Background: Because of the complexity of the trabeculated endocardial surface and tangential histologic sectioning, the differentiation of acute cellular rejection (ACR) from Quilty B lesions (QB) in endomyocardial biopsies (EMBs) is problematic. We hypothesized that the phenotype chemokine RANTES (regulated upon activation, normal T cell expressed and secreted) expression of infiltrating cells and the pattern of expression of transforming growth factor-beta (TGF-beta) may distinguish ACR from QB. In previous studies, the number of RANTES-positive cells and the expression of TGF-beta correlated with the severity of rejection.
Methods: We used immunohistochemical techniques to stain sections of human EMBs with only QB (n = 14) or with only ACR (International Society for Heart and Lung Transplantation Grades 1A and 1B, n = 7; Grades 3A and 3B, n = 7) for B (CD20) and T-lymphocytes (CD3), macrophages (CD68), RANTES, and TGF-beta expression. We graded the percentage of positive cells from 0 to 4 (1 = 1% to 25%; 2 = 26% to 50%; 3 = 51% to 75%, and 4 = 76% to 100%).
Results: When ACR was compared with QB, we found no difference in the proportion of myocardial B cells (0.9 +/- 0.3 vs 1.1 +/- 0.3, p = 0.17); however, we found a lesser proportion of T cells (1.8 +/- 0.5 vs. 2.8 +/- 0.9, p <0.01) but more macrophages (2.9 +/- 0.5 vs. 1.1 +/- 0.6, p < 0.0001) in ACR than in QB. We also found more RANTES-positive leukocytes in ACR vs. QB (2.8 +/- 1.3 vs. 1.9 +/- 0.9, p = 0.03). In QB, many endocardial vessels stained for TGF-beta (2.9 +/- 1.6). Myocardial vessels and injured myocytes in both ACR and QB expressed TGF-beta.
Conclusions: In ACR, although T-lymphocytes are numerous, more than 50% of infiltrating cells are macrophages and more than 50% express RANTES. In QB lesions, more than 50% of infiltrating cells are T-lymphocytes and less that 50% of leukocytes will express RANTES. B cells are present in both ACR and QB, but on average comprise only 25% of the cells present. Thus, a relatively simple immunohistochemical analysis of endomyocardial biopsies may be useful in distinguishing ACR from QB.