Objectives: We sought to compare long-term outcomes between percutaneous mitral valvuloplasty and open heart surgery.
Methods: The study evaluated 402 patients who underwent percutaneous mitral valvuloplasty and 159 patients who underwent open heart surgery between January 1, 1995, and December 31, 2000. The rates of cardiovascular death or repeated intervention (redo percutaneous mitral valvuloplasty or open heart surgery) were determined over a median follow-up of 109 months (mean + or - SD, 106 + or - 27). The therapeutic effects on adverse outcomes were estimated by the Cox proportional hazards model adjusting differences in the severity of illness before intervention. The effects of the cardiac rhythm and echocardiographic score were also tested.
Results: The observed (unadjusted) event-free survival was similar for both groups, and the hazard ratio for the clinical events after percutaneous mitral valvuloplasty as compared with after open heart surgery was 1.510 (95% confidence interval, 0.914-2.496; P = .1079). However, the adjusted hazard ratio was 3.729 (95% confidence interval, 1.963-7.082; P < .0001), showing a higher event-free survival in the open heart surgery group. The adjusted hazard ratio after percutaneous mitral valvuloplasty as compared with after open heart surgery in patients with echocardiographic scores of 8 or more and atrial fibrillation were 5.348 (95% confidence interval, 2.504-11.422; P < .001) and 3.440 (95% confidence interval, 1.805-6.555; P = .0002), respectively, whereas the hazard ratio in patients with echocardiographic scores less than 8 and normal sinus rhythm did not show differences.
Conclusions: Open heart surgery was associated with a higher adjusted rate of long-term event-free survival than percutaneous mitral valvuloplasty. Patients with high echocardiographic scores or atrial fibrillation showed better outcomes after open heart surgery.
Copyright 2010 The American Association for Thoracic Surgery. Published by Mosby, Inc. All rights reserved.