Safety of minimally invasive mitral valve surgery without aortic cross-clamp

Background. We developed a technique for open heart surgery through a small (5 cm) right-anterolateral thoracotomy without aortic cross-clamp. Methods. One hundred and ninety-five consecutive patients (103 male and 92 female), age 69 +/- 8 years, underwent surgery between January 2006 and July 2007. Mean preoperative New York Heart Association function class was 2.2 +/- 0.7. Thirty-five patients (18%) had an ejection fraction 0.35 or less. Cardiopulmonary bypass was instituted through femoral (176 of 195, 90%), axillary (18 of 195, 9%), or direct aortic (1 of 195, 0.5%) cannulation. Under cold fibrillatory arrest (mean temperature 28.2 degrees C) without aortic cross-clamp, mitral valve repair (72 of 195, 37%), mitral valve replacement (117 of 195, 60%), or other (6 of 195, 3%) procedures were performed. Concomitant procedures included maze (45 of 195, 23%), patent foramen ovale closure (42 of 195, 22%) and tricuspid valve repair (16 of 195, 8%), or replacement (4 of 195, 2%). Results. Thirty-day mortality was 3% (6 of 195). Duration of fibrillatory arrest, cardiopulmonary bypass, and "skin to skin" surgery were 88 +/- 32, 118 +/- 52, and 280 +/- 78 minutes, respectively. Ten patients (5%) underwent reexploration for bleeding and 44% did not receive any blood transfusions. Six patients (3%) sustained a postoperative stroke, eight (4%) developed low cardiac output syndrome, and two (1%) developed renal failure requiring hemodialysis. Mean length of hospital stay was 7 +/- 4.8 days. Conclusions. This simplified technique of minimally invasive open heart surgery is safe and easily reproducible. Fibrillatory arrest without aortic cross-clamping, with coronary perfusion against an intact aortic valve, does not increase the risk of stroke or low cardiac output. It may be particularly useful in higher risk patients in whom sternotomy with aortic clamping is less desirable.