Hybrid Revascularization Versus Coronary Artery Bypass Grafting

Hybrid Revascularization Versus Coronary Artery Bypass Grafting in Patients With Multi-vessel Coronary Artery Disease. A Randomized Clinical Trial.

Patients with multivessel coronary disease have a compromised blood flow in the hearts own vessels, due to stenoses, and they are often treated with a coronary artery bypass grafting surgery. This is a big heart operation where the sternum (breast plate) is split and the heart is once again given a sufficient blood flow, by redirecting blood to the compromised areas through grafts. Another method for treating coronary disease is with percutaneous coronary intervention, or an endovascular approach that can re-establish blood flow to the hearts vessels by lessening the stenoses. However, bypass surgery is still the treatment of choice when there are multiple vessels in the heart with significant disease. In this trial we want to examine if a less invasive approach, called hybrid revascularization, where we combine the best of both treatments is better than the current standard. The combination is a minimally invasive bypass surgery, called MIDCAB which makes a bypass graft through a key-hole surgery for the anterior vessel on the heart, and catheter treatment of all other vessels with significant stenoses. We will compare the hybrid revascularization with the conventional treatment with a full bypass surgery through a sternotomy. Our hypothesis is that a less invasive revascularization strategy gives a better clinical outcome than the current standard with coronary artery bypass surgery through a split sternum. To test our hypothesis we will conduct a multicenter randomized clinical trial.

Coronary artery disease is the most common cardiovascular disease and accounts for 16% of deaths worldwide. Complex coronary disease is often treated with coronary artery bypass surgery (CABG), but percutaneous coronary intervention (PCI) with drug eluting stents is used increasingly. Hybrid coronary revascularization was developed to combine the best parts of the two revascularization techniques. It consists of a minimally invasive off-pump LIMA to LAD grafting, often as a MIDCAB, and PCI with DES of all functionally significant non-LAD stenoses. Theoretically, patients may obtain the survival benefit from the LIMA-LAD graft as well as benefits from a less invasive approach and avoidance of vein graft failures. Hybrid coronary revascularization may therefore be considered a suitable alternative approach to multivessel disease. Hybrid coronary revascularization is to date only supported by very limited data. A literature search identified only three randomized trials on hybrid revascularization. Therefore, current guidelines state that hybrid coronary revascularization may be considered for specific patients at experienced centers. It is unclear whether the contemporary refined PCI techniques, in combination with the unsurpassed patency of the LIMA-graft to the LAD, translate into better outcomes for patients with multivessel coronary disease than conventional CABG does. The randomized trials on hybrid coronary revascularization and reviews of the procedure all state that the procedure is feasible but that larger randomized trials are needed to determine its future place in the treatment of multivessel coronary disease. By conducting a superiority, randomized trial on hybrid coronary revascularization versus CABG, it would give us the possibility to assess the beneficial and harmful effects of hybrid coronary revascularization. OBJECTIVE Our aim is to assess the beneficial and harmful effects of hybrid coronary revascularization versus coronary artery bypass grafting in patients with multivessel coronary artery disease. TRIAL DESIGN We plan to conduct a multicenter randomized superiority trial comparing hybrid coronary revascularization versus conventional CABG surgery in patients with multi-vessel coronary artery disease. As a multicenter trial, we aim at a collaboration with 15-20 centers for inclusion of patients. Participating centers will all be experienced with the MIDCAB procedure with endoscopic or robotic LIMA harvest. Thus requiring surgeons to have performed at least 40 MIDCAB cases and the center to have a yearly production of at least 25 MIDCABs. Patients will be randomized 1:1 to hybrid coronary revascularization or CABG. Blinding: Participants will not be blinded, but they will be instructed to withhold information of their allocation group when assessed at the follow up at 12 months. Blinding of the surgeons and cardiologists delivering the trial interventions will not be possible in the trial. All other investigators including treatment providers not included in the delivery of the trial interventions, outcome assessors, data managers, data safety monitoring committee, statisticians, and writers of the manuscript will be sought blinded. Screening: All patients with multi-vessel coronary disease referred for a CABG will be discussed during the heart team conference and considered for participation in the trial. DATA COLLECTION Method: All data will be collected to electronic CRFs. Timing: Baseline and pre-procedure data collection Demographic data and a medical history will be obtained, including indication for revascularization, risk factors for coronary artery disease (smoking status (current/former/non-smoker), diabetes, hypertension, hypercholesterolemia, relevant family history, prior coronary artery disease including PCI, presenting ECG changes and peak biomarker (Troponin I /T or CK-MB) levels. A list of all cardiac medications will be obtained including antiplatelet therapy, use of anticoagulation, statins, ACE-inhibitors, beta-blockers or aldosterone receptor antagonists. Pre-procedural angina status (CCS Class I-IV) and dyspnea status (NYHA Class I-IV) will be recorded, and the participant will be asked to complete both a Seattle Angina Questionnaire and EQ-5D-5L quality of life assessment. All participants will undergo testing with 12-lead electrocardiography (ECG) and transthoracic echocardiography as per standard-of-care. Pre-procedural laboratory tests (these must have been taken no more than 24 hours before randomization) will include assessment of high sensitive CRP, hemoglobin levels, leucocytes, platelets, serum creatinine, Troponin I/T and creatinine kinase-MB. Preoperative coronary angiography will be described according to the 17-segment American Heart Association model, with visual estimation of diameter stenoses, lesion complexity (Type A, B1/2 or C(50)) and pre-existing stents. Stenoses will be functionally assessed by FFR and/or iFR or any other diastolic indices values recorded. SYNTAX score will be assessed at the heart team conference. Procedural data collection (Hybrid) From the MIDCAB-part of the hybrid procedure: Data from the surgical part of the hybrid procedure will include procedural time, heparin dose, activated clotting time before heparin/after first dose heparin/and at the end of procedure, use of and size of shunts in the LAD, conversion to sternotomy, intraoperative/postoperative bleeding and reoperation due to bleeding. Graft flow will be measured intraoperative and documented as flow in ml/min and pulse index. From the PCI-part of the hybrid procedure: This will include arterial access site, procedural time, contrast and fluoroscopy (dose-area product (mGy·cm2)) use. Coronary angiography will be described according to the 17-segment American Heart Association model, with visual estimation of diameter stenoses, lesion complexity (Type A, B1/2 or C(50)) and corresponding FFR and/or iFR values recorded. Stent type, length and the number of treated vessels will be documented, along with the use of adjuncts such as intracoronary lithoplasty and rotational atherectomy. Use of intravascular imaging with either IVUS or/and OCT will be documented Procedural data collection (CABG) Data collection from the CABG procedure will include graft choice, endoscopic or open graft harvest, number of grafts and position, Y-grafts or proximal anastomoses, jump/snake or single grafts, graft flow (ml/min and PI), X-clamp time, ECC time, total operating time, intra/postoperative bleeding and reoperation due to bleeding. Post-procedural data collection A 12-lead ECG is to be taken at 24 hours. Levels of the CK-MB fraction of creatine kinase are to be measured at baseline and at 12 and 24 hours after procedures. Additional electrocardiograms and biomarker measurements are to be taken as required clinically to assess for recurrent ischemia or to evaluate adverse cardiac events. Data collection at follow-up All participants will have a clinical follow-up at 12 months. A review of medical records relevant to the primary, secondary and exploratory outcome measures will be conducted, whilst angina status and medications will be recorded. Follow-up at 12 months will include the Seattle Angina Questionnaire and EQ-5D-5L quality of life assessments. Cardiac CT may also be conducted at this time. Adverse events at 5 and 10 years will be sought through electronic patient journals. DATA MANAGEMENT Data will be managed centrally at the Copenhagen Trial Unit. The database will be continuously updated and checked for missing data/errors at 1-month intervals. All personal data for the participant will be pseudo anonymized, in order to protect confidentiality before, during and after the trial. All data will be made publically available after the end of the trial. STATISTICS Sample size and power estimation: With 524 participants in the intervention and control group respectively, we will have a power of 0.90 to detect an absolute risk reduction of 10% (corresponding to a relative risk reduction of 18.2%), with α=0.05, provided that the incidence of either all-cause mortality, a diagnosis of myocardial infarction, a diagnosis of stroke, or hospitalization in the control group is equal to 55%. Notably, a 10% absolute risk reduction corresponds to an incidence of either all-cause mortality, a diagnosis of myocardial infarction, a diagnosis of stroke, or hospitalization equal to 45% in the intervention group. Power estimations of secondary outcomes: Quality of life measured using the EQ-5D: With an alpha of 0.05 and a standard deviation of 25 points, we will have a power of 90.0% to confirm or reject a difference of five points on the EQ-5D scale. Angina symptoms measured using Seattle Angina Questionnaire: With an alpha of 0.05 and a standard deviation of 17 points, we will have a power of 99.7% to confirm or reject a difference of five points on the Seattle Angina Questionnaire. Days alive outside hospital within 90 days after procedure: With an alpha of 0.05 and a standard deviation of five days, we will have a power of 84.6% to confirm or reject a difference of one day on the outcome measuring days alive outside of hospital within 90 days. We have subtracted 15% of the sample size for this estimation based on presumed non-normally distributed data and hence the use of non-parametric statistics. Procedural complications: with an alpha of 0.05 and a proportion of 10% in the control group, we will have a power of 41.3% to confirm or reject a relative risk reduction of 30% on complications. Statistical method: A statistical analysis plan will be published before the analyses commence, in which we will provide a detailed description of the analysis of all primary, secondary and exploratory analyses. In short, all randomised participants will be included in all analysis (intention-to treat principle). Dichotomous outcomes will be analysed using logistic regression, continuous outcomes will be analysed using linear regression, , and count data will be analysed using van Elteren test. (49) Analysis of count data will only be adjusted for site. All other analyses will be adjusted for site and diabetes (yes/no) and continuous outcomes will additionally be adjusted for the baseline value.. We will secondly demonstrate the results of the primary outcome using win ratio with the outcomes prioritised as listed under 'Outcomes'. The primary conclusion will be based on the primary outcome. Secondary and exploratory outcomes are hypothesis generating only. We will handle missing data, and systematically assess if the underlying assumptions behind the analysis are violated, according to the recommendations by Jakobsen et al. (50) We will use a five-step procedure to assess if the thresholds for statistical and clinical significance are crossed. All analyses will be performed blinded by two independent statisticians with the two intervention groups concealed as 'A' and 'B'. QUALITY CONTROL The first couple of CRFs will be thoroughly examined by one of the primary investigators in order to ensure that all data is systematically collected. Hereafter, we will examine a random sample of CRFs, representing 2% of participants. If any major violations are identified, a larger sample of CRFs will be examined, and if a consistent problem is identified then all CRFs will be monitored. A trial monitor will be appointed before initiation of the trial. The trial will be monitored in accordance to SOP and ICH-GCP. The trial monitor will also check all CRFs for the acquirement of informed consent from participants. The SOP monitoring of clinical trials ensures that data are generated, recorded and reported in compliance with the protocol with GCP and any other regulatory requirements.

We intend to blind all possible trial personal. Participants and care providers can not be blinded due to obvious scaring on the patient and the ward that is responsible for patient care.

Revascularization in the form of a combined MIDCAB (LIMA-LAD graft trough a minimal invasive left anterior thoracotomy) and PCI (of all non-LAD stenoses)

Coronary revascularization is conducted as a combined MIDCAB (LIMA-LAD grafting through a minimally invasive anterior left thoracotomy) and PCI (of all non-LAD stenoses)

A composite outcome of either all-cause mortality, a diagnosis of spontaneous myocardial infarction, a diagnosis of stroke, or any unplanned hospitalization

A composite outcome of either all-cause mortality, a diagnosis of spontaneous myocardial infarction, a diagnosis of stroke, or any unplanned hospitalization

Procedural complications defined as periprocedural myocardial infarction, unplanned re-intervention, stroke, BARC 3-5 bleeding complications or sternal wound infection

Inclusion Criteria: Written informed consent ≥ 18 years of age Chronic coronary disease or acute coronary syndrome Functionally significant multivessel disease (FFR ≤ 0.80, non-hyperemic index ≤ 0.89 or ≥ 90% diameter stenosis by visual estimation) with indication for CABG. Patient case reviewed by both cardiac surgeon and cardiologist who in agreement find both CABG and hybrid revascularization technically feasible with an expectation of complete revascularization of all functionally-significant stenoses in the LAD and at least one other coronary artery ≥2.5mm in diameter or left main bifurcation stenosis with functionally-significant stenoses in both the LAD and left circumflex artery (LCX) Exclusion Criteria: Chronic kidney disease with estimated glomerular filtration rate < 20 mL/min/kg Pregnancy Left main stenosis without additional functionally significant stenoses in both the LAD and the LCX Contraindications to the use of dual antiplatelet therapy STEMI within 1 month Indication for acute revascularization

We plan to make all data publicly available in peer-review papers, after the trial has been finalized.