Non-invasive detection of cardiac rejection still remains a challenge after heart transplantation. We assessed troponin-T as a new serum marker to diagnose cardiac rejection. Twenty-five heart transplant patients (Berne) were monitored prospectively for up to 2 years, and compared to 89 retrospectively assessed patients (Stanford). Blood samples (392 Berne and 320 Stanford) were analyzed (creatine kinase, isoenzymes MB activity and MB mass, troponin-T and troponin-I). Regression analysis between the results of these blood samples and cardiac rejection grading from simultaneously performed endomyocardial biopsies was carried out. Troponin-T tests done in two different laboratories showed a good correlation (r = 0.91; P < 0.0001), whereas troponin-T versus troponin-I showed a lower correlation (r = 0.53; P < 0.0001). Troponin-T and -I in contrast to other enzymes were elevated for a longer period (up to 4 weeks before returning to baseline) after transplantation than during conventional cardiac surgery. Beyond 3 months the following correlations were found between troponin-T (new or old test) and the other enzymes (creatine kinase: r = 0.26, MB activity: r = 0.4, and MB mass: r = 0.68). The correlation between the degree of rejection and the enzyme release is poor, however, the best results were obtained for troponin-T (r = 0.22; P < 0.001). We found a low correlation between troponin-T and the degree of rejection beyond 3 months after heart transplantation. Despite a troponin-T elevation in some patients with rejection, the new test is not sensitive enough to be used alone for the non-invasive diagnosis of cardiac rejection.