Perioperative risk of redo aortic valve replacement.

BACKGROUND To evaluate the perioperative risk of redo aortic valve replacement (AVR). MATERIALS AND METHODS Sixty-three patients (53 males, 10 females) underwent redo AVR from 2001-2005. Forty-one bioprostheses had to be replaced for degeneration and/or paravalvular leakage, and 18 mechanical prostheses were changed because of thrombosis and/or paravalvular leakage. Four patients with a bicuspid aortic valve underwent a mechanical AVR after a primary reconstructive procedure. We compared the perioperative course of the redo AVR (group 2) with the primary procedure (group 1). RESULTS Fifty-two patients received a mechanical prosthesis and eight a biological one. Three patients underwent a refixation of the prosthesis for a paravalvular leakage. The durations of surgery (261.7 +/- 49.5 min vs. 191.7 +/- 31.6 min), cardiopulmonary bypass (130.3 +/- 37.1 min vs. 101.3 +/- 28.4 min), and cross-clamping (80.4 +/- 23.4 min vs. 66.4 +/- 20.6 min) were significantly longer in group 2 than in group 1. Forty-three patients had an uneventful postoperative course. There were 28 (8) postoperative complications in group 2 (1): Need for pacemaker insertion: 8 vs. 2; reexploration for bleeding: 4 vs. 1; temporary renal insufficiency: 3 vs. 4; cerebral confusion: 5 vs. 0; low cardiac output syndrome: 4 vs. 0; wound infection: 2 vs. 1; intestinal ischemia: 1 vs. 0. Four patients expired after redo AVR: two resulting from multiorgan failure, one suffered from an intestinal ischemia requiring bowel resection, and one expired as a result of an aortic rupture during resuscitation. CONCLUSION Conventional reoperative AVR is associated with an enhanced perioperative risk. Therefore these patients should be referred early for reoperation to avoid high-risk emergency operations with a significantly increased mortality.