Context.—Recurrent and de novo glomerulonephritis is an important cause of renal allograft failure, but estimates of its prevalence vary widely. One reason for such variability is the inconsistency with which electron microscopy and immunofluorescence are used in assessing renal allograft biopsies.
Objective.—To determine the prevalence of immune-complex deposits in all renal allograft biopsies performed during a 1-year period and to correlate their presence with clinical data.
Design.—Our center accessioned a total of 118 renal allograft biopsies during 1 year from 88 patients. All biopsies were examined by both electron microscopy and immunofluorescence in addition to conventional light microscopy. Patient and donor characteristics were obtained as well as follow-up data for a minimum of 26 months after the index biopsy.
Results.—Eight cases of immunoglobulin (Ig) A nephropathy were found (recurrent in 7 and de novo in 1). There were 9 instances of what we designate “IgM-positive immune deposits” without specific features of a recognized glomerulonephritis. To the best of our knowledge, the latter has not hitherto been described and may be part of a heterogeneous group of glomerulopathies. Other unexpected findings included de novo fibrillary glomerulonephritis and de novo membranous glomerulonephritis, the latter occurring at 3 months after engraftment.
Conclusions.—A high proportion (19.5%) of unselected renal allograft biopsies show immune-complex deposits both with and without a recognized glomerulopathy. These require both electron microscopy and immunofluorescence for detection. IgM-positive deposits of uncertain etiology are relatively frequent.
Immune-complex–mediated glomerulonephritis is considered 1 of the 3 leading causes of renal allograft loss (after chronic rejection and death with a functioning graft).1 Reports of its prevalence vary from 6% to 19.4%.1 This variation may be due to the restricted populations studied (eg, those with recurrent glomerulonephritis only or with proteinuria only1,2) or to the inclusion of patients with metabolic glomerulopathies.3 Incomplete graft biopsy examination has not been emphasized as a cause of such disparities, and the aim of our present study was to determine the prevalence and significance of graft glomerulonephritis as detected in unselected biopsies submitted for electron microscopy (EM) and immunofluorescence (IF).
MATERIALS AND METHODS
One hundred eighteen consecutive renal allograft biopsies from 88 patients, performed during a 1-year period (July 1999 to July 2000) at a single renal transplant center, were processed in a standard fashion and examined by light microscopy as well as by EM and IF. Glomerular electron-dense deposits, in nonsclerotic glomeruli without hyalinosis and with colocalized positive immunofluorescence for immunoglobulin, were designated “immune-complex (IC) deposits.” Patients included had both recent (3 months earlier) and long-term (up to 228 months earlier) transplants. All relevant data on donors and recipients were gathered at the time of biopsy and after follow-up for a minimum of 26 months. The indication for biopsy in most cases was a rapidly rising creatinine level, with suspected acute rejection or drug toxicity. In 17 instances, the biopsy was done for surveillance purposes in patients with stable grafts.
Twenty-three allograft biopsies from 23 patients showed glomerular IC deposits. These included 7 cases of recurrent immunoglobulin (Ig) A nephropathy (IgAN), 1 case of de novo IgAN, 2 cases of recurrent membranoproliferative glomerulonephritis (1 case each of type I and type II membranoproliferative glomerulonephritis), and 1 case each of de novo membranous glomerulonephritis and of de novo fibrillary glomerulonephritis. Two patients showed IgG mesangial deposits with weak C1q positivity (both had lupus nephritis in the native kidney). In addition, 9 patients showed IgM-positive mesangial deposits, with or without C3 positivity. These 9 represented a heterogeneous group, including 1 patient with recurrent focal segmental glomerulosclerosis, 1 with IgAN in the native kidney (but IgA was negative in the graft biopsy), and 3 patients with current (hepatitis B and C) or subsequent (BK and cytomegalovirus) viral infection. There were no cases of crescentic glomerulonephritis. There was no significant difference between the IC and the non-IC group with respect to the following: number of rejection episodes, age, HLA matching, and cadaveric versus living (related and unrelated) donor. Female sex and previous transplantation were slightly more common in the IC group, but this did not reach the level of statistical significance. After a minimum follow-up period of an additional 26 months, the mean serum creatinine level for the IC group was significantly higher than in the non-IC group (P < .001, using analysis of variance). There were 2 graft losses in the “IgM” group, 1 from BK nephritis and 1 from chronic allograft nephropathy. One other patient died from disseminated cytomegalovirus with a functioning graft. In the non-IC group, there were 4 deaths with functioning grafts and 3 graft failures due to rejection.
In the largest study of recurrent glomerulonephritis to date, EM was not used in approximately 50% of graft biopsies, and an unspecified number of cases did not have IF performed.1 This may have led to underestimation of both recurrent and de novo glomerulonephritis, such as recurrent IgAN. Without EM and IF, making the distinction between membranoproliferative glomerulonephritis and transplant glomerulopathy is very difficult. The utilization rates of IF and EM are particularly low in early graft biopsies, so the timing of onset of glomerulonephritis in allografts is also hard to estimate. In our biopsy-based study, on the other hand, all graft biopsies were submitted for light microscopy, EM, and IF.
Our results confirm the high rate of recurrent IgAN and show its early onset (4/8 patients were biopsied at less than 12 months after transplantation). In addition, we found 1 case each of de novo fibrillary glomerulonephritis and membranous glomerulonephritis, the latter occurring at 3 months after transplantation.
There were also 9 patients who had IgM-positive mesangial electron-dense immune deposits, in some cases with C3 positivity also (Table). On EM, the size of these deposits varied from small to very large (Figure) and resembled the deposits seen in IgAN. Only 1 of these cases resembled a recognized clinicopathologic entity (ie, recurrent focal segmental glomerulosclerosis, which was diagnosed in a second biopsy). The presence or subsequent development of viral infection (BK, cytomegalovirus, and hepatitis C and B) in 3 of the IgM group raises the possibility of antiviral antibody/viral antigen immune-complex deposits. Three biopsies from the IgM group showed only mesangial expansion by light microscopy, and the remainder showed no diagnostic light microscopic feature. Seven of the 9 occurred less than 1 year after transplantation. Four showed proteinuria greater than 500 mg/L/ d. The etiology of the transplant-associated IgM deposits is not clear. Possibilities include T-cell deficiency or mesangial overloading by antigen-antibody complexes (the antibody possibly directed against an unidentified microorganism). Rejection-related immune events are another possibility. However, we found no case of transplant glomerulopathy or of “glomerulitis” (which can be associated with acute humoral rejection) with glomerular immune deposits. We are now studying all our cases of IgM-positive glomerular IC deposits in allografts over a 10-year period and its association with graft function, infection, and recurrent disease (particularly IgAN).
In summary, the results of this comprehensive cross-sectional analysis of renal allograft biopsies suggests a high prevalence of immune-complex deposition (19.5% of all biopsies and 26% of all patients biopsied), often occurring early. We also report on a group of IgM-positive glomerular immune-complex deposits that, to our knowledge, has not been previously described.
The authors have no relevant financial interest in the products or companies described in this article.
Reprints: James Gough, MB, University of Calgary Department of Pathology, Foothills Medical Centre, 1403 29 St NW, Calgary, Alberta, Canada T2N 2T9 (email@example.com)