Minimally invasive, nonsternotomy approaches for valve procedures may reduce the risks associated with cardiac surgery after prior sternotomy and may improve outcomes. We analyzed our institutional experience to test this hypothesis.Methods
Between 1995 and 2002, 498 patients with previous cardiac operations via sternotomy underwent isolated valve surgery: 337 via median sternotomy (aortic = 160; mitral = 177) and 161 via mini-thoracotomy (aortic = 61; mitral = 100). Data were collected prospectively using the New York State Cardiac Surgery Report Form.Results
Preoperative incidences of congestive heart failure, renal disease, and nonelective procedures were higher in the sternotomy group. Hospital mortality was significantly lower with the minimally invasive approach, 5.6% (9/161) versus 11.3% (38/337) (univariate, p = 0.04). However, multivariate analysis (odds ratio: 95% confidence intervals, p value) revealed that chronic obstructive pulmonary disease (6.6: 1.4 to 3.1, p = 0.001), renal disease (4.1: 1.52 to 11.2, p = 0.01), cerebrovascular disease (2.2: 1.03 to 4.78, p = 0.04), and ejection faction <30% (1.5: 0.96 to 5.5, p = 0.06) were associated with increased mortality. While mean bypass time, cross-clamp times, and stroke rates were comparable between groups, patients undergoing minimally invasive valve surgery had no deep wound infections (0% vs 2.4%, p = 0.05), less need for blood products (p = 0.02), and shorter hospital stays (p = 0.009). Five-year survival was higher with minimally invasive techniques as compared to a sternotomy approach (92.4 ± 2% and 86.0 ± 2%, respectively, p = 0.08).Conclusions
Reoperative valve surgery can be safely performed using a nonsternotomy, minimally invasive approach, with at least equal mortality, less hospital morbidity, decreased hospital length of stay, and slightly favorable mid-term survival as compared to sternotomy.