Minimally Invasive CABG or Hybrid Coronary Revascularization for Multivessel Coronary Diseases: Which Is Best? A Systematic Review and Metaanalysis.

ABSTRACT

OBJECTIVES: Minimally invasive coronary revascularization (MICR) involves minimally invasive direct coronary artery bypass grafting (MIDCAB) and robotic-assisted coronary artery bypass grafting (RCABG), and hybrid coronary revascularization (HCR) aims to combine MICR/RCABG on left anterior descending (LAD) and percutaneous coronary interventions (PCI) on non-LAD lesions. We performed a systematic review and metaanalysis to compare clinical outcome after MICR and HCR. METHODS: A metaanalysis was carried out through searching PubMed, EMBASE, Web of Science, and Medline for comparative studies evaluating the primary and secondary clinical end points. RESULTS: A systematic literature search identified 8 observational studies that satisfied our inclusion criteria, including being suitable for metaanalysis; the studies were between 1990 and 2018 and included 1084 cases of HCR and 2349 cases of MICR. Metaanalysis of these studies showed that HCR was associated with a reduced need for ICU LOS (WMD -11.46 hours, 95% CI, -18.76 ~ -4.25, P = .02), to hospital time (WMD -1.34 hours, 95% CI, -2.42 to 0.26, P < .01), and blood transfusion (OR 0.43, 95% CI, 0.31-0.59, P < .00001) than MICR. Comparisons of individual components showed no significant difference in terms of in-hospital mortality, MACCE, shock, myocardial infarction (MI), long-term survival, total variable cost, and surgical complications (including renal failure, chest drainage, bleeding). CONCLUSIONS: HCR was noninferior to MICR in terms of in-hospital mortality, MACCE, shock, MI, long-term survival, total variable cost, and surgical complications (including renal failure, chest drainage, bleeding), whereas HCR was associated with a reduced need for ICU LOS, hospital time, and blood transfusion than MICR and less infection than MICR. Further randomized studies are warranted to corroborate these observational data.