Conservative Non-Invasive Versus Routine Invasive Management in Coronary Artery Bypass Surgery Patients with Non ST Elevation Elevation Acute Coronary Syndrome

Background: Coronary artery bypass graft (CABG) surgery is commonly performed in the NHS and worldwide based on historical evidence of benefit over medical therapy and more recent evidence of benefit over percutaneous coronary intervention (PCI). According to the national audit data from the Society of Cardiothoracic Surgeons and the British Cardiovascular Intervention Society, over 1 million CABG operations have been performed in the UK, with 20,000 - 25,000 operations performed annually since 1994. Although the number of isolated CABG operations performed annually has fallen slightly (e.g. 17,822 in 2010), CABG combined with valve surgery has increased. Since the 5 year survival rate after CABG is approximately 90%, several hundred thousand patients with prior CABG are likely alive in the UK. However, occlusive vein graft disease affects around 10% of patients within 12 months of surgery and affects at least two thirds of patients by 10 years. Furthermore, disease in the native coronary arteries may also deteriorate. Consequently, even though vein graft occlusion may initially be subclinical, angina eventually recurs in most patients with repeat revascularisation (almost always by PCI) required in nearly two thirds of patients by 12 years. Furthermore, when PCI is performed in these patients clinical outcomes in the longer term are poor because of restenosis.

Given the large number of CABG survivors in the UK, and the complexities of their clinical management, their cardiac morbidity and mortality is an increasing challenge in NHS hospitals.

Chest pain is the commonest reason for hospital admission in the UK and more than 1 in 10 hospital admissions with suspected acute myocardial infarction (MI) have a history of prior CABG and non-ST elevation acute coronary syndrome (NSTE-ACS) patients with prior CABG spend more time in hospital than those without a history of CABG. According to the Myocardial Ischaemia National Audit Project (MINAP), 47,162 patients were admitted with an NSTE-ACS in England and Wales during 2009/2010 including approximately 5000 patients with prior CABG. The median 30-day death rate of all NSTE-ACS patients in the UK in 2009/2010 was approximately 7.5% and clinical outcomes in CABG patients are much worse than in patients without prior CABG. However, since prior CABG status is not an independent predictor of mortality, the prognostic determinants are multifactorial and strongly influenced by age and co-morbidity.

There is an important lack of evidence to inform clinical decision making in symptomatic CABG patients hospitalised with NSTE-ACS. Firstly, most clinical trials of treatment strategies in ACS have excluded patients with prior CABG. The pivotal clinical trials which specifically compared invasive versus conservative management in unstable coronary syndromes, including Thrombolysis in Myocardial Infarction (TIMI) IIIb, Fragmin during Instability in Coronary Artery Disease (FRISC) II, Treat Angina With Aggrastat and Determine the Cost of Therapy With an Invasive or Conservative Strategy (TACTICS)-TIMI 18, and Randomized Intervention Treatment of Angina (RITA) 3, excluded patients with prior CABG, meaning evidence is lacking on which treatment strategy is most appropriate in this patient group. In the absence of trial data, guidelines for clinicians are based on observational studies, a subgroup analysis and expert opinion (level of evidence C). According to the UK National Institute of Clinical Excellence (NICE) Clinical Guideline 94, "in the absence of more research, consensus expert opinion is the best that can be achieved." Although CABG patients with acute ischaemia are often managed invasively, angiography commonly does not lead to revascularisation. PCI, or even repeat CABG, is technically challenging, complications are common and longer term outcomes are poor. This is especially the case for repeat PCI in vein grafts. Overall, invasive management of patients with prior CABG is technically challenging and health outcomes are influenced by the effectiveness of the procedure and complications (safety). Procedural success is harder to achieve because of advanced native coronary and graft disease and co-morbidity meaning the procedure is less often a success and the risks of complications and adverse events (e.g. stroke) are three-four fold higher than average. Since the internal mammary graft is almost always patent, redo surgery is rarely performed because sternotomy may damage the mammary graft (which can be a life-threatening complication) and current guidelines recommend PCI unless the mammary graft is occluded.

The effectiveness of a routine invasive approach in every-day clinical practice, which is the approach of many clinicians, is called into question because acute and long term outcomes are much worse than in patients without prior CABG. PCI in redo CABG carries a four-fold higher risk of death compared to the index procedure and complications associated with PCI include bleeding, stroke and contrast nephropathy. Redo CABG confers a much higher risk than percutaneous coronary intervention (PCI). Overall, increased age and co-morbidity are common and key drivers of long-term outcome. Outcomes after PCI in vein grafts or native vessel disease are poor. For example, in Wales, of 161 patients who had PCI after CABG, adverse cardiac events occurred in 14% of them by 13 months (mainly related to repeat PCI (12%). In the UK overall, of 33,652 PCIs performed in 1999/2000, 907 (2.7%) were performed in patients with prior CABG. Only 1 in 10 procedures were reported as successful and 1.3% and 1.0% of patients experienced in-hospital MI or death. Ten years on (2009/2010), 8.4% of all patients undergoing PCI had a history of prior CABG and procedure-related outcomes remained disappointingly similar (1 in 10 procedures unsuccessful, in-hospital death rate (1.0%), 30-day mortality 1.5%).

By contrast, the efficacy of contemporary optimal medical therapies in patients with prior CABG is at least as good as in patients without prior CABG. Furthermore, medical therapy is not associated with procedure-related complications and potentially may be safer than invasive management. In contrast to revascularisation trials, patients with prior CABG are usually not excluded from contemporary drug trials (e.g. PLATO) and the evidence for secondary prevention medical therapies in patients with prior CABG is reasonably robust. In fact, some therapeutic trials have specifically focused on post-CABG patients and have shown that statin therapy attenuates graft disease progression and improves patient outcomes. In ACS patients with prior CABG, health outcomes are also improved by treatment with contemporary anti-platelet drug therapies. New evidence-based medicines continue to emerge including more potent anti-platelet drugs such as ticagrelor and prasugrel and safer anti-coagulants, such as fondaparinux. Our strategy is also predicated on use of optimal anti-ischaemic therapies, including novel drugs such as ivabradine, as clinically appropriate.

PRACTICES AND UNCERTAINITIES ASSOCIATED WITH MANAGEMENT OF CABG PATIENTS WITH ACUTE ISCHEMIA: Since most patients with prior CABG and an acute NSTE-ACS will be intermediate-high risk, most patients will be considered for invasive management according to guideline recommendations. Registries indicate that invasive management only occurs in about half of these patients in the UK and abroad, and revascularisation only occurs in about half of those who have angiography. In fact, prior CABG status is an independent predictor against invasive management. Use of evidence-based drug therapies, such as aspirin, clopidogrel, ticagrelor is lower in ACS patients with prior CABG compared to in patients without prior CABG despite the fact clinical trial results suggest these drugs improve outcomes substantially in patients with prior CABG.

ISSUES AROUND THE STUDY:

PATIENT POPULATION: The study population will involve patients with an established history of coronary heart disease who have had previous CABG. Typically, CABG patients who are readmitted to hospital with recurrent angina or myocardial infarction do so at least one year after the operation (and usually 5-15 years afterwards). Therefore, these patients are usually familiar with the hospital environment and will always have had a coronary angiogram in the past since this is a requirement before CABG. Furthermore, many (e.g. at least one quarter) of these patients will have had an angioplasty in the past too. Some of these patients will have had angiograms and angioplasty procedures after the CABG also.

ISSUES RELATING TO ACUTE CORONARY SYNDROME: We will recruit patients with unstable angina or a non-ST elevation acute coronary syndrome (NSTE-ACS). These patients usually have a history of self limiting chest pain. Usually, the pain is controlled by initial medical therapy and these patients routinely wait on hospital cardiology wards often days at a time for an in-patient coronary/graft angiogram. Sometimes the angiography is deferred as an out-patient. Occasionally, angina recurs at rest or with minimal exertion (refractory ischaemia) and angiography is expedited. In the vast majority of patients, there will be ample time (minimum 4 h, usually > 24 h as appropriate) to discuss the Patient Information Sheet and obtain informed consent.

MANAGEMENT STRATEGIES: Clinical Guidelines from NICE and the European Society of Cardiology recognise that there is no evidence base to instructively inform the management of these patients thus patient and clinician preference usually inform decision-making. In this trial, patients who are invited to participate should be eligible for either treatment option. In this trial, the type of treatment in each strategy (medical or invasive) will be as per current practice in the NHS and there will be no additional study interventions. Usual care diagnostic tests including non-invasive stress testing and CT angiography may be performed if requested by the attending clinician, but these tests cannot influence the treatment group assignment of randomised patients.

HOSPITAL TYPE: A group of hospitals will be included that provide different forms of clinical services. The city hospitals in this trial will be the Western Infirmary and Glasgow Royal Infirmary. The District hospitals will be the Royal Alexandra Hospital and Royal Blackburn Hospital. One of the hospitals will have 'on-site' admission and catheterisation facilities (the Royal Blackburn Hospital) and the other hospitals participate in a regional model of care with transfer for invasive management to the regional cardiothoracic centre )the Golden Jubilee National Hospital). The standard clinical referral pathway in the NHS will be used in this trial.

OBJECTIVE OF THE PILOT: to prospectively gather information on about screening, recruitment, randomisation and initial clinical outcomes. Other than the randomisation to either medical or invasive management, the trial will essentially be observational.

RECRUITMENT: will take place during routine clinical care and will be lead by the attending clinicians who are responsible for the patients (rather than being lead by a research team).

OBJECTIVE OF THE REGISTRY: to understand the characteristics of NSTE-ACS patients with prior CABG who do not take part in the randomised trial. The reasons for not participating may be due to the presence of exclusion characteristics, patient preference, or physician preference. This information will be very important to understand feasibility and also whether or not our trial will enroll patients who are representative overall. Thus, we will record information on registry patients with prior CABG who are not randomised and follow them up in the same way as the trial patients (i.e. case note review, 1 year contact by telephone or hospital visit, electronic record linkage).

SPECIFIC AIMS RELEVANT TO DESIGN OF FUTURE DEFINITIVE CLINICAL TRIAL

PRIMARY AIMS:

1. To determine whether or not a strategy of invasive management is more effective in symptomatic patients with prior CABG compared to optimal medical therapy patients.
2. To determine whether optimal medical therapy is associated with fewer major complications and so might be safer than invasive management,
3. To prospectively quantify resource utilisation in the NHS with each treatment strategy, derive the difference in the quality-adjusted life years (QUALY) and hence evaluate comparative cost-effectiveness.

SECONDARY AIMS:

1) Quality of life (EQ-50 at baseline and 6 months); 2) components of the primary composite outcome; 3) Canadian Cardiovascular Society angina class; 4) secondary care costs.

RESEARCH HYPOTHESIS:

Unifying null hypothesis: Overall, we hypothesise that a routine invasive approach in NSTE-ACS patients with prior CABG will not be superior to a conservative non-invasive approach.

For the future multicentre definitive trial that we hope to undertake, we have adopted a pragmatic trial design. A non-inferiority trial would require a very large number of patients (and may not be ethical given the large equivalence margin), if a routine invasive approach proves not to be superior to a conservative non-invasive approach and our safety and health economic hypotheses prove correct then the trial results would support a conclusion that a routine invasive approach should not be adopted in the NHS.

Active Efficacy Hypothesis: Compared to a routine conservative non-invasive approach, an invasive approach in NSTE-ACS patients with prior CABG is associated with a lower rate of all-cause mortality, recurrent myocardial ischaemia or heart failure.

Active Safety hypothesis: Compared to an invasive approach, a conservative approach in NSTE-ACS patients with prior CABG is associated with a lower rate of adverse events related to safety during the index hospitalisation.

Health economics hypothesis: A conservative approach is associated with reduced resource utilisation compared to a routine invasive approach in NSTE-ACS patients with prior CABG.

PRIMARY AND SECONDARY ADVERSE EVENTS are defined in the Trial Protocol.

DATA MONITORING COMMITTEE - Data monitoring for safety will be coordinated by the Sponsor and Trials Unit (Pharmacovigilance).

CLINICAL EVENT COMMITTEE - A clinical event committee including 3 cardiologists independent to the research team and blind to treatment group assignment will adjudicate on the primary and secondary causes of adverse events. The case reviews will be supported with source documentation (e.g. ECGs, discharge letter, blood results), as appropriate. The adjudication decisions will be provided to the Sponsor and Clinical Trials Unit.