Prediction of outcome after mitral valve replacement in patients with symptomatic chronic mitral regurgitation. The importance of left atrial size.

BACKGROUND: The ability to predict outcome after mitral valve replacement remains limited in patients with symptomatic chronic mitral regurgitation. The aims of this study were to determine the preoperative predictors of postoperative cardiac-related mortality and to assess the additive prognostic value of tests performed in such patients. METHODS AND RESULTS: Accordingly, 176 patients (mean age, 57 +/- 14 years) who underwent mitral valve replacement were followed up for 3.8 +/- 0.5 years. Four categories of variables were analyzed to predict postoperative cardiac-related mortality: clinical, laboratory, two-dimensional echocardiographic (2DE), and cardiac catheterization. There were 39 cardiac-related deaths (29 due to congestive heart failure and 10 sudden). When the four categories were analyzed separately, two clinical, one laboratory, two 2DE, and one catheterization variable best predicted postoperative death. When these six variables were examined simultaneously, only three (one clinical and two 2DE) remained significant predictors of cardiac-related mortality: presence of pulmonary rales, left atrial size, and the ratio of left ventricular wall thickness to left ventricular cavity dimension in end systole. A model based on these three variables may predict cardiac-related death with considerable accuracy. Laboratory data did not add to clinical information for predicting death. 2DE variables provided significant additional information in this regard (p less than 0.001). Further addition of catheterization variables was not useful. Prognostic value did not change significantly when 50 patients with prior mitral valve surgery or 49 patients undergoing concomitant aortic valve replacement or coronary artery bypass surgery were excluded from analysis. CONCLUSIONS: We conclude that 1) measures of both left ventricular systolic function and left atrial size are equally important in predicting postoperative cardiac-related mortality in patients with symptomatic chronic mitral regurgitation undergoing mitral valve replacement; 2) left atrial size may be important because it reflects the "history" (severity and duration) of mitral regurgitation; 3) 2DE assessment of left atrial size and left ventricular function provides prognostic information that is significantly greater than that obtained from clinical and laboratory parameters alone; the addition of catheterization variables does not increase the prognostic value of the clinical and 2DE data.