A case of antiglomerular basement membrane glomerulonephritis complicated by membranous nephropathy

Abstract

The sequential or simultaneous presentation of anti-glomerular basement membrane (anti-GBM) glomerulonephritis with membranous nephropathy (MN) has been infrequently reported. Although the mechanism underlying MN superimposed on anti-GBM glomerulonephritis is unknown, the two entities are believed to be interrelated. We report the case of a 75-year-old woman diagnosed with rapidly progressive glomerulonephritis. Renal biopsy revealed crescentic glomerulonephritis with linear and granular staining of immunofluorescent IgG1 and IgG4 granular staining on the capillary loops. Electron microscopy revealed extensive subepithelial deposits. These findings suggested simultaneous development of anti-GBM glomerulonephritis and MN in this case. Serum phospholipase A2 receptor (PLA2R) antibody was negative. The patient was treated with prednisolone and plasma exchange, resulting in resolution of renal insufficiency and a decrease in urinary protein. The rapid decrease in urinary protein and absence of PLA2R antibody suggest that the mechanism of MN associated with anti-GBM glomerulonephritis differs from that of primary MN.

Introduction

Anti-glomerular basement membrane (anti-GBM) glomerulonephritis, a rare disease in Japan, is one of the causes of crescentic glomerulonephritis. This condition is known as Goodpasture’s syndrome when accompanied by lung lesions. Anti-GBM glomerulonephritis is induced by antibodies directed against the alpha-3 subunit of type IV collagen, a normal structural component of GBM. Immunofluorescence staining reveals linear staining of IgG in the capillary wall, predominantly subclass IgG1 [1].

Membranous nephropathy (MN) is a common cause of nephrotic syndrome. Immunofluorescence staining shows granular deposition of IgG in the capillary wall. In primary MN, the subclass of IgG is predominantly IgG4, while in secondary MN, staining for IgG1 and IgG2 is more strongly positive in addition to IgG4 staining [2]. The phospholipase A2 receptor (PLA2R) on the surface of podocytes is thought to be involved in the mechanism underlying primary MN. This receptor acts as an autoantigen serving as a target for circulating antibodies and leading to in situ immune complex formation [3].

Although isolated anti-GBM glomerulonephritis is a rare disease, many cases of anti-GBM glomerulonephritis with MN have been reported. This suggests that anti-GBM glomerulonephritis and MN are interrelated each other and that their coexistence is not a chance event.

Case report

A 75-year-old woman was referred to a local hospital with complaints of macrohematuria, fever, generalized fatigue, and appetite loss. Ten years before, she had been diagnosed with hypertension and treated with benidipine. She had shown no evidence of kidney dysfunction or other abnormality on urinalysis at a general health examination performed 5 months earlier. On initial presentation at the hospital, she had renal insufficiency, abnormal results on urinalysis, and an anti-GBM antibody serum level of 50 EU (ELISA, normal <10 EU). Rapidly progressive glomerulonephritis caused by anti-GBM glomerulonephritis was suspected, and she was referred to our hospital. At admission, her body temperature was 38.0 °C. There were no infectious signs on physical examination. Laboratory tests showed that the level of hemoglobin was 8.2 g/dl; white blood cell counts, 9200/μl; C-reactive protein (CRP), 7.89 mg/dl; creatinine, 1.77 mg/dl; creatinine clearance, 53.9 ml/min/1.73 m², urinary protein excretion, 0.93 g/day; and urinary sedimentation, >100 red blood cells/high-power field with casts. Serological examination revealed an anti-GBM antibody level of 73 EU. Myeloperoxidase-antineutrophil cytoplasmic antibody (ANCA), proteinase 3 ANCA, and antinuclear antibody were negative. Chest radiography and computed tomography showed normal lungs, and renal ultrasonography showed normal kidney size and shape. Renal biopsy was performed. Light microscopic findings showed a total of 24 glomeruli, 3 of which had global sclerosis; 8 had cellular crescents; 4 had fibrocellular crescents; 1 had fibrous crescents (Fig. 1). Obvious membranous change or mesangial proliferation was not observed in the light microscopy. Severe tubulointerstitial damage was detected, but there was no evidence of vasculitis. Immunofluorescence staining showed diffuse linear deposition of IgG and segmental linear deposition of C3 in the capillary loops. IgA, IgM, C4, C1q, and fibrinogen stainings were negative (Fig. 2). IgG subclass analysis showed linear and granular staining for IgG1, strong coarse granular staining for IgG4, and weak granular staining for IgG2 and IgG3 (Fig. 3). In electron microscopy, extensive subepithelial deposits and foot process effacement as well as cellular crescents and fibrin exudation in Bowman’s space were observed (Fig. 4). These findings were consistent with a diagnosis of anti-GBM glomerulonephritis with stage II MN, although serum PLA2R antibody was absent. In order to distinguish secondary MN, we additionally examined the esophagogastroduodenoscopy, immunoassay fetal occult blood test, hepatitis B and C as well as HIV tests, and antinuclear antibodies. The results of these examinations were all negative. There was no evidence of malignancy, infection, collagen disease, or drugs that induce secondary MN. The patient was treated with 40 mg/day of oral prednisolone (1 mg/kg/day) and two plasma exchanges. Renal function improved, with the serum creatinine level reaching 1.10 mg/dl with a decrease in CRP. The level of anti-GBM antibody rapidly decreased after plasma exchange and eventually disappeared. Proteinuria was alleviated, with urinary protein measuring 0.10 g/day after 6 weeks of treatment. The prednisolone dose was tapered to 30 mg/day at 4 weeks later, and the patient was discharged. At 5 months after discharge, she had no evidence of renal failure and no proteinuria or CRP (Fig. 5).

Fig. 1.

Light microscopy showing cellular and fibrocellular crescents with severe interstitial damage (periodic acid-Schiff staining ×200)

Fig. 2.

Immunofluorescence staining showing global linear deposition of IgG and C3 on capillary loops

Fig. 3.

IgG subclass analysis showing strong linear and granular staining of IgG1 (left small square shows higher magnification of IgG1), strong granular staining of IgG4 (right small square shows higher magnification of IgG4), weak granular staining of IgG2, and faint granular staining of IgG3 on capillary loops

Fig. 4.

Electron microscopy showing electron-dense subepithelial deposits in the glomerular basement mambrane, which indicate stage II membranous nephropathy. Global swelling of endothelial cells and extensive foot process effacement are shown

Fig. 5.

Clinical course. PSL predonisolone, UP urinary protein, Cr serum creatinine, CRP serum C-related protein

Discussion

In the previous literature, 30 reported cases of anti-GBM glomerulonephritis associated with MN were identified. In 7 cases, anti-GBM glomerulonephritis followed MN [4–10]; in 5 cases, MN followed anti-GBM glomerulonephritis [11–14]; in 18 cases, anti-GBM glomerulonephritis and MN developed simultaneously (Table 1) [1, 5, 12, 15–26]. A biphasic mechanism has been proposed to explain the pathophysiology of MN following anti-GBM glomerulonephritis and cases of simultaneous disease [1]. In the first phase, linear deposition of IgG, resulting from antibody binding to fixed structural antigens of the glomerular capillary wall, promotes upregulation of antigenic basement membrane components that are synthesized and secreted by podocytes. In the second phase, a multispecific antibody reacts with these basement membrane components, forming an immune complex in situ in the subepithelial space.

Table 1.

Anti-GBM glomerulonephritis and membranous nephropathy cases reported in previous studies

Reference	Age/sex	Type	Therapy	Outcome
4	49/M	MN → GBM	Steroids	Died
5	53/M	MN → GBM	Steroids, azathioprine	Hemodialysis
6	65/M	MN → GBM	Steroids, CP, PP	Died
7	54/M	MN → GBM	Steroids, PP	ESRD, transplant
8	57/?	MN → GBM	Steroids, CP, PP	Dialysis
9	50/F	MN → GBM	Steroids, CP, PP	Dialysis
10	50/M	MN → GBM	Steroids, CP, PP	Dialysis
11	22/F	GBM → MN	None	Recovery
12	20/M	GBM → MN	Cytotoxic drugs, PP	Recovery
12	18/F	GBM → MN	Steroids, CP, PP	Renal failure
13	20/M	GBM → MN	Steroids, CP	Recovery, proteinuria
14	17/M	GBM → MN	Antibiotics	Recovery
5	44/M	Simultaneous	Steroids	Dialysis
12	16/F	Simultaneous	Steroids, CP, PP	Recovery
12	47/M	Simultaneous		Dialysis
12	21/M	Simultaneous	Steroids, CP, PP	Recovery
12	20/F	Simultaneous	Steroids, CP, PP	Recovery
15	19/F	Simultaneous	Steroids, azathioprine	Dialysis
16	25/M	Simultaneous	Steroids, CP, PP	Died
17	17/M	Simultaneous	Steroids, CP	Recovery
18	22/F	Simultaneous
19	20/M	Simultaneous	Steroids, PP	Recovery
20	50/M	Simultaneous	Steroids, CP, PP	Recovery
21	54/F	Simultaneous	Steriods, PP	Dialysis
22	53/M	Simultaneous	Steroids, CP, PP	Dialysis
23	48/F	Simultaneous	Steroids, CP, PP	Dialysis
24	71/F	Simultaneous	Steroids	Renal insufficiency
25	49/M	Simultaneous	Steroids, CP, PP	Dialysis
26	59/M	Simultaneous	Steroids, CP, PP	Dialysis
27	59/M	Simultaneous	Steroids, CP, PP	Recovery
Our case	75/F	Simultaneous	Steroids, PP	Recovery

MN → GBM anti-GBM glomerulonephritis followed membranous nephropathy (MN), GBM → MN MN followed anti-GBM glomerulonephritis; simultaneous, simultaneous presentation of anti-GBM glomerulonephritis and MN, M male, F female, CP cyclophosphamide, PE plasma exchange

In our case, the absence of proteinuria before the onset of renal insufficiency was consistent with simultaneous onset of anti-GBM glomerulonephritis and MN. Indeed, the stage II MN might indicate that the onset of MN was earlier than the time of her clinical symptoms. Moreover, MN occasionally shows no abnormality on urinalysis. Therefore, our patient may have had anti-GBM glomerulonephritis following MN. However, because the kidney biopsy showed fibrous crescents, the onset of anti-GBM glomerulonephritis would have been before the onset of clinical symptoms; then there would be no discrepancy between the onset of anti-GBM glomerulonephritis and MN. Moreover, we usually think that the clinical course indicates the simultaneous onset of anti-GBM glomerulonephritis and MN.

Linear deposition of IgG1 is thought to indicate deposition of anti-GBM antibody, whereas the predominant granular IgG4 staining suggests deposition of in situ immune complexes. Hoshino et al. [23] reported that in patients with simultaneous anti-GBM glomerulonephritis and MN, the initial biopsy revealed linear deposition of IgG1 and granular deposition of IgG4, but the granular IgG4 deposits were not observed on repeat biopsy after remission.

Previous studies show that the outcome of MN following anti-GBM glomerulonephritis is generally favorable. Renal outcome is usually poor in simultaneous disease; a few patients fortunately recover renal function in simultaneous disease but tend to have an absence of urinary protein after treatment (Table 1). These clinical findings support the biphasic mechanism of anti-GBM glomerulonephritis and MN. In our patient, treatment resulted in complete remission, in contrast to the typical outcome in primary MN. Proteinuria resolved promptly after prednisolone therapy and plasma exchanges, concurrent with the disappearance of serum anti-GBM antibodies. We speculate that in the present case the damage by GBM was improved by the complete and early disappearance of anti-GBM antibodies and suppression of the antigenic membranous component production.

The biphasic mechanism of MN occurring before or concurrently with anti-GBM glomerulonephritis may be a different pathophysiology than that of primary MN.

In a report of 7 cases of anti-GBM glomerulonephritis following MN, it was hypothesized that cryptic type IV collagen epitopes that are immunogenic GBM antigens are more easily exposed in collagen hexamers lacking structural reinforcement, as expected in newly synthesized/remodeled GBM in the setting of MN [27]. All 7 patients progressed to end-stage renal failure and were treated by hemodialysis, suggesting that the anti-GBM antibody caused severe and irreversible damage to GBM with exposed type IV collagen epitopes in this type of glomerulonephritis.

The predominant granular IgG4 staining in our case is consistent with a diagnosis of primary MN [2]. However, the absence of serum PLA2R antibody, which is generally present in primary MN, may account for the observed difference in clinical course between MN associated with anti-GBM glomerulonephritis and primary MN.

In summary, we present a case of a patient with simultaneous occurrence of anti-GBM glomerulonephritis and MN whose renal function and proteinuria improved with treatment. Although the IgG4 subclass predominance resembled that of primary MN, the absence of PLA2R antibody and the better clinical outcome suggest that a substantial difference in pathophysiology exists between MN with anti-GBM glomerulonephritis and primary MN.