Type I membranoproliferative glomerulonephritis in a renal allograft: A recurrence induced by a cytomegalovirus infection?

CASE REPORT

Type I Membranoproliferative Glomerulonephritis in a Renal Allograft: A Recurrence Induced by a Cytomegalovirus Infection? Margret B. Andresdottir, MD, Karel J.M. Assmann, MD, PhD, Luuk B. Hilbrands, MD, PhD, and Jack F.M. Wetzels, MD, PhD ● A 40-year-old white woman with end-stage renal disease from idiopathic type I membranoproliferative glomerulonephritis (MPGN) developed proteinuria and renal dysfunction 7 weeks after cadaveric donor renal transplantation. At the same time, a primary cytomegalovirus (CMV) infection was diagnosed. Complement levels were low. A renal biopsy disclosed an acute exudative proliferative glomerulonephritis with influx of polymorphonuclear granulocytes (PMNs), with granular deposits of C3, C1q, IgG, and IgM. The immunofluorescence (IF) and electron microscopy (EM) findings were compatible with an early stage of a type I MPGN. CMV could not be detected in the glomeruli nor elsewhere in the kidney by IF or EM. The patient was treated with ganciclovir. In a renal biopsy 3 weeks later, the exudative lesions had disappeared, and some glomeruli now showed the characteristic lesions of a type I MPGN with an increase of mesangial cells and matrix, and reduplication of the glomerular basement membrane. Over the following period, repeated biopsies were performed. The activity of the glomerular inflammation and immune complex deposits paralleled the waxing and waning of the CMV viral load. After 10.5 months, the graft was removed because of a life-threatening systemic fungal infection. At that time, the CMV infection had cleared, and in the transplantectomy material, the membranoproliferative pattern of injury had disappeared, and in the glomeruli hardly any deposits were found. These data strongly suggest that a primary CMV virus infection can induce an apparent recurrence of type I MPGN. 娀 2000 by the National Kidney Foundation, Inc. INDEX WORDS: Type I membranoproliferative glomerulonephritis (MPGN); renal transplantation; recurrence; cytomegalovirus (CMV) infection.

T

YPE I membranoproliferative glomerulonephritis (MPGN) is an immune complex disease, characterized by mesangial matrix expansion or cell proliferation, reduplication of the glomerular basement membrane (GBM), and granular deposits of, particularly, C3 in the capillary wall and mesangium.1 The source of the antigens involved in this glomerular disease remains unknown in most patients. However, a growing number of cases are attributed to viral infections such as hepatitis B and hepatitis C,2-4 and it is likely that, with newer diagnostic tests, more infectious agents will be discovered that can lead to this pattern of glomerular injury. After renal transplantation in patients with type I MPGN, up to half of the patients experience a recurrence of the original disease.5 Thus far, viruses have not been implicated in the pathogenesis of such recurrences. We describe a patient with type I MPGN, in whom an apparent recurrence developed early after transplantation. In this patient, there was a temporal relationship between the onset of the glomerulonephritis and the development of a primary cytomegalovirus (CMV) infection. Additional findings further supported a role for CMV in the induction of the MPGN in this patient.

METHODS For light microscopy (LM), a portion of the kidney biopsy was fixed in Bouin’s solution, dehydrated, and embedded in Paraplast (Amstelstad, Amsterdam, The Netherlands). Twomicron sections were stained with hematoxylin and eosin, periodic acid-Schiff, and silver methenamine. For immunofluorescence (IF), kidney fragments were snap-frozen in liquid nitrogen, and 2-µm cryostat sections were incubated with fluorescein-labeled antisera directed to human immunoglobulin (Ig) G, IgM, IgA, C1q, C3, and fibrinogen. Furthermore, antisera directed to DNA-binding protein p52 of the CMV were used (DAKO, Copenhagen, Denmark). For electron microscopy (EM), a small piece of the biopsy was fixed in 2.5% glutaraldehyde dissolved in 0.1 mol/L sodium cacodylate buffer, pH 7.2, for 4 hours at 4°C, and washed in

From the Department of Internal Medicine, Division of Nephrology, and Department of Pathology, University Hospital Nijmegen, The Netherlands. Received June 14, 1999; accepted in revised form October 8, 1999. M.B.A. is supported by a grant from the Dutch Kidney Foundation (C96.1543). Address reprint requests to Margret B. Andresdottir, MD, Department of Internal Medicine, Division of Nephrology, University Hospital Nijmegen, PO Box 9101, 6500 HB Nijmegen, The Netherlands. E-mail: J.Wetzels@nefro. azn.nl

娀 2000 by the National Kidney Foundation, Inc. 1523-6838/00/3502-0002$3.00/0

American Journal of Kidney Diseases, Vol 35, No 2 (February), 2000: E6

1

2

ANDRESDOTTIR ET AL

the same buffer. The tissue fragments were postfixed in phosphate-buffered 2% OsO4 for 2 hours, dehydrated, and embedded in Epon 812. Ultrathin sections were cut in an ultratome (LKB producteer, Bromma, Sweden) and stained with 5% uranyl acetate for 45 minutes and with lead citrate for 2 minutes at room temperature. The sections were examined using a JEOL 1200 EX2 electron microscope (JEOL, Tokyo, Japan).

CASE REPORT A 40-year-old white woman with end-stage renal disease secondary to biopsy-proven idiopathic type I MPGN received a renal transplant from a three-antigen mismatched postmortal donor in January 1998, 4 years after initiation of peritoneal dialysis. Pretransplantation assessment of the patient showed absence of antibodies against CMV, hepatitis B, hepatitis C, and human immunodeficiency virus, whereas antibodies against Epstein-Barr virus were present. The donor had a positive serology for CMV. The basic immunosuppressive treatment consisted of cyclosporine (6 mg/kg, tapered to 4 mg/kg at 3 months), mycophenolate mofetil (1 g twice daily), and prednisone (100 mg/d for the first 3 days, tapered to 5 mg/d at 7 weeks). There was immediate diuresis, and the early postoperative period was uneventful. When the patient was discharged at the seventh day after transplantation, the serum creatinine level was 158 µmol/L, and there was no proteinuria. At 5 weeks after transplantation, the patient developed fever, abdominal pain, and elevated aspartate aminotransferase, alanine aminotransferase, and lactate dehydrogenase (max, 72, 113, and 821 E/L, respectively). The serum creatinine level was 121 µmol/L, and urinary protein excretion was 1.1 g/10 mmol creatinine. A CMV pp65 antigen test was positive (⬎100 cells positive per 150,000 granulocytes), and CMV IgG was still negative. A diagnosis of a primary CMV infection was made, a treatment course with intravenous ganciclovir was given, and mycophenolate mofetil was stopped. The clinical course was complicated by a rise in serum creatinine (189 µmol/L) and proteinuria (1.9 g/10 mmol creatinine). Complement C3 and C4 levels were low (C3, 340 mg/L [normal, 750 to 1,250]; C4, 100 mg/L [normal, 180 to 400]), and a small amount (100 mg/L) of type III cryoglobulins was found. A renal biopsy was performed. On LM, findings were compatible

with a diffuse endocapillary glomerulonephritis with pronounced influx of polymorphonuclear granulocytes (PMNs), but without reduplication of the GBM (Fig 1A). By IF, prominent granular deposits of C3, C1q, IgG, and IgM were present in the glomerular capillary wall and in the mesangium (Fig 1B). Subendothelial and mesangial electron-dense deposits were observed by EM, but humps could not be detected (Fig 1C). The foot processes of the podocytes were fused, and the capillary lumina were invaded by polymorphonuclear cells (PMNs) and an occasional monocyte. CMV could not be demonstrated by IF or EM, neither in the glomeruli nor elsewhere in the kidney. Although a postinfectious glomerulonephritis related to a viral infection could not be ruled out completely, we favored at this moment an early and acute recurrence of the type I MPGN, based on the following reasons: the comparable IF findings of the original and transplant glomerulonephritis, the lack of humps (EM) and the absence of CMV in the deposits (IF or EM). The patient was treated with high doses of oral corticosteroids. Her serum creatinine transiently rose to 299 µmol/L but subsequently stabilized at 149 µmol/L. However, a persistent nephrotic syndrome developed. The complement levels gradually normalized, and the cryoglobulins disappeared. Ten weeks after transplantation, the serum creatinine level rose to 245 µmol/L, and proteinuria reached 14.2 g/10 mmol creatinine. The CMV antigen test was positive (⬎100 cells positive per 150,000 granulocytes), and the CMV IgG antibody titer remained as low as 10. A second transplant biopsy was performed, which showed a membranoproliferative pattern of injury in 3 of the 14 glomeruli present with segmental increase of mesangial cells and matrix, endocapillary proliferation, and reduplication of the GBM. All other glomeruli appeared normal. The massive PMN influx seen in the first biopsy had completely disappeared. Only in the three glomeruli with MPGN lesions could deposits of C3, C1q, IgG, and IgM be observed, but the amounts were considerably less than in the previous biopsy. No or hardly any immune deposits were present in the nonaffected glomeruli. Furthermore, fibrin thrombi were seen in one glomerulus and an afferent arteriole. By EM, fusion of the podocytic foot processes was still present, but the amount of electron-dense deposits was reduced. No PMNs were detectable in the capillary lumen, and no humps were observed on

= Fig 1. LM, IF (complement 3), and EM of the first (A, B, C) and third renal transplant biopsy specimen (D, E, F), and renal transplant specimen (G, H, I). The first renal transplant biopsy (A, B, C) shows a diffuse endocapillary, exudative (many PMNs) proliferative glomerulonephritis (A: methenamine silver staining; original magnification ⴛ275) with predominant C3 deposits in the periphery of the capillary loops (B; original magnification ⴛ275). Ultrastructurally (C), subendothelial electron dense deposits are present (arrow). There is fusion of the podocytic (P) foot processes. The capillary lumina are occluded by mononuclear or neutrophilic cells (original magnification ⴛ8,000). In a few glomeruli of the third biopsy of the renal transplant (D, E, F), there is an increase in mesangial cells and matrix and segmental reduplication of the GBM (D, arrow; methenamine silver staining; original magnification ⴛ275). The other glomeruli were quite normal. In the affected glomeruli, fine granular deposits of C3 along the capillary loops were present (E; original magnification ⴛ275). The amount of deposits were less than in the first biopsy specimen. The normal-looking glomeruli did not contain C3 deposits. By EM (F), reduplication of the GBM containing electron dense deposits was seen (original magnification ⴛ13,500). This lesion is reminiscent of a focal MPGN. Renal transplant specimen (G, H, I): A mild influx of neutrophils is present in all glomeruli. In none of the glomeruli were characteristics of an MPGN seen (G; original magnification ⴛ275). Hardly any C3 deposits (H) could be observed in the glomeruli (G). C3 was present in small arteries (A, original magnification ⴛ275). EM (I; original magnification ⴛ3,000) shows a normal glomerulus with intact podocytic architecture, no reduplication of the GBM, and no electron dense deposits. Abbreviation: P, podocyte.

CMV INFECTION AND GLOMERULONEPHRITIS

3

Fig 1.

4

ANDRESDOTTIR ET AL

the epithelial side of the GBM. The fibrin deposits were interpreted as a sign of cyclosporine nephrotoxicity, and, therefore, cyclosporine was replaced by azathioprine. Because of the CMV antigenemia, a second course of intravenous ganciclovir was given. As a result of this treatment, the kidney function improved, and the CMV antigenemia subsided. At 13 weeks posttransplantation, the patient was admitted severely ill, with headache, neck pain, nausea, and vomiting. An analysis of the cerebrospinal fluid showed cryptococcus meningitis, and treatment with amphotericin B in combination with flucytosine was started, followed by oral therapy with fluconazole. After initial recovery, the patient was readmitted at 4.5 months posttransplantation, because of fever, muscle pain, and neutropenia. Again, the CMV antigen test was positive (70 cells positive per 100,000 granulocytes), with no significant rise in the CMV IgG titer. The C3 and C4 levels were normal. Because of the persistent nephrotic syndrome (7.7 g/10 mmol creatinine; serum albumin, 28 g/L), a third renal transplant biopsy was performed. Compared with the earlier biopsy specimens, there was a mild progression of the glomerular pathology, with still a focal membranoproliferative pattern of injury with a mild influx of predominantly mononuclear cells, confirmed by EM (Fig 1D, F). By IF, there was a pronounced deposition of C3, along with small amounts of IgM (Fig 1E). Treatment with oral ganciclovir was given once more. A week after the therapy was started, a rise in CMV IgG antibody titer with disappearance of the antigenemia was noted, and the ganciclovir treatment was stopped. Seven months posttransplantation, a fourth transplant biopsy was performed because of persistent nephrotic syndrome. At that time, the patient had a CMV IgG antibody titer of 94 E/mL. This biopsy specimen showed only minor signs of a membranoproliferative pattern of injury, less severe than in the previous biopsy and accompanied by a few deposits of C3 and IgM. CMV was still not

detectable in the kidney. The foot processes as seen by EM were only segmentally fused. Electron-dense deposits could not be found along the capillary wall or in the mesangium. The clinical course thereafter was complicated by continuous ambulatory peritoneal dialysis peritonitis and the development of an intra-abdominal infection with aspergillus fumigatus. Because of this life-threatening situation, a transplantectomy was performed 10.5 months posttransplantation, and the immunosuppressive therapy was stopped. The transplantectomy specimen showed only a mild influx of PMN, with no membranoproliferative characteristics as seen in the previous biopsy specimens. Especially reduplication of the GBM was completely absent (Fig 1G). Only minor deposits of IgM and C3 could be detected by IF (Fig 1H). The foot processes of the podocytes were intact, and the GBM had a normal appearance (Fig 1I). Only a few glomeruli showed sclerosis, hyalinosis, and adhesions of the glomerular capsule to Bowman’s capsule compatible with secondary focal glomerulosclerosis. The morphological findings are summarized in Table 1.

DISCUSSION

We describe a renal transplant patient who developed an acute endocapillary proliferative glomerulonephritis that evolved into a focal MPGN suggestive of a recurrent disease. Remarkably, the glomerulonephritis finally recovered almost completely. In this patient, the onset of the glomerulonephritis concurred with the development of a primary CMV infection. Follow-up biopsies have allowed us to study the morphological evolution of the glomerular abnormalities in

Table 1. Biopsy Findings in a Patient With Type I MPGN, Before and After Transplantation IF Time of Biopsy

I Native kidney II Transplant bx 7 wk after tx

10 wk after tx

LM

IgG

IgM

IgA

C1q

C3

Fibrinogen

CMV

Focal type I MPGN (crescents 60-70%)

1⫹

2⫹

⫺

2⫹

3⫹

2⫹

ND

Diffuse endocapillary exudative proliferative GN Focal MPGN (mild)

1-2⫹

1⫹

⫺

2⫹

3⫹

2⫹

⫺

1⫹/⫺

⫾/⫺

⫺

⫾/⫺

2⫹/⫺

⫾/⫺

⫺

22 wk after tx

Focal MPGN (moderate)

⫺

1⫹

⫺

⫺

2⫹

⫾

⫺

28 wk after tx

Focal MPGN (mild)

⫺

⫾

⫺

⫺

⫾/⫺

⫺

⫺

Mild diffuse endocapillary exudative proliferative GN and FGS

⫺

⫾

⫺

⫺

⫾

⫺

⫺

III Transplantectomy 10.5 mo

EM

Subendothelial deposits Subendothelial deposits in all glomeruli Subendothelial deposits in a few glomeruli Subendothelial deposits in a few glomeruli Minor subendothelial deposits No deposits

Abbreviations: ND, not done; Tx, transplantation; GN, glomerulonephritis; FGS, focal segmental glomerulosclerosis.

CMV INFECTION AND GLOMERULONEPHRITIS

relation to clinical parameters. Although the patient had type I MGPN as the original renal disease, the data strongly suggest that the development of the glomerulonephritis posttransplantation was related to the CMV infection rather than a mere ‘‘idiopathic’’ recurrence of the original disease. First, the glomerulonephritis developed early (within 7 weeks) after transplantation, with heavy proteinuria. In contrast, in our series of 12 patients with recurrent type I MPGN, the recurrence occurred at a median of 20 months after cadaveric transplantation, and proteinuria at onset was mild and only gradually increased. Second, we have found that the likelihood of recurrent type I MPGN was dependent on the HLA genotype, recipients with the B8 DR3 haplotype being at greatest risk.5 Our patient was B8 DR3 negative. Furthermore, the evolution of the glomerular lesions with disappearance of the immune deposits argues against a mere recurrent type I MPGN. We are unaware that a recurrent type I MPGN can resolve spontaneously after renal transplantation. In our series, all patients with recurrent type I MPGN followed a progressive course toward end-stage renal disease. In one patient, we had follow-up biopsy material that documented the progression from immune complex deposition without LM changes to the full-blown picture of type I MPGN, 2 years later. Schu¨rch et al6 also have documented the course of recurrent MPGN in one patient. In their patient, proteinuria developed 6 months after transplantation. The renal biopsy specimen showed, analogous to our patient, immune complex deposition in the mesangium and, to a lesser extent, in the subendothelial space, without glomerular cell proliferation. Almost a year later, there was a progression of the immune complex deposition, and cell proliferation was seen. Several data suggest that in our patient, the CMV infection triggered the development of the glomerulonephritis. First, there was the close temporal relationship. The glomerulonephritis was diagnosed within 2 weeks after the onset of the CMV antigenemia. At this time, complement levels were low and cryoglobulins were present. The renal biopsy specimen disclosed an acute exudative glomerulonephritis with influx of mainly PMNs and a granular deposition of immune complexes in the mesangium and the sub-

5

endothelial space.This is a well-known scenario for the development of an immune complex glomerulonephritis. The fact that the CMV itself was not found in the glomerulus does not exclude its participation in the pathogenesis of the glomerulonephritis. Likewise, hepatitis C virus has never been found in the glomerulus of patients with type I MPGN, although there is compelling evidence that the hepatitis C virus can trigger such a glomerulonephritis.3,7-9 Viral antigens may not necessarily be involved, because the direct cytopathogenetic effect of the viruses can result in a release of sequestered tissue antigens into the circulation with subsequent host antibody formation. In that case, the circulating immune complexes would not include the virus itself or antiviral antibodies. Our patient had long-lasting, episodic antigenemia, which increases the chance of developing an overt immune complex glomerular disease.10 Another argument for a role of the CMV in the genesis of this patient’s glomerular disease was that the activity of the glomerular lesions paralleled the clinical and serological activity of the CMV infection. In the end, after disappearance of the antigenemia and the appearance of CMV IgG antibodies, histology of the transplantectomy specimen showed a complete resolution of the membranoproliferative pattern of glomerular injury and the disappearance of immune deposits in the glomeruli. Only a mild influx of PMNs was seen. This influx of PMNs could on morphological or clinical grounds not be related to the initial glomerulonephritis but might be a consequence of the systemic fungal infection with accompanying leukocytosis. We are aware of one case report suggesting a relation between a CMV infection and recurrent type I MPGN. This patient, reported by Birk and Chavers,11 developed a recurrent type I MPGN at 3 months after transplantation. This patient had a primary CMV infection, and in the renal biopsy specimen, intracytoplasmic CMV inclusions were seen by EM in the glomerular cells. These CMV inclusions were markedly reduced in a follow-up biopsy after treatment with ganciclovir therapy and discontinuation of azathioprine. No information is given on the evolution of the glomerular histological abnormalities, and the authors do not address the relationship between the recurrent MPGN and the CMV inclusions. The fact

6

that the patient has retained normal graft function for 6 years could indicate a regression of the glomerulonephritis. It is remarkable that our, as well as Birk’s, case of CMV infection in relation to the development of type I MPGN after transplantation occurred in patients with type I MPGN as their original disease. Therefore, it is possible that CMV will only trigger type I MPGN after transplantation in susceptible patients. Admittedly, we cannot exclude that the glomerulonephritis in our patient was caused by CMV independently from the original renal disease. In this respect, the evolution of the histological pattern and the clinical picture would fit with a postinfectious glomerulonephritis, going through an acute exudative stage with influx of inflammatory cells, through a chronic stage with reduplication of the basement membranes, and eventually to a resolution of the glomerular abnormalities. However, the IF findings of especially the first transplant biopsy lacked some of the characteristics of postinfectious glomerulonephritis. By IF, the deposits were positive for C3, IgG, IgM, and C1q, whereas the deposits of postinfectious glomerulonephritis usually contain C3 with minor IgG but lack C1q. On EM, the deposits were restricted to the subendothelial space and mesangial area. The characteristic subepithelial humps seen in most patients with poststreptococcal glomerulonephritis were never observed. CMV infection has been implicated in the pathogenesis of glomerulonephritis/glomerulopathy after renal transplantation. Richardson et al12 reported a distinct glomerular lesion in four renal transplant patients characterized by distended, occluded capillary loops with endothelial cell swelling or necrosis and accumulation of a few mononuclear cells, but not PMNs, and fibrillary material in glomerular capillaries. There was a slight increase in mesangial matrix and cells, and the IF showed granular deposits of IgM and C3 and, to a lesser extent, IgG in the glomerular capillary wall. The authors suggested that the glomerular lesion was associated with CMV, because the lesion was only seen in patients with CMV viremia. Also, the IF staining for CMV was positive in the glomerular capillary wall in one patient. Furthermore, increased immunosuppression was associated with worsening of renal function, whereas decreased immunosuppression appeared to be associated with improve-

ANDRESDOTTIR ET AL

ment. Others have, however questioned the association between CMV infection and this lesion, because similar glomerular findings occur in the absence of evidence of CMV infection.13-15 Detwiler et al16 described a patient with a de novo necrotizing glomerulonephritis that resolved after treatment of CMV. Furthermore, CMV inclusions were found in the glomerular endothelial cells of this patient, and these inclusions likewise disappeared. Because of the absence of intraglomerular immune complexes, the most likely pathophysiological mechanism was thought to be CMV-induced direct injury to glomerular cells. Finally, Rao et al17 described a patient with a de novo immunotactoid glomerulonephritis, also resolving on treatment of the CMV infection. In this patient, no CMV was found in the glomerulus. Thus, the glomerular lesions described with the involvement of CMV are quite diverse. In conclusion, our data strongly suggest that CMV infection can induce type I MPGN, mimicking a recurrence of the original renal disease. Our patient shows that the glomerular lesions can heal if the infectious agent is controlled. It is important to realize that in such patients, the treatment of the immune complex glomerulonephritis does not depend on broad-based immunosuppression but rather on effective antiviral therapy. ACKNOWLEDGMENT We acknowledge H. Dijkman for his skillful help in performing the histology and electron microscopy and in preparing the morphological pictures.

REFERENCES 1. Glassock RJ, Cohen AH, Adler SG: Primary glomerular diseases, in Brenner BM (ed.): Brenner and Rector’s The Kidney. Philadelphia, PA, Saunders, 1996, pp 1392-1497 2. Johnson RJ, Couser WG: Hepatitis B infection and renal disease: Clinical, immunopathogenetic and therapeutic considerations. Kidney Int 37:663-676, 1990 3. Johnson RJ, Gretch DR, Yamabe H, Hart J, Bacchi C, Hartwell P, Couser WG, Corey L, Wener MH, Alpers CE, Willson R: Membranoproliferative glomerulonephritis associated with hepatitis C virus infection. N Engl J Med 328:465-470, 1993 4. Morales JM, Campistol JM, Andres A, Rodicio JL: Glomerular diseases in patients with hepatitis C virus infection after renal transplantation. Curr Opin Nephrol Hypertens 6:511-515, 1997 5. Andresdottir MB, Assmann KJM, Hoitsma AJ, Koene RAP, Wetzels JFM: Recurrence of type I membranoprolifer-

CMV INFECTION AND GLOMERULONEPHRITIS

ative glomerulonephritis after renal transplantation. Transplantation 63:1-5, 1997 6. Schurch W, Leski M, Hinglais N: Evolution of recurrent lobular glomerulonephritis in a human kidney allotransplant. Virchows Arch A 355:66-84, 1972 7. Roth D, Cirocco R, Zucker K, Ruiz P, Viciana A, Burke A, Carreno M, Esquenazi V, Miller J: De novo membranoproliferative glomerulonephritis in hepatitis C virus-infected renal allograft recipients. Transplantation 59: 1676-1682, 1995 8. Cruzado JM, Gil-Vernet S, Ercilla G, Seron D, Carrera M, Bas J, Torras J, Alsina J, Grinyo JM: Hepatitis C virus-associated membranoproliferative glomerulonephritis in renal allografts. J Am Soc Nephrol 7:2469-2475, 1996 9. Gallay BJ, Alpers CE, Davis CL, Schultz MF, Johnson RJ: Glomerulonephritis in renal allografts associated with hepatitis C infection: A possible relationship with transplant glomerulopathy in two cases. Am J Kidney Dis 26:662-667, 1995 10. Glassock RJ: Immune complex-induced glomerular injury in viral diseases: An overview. Kidney Int 40:S5-S7, 1991 (suppl 35) 11. Birk PE, Chavers BM: Does cytomegalovirus cause glomerular injury in renal allograft recipients? J Am Soc Nephrol 8:1801-1808, 1997

7

12. Richardson WP, Colvin RB, Cheeseman SH, TolkoffRubin NE, Herrin JT, Cosimi AB, Collins AB, Hirsch MS, McCluskey RT, Russel PS, Rubin RH: Glomerulopathy associated with cytomegalovirus viremia in renal allografts. N Engl J Med 305:57-63, 1981 13. Herrera GA, Alexander RW, Cooley CF, Luke RG, Kelly DR, Curtis JJ, Gockerman JP: Cytomegalovirus glomerulopathy: A controversial lesion. Kidney Int 29:725-733, 1986 14. Spencer ES, Jepsen FL: Glomerulopathy in renal allografts from patients with and without active cytomegalovirus infection. Danish Med Bull 36:300-301, 1989 15. Olsen S, Spencer E, Cockfield S, Marcussen N, Solez K: Endocapillary glomerulitis in the renal allograft. Transplantation 59:1421-1425, 1995 16. Detwiler RK, Singh HK, Bolin P, Jennette JC: Cytomegalovirus-induced necrotizing and crescentic glomerulonephritis in a renal transplant patient. Am J Kidney Dis 32:820-824, 1998 17. Rao KV, Hafner GP, Crary GS, Anderson WR, Crosson JT: De novo immunotactoid glomerulopathy of the renal allograft: Possible association with cytomegalovirus infection. Am J Kidney Dis 24:97-103, 1994