Aim: 1. To evaluate the ability of TDI measurements to predict unsuccessfull postoperative cardiac rehabilitation (CR) after aortic valve replacement (AVR), comparing aortic stenosis (AS) to aortic regurgitation (AR).
2. To determine the relation between left ventricular (LV) function assessed by TDI and exercise capacity and whether the diastolic and/or systolic dysfunction participate in unsuccessfull CR.
3. To define the independent predictors for unsuccessfull CR and for medium term prognosis/evolution in these pts.
Material and method: Prospective randomized study on 1212 pts undergoing AVR for AS (676pts) or AR (536pts) who underwent postoperative CR (12 weeks of supervised exercise training 3 times/ week). A complete echocardiography (including TDI) was performed before the CR and weekly till 3 months postoperatively. The primary endpoints were the composites of changes in peak oxygen uptake and in physical capacity.
Stastistical analysis used SYSTAT and SPSS for regression analysis and for the relative risks and correlation coefficient calculations.
Results: 1. The percent of the pts with a favourable effect of CR was higher in those with normal LV diastolic filling pattern (LVDFP) (95,09%) compared with pts with restrictive LV filling (77.33%, p<0.0001). The presence of a restrictive LVDFP increased the risk for unsuccessfull CR by 8.2 fold.
2. Before and during the CR, the E/E' ratio was higher in the failed group than in the successfull group. The cut-off value to predict CR failure (obtained from ROC), gave an E/E' ratio before CR of 14.5 with a sensitivity of 77% and a specificity of 95.8% (p<0.01).The LVEF did not differ between the two groups, whereas left and right ventricular TDI velocities were greater in successfull group. E' was lower in the failed group and increased during CR in the successfull group while no change occurred in the failed group.
3. Regression analysis has identified as independent predictors for unsuccessfull CR: E/E' >14.5, late systolic myocardial motion (Sm) <5.5 cm/sec, restrictive LVDFP, age >75 years, AR and LV endsystolic diameter (LVESD) >57 mm.
Conclusions: 1. The presence of a restrictive LVDFP in AVR pts is an independent predictor for unsuccessfull CR, the LV diastolic function being a more reliable parameter for prognosis appreciation than LV systolic function. 2. Sm and E/E' ratio were the only TDi variables independently predicting exercise capacity and related to success of CR program (p<0.05). 3. The independent predictors for unsuccessfull CR were: E/E' >14.5, Sm <5.5 cm/sec, restrictive LVDFP, age >75 years, AR and LVESD >57 mm.