Clinicopathological study of originally non-lupus “full-house” nephropathy

Abstract

Background: Glomerular “full-house” immunofluorescence staining commonly indicates lupus nephritis. However, some non-lupus nephropathy also can present with a “full-house” immunofluorescence pattern mimicking lupus nephritis. The goal of this study is to define the clinicopathological spectrum of originally non-lupus “full-house” nephropathy. Methods: Records of 24 patients with “full-house” nephropathy in the absence of clinical or serological evidence of systemic lupus erythematosus (SLE) at the time of renal biopsy were abstracted for demographics, clinical presentation, laboratory data, renal biopsy findings, and clinical follow-up. Results: The clinicopathological diagnoses included membranous glomerulonephritis (GN) (46%), IgA nephropathy (21%), membranoproliferative GN (12.5%), postinfectious GN (12.5%), C1q nephropathy (4%), and unclassified mesangial GN (4%). No one had endothelial tubuloreticular inclusions. One patient originally diagnosed as IgA nephropathy developed anti-DNA antibody and another one patient with membranous GN developed hypocomplementemia 8 months and 10 months after renal biopsy, respectively. The two patients also developed clinical symptoms of lupus subsequently. Conclusions: There was a broad spectrum of glomerular histological findings in non-lupus “full-house” nephropathy. The possibility of “full-house” nephropathy preceding the emergence of overt systemic lupus erythematosus remained to be elucidated.

Keywords:

• “full-house” immunofluorescence
• glomerulonephritis
• lupus nephritis
• renal biopsy

INTRODUCTION

“Full-house” immune complex deposits is well known as a characteristic feature of lupus nephritis.^1,2 However, similar to the endothelial tubuloreticular inclusions characteristic of but not pathognomonic for lupus nephritis, the “full-house” immunofluorescence pattern can be occasionally found in other glomerular diseases. According to the American Rheumatism Association, at least four clinical and serological criteria are required for the diagnosis of systemic lupus erythematosus (SLE).³ If only isolated “full-house” nephropathy is present, the diagnosis of SLE cannot be established. Nevertheless, some anecdotal reports showed that the appearance of other clinical and serological features of SLE may be delayed by several years after the presence of nephropathy with a “full-house” immunofluorescence pattern and suggested that “full-house” nephropathy may be the first symptom of SLE.^4,5 As a result, “full-house” nephropathy in patients without other clinical or serological evidence of SLE may confound the diagnosis in relation to other immune complex-mediated glomerular diseases. Therefore, we conducted this study to elucidate the clinicopathological spectrum of originally non-lupus “full-house” nephropathy. Very few patients have been described in the literature and this report could be the largest series.

METHODS

Computerized records of the Department of Pathology, Changhua Christian Medical Center, were originally searched to identify all renal biopsies showing “full-house” nephropathy, which was diagnosed when immunofluorescence studies showed glomerular staining simultaneously positive for IgA, IgG, IgM, C3, and C1q with ≥1+ intensity. Only cases showing granular staining in the glomerular capillary walls, mesangial areas, or both in all glomeruli (excluding globally sclerotic glomeruli) were included; and segmental and blotchy staining typical of hyalinosis lesions was not considered. For cases meeting these criteria, the clinical records and laboratory data received with the biopsy were reviewed. Patients with no clinical or serological evidence of SLE at the time of renal biopsy were defined as having originally non-lupus “full-house” nephropathy and were enrolled in this study.

Renal biopsies were triaged and processed for light microscopy, immunofluorescence, and electron microscopy. For light microscopy, sections were fixed in 10% buffered formaldehyde and embedded in paraffin and were analyzed after hematoxylin-eosin, periodic acid-Schiff, and Masson trichrome stains. Direct immunofluorescence studies were routinely performed on cryostat sections of unfixed tissue, using antisera against IgG, IgA, IgM, C3, and C1q. For electron microscopy, the tissue was fixed in 3% buffered glutaraldehyde and embedded in epoxy resin. Thin sections were routinely contrasted with uranyl acetate and lead citrate. The extent of mesangial and endocapillary hypercellularity was graded as focal (involving <50% of glomeruli) and diffuse (involving ≥50% of glomeruli). The intensity of immunofluorescence staining was graded on a scale of 0–3+. The electron microscopy was reviewed with the specific intention of detecting any tubuloreticular inclusions in endothelial cells. The clinical records were also reviewed to determine whether any clinical or serological evidence of SLE had become apparent on follow-up.

RESULTS

A total of 1244 biopsies, not including repeat biopsies, were evaluated at our center from January 2000 to July 2009, of which “full-house” nephropathy was observed in 94 patients. Among them, 70 cases satisfied at least 4 American Rheumatism Association criteria for the diagnosis of SLE. The remaining 24 patients with no clinical or serological evidence of SLE at the time of renal biopsy were enrolled in this study.

The clinical and histological features of the 24 patients with originally non-lupus “full-house” nephropathy are summarized in Tables 1 and 2. There were 12 males and 12 females. The mean age was 47 years (range 14–79 years). Renal biopsy showed various types of glomerulonephritis (GN). The morphological diagnoses were decided on the basis of the more prevalent and diffuse glomerular lesions and included 11 (46%) cases of membranous GN, 9 (37%) of diffuse proliferative GN, and 4 (17%) of focal mesangial GN. In cases of membranous GN, mild mesangial proliferation was present in five patients. Hyaline thrombus was observed in one patient (case no. 15). In cases of diffuse proliferative GN, prominent endocapillary proliferation was seen in six patients and prominent mesangial proliferation was observed in three patients. Glomerular neutrophil infiltration was present in six patients. Mesangial interposition associated with double-contoured glomerular basement membrane was seen in three patients. Four patients had glomerular crescents of which three had diffuse crescentic GN (defined as ≥50% of glomeruli affected by crescent). Hyaline thrombus was found in one patient (case no. 13). In cases of focal mesangial GN, one patient (case no. 5) had glomerular neutrophil infiltration and hyaline thrombus. None had “wire loop” appearance on light microscopy. By immunofluorescence, all patients had diffuse and significant staining (≥1+ intensity graded on a 0–3+ scale) for IgA, IgG, IgM, C3, and C1q. Among them, 14 were limited to glomerular capillary loops, 6 were limited to mesangium, and 4 were in both. On electron microscopy study, large subepithelial hump-shaped deposits characteristic of postinfectious GN were present in three patients (case no. 10, 11, 12). No biopsies were found to have endothelial tubuloreticular inclusions.

Table 1. Clinical features of 24 patients with originally non-lupus “full-house” nephropathy

Case no.	Age/Gender	Clinicopathological diagnosis	Clinical presentation	Serological study (positive finding)	Treatment	Follow-up (months)	Development of S/S of SLE
1	57/M	IgA nephropathy	Rapid progressive renal failure	–	MPT/CYC/PRED	18	–
2	79/M	IgA nephropathy	Rapid progressive renal failure	–	MPT/CYC/PRED	12	–
3	14/F	IgA nephropathy	Nephrotic syndrome	–	PRED	20	–
4	28/F	IgA nephropathy	Chronic renal insufficiency	–	ACEI	42	–
5	14/F	IgA nephropathy	Non-nephrotic proteinuria	–	PRED	8	Malar rash, anti-DNA Ab
6	30/M	Unclassified mesangial GN	Non-nephrotic proteinuria	–	PRED	10	–
7	77/M	Membranoproliferative GN	Chronic renal insufficiency	Hypocomplement	ACEI	7	–
8	61/M	Membranoproliferative GN	Nephrotic syndrome	Hypocomplement	PRED	84	–
9	35/M	Membranoproliferative GN	Rapid progressive renal failure	Hypocomplement	MPT/PRED	12	–
10	30/M	Postinfectious GN  (MSSA-related IE)	Acute nephritic syndrome	Hypocomplement, CG	Antibiotic	24	–
11	59/M	Poststreptococcal GN	Acute nephritic syndrome	Hypocomplement, ASLO	Supportive	15	–
12	23/F	Poststreptococcal GN	Nephrotic syndrome	Hypocomplement, ASLO	ACEI	6	–
13	23/F	C1q nephropathy	Nephrotic syndrome	–	PRED/AZA	12	–
14	51/M	Membranous GN	Nephrotic syndrome	–	PRED/CYC/ARB	12	–
15	76/F	Membranous GN	Nephrotic syndrome	–	ACEI	7	–
16	58/F	Membranous GN	Nephrotic syndrome	–	PRED/CS/ARB	16	–
17	35/F	Membranous GN	Non-nephrotic proteinuria	–	PRED/AZA	10	Hypocomplement, arthritis, hair loss
18	20/F	Membranous GN  (HBV-related)	Nephrotic syndrome	HBsAg, HBeAg	Interferon	60	–
19	40/M	Membranous GN	Nephrotic syndrome	–	PRED/CYC/ACEI	72	–
20	72/M	Membranous GN	Nephrotic syndrome	–	ACEI	8	–
21	59/F	Membranous GN	Nephrotic syndrome	–	PRED/CYC	6	–
22	64/F	Membranous GN	Non-nephrotic proteinuria	–	PRED/CYC/ACEI	30	–
23	61/M	Membranous GN	Nephrotic syndrome	–	PRED/AZA	24	–
24	69/F	Membranous GN	Nephrotic syndrome	–	PRED/CYC	60	–

Note: M, male; F, female; MSSA, methicillin-sensitive Staphylococcus aureus; IE, infective endocarditis; CG, cryoglobulinemia; ASLO, antistreptolysin O; S/S, symptom and sign; HBV, hepatitis B virus; MPT, methylprednisolone pulse therapy; PRED, oral prednisone; CYC, cyclophosphamide; AZA, azathioprine; CS, cyclosporine; ACEI, angiotensin-converting enzyme inhibitors; ARB, angiotensin II receptor blocker.

Table 2. Histological features of 24 patients with originally non-lupus “full-house” nephropathy

Case no.	Light microscopy	Immunofluorescence^a	Electron microscopy	Classification as in SLE^b	A/C index
1	Diffuse proliferative GN with crescents	IgA(2+), IgG(2+), IgM(2+), C3(2+), C1q(1+)/Mes	Mesangial deposits	IV	9/4
2	Diffuse proliferative GN with crescents	IgA(2+), IgG(1+), IgM(1+), C3(2+), C1q(1+)/Mes	Mesangial deposits	IV	10/4
3	Focal mesangial GN	IgA(1+), IgG(1+), IgM(1+), C3(1+), C1q(1+)/Mes	Mesangial deposits	II	0/2
4	Focal mesangial GN	IgA(2+), IgG(2+), IgM(2+), C3(2+), C1q(2+)Mes	Mesangial deposits	II	0/3
5	Focal mesangial GN	IgA(1+), IgG(1+), IgM(1+), C3(1+), C1q(1+)/Mes	Mesangial deposits	II	4/1
6	Focal mesangial GN	IgA(1+), IgG(2+), IgM(2+), C3(2+), C1q(1+)Mes	Measngial deposits	III	2/2
7	Diffuse proliferative GN with double-contoured GBM	IgA(1+), IgG(1+), IgM(1+), C3(2+), C1q(1+)/Mes, GCW	Subendothelial and mesangial deposits	IV	1/3
8	Diffuse proliferative GN with double-contoured GBM	IgA(2+), IgG(2+), IgM(1+), C3(1+), C1q(2+)/Mes, GCW	Subendothelial and mesangial deposits	IV	3/4
9	Diffuse proliferative GN with crescents	IgA(1+), IgG(2+), IgM(1+), C3(2+), C1q(1+)/Mes, GCW	Subendothelial and mesangial deposits	IV	7/5
10	Diffuse proliferative GN	IgA(1+), IgG(1+), IgM(1+), C3(2+), C1q(1+)/GCW	Subepithelial humps	IV	6/0
11	Diffuse proliferative GN	IgA(2+), IgG(1+), IgM(2+), C3(3+), C1q(2+)/GCW	Subepithelial humps	IV	6/0
12	Diffuse proliferative GN	IgA(2+), IgG(2+), IgM(1+), C3(2+), C1q(1+)/GCW	Subepithelial humps	IV	6/1
13	Diffuse proliferative GN	IgA(1+), IgG(2+), IgM(1+), C3(2+), C1q(3+)/Mes, GCW	Subendothelial and mesangial deposits	V	6/1
14	Membranous GN	IgA(1+), IgG(2+), IgM(1+), C3(2+), C1q(1+)/GCW	Subepithelial deposits	V	0/1
15	Membranous GN with hyaline thrombus	IgA(1+), IgG(2+), IgM(1+), C3(2+), C1q(1+)/GCW	Intramembranous deposits	V	1/5
16	Membranous GN	IgA(1+), IgG(2+), IgM(1+), C3(2+), C1q(1+)/GCW	Subepithelial deposits	V	0/1
17	Membranous and focal mesangial GN	IgA(1+), IgG(2+), IgM(1+), C3(1+), C1q(1+)/GCW	Subepithelial deposits	V	0/1
18	Membranous GN	IgA(1+), IgG(3+), IgM(2+), C3(2+), C1q(3+)/GCW	Subepithelial deposits	V	0/2
19	Membranous and focal mesangial GN	IgA(1+), IgG(2+), IgM(1+), C3(2+), C1q(1+)/GCW	Subepithelial deposits	V	0/2
20	Membranous and focal mesangial GN	IgA(1+), IgG(1+), IgM(1+), C3(2+), C1q(2+)/GCW	Subepithelial deposits	V	0/3
21	Membranous and focal mesangial GN	IgA(1+), IgG(1+), IgM(1+), C3(1+), C1q(1+)/GCW	Subepithelial deposits	V	0/1
22	Membranous GN	IgA(1+), IgG(3+), IgM(1+), C3(2+), C1q(2+)GCW	Subepithelial deposits	V	0/1
23	Membranous GN	IgA(1+), IgG(1+), IgM(1+), C3(1+), C1q(1+)/GCW	Subepithelial deposits	V	0/0
24	Membranous and focal mesangial GN	IgA(1+), IgG(2+), IgM(1+), C3(1+), C1q(1+)/GCW	Subepithelial deposits	V	0/3

Notes:^aStaining intensity graded on a 0 to 3+ scale; ^bBased on the revisited classification of glomerulonephritis in SLE published by Weening et al. in 2004²⁰.

GN, glomerulonephritis; GBM, glomerular basement membrane; Mes, mesangium; GCW, glomerular capillary wall; A/C, activity/chronicity.

On clinical presentation, 13 patients had nephrotic syndrome with a mean proteinuria of 5.6 ± 2.2 g/24 h (range 3.6–9.8 g/24 h), 4 had non-nephrotic proteinuria, 3 had rapidly progressive renal failure, 2 had acute nephritic syndrome, and 2 had chronic renal insufficiency. Hypocomplementemia was present in six patients. None had anti-nuclear and anti-double-stranded DNA antibodies. In all patients, the final diagnoses were made on a clinicopathological correlation and included 11 (46%) cases of membranous GN, 5 (21%) of IgA nephropathy, 3 (12.5%) of type I membranoproliferative GN, 3 (12.5%) of postinfectious GN, 1 (4%) of C1q nephropathy, and 1 (4%) of unclassified mesangial GN. Treatment was individual and toward the final diagnosis.

After a mean follow-up of 24 months (range 6–84 months), one patient (case no. 5) originally diagnosed as IgA nephropathy developed malar rash and anti-double-stranded DNA antibody 8 months after renal biopsy. Another one patient (case no. 17) with membranous GN developed arthritis, hair loss, and hypocomplementemia 10 months after renal biopsy. Subsequently, the two patients were diagnosed and treated as lupus nephritis by their clinicians. None of the remaining 22 patients had developed any clinical or serological evidence of SLE.

DISCUSSION

The “full-house” immunofluorescence staining pattern occurs most commonly in lupus nephritis. Although “full-house” nephropathy is a characteristic feature of lupus nephritis, it also can be found in other types of GN. There have rarely been reported in IgA nephropathy, membranoproliferative GN, and postinfectious GN.^6–8 Our report showed that “full-house” nephropathy could be seen in various types of GN. These included membranous GN (accounted for 9% of idiopathic cases identified at our center), IgA nephropathy (3% of idiopathic cases), and type I membranoproliferative GN (25% of idiopathic cases). Furthermore, 3 out of 27 (11%) patients diagnosed as having postinfectious GN at our center also presented with a “full-house” immunofluorescence pattern. One patient had predominant C1q mesangial deposition (less-intense staining for IgA, IgG, and IgM) consistent with previous descriptions of C1q nephropathy.^9,10

The association of postinfectious GN with “full-house” nephropathy has rarely been reported. To our knowledge, only one patient with infective endocarditis reported to have “full-house” nephropathy in English literature.⁸ On immunofluorescence microscopy, the typical finding of postinfectious GN is C3 deposition, often with IgG and occasionally with IgM. IgA deposition is usually not predominant. Recently, however, there have been several reports of postinfectious GN with IgA-dominant immune complex deposition, mostly related to staphylococcal infection.¹¹ It has been speculated that bacterial superantigens played an important role in the pathogenesis of this unique form of postinfectious GN.¹² C1q deposition is also unusual in postinfectious GN. Autoantibodies against complement C1q (anti-C1q) have been found to strongly correlate with active lupus nephritis. It was also found to correlate with its deposition in the glomeruli.¹³ Recent data suggested that anti-C1q might also correlate with more severe forms of acute poststreptococcal GN and pointed to a potential pathogenic role of anti-C1q in acute poststreptococcal GN.¹⁴ However, the correlation between anti-C1q and its glomerular deposition in postinfectious GN remains to be determined.

It has been argued whether “full-house” nephropathy precedes the development of overt lupus. The diagnosis of lupus nephritis is often straightforward. Patients typically have clinical evidence of extrarenal lupus with diagnostic serological changes. The renal histology of lupus nephritis is very variable, but certain features are usually found only in lupus nephritis. These include a “full-house” immunofluorescence pattern, endothelial tubuloreticular inclusions on electron microscopy, and membranous GN with mesangial deposits.^1,2 However, a variant has been described, sometimes called “seronegative lupus nephritis.” This term is usually used to describe patients in whom the renal histology is typical of lupus nephritis, yet there is no clinical or serological evidence of SLE at the time of renal biopsy. It has been proposed that a significant proportion of patients with such “seronegative lupus nephritis” will develop overt SLE in due course.^4,5 As a consequence, many such patients are treated with immunosuppressive regimens similar to those proposed for overt lupus nephritis with equivalent histological severity, often including corticosteroids and cytotoxic agents. However, published evidence to support these approaches is limited, and not all studies have shown that such patients will progress to overt SLE.

Little studies have been published on this issue in the last 20 years. Cairns et al.¹⁵ reported 11 patients diagnosed as “seronegative lupus nephritis”, 4 of whom developed overt lupus over 1–7 years of follow-up. Jones and Magil¹⁶ reported five patients, none of whom developed overt lupus over 0.1–4.8 years. Adu et al.¹⁷ reported 17 patients of whom 5 developed definite evidence of lupus and 4 were probable lupus over 1–10 years. Enriquez et al.¹⁸ reported three patients who did not develop overt lupus over 1–2 years. Gianviti et al.⁴ reported 17 patients of whom 1 developed overt lupus and 2 developed autoantibodies over 3–10 years. Recently, Sharman et al.¹⁹ reported 9 patients, none of whom developed lupus over 0.1–9 years, and furthermore, they felt C1q nephropathy rather than lupus nephritis should be considered in these patients. In our report, two patients (case no. 5, 17) developed evidences of lupus. However, they fulfilled only 2–3 items of American Rheumatism Association criteria. We felt case no. 5 should be considered as having SLE because she developed two of the most important features of lupus: nephritis and autoantibodies. Although case no. 17 had no serological evidence of lupus, SLE was still the most probable diagnosis based on the following features: immunological phenomenon (hypocomplementemia), nature of systemic involvement (kidney, skin, and joint), and predictive renal biopsy finding (“full-house” nephropathy). In summary, of the 86 patients with “seronegative lupus nephritis” described in the literature (including our report), we identified 59 patients who met the criteria of “full-house” nephropathy. Of the 59 patients, only 7 patients developed clinical or serological evidence of SLE during follow-up (Table 3).

Table 3. Data of patients with originally non-lupus “full-house” nephropathy described in the literature

Authors [reference]	Case no.	Renal biopsy (case number)	Case number of development of SLE (time after presentation)
Cairns et al.^15	1	Membranous GN (1)	1 (3 months)
Jones and Magil^16	5	Mesangial GN (5)	None
Adu et al.^17	3	Mesangial GN (1) Membranous GN (2)	1 (2 years)
Enriquez et al.^18	3	Mesangial GN (3)	None
Gianviti et al.^4	17	Diffuse proliferative GN (8) Membranous GN (4) Membranoproliferative GN (3) Focal segmental GS (1) Focal GN (1)	3 (3, 5, 10 years)
Sharman et al.^19	6	Diffuse mesangial GN (3) Membranous GN (2) Focal segmental GN (1)	None
Present study	24	Membranous GN (11) Diffuse proliferative GN (6) Membranoproliferative GN (3) Focal mesangial GN (4)	2 (8, 10 months)

Note: GN, glomerulonephritis; GS, glomerulosclerosis; S/S, symptom and sign; SLE, systemic lupus erythematosus.

In conclusion, there was a broad spectrum of glomerular histological findings in non-lupus “full-house” nephropathy. Although the possibility of “full-house” nephropathy preceding the emergence of overt SLE remained uncertain, we suggested that patients with originally non-lupus “full-house” nephropathy still should be kept under surveillance for the appearance of symptoms and/or autoantibodies suggestive of SLE.

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.