A semiquantitative approach to the evaluation of acute cardiac allograft rejection at endomyocardial biopsy

Background: Histopathologic criteria for grading of acute cardiac allograft rejection are focused on the most severe lesion that is recognized among the myocardial fragments provided by each endomyocardial biopsy specimen. Considering the distribution of rejection lesions among all the fragments improved the accuracy in characterizing the severity of rejection in pathologic studies. This study was undertaken to verify the usefulness of a semiquantitative evaluation of endomyocardial biopsy specimens, consisting of the calculation of the proportion of fragments showing rejection in the clinical setting. Methods: Of the 2386 biopsy specimens obtained during the first posttransplantation year in 168 consecutive cardiac allograft recipients, 290 biopsy specimens constituted by ≥3 adequate fragments and showing rejection not followed by treatment (n = 159) or being the first biopsy specimen prompting treatment with augmented immunosuppression for that rejection episode (n = 131) were selected. These biopsy specimens (index biopsy specimens) were grouped according to whether rejection was present in ≤33%, >33% to ≤67%, and >67% of the fragments. The rejection grade (according to the standardized grading system) and the proportion of fragments positive for rejection were correlated with the occurrence of clinical symptoms and signs of rejection at index biopsy and with the results of the next biopsy. Results: Rejections graded ≥3A were more frequently symptomatic (36% vs 9% for those graded 33 to ≤67%: 13 of 99, 13%; >67%: 19 of 67, 28% [p < 0.0001]). Spontaneous resolution after untreated biopsies was more frequent in focal (grade 1A and 2) than in diffuse mild (1B) rejections (68% vs 38% [p < 0.04]), whereas progression to grade 3A or greater was less frequent (4% vs 27% [p < 0.01]). Increasing proportions of positive fragments were associated with lower frequencies of spontaneous resolution (p < 0.05) and higher frequencies of worsening (9%, 22%, 43% [p < 0.009]) or progression to grade 3A or greater (2%, 6%, 28% [p < 0.005]). Complete resolution after treatment was less frequent for increasing proportions of positive fragments at index biopsy (80%, 66%, 49% [p < 0.05]). Conclusions: Diffuse versus focal rejection pattern and the proportion of positive fragments seem to be clinically relevant in terms of occurrence of symptoms, spontaneous evolution, and response to treatment.