A propensity matched analysis of robotic, minimally invasive, and conventional mitral valve surgery.

OBJECTIVES: Institutional studies suggest robotic mitral surgery may be associated with superior outcomes. The objective of this study was to compare the outcomes of robotic, minimally invasive (mini), and conventional mitral surgery.
METHODS: A total of 2300 patients undergoing non-emergent isolated mitral valve operations from 2011 to 2016 were extracted from a regional Society of Thoracic Surgeons database. Patients were stratified by approach: robotic (n=372), mini (n=576) and conventional sternotomy (n=1352). To account for preoperative differences, robotic cases were propensity score matched (1:1) to both conventional and mini approaches.
RESULTS: The robotic cases were well matched to the conventional (n=314) and mini (n=295) cases with no significant baseline differences. Rates of mitral repair were high in the robotic and mini cohorts (91%), but significantly lower with conventional (76%, P<0.0001) despite similar rates of degenerative disease. All procedural times were longest in the robotic cohort, including operative time (224 vs 168 min conventional, 222 vs 180 min mini; all P<0.0001). The robotic approach had comparable outcomes to the conventional approach except there were fewer discharges to a facility (7% vs 15%, P=0.001) and 1 less day in the hospital (P<0.0001). However, compared with the mini approach, the robotic approach had more transfusions (15% vs 5%, P<0.0001), higher atrial fibrillation rates (26% vs 18%, P=0.01), and 1 day longer average hospital stay (P=0.02).
CONCLUSION: Despite longer procedural times, robotic and mini patients had similar complication rates with higher repair rates and shorter length of stay metrics compared with conventional surgery. However, the robotic approach was associated with higher atrial fibrillation rates, more transfusions and longer postoperative stays compared with minimally invasive approach. © Article author(s) (or their employer(s) unless otherwise stated in the text of the article) 2018. All rights reserved. No commercial use is permitted unless otherwise expressly granted.