Banff scoring of kidney allograft biopsies: "Manual" application vs software-assisted sign-out.

OBJECTIVES: Pathologists interpreting kidney allograft biopsies using the Banff system usually start by recording component scores (eg, i, t, cg) using histopathologic criteria committed to memory. Component scores are then melded into diagnoses using the same manual/mental processes. This approach to complex Banff rules during routine sign-out produces a lack of fidelity and needs improvement. METHODS: We constructed a web-based "smart template" (software-assisted sign-out) system that uniquely starts with upstream Banff-defined additional diagnostic parameters (eg, infection) and histopathologic criteria (eg, percent interstitial inflammation) collectively referred to as feeder data that is then translated into component scores and integrated into final diagnoses using software-encoded decision trees. RESULTS: Software-assisted sign-out enables pathologists to (1) accurately and uniformly apply Banff rules, thereby eliminating human inconsistencies (present in 25% of the cohort); (2) document areas of improvement; (3) show improved correlation with function; (4) examine t-Distributed Stochastic Neighbor Embedding clustering for diagnosis stratification; and (5) ready upstream incorporation of artificial intelligence-assisted scoring of biopsies. CONCLUSIONS: Compared with the legacy approach, software-assisted sign-out improves Banff accuracy and fidelity, more closely correlates with kidney function, is practical for routine clinical work and translational research studies, facilitates downstream integration with nonpathology data, and readies biopsy scoring for artificial intelligence algorithms.

Standardizing the histological assessment of late posttransplantation biopsies from pediatric liver allograft recipients.

Excellent short-term survival after pediatric liver transplantation (LT) has shifted attention toward the optimization of long-term outcomes. Despite considerable progress in imaging and other noninvasive modalities, liver biopsies continue to be required to monitor allograft health and to titrate immunosuppression. However, a standardized approach to the detailed assessment of long-term graft histology is currently lacking. The aim of this study was to formulate a list of histopathological features relevant for the assessment of long-surviving liver allograft health and to develop an approach for assessing the presence and severity of these features in a standardized manner. Whole-slide digital images from 31 biopsies obtained ≥4 years after transplantation to determine eligibility for an immunosuppression withdrawal trial were selected to illustrate a range of typical histopathological findings seen in children with clinically stable grafts, including those associated with alloantibodies. Fifty histological features were independently assessed and, where appropriate, scored semiquantitatively by six pathologists to determine inter- and intraobserver reproducibility of the histopathological features using unweighted and weighted kappa statistics; the latter metric enabled distinction between minor and major disagreements in parameter severity scoring. Weighted interobserver kappa statistics showed a high level of agreement for various parameters of inflammation, interface activity, fibrosis, and microvascular injury. Intraobserver agreement for these features was even more substantial. The results of this study will help to standardize the assessment of biopsies from long-surviving liver allografts, aid the recognition of important histological features, and facilitate international comparisons and clinical trials aiming to improve outcomes for children undergoing LT.

Late graft dysfunction after pediatric heart transplantation is associated with fibrosis and microvasculopathy by automated, digital whole-slide analysis.

BACKGROUND: Histopathologic features of late graft dysfunction (LGD) in endomyocardial biopsies (EMBs) after pediatric heart transplantation (HT) have been incompletely described and rarely quantified. We employed automated, morphometric analysis of whole-slide EMB images to objectively quantify fibrosis and microvasculopathy after pediatric HT. METHODS: Nine recipients with clinical LGD were matched with controls on age, listing diagnosis, crossmatch and time since HT. Fibrosis was quantified as percent tissue area with fibrosis and capillary density as capillaries per unit area, number of capillary "neighbors" within 30 µm of each myocyte and myocyte-to-nearest-capillary diffusion distance. Clinical data, including all EMB reports, were also reviewed. RESULTS: The groups were well matched for age at HT (median 4.0 vs 3.1 years), listing diagnosis (50% congenital heart disease for each), positive crossmatch (11% each) and days post-HT (2,628 vs 2,894, p = 0.69). Despite a similar number of previous EMBs (median 23 each, p = 0.43), areas occupied by fibrosis were greater in LGD cases (44.5% vs 23.2%, p = 0.012). Capillary number/area data were not statistically different between LGD cases and controls (378/mm2 vs 559/mm2, p = 0.57), but LGD cases more commonly had zero capillary neighbors (35% vs 20%, p = 0.02) and greater myocyte-to-nearest-capillary distances (27.1 µm vs 18.7 µm, p = 0.005). Cumulative rejection history correlated with fibrosis (r = 0.49, p = 0.039) and myocyte-to-nearest-capillary distance (r = 0.5, p = 0.036). CONCLUSIONS: LGD after pediatric HT is associated with previous rejection and characterized histologically by fibrosis and microvasculopathy, which are not readily appreciated by traditional semi-quantitative EMB analysis. Software-assisted EMB analysis may enable greater pathophysiologic understanding of LGD and identification of targets for future study and intervention.