Abstract 4372084: Robot-Assisted Mitral Valve Repair versus Conventional Sternotomy versus Minimally Invasive Procedure for Degenerative Mitral Regurgitation: A Systematic Review and Network Meta-Analysis of Intra-Hospital Mortality and Myocardial Infarction

Abstract

Introduction: Degenerative mitral regurgitation (DMR) is a common valvular heart disease requiring surgical intervention. While conventional sternotomy (CS) has been the traditional approach, minimally invasive procedures (MIP) and robot-assisted surgery (RAS) have emerged as alternatives. However, comprehensive comparisons of their perioperative safety profiles remain limited. This network meta-analysis (NMA) evaluates intra-hospital mortality and myocardial infarction (MI) rates among these three surgical approaches. Research Question: What are the comparative outcomes of CS, MIP, and RAS in terms of intra-hospital mortality, intra-hospital MI, intra-hospital acute renal failure (ARF), and length of hospital stay (LOS) in patients with DMR? Methodology: We conducted a Bayesian NMA using a random-effects model, synthesising data from 15 studies (n=12,247 patients). The analysis included direct and indirect comparisons between RAS, CS, and MIP. Primary outcomes were 30-day mortality and stroke, with secondary outcomes of ARF and LOS. We calculated odds ratios (OR) with 95% credible intervals (CrI) and ranked treatments using surface under the cumulative ranking curve (SUCRA) probabilities. Multi-criteria decision analysis (MCDA) was used to rank interventions on the basis of all outcomes. Results: Our NMA included 11 studies involving 12,247 patients. RAS demonstrated superior performance across all measured endpoints, with a 37% reduction in 30-day mortality compared to CS (OR 0.63, 95% CrI 0.41-0.97) and a 17% reduction versus MIP (OR 0.83, 95% CrI 0.55-1.25) (SUCRA= RAS: 80.3%, MIP: 57.1%, CS: 12.6%). For renal complications, RAS showed a 39% risk reduction versus CS (OR 0.61, 95% CrI 0.41-0.90) and 36% versus MIP (OR 0.64, 95% CrI 0.43-0.94), with SUCRA probabilities of 88.9%, 54.7%, and 6.4%, respectively. Myocardial infarction rates did not differ significantly between techniques (RAS vs CS OR 0.70, 95% CrI 0.35-1.41), though RAS maintained the highest probability of being safest (SUCRA 74.6%). Conclusion: This comprehensive NMA demonstrates that RAS for degenerative mitral valve repair offers the most favourable perioperative outcomes, with significant reductions in mortality, renal complications compared to CS. MIP present a viable alternative when robotics is unavailable. These findings support the adoption of robotic approaches where feasible.