Association between amount of biventricular pacing and heart failure status measured by a multisensor implantable de fi brillator algorithm

BACKGROUND Achieving a high biventricular pacing percentage (BiV%) is crucial for optimizing outcomes in cardiac resynchronization therapy (CRT). The HeartLogic index, a multiparametric heart failure (HF) risk score, incorporates implantable cardioverter-defibrillator (ICD)-measured variables and has demonstrated its predictive ability for impending HF decompensation. OBJECTIVE This study aimed to investigate the relationship between daily BiV% in CRT ICD patients and their HF status, assessed using the HeartLogic algorithm. METHODS The HeartLogic algorithm was activated in 306 patients across 26 centers, with a median follow-up of 26 months (25th - 75th percentile: 15 -37). RESULTS During the follow-up period, 619 HeartLogic alerts were recorded in 186 patients. Overall, daily values associated with the best clinical status (highest first heart sound, intrathoracic impedance, patient activity; lowest combined index, third heart sound, respiration rate, night heart rate) were associated with a BiV% exceeding 99%. We identified 455 instances of BiV% dropping below 98% after consistent pacing periods. Longer episodes of reduced BiV% (hazard ratio: 2.68; 95% CI: 1.02 -9.72; P = . 045) and lower BiV% (hazard ratio: 3.97; 95% CI: 1.74 -9.06; P=. 001) were linked to a higher risk of HeartLogic alerts. BiV% drops exceeding 7 days predicted alerts with 90% sensitivity (95% CI [74% -98%]) and 55% specificity (95% CI [51% -60%]), while BiV % < 96% predicted alerts with 74% sensitivity (95% CI [55% - 88%]) and 81% specificity (95% CI [77% -85%]). CONCLUSION A clear correlation was observed between reduced daily BiV% and worsening clinical conditions, as indicated by the HeartLogic index. Importantly, even minor reductions in pacing percentage and duration were associated with an increased risk of HF alerts.