Background and aims of the study: Reoperative mitral surgery via sternotomy can be associated with significant complications, including excessive blood loss and injuries to the heart, great vessels and patent coronary artery grafts. The right antero-lateral thoracotomy offers excellent exposure with less risk from re-entry.
Materials and methods: Between 1982 and 1992, 221 patients had repeat mitral valve procedures at our institution. Fifteen of these 221 underwent mitral valve replacement via right thoracotomy. Indications for surgery in each group included bioprosthetic valve failure, paravalvular leak and bacterial endocarditis. Fifteen patients having reoperative mitral valve surgery via right thoracotomy approach were compared with a control group of 33 patient who underwent surgery via repeat sternotomy. All thoracotomy patients underwent mitral replacement or repair with ventricular fibrillation without aortic cross-clamping. Operative time, cardiopulmonary bypass time, requirement for inotropic support, blood loss within the first six postoperative hours, number of blood units transfused, length of ICU stay, days to discharge, and 30-day survival were compared between the two groups. In addition, the preoperative PaO2/FiO2 (P/F) ratio was evaluated as a prognostic indicator.
Results: Bypass time (162 +/- 43 min thoracotomy group vs. 131 +/- 34 min sternotomy group), operative time (389 +/- 100 min thoracotomy group vs. 450 +/- 25 min sternotomy group), ICU stay (6 +/- 8 days thoracotomy group vs. 5 +/- 6 days sternotomy group), P/F ratio (352 +/- 142 thoracotomy group vs. 423 +/- 108 sternotomy group), and 30-day survival (93% thoracotomy group vs. 91% sternotomy group) were not found to be significantly different between groups. Of great significance was the reduction in blood loss (277 +/- 152 ml thoracotomy vs. 651 +/- 504 ml sternotomy, p < 0.05) and blood transfused (2.0 +/- 1.7 units thoracotomy vs. 6.5 +/- 3.3 units sternotomy, p < 0.01) with the thoracotomy approach. Also of significance was a reduction in frequency with which significant inotropic support was needed to separate from cardiopulmonary bypass (26% vs. 63%, p < 0.05). Despite decreased access to the heart for de-airing maneuvers, no cerebrovascular events whatsoever were noted with the thoracotomy approach.
Conclusion: The right thoracotomy approach is recommended for redo mitral valve surgery. Despite these advantages, severe pulmonary dysfunction (as indicated by a P/F ratio less than 300) correlated with a prolonged hospital course in four thoracotomy patients; such patients should have repeat sternotomy.