REVascularisation of Ischaemic VEntricular Dysfunction (REVIVED): a randomised comparison of percutaneous coronary intervention (with optimal medical therapy) vs. optimal medical therapy alone for heart failure secondary to coronary disease

DESIGN: prospective, randomised, controlled, multi-centre trial HEALTH TECHNOLOGY BEING ASSESSED: It is unclear whether any form of revascularisation is beneficial in heart failure due to coronary artery disease (CAD). Coronary artery bypass graft surgery (CABG) has been beneficial in previous registries but the increased mortality associated with surgery may offset the benefits. Percutaneous Coronary Intervention (PCI) is a lower risk procedure and is the most widely employed treatment for CAD, with 80,000 procedures carried out annually in the UK, at a ratio of 3.25 to 1 in relation to CABG. The two main indications for PCI at present are acute myocardial infarction (or acute coronary syndrome) and stable exertional angina. Treatment of severe left ventricular impairment (in the absence of limiting angina) is not an established indication for PCI and is the subject of the proposed study. TARGET POPULATION: Patients with poor LV function (EF=30%) significant CAD (Jeopardy Score =6) AND viable myocardium (=30% of dysfunctional segments) will be eligible and those with significant angina (=CCS class 3) or a recent myocardial infarction (< 6 weeks) will be excluded. SETTING: 25-30 tertiary care cardiac units in the UK MEASUREMENT OF OUTCOMES/COSTS:The primary outcome measure is the composite of all-cause death and hospitalisation due to heart failure. Major Secondary outcome measures include Quality of Life scores (EQ5D, KCCQ), NHYA functional class and LV function at 6 months and 1 year. Other outcome measures include the need for further revascularisation, cardiovascular death or acute MI and NHS resource use. An analysis of cost effectiveness will be carried out based on the evidence collected within REVIVED-BCIS2 and existing data. SAMPLE SIZE: The predicted occurrence of death or hospitalisation for heart failure at two years is 36% in the OMT group. The primary endpoint in REVIVED-BCIS2 will be measured over the entire trial duration (i.e. up to 66 months for some patients) with a minimum follow-up duration of two years, thus increasing the number of events. A trial of 700 (350 in each group) would have over 85% power to detect a hazard ratio of 0.7 (a 30% relative reduction in the hazard) at 5% significance allowing for up to 5% losses by the end of follow-up and increasing recruitment over time. For illustrative purposes this represents a reduction to 27% of patients with an event in the PCI group at two years. These calculations are based on patient accrual for 42 months and minimum follow-up of 24 months (range 24-66 months). The above predicted event rates are conservative in relation to the existing literature, and take into account the possibility of patients randomised to OMT subsequently undergoing PCI (see below). If a higher event rate is found in the OMT group or patient recruitment rates exceed expectation early in the trial (thus providing a longer duration of follow-up in a larger proportion of patients), the study would have greater power to detect a hazard ratio of 0.7, or alternatively, provide over 85% power to detect smaller differences in treatment effect. Although a smaller treatment effect may be clinically significant, this would have a major impact on sample size, which in turn may affect the feasibility of completing the trial within the proposed timescale and resources. The hazard ratio of 0.7 used in the power calculation is pragmatic, while being clinically meaningful and is in line with the magnitude of benefit observed across other treatment modalities in this population. The study is expected to have very good power to detect differences in Quality of Life (one of the major secondary outcomes). PROJECT TIMETABLE: Start date – 1st June 2013. Recruitment period – 42 months (1st September 2013 to 1st March 2017). Minimum follow-up - 2 years (range 24-66 months) Recruitment rate – Approximately 8 patients per centre per year across 25-30 centres. Estimated date of first presentation of results - Q4 2019