Abstract

OBJECTIVE--To determine which factors predict the occurrence of mitral regurgitation after balloon dilatation of the mitral valve for rheumatic stenosis. DESIGN--Analysis of a case series of patients with rheumatic mitral valve stenosis who had had successful balloon dilatation of the mitral valve. SETTING--A tertiary care centre with an experience of over 150 balloon dilatations of the mitral valve. PATIENTS--70 young patients with non-calcified rheumatic mitral stenosis, who had undergone successful balloon dilatation of the mitral valve. No patient had mitral regurgitation or atrial fibrillation before dilatation. INTERVENTION--Dilatation of the mitral valve by the transvenous, transatrial double balloon technique. MAIN OUTCOME MEASURE--Development of mitral regurgitation after balloon dilatation of the mitral valve and its relation to age, mitral valve area before dilatation and after dilatation, the degree of mitral subvalvar pathology, and the size of balloon used for dilatation. RESULTS--In 10 patients (14%) mitral regurgitation developed after balloon dilatation of the mitral valve. No statistically significant differences were found between patients who did not develop regurgitation and those who did in terms of age (mean (SD)) (19.9 (6.46) v 19.4 (5.5)), mitral valve area before dilatation (1.05 (0.33) v 0.94 (0.4) cm2) and after dilatation (2.52 (1.06) v 2.45 (1.1) cm2), mitral subvalvar pathology assessed by the mitral subvalvar distance ratio (0.116 (0.03) v 0.118 (2.32), or balloon diameter corrected for body surface area (21.37 (3.5) v 20.57 (2.32) mm/m2. CONCLUSIONS--In this subset of children and young adults with non-calcified mitral stenosis, none of the morphological, technical, or patient characteristics studied predicted the development of mitral regurgitation after balloon dilatation. The low incidence of mitral regurgitation may have reduced the discriminatory power of this study. None the less, the means and standard deviation for each factor in each group suggest that even in a larger sample size the variables would have little predictive capacity.