Author:

Keywords:

Science & Technology, Life Sciences & Biomedicine, Surgery, Transplantation, 11 Medical and Health Sciences, 3202 Clinical sciences, 3204 Immunology

Abstract:

BACKGROUND The relative impact of different types of T-cell mediated rejection (borderline changes [BAR], grade 1 or grade 2-3 T-cell mediated rejection [TCMR]) and of antibody-mediated rejection (acute [aABMR] or chronic [cABMR]) on kidney transplant is currently not well known. METHODS We followed all 1197 renal allograft recipients transplanted at a single center between 1991 and 2001. During follow-up (mean 14.5±2.80 years post-transplant), 1365 post-transplant renal allograft indication biopsies were performed in 738 recipients. All biopsies were rescored according to the current Banff classification. Risk models for graft loss were calculated by multivariate Cox proportional hazards analysis. RESULTS During follow-up, 479 (35.1%) biopsies showed TCMR according to the Banff classifi cation (157 grade 1, 318 grade 2 and 4 grade 3), 209 (15.3%) had histology of aABMR (C4d positive with microcirculation inflammation), and 107 (7.84%) had cABMR (transplant glomerulopathy). When diagnosed within the first year after transplantation, only cABMR was associated with death-censored graft survival (adjusted HR 2.62 [1.05-6.53]; p=0.04); BAR, TCMR and aABMR in the first year after transplantation did not associate with graft outcome. If diagnosed after the first year after transplantation, both BAR (adjusted HR of 1.74 [1.02-2.97](p=0.04), TCMR (adjusted HR 3.41 [2.14-5.43], p<0.0001), aABMR (adjusted HR 3.00 [1.70-5.30], p=0.0001) and cABMR (adjusted HR 3.17 [1.34-7.51], p=0.003) were significantly and independently associated with death-censored graft survival. Of all graft failures later than 1 year post-transplant (N=238), 20.5% were preceded by transplant glomerulopathy, and 41.0% were preceded by aABMR or cABMR. There was no difference in outcome between grade 1 and grade 2-3 TCMR. CONCLUSION Not all rejection episodes have the same impact on graft outcome. Early rejection likely represents a non-specific response to injury (e.g. ischemia-reperfusion injury) and comprises no long-term risks if treated appropriately. Late rejection (including borderline changes) is an important predictor of graft outcome, and its underlying pathophysiology is likely less benign (e.g. insufficient suppression of allo-immunity, or immunosuppressive drug non-adherence). Interestingly, this time-dependency not only applies to T-cell mediated rejection, but also to antibody-mediated changes. Finally, diagnosis of chronic antibody-mediated rejection portends the major predictor of outcome, irrespective of the timing of occurrence.