Heart rate score and outcomes in ICD patients: insights from the prospective randomized INTRINSIC RV trial

Background Heart rate score (HRSc), the percentage of atrial sensed and paced beats in the largest 10 beat/min bin of a device histogram and mean intrinsic heart rate (MIHR), predicted survival in nonrandomized studies of implantable defibrillator (ICD) patients. We evaluated whether HRSc and MIHR independently predicted mortality and heart failure (HF) hospitalization in the prospective, randomized, controlled INTRINSIC RV trial. Methods and results The INTRINSIC RV trial enrolled 1530 patients receiving dual-chamber ICDs. This analysis involves patients ( n = 1471) for whom MIHR and HRSc data were available. Mean follow-up was 10.4 months. The relationship between pre-randomization MIHR and HRSc on the primary endpoint of all-cause mortality and HF hospitalization was assessed using multivariate regression and Cox modeling. As categorical variables, MIHR > 70 bpm and HRSc > 70% were considered high. Results The median baseline MIHR and HRSc were 74 (IQR = 16) and 50% (IQR = 20) respectively. As a continuous variable, for every 1% increase in HRSc, death/HF hospitalization increased by 1% (95%CI: 1.002–1.017; p = 0.01). Regression analysis showed baseline MIHR was associated with HRSc ( p = 0.01); for every 1 beat/min increase in MIHR, HRSc increased by 1.8%. A MIHR > 70 bpm and HRSc ≥ 70% predicted, but were independently associated with, the primary endpoint (HR: 1.39; 95%CI: 1.10–1.76, p = 0.005 for MIHR and HR: 1.654; 95%CI: 1.11–2.46, p = 0.01 for HRSc). Male gender (HR: 0.75), history of HF (HR: 1.29), and atrial fibrillation (HR: 1.37) also predicted death/hospitalization in the Cox model. Conclusions In this large, prospectively studied ICD population, HRSc was a robust and independent predictor of death/HF hospitalization. High MIHR and high HRSc were associated but each predicted outcomes independently.