A comparison of myocardial perfusion and rejection in cardiac transplant patients

Although histological evaluation of the cardiac tissue is the current gold standard for evaluation of rejection, we hypothesized that cardiac perfusion MRI is a safe non-invasive method that correlates tissue blood flow changes with biopsy proven rejection in the cardiac transplant patient. In a retrospective study from 1984-2001, 83 patients underwent 135 MR Gd-DTPA imaging studies. In 8 patients (9%), biopsies graded 2 or higher (by ISHLT criteria) provided evidence of rejection. Patients were age and sex matched to 11 non-rejected controls for imaging analysis. Time-signal intensity curves generated for a mid-ventricle LV short axis slice during rest and adenosine stress allowed determination of myocardial blood flow (MBF, ml/min/gm). ROC curve analysis by SPSS allowed estimation of sensitivity and specificity. At rest, there was no difference in MBF between patients with prior rejection vs. those without (1.18 +/- 0.26 vs. 1.16 +/- 0.29). At stress there was a decrease in MBF for patients with prior rejection episodes (3.27 +/- 0.74) compared to no rejection (3.60 +/- 0.72), P = 0.067). The area under the ROC curve was 0.82, with specificity and sensitivity of 75% and 81%, respectively. This study suggests that perfusion MR imaging can be used in assessing the cardiac transplant patient for rejection related microvascular changes. The high specificity and sensitivity recorded from the ROC curve illustrates the potential utility of this diagnostic test for future studies.