Background: Heart transplant recipients require frequent myocardial biopsies to screen for acute rejection. The purpose of this study was to identify demographic and echocardiographic factors associated with transplant rejection and develop a predictive model, which may reduce the number of cardiac biopsies.
Methods: From January 1998 to December 2001, we performed 406 echocardiographic studies on 264 heart transplant patients who had biopsies performed on the same day. Two-dimensional, pulsed and tissue Doppler echocardiographic variables were compared between patients with and without rejection, and their predictive ability for detecting rejection was determined by uni- and multivariate analyses.
Results: In 268 biopsies there was no significant rejection (ISHLT Grade <==II), whereas 138 showed rejection (ISHLT Grade > or =IIIa). By multivariate analysis, pericardial effusion, isovolumic relaxation time (IVRT) <90 milliseconds and mitral inflow E/A ratio >1.7, diameter of inferior vena cava and duration of pulmonary vein atrial reversal were independently associated with rejection. Because the odds ratios were similar for all 5 predictors, a simplified model was developed based on the sum of the number of abnormal predictors present (0 to 5). The probability of rejection increased from 15.9%, in the absence of any predictor, to 39.7%, 52.0% and 71.1%, if 1, 2 or 3 predictors were present, respectively.
Conclusions: Recipient age, pericardial effusion, IVRT and ratio of pulsed Doppler E/A are significant predictors of acute cardiac allograft rejection. However, no single predictor or combination of predictors were powerful enough to eliminate surveillance endomyocardial biopsies.