Minimally invasive direct access mitral valve surgery.

We reviewed our experience with minimally invasive direct-access mitral valve surgery in 207 patients through February 1999. Three patients underwent associated procedures, a coronary artery bypass graft (CABG) with right internal mammary artery to right carotid artery (RIMA-RCA), a left ventricular outflow tract (LVOT) debridement for endocarditis, and a primum atrial septal defect (ASD) repair, and were excluded from analysis. Of the 204 remaining patients, 120 (59%) patients were men, aged 58.7 +/- 13.2 years, functional class of 2.3 +/- 0.5. The cause was myxomatous in 162 (79%) patients, rheumatic in 28 (14%) patients, endocarditis in 8 (4%) patients, congenital in 3 (2%) patients, and ischemic in 3 (2%) patients. Mean preoperative EF was 60% +/- 10%, with 184 (90%) patients showing ejection fraction (EF) greater than 50%. The valve was approached through a 5- to 8-cm right parasternal (n = 180, 88%) or right inframammary (n = 24, 12%) incision. One hundred nineteen (58%) patients had open femoral artery-femoral vein cannulation, and 85 (42%) patients had direct cannulation of the aorta and percutaneous cannulation of the femoral vein. One hundred seventy (83%) patients underwent successful valve repair, and 34 (17%) patients required valve replacement. The mean duration of aortic clamping and cardiopulmonary were, respectively, 100 +/- 34 and 146 +/- 44 minutes. There were 2 (1%) surgical deaths. Nonfatal perioperative complications included 3 (1.5%) ascending aortic complications, 3 (1.5%) reoperations for bleeding, 4 (2%) strokes, 2 (1%) transient ischemic attacks (TIAs), 2 (1%) myocardial infarctions, 3 (1.5%) pericardial effusions requiring drainage, 9 (4.5%) vascular complications, and 3 (1.5%) wound complications. Mean length of stay (LOS) was 6.1 +/- 3 days, with 63 (31%) patients being discharged in less than 5 days. One hundred twenty-nine (63%) patients did not require blood transfusions. Follow-up was complete in 165 (81%) patients, with mean follow-up of 13.2 +/- 8 months. Late complications included 1 (0.5%) myocardial infarction, 3 (1.5%) reoperations, all converting repairs to replacements, 3 (1.5%) wound hernias requiring reoperation and repair with mesh, 5 (2.5%) thromboembolic events, and 3 (1.5%) deaths of suicide, pneumonia, and sudden death, respectively. Mean follow-up New York Heart Association (NYHA) functional class was 1.2 +/- 0.5. We conclude that minimally invasive direct-access mitral valve surgery is safe, effective, and applicable for most patients presenting for isolated mitral valve surgery. We now consider it the standard of care for selected patients.