The Utility of 1- and 3-Month Protocol Biopsies on Renal Allograft Function: A Randomized Controlled Study


Corresponding author: Dr. Vivekanand Jha,


Identification of pathological events in the renal allograft using protocol biopsies at predetermined time intervals may yield useful information and improve outcomes. We examined the influence of decisions taken on the basis of 1- and 3-month protocol biopsies findings on 1-year renal allograft function in a prospective randomized study. Out of 102 living-donor allograft recipients, 52 were randomized to undergo protocol biopsies and 50 controls had only indicated biopsies. All acute rejection (AR) episodes (clinical and subclinical) were treated. Calcineurin inhibitor (CNI) dose adjustments were made on clinical judgment. Baseline recipient and donor characteristics, immunosuppressive drug usage, HLA matches and 2-h cyclosporine levels were similar in both groups. At 1 and 3 months, protocol biopsies revealed borderline (BL) changes in 11.5% and 14% patients, AR in 17.3% and 12% and chronic allograft nephropathy (CAN) in 3.8% and 10%. The incidence of clinically evident AR episodes was similar in the two groups, but biopsy group had lower serum creatinine at 6 months (p = 0.0003) and 1 year (p < 0.0001). The renal functions were similar in those with normal histology and BL changes. Protocol biopsies are helpful in detecting subclinical histological changes in the graft and improving short-term renal allograft function.


Renal allograft biopsy is most accurate method for identification of pathophysiologic events that have a bearing on short- and long-term graft outcomes. Allograft biopsies are traditionally performed in the setting of acute or chronic graft dysfunction (1). Recent years have seen increasing recognition that a measurable decrease in renal function does not always accompany acute rejection (AR) or chronic allograft nephropathy (CAN), now called ‘interstitial fibrosis and tubular atrophy, not otherwise specified’. This has given rise to the practice of biopsying stable allografts at predefined posttransplant periods (protocol biopsies). Several studies have shown that protocol biopsies allow timely detection of AR, CAN and calcineurin inhibitor (CNI) nephrotoxicity. Therapeutic interventions carried out as result of information provided by these biopsies, especially treatment of subclinical AR, may improve clinical outcomes (2). Serial biopsies also allow evaluation of progression of histological lesions and provide an insight into the pathogenesis of chronic graft loss. The safety of this procedure has been documented in several series (3,4).

Despite encouraging reports on the value of protocol biopsies in management of kidney transplant recipients, there is a scarcity of randomized controlled trials that have evaluated the utility of this approach on short- or long-term outcomes. To the best of our knowledge, only one such trial has been published in its full form (5). Furthermore, most reports have come from programs where a majority of the grafts are obtained from deceased donors. Whether this information can be directly applied to recipients of kidneys from living related donors has not been examined.

We conducted a prospective randomized study with the aim of evaluating the utility of protocol biopsies in a living related transplant program and to ascertain whether proactive decisions taken on the basis of findings in these biopsies, including treatment of subclinical AR in the first 3 months after transplantation, would have a beneficial effect on renal allograft function.

Patients and Methods

The study was conducted at the Postgraduate Institute of Medical Education and Research, Chandigarh, India. The study protocol was approved by the Institute Ethics Committee. All patients who received a living donor kidney transplant over a 1-year period (July 2004 to June 2005) and had a stable graft function at 1 month were eligible for inclusion. Patients who were not receiving their first transplant and those who received cadaver grafts were excluded. Patients were informed about the study well before transplantation. All those who met the inclusion criteria and consented to participate were randomized to two groups; patients were not excluded on any other grounds.

In addition to intraoperative biopsy, biopsies were performed in one group (Group I) at the end of 1 and 3 months (±1 week) after transplantation. Group II patients only had the intraoperative biopsy. Patients in both groups also underwent biopsies whenever clinically indicated. The major trigger for such biopsies was acute graft dysfunction as defined by persistent increase of serum creatinine by >25% from baseline after excluding anatomic and surgical causes. The primary study endpoints were 6-month and 1-year serum creatinine values.

All patients received a triple drug immunosuppressive regime consisting of a CNI (cyclosporine or tacrolimus), antimetabolite (mycophenolate mofetil or azathioprine) and prednisolone. Two patients in the study group and four in control group received two doses of antiinterleukin-2 receptor antibody (basiliximab) on day 0 and 4. The choice of immunosuppressive agents was dictated by the financial status of the patients. Cyclosporine dose was titrated so as to achieve a 2-h (C2) level of 1200–1600 ng/mL in the first 2 months and 800–1200 ng/mL thereafter; and tacrolimus to keep the trough level at 10–12 ng/mL in the first month, 6–10 ng/mL in 2–3 months and 5–8 ng/mL thereafter. In view of universal donor and recipient seropositivity, no prophylaxis was given for CMV infection.

All posttransplant biopsies were done in the morning under ultrasound guidance using spring loaded 16-gauge biopsy gun under aseptic conditions and local anesthesia with 2% lignocaine. Two cores were obtained. Patients were confined to bed for the rest of the day. The stay was prolonged in the case of biopsy related complication, such as gross hematuria or perigraft hematoma until they become stable. Biopsies were analyzed by light microscopy by two experienced pathologists (K.J. and R.N.) and scored using the Banff's schema. C4d staining was not performed routinely.

All histologically proven (as classified by Banff schema) AR episodes (clinical and subclinical) were treated with three daily intravenous pulses of 500 mg of methylprednisolone. Patients with borderline (BL) changes did not receive any specific treatment, except CNI dose adjustment to maintenance blood levels in upper range of target values, and were followed up. Patients with evidence of CNI toxicity on indicated and protocol biopsies underwent 25–50% reduction in CNI dose. Additional CNI dose adjustments were made based on clinician judgment. The diagnosis of BK nephropathy was confirmed by immunohistochemistry. All patients were followed weekly for first 3 months, monthly for 6 months and 3 monthly thereafter.

The SPSS statistical package (v11.0.1) was used for data analysis. Student's t-test and Fisher exact test were used to compare continuous and categorical data, respectively. A p-value of less than 0.05 was considered statistically significant. Multiple linear regression analysis was used to measure the effect of donor age, HLA mismatches, mycophenolate mofetil (MMF) use and protocol biopsy on outcome measures.


A total of 102 patients were enrolled in the study, 52 were randomized to the study group and 50 served as controls. The baseline demographic characteristics of both groups are given in Table 1. All patient parameters were comparable between groups. The mean C2 levels the end of 1,3,6 and 12 months were comparable between groups. A total of 16 patients (13 and 3 in Groups I and II, respectively) were switched over from MMF to azathioprine 6–38 weeks after transplantation because of financial reasons; eight and two switches were made before the end of 3 months in Groups I and II, respectively. A total of 45 Group I and 46 Group II patients required antihypertensive drugs. The blood pressure values were comparable between groups at various time points (data not shown).

Table 1.  Baseline characteristics of the study population
 Group I (n = 52)Group II (n = 50)P value
  1. M = male; F = female; CIT = cold ischemia time; HLA = human leukocyte antigen; MMF = mycophenolate mofetil.

Recipient age (years)31.1 ± 1031.9 ± 100.69
Sex ratio (males:females)46:643:70.71
Basic disease
 Chronic glomerulonephritis19210.61
 Chronic interstitial nephritis 9 7 
 Diabetic nephropathy 2 3 
 Unknown and others2220 
Donor characteristics
Age (years)41.9 ± 1040.3 ± 130.49
Relationship of donor with recipient
 Spouse 711 
 Second-degree relative 2 2 
 Offspring 0 2 
Sex ratio (males:females)13:3913:370.90
Cold ischemia time (min)56.38 ± 1159 ± 200.42
HLA mismatches2.13 ± 0.61.95 ± 0.50.10
Induction with basiliximab2 (3.8%)4 (8%)0.37
Type of calcineurin inhibitor 0.43
 Tacrolimus 2 40.07
Type of antimetabolite3322 
2-h cyclosporine blood level (ng/mL)
 1 month1310.42 ± 367.21338.82 ± 391.950.71
 3 months1092.13 ± 310.4    1108 ± 297.4220.79
 6 (±1) months  936.35 ± 235.40919.012 ± 199.050.7 
 12 (±2) months  864.44 ± 181.28 836.44 ± 114.220.36

The only abnormalities noted in the 0-h biopsies were mild interstitial fibrosis involving <5% of cortex in six Group I and three Group II patients and mild arteriolar hyaplinosis in two Group I cases. None of these changes showed any progression on subsequent protocol biopsies.

Details of the protocol biopsy findings are given in Table 2. The third-month protocol biopsy could not be done in two patients as one did not report at the scheduled time and another had died of disseminated cytomegalovirus infection at 10 weeks.

Table 2.  Histological findings in protocol renal allograft biopsies 1 month and 3 months after transplantation
 1 month (n = 52)3 months (n = 50)
  1. 1One patients each had concomitant acute rejection and borderline changes.

Normal35 (67.3%) 33 (66%) 
Borderline changes6 (11.5%)7 (14%)
Acute rejection9 (17.3%)6 (12%)
Chronic allograft nephropathy2 (3.8%)  5 (10%)1
Cyclosporine toxicity01 (4%) 

Three patients with normal first-month protocol biopsies developed clinical AR in the second month. Graft biopsy showed moderate tubulitis (Banff 1a-2, 1b-1); all three responded to a 3-day course of methylprednisolone. The third-month biopsies were normal.

Table 3 shows the 3-month protocol biopsy findings in the different patients divided according to their 1-month biopsy class. Of the nine patients with subclinical AR in first month protocol biopsy, three continued to exhibit AR even in the third month despite having received treatment, and one showed BL changes. Cyclosporine was changed to tacrolimus in three, and dose was increased in the other patient. MMF was added in two cases. The other four had reverted to normal. There was no statistically significant difference in the third month protocol biopsy findings between patients with BL and normal histopathologic findings in the first month biopsy. Of the patients who had switched from MMF to Aza, one showed BL changes on the third-month protocol biopsy. Protocol biopsy findings led to immediate treatment changes in a total of 9 out of 52 (17%) 1-month and 8 of 50 (16%) 3-month biopsies.

Table 3.  Comparison of first and third month protocol biopsy findings
Banff grades on the first month biopsyBanff grades at third monthCyclosporine toxicity (n = 1)
Normal (n = 33)Borderline (n = 7)Acute cellular rejection (n = 6)Chronic allograft nephropathy (n = 5)1
  1. 1One patient each had concomitant AR and BL changes.

  2. 2Two patients did not undergo the third month biopsy.

Normal (n = 33)2264201
Borderline changes (n = 6)32110
Acute rejection (n = 9)41320
Chronic allograft nephropathy (n = 2)00020

There were 10 nonprotocol biopsies in nine patients the study group and 12 in control group for episodes of graft dysfunction. Details of findings are given in Table 4. In Group I, seven patients showed AR, six responded to methylprednisolone, whereas the seventh patient had steroid-resistant rejection (confirmed by repeat biopsy) and needed thymoglobulin. The one-month protocol biopsy was normal in this patient. In the control group, nine rejection episodes were noted in eight patients. All but one responded completely to IV methylprednisolone whereas in one patient, the response was incomplete. Group II had a higher frequency of CAN, but the numbers were too small to draw a meaningful conclusion. MMF was withdrawn and cyclosporine dose reduced by 25% in the lone patient with BK-nephropathy, and the graft function remained stable.

Table 4.  Nonprotocol biopsy findings in the two groups
FindingGroup I (n = 10)Group II (n = 12)p-Value
  1. 1Three patients had concomitant chronic allograft nephropathy and one had BKV nephropathy.

Acute rejection8 91>0.05
BK virus nephropathy01>0.05
Cyclosporine nephrotoxicity12>0.05
Chronic allograft nephropathy14>0.05

A total of three patients in Group I and two in Group II had suffered a rejection episode in the first month, before the first protocol biopsy was due. All the episodes were steroid responsive. Of the three Group I patients, one showed CAN on the first and third month protocol biopsies.

A significant difference was noted between the groups in terms of the primary outcome measure, that is the serum creatinine values at 6 months (p = 0.0003) and 1 year (p < 0.0001) (Table 5). The estimated GFR values using the abbreviated modification of diet in renal disease (MDRD) formula were also significantly higher in Group I at 6 (p = 0.0036) and 12 (p = 0.0029) months (Figure 1). On regression analysis, only protocol biopsies had a significant negative correlation with serum creatinine values at 6 months (p < 0.001) and 1 year (p < 0.0001).

Table 5.  Serum creatinine (mg/dL) at 1, 6 and 12 months in the two groups
TimeGroup IGroup IIp-Value
1 month1.18 ± 0.211.19 ± 0.240.82 
6 months1.28 ± 0.331.55 ± 0.39  0.0003
12 months1.20 ± 0.331.52 ± 0.41<0.0001
Figure 1.

Graph showing glomerular filtration rates calculated by abbreviated MDRD equation in the two groups at 1, 6 and 12 months (*p < 0.001, **p < 0.0001 vs. Group II).

We compared patients who received MMF with those who got azathioprine in terms of the 6-month and 1-year graft function and histology findings, and found no difference. The 6-month creatinine values in patients who received MMF and azathioprine were 1.26 ± 0.37 and 1.32 ± 0.28 mg/dL, respectively, (p = 0.51) and the corresponding 1-year values were 1.14 ± 0.43 and 1.33 ± 0.38 mg/dL (p = 0.10). The 6-month and 1-year creatinine values in those showing BL changes (not treated with antirejection therapy) were similar to those with normal histology (data not shown).

A total of four patients developed gross hematuria following protocol biopsies. In three, it was minor and resolved within a few hours whereas the in the fourth, it was accompanied by passage of clots which took 48 h to clear. All patients remained hemodynamically stable and none required blood transfusion.

A total of five Group I and six Group II patients developed infections. One of the Group I patients developed postoperative wound infection that healed with therapy, but later he developed disseminated CMV disease with superimposed pneumonia following treatment of steroid resistant rejection with thymoglobulin and died. Three patients with urinary tract infection and another with wound infection recovered with antibiotic therapy. The infections encountered in control group were urinary tract infection (three cases), wound infection, esophageal candidiasis and lymph node tuberculosis (one each). All responded to appropriate antimicrobial therapy.


This study establishes the utility of protocol biopsies in recipients of living related donor grafts through a randomized controlled trial. Graft function (as reflected by serum creatinine and MDRD eGFR) was superior at 6 months and 1 year amongst patients who underwent protocol biopsies, but there was no difference in the incidence of clinical AR episodes. Protocol biopsies are useful in detecting subclinical rejection (SCR) in early posttransplant period (6–13). In one study, SCR on 14-day protocol biopsy predicted poorer 10-year graft survival (14). Treatment of SCR has also been shown to favorably influence progression of interstitial fibrosis and tubular atrophy at 1 year (15), and it has been suggested that early intervention may improve long-term outcomes (2). In a pivotal trial, Rush et al. (5) randomized 72 patients to undergo protocol biopsies at 1, 2, 3, 6 and 12 months (biopsy group), or at 6 and 12 months only (control group). Patients in the biopsy arm of the study had a significant decrease in AR episodes, a reduced 6-month chronic tubulointerstitial score, and a lower 2-year serum creatinine. Patients without any rejection were less likely to show doubling of serum creatinine at 2 years than those with BL changes or AR. Serum creatinine did not increase and interstitial fibrosis was less in those treated for SCR.

Patients in our study were treated predominantly with cyclosporine and azathioprine. Our figures of subclinical AR in 17.3% and 12% of patients at 1 and 3 months and BL findings in another 11.5% and 14% patients at the same time points are lower than most previous studies of cyclosporine treated patients. The reported prevalence of Banff grade 1a or higher rejection has varied from 13 to 25% at 1–2 weeks, 11 to 43% at 1–2 months, 3 to 31% at 2–3 months and 4 to 50% at 1 year (6,12,16). The notable exception is the report of Roberts et al. (8) that documented subclinical AR in 13% and 8% at 7 and 28 days and BL changes in 12% and 16% at the same time points in cyclosporine-treated patients. Over a 10-year period, Nankivell et al. (17) prospectively evaluated 961 kidney-transplant biopsy specimens obtained from 120 recipients with type 1 diabetes, all but one of whom had received kidney–pancreas transplants. Clinically significant acute lymphocytic infiltration and tubulitis occurred in 61% at 1 month, 46% at 3 months, 26% at 1 year and 18% after 1 year.

Possible explanations for the variable incidence of SCR include differences in inclusion criteria, immunologic risk, delayed graft function (DGF), ethnicity and baseline immunosuppression. All our patients received grafts from living related donors; most had good HLA matches and none experienced DGF. DGF has been found to be an important risk factor for both clinical and SCR (6,18). Timing of biopsies is also a factor; the incidence of SCR is highest in 1-month biopsies and comes down at 3 months. CNI-treated grafts free of BKV infection virtually show no interstitial infiltration after 1 year (19).

Baseline immunosuppression is perhaps the most important factor determining the incidence of SCR. Tacrolimus and MMF treated patients invariably do better than their counterparts on cyclosporine and azathioprine (10). Tacrolimus use is associated with reduced incidence of SCR (4,11,17,20,21), lower acute Banff scores (22,23) and 1-year serum creatinine (21). Seron et al. (24) showed lower number of CD45, CD3 and CD68 positive cells in protocol biopsies from patients on tacrolimus. In view of this emerging data, whether tacrolimus and MMF treated patients will also benefit from protocol biopsies as shown in the current study will need to be examined afresh.

Of the nine patients who showed AR in the first month protocol biopsy, 33% continued to show AR even in the third month biopsy despite receiving treatment. Similarly, of the patients who showed BL changes at first month, one-third continued to show BL lesions even at 3 months. This phenomenon has been well described in literature (25).

The prevalence of CAN was also much lower than that reported so far, with the reported figures varying between 25% and 35% at 3 months (2,26–28). The factors that could account for these differences include differences in the time of biopsies and donor source (living related vs. deceased). The salutary effect of short cold ischemia times on the development of CAN has been documented (26,29). Also, the donor population in our study was relatively young, with a mean age of 42 years. In a study of paired allograft biopsies performed at 1 and 6 months, the rate of progression of CAN was about 25% slower in recipients of grafts from donors less than 55 years than those who got grafts from older donors (30). Oritz et al. (31) noted higher chronic allograft damage index scores in biopsies of recipients of grafts from older donors.

Mild tubulitis was present in about 10% of biopsies with otherwise normal histology, and in about 4% of cases with CAN. This is in agreement with the results of previous studies that reported that a substantial number of stable grafts display mild tubulitis (16,32,33) and also confirms that this lesion can be observed in biopsies with CAN. We did not treat this group.

In our study, the 6 month and 1 year creatinine values in patients with BL changes were similar to those with normal biopsies, but the small number of cases makes a meaningful conclusion difficult. Data regarding the utility of treating of BL changes are scarce and unclear. Often this lesion is clubbed with AR for purpose of analysis (25). In one study (14), untreated patients with BL changes performed better than those with SCR. Kee at al. (25) noted persistent SCR in 46% of cases despite treatment with pulse steroids, but treatment at 1 month was associated with lower acute Banff scores at 3 months. All patients including those with BL lesions had been treated in their study. Other studies (34), however, reported no significant differences in the outcome of patients with CAN with or without mild tubulitis. Misclassification of AR as BL changes and vice versa is also possible. To avoid this, Nankivell and Chapman (19) have suggested empiric treatment for both types of lesions. Another approach could be to strengthen baseline immunosuppression in such cases. In view of the lower incidence of SCR in tacrolimus-treated patients (17,20), a policy of switching over to this agent in those on cyclosporine has been advocated.

The exact way in which use of protocol biopsies helps in improvement of GFRs is difficult to pinpoint. In addition to allowing detection and treatment of histological abnormalities such as SCR and dose adjustments for potentially nephrotoxic CNIs, protocol biopsy findings might lead to other subtle changes in local management practices that contribute to preservation of GFR. Seikku et al. (35) commented on the value of tailoring of immunosuppressive therapy on the basis of findings from protocol biopsies. Only one patient in our study showed BK nephropathy on protocol biopsy. In a retrospective study (36), BK nephropathy was found in 7% patients on protocol and indicated biopsies. These patients lost their grafts more frequently than those without BKN. In an earlier study (37), we have documented BK nephropathy in 9% of all our patients.

A minor caveat of the study is the higher number of patients who initially received MMF in Group I, although the difference did not reach statistical significance. The decision to use MMF was based purely on financial reasons, that is patients' ability to afford the increased cost of therapy. A significant number of patients requested a change to azathioprine because of financial reasons. Moreover, neither the histologic findings nor the 6-month and 1-year creatinine values were different between groups who received MMF or azathioprine. Also, it needs pointing out that although we used the MDRD equation to estimate GFR, it has not been validated in Indian subjects.

To conclude, we show that protocol biopsies in the early posttransplant period can facilitate the detection of subclinical histological changes such as AR and CAN, and help in preservation of GFR in cyclosporine-treated living-related donor transplant recipients. The exact cause of this beneficial effect is difficult to ascertain, but could be due to a combination of treatment of SCR, modification of immunosuppressive drug dosage and other subtle changes in management practice.