ISCHEMIA-EXTEND: early invasive strategy did not change 11-year mortality in 5179 coronary patients

Long-term follow-up results of ISCHEMIA, an international trial in chronic coronary disease, were published in JAMA. The trial randomized 5179 participants at 320 sites in 37 countries beginning August 2012, and follow-up continued through December 2025. All-cause mortality at 11 years did not differ significantly between an initial invasive strategy and an initial conservative strategy jamanetwork.comJAMAjamanetwork.comOpen source ↗.

Design and participants

Participants had chronic coronary disease with moderate or severe ischemia and a left ventricular ejection fraction of 35% or more. Patients with left main coronary artery disease, unacceptable angina or a recent acute coronary syndrome were not included. One group received guideline-directed medical therapy plus coronary angiography and revascularization when feasible. The other received guideline-directed medical therapy alone, with invasive management if medical therapy failed.

Median age was 64 years (IQR, 58-70); 23% of participants were female and 42% had diabetes. Median ejection fraction was 60% (IQR, 55%-65%). Data were analyzed by intention to treat, using cumulative incidence estimates, Cox and Fine-Gray models, bayesian analysis and restricted mean survival time.

Mortality results

Median follow-up was 9 years (range, 1 day to 13.4 years). In total, 1071 deaths occurred, 782 of them during extended follow-up: 672 cardiovascular, 359 noncardiovascular and 40 unclassified jamanetwork.comJAMAjamanetwork.comOpen source ↗.

All-cause mortality at 11 years was 25.8% in the invasive group and 26.9% in the conservative group (estimated difference, -1.0 percentage point; 95% CI, -4.2 to 2.1; adjusted hazard ratio, 0.97; 95% CI, 0.86-1.09). In the bayesian analysis, the posterior probability of a survival benefit of more than 1 percentage point for the invasive strategy was 18%.

Cardiovascular mortality at 11 years was 15.8% with the invasive strategy and 17.1% with the conservative one (adjusted hazard ratio, 0.88; 95% CI, 0.75-1.02). Noncardiovascular mortality was 9.1% and 8.1%, respectively (adjusted hazard ratio, 1.18; 95% CI, 0.96-1.45). Eleven-year restricted mean survival differed by 14 days (95% CI, -41 to 70 days), which was not significant.

The authors concluded that long-term all-cause mortality in patients with high-risk chronic coronary disease was not significantly different after routine invasive versus initial conservative management. The trial is registered at ClinicalTrials.gov, NCT04894877 jamanetwork.comJAMAjamanetwork.comOpen source ↗.