Long-Term Survival After Initial Invasive or Conservative Management of Stable Coronary Disease in Patients With Advanced Chronic Kidney Disease: ISCHEMIA-CKD EXTEND

BACKGROUND: In the ISCHEMIA-CKD trial (International Study of Comparative Health Effectiveness With Medical and Invasive Approaches-Chronic Kidney Disease), at a median follow-up of 2.2 years, the primary composite outcome of death or nonfatal myocardial infarction was similar between an initial invasive strategy (INV) compared with an initial conservative strategy (CON) in patients with moderate or severe myocardial ischemia and advanced chronic kidney disease. However, the long-term outcomes comparing the 2 strategies are unknown. METHODS: Participants in the ISCHEMIA-CKD trial were randomly assigned to either INV, involving coronary angiography and revascularization when feasible, added to guideline-directed medical therapy or to CON with guideline-directed medical therapy alone and invasive procedures reserved for failure of guideline-directed medical therapy. After the active intervention trial phase, long-term vital status follow-up was conducted through participant contact by participating sites and by health record and centralized death database searches. The primary outcome for the extended follow-up phase was all-cause mortality. Secondary outcomes were cause-specific mortality, including cardiovascular and noncardiovascular mortality. RESULTS: Among the 777 participants randomized (388 to INV and 389 to CON), over a median of 8 years of follow-up, vital status retention among survivors was 95%. There was a total of 429 deaths (55%). Cumulative all-cause mortality (57.4% versus 56.8%; adjusted hazard ratio, 1.05 [95% CI, 0.87-1.28]) or cardiovascular mortality (40.3% versus 38.4%; adjusted hazard ratio, 1.06 [95% CI, 0.84-1.34]) at 8 years did not differ between INV and CON. The restricted mean survival time was 1941 days (1834-2047 days) for INV versus 1990 days (1885-2094 days) for CON, with an adjusted estimated difference of -99 days (-245 to 46). The posterior probability of at least a 1 percentage point lower or higher death with INV at the 8-year median follow-up was 28% and 52%, respectively. Similarly, the posterior probability of at least a 1 percentage point lower or higher cardiovascular mortality with INV was 23% and 56%, respectively. CONCLUSIONS: In patients with at least moderate ischemia and advanced chronic kidney disease, there was no significant long-term benefit of an initial INV at reducing the risk of all-cause or cardiovascular mortality compared with an initial conservative strategy. REGISTRATION: URL: https://www.clinicaltrials.gov; Unique identifier: NCT01985360.