Effect of SAcubitril/Valsartan on left vEntricular ejection fraction and on the potential indication for Implantable Cardioverter Defibrillator in primary prevention: the SAVE-ICD study.

Abstract

PURPOSE\nSacubitril/valsartan has been associated with a positive reverse left ventricular remodelling in patients with heart failure with reduced ejection fraction (HFrEF). These patients may also benefit from an ICD implant. We aimed to assess EF improvement after 6\xa0months of treatment with sacubitril/valsartan, evaluating when ICD as primary prevention was no longer indicated.\n\n\nMETHODS\nMulticentre, observational, prospective study enrolling all consecutive patients with HFrEF and EF\u2009≤\u200935% with an ICD as primary prevention and starting treatment with sacubitril/valsartan (NCT03935087). Resynchronization therapy and patients experiencing appropriate ICD therapies before sacubitril/valsartan were excluded.\n\n\nRESULTS\nTwo-hundred-and-thirty patients were enrolled (73.9% males, mean age 64.3\u2009±\u200912.1\xa0years) After 6\xa0months of treatment, a reduction in left ventricular end-diastolic and end-systolic volumes was noted and LVEF increased from 28.3\u2009±\u20095.6% to 32.2\u2009±\u20096.5% (p\u2009<\u20090.001). At 6\xa0months, a non-ischemic aetiology of cardiomyopathy and a final dose of sacubitril/valsartan\u2009>\u200924/26\xa0mg twice daily were associated with a higher probability of an absolute increase of\u2009>\u20095% in LVEF. A total of 5.3% of primary prevention patients still had an arrhythmic event in the first 6\xa0months after treatment with sacubitril/valsartan started.\n\n\nCONCLUSIONS\nSacubitril/valsartan improves systolic function in HFrEF, mainly due to reverse left ventricular remodelling. Improvement in EF after 6\xa0months of treatment could help prevent ICD implantation in nearly one out of four patients, with important clinical and economic implications. However, the risk of sudden cardiac death in this recovered HFrEF population has not been thoroughly studied, and the present data should be interpreted only as hypothesis-generating.