Banff criteria as predictors of outcome following acute renal allograft rejection

The Banff classification of renal allograft rejection grades acute tubulointerstitial rejection (AIR) by severity of tubulitis and acute vascular rejection (AVR) by severity of arteritis. The intensity of tubulitis has not, however, been demonstrated to be of prognostic value and other features such as glomerulitis and eosinophil infiltration are of uncertain significance. This study was performed in order to determine the clinical value of this pathological classification. Banff criteria were correlated with outcome in 134 consecutive graft recipients transplanted in our unit over a 3-year period (1994 1996) who experienced at least one biopsy-confirmed acute rejection episode. Of 197 biopsies performed for the diagnosis of rejection, 177 contained at least one artery and were suitable for Banff grading. Tissue eosinophil counts were available for 101 biopsies. Clinical severity of rejection was classified as mild (fully responsive to pulse steroid therapy), moderate (partially steroid responsive) and severe (steroid unresponsive/requiring ATG therapy). Graft failure ensued in 18 of 58 patients with AVR compared with 10 of 65 patients with AIR (P= < 0.05). Clinical severity of rejection correlated with the presence of arteritis, but not severity of tubulitis; rejections graded I, IIA and IIB according to the Banff' 93 classification were clinically severe in 3/68 (4%), 2/28 (7%) and 15/67 (22%) respectively (P= 10 eosinophils/mm2) was present in 18 of 33 patients who had at least one episode of AVR (v1/2), compared with 11 of 45 patients who suffered only AIR (P=