all-cause mortality, heart transplantation or left ventricular assist device implantation.

Results

From 667 patients, 347 (52%) underwent cardiac resynchronization therapy-pacemaker (CRT-P), 320 (48%) cardiac resynchronization therapy-defibrillator (CRT-D) implantations. During the median follow up time of 4.3 years, 306 (46%) patients reached the primary endpoint (CRT-D 37% vs. CRT-P 63%; p < 0.001). CRT-D patients were younger (64 vs. 69 years; p < 0.001), infrequently females (26 vs. 39%; p < 0.001), and had a lower ejection fraction (27 vs. 29%; p < 0.01) compared to CRT-P patients. After GRS calculation, patients were dichotomized by low (< 3) and high (≥ 3) scores. CRT-D patients with low GRS showed a mortality benefit compared to CRT-P (HR 0.68; 95% CI 0.48–0.96; p = 0.03), high-risk patients did not (HR 0.84; 95% CI 0.62–1.13; p = 0.26).

Conclusion

In our non-ischemic cohort, patients with low GRS showed a clear long-term mortality benefit by adding ICD to CRT, however, in high-risk patients no further benefit could be observed.