In Press, 1995, Transplantation Proceedings copyright by Appleton and Lange.

Follow-up of Patients with Borderline Changes in Renal Allograft Biopsies. Clinical outcome and evaluation of other histological features in addition to tubulitis.

John C. Papadimitriou, Cinthia B. Drachenberg, Leslie Anderson, Stephen T. Bartlett, Lynt B. Johnson, David K. Klassen, Edward Hoehn-Saric, Mattew R. Weir, Eugene J. Schweitzer M.D.

University of Maryland School of Medicine, Departments of Pathology, Surgery and Medicine, Baltimore MD 21201


Address correspondence to :

                         John C. Papadimitriou,M.D.,Ph.D.
                         University of Maryland Hospital
                         Department of Pathology
                         22 South Greene St.
                         Baltimore, MD 21201
                         Phone: (410) 328 5560
                         Fax: (410) 328 5508



The Banff working classification of kidney transplant pathology [1] defined borderline changes as a "category with no intimal arteritis present, but only mild or moderate focal mononuclear cell infiltration with foci of mild tubulitis (1-4 mononuclear cells/tubular cross section)". The recommendation for these patients is to institute "no treatment, or treat other entity".

In this report we discuss a group of borderline biopsies from patients biopsied for acute allograft dysfunction who received antirejection treatment that appeared to have been beneficial in their clinical course.



Material derived from the kidney transplant recipients treated at the University of Maryland, Baltimore was studied. At this program, management of kidney transplant recipients follows the generally accepted guidelines with periodical serum creatinine (Cr) assessment. If the Cr rises >25% over the baseline, the patient is hydrated, cyclosporine (CsA) toxicity is excluded and possible mechanical and infectious problems are ruled out. If these common causes of increasing Cr are not confirmed, a percutaneous renal biopsy is obtained in order to evaluate the presence of rejection. In other cases biopsies are performed if the kidney function although stable remains suboptimal. Transplant biopsies are also obtained routinely during major abdominal operations if it is not otherwise contraindicated.

In the context of acute allograft dysfunction, if there is a strong clinical suspicion and/or pathologic evidence of acute rejection, treatment with pulse steroids, ATGAM or OKT3 are initiated. Steroids are given initially in cases of mild to moderate rejection. ATGAM or OKT3 are used if no response to steroids is seen within 48-72 hours, or initially for cases of moderate to severe acute rejection. Cases with borderline changes are also treated in that fashion if there is strong clinical suspicion for rejection.

Three hundred fifty one biopsies from 170 renal allografts were graded according to the Banff criteria by two transplant pathologists who were blinded as to the identity and any other clinical information about the patients. Classified following the Banff Working Classification for Rejection, 81 (23%) had "borderline changes", 218 had various degrees of rejection and 52 had no histologic evidence of rejection. The pertinent clinical information including rejection treatments, intercurrent illnesses, and Cr levels were then recorded.

Percentage of biopsy surface (0-100%) with mononuclear infiltrates, degree of eosinophilic infiltrates (0-3+) and degree of polymorphonuclear leukocytes (0-3+) per biopsy were compared in the 81 biopsies with borderline changes vs. 79 biopsies showing clear cut mild rejection according to the Banff scheme. We have previously shown that these three morphological features correlate in various degrees with tubulitis [3].



Ninety six percent of the biopsies with "borderline" changes (78/81) were from patients with a mean rise in Cr of 1.1 +/- 0.1 mg/dl over the baseline. The three remaining biopsies were obtained at the time of exploratory laparotomy. The latter patients were not treated for rejection and continued having stable graft function.

Otherwise, most patients with "borderline changes" were treated for acute rejection (61/78,78%). Seventeen patients were not treated because in addition to the borderline changes they had extensive chronic rejection. The mean Cr in patients treated for rejection was 2.3 +/-0.1 mg/dl at baseline, 3.4 +/-0.1 mg/dl at the time of biopsy (p<0.001 vs. baseline), and 2.8 +/-0.2 mg/dl one month after treatment (p<0.01 vs. Cr at biopsy). From the 78 patients treated for rejection a group of 33 was identified in which no other associated pathologic conditions were identified. In this group the mean Cr was 2.0 +- 0.1 at baseline, 3.3 +-0.2 at the time of biopsy (p < 0.001 vs. baseline) and 2.2 +-0.1 one month after treatment (p <0.001 vs. Cr at biopsy).

Follow up biopsies obtained within one month of the initial borderline biopsy in 24 cases showed borderline changes in 21%, mild acute rejection in 33%, and moderate to severe acute rejection in 46%.

Comparison between the histologic features observed in biopsies with borderline changes and mild rejection showed qualitative similarities between these two groups: In both cases many of the lymphocytes showed an "activated" phenotype (cleaved nuclei, prominent nucleoli, etc.), interstitial edema, tubular epithelial reactive changes, and occasionally peritubular capillary dilatation. The quantitative features, however, were significantly different. The mean percentage of biopsy surface covered by inflammatory infiltrates was 19.9% in borderline changes vs. 53.7% in the mild rejection cases (p <.005). The degree of infiltration by eosinophils was 1.7 in the mild rejection biopsies and 1.07 in the borderline cases (p <.005), whereas the degree of neutrophil infiltration was 1.09 and 0.26 respectively (p <.005).



The Banff working classification of renal allograft rejection represents an extremely useful tool to standardize the biopsy interpretation of renal transplant biopsies. The principles of the classification have been welcomed equally by pathologists and clinicians in our program where the results from each kidney biopsy are included in the daily computer print-out following the Banff numerical system. We are, however, confronted relatively frequently with biopsies from patients with acute deterioration of the graft function that can not be attributed to any other intercurrent condition but rejection, which show only borderline changes. Others have previously voiced concern that the "borderline" biopsy is indicative of early mild acute rejection [2] and suggested that withholding treatments would allow progression of the rejection reaction. Our data are consistent with this concept. In our program the tendency is to treat patients with "borderline" biopsies for rejection and the data in this communication document a beneficial effect of treatment in these patients.

Our results are further indicative of the value of the Banff schema, with the borderline category appearing to be a distinct entity that differs from both the normal condition of non-rejection as well as mild rejection. The interpretation of these findings, however, is heavily dependent on the clinical situation in our cases. In the setting of stable graft function, "borderline" biopsies are viewed with caution and the patients closely followed but not treated for rejection. In the setting of deteriorating graft function, "borderline" is interpreted as a very mild form in the spectrum of acute rejection and treated accordingly.



  1. Solez K, Axelsen RA, Benediktsson H, et al.: International standardization of criteria for the histologic diagnosis of renal allograft rejection: The Banff working classification of kidney transplant pathology. Kidney Int 1993;44:411.


  2. Dunnill MS: Histopathology of renal allograft rejection. In: Morris PJ, ed. kidney transplantation: Principles and Practice. Saunders, Philadelphia, 1994:266.


  3. Papadimitriou JC, Klassen DK, Hebel JR, Drachenberg CB: Quantitative evaluation of ancillary morphological parameters in the assessment of rejection in renal transplant biopsies. Int J Surg Path 1995;.


Return to
Banff Presentations


Last Modified: April 03, 1996 11:03:59 AM