Reoperative mitral valve surgery in the UK: 20-year trends and early outcomes

INTRODUCTION: Mitral valve surgery has undergone significant expansion and improvement in early in-hospital mortality over the last 20 years. At the same time, a considerable decline seems to have occurred in reoperative mitral valve surgery, which remains a relatively high-risk procedure. We investigated the trends, early outcomes and predictors of complications in patients undergoing RMV over 20 years in the United Kingdom. METHODS: The UK National Adult Cardiac Surgery Audit database was used to retrospectively analyse patients who underwent reoperative mitral valve surgery (n = 1239). These were then compared with a cohort of patients who underwent primary mitral valve surgery (n = 47,213) during the same period, from November 1, 1998, to March 30, 2019. We investigated in-hospital outcomes and examined predictors of adverse outcomes in patients undergoing reoperative mitral valve surgery. RESULTS: The proportion of reoperative mitral valve surgery out of all mitral procedures decreased significantly between 1999 and 2001, from 7.31 % to 3.44 %. After this time point, the percentage remained relatively steady, ranging from 1.8 % to 3.45 %. Of the 1239 reoperative mitral valve surgery, most were replacements 1127 (91 %). When compared to primary mitral valve surgery, reoperative mitral valve surgery were associated with a significantly increased mortality (13 % vs 5.2 %), requirement for dialysis (14 % vs 5.2 %), cerebrovascular event (CVE,2.4 % vs 1.2 %), return to theatre for bleeding (8.3 % vs 5.3 %) and increased length of stay (15 vs 11 days) (all, p < 0.001). After adjustment for baseline characteristics, reoperative mitral valve surgery was still associated with increased mortality (OR1.92; CI1.53-2.43; p < 0.001), CVE (OR1.32, CI1.05-1.65; p = 0.027) and dialysis (OR1.91; CI1.5-2.42, p < 0.001), compared to primary mitral valve surgery. The predictors for increased in-hospital mortality in the reoperative mitral valve surgery cohort were old age (OR 1.4, CI 1.00-1.08), creatinine >200 μmol/L (OR 3.86, CI 1.93-7.70, p < 0.001), need for emergency surgery (OR 2.18, CI 1.06-4.47, p = 0.034) and increased aortic cross-clamp time (OR 1.02, CI 1:00-1.03). CONCLUSION: The incidence of reoperative mitral valve surgery has decreased in the UK over the first decade, most notably after 2001. Despite some variations, the volume remained steady in the second decade. Reoperative mitral valve surgery mortality and associated morbidity remain relatively high.