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

Proliferation Rate of Cells in the Interstitial Infiltrate in Acute Kidney Allograft Rejection

Steen Olsen and Hans Erik Hansen

University Hospital of Ćarhus, Denmark, Institute of Pathology, Kommunehospitalet and Department of Nephrology C, Skejby Hospital, Ćarhus, Denmark.


Address Reprint Requests to:	Professor Steen Olsen, M.D.
				Institute of Pathology
				8000 Ćarhus C, Denmark
Key Words: Renal allograft; Rejection; Proliferation rate; Ki-67 antigen


MIB-1 is an monoclonal antibody raised against recombinant fractions of the Ki-67 gene [1, 2]. The antibody reacts with all phases of the cell cycle except G0. It is effective in formaldehyde fixed paraffin wax embedded material and has been found to be the best proliferation marker for routine use in archival material [3].
We have used this method to study the proliferation rate in the interstitial infiltrate in renal graft biopsies from the first 90 days post transplant.



From our files were selected at random 24 graft biopsies performed during the first 90 days post transplant from 24 renal allografts during rejection episodes of histological grades 1, 2A, 2B and 3 according to the Banff system [4]. We also studied 8 protocol biopsies taken 3 weeks post transplant from well functioning grafts (patients without need of dialysis, serum creatinine < 150 ”Mol/l) as well as 8 baseline biopsies performed after the opening of the anastomoses at the transplantation.

Standard immunosuppression was cyclosporine, azathioprine and prednisolone. Rejection episodes were treated with methylprednisolone and monoclonal antibodies (OKT 3).
Formalin fixed renal biopsies embedded in paraffin wax and cut into serial sections at 2 ”m were immunostained according to the method of Catoretti et al [1]. Mouse monoclonal antibody MIB-1 against nuclear antigen Ki-67 was obtained from IMMUNOTECH, Marseille, France. The tissue sections were microwave treated with four cycles of 5 min each at 800 W. Lymph node was used as positive control for each batch of staining and a negative control (staining procedure without antibody) was added on each stained section.

Counting of cell nuclei of all as well as MIB-1 positive mononuclear leucocytes in the cortical interstitium was performed separately in dense, localized cell infiltrates and in tissue outside these infiltrates. The relative area of infiltrates was estimated by the point count method using total cortical area as reference. The counting was done with an unbiased counting frame [5] in a video image system using the GRID Stereological Package, Interactivision Inc., Silkeborg, Denmark. For statistical analyses the Wilcoxon signed rank method was used.



The results are shown in Table 1. The biopsies from the control groups (baseline biopsies and protocol biopsies 3 weeks post transplant) did not contain localized cell infiltrates. A slight tendency to higher numbers of interstitial cells in biopsies from functioning grafts at 3 weeks post transplant compared with baseline biopsies was not significant. There was a significant increase in the number of diffuse interstitial cells as well as of MIB-1 positive cells per mm2 and of the per cent of positive cell nuclei in acute rejection of all grades (p < 0.01) compared with the control groups , but no difference between the different grades of rejection. The area of focal interstitial infiltrates was not significantly different in the different grades. Six grafts which were not functioning or removed within one year because of rejection had values at biopsy which was in the same range as in biopsies from grafts which were functioning 1 year post transplant.

Mononuclear cells infiltrating in tubules (tubulitis) had MIB-1 positive nuclei in approximately the same relative number as cells in the infiltrates, but exact quantitation was not possible due to uncertainty of the precise location of some of these cells in relation to the tubular basement membrane in the immune staining.



Infiltration of the interstitium with mononuclear cells is a characteristic feature in renal allograft rejection, already known from experimental work before clinical allotransplantation had begun. The intensity or extent of the infiltrate has, however, no or only negligible predictive value for graft prognosis [6, 7, 8, 9] and the score for interstitial cell infiltration has not been included in the criteria for grading according to the Banff system [4]. Using antibodies against the proliferation cell nuclear antigen (PCNA), a method, which is considered to be less specific than MIB-1 [3], Thom et al. found that the percentage of PCNA-positive infiltrating cells was higher in clinical rejection than in non-rejection biopsies [10].

Our study was performed to investigate if proliferation of infiltrating cells is related to histological rejection grade and to the fate of the graft. The appearance of a reliable method to mark proliferating cells in formalin fixed, paraffin embedded material made it possible to perform this investigation on a historical material.

The data from our quantitative study show that the number of interstitial cells per square mm in diffuse infiltrates is almost doubled in rejection compared with controls and that concentration of mononuclear cells is increased with a factor of ten in localized infiltrates. The percentage of cells in division is significantly increased from a level of 2.4 - 3 % in controls to more than 20 % in grade 2 and 3. We found no significant relation between these cell concentrations or proliferation rates with increasing Banff grades. This is interesting in view of the correlation which has been found between clinical and histological rejection [4, 8](4, 8).

Rejection treatment led to reversal of clinical rejection in all episodes except in 6 patients who lost their grafts within 1 year post transplant due to rejection. These patients had all had rejection episodes graded 2B or more. In the biopsies studied from patients who lost their grafts the number of infiltrating cells in diffuse as well as localized infiltrates and the relative number of proliferating cells were all within the same range as seen in biopsies from patients with functioning grafts after one year.

We conclude that counting of cell concentrations and proliferation indices do not add information to the diagnosis or grading of rejection which can not be achieved by the current histological Banff system. Decisive for prediction of poor graft prognosis is transition from grade 2A to 2B as previously shown by our group [4].



  1. G. Cattoretti, et al., J Pathol 168, 357-363 (1992).


  2. G. Key, et al., J Clin Pathol 46, 1080-1084 (1993).


  3. D. Rose, P. Maddox, D. Brown, J Clin Pathol 47, 1010-1014 (1994).


  4. K. Solez, et al., Transplant Proc 27, 1009 - 1011 (1995).


  5. H. Gundersen, et al., APMIS 96, 379 - 394 (1988).


  6. B. Herbertson, D. Evans Calne, RY, A. Banerjee, Histopathology 13, 161 - 178 (1977).


  7. A. Hsu, G. Arbus, E. Noriega, J. Huber, Clin Nephr 5, 260 - 265 (1976).


  8. L. Gaber, L. Moore, r. Alloway, S. Flax, A. Gaber, Transplant Proc 27, 1019 (1955).


  9. H. Kjaer, H. Hansen, S. Olsen, Clin Nephrol 13, 58 - 63 (1980).


  10. M. Thom, A. Palmer, V. Catelli, T. Cook, Nephr Dial Transpl 9, 153 - 155 (1994).




Please note your browser must be capable of handling Tables. Newer versions of the Netscape browser are recommended.


Table 1:

  n Diffuse
pos %
pos %
area %
Baseline 8 384 ±131 9.4 ±2.6 2.4 - -   0
3 w plus fx 8 429 ±95 13.0 ±10.2 3.0 - -   0
Acute Rejection Grade 1 8 706 ±251 96.9 ±90.9 13.7 7855 ±2947 723 ±807 9.2 12.6 ±7.6
Acute Rejection Grade 2A 6 649 ±423 222 ±173 23.3 6969 ±2534 950 ±1014 13.6 4.0 ±2.8
Acute Rejection Grade2B +3 10 691 ±255 143 ±75.9 20.7 7936 ±2169 987 ±1022 11.3 10.5 ±9.7
All 40              
Rejected 1 year
(all grade 2B or 3)
6 809 ±207 119 ±130 14.7 7962 ±2584 711 ±321 8.9 13.6 ±10.0

Click here to go back.


Return to Banff Presentations


Last Modified: April 03, 1996 11:25:50 AM