Pathologic Interpretation of Transbronchial Biopsy for Acute Rejection of Lung Allograft Is Highly Variable

Despite the standardization of pathologic grading of acute rejection in transbronchial lung biopsies following lung transplantation, the reproducibility of pathologic diagnosis has not been adequately evaluated. To determine the interobserver variability for pathologic grading of acute rejection, 1566 biopsies from 845 subjects in the Lung Allograft Rejection Gene Expression Observational study were regraded by a pathology panel blinded to the original diagnosis and compared to the grade of acute rejection assigned by individual center pathologists. The study panel confirmed 49.1% of center pathologists’ A0 grades, but upgraded 5.7% to A1 and 2.7% to grade ≥ A2 rejection; 42.5% were regraded as AX. Of 268 grade A1 samples, 21.2% were confirmed by the pathology panel; 18.7% were upgraded to ≥ A2 and 35.8% were downgraded to A0 with 24.3% being regraded as AX. Lastly, 53.5% of ≥ A2 cases were confirmed, but 15.7% were downgraded to grade A0 and 18.4% cases to A1, while 12.4% were regraded as AX. The kappa value for interobserver agreement was 0.183 (95%CI 0.147–0.220, p < 0.001). The results for B grade interpretation were similar. Suboptimal sampling is common and a high degree of variability exists in the pathologic interpretation of acute rejection in transbronchial biopsies. Despite the standardization of pathologic grading of acute rejection in transbronchial lung biopsies following lung transplantation, the reproducibility of pathologic diagnosis has not been adequately evaluated. To determine the interobserver variability for pathologic grading of acute rejection, 1566 biopsies from 845 subjects in the Lung Allograft Rejection Gene Expression Observational study were regraded by a pathology panel blinded to the original diagnosis and compared to the grade of acute rejection assigned by individual center pathologists. The study panel confirmed 49.1% of center pathologists’ A0 grades, but upgraded 5.7% to A1 and 2.7% to grade ≥ A2 rejection; 42.5% were regraded as AX. Of 268 grade A1 samples, 21.2% were confirmed by the pathology panel; 18.7% were upgraded to ≥ A2 and 35.8% were downgraded to A0 with 24.3% being regraded as AX. Lastly, 53.5% of ≥ A2 cases were confirmed, but 15.7% were downgraded to grade A0 and 18.4% cases to A1, while 12.4% were regraded as AX. The kappa value for interobserver agreement was 0.183 (95%CI 0.147–0.220, p < 0.001). The results for B grade interpretation were similar. Suboptimal sampling is common and a high degree of variability exists in the pathologic interpretation of acute rejection in transbronchial biopsies.