Reproducibility of the Banff classification in subclinical kidney transplant rejection

Abstract

The Banff classification for kidney allograft pathology has proved to be reproducible, but its inter and intraobserver agreement can vary substantially among centres. The aim of this study was to evaluate Banff reproducibility of surveillance renal allograft biopsies among renal pathologists from different transplant centres. This study included 32 renal transplant patients with stable graft function. Biopsies were performed 2 and 12 months post-transplant. Histology was interpreted according to the Banff schema by three renal pathologists, and inter and intraobserver agreement were measured. The best reproducibility was obtained for the presence or absence of acute rejection (AR), with kappa values ranging from moderate (kappa = 0.47; p = 0.006) to good (kappa = 0.72; p = 0.0001). However, the agreement for 'suspicious for AR' category was poor between all observers. For scoring and grading interstitial inflammation and intimal arteritis the agreement were poor and moderate, respectively. Reproducibility for the presence or absence of chronic allograft nephropathy (CAN) was heterogeneous, ranging from poor (kappa = 0.13; p = NS) to moderate (kappa = 0.56; p = 0.007). Scoring chronic changes such as fibrous intimal thickening gave a reasonable interobserver agreement. Intraobserver reproducibility was good for presence or absence of AR, but was poor for the diagnosis of CAN. In conclusion, histologic analysis of stable renal allografts based on Banff criteria showed a good agreement for the diagnosis of AR and a reasonable kappa for CAN, but reproducibility for scoring and grading showed a substantial interobserver variation.

The Significance of Subclinical Rejection and the Value of Protocol Biopsies

“…Interobserver agreement for acute rejection (j = 0.77) and scores of glomerulitis, intimal arteritis, interstitial infiltrates and tubulitis were good (17). Similar results were seen by Veronese et al for acute rejection (j = 0.47-0.72); however, agreement for borderline rejection and reproducibility for other scores and grades showed a substantial interobserver variation (18). The subcategorization of SCR into "acute" and "borderline" by the Banff system is also problematic, as the severity of rejection critically hinges on the tubulitis score, which in turn controls the final rejection grade, and defaults to the most severe lesion.…”

Fibrosis with Inflammation at One Year Predicts Transplant Functional Decline

“…Although assignment of i Ն1 may vary for individual pathologists, 11,41,42,45 the results of IHC, TLDA, and microarrays confirmed that biopsies assigned to IFϩi had distinct characteristics compared with those interpreted as having normal histology or IF alone. Targeted reverse transcriptase-PCR analysis demonstrated elevated expression of multiple innate and adaptive immune mediators consistent with tissue injury response, Th1-type T cell response, and suppression of counterregulatory pathways.…”

Renal Graft Fibrosis and Inflammation Quantification by an Automated Fourier–Transform Infrared Imaging Technique

“…26 It has been reported that visual assessment of IF with scoring (IFTA) according to the Banff classification suffers from poor interobserver reproducibility. 3,27 Moreover, the use of a small number of grades to describe the severity of individual histologic injuries may lack sensitivity during the early stage of IFTA. 28 To improve reproducibility and the performance of IF quantification, several techniques for computerized image analysis of IF in renal biopsies (stained by red Sirius or Masson trichrome) have been developed over the last decade.…”