- Email: [email protected]
Characteristic tubulointerstitial nephritis in IgG4-related disease
Human Pathology (2012) 43, 536–549
www.elsevier.com/locate/humpath
Original contribution
Characteristic tubulointerstitial nephritis in IgG4-related disease☆,☆☆ Yutaka Yamaguchi MD a , Yukiko Kanetsuna MD b , Kazuho Honda MD c , Nobuaki Yamanaka MD d , Mitsuhiro Kawano MD e , Michio Nagata MD f,⁎ on behalf of the Japanese study group on IgG4-related nephropathy a
Yamaguchi's Pathology Laboratory, Matsudo, Chiba, 270-2231 Japan Department of Pathology, The Jikei University Kashiwa Hospital, Kashiwa, Chiba, 277-8567 Japan c Department of Pathology, Tokyo Women's Medical University, Tokyo, 162-8666 Japan d Tokyo Kidney Research Institute, Tokyo, 113-8602 Japan e Division of Rheumatology, Department of Internal Medicine, Graduate School of Medical Science, Kanazawa University, Kanazawa, 920-8641 Japan f Department of Kidney and Vascular Pathology, Graduate School of Comprehensive Human Sciences, University of Tsukuba, Tsukuba, 305-8575 Japan b
Received 31 March 2011; revised 28 May 2011; accepted 1 June 2011
Keywords: IgG4-related disease; Tubulointerstitial nephritis; Tubulointerstitial deposition; Interstitial fibrosis/sclerosis
Summary Nephropathy associated with IgG4-related disease is characterized by tubulointerstitial nephritis. To better identify its pathology, the present study analyzed clinicopathologic features of IgG4-related tubulointerstitial nephritis cases from across Japan. Sixteen cases were identified as IgG4-related nephropathy using the criterion of high serum IgG4 levels (N135 mg/dL) with abnormal kidney computed tomography or elevated serum creatinine levels. Male predominance (75%) and advanced age (average, 62.0 years) were noted. Eight cases displayed no autoimmune pancreatitis. Renal computed tomography abnormalities were found in 12 of 13 cases examined. Renal dysfunction was found in 15 of 16 cases at biopsy. Distinctive features of tubulointerstitial lesions included (1) well-demarcated borders between involved and uninvolved areas; (2) involvement of the cortex and medulla, often extending beyond the renal capsule and with occasional extension to retroperitoneal fibrosis; (3) interstitial inflammatory cells comprising predominantly plasma cells and lymphocytes, with a high prevalence of IgG4-positive cells often admixed with fibrosis; (4) peculiar features of interstitial fibrosis resembling a “bird's-eye” pattern comprising fibrosis among inter–plasma cell spaces; and (5) deposits visible by light and immunofluorescent microscopy in the tubular basement membrane, Bowman capsule, and interstitium that are restricted to the involved portion, sparing normal parts. Ultrastructural analysis revealed the presence of myofibroblasts with intracellular/pericellular collagen accompanied by plasma cell accumulation from an early stage. Histology could not discriminate between IgG4-related tubulointerstitial nephritis with and without autoimmune pancreatitis. In conclusion, the distinctive histologic features of IgG4related tubulointerstitial nephritis can facilitate the differential diagnosis of tubulointerstitial nephritis, even without autoimmune pancreatitis or an abnormal computed tomography suggesting a renal tumor. © 2012 Elsevier Inc. All rights reserved.
☆
Conflict of interest: none. Disclosures: none. ⁎ Corresponding author. E-mail address: [email protected] (M. Nagata). ☆☆
0046-8177/$ – see front matter © 2012 Elsevier Inc. All rights reserved. doi:10.1016/j.humpath.2011.06.002
Tubulointerstitial nephritis in IgG4-related disease
537
1. Introduction
2. Subjects and methods
IgG4-related disease is a systemic disease associated with high serum IgG4 levels that was first identified in Japan and is now emerging worldwide [1-4]. IgG4-related disease encompasses various uncommon clinical manifestations, including autoimmune pancreatitis (AIP), Mikulicz disease, and Küttner disease, despite a relatively common background pathology: interstitial fibrosis/sclerosis and lymphoplasmacytic infiltration [4-9]. IgG4-related disease is associated with high IgG4 serum levels in approximately 70% to 80% of cases. The kidney is an organ particularly affected by IgG4-related disease, and chronic renal failure occasionally develops [10-14]. In practice, IgG4-related nephropathy is occasionally found by abnormal computed tomography (CT), revealing a tumor that is sometimes treated by performing an unnecessary nephrectomy [15]. Among several forms of IgG-related disease in the kidney (IgG4-related nephropathy), previous reports have documented that tubulointerstitial nephritis (TIN) is common and may determine a prognosis [10-12,16-22]. IgG4-related TIN is often accompanied by AIP; however, several cases without AIP have also been reported [6,14,16,18]. Generally, the histology of TIN is fairly nonspecific; thus, the diagnosis of IgG4-related TIN, particularly in cases without AIP, is problematic. Because steroid therapy has been found to be effective in AIP [1,4,23,24], a proper diagnosis of IgG4-related TIN is very important so that steps can be taken to halt disease progression. Several case reports have described the histologic features of IgG4-related TIN. Most were single or a few cases associated with AIP and were subjected to renal biopsy, revealing various features ranging from mild interstitial inflammation to diffuse tubulointerstitial fibrosis [10-14,17-22]. Some reports have demonstrated IgG4 immunostaining, but most did not refer to the relationship between the presence of IgG4-positive cells and disease progression because of the small number of cases and the inconsistent severity of the interstitial injury. In particular, whether the different pathologic profiles noted in previous reports are based on different stages of the same disease or are histologic variants has not been determined. Recently, Saeki et al [25] documented clinicopathologic characteristics of IgG4-related TIN. Although its pathology was briefly summarized in their publication, detailed morphological considerations, including histologic staging and features of progression, were not well discussed. The present study analyzed 16 cases of IgG4-related TIN from across Japan (including 4 cases [cases 5, 8, 11,and 13] that appeared in the report of Saeki et al) to better define the clinical and pathologic characteristics of this condition, facilitate its diagnosis, and discuss its pathogenesis.
2.1. Subjects The study group collected 16 cases from across Japan. The diagnostic criterion for IgG4-related nephropathy was the presence of either AIP with renal dysfunction or elevated serum IgG4 levels (N135 mg/dL) without AIP. The diagnosis of AIP was made according to Asian criteria, as previously reported [26]. Patients with clinically suspected Castleman disease or with high titers of anti–double-stranded DNA, anti-Sm, or antiSjogren syndrome A were not included. Renal dysfunction was defined by more than 1.1 mg/dL of serum creatinine. Among the 16 cases, 4 (cases 5, 8, 11, 13) had appeared in the recent report by Saeki et al [25]; the remaining 12 cases were newly enrolled and had not been published. The study was approved by the ethical committees of Kanazawa University.
2.2. Histopathology Renal biopsy samples were processed by standard techniques for light microscopy (LM), immunostaining, and electron microscopy (EM). Immunostaining was performed either with immunofluorescence (IF, 5 cases with frozen sections) or with peroxidase-antiperoxidase methods (11 cases with paraffin sections) using polyclonal antibodies against IgG, IgA, IgM, C3, C1q, and κ and λ light chains (Dako, Kyoto, Japan). EM was performed in 10 cases, using standard techniques.
2.3. Immunohistochemistry for IgG subclass staining IgG subclass staining was performed in all cases by a standard method using an avidin-biotin immunoperoxidase technique with antigen retrieval by microwave. The primary antibodies were mouse monoclonals against each IgG subclass (IgG1, IgG2, IgG3, and IgG4) (Zymed Laboratories Inc, San Francisco, CA), applied at a 1:100 dilution.
2.4. Statistics All data are presented as absolute numbers or means ± SE. Statistical differences between groups were evaluated using Student unpaired t test. P b .05 was deemed to indicate statistical significance.
3. Results 3.1. Clinical features All cases consisted of Japanese patients, most of whom were men (75%) and older than 50 years (average, 62.0 years; Table 1). Renal dysfunction was found in 15 cases (94%), and
538
Y. Yamaguchi et al.
Table 1 Case
With 1 2 3 4 5 6 7 8
Clinical data Kidney CT
IgG (mg/dL), normal value (800-1600 mg/dL)
Low C3 SCr IgG4 (mg/dL), (mg/dL) normal value (48-105 mg/dL)
Proteinuria (g/d)
Salivary gland swelling
Others
M M F M M M M M
Attenuation Attenuation Swelling NE NE Nodular Swelling Attenuation
2947 3570 1706 6328 4110 3019 1569 3935 3398 ± 1510
2120 602 142 781 1340 936 300 185 801 ± 671
+ + NE NE + NE + −
1.9 1.2 1.05 3.56 2.2 0.84 1.1 6.17 2.3 ± 1.8
0.15 0.4 14 (NS) 1.1 0.55 − − 1.38 0.72 ± 0.5
+ + − − − − − −
DM
F F M M F M M M
Swelling NE Attenuation Swelling Nodular Normal Nodular Nodular
3143 2864 5040 3639 2850 4292 5510 4027 3920 ± 991
1520 905 1780 2110 1470 260 980 1510 1311 ± 579
− − − − − + NE +
2.7 1.38 1.8 2.2 1.15 1.4 1.3 2.29 1.8 ± 0.6
0.93 0.21 0.49 − − 0.67 − 0.65 0.65 ± 0.6
+ − + − − − − +
Age Sex
AIP 45 59 60 67 69 72 75 78
Without AIP 9 49 10 54 11 54 12 56 13 58 14 64 15 64 16 68
Lung tumor M-protein
HD
Liver tumor
Hydronephrosis
Abbreviations: NE, not examined; SCr, serum creatinine; DM, diabetes mellitus; HD, hemodyalisis; NS, nephrotic syndrome. NOTE. IgG4 levels in 3 cases among AIP group were not examined at biopsy because of poststeroid therapy for AIP.
proteinuria (N400 mg/d) was found in 10, including 1 case with nephrotic syndrome, as identified by the presence of membranous glomerulonephritis. Abnormal kidney CT, including swelling and a nodular or an irregular pattern, was found in 12 (92%) of 13 cases examined (Fig. 1). Extrarenal comorbidities were composed of AIP (8 cases), diabetes mellitus (2 cases), and hydronephrosis caused by peritoneal fibrosis (1 case). Salivary gland abnormalities were found in 7 cases, with swelling (4 cases) or accumulation on
gallium scintigraphy without noticeable swelling (1 case). CT-detected pseudotumors were found in 2 cases, in the lung and the liver, respectively. The liver biopsy showed accumulation of IgG4-positive plasma cells in the tumor. Serum IgG and IgG4 levels were measured in all cases, revealing high serum IgG levels (mean, 3659.3 mg/dL) and high serum IgG4 levels (mean, 1162.9 mg/dL). A serum hypo-C3 level was found in 6 of 12 cases examined. When we stratified the cases into 2 groups, with and without AIP, no statistical difference in IgG, IgG4, serum creatinine level, or proteinuria between the 2 groups was noted, as shown in Table 1.
3.2. Pathologic findings Morphological findings are summarized in Table 2 (LM, 16 cases), Table 3 (EM, 10 cases), and Table 4 (immunohistochemistry, 16 cases).
Fig. 1 Representative enhanced CT scan in case 6 reveals multiple nodular lesions with frequent effacement in the bilateral kidneys. Note that the enhanced area includes both the cortex and medulla. Retroperitoneal fibrosis is also present.
3.2.1. Tubulointerstitium The most distinctive and common histology observed in all samples by LM was interstitial inflammatory cell infiltration and fibrosis (Fig. 2A), with the 2 often admixed as various features. The margins of such interstitial lesions were relatively well defined (Fig. 2B). Although generally unusual for TIN, interstitial fibrosis in this disease involved the deep renal medulla (Fig. 2C). In addition, fibrosis extended to the capsule and perirenal areas in 6 cases, occasionally spreading into and effacing the retroperitoneal fibrosis (Fig. 2D). In the areas of interstitial fibrosis,
Tubulointerstitial nephritis in IgG4-related disease Table 2 Case
Light microscopic findings
Lymphoplasmacytic infiltration
With AIP 1 ++ 2 +++ 3 +++ 4 +++ 5 ++ 6 ++ 7 +++ 8 ++ Without AIP 9 +++ 10 +++ 11 ++ 12 ++ 13 +++ 14 +++ 15 + 16 +++
Table 3
539
Eosinophilia
Interstitial fibrosis
Stages
Bird's eye
Deposition
++ ++ ++ ++ + − ++ −
+++ ++ +++ +++ +++ ++ + +++
++ ++ + − + ++ + ++
+++ +++ +++ +++ ++ +++ + +++
Other infiltration
Tubular wall
Bowman capsule
Perirenal
Medulla
B+C A+B+C A+B+C B+C C+D A+B+C A+B C+D
+++ ++ +++ +++ ++ ++ + ++
− + + − + − − +
− − − − − − − +
− − − − − + − −
− − − + − − − +
C A+C C+D C+D B+C B+C A+B+C B+C
+++ +++ ++ +++ ++ +++ + ++
++ ++ ++ − + + − +
+ ++ + − − + − −
− + − − + − + ++
− − − + − − − −
Electron microscopic findings
Case
Pericellular fibrosis Epi
With AIP 1 + 2 ++ 3 + 4 ++ 5 ++ 8 ++ Without AIP 9 ++ 10 − 12 + 14 ++
Pericellular fibrosis
Im
Mes
Glomerulus
Interstitium
TBM
Bowman's capsule
Vascular
− − ++ − − +
− − − − − −
− − − − − −
+ + − ++ ++ −
++ ++ ++ +++ ++ +
− − ++ − − ++
− − + − + −
− − − ++
− − − ++
− − − ++
+ − + +
+ ++ + ++
− ++ − +
− + − +
Abbreviations: Epi, subepithelium; Im, intramembrane; Mes, mesangium; TBM, tubular basement membrane.
Table 4
Immunofluorescence findings
Case
IgG
IgG1
IgG2
IgG3
IgG4
IgA
IgM
κ/λ
C3
C4
C1q
1 9 10 14 15
G, B, T, I T, I G, B, T, I B, T, I T, I
G, B, T, I T, I G, B, T, I B, T, I T, I
None None G, B, T, I None None
None T, I G, B, T, I B, T, I None
G, B, T, I B, T, I G, B, T, I B, T, I T, I
None None None T, I None
None None None None None
G, B, T, I T, I G, B, T, I G, B, T, I T, I
None B, T, I B, T T, I None
None None B, T None None
None None B, T T, I None
Abbreviations: G, glomerular basement membrane; B, Bowman capsule; T, TBM; I, interstitium.
540
Y. Yamaguchi et al.
A
B
C
D
Fig. 2 Representative light microscopic findings of renal biopsy. (A), Marked tubular attenuation with diffuse interstitial infiltration and fibrosis (PAS staining). This is the most common histologic feature noted in our series. (B), Focal tubulointerstitial involvement in the cortex and medulla showing clear demarcation (dashed line) between involved and uninvolved areas (PAS). (C), Extension of fibrosis into the deep medulla. Note tubular attenuation and focal inflammatory cell accumulation on a background of connective tissue deposition (Massons trichrome stain). (D), Subcapsular and perirenal involvement (upper) in a case with retroperitoneal fibrosis. Occasional lymph follicular features in the extrarenal fibrosis are associated (PAS). Original magnification: A, ×20; B, ×10; C, ×20; D, ×20.
Tubulointerstitial nephritis in IgG4-related disease
541
A
B
C
D
E
F
Fig. 3 Various tubulointerstitial changes classified as stages (A to D). (A), Stage A: lymphoid cell–predominant interstitial infiltration with minimal tubulitis (hematoxylin and eosin [H&E] staining). (B), Stage B: plasma cell–predominant infiltration and scattered eosinophils with diffuse connective tissue deposition (H&E). (C), Stage C: small, swollen plasma cell nests encircled by collagenous tissue (PAS). (D), Stage D: diffuse collagen deposition widens the intertubular spaces. Note peculiar features showing nonatrophic tubuli, despite progressive fibrosis. (E), Large magnification of characteristic morphology of (C) showing a bird's-eye pattern (F). Original magnification: A, ×100; B, ×40; C, ×200; D, ×40; E, ×400.
542
Y. Yamaguchi et al.
Tubulointerstitial nephritis in IgG4-related disease inflammatory cell infiltrates often expanded into the spaces between the tubules and peritubular capillaries. The cellular components of inflammatory infiltrates comprised plasma cells, lymphocytes, and variable numbers of eosinophils. The cellular composition of the inflammatory cells altered with progression of the fibrosis (Fig. 3A-D). Initially, when fibrosis is less apparent, lymphocytes as well as plasma cells are the predominant component, whereas plasma cells become more prominent in parallel with growing fibrosis. Inflammatory infiltrates attenuate when fibrosis extends. Fibrosis and inflammatory infiltrates were associated in most cases, but inflammatory infiltrates tended to accumulate at the margins of the solid lesions. Small nests of plasma cells or individual plasma cells were commonly encased by fibrosis. The features resembled a maple wood grain pattern called “bird's eye” (Fig. 3E and F). Such “bird'seye” fibrosis was observed in all cases (100%). Lymphoid follicles were detected in 9 patients (45%; data not shown). Because the degree of fibrosis was variable among the samples, we tentatively classified the stage of fibrosis as follows: stage A, active cellular infiltration with little fibrosis; stage B, active cellular infiltration with mild but distinct interstitial fibrosis; stage C, interstitial fibrosis dominant with mild cellular infiltration; and stage D, advanced interstitial fibrosis with little cellular infiltration (Fig. 4). In 15 cases (94%), different stages could be identified in a single specimen. Given that the interstitium was replaced by fibrosis, peritubular capillary density was markedly reduced. Stages were better characterized by EM (Fig. 5). In the early stages, A to B (Fig. 5A), interstitial fibrosis revealed bundles of fine fibrils that were sparse among the infiltrating cells; however, with more advanced fibrosis, such as in stage C, fibrous bundles encased each interstitial inflammatory cell (Fig. 5B). In the advanced stage D, fibrous bundles became thick, and interstitial infiltrating cells were atrophic (Fig. 5C). Note that fibrils seemed to be low-dense and distinct fibers compared with those of earlier stages. On higher magnification, distinct collagen bundles were observed around the infiltrated fibroblastic cells, which were intermingled with inflammatory cells (Fig. 5D). Additional characteristics noted in our series were nodular hyaline deposits on the tubular basement membrane (TBM) of atrophic tubules, found in 10 cases (63%; Fig. 6A), and under Bowman epithelium in 5 cases (31%). This lesion is sufficiently recognizable by LM, and the deposits are limited to the involved area, as revealed by Masson trichrome staining. By EM, electron-dense deposits were seen in TBMs in 10 of 10 cases analyzed, and in Bowman capsules in 4 cases (Table 3). Deposits were prominent in the TBMs of atrophic
543 tubules with a coarse granular pattern (Fig. 6B). Electrondense deposits in the interstitium were detected in 7 cases. Dense deposits were occasionally noted in the perivascular areas. There were no distinct substructures in the deposits. In glomeruli, subepithelial deposits were detected in 3 cases, including 2 cases of membranous glomerulonephritis. Immunofluorescence analysis was available in 5 cases, and others were examined with peroxidase-antiperoxidase methods in paraffin sections, the latter being inadequate to estimate interstitial immune deposition. In the 5 cases for which IF was available, all showed IgG, IgG1, and IgG4 deposition in the TBM and interstitium. IgG3 was found in 3 cases, whereas IgG2 was noted in 1. C3 was found in 3 cases, in interstitium, Bowman capsule, and the TBM. C4 and C1q were detected in 1 and 2 cases, respectively (Fig. 7). κ or λ light chains were equally stained, suggesting polyclonal immunoglobulin deposition (Table 4 and Fig. 8). The average proportion of IgG1-, IgG2-, IgG3-, or IgG4-positive plasma cells was 24.3%, 4.9%, 22.3%, or 48.5%, respectively, in 16 cases examined (Table 5). In stage A samples, predominant IgG4 immunostaining was evident among infiltrated plasma cells. The IgG subclass population in cases with and without AIP showed no statistical differences. 3.2.2. Glomerulus Glomeruli generally showed minor abnormalities by LM. Glomerular collapse, occasionally showing presumable atubular glomeruli and global sclerosis, was observed in areas of extensive interstitial fibrosis. Two cases had concomitant membranous glomerulonephritis, as revealed by EM. 3.2.3. Vasculature Vascular changes seemed to be associated with interstitial changes. In portions with advanced fibrosis, small arteries or arterioles revealed loss of vascular smooth muscle cells showing mucoid features. Onion skinning was also noted at the perivascular interstitium. Five cases showed phlebitis and/or interlobular venous sclerosis, but phlebosclerosis was not prominent. Cellular infiltration, fibrosis, and rupture of the external elastic lamina around the interlobular arteries and small arteries were observed, suggesting perivasculitis. These vascular changes occasionally comprised medial fibrosis and luminal narrowing.
4. Discussion The present study characterized the pathologic features of 16 cases from across Japan that may prove useful in
Fig. 4 Development of interstitial fibrosis and inflammatory retraction as revealed by the periodic acid methenamine silver stain (PAM stain). (A), Incipient appearance of PAM-positive thin fibers among inflammatory infiltrates corresponding to stage A. (B), Characteristic cell nests and PAM-positive fibrous bundle pattern demonstrate the bird's-eye pattern. (C), Increased interstitial PAM-stained fibrosis and pyknotic cell nuclei with scant cytoplasm. (D), Thick interstitial PAM-stained bundles among tubular spaces. Note that inflammatory infiltrates are minimal and tubuli are not atrophic. Original magnification: A, ×100; B, ×100; C, ×40; D, ×100.
544
Y. Yamaguchi et al.
PC PC
PC
MAC
PC
PC
A
B
FIB PC
PC
FIB
PC
MAC C
D
Tubulointerstitial nephritis in IgG4-related disease distinguishing IgG4-related TIN from other types of TIN. First, a relatively well-demarcated border between involved and uninvolved areas was present. Second, lesions were not restricted to the cortex but often involved the deep medulla and the renal capsule as well. The latter occasionally extended to the peritoneal fibrosis. Third, plasma cell– predominant infiltration was frequently and prominently admixed with fibrosis in varying proportions—even in a single sample—with highest prevalence in the IgG4positive cell fraction. Fourth, bird's-eye fibrosis was observed in all cases and is unique to this disease. Fifth, deposits in the TBM, Bowman capsule, and interstitium were present and were sufficiently identifiable by LM with periodic acid-Schiff (PAS) and Masson trichrome staining. These characteristics were similarly observed in cases with or without AIP, thus supporting an LM-based differential diagnosis from other types of TIN. Further studies testing interobserver validation and reproducibility should be done. Although glomerular changes are sometimes associated with TIN in IgG4-related nephropathy [12,13,16,20], the histologic findings in our series showed that the tubulointerstitial changes are the main cause of progressive renal dysfunction. Generally, TIN is defined by inflammatory infiltrates in the interstitium with frequent tubulitis [27]. Tubulointerstitial inflammation in IgG4-related TIN is composed mainly of lymphocytes and plasma cells with various numbers of eosinophils, but neutrophil infiltration is unremarkable [1,9,10,17,24]. Notably, the cases in our series very seldom manifested tubulitis, even in stage A, showing predominant inflammatory cell infiltration. In AIP, plasma cell–predominant infiltration, particularly frequent IgG4-positive cell infiltration, is an important diagnostic feature [2-4]. Likewise, IgG4-positive cells in the interstitium are typical features, provoking TIN [10-14,17-22]. Immunohistochemistry for IgG subclass–positive cells showed an elevation of the IgG4-positive cell population to 48.5%, on average. Interestingly, IgG4-positive cells remained as interstitial fibrosis progressed in stages C and D. Although IgG4positive cells are substantially increased, IgG3-positive cells are also increased relative to normal kidney tissue, in which IgG3-positive cells are largely absent. This is the first report on IgG subclass populations in IgG4-related TIN; an involvement of other IgG subclass–positive cells in this type of TIN needs further investigation. Steroid treatment has been reported to be effective in IgG4-related TIN. Although follow-up data are not shown
545
A
TC PC
PC
B Fig. 6 Tubulointerstitial depositions. (A), Apparent deposition (red) in the thickened TBM by Masson trichrome staining. (B), Granular electron-dense deposits (arrow) are accumulated in the thickened TBM but not apparent in the tubular cytoplasm. Note that the scattered particles of deposits are also observed in the interstitium. Original magnification: A, ×400; B, ×100. PC, plasma cell; TC, tubular cell.
in the present article, serial biopsy in case 15 revealed decreased IgG4-positive cell infiltration (from 41.5% to 18.2%) with progressive fibrosis after steroid treatment. However, IgG4-positive cell infiltration increased again (to 57%) after clinical recurrence after steroid discontinuation. More interestingly, reappearance of the bird's-eye pattern accompanied this change (data not shown). These findings imply that the various histologic features categorized by us as stages A to D reflect different disease stages, rather than histologic variations. In addition, we surmise that serum
Fig. 5 Electron microscopic findings of different stages of fibrosis. (A), Stage A: thin fibrous deposition among lymphoplasmacytoid cells. (B), Stage B: fibrous deposition has become more extensive, and plasma cells and eosinophils have abundant cytoplasm with granules. (C), Stage C: deposition of intercellular fibrous bundles is thicker, and numbers of mononuclear cells are reduced and atrophic. (D), The presence of fibroblasts among the inflammatory cells is accompanied by dense fibrous bundles of interstitial-type collagen (all figures, original magnification ×1000). PC indicates plasma cells; FIB, fibroblasts; MAC, macrophage; arrow, collagen. Original magnification: A, ×2400; B, ×3600; C, ×1200; D, ×4800.
546
Y. Yamaguchi et al.
A
B
Fig. 7 IgG4 immunostaining. (A), IgG4-positive cells among severe lymphoplasmacytic infiltration of Stage A. Note that relatively frequent IgG4-positive cells account for appro×imately 40% of the total infiltrating cells. (B), Larger magnification of IgG4-positive cells showing predominant cytoplasmic localization. Original magnification: A, ×100; B, ×400.
IgG4 levels may reflect a systemic increase of IgG4-positive cells because of a steroid-induced attenuating disease activity owing to suppression of IgG4-positive cells. The accumulation of rebiopsied cases with steroid treatment is needed to investigate this hypothesis. The kidney is one of the susceptible organs in IgG4related disease, although the underlying mechanisms are still unknown. Given the elevated plasma IgG4 levels, significant elevation of IgG4-positive cells, and electron-dense deposition in the pancreas, Deshpande et al [24] suggested an immune complex–mediated mechanism in AIP. Similarly, in the kidney, Cornell et al [10] found deposits in the TBM as well as in glomeruli in some cases, also suggestive of an immune complex disease. This is consistent with previous reports showing an association of the circulatory immune complex with nephropathy [13,19,28]. Our series of cases revealed interstitial deposition in all cases in which EM was performed. However, it is still unclear whether interstitial deposits are immune complexes. Complement activation is one factor suggesting that deposition results from immune complex formation. Our IF study showed that IgG4 deposition in Bowman capsule, TBM, and interstitium was associated with C3 deposition. In addition, IgG4 deposition was associated with IgG and IgG1, and occasionally with IgG3. Notably, IgG4 alone has been reported to insufficiently activate the complement cascade [28]. Thus, it may be that complement activation is based on the presence of IgG1 or IgG3, but not IgG4.
The mechanism of intrarenal IgG4 deposition remains unexplained. Curiously, we noticed that deposits are limited to the affected tubulointerstitium, where IgG4-positive cells are present, but are not observed in the uninvolved portions. In addition, deposition is histologically associated with plasma cell infiltration. This suggests that interstitial IgG4 deposits may be derived from local production. Because IgG4 per se may not filtered in the glomeruli and cases with these deposits are not associated with glomerulopathy, tubular deposition may not be derived from IgG4 filtered in glomeruli and absorbed by the tubuli or Bowman epithelium. Accumulation of IgG4 in the Bowman capsule and TBM may require another mechanism. Although interstitial fibrosis is associated with plasma cell infiltration as well as deposition, the present study cannot determine how deposition or plasma cell infiltration promotes progressive fibrosis. Characteristically observed as a bird's-eye pattern, fibrosis commences in the active inflammatory stage A as revealed by EM. In addition, fibrosis grows in the intercellular spaces among viable inflammatory cells. Electron microscopy in the present study clearly identified fibroblastic or myofibroblastic cells in interstitial-type collagen in both the cytoplasm and pericellular spaces. This suggests that fibroblasts are the culprits causing fibrosis. Interstitial fibrosis in typical TIN is based on tubulitis; however, it is unusual in IgG4-related TIN. This implies that progressive interstitial fibrosis in IgG4-related TIN may be mechanistically different from
Tubulointerstitial nephritis in IgG4-related disease
547
IgG
IgG3
IgG4
C3
C4
C1q
Fig. 8 Immunofluorescence. Serial sections from case 10. In the upper panel, IgG, IgG3, and IgG4 reveal similar distributions. Notably, strong accumulation was observed in the Bowman capsule and TBM. Interstitial deposits are also observed. Lower panel shows complement deposition. Granular pattern deposition of C3 is found in some TBM and interstitium. C4 and C1q are faint but distinctly observed in TBM, but not in interstitium. Magnification: ×200.
that in TIN in general. As shown in our study, not only IgG4- but also IgG1- and IgG3-positive cells are substantially present in the interstitium. The local immune network, not restricted to IgG4, may be involved in fibrosis. Further study is required to identify the mechanism of fibrosis in IgG4-related TIN, particularly from the perspective of therapeutic strategy. IgG4-related disease is a systemic disease, and TIN is not always associated with AIP. In particular, cases with abnormal kidney CT without AIP are sometimes misdiagnosed as renal tumors, and nephrectomy is performed. In such cases, pathologic evaluation only shows TIN or features of inflammatory pseudotumors. In this regard, the present study describes the characteristic histopathology of IgG4-related TIN, which is useful for differentiating TIN in general as well as making a diagnosis of TIN lesions in
nontumor cases. In addition, we provide the first evidence that renal pathology in IgG4-related TIN with or without AIP is otherwise identical. Our observations should benefit differential diagnosis using renal biopsy, which is used in a variety of diseases.
5. Conclusion This series of IgG4-related nephropathy cases in Japan show its distinctive pathology in relation to TIN. The features may be useful in identifying cases with IgG4related nephropathy that are not associated with AIP or with systemic sclerosing diseases. The renal histology of IgG4-related nephropathy will be better clarified by accumulating more cases, particularly with repeat biopsies,
548 Table 5 Case
Y. Yamaguchi et al. Percent positive cells in IgG subclasses staining IgG1
With AIP 1 28.2 2 22.1 3 32.1 4 18.4 5 13.3 6 33.4 7 42.3 8 23.0 27.1 ± 9.3 Without AIP 9 16.9 10 22.8 11 33.7 12 11.0 13 15.7 14 32.3 15 20.5 16 23.0 22.0 ± 7.9
IgG2
IgG3
IgG4
11.6 0.1 0.9 0.0 8.1 8.6 8.1 7.8 5.3 ± 4.6
16.9 31.4 26.5 32.4 32.2 11.4 8.1 28.5 22.7 ± 9.8
43.3 46.4 40.5 49.2 46.4 46.6 41.4 40.5 44.8 ± 3.3
1.0 0.0 11.0 1.1 4.1 0.0 0.0 15.7 4.1 ± 6.0
32.9 23.0 10.0 24.2 28.9 26.2 4.7 19.0 21.1 ± 9.5
49.2 54.2 45.3 63.7 51.3 41.5 74.8 42.3 52.8 ± 11.4
and the actual pathogenetic mechanisms of IgG4-related disease will also require further attention.
Acknowledgment The present study was made possible through case submissions and discussion by members of the Japanese study group on IgG4-related nephropathy. We thank Drs Kunio Morozumi, Asami Takeda (Nagoya Daini Red Cross Hospital), Kousaku Nitta (Tokyo Women's Medical University), Yasunori Utsunomiya (The Jikei University), Kenjiro Kimura (St.Mariana Medical University), Takako Saeki (Nagaoka Red Cross Hospital), Naobumi Mise (Mitsui Memorial Hospital), Shinichiro Tsunoda (Yokohama Sakaekyousai Hospital), Yasuhiro Amagasaki and Nobuko Nakamura (Saiseikai Yokohamashi Nanbu Hospital), and Akira Shimizu (Nippon Medical School) for submitting the cases and critical discussion. We also thank Shigeru Horita, Hideki Nakayama, and Mayuko Ohno, (Tokyo Women's Medical University), and Naoshi Imai (Niigata University School of Medicine) for their excellent support in pathologic analyses. The content of the present article has been orally presented in domestic meetings but has not appeared in any formal publication. The English grammar of this document has been checked by at least 2 professional editors, both native speakers of English. For a certificate, please see: http:// www.textcheck.com/certificate/WrzL9l.
References [1] Hamano H, Kawa S, Horiuchi A, et al. High serum IgG4 concentrations in patients with sclerosing pancreatitis. N Engl J Med 2001;344:732-8.
[2] Hamano H, Kawa S, Ochi Y, et al. Hydronephrosis associated with retroperitoneal fibrosis and sclerosing pancreatitis. Lancet 2002;359: 1403-4. [3] Kamisawa T, Funata N, Hayashi Y, et al. A new clinicopathological entity of IgG4-related autoimmune disease. J Gastroenterol 2003;38: 982-4. [4] Finkelberg DL, Sahani D, Deshpande V, Brugge WR. Autoimmune pancreatitis. N Engl J Med 2006;355:2670-6. [5] Kitagawa S, Zen Y, Harada K, et al. Abundant IgG4-positive plasma cell infiltration characterizes chronic sclerosing sialadenitis (Kuttner's tumor). Am J Surg Pathol 2005;29:783-91. [6] Shimoyama K, Ogawa N, Sawaki T, et al. Umehara: a case of Mikulicz's disease complicated with interstitial nephritis successfully treated by high-dose corticosteroid. Mod Rheumatol 2006;16:176-82. [7] Nakamura H, Wada H, Origuchi T, et al. A case of IgG4-related autoimmune disease with multiple organ involvement. Scand J Rheumatol 2006;35:69-71. [8] Saeki T, Saito A, Hiura T, et al. Lymphoplasmacytic infiltration of multiple organs with immunoreactivity for IgG4: IgG4-related systemic disease. Intern Med 2006;45:163-7. [9] Aoki A, Sato K, Itabashi M, et al. A case of Mikulicz's disease complicated with severe interstitial nephritis associated with IgG4. Clin Exp Nephrol 2009;13:367-72. [10] Cornell LD, Chicano SL, Deshpande V, et al. Pseudotumors due to IgG4 immune-complex tubulointerstitial nephritis associated with autoimmune pancreatocentric disease. Am J Surg Pathol 2007;31: 1586-97. [11] Murashima M, Tomaszewski J, Glickman JD. Chronic tubulointerstitial nephritis presenting as multiple renal nodules and pancreatic insufficiency. Am J Kidney Dis 2007;49:7-10. [12] Morimoto J, Hasegawa Y, Fukushima H, Uesugi N, Hisano S, Kaneoka H. Membranoproliferative glomerulonephritis-like glomerular disease and concurrent tubulointerstitial nephritis complicating IgG4-related autoimmune pancreatitis. Intern Med 2009;48:157-62. [13] Katano K, Hayatsu Y, Matsuda T, et al. Endocapillary proliferative glomerulonephritis with crescent formation and concurrent tubulointerstitial nephritis complicating retroperitoneal fibrosis with a high serum level of IgG4. Clin Nephrol 2007;68:308-14. [14] Saida Y, Homma N, Hama H, et al. A case of IgG4-related tubulointerstitial nephritis showing the progression of renal dysfunction after a cure for autoimmune pancreatitis. Nippon Jinzo Gakkai Shi 2010;52:73-9. [15] Rudmik L, Trpkov K, Nash C, et al. Autoimmune pancreatitis associated with renal lesions mimicking metastatic tumours. CMAJ 2006;175:367-9. [16] Saeki T, Imai N, Ito TT, Yamazaki H, Nishi S. Membranous nephropathy associated with IgG4-related systemic disease and without autoimmune pancreatitis. Clin Nephrol 2009;71:173-8. [17] Saeki T, Saito A, Yamazaki H, et al. Tubulointerstitial nephritis associated with IgG4-related systemic disease. Clin Exp Nephrol 2007;11:168-73. [18] Saeki T, Nishi S, Yamazaki H, et al. Renal lesions in IgG4-related systemic disease. Intern Med 2007;46:1365-71. [19] Takeda S, Haratake J, Kasai T, Takaeda C, Takazakura E. IgG4associated idiopathic tubulointerstitial nephritis complicating autoimmune pancreatitis. Nephrol Dial Transplant 2004;19:474-6. [20] Uchiyama-Tanaka Y, Mori Y, Kimura T, et al. Acute tubulointerstitial nephritis associated with autoimmune-related pancreatitis. Am J Kidney Dis 2004;43:18-25. [21] Yoneda K, Murata K, Katayama K, et al. Tubulointerstitial nephritis associated with IgG4-related autoimmune disease. Am J Kidney Dis 2007;50:455-62. [22] Watson SJ, Jenkins DA, Bellamy CO. Nephropathy in IgG4-related systemic disease. Am J Surg Pathol 2006;30:1472-7. [23] Naitoh I, Nakazawa T, Ohara H, et al. Autoimmune pancreatitis associated with various extrapancreatic lesions during a long-term clinical course successfully treated with azathioprine and corticosteroid maintenance therapy. Intern Med 2009;48:2003-7.
Tubulointerstitial nephritis in IgG4-related disease [24] Deshpande V, Chicano S, Finkelberg D, et al. Autoimmune pancreatitis: a systemic immune complex mediated disease. Am J Surg Pathol 2006;30:1537-45. [25] Saeki T, Nishi S, Imai N, et al. Clinicopathological characteristics of patients with IgG4-related tubulointerstitial nephritis. Kidney Int 2010;78:1016-23. [26] Otsuki M, Chung JB, Okazaki K, et al. Research Committee of Intractable Pancreatic Diseases provided by the Ministry of Health, Labour and Welfare of Japan and the Korean Society of Pancreatobiliary Diseases: Asian diagnostic criteria for autoimmune pancreatitis:
549 consensus of the Japan-Korea Symposium on Autoimmune Pancreatitis. J Gastroenterol 2008;43:403-8. [27] D'Agati VD, Charles Jennette J, Silva FG. Noninfectious tubulointerstitial nephritis. In: King DW, editor. Atlas of nontumor pathology First Series Fascicle 4 Non-Neoplastic Kidney Diseases. Maryland: ARP Press; 2005. p. 573-601. [28] Doi T, Kanatsu K, Mayumi M, Hamashima Y, Yoshida H. Analysis of IgG immune complexes in sera from patients with membranous nephropathy: role of IgG4 subclass and low-avidity antibodies. Nephron 1991;57:131-6.
www.elsevier.com/locate/humpath
Original contribution
Characteristic tubulointerstitial nephritis in IgG4-related disease☆,☆☆ Yutaka Yamaguchi MD a , Yukiko Kanetsuna MD b , Kazuho Honda MD c , Nobuaki Yamanaka MD d , Mitsuhiro Kawano MD e , Michio Nagata MD f,⁎ on behalf of the Japanese study group on IgG4-related nephropathy a
Yamaguchi's Pathology Laboratory, Matsudo, Chiba, 270-2231 Japan Department of Pathology, The Jikei University Kashiwa Hospital, Kashiwa, Chiba, 277-8567 Japan c Department of Pathology, Tokyo Women's Medical University, Tokyo, 162-8666 Japan d Tokyo Kidney Research Institute, Tokyo, 113-8602 Japan e Division of Rheumatology, Department of Internal Medicine, Graduate School of Medical Science, Kanazawa University, Kanazawa, 920-8641 Japan f Department of Kidney and Vascular Pathology, Graduate School of Comprehensive Human Sciences, University of Tsukuba, Tsukuba, 305-8575 Japan b
Received 31 March 2011; revised 28 May 2011; accepted 1 June 2011
Keywords: IgG4-related disease; Tubulointerstitial nephritis; Tubulointerstitial deposition; Interstitial fibrosis/sclerosis
Summary Nephropathy associated with IgG4-related disease is characterized by tubulointerstitial nephritis. To better identify its pathology, the present study analyzed clinicopathologic features of IgG4-related tubulointerstitial nephritis cases from across Japan. Sixteen cases were identified as IgG4-related nephropathy using the criterion of high serum IgG4 levels (N135 mg/dL) with abnormal kidney computed tomography or elevated serum creatinine levels. Male predominance (75%) and advanced age (average, 62.0 years) were noted. Eight cases displayed no autoimmune pancreatitis. Renal computed tomography abnormalities were found in 12 of 13 cases examined. Renal dysfunction was found in 15 of 16 cases at biopsy. Distinctive features of tubulointerstitial lesions included (1) well-demarcated borders between involved and uninvolved areas; (2) involvement of the cortex and medulla, often extending beyond the renal capsule and with occasional extension to retroperitoneal fibrosis; (3) interstitial inflammatory cells comprising predominantly plasma cells and lymphocytes, with a high prevalence of IgG4-positive cells often admixed with fibrosis; (4) peculiar features of interstitial fibrosis resembling a “bird's-eye” pattern comprising fibrosis among inter–plasma cell spaces; and (5) deposits visible by light and immunofluorescent microscopy in the tubular basement membrane, Bowman capsule, and interstitium that are restricted to the involved portion, sparing normal parts. Ultrastructural analysis revealed the presence of myofibroblasts with intracellular/pericellular collagen accompanied by plasma cell accumulation from an early stage. Histology could not discriminate between IgG4-related tubulointerstitial nephritis with and without autoimmune pancreatitis. In conclusion, the distinctive histologic features of IgG4related tubulointerstitial nephritis can facilitate the differential diagnosis of tubulointerstitial nephritis, even without autoimmune pancreatitis or an abnormal computed tomography suggesting a renal tumor. © 2012 Elsevier Inc. All rights reserved.
☆
Conflict of interest: none. Disclosures: none. ⁎ Corresponding author. E-mail address: [email protected] (M. Nagata). ☆☆
0046-8177/$ – see front matter © 2012 Elsevier Inc. All rights reserved. doi:10.1016/j.humpath.2011.06.002
Tubulointerstitial nephritis in IgG4-related disease
537
1. Introduction
2. Subjects and methods
IgG4-related disease is a systemic disease associated with high serum IgG4 levels that was first identified in Japan and is now emerging worldwide [1-4]. IgG4-related disease encompasses various uncommon clinical manifestations, including autoimmune pancreatitis (AIP), Mikulicz disease, and Küttner disease, despite a relatively common background pathology: interstitial fibrosis/sclerosis and lymphoplasmacytic infiltration [4-9]. IgG4-related disease is associated with high IgG4 serum levels in approximately 70% to 80% of cases. The kidney is an organ particularly affected by IgG4-related disease, and chronic renal failure occasionally develops [10-14]. In practice, IgG4-related nephropathy is occasionally found by abnormal computed tomography (CT), revealing a tumor that is sometimes treated by performing an unnecessary nephrectomy [15]. Among several forms of IgG-related disease in the kidney (IgG4-related nephropathy), previous reports have documented that tubulointerstitial nephritis (TIN) is common and may determine a prognosis [10-12,16-22]. IgG4-related TIN is often accompanied by AIP; however, several cases without AIP have also been reported [6,14,16,18]. Generally, the histology of TIN is fairly nonspecific; thus, the diagnosis of IgG4-related TIN, particularly in cases without AIP, is problematic. Because steroid therapy has been found to be effective in AIP [1,4,23,24], a proper diagnosis of IgG4-related TIN is very important so that steps can be taken to halt disease progression. Several case reports have described the histologic features of IgG4-related TIN. Most were single or a few cases associated with AIP and were subjected to renal biopsy, revealing various features ranging from mild interstitial inflammation to diffuse tubulointerstitial fibrosis [10-14,17-22]. Some reports have demonstrated IgG4 immunostaining, but most did not refer to the relationship between the presence of IgG4-positive cells and disease progression because of the small number of cases and the inconsistent severity of the interstitial injury. In particular, whether the different pathologic profiles noted in previous reports are based on different stages of the same disease or are histologic variants has not been determined. Recently, Saeki et al [25] documented clinicopathologic characteristics of IgG4-related TIN. Although its pathology was briefly summarized in their publication, detailed morphological considerations, including histologic staging and features of progression, were not well discussed. The present study analyzed 16 cases of IgG4-related TIN from across Japan (including 4 cases [cases 5, 8, 11,and 13] that appeared in the report of Saeki et al) to better define the clinical and pathologic characteristics of this condition, facilitate its diagnosis, and discuss its pathogenesis.
2.1. Subjects The study group collected 16 cases from across Japan. The diagnostic criterion for IgG4-related nephropathy was the presence of either AIP with renal dysfunction or elevated serum IgG4 levels (N135 mg/dL) without AIP. The diagnosis of AIP was made according to Asian criteria, as previously reported [26]. Patients with clinically suspected Castleman disease or with high titers of anti–double-stranded DNA, anti-Sm, or antiSjogren syndrome A were not included. Renal dysfunction was defined by more than 1.1 mg/dL of serum creatinine. Among the 16 cases, 4 (cases 5, 8, 11, 13) had appeared in the recent report by Saeki et al [25]; the remaining 12 cases were newly enrolled and had not been published. The study was approved by the ethical committees of Kanazawa University.
2.2. Histopathology Renal biopsy samples were processed by standard techniques for light microscopy (LM), immunostaining, and electron microscopy (EM). Immunostaining was performed either with immunofluorescence (IF, 5 cases with frozen sections) or with peroxidase-antiperoxidase methods (11 cases with paraffin sections) using polyclonal antibodies against IgG, IgA, IgM, C3, C1q, and κ and λ light chains (Dako, Kyoto, Japan). EM was performed in 10 cases, using standard techniques.
2.3. Immunohistochemistry for IgG subclass staining IgG subclass staining was performed in all cases by a standard method using an avidin-biotin immunoperoxidase technique with antigen retrieval by microwave. The primary antibodies were mouse monoclonals against each IgG subclass (IgG1, IgG2, IgG3, and IgG4) (Zymed Laboratories Inc, San Francisco, CA), applied at a 1:100 dilution.
2.4. Statistics All data are presented as absolute numbers or means ± SE. Statistical differences between groups were evaluated using Student unpaired t test. P b .05 was deemed to indicate statistical significance.
3. Results 3.1. Clinical features All cases consisted of Japanese patients, most of whom were men (75%) and older than 50 years (average, 62.0 years; Table 1). Renal dysfunction was found in 15 cases (94%), and
538
Y. Yamaguchi et al.
Table 1 Case
With 1 2 3 4 5 6 7 8
Clinical data Kidney CT
IgG (mg/dL), normal value (800-1600 mg/dL)
Low C3 SCr IgG4 (mg/dL), (mg/dL) normal value (48-105 mg/dL)
Proteinuria (g/d)
Salivary gland swelling
Others
M M F M M M M M
Attenuation Attenuation Swelling NE NE Nodular Swelling Attenuation
2947 3570 1706 6328 4110 3019 1569 3935 3398 ± 1510
2120 602 142 781 1340 936 300 185 801 ± 671
+ + NE NE + NE + −
1.9 1.2 1.05 3.56 2.2 0.84 1.1 6.17 2.3 ± 1.8
0.15 0.4 14 (NS) 1.1 0.55 − − 1.38 0.72 ± 0.5
+ + − − − − − −
DM
F F M M F M M M
Swelling NE Attenuation Swelling Nodular Normal Nodular Nodular
3143 2864 5040 3639 2850 4292 5510 4027 3920 ± 991
1520 905 1780 2110 1470 260 980 1510 1311 ± 579
− − − − − + NE +
2.7 1.38 1.8 2.2 1.15 1.4 1.3 2.29 1.8 ± 0.6
0.93 0.21 0.49 − − 0.67 − 0.65 0.65 ± 0.6
+ − + − − − − +
Age Sex
AIP 45 59 60 67 69 72 75 78
Without AIP 9 49 10 54 11 54 12 56 13 58 14 64 15 64 16 68
Lung tumor M-protein
HD
Liver tumor
Hydronephrosis
Abbreviations: NE, not examined; SCr, serum creatinine; DM, diabetes mellitus; HD, hemodyalisis; NS, nephrotic syndrome. NOTE. IgG4 levels in 3 cases among AIP group were not examined at biopsy because of poststeroid therapy for AIP.
proteinuria (N400 mg/d) was found in 10, including 1 case with nephrotic syndrome, as identified by the presence of membranous glomerulonephritis. Abnormal kidney CT, including swelling and a nodular or an irregular pattern, was found in 12 (92%) of 13 cases examined (Fig. 1). Extrarenal comorbidities were composed of AIP (8 cases), diabetes mellitus (2 cases), and hydronephrosis caused by peritoneal fibrosis (1 case). Salivary gland abnormalities were found in 7 cases, with swelling (4 cases) or accumulation on
gallium scintigraphy without noticeable swelling (1 case). CT-detected pseudotumors were found in 2 cases, in the lung and the liver, respectively. The liver biopsy showed accumulation of IgG4-positive plasma cells in the tumor. Serum IgG and IgG4 levels were measured in all cases, revealing high serum IgG levels (mean, 3659.3 mg/dL) and high serum IgG4 levels (mean, 1162.9 mg/dL). A serum hypo-C3 level was found in 6 of 12 cases examined. When we stratified the cases into 2 groups, with and without AIP, no statistical difference in IgG, IgG4, serum creatinine level, or proteinuria between the 2 groups was noted, as shown in Table 1.
3.2. Pathologic findings Morphological findings are summarized in Table 2 (LM, 16 cases), Table 3 (EM, 10 cases), and Table 4 (immunohistochemistry, 16 cases).
Fig. 1 Representative enhanced CT scan in case 6 reveals multiple nodular lesions with frequent effacement in the bilateral kidneys. Note that the enhanced area includes both the cortex and medulla. Retroperitoneal fibrosis is also present.
3.2.1. Tubulointerstitium The most distinctive and common histology observed in all samples by LM was interstitial inflammatory cell infiltration and fibrosis (Fig. 2A), with the 2 often admixed as various features. The margins of such interstitial lesions were relatively well defined (Fig. 2B). Although generally unusual for TIN, interstitial fibrosis in this disease involved the deep renal medulla (Fig. 2C). In addition, fibrosis extended to the capsule and perirenal areas in 6 cases, occasionally spreading into and effacing the retroperitoneal fibrosis (Fig. 2D). In the areas of interstitial fibrosis,
Tubulointerstitial nephritis in IgG4-related disease Table 2 Case
Light microscopic findings
Lymphoplasmacytic infiltration
With AIP 1 ++ 2 +++ 3 +++ 4 +++ 5 ++ 6 ++ 7 +++ 8 ++ Without AIP 9 +++ 10 +++ 11 ++ 12 ++ 13 +++ 14 +++ 15 + 16 +++
Table 3
539
Eosinophilia
Interstitial fibrosis
Stages
Bird's eye
Deposition
++ ++ ++ ++ + − ++ −
+++ ++ +++ +++ +++ ++ + +++
++ ++ + − + ++ + ++
+++ +++ +++ +++ ++ +++ + +++
Other infiltration
Tubular wall
Bowman capsule
Perirenal
Medulla
B+C A+B+C A+B+C B+C C+D A+B+C A+B C+D
+++ ++ +++ +++ ++ ++ + ++
− + + − + − − +
− − − − − − − +
− − − − − + − −
− − − + − − − +
C A+C C+D C+D B+C B+C A+B+C B+C
+++ +++ ++ +++ ++ +++ + ++
++ ++ ++ − + + − +
+ ++ + − − + − −
− + − − + − + ++
− − − + − − − −
Electron microscopic findings
Case
Pericellular fibrosis Epi
With AIP 1 + 2 ++ 3 + 4 ++ 5 ++ 8 ++ Without AIP 9 ++ 10 − 12 + 14 ++
Pericellular fibrosis
Im
Mes
Glomerulus
Interstitium
TBM
Bowman's capsule
Vascular
− − ++ − − +
− − − − − −
− − − − − −
+ + − ++ ++ −
++ ++ ++ +++ ++ +
− − ++ − − ++
− − + − + −
− − − ++
− − − ++
− − − ++
+ − + +
+ ++ + ++
− ++ − +
− + − +
Abbreviations: Epi, subepithelium; Im, intramembrane; Mes, mesangium; TBM, tubular basement membrane.
Table 4
Immunofluorescence findings
Case
IgG
IgG1
IgG2
IgG3
IgG4
IgA
IgM
κ/λ
C3
C4
C1q
1 9 10 14 15
G, B, T, I T, I G, B, T, I B, T, I T, I
G, B, T, I T, I G, B, T, I B, T, I T, I
None None G, B, T, I None None
None T, I G, B, T, I B, T, I None
G, B, T, I B, T, I G, B, T, I B, T, I T, I
None None None T, I None
None None None None None
G, B, T, I T, I G, B, T, I G, B, T, I T, I
None B, T, I B, T T, I None
None None B, T None None
None None B, T T, I None
Abbreviations: G, glomerular basement membrane; B, Bowman capsule; T, TBM; I, interstitium.
540
Y. Yamaguchi et al.
A
B
C
D
Fig. 2 Representative light microscopic findings of renal biopsy. (A), Marked tubular attenuation with diffuse interstitial infiltration and fibrosis (PAS staining). This is the most common histologic feature noted in our series. (B), Focal tubulointerstitial involvement in the cortex and medulla showing clear demarcation (dashed line) between involved and uninvolved areas (PAS). (C), Extension of fibrosis into the deep medulla. Note tubular attenuation and focal inflammatory cell accumulation on a background of connective tissue deposition (Massons trichrome stain). (D), Subcapsular and perirenal involvement (upper) in a case with retroperitoneal fibrosis. Occasional lymph follicular features in the extrarenal fibrosis are associated (PAS). Original magnification: A, ×20; B, ×10; C, ×20; D, ×20.
Tubulointerstitial nephritis in IgG4-related disease
541
A
B
C
D
E
F
Fig. 3 Various tubulointerstitial changes classified as stages (A to D). (A), Stage A: lymphoid cell–predominant interstitial infiltration with minimal tubulitis (hematoxylin and eosin [H&E] staining). (B), Stage B: plasma cell–predominant infiltration and scattered eosinophils with diffuse connective tissue deposition (H&E). (C), Stage C: small, swollen plasma cell nests encircled by collagenous tissue (PAS). (D), Stage D: diffuse collagen deposition widens the intertubular spaces. Note peculiar features showing nonatrophic tubuli, despite progressive fibrosis. (E), Large magnification of characteristic morphology of (C) showing a bird's-eye pattern (F). Original magnification: A, ×100; B, ×40; C, ×200; D, ×40; E, ×400.
542
Y. Yamaguchi et al.
Tubulointerstitial nephritis in IgG4-related disease inflammatory cell infiltrates often expanded into the spaces between the tubules and peritubular capillaries. The cellular components of inflammatory infiltrates comprised plasma cells, lymphocytes, and variable numbers of eosinophils. The cellular composition of the inflammatory cells altered with progression of the fibrosis (Fig. 3A-D). Initially, when fibrosis is less apparent, lymphocytes as well as plasma cells are the predominant component, whereas plasma cells become more prominent in parallel with growing fibrosis. Inflammatory infiltrates attenuate when fibrosis extends. Fibrosis and inflammatory infiltrates were associated in most cases, but inflammatory infiltrates tended to accumulate at the margins of the solid lesions. Small nests of plasma cells or individual plasma cells were commonly encased by fibrosis. The features resembled a maple wood grain pattern called “bird's eye” (Fig. 3E and F). Such “bird'seye” fibrosis was observed in all cases (100%). Lymphoid follicles were detected in 9 patients (45%; data not shown). Because the degree of fibrosis was variable among the samples, we tentatively classified the stage of fibrosis as follows: stage A, active cellular infiltration with little fibrosis; stage B, active cellular infiltration with mild but distinct interstitial fibrosis; stage C, interstitial fibrosis dominant with mild cellular infiltration; and stage D, advanced interstitial fibrosis with little cellular infiltration (Fig. 4). In 15 cases (94%), different stages could be identified in a single specimen. Given that the interstitium was replaced by fibrosis, peritubular capillary density was markedly reduced. Stages were better characterized by EM (Fig. 5). In the early stages, A to B (Fig. 5A), interstitial fibrosis revealed bundles of fine fibrils that were sparse among the infiltrating cells; however, with more advanced fibrosis, such as in stage C, fibrous bundles encased each interstitial inflammatory cell (Fig. 5B). In the advanced stage D, fibrous bundles became thick, and interstitial infiltrating cells were atrophic (Fig. 5C). Note that fibrils seemed to be low-dense and distinct fibers compared with those of earlier stages. On higher magnification, distinct collagen bundles were observed around the infiltrated fibroblastic cells, which were intermingled with inflammatory cells (Fig. 5D). Additional characteristics noted in our series were nodular hyaline deposits on the tubular basement membrane (TBM) of atrophic tubules, found in 10 cases (63%; Fig. 6A), and under Bowman epithelium in 5 cases (31%). This lesion is sufficiently recognizable by LM, and the deposits are limited to the involved area, as revealed by Masson trichrome staining. By EM, electron-dense deposits were seen in TBMs in 10 of 10 cases analyzed, and in Bowman capsules in 4 cases (Table 3). Deposits were prominent in the TBMs of atrophic
543 tubules with a coarse granular pattern (Fig. 6B). Electrondense deposits in the interstitium were detected in 7 cases. Dense deposits were occasionally noted in the perivascular areas. There were no distinct substructures in the deposits. In glomeruli, subepithelial deposits were detected in 3 cases, including 2 cases of membranous glomerulonephritis. Immunofluorescence analysis was available in 5 cases, and others were examined with peroxidase-antiperoxidase methods in paraffin sections, the latter being inadequate to estimate interstitial immune deposition. In the 5 cases for which IF was available, all showed IgG, IgG1, and IgG4 deposition in the TBM and interstitium. IgG3 was found in 3 cases, whereas IgG2 was noted in 1. C3 was found in 3 cases, in interstitium, Bowman capsule, and the TBM. C4 and C1q were detected in 1 and 2 cases, respectively (Fig. 7). κ or λ light chains were equally stained, suggesting polyclonal immunoglobulin deposition (Table 4 and Fig. 8). The average proportion of IgG1-, IgG2-, IgG3-, or IgG4-positive plasma cells was 24.3%, 4.9%, 22.3%, or 48.5%, respectively, in 16 cases examined (Table 5). In stage A samples, predominant IgG4 immunostaining was evident among infiltrated plasma cells. The IgG subclass population in cases with and without AIP showed no statistical differences. 3.2.2. Glomerulus Glomeruli generally showed minor abnormalities by LM. Glomerular collapse, occasionally showing presumable atubular glomeruli and global sclerosis, was observed in areas of extensive interstitial fibrosis. Two cases had concomitant membranous glomerulonephritis, as revealed by EM. 3.2.3. Vasculature Vascular changes seemed to be associated with interstitial changes. In portions with advanced fibrosis, small arteries or arterioles revealed loss of vascular smooth muscle cells showing mucoid features. Onion skinning was also noted at the perivascular interstitium. Five cases showed phlebitis and/or interlobular venous sclerosis, but phlebosclerosis was not prominent. Cellular infiltration, fibrosis, and rupture of the external elastic lamina around the interlobular arteries and small arteries were observed, suggesting perivasculitis. These vascular changes occasionally comprised medial fibrosis and luminal narrowing.
4. Discussion The present study characterized the pathologic features of 16 cases from across Japan that may prove useful in
Fig. 4 Development of interstitial fibrosis and inflammatory retraction as revealed by the periodic acid methenamine silver stain (PAM stain). (A), Incipient appearance of PAM-positive thin fibers among inflammatory infiltrates corresponding to stage A. (B), Characteristic cell nests and PAM-positive fibrous bundle pattern demonstrate the bird's-eye pattern. (C), Increased interstitial PAM-stained fibrosis and pyknotic cell nuclei with scant cytoplasm. (D), Thick interstitial PAM-stained bundles among tubular spaces. Note that inflammatory infiltrates are minimal and tubuli are not atrophic. Original magnification: A, ×100; B, ×100; C, ×40; D, ×100.
544
Y. Yamaguchi et al.
PC PC
PC
MAC
PC
PC
A
B
FIB PC
PC
FIB
PC
MAC C
D
Tubulointerstitial nephritis in IgG4-related disease distinguishing IgG4-related TIN from other types of TIN. First, a relatively well-demarcated border between involved and uninvolved areas was present. Second, lesions were not restricted to the cortex but often involved the deep medulla and the renal capsule as well. The latter occasionally extended to the peritoneal fibrosis. Third, plasma cell– predominant infiltration was frequently and prominently admixed with fibrosis in varying proportions—even in a single sample—with highest prevalence in the IgG4positive cell fraction. Fourth, bird's-eye fibrosis was observed in all cases and is unique to this disease. Fifth, deposits in the TBM, Bowman capsule, and interstitium were present and were sufficiently identifiable by LM with periodic acid-Schiff (PAS) and Masson trichrome staining. These characteristics were similarly observed in cases with or without AIP, thus supporting an LM-based differential diagnosis from other types of TIN. Further studies testing interobserver validation and reproducibility should be done. Although glomerular changes are sometimes associated with TIN in IgG4-related nephropathy [12,13,16,20], the histologic findings in our series showed that the tubulointerstitial changes are the main cause of progressive renal dysfunction. Generally, TIN is defined by inflammatory infiltrates in the interstitium with frequent tubulitis [27]. Tubulointerstitial inflammation in IgG4-related TIN is composed mainly of lymphocytes and plasma cells with various numbers of eosinophils, but neutrophil infiltration is unremarkable [1,9,10,17,24]. Notably, the cases in our series very seldom manifested tubulitis, even in stage A, showing predominant inflammatory cell infiltration. In AIP, plasma cell–predominant infiltration, particularly frequent IgG4-positive cell infiltration, is an important diagnostic feature [2-4]. Likewise, IgG4-positive cells in the interstitium are typical features, provoking TIN [10-14,17-22]. Immunohistochemistry for IgG subclass–positive cells showed an elevation of the IgG4-positive cell population to 48.5%, on average. Interestingly, IgG4-positive cells remained as interstitial fibrosis progressed in stages C and D. Although IgG4positive cells are substantially increased, IgG3-positive cells are also increased relative to normal kidney tissue, in which IgG3-positive cells are largely absent. This is the first report on IgG subclass populations in IgG4-related TIN; an involvement of other IgG subclass–positive cells in this type of TIN needs further investigation. Steroid treatment has been reported to be effective in IgG4-related TIN. Although follow-up data are not shown
545
A
TC PC
PC
B Fig. 6 Tubulointerstitial depositions. (A), Apparent deposition (red) in the thickened TBM by Masson trichrome staining. (B), Granular electron-dense deposits (arrow) are accumulated in the thickened TBM but not apparent in the tubular cytoplasm. Note that the scattered particles of deposits are also observed in the interstitium. Original magnification: A, ×400; B, ×100. PC, plasma cell; TC, tubular cell.
in the present article, serial biopsy in case 15 revealed decreased IgG4-positive cell infiltration (from 41.5% to 18.2%) with progressive fibrosis after steroid treatment. However, IgG4-positive cell infiltration increased again (to 57%) after clinical recurrence after steroid discontinuation. More interestingly, reappearance of the bird's-eye pattern accompanied this change (data not shown). These findings imply that the various histologic features categorized by us as stages A to D reflect different disease stages, rather than histologic variations. In addition, we surmise that serum
Fig. 5 Electron microscopic findings of different stages of fibrosis. (A), Stage A: thin fibrous deposition among lymphoplasmacytoid cells. (B), Stage B: fibrous deposition has become more extensive, and plasma cells and eosinophils have abundant cytoplasm with granules. (C), Stage C: deposition of intercellular fibrous bundles is thicker, and numbers of mononuclear cells are reduced and atrophic. (D), The presence of fibroblasts among the inflammatory cells is accompanied by dense fibrous bundles of interstitial-type collagen (all figures, original magnification ×1000). PC indicates plasma cells; FIB, fibroblasts; MAC, macrophage; arrow, collagen. Original magnification: A, ×2400; B, ×3600; C, ×1200; D, ×4800.
546
Y. Yamaguchi et al.
A
B
Fig. 7 IgG4 immunostaining. (A), IgG4-positive cells among severe lymphoplasmacytic infiltration of Stage A. Note that relatively frequent IgG4-positive cells account for appro×imately 40% of the total infiltrating cells. (B), Larger magnification of IgG4-positive cells showing predominant cytoplasmic localization. Original magnification: A, ×100; B, ×400.
IgG4 levels may reflect a systemic increase of IgG4-positive cells because of a steroid-induced attenuating disease activity owing to suppression of IgG4-positive cells. The accumulation of rebiopsied cases with steroid treatment is needed to investigate this hypothesis. The kidney is one of the susceptible organs in IgG4related disease, although the underlying mechanisms are still unknown. Given the elevated plasma IgG4 levels, significant elevation of IgG4-positive cells, and electron-dense deposition in the pancreas, Deshpande et al [24] suggested an immune complex–mediated mechanism in AIP. Similarly, in the kidney, Cornell et al [10] found deposits in the TBM as well as in glomeruli in some cases, also suggestive of an immune complex disease. This is consistent with previous reports showing an association of the circulatory immune complex with nephropathy [13,19,28]. Our series of cases revealed interstitial deposition in all cases in which EM was performed. However, it is still unclear whether interstitial deposits are immune complexes. Complement activation is one factor suggesting that deposition results from immune complex formation. Our IF study showed that IgG4 deposition in Bowman capsule, TBM, and interstitium was associated with C3 deposition. In addition, IgG4 deposition was associated with IgG and IgG1, and occasionally with IgG3. Notably, IgG4 alone has been reported to insufficiently activate the complement cascade [28]. Thus, it may be that complement activation is based on the presence of IgG1 or IgG3, but not IgG4.
The mechanism of intrarenal IgG4 deposition remains unexplained. Curiously, we noticed that deposits are limited to the affected tubulointerstitium, where IgG4-positive cells are present, but are not observed in the uninvolved portions. In addition, deposition is histologically associated with plasma cell infiltration. This suggests that interstitial IgG4 deposits may be derived from local production. Because IgG4 per se may not filtered in the glomeruli and cases with these deposits are not associated with glomerulopathy, tubular deposition may not be derived from IgG4 filtered in glomeruli and absorbed by the tubuli or Bowman epithelium. Accumulation of IgG4 in the Bowman capsule and TBM may require another mechanism. Although interstitial fibrosis is associated with plasma cell infiltration as well as deposition, the present study cannot determine how deposition or plasma cell infiltration promotes progressive fibrosis. Characteristically observed as a bird's-eye pattern, fibrosis commences in the active inflammatory stage A as revealed by EM. In addition, fibrosis grows in the intercellular spaces among viable inflammatory cells. Electron microscopy in the present study clearly identified fibroblastic or myofibroblastic cells in interstitial-type collagen in both the cytoplasm and pericellular spaces. This suggests that fibroblasts are the culprits causing fibrosis. Interstitial fibrosis in typical TIN is based on tubulitis; however, it is unusual in IgG4-related TIN. This implies that progressive interstitial fibrosis in IgG4-related TIN may be mechanistically different from
Tubulointerstitial nephritis in IgG4-related disease
547
IgG
IgG3
IgG4
C3
C4
C1q
Fig. 8 Immunofluorescence. Serial sections from case 10. In the upper panel, IgG, IgG3, and IgG4 reveal similar distributions. Notably, strong accumulation was observed in the Bowman capsule and TBM. Interstitial deposits are also observed. Lower panel shows complement deposition. Granular pattern deposition of C3 is found in some TBM and interstitium. C4 and C1q are faint but distinctly observed in TBM, but not in interstitium. Magnification: ×200.
that in TIN in general. As shown in our study, not only IgG4- but also IgG1- and IgG3-positive cells are substantially present in the interstitium. The local immune network, not restricted to IgG4, may be involved in fibrosis. Further study is required to identify the mechanism of fibrosis in IgG4-related TIN, particularly from the perspective of therapeutic strategy. IgG4-related disease is a systemic disease, and TIN is not always associated with AIP. In particular, cases with abnormal kidney CT without AIP are sometimes misdiagnosed as renal tumors, and nephrectomy is performed. In such cases, pathologic evaluation only shows TIN or features of inflammatory pseudotumors. In this regard, the present study describes the characteristic histopathology of IgG4-related TIN, which is useful for differentiating TIN in general as well as making a diagnosis of TIN lesions in
nontumor cases. In addition, we provide the first evidence that renal pathology in IgG4-related TIN with or without AIP is otherwise identical. Our observations should benefit differential diagnosis using renal biopsy, which is used in a variety of diseases.
5. Conclusion This series of IgG4-related nephropathy cases in Japan show its distinctive pathology in relation to TIN. The features may be useful in identifying cases with IgG4related nephropathy that are not associated with AIP or with systemic sclerosing diseases. The renal histology of IgG4-related nephropathy will be better clarified by accumulating more cases, particularly with repeat biopsies,
548 Table 5 Case
Y. Yamaguchi et al. Percent positive cells in IgG subclasses staining IgG1
With AIP 1 28.2 2 22.1 3 32.1 4 18.4 5 13.3 6 33.4 7 42.3 8 23.0 27.1 ± 9.3 Without AIP 9 16.9 10 22.8 11 33.7 12 11.0 13 15.7 14 32.3 15 20.5 16 23.0 22.0 ± 7.9
IgG2
IgG3
IgG4
11.6 0.1 0.9 0.0 8.1 8.6 8.1 7.8 5.3 ± 4.6
16.9 31.4 26.5 32.4 32.2 11.4 8.1 28.5 22.7 ± 9.8
43.3 46.4 40.5 49.2 46.4 46.6 41.4 40.5 44.8 ± 3.3
1.0 0.0 11.0 1.1 4.1 0.0 0.0 15.7 4.1 ± 6.0
32.9 23.0 10.0 24.2 28.9 26.2 4.7 19.0 21.1 ± 9.5
49.2 54.2 45.3 63.7 51.3 41.5 74.8 42.3 52.8 ± 11.4
and the actual pathogenetic mechanisms of IgG4-related disease will also require further attention.
Acknowledgment The present study was made possible through case submissions and discussion by members of the Japanese study group on IgG4-related nephropathy. We thank Drs Kunio Morozumi, Asami Takeda (Nagoya Daini Red Cross Hospital), Kousaku Nitta (Tokyo Women's Medical University), Yasunori Utsunomiya (The Jikei University), Kenjiro Kimura (St.Mariana Medical University), Takako Saeki (Nagaoka Red Cross Hospital), Naobumi Mise (Mitsui Memorial Hospital), Shinichiro Tsunoda (Yokohama Sakaekyousai Hospital), Yasuhiro Amagasaki and Nobuko Nakamura (Saiseikai Yokohamashi Nanbu Hospital), and Akira Shimizu (Nippon Medical School) for submitting the cases and critical discussion. We also thank Shigeru Horita, Hideki Nakayama, and Mayuko Ohno, (Tokyo Women's Medical University), and Naoshi Imai (Niigata University School of Medicine) for their excellent support in pathologic analyses. The content of the present article has been orally presented in domestic meetings but has not appeared in any formal publication. The English grammar of this document has been checked by at least 2 professional editors, both native speakers of English. For a certificate, please see: http:// www.textcheck.com/certificate/WrzL9l.
References [1] Hamano H, Kawa S, Horiuchi A, et al. High serum IgG4 concentrations in patients with sclerosing pancreatitis. N Engl J Med 2001;344:732-8.
[2] Hamano H, Kawa S, Ochi Y, et al. Hydronephrosis associated with retroperitoneal fibrosis and sclerosing pancreatitis. Lancet 2002;359: 1403-4. [3] Kamisawa T, Funata N, Hayashi Y, et al. A new clinicopathological entity of IgG4-related autoimmune disease. J Gastroenterol 2003;38: 982-4. [4] Finkelberg DL, Sahani D, Deshpande V, Brugge WR. Autoimmune pancreatitis. N Engl J Med 2006;355:2670-6. [5] Kitagawa S, Zen Y, Harada K, et al. Abundant IgG4-positive plasma cell infiltration characterizes chronic sclerosing sialadenitis (Kuttner's tumor). Am J Surg Pathol 2005;29:783-91. [6] Shimoyama K, Ogawa N, Sawaki T, et al. Umehara: a case of Mikulicz's disease complicated with interstitial nephritis successfully treated by high-dose corticosteroid. Mod Rheumatol 2006;16:176-82. [7] Nakamura H, Wada H, Origuchi T, et al. A case of IgG4-related autoimmune disease with multiple organ involvement. Scand J Rheumatol 2006;35:69-71. [8] Saeki T, Saito A, Hiura T, et al. Lymphoplasmacytic infiltration of multiple organs with immunoreactivity for IgG4: IgG4-related systemic disease. Intern Med 2006;45:163-7. [9] Aoki A, Sato K, Itabashi M, et al. A case of Mikulicz's disease complicated with severe interstitial nephritis associated with IgG4. Clin Exp Nephrol 2009;13:367-72. [10] Cornell LD, Chicano SL, Deshpande V, et al. Pseudotumors due to IgG4 immune-complex tubulointerstitial nephritis associated with autoimmune pancreatocentric disease. Am J Surg Pathol 2007;31: 1586-97. [11] Murashima M, Tomaszewski J, Glickman JD. Chronic tubulointerstitial nephritis presenting as multiple renal nodules and pancreatic insufficiency. Am J Kidney Dis 2007;49:7-10. [12] Morimoto J, Hasegawa Y, Fukushima H, Uesugi N, Hisano S, Kaneoka H. Membranoproliferative glomerulonephritis-like glomerular disease and concurrent tubulointerstitial nephritis complicating IgG4-related autoimmune pancreatitis. Intern Med 2009;48:157-62. [13] Katano K, Hayatsu Y, Matsuda T, et al. Endocapillary proliferative glomerulonephritis with crescent formation and concurrent tubulointerstitial nephritis complicating retroperitoneal fibrosis with a high serum level of IgG4. Clin Nephrol 2007;68:308-14. [14] Saida Y, Homma N, Hama H, et al. A case of IgG4-related tubulointerstitial nephritis showing the progression of renal dysfunction after a cure for autoimmune pancreatitis. Nippon Jinzo Gakkai Shi 2010;52:73-9. [15] Rudmik L, Trpkov K, Nash C, et al. Autoimmune pancreatitis associated with renal lesions mimicking metastatic tumours. CMAJ 2006;175:367-9. [16] Saeki T, Imai N, Ito TT, Yamazaki H, Nishi S. Membranous nephropathy associated with IgG4-related systemic disease and without autoimmune pancreatitis. Clin Nephrol 2009;71:173-8. [17] Saeki T, Saito A, Yamazaki H, et al. Tubulointerstitial nephritis associated with IgG4-related systemic disease. Clin Exp Nephrol 2007;11:168-73. [18] Saeki T, Nishi S, Yamazaki H, et al. Renal lesions in IgG4-related systemic disease. Intern Med 2007;46:1365-71. [19] Takeda S, Haratake J, Kasai T, Takaeda C, Takazakura E. IgG4associated idiopathic tubulointerstitial nephritis complicating autoimmune pancreatitis. Nephrol Dial Transplant 2004;19:474-6. [20] Uchiyama-Tanaka Y, Mori Y, Kimura T, et al. Acute tubulointerstitial nephritis associated with autoimmune-related pancreatitis. Am J Kidney Dis 2004;43:18-25. [21] Yoneda K, Murata K, Katayama K, et al. Tubulointerstitial nephritis associated with IgG4-related autoimmune disease. Am J Kidney Dis 2007;50:455-62. [22] Watson SJ, Jenkins DA, Bellamy CO. Nephropathy in IgG4-related systemic disease. Am J Surg Pathol 2006;30:1472-7. [23] Naitoh I, Nakazawa T, Ohara H, et al. Autoimmune pancreatitis associated with various extrapancreatic lesions during a long-term clinical course successfully treated with azathioprine and corticosteroid maintenance therapy. Intern Med 2009;48:2003-7.
Tubulointerstitial nephritis in IgG4-related disease [24] Deshpande V, Chicano S, Finkelberg D, et al. Autoimmune pancreatitis: a systemic immune complex mediated disease. Am J Surg Pathol 2006;30:1537-45. [25] Saeki T, Nishi S, Imai N, et al. Clinicopathological characteristics of patients with IgG4-related tubulointerstitial nephritis. Kidney Int 2010;78:1016-23. [26] Otsuki M, Chung JB, Okazaki K, et al. Research Committee of Intractable Pancreatic Diseases provided by the Ministry of Health, Labour and Welfare of Japan and the Korean Society of Pancreatobiliary Diseases: Asian diagnostic criteria for autoimmune pancreatitis:
549 consensus of the Japan-Korea Symposium on Autoimmune Pancreatitis. J Gastroenterol 2008;43:403-8. [27] D'Agati VD, Charles Jennette J, Silva FG. Noninfectious tubulointerstitial nephritis. In: King DW, editor. Atlas of nontumor pathology First Series Fascicle 4 Non-Neoplastic Kidney Diseases. Maryland: ARP Press; 2005. p. 573-601. [28] Doi T, Kanatsu K, Mayumi M, Hamashima Y, Yoshida H. Analysis of IgG immune complexes in sera from patients with membranous nephropathy: role of IgG4 subclass and low-avidity antibodies. Nephron 1991;57:131-6.