Significance of the banff borderline biopsy

Significance of the Banff Borderline Biopsy Eugene J. Schweitzer, MD, Cinthia B. Drachenberg, MD, Leslie Anderson, CCRC, John C. Papadimetriou, MD, Paul C. Kuo, MD, Lynt B. Johnson, MD, David K. Klassen, MD, Edward Hoehn-Saric, MD, Matthew R. Weir, MD, and Stephen T. Bartlett, MD 0 In the Banff classification of kidney transplant pathology the “borderline changes” category falls short of a diagnosis of mild acute rejection, with the recommendation that no treatment is a possible clinical approach. We reviewed the clinical course of patients whose renal allograft biopsies showed “borderline changes” to determine how often these histologic findings actually represented acute rejection. Between January 1992 and June 1994, 361 biopsy specimens were obtained from 170 renal allografts and graded according to the Banff criteria. Eightyone biopsy specimens were classified as “borderline changes” (23%). Df these, 59 had Banff scores of il, tl, and ~0; the remaining 22 had scores of i2, tl, and v0 (i = interstitial infiltrate, t = tubulitis, and v = vasculitis). Medical record review showed that nearly all the “borderline” biopsies had been performed because of an elevated creatinine (Cr; 78 of 81 [sS%]), with a mean increase of 1.1 2 0.1 mg/dL @SE) over baseline. Most of the patients with “borderline changes” and elevated Cr were treated for acute rejection (81 of 78 [78%]); some with pulse steroids alone (29 of 81 [48%]), the rest with antilymphocyte antibody (32 of 81 [52%]). Among all 61 patients with “borderline” biopsies treated for rejection, 26 had a complete response (43%), 17 had a partial response (28%), and 18 had no response (30%). Interpretation of these changes in Cr, however, was confounded by intercurrent conditions in 28 of the patients. A group of 33 patients was therefore identified in whom a “borderline changes” biopsy was obtained, who were treated for rejection, and in whom all other identifiable causes of elevated Cr other than possible acute rejection had been systematically eliminated from consideration. In this group the mean Cr was 2.0 + 0.1 mg/dL at baseline, 3.3 2 0.2 mg/dL at the time of biopsy, and 2.2 ? 0.1 mg/dL 1 month after treatment (P < 0.001 Cr at biopsy v Cr 1 month later). Among these 33 patients, lg had a complete response (68%), 10 had a partial response (30%), and four had no response (12%). Therefore, the Cr in 88% of the patients in this group was lower 1 month after treatment for rejection than it was at the time of the biopsy. Follow-up biopsies were performed within 1 month of the “borderline” biopsy in 24 cases; these showed “borderline changes” (five of 24 [21%]), mild acute rejection (eight of 24 [33%]), or moderate to severe acute rejection (11 of 24 [46%]). We conclude that in the clinical setting of deteriorating renal graft function with mild elevation of serum Cr, the “borderline changes” biopsy frequently represents acute rejection. Antirejection treatment is therefore appropriate in the majority of cases. The reader should bear in mind that the current study is retrospective, with no control group. The risk of loosely interpreting these data is that some patients will be treated without due cause. Banff “borderline changes” should be used as part of an algorithm, but not the sole criterion, for therapeutic decision making. 0 1996

by the National

Kidney

Foundation,

INDEX WORDS: Kidney; transplant;

Inc.

biopsy; Banff; borderline.

P

ERCUTANEOUS biopsy has become a routine part of the evaluation of renal allograft dysfunction. One of the most important biopsy findings is the presence or absence of acute rejection. Certain findings, such as vasculitis, constitute clear evidence of acute rejection.‘” The significance of other findings, such as low levels of mononuclear cellular infiltration, remains in dispute. For instance, it has been shown that inflammatory infiltrates can occur in stable, wellfunctioning allografts.‘34‘6 The threshold level of interstitial infiltrate that constitutes acute rejection has not been well defined. The Banff classification of kidney transplant pathology has been developed to standardize biopsy interpretation, thereby facilitating use of the biopsy as an endpoint in clinical trials.7 In the Banff schema, the “borderline changes” category falls short of the threshold for mild acute rejection (which requires more than four lympho-

From the Departments of Surgery, Medicine, and Pathology, Universiry of Maryland School of Medicine, Baltimore, MD. Received November 7, 1995; accepted in revisedform May 29, 1996. Address reprint requests to Eugene .I. Schweitzer, MD, Department of Surgery, 29 S Greene Street #200, Baltimore, MD 21201. 0 1996 by the National Kidney Foundation, Inc. 0272~6386/96/2804-001.5$3.00/O

American

1996:

Journal

of Kidney

Diseases,

Vol 28, No 4 (October),

cytes per tubular cross-section), with the recommendation that no treatment is a possible clinical approach. We noted a number of patients in our practice with “borderline” biopsies whose clinical course was similar to one of acute rejection. In these cases, the “borderline” biopsy was the only finding in a search for the cause of a sudden increase in serum creatinine (Cr), and the Cr improved in response to antirejection therapy. We therefore reviewed the clinical course of all our

pp 585-588

585

586

SCHWEITZER

patients whose renal allograft biopsies showed ‘ ‘borderline changes’ ’ to determine how often these histologic findings actually represented acute rejection. PATIENTS

AND METHODS

Pathologic material and clinical data were obtained from the records of kidney transplant recipients treated at the University of Maryland (Baltimore). In this transplant program, management of kidney recipients follows standard clinical practice. In brief, Cr is assayed three times per week for 3 months, then weekly for 3 months, then monthly. If the Cr increases significantly (<25% over baseline), the patient is aggressively hydrated, cyclosporine or tacrolimus toxicity is excluded with blood trough levels, a renal ultrasound is obtained to exclude mechanical problems, and evidence of infection is sought. If other common causes of an increase in Cr are thereby excluded, a percutaneous renal biopsy under ultrasound guidance is obtained to rule out acute rejection. Biopsies are occasionally obtained in the absence of deteriorating renal function, as in cases in which kidney function remains stable but suboptimal. We also routinely biopsy kidneys during other major abdominal operations if it is safe and convenient. If there is a strong clinical suspicion or pathologic evidence of acute rejection, patients are treated with pulse steroids, ATGAM (UpJohn, Kalamazoo, MI), or OKT3 (Ortho Biotech, Raritan, NJ). Steroids are given initially when the rejection is considered mild to moderate. ATGAM or OKT3 is given in cases that fail to show a response to steroids within 48 to 72 hours, and as initial therapy for cases of moderate to severe acute rejection. Between January 1992 and July 1994, 351 biopsy specimens were obtained from 170 renal allografts (30 living donor, 140 cadaver donor). Four hematoxylin-eosin-stained slide and one periodic acid-Schiff-stained slide were examined for each biopsy, with three sections mounted per slide. All the biopsies were graded according to the Banff criteria by two transplant pathologists who were blinded to the identity and status of the patients. The biopsies were scored according to the degree of mononuclear cell infiltration, tubulitis, glomerulitis, intimal arteritis, hyaline and fibrous arteriolar thickening, interstitial fibrosis, and tubular atrophy. They were then classified into Banff diagnostic categories.’ Eighty-one biopsies were classified as “borderline changes” (23%), ie, specimens with only mild or moderate focal mononuclear cell infiltration with foci of mild tubulitis (one to four mononuclear cells/tubular cross-section). Among these, 59 had Banff scores of il, tl, and v0; the remaining 22 had scores of i2, tl, and v0 (i = interstitial infiltrate, t = tubulitis, v = vasculitis). The medical records of the patients with “borderline” biopsies were then reviewed. Rejection treatments were recorded, as well as the presence of documented or suspected intercurrent illnesses that might have effected Cr. Baseline Cr was defined as the average of the last two Cr levels during the last period of stable graft function. A “complete” response to rejection treatment was defined as a return of Cr to within 0.4 mg/dL of baseline within 1 month; a “partial” response was defined as a Cr at 1 month less than that at biopsy, but greater than 0.4 mg/dL of baseline; “no re-

ET AL

sponse’ ’ was defined as a l-month Cr greater than or equal to that at biopsy. A relatively long period of 1 month after the diagnosis of rejection was selected for the patients to attain their new baseline. This was done because it was noted that the Cr of some patients does not stabilize for several weeks, especially if the rejection episode is particularly severe, requires a prolonged course of rejection treatment, or involves adjustments in cyclosporine or tacrolimus doses. Creatinine levels were expressed as the mean value for the group 2 SEM, and were compared with paired r-tests. Pearson’s coefficient of correlation was calculated to relate Banff scores to the response to rejection therapy. RESULTS

Nearly all the “borderline” biopsies had been performed because of an elevated Cr (78 of 81 biopsies [96%]), with a mean increase of 1.l + 0.1 mg/dL over baseline. The three biopsies that were not performed for an elevated Cr were done as an incidental procedure during exploratory laparotomy. These patients continued to have stable, satisfactory graft function despite the fact that they were not treated for rejection. Most of the patients with “borderline changes” and elevated Cr were treated for acute rejection (61 of 78 [78%]); the 17 patients who were not treated for acute rejection had significant chronic transplant nephropathy in addition to the “borderline changes.” In each case it was felt that the mild tubulitis of the “borderline” biopsy was overshadowed by the chronic changes and that the possible benefit from treatment of acute rejection was outweighed by the risk. Some of the patients treated for rejection were given pulse steroids alone (32 of 61 [52%]); the remaining patients received antilymphocyte antibody (29 of 61 [48%]; 11 ATGAM and 18 OKT3), either initially for moderate to severe rejection or after failure to respond to steroids. The mean Cr in patients treated for rejection was 2.3 t 0.1 mg/dL at baseline, 3.4 ? 0.1 mg/dL at the time of biopsy (P < 0.001 u baseline), and 2.8 + 0.2 mg/dL 1 month after treatment (P < 0.01 u Cr at biopsy). Among all 61 patients with “borderline” biopsies treated for rejection, 26 had a complete response (43%), 17 had a partial response (28%), and 18 had no response (30%). There was no correlation between the response to rejection treatment and either the grade of Banff chronic transplant nephropathy or with the Banff scores for interstitial inflammation (Pearson correlation coefficient, r = 0.01 for both). Interpretation of these changes in Cr was con-

BANFF

BORDERLINE

BIOPSY

founded by intercurrent conditions in 28 of the patients. The major intercurrent conditions included sepsis (26%), chronic rejection (22%), renal artery stenosis (22%), cyclosporine toxicity (18%), ureteral obstruction (4%), acute tubular necrosis (4%) and severe congestive heart failure (4%). In this group the mean Cr was 2.7 ? 0.3 mg/dL at baseline, 3.5 + 0.2 mg/dL at the time of biopsy (P < 0.02 IJ baseline), and 3.5 + 0.3 mg/dL 1 month after treatment (P = NS u Cr at biopsy). Among these 28 patients, seven had a complete response (25%), seven had a partial response (25%), and 14 had no response (50%). Eliminating the patients with intercurrent conditions allowed for identification of a group of 33 patients who had a “borderline changes” biopsy, were treated for rejection, and had no identifiable cause of elevated Cr other than possible acute rejection. In this group the mean Cr was 2.0 2 0.1 mg/dL at baseline, 3.3 + 0.2 mg/dL at the time of biopsy (P < 0.001 u baseline), and 2.2 ? 0.1 mg/dL 1 month after treatment (P < 0.001 u Cr at biopsy). Among these 33 patients, 19 had a complete response (58%), 10 had a partial response (30%), and four had no response (12%). Therefore, the Cr in 88% of the patients in this group was lower 1 month after treatment for rejection than it was at the time of biopsy. This indicates that most of the patients with “borderline” biopsies and no other identifiable cause for elevated Cr responded at least partially to rejection treatment. Follow-up biopsies were performed within 1 month of the “borderline” biopsy in 24 cases, due either to doubt about the diagnosis or to failure to respond to therapy. These showed “borderline changes” (five of 24 [21%]), mild acute rejection (eight of 24 [33%]), or moderate to severe acute rejection (11 of 24 [46%]). DISCUSSION

Some investigators have expressed concern that the Banff “borderline” biopsy is indicative of early rejection and that withholding treatment would allow progression of the rejection process.’ Our data support this view. When the kidney biopsy is performed because of an elevated Cr, we have been inclined to treat patients with “borderline changes” for acute rejection, especially when other causes of elevated Cr have been eliminated. We have found that most respond

587

with a decrease in Cr, suggesting that acute rejection is being successfully treated. Furthermore, we found that most of our follow-up biopsies showed convincing evidence of ongoing rejection, indicating that the prior “borderline” biopsy was probably a manifestation of the same acute rejection process. The Banff schema has already become an invaluable tool for clinical research studies. The reproducibility of the scoring system is good,’ and it furnishes a standard means to categorize and numerically quantify pathologic changes. This provides homogeneity of biopsy interpretation among pathologists from multicenter trials and allows statistical analysis of biopsy results. Unfortunately, “borderline changes” can be correctly interpreted at times as a variant of normal and at other times as a variant of acute rejection. The correct interpretation is critical, since the decision to administer treatment for rejection depends on it. It is our practice to view the “borderline” biopsy from the clinical context where it was obtained. In the setting of stable graft function, we interpret the “borderline” biopsy as a variant of normal. In this case, we do not treat for rejection, but remain vigilant for signs of deteriorating graft function. In the setting of deteriorating graft function, we interpret the biopsy as a low point along the spectrum of acute rejection and treat it accordingly. While clinical information assists in interpreting the “borderline” biopsy, the decision to treat for rejection often remains difficult. For example, it is well known that needle biopsies may not be representative of the pathologic changes in the rest of the kidney.” Does the biopsy happen to exhibit the least, the most, or the average inflammatory infiltrate and tubulitis in the kidney? Furthermore, it may be true that the “borderline changes” often represent smoldering acute rejection, even in the setting of stable graft function. Does a change in the serum Cr provide a sensitive enough indicator for tubular injury by low-grade rejection? If not, will failure to treat “borderline” changes hasten the progression to chronic transplant nephropathy? Although we do not routinely perform immunohistologic staining of our kidney biopsy specimens, other investigators have examined the phenotypes of graft-infiltrating cells. Predominance of certain phenotypes, such as CD8 and isoforms

588

SCHWEITZER ET AL

of CD45 has been associated with acute rejection.” This type of analysis has been used to distinguish acute rejection from cyclosporine toxicity” and also might be helpful in interpreting the ambiguous “borderline” biopsy. In summary, Banff “borderline changes” can be seen in protocol biopsies of patients with stable graft function and might then be considered a variant of normal. In these cases we agree with the practice of other centers that withhold rejection treatment while remaining vigilant for signs of deterioration of function. However, most “borderline” biopsies are found in patients who have an increasing Cr. In this setting, the data in this study suggest that the “borderline changes” biopsy often represents acute rejection. We believe that intercurrent causes of an elevated Cr should be rapidly identified and treated. If acute rejection is still suspected, then another biopsy or antirejection treatment is appropriate. The reader should bear in mind that the current study is retrospective, with no control group. Furthermore, our conclusions are based in part on the assumption that an improvement in renal function after antirejection therapy was, in our series, due to successful treatment of rejection, not to resolution of some other unidentified problem. While valuable information can be obtained from this type of study, overinterpretation is a potential pitfall. The risk of loosely interpreting these data is that groups of patients will be treated without due cause. Banff “borderline changes” should be used as part of an algorithm, but not the sole criterion, for therapeutic decision making. REFERENCES 1. Burdick JF, Beschomer WE, Smith WJ, Mcgraw D, Bender WL, Williams GM: Characteristics of early routine renal allograft biopsies. Transplantation 38:679-684, 1984

2. Sibley RK, Rynasiewicz J, Ferguson RM, Fryd D, Sutherland DE, Simmons RL, Najarian JS: Morphology of cyclosporine nephrotoxicity and acute rejection in patients immunosuppressed with cyclosporine and prednisone. Surgery 94:225-234, 1983 3. Sanfilippo F, Kolbeck PC, Vaughn WK, Bollinger RR: Renal allograft cell infiltrates associated with irreversible rejection. Transplantation 40:679-685, 1985 4. Rush DN, Henry SF, Jeffery JR, Schroeder TJ, Gough J: Histological findings in early routine biopsies of stable renal allograft recipients. Transplantation 57:208-211, 1994 5. McWhinnie DL, Thompson JF, Taylor HM, Chapman JR, Bottom EM, Carter NP, Wood RF, Morris PJ: Morphometric analysis of cellular infiltration assessedby monoclonal antibody labeling in sequential human renal allograft biopsies. Transplantation 2:352-358, 1986 6. Matas AJ, Sibley R, Mauer SM, Kim Y, Sutherland DE, Simmons RL, Najarian JS: Pre-discharge, post-transplant kidney biopsy does not predict rejection. J Surg Res 32:269274, 1982 7. Solez K, Axelsen RA, Benediktsson H, Burdick JF, Cohen AH, Colvin RB, Croker BP, Droz D, Dunnill MS, Halloran PF: International standardization of criteria for the histologic diagnosis of renal allograft rejection: The Banff Working Classification of Kidney Transplant Pathology. Kidney Int 44:41 l-422, 1993 8. Dunnill MS: Histopathology of renal allograft rejection, in Morris PJ (ed): Kidney Transplantation: Principles and Practice. Philadelphia, PA, Saunders, 1994, pp 266-285 9. Marcussen N, Olsen TS, Benediktsson H, Racusen L, Solez K: Reproducibility of the Banff classification of renal allograft pathology. Inter- and intraobserver variation. Transplantation 60:1083-1089, 1995 10. Sorof JM, Vartanian RK, Olson JL, Tomlanovich SJ, Vincenti FG, Amend WJC: Histological concordance of paired renal allograft biopsy cores. Transplantation 60:12151219, 1995 11. Ibrahim S, Dawson DV, Sanfilippo F: Predominant infiltration of rejecting human renal allografts with T cells expressing CD8 and CD45RO. Transplantation 59:724-728, 1995 12. Platt JL, Ferguson RM, Sibley RK, Gajl-Peczalska KJ, Michael AF: Renal interstitial cell populations in cyclosporine nephrotoxicity. Transplantation 36:343-346, 1983