2179Clinical significance of device-derived activity in ICD and CRT-D patients - Data from MADIT-RIT

Abstract

Abstract Background The Multicenter Automatic Defibrillator Implantation Trial – Reduce Inappropriate Therapy (MADIT-RIT) enrolled 1500 patients and showed that novel ICD programming reduced inappropriate therapy and improved survival. However, the role of device-derived patient activity to predict mortality is not known. Methods In 1500 patients enrolled in MADIT-RIT, device-derived patient activity was captured daily. Device-derived activity was averaged for the first 30 days following randomization, and utilized in this study to predict mortality post-30 days. Kaplan-Meier survival analysis and multivariate Cox proportional hazards regression models were used to evaluate all-cause mortality by 30-day device derived patient activity quintiles, and as a 3-level function of 30-day device derived patient activity (Q1, Q2–3, Q4–5). Results There were a total of 1463 patients with data available (98%), 66 of them died during the follow-up post-30 days. Patients in the lowest quintile (Q1: 4%∼1 hour daily activity) of device-derived activity were older, they were more often female, and they more often had diabetes and NYHA class III HF symptoms. Patients in the lowest quintile of 30-day device derived median activity (1 hour daily activity) had the highest risk of mortality, 15% in 2 years as compared to Q2–3 (1–2 hours daily activity, 8–7% 2-year mortality), and Q4–5 (>2 hours daily activity, 2–3% 2-year mortality) (Figure, p<0.001 for the overall duration). Each quintile decrease in device-derived 30-day median patient activity was associated with a significant, 41% increase in mortality (HR=1.41, 95% CI: 1.15–1.71, p=0.001). Patients with the lowest level of 30-day median patient activity (Q1) had 4.13-times higher risk of mortality as compared to the highest level of activity patients, Q4–5 (HR=4.13, 95% CI: 1.89–9.03, p<0.001). Patients with intermediate levels of activity (Q2–3) still had a 2.8-fold increase in death as compared to the highest activity level cohort of patients (HR=2.79, 95% CI: 1.31–5.91, p=0.008). Figure 1 Conclusions Device-derived 30-day median patient activity predicted subsequent all-cause mortality in ICD and CRT-D patients enrolled in MADIT-RIT. Patients with low and moderate levels of 30-day device-derived median patient activity (less than 2 hours daily activity) were at a significantly higher risk of death, and these cohorts warrant further investigation and management to improve outcomes. Acknowledgement/Funding MADIT-RIT was funded by an unrestricted research grant from Boston Scientific to the University of Rochester.