Transplant glomerulitis is an active form of glomerular injury associated with suboptimal graft outcome, inadequate histologic reproducibility, and poorly understood pathogenesis. Using a modified pathologic schema where glomerular inflammation is defined by the presence of five or more leukocytes per glomerulus, we sought to assess the reproducibility of transplant glomerulitis and to prospectively investigate the pathogenesis of glomerular inflammation.Methods
Our cohort includes 59 kidney transplant recipients who underwent 60 “for cause” allograft biopsies. In addition to light microscopy, the majority of the biopsies were assessed using immunohistochemistry, immunofluorescence, and electron microscopy studies. Biopsies were classified as noninflamed (n=21), inflamed (borderline changes or above) without glomerulitis (n=21), and transplant glomerulitis (n=18). Peripheral blood was collected on the day of biopsy and cytokines secreted by peripheral blood mononuclear cells (PBMCs) were measured ex vivo.Results
Our modified schema had higher inter-observer agreement for detecting glomerulitis than that of the current Banff schema. Biopsies with glomerulitis showed ultrastructural signs of glomerular capillary wall remodeling. In contrast to other anatomic compartments, intraglomerular leukocytes in glomerulitis group consisted largely of monocytes. Patients with glomerulitis had high levels of IL-6 and IL-1β secreted by PBMCs. Furthermore, the percentage of inflamed glomeruli and the number of intraglomerular monocytes showed independent association with IL-6 and IL-1β levels, which tended to correlate with subsequent estimated glomerular filtration rate decline.Conclusions
Inter-observer reproducibility of transplant glomerulitis can be improved by using more stringent histologic criteria. Glomerular inflammation correlates with endothelial injury, monocyte influx, and IL-6 and IL-β secretion by circulating immune cells.