Full sternotomy and minimal access approaches for surgical aortic valve replacement: a multicentre propensity-matched study

OBJECTIVES Surgical aortic valve replacement (AVR) can be performed via a full sternotomy or a minimal access approach (mini-AVR). Despite long-term experience with the procedure, mini-AVR is not routinely adopted. Our goal was to compare contemporary outcomes of mini-AVR and conventional AVR in a large multi-institutional national cohort. METHODS A total of 5801 patients from 10 different centres who had a mini-AVR (2851) or AVR (2950) from 2011 to 2017 were evaluated retrospectively. Standard aortic prostheses were used in all cases. The use of the minimally invasive approach has increased over the years. The primary outcome is the incidence of 30-day deaths following mini-AVR and AVR. Secondary outcomes are the occurrence of major complications following both procedures. Propensity-matched comparisons were performed based on the multivariable logistic regression model. RESULTS In the overall population patients who had AVR had an increased surgical risk based on the EuroSCORE, and the 30-day mortality rate was higher (1.5% and 2.3% in mini-AVR and AVR, respectively; P = 0.048). Propensity scores identified 2257 patients per group with similar baseline profiles. In the matched groups, patients who had mini-AVR, despite longer cardiopulmonary bypass (81 ± 32 vs 76 ± 28 min; P = 0.004) and cross-clamp (64 ± 24 vs 59 ± 21 min; P ≤ 0.001) times, had lower 30-day mortality rates (1.2% vs 2.0%; P = 0.036), reduced low cardiac output (0.8% vs 1.4%; P = 0.046) and reduced postoperative length of stay (9 ± 8 vs 10 ± 7 days; P = 0.004). Blood transfusions (36.4% vs 30.8%; P ≤ 0.001) and atrial fibrillation (26.0% vs 21.5%, P ≤ 0.001) were higher in patients who had the mini-AVR. CONCLUSIONS In a large multi-institutional recent cohort, minimal access approach aortic valve replacement is associated with reduced 30-day mortality rates and shorter postoperative lengths of stay compared to standard sternotomy. A prospective randomized trial is needed to overcome the possible biases of a retrospective study.