NON-INVASIVE VERSUS INVASIVE MANAGEMENT IN PATIENTS WITH PRIOR CORONARY ARTERY BYPASS SURGERY WITH A NON-ST SEGMENT ELEVATION ACUTE CORONARY SYNDROME: COMPARISONS BETWEEN THE RANDOMIZED CONTROLLED PILOT TRIAL AND REGISTRY

Background: Elderly patients with multi-morbidity are often under-represented in clinical trials. The CABG-ACS pilot trial (NCT01895751) prospectively assessed reasons for entering the trial or registry and subsequent outcomes. Methods: Patients with a non-ST segment elevation acute coronary syndrome (NSTE-ACS) and prior coronary artery bypass graft (CABG) admitted to 4 hospitals were randomized to invasive or non-invasive management. Non-randomized patients entered a follow-up registry. The primary efficacy outcome was a composite of all-cause death, rehospitalization for refractory ischemia/angina, myocardial infarction (MI) and hospitalization due to heart failure. The primary safety outcome was the composite of bleeding, stroke, procedure-related MI and worsening renal function. A blinded Clinical Event Committee independently assessed events. EuroQol-5 Dimensions (EQ-5D) was assessed at 6 monthly intervals for ≥18 months. Results: 217 patients with prior CABG and unplanned hospitalization for suspected ACS were screened. 84 subjects did not consent (≥1 reasons): 43 not NSTE-ACS, 35 unsuitable for invasive management, 9 refractory ischemia, 3 unable to consent. Of 133 eligible subjects, 60 (mean±SD age 71±9 years, 28% female) entered the trial and 73 (age 72±10 years, 27% female) entered the registry (preferences: physician 79%, patient 40% or both 18%). Compared to trial patients, registry patients had significantly more valve disease, lower hemoglobin, worse New York Heart Association class and higher frailty index. Baseline EQ-5D, medications and left internal mammary artery grafts were similar. Registry patients had significantly more medication changes due to recurrent angina and more urgent inpatient invasive procedures. The primary efficacy outcome occurred in 49% registry vs. 43% trial patients (HR (95% CI) 1.12 (0.77,1.63); p=0.601). Primary safety outcomes were similar (22% registry vs. 28% trial; HR 0.76 (0.42,1.38); p=0.425). EQ-5D health status was lower in the registry at 6 months (p=0.011) but not at 1 year (p=0.068). Conclusion: Compared to trial patients, the registry had excess morbidity but their longer term health outcomes were similar.