Immune-Complex Mediated Mesangial Proliferative Glomerulonephritis with Full-House Pattern Observed During Treatment of Immune Thrombocytopenic Purpura

Abstract

This study reports a rare case of immune-complex mediated mesangial proliferative glomerulonephritis (ICGN) with a full-house pattern in a 56-year-old Japanese man, observed during the treatment of immune thrombocytopenic purpura (ITP). Because of persistent complement deficiency and worsening of kidney function, he was treated with prednisolone, and his urinary findings improved markedly. However, as the complement titers were still low, mycophenolate mofetil was also prescribed, which normalized complement levels. Production of anti-platelet antibodies is considered to be involved in the etiology of ITP. Although little is known about the mechanism by which ITP causes glomerulonephritis, including ICGN, glomerular deposition of circulating immune complexes synthesized by antiplatelet antibodies may be involved. This case shows full-house nephropathy, suggesting the involvement of immune complexes, which in turn, suggested an association between ITP and glomerulonephritis.

Background

Immune thrombocytopenic purpura (ITP) is primarily diagnosed by exclusion, because there is no established disease-specific test. ITP is diagnosed if a patient has thrombocytopenia (<10 × 10⁴/µL), but erythrocytes (excluding anemia caused by bleeding or chronic iron deficiency) and leukocytes are normal, and other possible diseases that cause thrombocytopenia are excluded. Bleeding symptoms, bone marrow biopsy, and platelet-associated IgG (PAIgG) may help diagnose ITP, but none of them are specific. ¹ ITP is an autoimmune disease caused by the destruction of platelets in the periphery and their inappropriate production in the bone marrow. Glomerulonephritis such as lupus nephritis, IgA nephropathy, and membranous nephropathy (MN) have been reported as causes of ITP-related nephritis,^2-4 with lupus nephritis being particularly likely to occur. Continued production of autoantibodies, such as antinuclear antibodies, during the course of ITP may lead to glomerulonephritis, although precise mechanism remains unclear.

In this study, we report a rare case of immune-complex mediated mesangial proliferative glomerulonephritis (ICGN) and full-house pattern of immunofluorescence (IF) observed during ITP treatment, which was eventually treated with prednisolone (PSL) and mycophenolate mofetil (MMF).

Case Presentation

A 56-year-old Japanese man underwent a detailed examination because of a low platelet count (7.3 × 10⁴/µL) identified during a physical examination that was conducted in 2012. He had purpura, PAIgG was high (52 ng/10⁷ cell), but no findings suggested other diseases inducing thrombocytopenia, including a bone marrow biopsy. He was then diagnosed with ITP.

He started treatment with PSL 45 mg/day in December 2013 because his platelet count dropped to 1.8 × 10⁴/µL. The PSL dose was gradually reduced and was terminated in May 2015. During this period, the patient had steroid diabetes and was treated with nateglinide, vildagliptin, and metformin hydrochloride. He started treatment with eltrombopag 12.5 mg in July 2015 because his platelet count dropped to 1.5 × 10⁴/µL again. Eltrombopag was terminated in December 2015 because the platelet count recovered to 16.3 × 10⁴/µL, but was restarted in March 2017 because of the decreased platelet count (1.7 × 10⁴/µL). It was switched to rituximab (375 mg/m² per week), which was administered four times. Platelet count improved to 8.1 × 10⁴/µL. He took diazepam for 3 months for neuropathy. Because he had kidney dysfunction (serum creatinine (sCr) 1.13 mg/dL), hematuria (2+), and proteinuria (+) in July 2018, chronic glomerulonephritis was suspected. Since hematuria and proteinuria were persistent, he was admitted for kidney biopsy in January 2019. Physical examination on admission were as follows: height, 176.4 cm; weight, 85.7 kg; body mass index, 27.5 kg/m²; body temperature, 36.4°C; blood pressure, 132/78 mmHg; heart rate, 70 beats/minute with sinus rhythm. No butterfly erythema, discoid rash, joint pain, muscular pain, photo-sensitivity, oral ulcers or slapped-cheek appearance were observed. Clinical laboratory data were as follows (Table 1): white blood cell count, 6.9 × 10³/μL; neutrophil, 54.6%; hemoglobin, 13.7 g/dL; platelet count, 5.5 × 10⁴/μL; sCr, 1.22 mg/dL; estimated glomerular filtration rate, 49.4 mL/min/1.73 m²; IgG, 1155 mg/dL; IgA, 195 mg/dL; IgM, 199 mg/dL; C3, 66 (normal range, 65-135) mg/dL; C4, 14 (13-35) mg/dL; CH50, 16.5 (30-46) U/mL; and antinuclear antibody, <40 fold; cryoglobulin, (−). His urinalysis revealed 20-29 red blood cells/high-power fields and proteinuria of 0.88 g/gCr. No serositis was observed.

Table 1.

Laboratory data on admission.

Paraclinical	result	Reference range	Paraclinical	result	Reference range
Urinalysis			TIBC	335 μg/dL	235-385
Protein	(2+)	(−)	Ferritin	75.5 ng/dL	23-250
Occult blood	(3+)	(−)	CRP	⩽ 0.3 mg/dL	⩽0.3
Red blood cell	20-29/HPF	< 1	Immunology
Urine chemistry			IgG	1155 mg/dL	870-1700
PCR	0.88 g/gCr	<0.15	IgA	195 mg/dL	110-410
β2-MG	59 IU/L	⩽4.2	IgM	199 mg/dL	35-220
NAG	10.7 μg/L	⩽200	C3	66 mg/dL	65-135
Urinary protein fraction			C4	14 mg/dL	13-35
Alb	75.3%		CH50	16.5 U/mL	31-58
α1-globulin	4.9%		RF	<3 IU/mL	⩽20
α2-globulin	4.9%		ANA	<40	<40
β-globulin	7.5%		Anti-ds-DNA antibody	4.7 U/mL	⩽12.0
γ-globulin	7.4%		Anti-Sm antibody	(−)	(−)
BJP	(−)	(−)	aCL	9 U/mL	<10
Complete blood cell count			LAC	1.2	<1.3
WBC	6900/μL	3300-8600	PB-IgG	(−)	(−)
Neutrophil	54.6%	38.5-80.5	MPO-ANCA	<1.0 U/mL	<3.5
Lymphocyte	28.2%	16.5-49.5	PR3-ANCA	<1.0 U/mL	<3.5
Others	17.2%	0-21.0	Anti-GBM antibody	<2.0 U/mL	<3.0
Hb	13.7 g/dL	13.7-16.8	Anti-SS-A antibody	<1.0 U/mL	<10.0
Ht	41.4%	4.0-5.0	Anti-SS-B antibody	<1.0 U/mL	<10.0
Plt	5.5 × 104/μL	15.0-40.0	Anti-Scl-70 antibody	<2.0 U/mL	<10.0
Blood chemistry			Cryoglobulin	(−)	(−)
TP	6.8 g/dL	6.5-8.2	DCT	(−)	(−)
Alb	4.3 g/dL	3.8-5.2	FLC (κ/λ ratio)	1.294	0.26-1.65
T-bil	0.67 mg/dL	0.2-1.2	κ chain	30.4 mg/L	3.3-19.4
AST	17 U/L	8-30	λ chain	23.5 mg/L	5.7-26.3
ALT	16 U/L	5-35	Infection
ALP	235 U/L	100-340	HBs antigen	(−)	(−)
LDH	198 U/L	100-225	HCV antibody	(−)	(−)
γ-GTP	14 U/L	7-70	RPR	(−)	(−)
CK	177 U/L	⩽160	TPHA	(−)	(−)
Glu	108 mg/dL	65-110	HIV	(−)	(−)
TC	221 mg/dL	130-230	Coagulation
LDL	155 mg/dL	⩽140	APTT	26.4 sec	24.0-32.0
TG	135 mg/dL	30-150	APTT (control)	30.1 sec
BUN	22 mg/dL	8-20	PT-INR	0.97	0.9-1.1
Cr	1.22 mg/dL	0.61-1.13	Endocrine
eGFR	49.4 mL/min/1.73 m2		TSH	0.83 μIU/mL	0.61-4.68
Uric acid	7.6 mg/dL	2.7-8.0	Free T4	0.94 ng/dL	0.76-1.65
Sodium	141 mmol/L	135-147	BNP	⩽5.10 pg/mL	⩽ 18.4
Potassium	4.3 mmol/L	3.5-5.0	Venous blood gas
Chloride	108 mmol/L	98-108	pH	7.322
Calcium	8.4 mg/dL	8.5-10.3	PaCO2	46.5 Torr
Phosphorus	2.6 mg/dL	2.3-4.5	${\mathrm{HCO}}_3^{-}$	23.4 mmol/L
Fe	148 μg/dL	80-160	BE	−2.5 mmol/L

Urinalysis: Alb, albumin; β2-MG, β2-microglobulin; BJP, Bence jones protein; g/gCr, g per g creatinine; HPF, high-power field; NAG, N-acetyl-β-D-glucosaminidase.

Complete blood cell count: Hb, hemoglobin; Ht, hematocrit; Plt, platelets; WBC, white blood cell.

Blood chemistry: ALP, alkaline phosphatase; ALT, alanine aminotransferase; AST, aspartate aminotransferase; BUN, blood urea nitrogen; CK, creatine kinase; Cr, creatinine; CRP, C-reactive protein; eGFR, estimated glomerular filtration rate; Fe, Ferrum; Glu, glucose; LDH, lactate dehydrogenase; LDL, low-density lipoprotein; T-bil, total bilirubin; TC, total cholesterol; TG, triglyceride; TIBC, total iron binding capacity; TP, total protein; γ-GTP, gamma-glutamyltranspeptidase.

Immunology: aCL, anticardiolipin antibody; ANA, antinuclear antibody; Anti-ds-DNA antibody, anti-double stranded DNA antibody; Anti-GBM, antibody antiglomerular basement membrane antibody; Anti-Scl-70 antibody, antitopoisomerase antibody; Anti-Sm antibody, anti-Smith antibody; Anti-SS-A antibody, anti-Sjogren syndrome A antibody; Anti-SS-B antibody, anti-Sjogren syndrome B antibody; C3, complement factor 3; C4, complement factor 4; CH50, complement factor 50; DCT, direct Coombs test; FLC, free light chain; IgA, immunoglobulin A; IgG, immunoglobulin G; IgM, immunoglobulin M; LAC, lupus anticoagulant; MPO-ANCA, myeloperoxidase antineutrophil cytoplasmic antibody; PB-IgG, platelet-bindable IgG; PR3-ANCA, proteinase antineutrophil cytoplasmic antibody; RF, rheumatoid factor.

Infection: HBs antigen, hepatitis B surface antigen; HCV, antibody hepatitis C virus antibody; HIV, Human Immunodeficiency Virus.; RPR, rapid plasma regain test; TPHA, Treponema pallidum hemagglutination test.

Coagulation: APTT, activated partial thromboplastin time; PT-INR, prothrombin time-international normalized ratio.

Endocrine: BNP, brain natriuretic hormone; fT4, free thyroxine; TSH, thyroid-stimulating hormone.

Venous blood gas: BE, base excess; ${\mathrm{HCO}}_3^{-}$ , hydrogen bicarbonate; PaCO₂, partial pressure of carbon dioxide.

A kidney biopsy was performed and a specimen containing 25 glomeruli was obtained. Under a light microscope (LM), periodic acid-Schiff staining revealed mesangial cell proliferation and total circumferential thickening of glomerular basement membrane (shown in Figure 1a and b). Under an electron microscope (EM), electron-dense deposits (EDD) were detected in the mesangial and subendothelial regions (shown in Figure 1c and d). No thrombi with immune complexes were seen on LM or EM. Mild tubular atrophy and lymphocytic infiltration into the interstitial space were observed. The IF results showed granular deposition of IgG, IgA, IgM, C3, C4, and C1q in the mesangial region (shown in Figure 2). Based on these results, the patient was diagnosed with immune-complex mediated mesangioproliferative glomerulonephritis. Because of persistent complement deficiency and worsening of kidney function, he subsequently received treatment with PSL 40 mg/day. The proteinuria and hematuria disappeared. However, the complement levels remained low (shown in Figure 3). In September 2019, MMF was started at 500 mg/day, and its dose eventually reached 2000 mg/day. The levels of complement factors improved to 78 mg/dL for C3, 13 mg/dL for C4, and 28.7 U/mL for CH50. sCr level improved from 1.49 mg/dL to 1.06 mg/dL. Platelet count improved to 20.3 × 10⁴/µL.

Figure 1.

Representative light microscopy and electron microscopy on kidney biopsy specimen: (a) mesangial hypercellularity (arrow) by light microscope (periodic acid-Schiff staining, ×400), (b) total circumferential thickening of glomerular basement membrane (arrow) by light microscope (periodic acid-Schiff staining, ×400), (c) electron-dense deposits (EDD) in the mesangial region (arrow) by electron microscope (×1500), and (d) EDD in the subendothelial region (arrow) by electron microscope (×1000).

Figure 2.

Immunohistochemistry of the glomerulus (immunofluorescence staining, ×400).

Figure 3.

Clinical course of the patient after the initiation of treatment for membranoproliferative glomerulonephritis.

Discussion

We encountered a 56-year-old Japanese man who had proteinuria and hematuria during treatment for ITP, who was diagnosed with ICGN by kidney biopsy. As kidney biopsy confirmed immune-complex mediated mesangioproliferative glomerulonephritis, we first considered IgA nephropathy, as it has been reported that IgA nephropathy can be associated with ITP. However, IF shows a full house pattern and IgA nephropathy is unlikely. IF showed IgM-predominant deposits with C1q, suggesting cryoglobulin nephropathy. However, serum cryoglobulin was negative, and neither cryoglobulin thrombi in LM nor short microtubules/cylinders in EM was observed. Systemic lupus erythematosus is another disease that has been reported to be involved in ITP. Although IF showed full-house pattern, no autoantibodies or symptoms can support the diagnosis of systemic lupus erythematosus (SLE). Moreover, lupus nephritis usually showed IgG-predominant IF pattern, which was different from the present case. Secondary nonlupus full-house nephropathy due to infections and drugs is also considered. However, he was HIV negative and did not have characteristic signs of parvovirus, such as a slapped cheek appearance. The patient had taken PSL, eltrombopag, rituximab, nateglinide, vildagliptin, metformin hydrochloride and diazepam, but, to best of our knowledge, there are no report of these drugs causing ICGN. Rituximab administration might have masked the development of nonlupus full-house nephropathy. However, urine tests were not performed before rituximab administration, we cannot evaluate influence of rituximab on nephropathy in this case.

ITP is an autoimmune disease caused by the destruction of platelets in the periphery and their inappropriate production in the bone marrow. The etiology of ITP is well known, and it involves the recognition of membrane GP complexes (primarily GPIb/IX and GPIIb/IIIa) by anti-platelet and anti-angiocyte antibodies. Opsonized platelets are phagocytized by macrophages in the spleen and liver via the Fcγ receptor. Activation of the classical complement pathway also causes platelet destruction via opsonization and complement-dependent cell injury. Platelet destruction also occurs via FcγR-independent pathways, but its underlying mechanisms are unknown.^5,6 In ITP, autoantibodies such as antinuclear, anti-double stranded DNA antibody, anticardiolipin antibodies, and antiplatelet antibodies may be produced. The continued production of these autoantibodies during the ITP course can also lead to other autoimmune diseases. ³

Lupus nephritis, IgA nephropathy, and MN have been reported to cause ITP-associated nephritis.^{2 -4} There are no reports of ITP complicating mesangial proliferative glomerulonephritis with hypocomplementemia. One case report showed that hypocomplementemic mesangial proliferative glomerulonephritis progressed to focal membranoproliferative glomerulonephritis. ⁷ Hypocomplementemia was also observed in this case, and the possibility of progression to membranoproliferative glomerulonephritis (MPGN) could not be ruled out in the future. To the best of our knowledge, only two case reports have explored the relationship between MPGN and ITP: one report was based on a patient with MPGN, ITP, and autoimmune hemolytic anemia, ⁸ and the other was based on a patient who developed MPGN while receiving hepatitis C virus treatment with rituximab. ⁹

A case of MPGN with immunoglobulin-positive is classified as immune complex-related MPGN. ¹⁰ Immune complex-related MPGN can be caused by cryoglobulinemia, M-proteinemia, infections, and autoimmune diseases such as SLE, rheumatoid arthritis, Sjogren’s syndrome, and IgA nephropathy. The following mechanisms have been proposed to explain the pathophysiology of MPGN. Immune complexes activate the classical pathway, thereby acutely damaging glomerular capillaries and mesangium, and cause an influx of inflammatory cells, leading to proliferative glomerular changes. The glomerular repair response is characterized by the generation of neovascular membranes by endothelial and mesangial cells, which trap immune complexes and form a double contour. In addition, mesangial expansion occurs due to an increase in the number of mesangial cells, the levels of mesangial matrix, and mononuclear cell infiltration. ¹¹ Lande et al. ⁴ proposed three underlying mechanisms for ITP caused by MN: (a) reaction of circulating antiplatelet antibodies with native antigens of the glomerular capillary wall, (b) reaction of circulating antiplatelet antibodies with transplanted glomerular antigens after their release following platelet destruction, and (c) deposition of circulating immune complexes formed by antiplatelet antibodies on glomeruli. Therefore, immune complexes formed by antiplatelet antibodies might also lead to the development of ICGN or MPGN.

Factors that are reported to be associated with a poor kidney prognosis in MPGN include nephrotic syndrome, elevated sCr level, and hypertension. ¹² MPGN is generally treated using steroids and immunosuppressive drugs such as cyclosporine, but evidence for their effectiveness in adults is insufficient. Small observational studies have suggested potential benefit of using MMF, especially when prescribed with oral glucocorticoids. ¹³ This patient had ICGN, not MPGN, but was treated according to MPGN because of worsening kidney function. Although kidney function was preserved due to treatment with PSL and MMF, the patient experienced a slight kidney dysfunction. Patients initially diagnosed with non-lupus full-house nephropathy may develop SLE during the course of their disease. Although SLE and non-lupus full-house nephropathy are different clinical entities, the outcomes are similar. ¹⁴ In this case, the possibility of developing SLE in the future cannot be excluded and requires careful follow-up.

Conclusions

This case report describes a rare case of ICGN with full-house pattern during the treatment of immune thrombocytopenic purpura. Although the mechanism that links ITP and ICGN is unknown, glomerular deposition of circulating immune complexes formed by antiplatelet antibodies may be involved in this association, and it is necessary to confirm the causal relationship between ITP and ICGN in the future.