• Anti-HLA antibody;
  • donor-specific antibody;
  • kidney;
  • late antibody-mediated rejection;
  • transplantation;
  • transplant glomerulopathy


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

To define the relative frequency of phenotypes of transplant glomerulopathy, we retrospectively reviewed the findings in 1036 biopsies for clinical indications from 1320 renal transplant patients followed in our clinics between 1997 and 2005. Transplant glomerulopathy, defined by double contours of glomerular basement membranes (D), was diagnosed in 53 biopsies (5.1%) from 41 patients (3.1%) at a median of 5.5 years post-transplant (range 3.8–381 months). In cases with D, we studied the frequency of circulating anti-HLA alloantibody (A), peritubular capillary basement membrane multilayering (B) and peritubular capillary C4d deposition (C). B was present in 48 (91%) of D biopsies. C4d staining by indirect immunofluorescence was detected in 18 of 50 D biopsies studied (36%). By Flow PRA® Screening or ELISA, A was detected in 33 (70%) in 47 D cases with available sera, of which 28/33 or 85% were donor-specific. Class II (13/33) or class I and II (17/33) were more common than class I (3/33) antibodies. Thus 73% of transplant glomerulopathy has evidence of alloantibody-mediated injury (A and/or C), with ABCD and ABD being the common phenotypes in biopsies for cause. The remaining 27%, mostly BD, may be a different disease or a stage in which A and C are undetectable.


antibody-mediated rejection


T-cell mediated rejection


transplant glomerulopathy


glomerular basement membranes


peritubular capillary basement membrane multilayering


peritubular capillary


donor-specific antibody


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Late kidney allograft loss remains common (1), with a rising recognition that alloantibody plays a role in this problem (2). One important phenotype of late kidney deterioration is transplant glomerulopathy (TG), a term introduced by Zollinger et al. (3). Glomerular changes peculiar to transplanted kidneys manifesting with proteinuria and/or graft dysfunction in patients in whom the original disease was not glomerulonephritis were first observed in early 1960s (4–9). TG is characterized by double contours of glomerular basement membranes (GBM), often accompanied by increased mesangial matrix (10). TG features evolve over time. The earliest change is diffuse swelling of endothelial and mesangial cells with diminished patency of the capillary loops, progressing to double contours of GBM and an increase in mesangial matrix, sometimes with segmental sclerosis (11). Immunofluorescence is typically negative or may show nonspecific IgM and C3 (3,11–13). Electron microscopy shows widening of the lamina rara interna with subendothelial accumulation of electron lucent material, accompanied by formation of new basement membrane-like material with or without mesangial cell interposition, conferring a reduplicated appearance of the GBM (11,14,15).

Recent evidence increasingly supports the association of some cases of TG with alloantibody and C4d deposits in peritubular capillaries (PTC) (16–20). Some early reports related the severity of glomerular changes to the degree of HLA-incompatibility (3,21) and the number of rejection episodes (3), suggesting that TG resulted from repeated or sustained endothelial cell damage and regeneration. Mauiyyedi et al. (16) demonstrated deposition of C4d in PTC in 61% of 38 cases with TG and/or chronic allograft arteriopathy; most C4d positive cases had anti-donor HLA antibodies (88%). Palomar et al. (22) showed that patients with TG often had anti-HLA antibodies. We reported C4d deposition in PTC in 25% of 24 patients with TG but in none with recurrent IgA nephropathy (19). Sijpkens et al. (20) found C4d deposits in the glomerular capillary walls in 10/11 biopsies with TG. The association of TG with severe peritubular capillary basement membrane multilayering (PTCBMML) was initially reported by Monga et al. (23–27) Indeed, the glomeruli and PTC show similar basement membrane abnormalities with thickening and multilamination, which are regarded as markers of endothelial cell injury and repair (16,19,23–25,27,28).

Thus the emerging picture is of an association between the presence of alloantibody, capillary C4d deposition, TG and PTCBMML (29). However, the degree of associations among these features is currently unknown. Thus we performed a retrospective study to assess the overall frequency and associations of TG with or without anti-HLA antibodies. We aimed to estimate the frequency of TG, its associations and the common combinations in a renal transplant population. As the unique provider of transplant follow-up services in a large and relatively isolated geographic region, with a long-standing policy of biopsying all deteriorating kidney transplants, we reasoned that such a study help to define the extent of the TG problem and its associations with alloantibody.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Renal transplant patients

During the period from January 1997 to December 2005, a total of 1320 renal transplant patients were followed up at the University of Alberta Hospitals (Table 1). During this period, 1036 post-transplant renal allograft biopsies performed for unexplained deterioration of renal function and/or persistent proteinuria >1+ were included in the present study.

Table 1.  The denominator of the study: A total of 1320 renal transplant patients being followed from January 1997 to December 2005, were investigated for the presence of transplant glomerulopathy
YearNew transplantsTransfer inTransfer outFailuresTotal followed (n)
  1. 1Total number of patients who were followed at University of Alberta Hospital between January 1, 1997, and December 31, 2005: n = 583+ (new transplants 1998 to 2005 + transfer in 1998 to 2005).

1997117 212 36 583
1998 7910 7 36 628
1999 6912 4 25 680
2000 79 2 8 36 717
2001 9014 7 44 771
2002 81 3 9 32 807
2003 82 6 9 22 852
2004 971316 18 929
2005 94 613 26 989

The electronic database of Department of Laboratory Medicine and Pathology was searched for biopsies with glomerular double contours using different key words. Clinically, we recorded recipient age and gender, donor age and gender, post-transplantation day at biopsy, serum creatinine at biopsy, proteinuria, type of immunosuppressive drugs, HLA-mismatches, history of biopsy proven acute rejection episodes, and status of graft. The standard immunosuppression during this period was cyclosporine, prednisone and MMF or azathioprine. In 2001, tacrolimus replaced cyclosporine as the standard calcineurin inhibitor. The biopsies were evaluated by light microscopy, immunofluorescence, and electron microscopy. The histopathologic lesions, C4d deposition and alloantibody in TG biopsies were compared to a group of non-TG biopsies, matched for post-transplant time.


The biopsies were diagnosed and were scored according to the Banff classification (10,30). TG was diagnosed by light microscopy based on double contours of GBM (10) and was supported by immunofluorescence studies, which showed mesangial IgM and/or C3 or negative immunofluorescence findings.

Peritubular capillaritis in TG biopsies was graded according to the quantitative criteria proposed by Ian Gibson (29). At least 10% of cortical area with PTC inflammation is accepted as a threshold for grading of peritubular capillaritis. The extent of peritubular capillaritis was documented, either as focal (10%–50% of cortical area) or diffuse (>50% of cortical area). The type of luminal inflammatory cells was also noted.

Electron microscopy

Electron microscopy was performed in all biopsies after 3 months post-transplant to support the diagnosis of TG, to assess PTCBMML, and to exclude immune complex glomerulonephritis. The cases with double contours and concurrent immune complex deposits (n = 29) or microvascular changes consistent with thrombotic microangiopathy (TMA) (n = 2) were excluded from the study. PTCBMML was examined on electron micrographs and the highest number of basement membrane layers found on the overall PTC circumference was recorded. PTCBMML was also graded as; mild (2–4 layers), moderate (5–6 layers) or severe (≥7 layers) (25).

Indirect Immunofluorescence Staining for C4d

C4d staining was done in all biopsies with available frozen tissue. Briefly, murine monoclonal anti-human C4d 100 μL (Quidel Corporation, San Diego, CA), then fluorescent antisera (Cy™2-conjugated AffiniPure Goat Antimouse IgG, Jackson ImmunoResearch Laboratories, Inc, West Grove, PA) was added to the frozen sections (19). Biopsies from patients with previously documented diffuse positive C4d in PTC served as positive controls. The linear staining of PTC for C4d with an intensity of mild or greater was graded as diffuse (estimated >50% of sampled capillaries) or focal (1%–50% of sampled capillaries) positive (30). The C4d staining on biopsies with positive alloantibody and negative C4d was repeated twice.

Anti-HLA antibody analysis

Recipient sera taken in the peribiopsy period (median 1 day) were screened for HLA class I and class II antibodies by FlowPRA® (n = 72) or ELISA (QUIKSCREEN® I GTI® WI) (n = 5). If positive, antibody specificities were determined by FlowPRA® Specific class I and or II and or FlowPRA® Single Antigen I and II beads (One Lambda Canoga Park, CA). Beads were analyzed on a BD FACSCalibur™ cytometer. Antibodies specificities to HLA-A, B DRβ1 DRβ3 DRβ4 DRβ5 DQβ1 were evaluated.


The study was covered by approval (Protocol #5299) from the Institutional Review Board (Health Research Ethics Board, University of Alberta, Edmonton, Alberta, Canada).

Statistical analysis

Data analyses were performed using SPSS 15.0 (Chicago, IL, USA) and GraphPad Prism 4 (San Diego, CA). The relationship between variables was investigated using Spearman's correlation, Mann-Whitney U test, independent-samples T test, Chi-Square test, and Fisher's exact test, where applicable. Cumulative graft survival analyses were performed with the Kaplan–Meier method using log rank test. The end points for graft failure were defined as return to dialysis or patient death with functioning graft. The level of significance was set at p < 0.05. All values are given as mean ± standard deviation unless noted otherwise.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

The prevalence and patient demographics of TG

All 1320 kidney transplant patients followed between 1997 and 2005 (Table 1) were included in the denominator to estimate the prevalence of TG, not only the patients that were biopsied. Of the 1036 biopsies performed in these patients, TG was diagnosed in 41 (3.1%) patients and 53 (5.1%) biopsies. Thus the estimated prevalence of TG in our renal transplant population is 3.1% of all patients who were observed in this time.

The demographics of the 41 patients (21 males, 20 females) with TG are given in Table 2. Of the 37/41 patients with available HLA matching data, none was HLA identical. The mean HLA-A-B-DR mismatches was 3.8 ± 1.4 (range 1–6). In the 37 of 41 patients with information available on rejection history, 20 (54%) had experienced rejection episodes prior to the diagnosis of TG. Of 35 biopsies in these 20 patients, 26 (74%) showed clinical T-cell mediated rejection (TCMR) (including borderline, Banff type ≥IA) and 9 (26%) showed antibody-mediated rejection (ABMR) (C4d+). TG was diagnosed a median of 66 months post-transplant (range 3.8–381) in biopsies obtained because of proteinuria and/or acute or chronic renal dysfunction, with a mean serum creatinine at biopsy of 177 ± 99.8 μmol/L. Of 53 biopsies with TG, proteinuria (≥1+ by dipstick) at time of biopsy was present in 32, trace in 4, negative in 11 and unknown in 6 cases.

Table 2.  Demographic and clinical characteristics of patients (n = 41) with transplant glomerulopathy diagnosed by biopsy (n = 53)
 Mean ± SDMinimum–Maximum
  1. 1Donor age and gender, and HLA mismatches are not known in 4 of 41 transplants.

  2. 2Included are biopsy proven clinical acute rejection episodes. Prior acute rejection history is known in 37 of 41 transplants. TCMR = T-cell mediated rejection; ABMR = antibody-mediated rejection.

Recipient factors
 Age at biopsy (years)47 ± 1319–73
 Gender (M/F)21/20 
Donor factors
 Age at donation (years)1 37 ± 14.72–61.5
 Gender (M/F/unknown)122/15/4 
Transplantation factors
 Number of transplantation (1/2)35/6 
 Deceased/Living donor (n)29/12 
 HLA-A-B-DR mismatches1 (n)3.8 ± 1.41–6
Prior acute rejection2, n (% of patients)20 of 37 patients (54%) 
 Prior rejection, one/multiple (n)12/8 of 20 patients 
 Total number of biopsies with prior rejection35 
 Borderline, n (%)1 (3%) 
 TCMR, n (%)25 (71%) 
 ABMR, n (%)9 (26%) 
Characteristics at time of biopsy
 Post-transplantation time (median, mos)66 ± 733.8–381
 Serum creatinine (median, μmol/L) 177 ± 99.8108–647
 Proteinuria (≥+1, dipstick) (yes/trace/no/unknown)32/4/11/6 
 Urine protein (dipstick)
 +1, n6 
 >+1, n26 

Histopathology of TG

Of the 53 TG biopsies, 19 were mild (cg1), 24 moderate (cg2) and 10 severe (cg3) (Table 3). Of these, 46 biopsies showed increase in mesangial matrix: 1 mm3, 28 mm2, 17 mm1, 7 mm0. A representative figure for TG by light microscopy is given in Figure 1A. Of 53 TG biopsies, 48 (91%) showed fibrous intimal thickening in arteries (cv). Transplant glomerulitis (g) was present in 19 (35%) of 53 biopsies. Peritubular capillaritis was present in 37 (70%) TG cases either diffuse in 14 (27%) or focal in 23 (43%).

Table 3.  Histopathology and C4d staining of allograft biopsies (n = 53) with transplant glomerulopathy
 Mean ± SD
  1. 1Peritubular capillaritis were scored from 0 to 3, as described in Materials and Methods (29).

  2. 2C4d staining was available in 50 biopsies.

  3. MN = mononuclear inflammatory cells; PTC = peritubular capillaries; PTCBMML = peritubular capillary basement membrane multilayering.

Glomerular double contours (cg)1.8 ± 0.7
Increase in mesangial matrix (mm)1.4 ± 0.7
Interstitial fibrosis (ci)1.8 ± 0.7
Tubular atrophy (ct)1.7 ± 0.7
Arterial fibrous intimal thickening (cv)1.6 ± 0.8
Hyaline arteriolar thickening (ah)1.8 ± 1.0
Glomerulitis (g)0.6 ± 0.8
Interstitial inflammation (i)1.3 ± 0.8
Tubulitis (t)0.5 ± 0.8
Intimal arteritis (v)0.2 ± 0.5
Peritubular capillaritis11.5 ± 1.1
Peritubular capillaritis, inflammatory cell type, n (%)
 Neutrophils (>50%) and MN cells2 (4%)
 Neutrophils (<50%) and MN cells23 (43%)
 MN cells only12 (23%)
 Diffuse positive in PTC, n (%)13 (26%)
 Focal positive in PTC, n (%)5(10%)
 Negative, n (%)32 (64%)
 No PTCBMML, n (%)5 (9%)
 2–4 Layers, n (%)17 (32%)
 5–6 Layers, n (%)19 (36%)
 ≥7 Layers, n (%)12 (23%)
 Number of layers, median5.0 ± 2.2

Figure 1. (A) Transplant glomerulopathy with numerous double contours (arrows) in glomerular basement membranes and a moderate increase in mesangial matrix (periodic acid-Schiff, original magnification x600). (B) Transplant glomerulopathy by electron microscopy: There is electron lucent widening of the subendothelial space and formation of subendothelial new basement membrane layers (arrowheads) (Uranyl acetate-lead citrate, original magnification, x8000). (C) Peritubular capillary basement membrane multilayering with up to nine layers (arrows) (Electron microscopy, Uranyl acetate-lead citrate, original magnification, x3500). (D) Diffuse C4d positivity in peritubular capillaries (Immunofluorescence, Cy™2, original magnification x400).

Download figure to PowerPoint

By electron microscopy, TG was diagnosed by the presence of widening of the subendothelial space by deposition of electron lucent material and reduplication and/or multilamination of GBM, with or without mesangial cell interposition (Figure 1B).

Anti-HLA antibodies in cases with TG

Of the 53 biopsies, 47 had available sera for anti-HLA analysis (Table 4). Alloantibody was detected in 33 (70%), either against class I HLA (n = 3) or class II HLA (n = 13) or both (n = 17). Of the 33 with anti-HLA antibodies, 28 had donor-specific antibodies (DSA) against either class I (n = 4) or class II (n = 14), or both (n = 10).

Table 4.  The status of circulating anti-HLA antibodies in 47 of 53 transplant glomerulopathy biopsies with available HLA analysis in the peribiopsy period
  1. 1The cases with positive panel-reactive antibodies were tested for HLA antibody specificities using FLOWPRA® Specific Antibody Detection Test (One Lambda, Inc).

Anti-HLA panel-reactive antibodies
 Negative14 (30%)
 Positive33 (70%)
  − Anti-HLA class I antibodies3
  − Anti-HLA class II antibodies13
  − Both class I and class II antibodies17
Donor specific anti-HLA antibodies1
 Negative 5 of 33 (15%)
 Positive28 of 33 (85%)
  − Anti-HLA class I antibodies4
  − Anti-HLA class II antibodies14
  − Both class I and class II antibodies10

The presence of peritubular capillary basement membrane multilayering in cases with TG

PTCBMML was present in 48 (91%) of 53 TG biopsies (median 5.0 ± 2.2 layers) (Table 2). Of these, 17 showed mild (2–4 layers), 19 moderate (5–6 layers), 12 severe (≥7 layers) PTCBMML (Figure 1C). The degree of PTCBMML did not correlate with cg score or other Banff scores, C4d deposition (Table 5), presence of anti-HLA antibody, post-transplantation day, serum creatinine at biopsy, proteinuria, HLA mismatches, recipient age, donor age and prior acute rejection (p > 0.05).

Table 5.  The status of C4d deposition and PTCBMML in 50 of 53 TG biopsies with available C4d staining
 C4d staining in PTC
Diffuse+ (n = 13)Focal+ (n = 5)Negative (n = 32)
  1. For statistical analysis the diffuse/focal C4d positivity, no PTCBMML/2–4 layers, and 5–6/≥7 layers were combined. Fisher's exact test, p > 0.05.

No PTCBMML, n10 4
2–4 Layers, n4110
5–6 Layers, n5310
≥7 Layers, n31 8

We examined the morphology of PTC endothelium by electron microscopy: almost all biopsies showed signs of endothelial injury, including cytoplasmic swelling, microvillous projections, decrease or loss of fenestrations, and partial detachment from the basement membrane (Figure 2A–B). Some PTCs showed high endothelial venule-like changes (Figure 2C). Features of endothelial apoptosis and necrosis were rarely identified (Figure 2A). The presence of peritubular capillaritis with leukocyte aggregation was confirmed (Figure 2D).


Figure 2. The ultrastructural changes of peritubular capillary endothelium in transplant glomerulopathy biopsies, including (A) luminal microvilli formation (arrows), apoptotic bodies (insert), (B) cytoplasmic swelling, loss of fenestrations and partial detachment from the basement membrane (arrow), (C) high endothelial venule-like change with thickened endothelium and (D) aggregation of mononuclear inflammatory cells with adhesion to nonfenestrated endothelium. Some of the luminal inflammatory cells are necrotic. Note the accompanying basement membrane splitting and multilayering (arrowheads) (Uranyl acetate-lead citrate, original magnification, A: x3500, B–D: x4000).

Download figure to PowerPoint

C4d deposition in PTC in cases with TG

C4d staining was performed in 50 TG biopsies and present in 18 (36%) with either diffuse staining in 13 (26%) (Figure 1D) or focal staining in 5 (10%) in PTC. Glomerular capillary loop staining for C4d was observed in 4 (8%) of 50 cases with concurrent PTC C4d deposition. C4d deposition in PTC was strongly associated with the presence of anti-HLA antibody (p = 0.004). All diffuse C4d positive cases (n = 13) had circulating anti-HLA antibodies. Of the five focal C4d positive cases, four had anti-HLA antibody, but one did not. C4d deposition did not correlate with individual Banff scores and other clinical features (p > 0.05).

TG associated with anti-HLA antibody, PTCBMML, and C4d

The association of TG (‘D’) with anti-HLA antibody (‘A’), PTCBMML (‘B’) and C4d (‘C’) is summarized in Table 6. Of the 45 biopsies with TG (“D”) in which both C4d staining and anti-HLA testing were available, 16 showed ‘ABCD’, 13 ‘ABD’, 1 ‘ACD’, 2 ‘AD’, and 1 ‘BCD’ phenotype. Ten showed ‘BD’ and two showed ‘D’ alone. Most cases of TG (73%, 33 of 45) showed features of ABMR (ABCD, ABD, ACD, AD, BCD). The remaining 27% (12 of 45 TG) did not show C4d and anti-HLA antibody at time of biopsy. Thus the common phenotypes were ABCD, ABD and BD. No biopsy showed a ‘CD’ phenotype. The distribution of anti-HLA antibody, PTCBMML, C4d and histologic lesions in TG biopsies is shown in Figure 3.

Table 6.  The association of TG (D) with anti-HLA antibody (A), PTCBMML (B) and C4d (C) in 45 cases in which both C4d staining and anti-HLA testing were available
TG phenotypen (%)
‘ABCD’16 (36)
‘ABD’13 (29)
‘ACD’1 (2)
‘AD’2 (4)
‘BCD’1 (2)
‘BD’10 (22)
‘D’2 (4)

Figure 3. The distribution of histologic lesions including cg, mm, g, peritubular capillaritis (ptc), and cv, peritubular capillary basement membrane multilayering (PTCBMML), C4d deposition and status of circulating anti-HLA antibodies in cases with transplant glomerulopathy (n = 45).

Download figure to PowerPoint

We compared 53 TG biopsies to 31 post-transplant time matched non-TG biopsies. The pathologic diagnoses in non-TG biopsies included TCMR (n = 8), calcineurin inhibitor toxicity (CNIT) (n = 8), glomerulonephritis (n = 5), borderline (n = 4), BK (n = 2), acute ABMR (n = 1) and other (n = 3). Compared to the non-TG biopsies, TG biopsies were significantly associated with C4d deposition (36% vs. 3%, p = 0.001) and presence of circulating alloantibody (70% vs. 43%, p = 0.01). However, the extent of PTCBMML (91% vs. 74%) and cv scores (1.6 ± 0.8 vs. 1.5 ± 0.7) was not different in time matched TG and non-TG biopsies (p > 0.05), indicating that both lesions are common in late allograft biopsies for cause (Table 7). The PTCBMML was not associated with alloantibody in both TG and non-TG biopsies (p > 0.05).

Table 7.  Comparison of clinico-pathologic features in TG biopsies vs. non-TG biopsies
  1. 1Anti-HLA antibody analysis was available in 47 of 53 TG biopsies and 30 of 31 non-TG biopsies in the peribiopsy period.

  2. PTCBMML = peritubular capillary basement membrane multilayering.

Post-transplant time (median, mos)66 ± 7369 ± 65NS
Recipient age at biopsy (years)47 ± 1349 ± 12NS
Donor age at donation (years)  37 ± 14.748 ± 130.001
Cg1.8 ± 0.70 ± 00.000
mm1.4 ± 0.70.5 ± 0.70.000
Cv1.6 ± 0.81.5 ± 0.7NS
 Number of layers, median5.0 ± 2.24.0 ± 2.4NS
 No PTCBMML, n (%)5 (9%)8 (26%) 
 2–4 Layers, n (%)17 (32%)9 (29%) 
 ≥5 Layers, n (%)31 (59%)14 (45%)  
C4d in peritubular capillaries18 (36%)1 (3%) 0.001
Anti-HLA panel-reactive Ab 1
 Positive, n33130.01 
 Negative, n1417 
Donor-specific anti-HLA Ab
 Positive, n2870.03 

C4d and anti-HLA antibody on historic samples from TG patients

The TG patients with available historic serum and biopsy samples are tested for panel reactive antibodies and C4d deposition. Of 27 TG patients with available historic sera (median 39 ± 853 days post-transplantation), 14 (52%) patients had anti-HLA antibodies. Of 14 TG patients with available previous biopsies and frozen tissue (n = 40 biopsies), 6 allografts (43%) showed C4d deposits in PTC prior to the diagnosis of TG. The patients with historic C4d (6/6) or alloantibody (13/14) developed TG with antibody/C4d in the subsequent biopsy.

TG with no alloantibody or C4d: possible associations with other diseases

TG biopsies with no antibody or C4d showed evidence for chronic CNIT with more ah lesions in comparison to TG with antibody/C4d (9 of 12, 75% vs. 18 of 33, 54%). The average ah score was higher in TG with no alloantibody or C4d (1.9 ± 1.1 vs. 1.5 ± 1) but this difference did not reach a statistical significance (p > 0.05). None of the TG biopsies showed isometric vacuolation, viral infection or TMA (the biopsies with microvascular thrombosis and cg excluded from this study). However, one TG patient had TMA in previous biopsies suggesting TMA as an underlying etiology in this particular case.

Of 12 TG cases with no antibody or C4d, 4 allografts had previous biopsies (n = 11) and none of these showed C4d deposition. Historic sera was available in 7 of 12 TG with no A or C: six had no alloantibodies and one had historic donor-specific class II antibodies (10 years before the TG biopsy) but did not have a prior biopsy sample.

TG with ABMR features vs. TG without ABMR features

We compared the clinico-pathological parameters in TG with alloantibody and/or C4d (n = 33) vs. TG with no alloantibody or C4d (n = 12). The glomerulitis (p = 0.03) and peritubular capillaritis (p = 0.05) scores were higher in TG with ABMR features (Table 8). The ultrastructural changes of PTC endothelium were similar in C4d positive TG and C4d negative TG biopsies. The other Banff scores, PTCBMML, recipient age, donor age, number of transplants, number of HLA mismatches, serum creatinine and proteinuria at biopsy were similar in TG with alloantibody and/or C4d and TG with no alloantibody or C4d (p > 0.05).

Table 8.  The quantification of peritubular capillary aggregation of inflammatory cells (peritubular capillaritis) and glomerulitis in TG biopsies with alloantibody/C4d vs. TG biopsies with no alloantibody or C4d
 TG with Ab/C4dTG with no Ab/C4d
  1. 1Glomerulitis and peritubular capillaritis were scored from 0 to 3, as described in Materials and Methods (29).

  2. MN = mononuclear inflammatory cells.

  3. Mann-Whitney test or Chi-Square test, ** p = 0.03, * p = 0.05.

Glomerulitis, mean score1 0.7 ± 0.8**0.2 ± 0.6
Peritubular capillaritis, mean score11.7 ± 1.0*1.0 ± 1.1
Peritubular capillaritis, extent1
 Absent, n (%)7 (21)6 (50)
 Present, n (%),26 (78)*6 (50)
 Diffuse/Focal, n7/194/2
Inflammatory cell type, n
 MN cells only10 of 261 of 6
 Neutrophils (<50%) and MN cells16 of 263 of 6
 Neutrophils (>50%) and MN cells02 of 6
PTC dilatation, n (%)15 (45)5 (42)

We explored whether the presence of alloantibody/C4d in TG influences graft survival. At 15 years post-transplant, cumulative death-censored graft survival was 35% in TG with no alloantibody/C4d vs. 40% in TG with alloantibody/C4d (log-rank test, p = 0.85) (Figure 4). Similar results were obtained when patient death was not censored (p > 0.05).


Figure 4. Cumulative death-censored graft survival curves in patients who developed TG with alloantibody/C4d vs. TG with no alloantibody or C4d (log rank test, p = 0.85).

Download figure to PowerPoint

We analyzed whether TG was associated with a history of previous episodes of TCMR or ABMR. Of 30 TG with ABMR features (A or C), 17 (57%) had no prior rejection, 9 (30%) prior TCMR, and 4 (13%) prior early ABMR. The median interval between the diagnosis of TG and prior ABMR was 12.1 ± 10 months. Of 11 TG cases without ABMR features (no A or C), 6 (55%) had no prior rejection, 5 (45%) prior TCMR, but none had prior ABMR. The TG cases with A or C had more ABMR episodes in the past, but this observation did not reach a statistical significance (Figure 5) (p > 0.05). The number of prior rejection episodes was not different between two groups (p > 0.05).


Figure 5. Comparison of prior acute rejection in transplant glomerulopathy (TG) with anti-HLA antibody/C4d vs. TG without anti-HLA antibody/C4d. TCMR = T-cell mediated rejection; ABMR = antibody-mediated rejection.

Download figure to PowerPoint

TG with proteinuria vs. TG without proteinuria

TG biopsies with proteinuria (n = 32) had higher cg and mm score in comparison to TG with no proteinuria (n = 11) (p < 0.05). Otherwise, circulating anti-HLA antibody, C4d deposition in PTC, clinical and other pathological features were similar in both TG groups with or without proteinuria (p > 0.05) (Table 9).

Table 9.  The presence of anti-HLA antibody, C4d deposition, and clinico-pathologic features in TG with proteinuria vs. no proteinuria
 TG, proteinuria (+)TG, proteinuria (−)p
  1. 1Anti-HLA analysis was available in 41 biopsies.

  2. PTC = peritubular capillaries; PTCBMML = peritubular capillary basement membrane multilayering.

Number of biopsies3111 
Recipient age at biopsy (years)46 ± 1247 ± 10
Donor age at donation (years)37 ± 1528 ± 15
Number of transplantation1.1 ± 0.31.3 ± 0.5
Deceased/Living donor (n)21/1110/1
HLA-A-B-DR mismatches (n)3.6 ± 1.34.4 ± 1.5
Prior acute rejection (n)15 of 326 of 11
Post-transplantation time (median, mos)76 ± 8391 ± 51
Serum creatinine (median, μmol/L)177 ± 104175 ± 63 
Glomerular double contours (cg) 2 ± 0.61.6 ± 0.9 .003
Increase in mesangial matrix (mm)1.5 ± 0.81.2 ± 0.4.02
Interstitial fibrosis (ci)1.8 ± 0.71.5 ± 0.5
Tubular atrophy (ct)1.8 ± 0.71.3 ± 0.6
Arterial fibrous intimal thickening (cv)1.7 ± 0.81.3 ± 0.6
Hyaline arteriolar thickening (ah)1.7 ± 1.11.5 ± 0.9
Glomerulitis (g)0.5 ± 0.80.4 ± 0.8
Interstitial inflammation (i)1.3 ± 0.91.1 ± 0.7
Tubulitis (t)0.5 ± 0.80.5 ± 0.6
Intimal arteritis (v)0.2 ± 0.40.1 ± 0.6
C4d deposits in PTC
 Diffuse positive, n84
 Focal positive, n4
 Negative, n197 
 PTCBMML present, n2711
 PTCBMML absent, n5
Anti-HLA panel-reactive Ab 1
 Positive, n218
 Negative, n102 
Donor-specific anti-HLA Ab
 Positive, n177


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

We studied the frequency and clinico-pathological characteristics of TG, and its associations with anti-HLA antibodies, capillary C4d deposition, and PTCBMML. The 3.1% estimated prevalence of TG in our renal transplant population was similar to previous reports of 1.6% to 7% (11,19,20,31,32). The common phenotypes of TG were ABCD, ABD and BD. The main finding in the current study is that the majority of biopsies for cause with TG (73%) had evidence of antibody-mediated injury, with C4d deposition and/or circulating anti-HLA antibodies (ABCD, ABD, AD, ACD, BCD). Of TG biopsies with available sera, 33 of 47 patients had anti-HLA antibodies, a majority of which were donor specific. Positive C4d staining in TG cases was strongly correlated with anti-HLA antibodies. There were no cases of TG with C4d positivity and no alloantibody. The incidence of C4d deposition in TG was lower than the incidence of circulating alloantibodies, indicating that C4d deposition along capillaries may be negative or fluctuating. Thus C4d negativity does not necessarily exclude alloantibody-mediated glomerular damage. (In acute ABMR, loss of C4d from the tissue has been documented as early as 8 days after the treatment (33).) Our results indicate that the features of late ABMR may be present at different time points, eventually leading to chronic graft injury.

The method of calculating the prevalence of TG in our series is a conservative estimate, since we assumed all patients followed for even brief periods of time in our clinics as the denominator, including in the denominator even patients with no biopsies. It is possible that there may be silent TG cases which were not biopsied. A protocol biopsy study has found that TG is present in 5.4% of clinically stable allografts by 10 years after transplantation (34). The frequency of TG may be much higher in patients with late allograft dysfunction: Ivanyi et al. (27) reported the incidence of TG as 57% of 91 biopsies from such patients. In our series, TG occurred at a median of 5.5 years with a minimum of 3.8 months after transplantation. In previous reports, TG was diagnosed as early as 10 weeks post-transplant (3), with a median of 8.3 years in a cohort of allograft biopsies from 1111 kidney transplants (20).

Our observation that 70% of TG cases had circulating anti-HLA antibodies in the peri-biopsy period, with C4d staining being in the anti-HLA-positive cases, confirms the major role of alloantibodies in the pathogenesis of TG. The reason for many anti-HLA positive cases being C4d negative is not clear: using >50% as a cut-off for diffuse PTC staining, we observed that 26% of TG biopsies had diffuse and 10% focal C4d deposition in PTC. Furthermore, C4d deposits and alloantibody in historic samples preceded the development of TG, as previously shown by Regele et al. (17). The incidence of C4d in TG in our series confirms our previous result (with some overlap in the cases) but is lower than in some studies (16–20). On the other hand, the reported incidence of C4d positivity in TG is quite variable, probably related to criteria for biopsing patients as well as differences in C4d detection methods, e.g. the type of anti-C4d antibodies (monoclonal or polyclonal) and the defining criteria for C4d positivity. Al Aly et al. (35) observed no C4d staining in 20 TG biopsies, but they did not study anti-donor antibodies. Our data showed that about half (15 of 27) of the C4d negative TG cases had circulating alloantibodies in the peribiopsy period.

The remarkably strong association of D with B - 91% of TG biopsies had PTCBMML, again underlines the probability that these lesions share a common pathogenesis. TG and PTCBMML both show basement membrane thickening and multilayering, which are regarded as markers of endothelial cell injury and repair. We previously observed that PTCBMML was significantly increased in TG but not in recurrent IgA nephropathy (19). Drachenberg et al. (36) showed that TG was mostly associated with severe PTCBMML (more than 6 layers), whereas lesser degrees of these changes (mostly 2–3 layers) were observed in transplants and in native kidneys with other types of glomerulopathies, diabetes and hypertension. In our series, the incidence and severity of PTCBMML in non-TG biopsies was also high, indicating that PTCBMML is common in late allografts. On the other hand, this analysis should be interpreted with caution, because the number of suitable time matched non-TG biopsies was relatively small.

The 91% association between fibrous intimal thickening in arteries and TG suggests that alloantibody-mediated vascular injury can trigger fibrous intimal thickening. On the other hand, time matched non-TG biopsies also had high cv scores. Recent experimental studies by Smith et al. (37) in long-surviving kidney allografts in nonhuman primates found that alloantibody responses and TG are often accompanied by a similar arteriopathy.

The fact that 54% of patients with TG in the current study had experienced prior TCMR (44%) or ABMR (10%) episodes makes us suspect that previous rejection is probably a clinically significant risk for TG. Although small numbers preclude statistical analysis, the TG cases with antibody plus C4d had more prior ABMR episodes, compatible with a recent report that early ABMR predicts increased risk of TG (38). Of interest, no TG cases occurred in HLA identical transplants. To our knowledge, TG has not been reported in ABO compatible allografts from HLA-identical siblings, although a role for non-HLA immunity in such recipients with panel reactive antibodies is suggested (39).

The remaining 27% of TG without evidence of alloantibody-mediated injury (BD and D) could represent a different disease mechanism (e.g. TCMR) or a stage in which ABMR features have disappeared. The TG biopsies with no antibody or C4d had more arteriolar hyalinosis suggesting a link between some TG cases and CNIT, but there is no proof of statistical significance. Although the data on historic samples is not supportive to call BD, D cases as an inactive stage of ABMR, we still cannot entirely rule out this possibility. Two recent studies found no differences in incidence of C4d deposition in TG versus biopsies without TG, although at least 50% of TG cases had C4d staining in both reports (40,41). Akalin et al. (41) showed glomerular infiltration by CXCR3+ ICOS+ activated T cells in grafts with TG, suggesting that effector T-cell response to glomerular antigens can result in TG. However, the presence of PTC C4d deposits in 4 of 5 TG biopsies suggests that alloantibody may have been the real culprit in these cases.

The D phenotype as defined in Banff may underestimate the problem of late ABMR, because other phenotypes may be associated with TG. In our series, we identified 29 additional biopsies with glomerular double contours and concurrent immune deposits. We suspect that some represent an emerging phenotype in which double contours of TG are associated with coincidental immune complex deposits, i.e. TG associated with IgA nephropathy. It is also possible that anti-HLA antibodies may trigger the formation of immune complex deposits. Other possible non-TG phenotypes may emerge as consequences of alloantibody-mediated injury, such as progressive fibrosis or fibrous intimal thickening of small arteries remain to be tested. The further refinement of our understanding of the phenotypes of late kidney transplant deterioration will require careful clinical observations of deteriorating transplants and an aggressive biopsy policy, to determine the extent to which the current criteria describe the problem of late graft deterioration, the sequence of events and above all the possibilities for intervention.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

The authors wish to thank Dr. Sakarn Bunnag, Ms. Susan Huygen and Dr. Zija Jacaj for assistance with retrieval of clinical records; Dr. Sita Gourishankar for outcome data; and Dr. Deborah James (Department of Medicine, University of Alberta) and Dr. Bruce Kaplan (Department of Medicine and Pharmacology, Multiorgan Transplant Institute, University of Illinois) for critical reading of the manuscript.

This research has been supported by funding and/or resources from Genome Canada, Genome Alberta, the University of Alberta, the University of Alberta Hospital Foundation, Roche Molecular Systems, Hoffmann-La Roche Canada Ltd., Alberta Innovation & Science, the Roche OrganTransplant Research Foundation, the Kidney Foundation of Canada, and Astellas Canada. Dr. Halloran also holds a Canada Research Chair in Transplant Immunology and the Muttart Chair in Clinical Immunology.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  • 1
    Meier-Kriesche HU, Schold JD, Srinivas TR, Kaplan B. Lack of improvement in renal allograft survival despite a marked decrease in acute rejection rates over the most recent era. Am J Transplant 2004; 4: 378383.
  • 2
    Terasaki PI. Humoral theory of transplantation. Am J Transplant 2003; 3: 665673.
  • 3
    Zollinger HU, Moppert J, Thiel G, Rohr HP. Morphology and pathogenesis of glomerulopathy in cadaver kidney allografts treated with antilymphocyte globulin. Curr Top Pathol 1973; 57: 148.
  • 4
    Hamburger J, Crosnier J, Dormont J. Observations in patients with a well-tolerated homotransplanted kidney: Possibility of a new secondary disease. Ann N Y Acad Sci 1964; 120: 558577.
  • 5
    Starzl TE, Marchior TL, Faris TD et al. Problems and prognosis of chronically surviving patient after renal homotransplantation. Ann NY Acad Sci 1966; 129: 598614.
  • 6
    Hamburger J, Crosnier J, Dormont J. Experience with 45 renal homotransplantations in man. Lancet 1965; 17: 985992.
  • 7
    Porter KA, Dossetor JB, Marchioro TL et al. Human renal transplants: I. Glomerular changes Lab Invest 1967; 16: 153181.
  • 8
    Peterson EW, Mcphaul JJ, Mcintosh DA. Serial histologic alterations in human renal homotransplants. Am J Clin Pathol 1966; 45: 521532.
  • 9
    Hume DM, Lee HM, Williams GM et al. Comparative results of cadaver and related donor renal homografts in man and immunologic implications of outcome of second and paired transplants. Ann Surg 1966; 164: 352397.
  • 10
    Racusen LC, Solez K, Colvin RB et al. The Banff 97 working classification of renal allograft pathology. Kidney Int 1999; 55: 713723.
  • 11
    Maryniak RK, First MR, Weiss MA. Transplant glomerulopathy: Evolution of morphologically distinct changes. Kidney Int 1985; 27: 799806.
  • 12
    Andresdottir MB, Assmann KJM, Koene RAP, Wetzels JFM. Immunohistological and ultrastructural differences between recurrent type I membranoproliferative glomerulonephritis and chronic transplant glomerulopathy. Am J Kidney Dis 1998; 32: 582588.
  • 13
    Porter KA, Andres GA, Calder MW et al. Human Renal Transplants .2. Immunofluorescent and immunoferritin studies. Lab Invest 1968; 18: 159171.
  • 14
    Hsu HC, Suzuki Y, Churg J, Grishman E. Ultrastructure of transplant glomerulopathy. Histopathology 1980; 6: 351367.
  • 15
    Mihatsch MJ, Nickeleit V, Gudat F. Morphologic criteria of chronic renal allograft rejection. Transplant Proc 1999; 31: 12951297.
  • 16
    Mauiyyedi S, Pelle PD, Saidman S et al. Chronic humoral rejection: Identification of antibody-mediated chronic renal allograft rejection by C4d deposits in peritubular capillaries. J Am Soc Nephrol 2001; 12: 574582.
  • 17
    Regele H, Bohmig GA, Habicht A et al. Capillary deposition of complement split product C4d in renal allografts is associated with basement membrane injury in peritubular and glomerular capillaries: A contribution of humoral immunity to chronic allograft rejection. J Am Soc Nephrol 2002; 13: 23712380.
  • 18
    Mroz A, Durlik M, Cieciura T, Lao M. Complement split product C4 as the indicator of immunological activity in chronic allograft rejection. Pol Merkur Lekarski 2003; 15: 363365.
  • 19
    Vongwiwatana A, Gourishankar S, Campbell PM, Solez K, Halloran PF. Peritubular capillary changes and C4d deposits are associated with transplant glomerulopathy but not IgA nephropathy. Am J Transplant 2004; 4: 124129.
  • 20
    Sijpkens YW, Joosten SA, Wong MC et al. Immunologic risk factors and glomerular C4d deposits in chronic transplant glomerulopathy. Kidney Int 2004; 65: 24092418.
  • 21
    Hood B, Olander R, Nagy Z, Bergentz SE. Glomerulopathy in the transplanted kidney. Scand J Urol/Nephrol 1970; 4: 135142.
  • 22
    Palomar R, Lopez-Hoyos M, Pastor JM et al. Impact of HLA antibodies on transplant glomerulopathy. Transplant Proc 2005; 37: 38303832.
  • 23
    Monga G, Mazzucco G, Novara R, Reale L. Intertubular capillary changes in kidney allografts: An ultrastructural study in patients with transplant glomerulopathy. Ultrastruct Pathol 1990; 14: 201209.
  • 24
    Monga G, Mazzucco G, Messina M, Motta M, Wuaranta S, Novara R. Intertubular capillary changes in kidney allografts: A morphologic investigation on 61 renal specimens. Mod Pathol 1992; 5: 125130.
  • 25
    Ivanyi B, Fahmy H, Brown H, Szenohradszky P, Halloran PF, Solez K. Peritubular capillaries in chronic renal allograft rejection: A quantitative ultrastructural study. Human Pathol 2000; 31: 11291138.
  • 26
    Mauiyyedi S, Nelson C, Tolkoff-Rubin N, Cosimi AB, Schneeberger EE, Colvin RB. Peritubular capillary lamination. A marker of antibody mediated chronic rejection of renal allografts. Mod Pathol 2000; 13:176A.
  • 27
    Ivanyi B, Kemeny E, Szederkenyi E, Marofka F, Szenohradszky P. The value of electron microscopy in the diagnosis of chronic renal allograft rejection. Mod Pathol 2001; 14: 12001208.
  • 28
    Gough J, Yilmaz A, Miskulin D et al. Peritubular capillary basement membrane reduplication in allografts and native kidney disease: A clinicopathologic study of 278 consecutive renal specimens. Transplantation 2001; 71: 13901393.
  • 29
    Solez K, Colvin RB, Racusen LC et al. Banff ‘05 meeting report: Differential diagnosis of chronic allograft injury and elimination of chronic allograft nephropathy (“CAN”). Am J Transplant 2007; 7: 518526.
  • 30
    Racusen LC, Colvin RB, Solez K et al. Antibody-mediated rejection criteria - An addition to the Banff 97 classification of renal allograft rejection. Am J Transplant 2003; 3: 708714.
  • 31
    Suri DL, Tomlanovich SJ, Olson JL, Meyer TW. Transplant glomerulopathy as a cause of late graft loss. Am J Kidney Dis 2000; 35: 674680.
  • 32
    Habib R, Broyer M. Clinical significance of allograft glomerulopathy. Kidney Int Suppl 1993; 43: S95S98.
  • 33
    Collins AB, Schneeberger EE, Pascual MA et al. Complement activation in acute humoral renal allograft rejection: Diagnostic significance of C4d deposits in peritubular capillaries. J Am Soc Nephrol 1999; 10: 22082214.
  • 34
    Nankivell BJ, Borrows RJ, Fung CL, O'Connell PJ, Allen RD, Chapman JR. The natural history of chronic allograft nephropathy. N Engl J Med 2003; 349: 23262333.
  • 35
    Al Aly Z, Yalamanchili P, Cortese C, Salinas-Madrigal L, Bastani B. C4d peritubular capillary staining in chronic allograft nephropathy and transplant glomerulopathy: An uncommon finding. Transpl Int 2005; 18: 800805.
  • 36
    Drachenberg CB, Steinberger E, Hoehn-Saric E et al. Specificity of intertubular capillary changes: Comparative ultrastructural studies in renal allografts and native kidneys. Ultrastruct Pathol 1997; 21: 227233.
  • 37
    Smith RN, Kawai T, Boskovic S et al. Chronic antibody mediated rejection of renal allografts: Pathological, serological and immunologic features in nonhuman primates. Am J Transplant 2006; 6: 17901798.
  • 38
    Gloor JM, Cosio FG, Rea DJ et al. Histologic findings one year after positive crossmatch or ABO blood group incompatible living donor kidney transplantation. Am J Transplant 2006; 6: 18411847.
  • 39
    Opelz G. Non-HLA transplantation immunity revealed by lymphocytotoxic antibodies. Lancet 2005; 365: 15701576.
  • 40
    Nickeleit V, Zeiler M, Gudat F, Thiel G, Mihatsch MJ. Detection of the complement degradation product C4d in renal allografts: Diagnostic and therapeutic implications. J Am Soc Nephrol 2002; 13: 242251.
  • 41
    Akalin E, Dikman S, Murphy B, Bromberg JS, Hancock WW. Glomerular infiltration by CXCR3+ ICOS+ activated T cells in chronic allograft nephropathy with transplant glomerulopathy. Am J Transplant 2003; 3: 11161120.