Triple valve surgery in the modern era: short- and long-term results from a single centre

Abstract

Triple valve surgery (TVS) is still a challenge for surgeons because of prolonged cardiopulmonary bypass (CPB) and myocardial ischaemic times. The reported operative mortality rate for TVS ranges between 2.5 and 25%; long-term survival is also diminished, with reported survival rates at 5 and 10 years of 75-82 and 61-75%, respectively. The objective of our study is to define early and late clinical outcomes, reporting the initial experience in the treatment of triple valve disease through a minimally invasive approach. A retrospective, observational, cohort study was undertaken of prospectively collected data on 106 patients who underwent TVS at our institution between October 2001 and June 2013. A total of 101 procedures were done through the standard median sternotomy; however, in 5 patients, the surgical procedure was carried out through a right minithoracotomy. Univariate analysis was performed to identify predictors of early and late survival. The in-hospital mortality rate was 5.6% (6 of 107 patients). Predictors of early mortality were: previous cardiac surgery [odds ratio (OR) 4, 95% confidence interval (CI) 1.08-5.2, P = 0.04], preoperative left ventricular ejection fraction (LVEF) (OR 0.9, 95% CI 0.8-1.1, P = 0.003), prolonged CPB time (OR 1.02, 95% CI 1.01-1.04, P = 0.01) and postoperative pulmonary complications (OR 8, 95% CI 5.8-41, P = 0.0001). Five- and 10-year survival rates were 85 ± 3 and 65 ± 9%, respectively. In univariate analysis, diabetes [hazard ratio (HR) 2.5, 95% CI 1-6.2, P = 0.045], preoperative dialysis (HR 3, 95% CI 2-4.7, P = 0.001), unstable angina (HR 4.8, 95% CI 1-18, P = 0.03), preoperative LVEF (HR 0.9, 95% CI 0.8-1.1, P = 0.02), concomitant coronary artery bypass grafting (CABG) (HR 2.5, 95% CI 1.5-5.7, P = 0.006), prolonged CPB time (HR 1.02, 95% CI 1.01-1.13, P = 0.006), postoperative pacemaker (PMK) implantation (HR 6.2, 95% CI 1.3-18, P = 0.01) and postoperative pulmonary complications (HR 3.3, 95% CI 2.1-7.3, P = 0.002) were found to be significant predictors of late mortality following TVS. The freedom rates from valve-related complications and reoperation at 10 years were 95 ± 2 and 97 ± 2%, respectively. The 10-year freedom rates from thromboembolism and anticoagulation-related haemorrhage were 88 ± 5 and 88 ± 4%, respectively. TVS offers encouraging short-term and long-term patient survival; these good results after TVS in patients with advanced valvular heart disease justify aggressive surgical therapy in these patients. TVS with a minimally invasive approach is feasible and could be another treatment option.