Is It Time to Consider a “Time-Out” Before Primary Prevention Implantable Cardioverter-Defibrillator in Currently or Recently Hospitalized Older Patients with Heart Failure?

Background Trajectories of mortality after primary implantable cardioverter-defibrillator (ICD) placement for older heart failure (HF) patients during or soon after acute hospitalization have not been assessed. Objectives To compare trajectories of mortality after primary ICD placement during or soon after acute cardiac or non-cardiac hospitalization. Methods We identified older HF patients with primary ICD using 20% Medicare data (2008-2018). Placement settings were: 1) ‘current-H’ during current hospitalization, 2) ‘recent-H’ within 90 days of hospitalization, or 3) ‘chronic stable’. Hospitalization was categorized cardiac vs. non-cardiac. Interval mortality and hazard ratios (HRs) using Cox regression were estimated at 0-30 days, 31-90 days, and 91-365 days after ICD placement. Results Of 61,710 patients (mean age 76; 35% female; 85% White), 19%, 25%, and 56% had ICDs in ‘current-H’, ‘recent-H’, and ‘chronic stable’ settings. Mortality rates (per 100 person-years) were highest during 0-30 days with 38(34-42) and 22(19-24) for ‘current-H’ and ‘recent-H’, which declined to 21(20-22) and 16(15-17) during 91-365 days, respectively. Compared to ‘chronic stable’, HRs were greatest during 0-30 days post-ICD placement (‘current-H’ = 5.5[4.5-6.8], ‘recent-H’ = 3.4[2.8-4.2]) and decreased during 91-365 days (‘current-H’ = 2.0[1.8-2.1], ‘recent-H’ = 1.6[1.5-1.7]). HR pattens were similar for cardiac and non-cardiac hospitalization. Conclusion Primary ICD placement during or soon after hospitalization for any reason was associated with worse mortality with diminishing risks after 90 days. Hospitalization likely identifies a sicker population in whom early mortality with or without ICD may be higher. Our results support careful consideration regarding ICD placement during the 90 days following hospitalization.