Published ahead of print on January 17, 2007
Clin J Am Soc Nephrol 2: 222-230, 2007
© 2007 American Society of Nephrology
doi: 10.2215/CJN.01790506
Granulomatous Interstitial Nephritis
Nicola Joss*,
Scott Morris
,
Barbara Young
, and
Colin Geddes*
* Renal Unit; Pathology Department, Western Infirmary; and Renal Unit, Glasgow Royal Infirmary, Glasgow, Scotland
Address correspondence to: Dr. Nicola Joss, Renal Unit, Western Infirmary, Glasgow, G11 6NT, Scotland. Phone: +44-141-211-2000; Fax: +44-141-211-1711; E-mail: nicola.joss{at}northglasgow.scot.nhs.uk
 |
Abstract
|
|---|
Granulomatous interstitial nephritis (GIN) is a rare histologic diagnosis. This series reports the presenting features, associated conditions, treatment, and outcome of patients with a diagnosis of GIN in Glasgow during a 15-yr period and compares this with the available literature. Eighteen cases were identified: Five cases were associated with sarcoidosis, two were associated with tubulointerstitial nephritis and uveitis, two were associated with medication, and nine were idiopathic. Patients presented with advanced renal failure (median estimated creatinine clearance 21 ml/min) and minimal proteinuria (urine albumin-to-creatinine ratio 9.9 mg/mmol). Sixteen patients were treated with prednisolone for a mean of 25 mo. Six patients relapsed with reduction in prednisolone dosage, and four patients required steroid-sparing agents. During the mean follow-up of 45 mo, renal function improved or stabilized in 17 patients; the rate of improvement in renal function was most marked in the first year after diagnosis with a gain in function of +1.9 ml/min per mo. The median estimated creatinine clearance at final visit was 56 ml/min. One patient required renal replacement therapy at diagnosis but recovered renal function with treatment. No patient required long-term renal replacement therapy. There was no correlation between the degree of fibrosis or inflammation on biopsy and renal outcome, and the features on biopsy did not help to determine the cause of GIN. GIN is a treatable cause of renal failure that highlights the value of renal biopsy in patients who present with renal failure even when there is minimal proteinuria. The rarity of GIN demonstrates the need for systematic data collection.
|
Introduction
|
|---|
Granulomatous interstitial nephritis (GIN) is a rare histologic diagnosis that is present in between 0.5 and 0.9% of native renal biopsies (1,2) and 0.6% of renal transplant biopsies (3). GIN has been associated with medication, infections, sarcoidosis, crystal deposits, paraproteinemia, and Wegener’s granulomatosis and also is seen in an idiopathic form. Medicines implicated include anticonvulsants, antibiotics, nonsteroidal anti-inflammatory drugs, allopurinol, and diuretics. Mycobacteria and fungi are the main infective causes and seem to be the main causative factor in cases in renal transplants (3,4). Few studies have analyzed the natural history and treatment of GIN; the largest included only seven patients (5). The purpose of this study was to analyze clinical features and outcome of patients who had GIN and presented in Glasgow between 1990 and 2004 and to compare our data with the four largest published series (1,5–7).
|
Materials and Methods
|
|---|
Patients with a diagnosis of GIN were identified by a search of electronic patient records of all patients who have attended Glasgow renal units and associated satellite clinics since 1990. Hemoglobin, serum creatinine, serum calcium concentrations, peripheral blood eosinophil count, and quantified proteinuria at diagnosis were retrieved from the electronic patient records along with demographic data. The serum angiotensin-converting enzyme (ACE) concentration and antineutrophil cytoplasmic antibody (ANCA) were obtained if performed. The upper limit of normal for ACE concentration was 88 U/L. Estimated creatinine clearance (ECC) was calculated using the Cockcroft and Gault formula (8), and rate of change in renal function was calculated using linear regression from the slope of the plot of ECC versus time. A minimum of four measurements of ECC was required. Chest radiograph (CXR) reports were obtained.
We divided patients into four groups: Idiopathic, drug related, sarcoidosis, and tubulointerstitial nephritis and uveitis (TINU) on the basis of history and clinical investigations. Therapy that was administered to treat GIN was recorded, and the dosage of prednisolone was calculated as mg/kg per d averaged over the first 3 mo of treatment. A relapse was diagnosed on the basis of deteriorating renal function that responded promptly to an increase in dosage of prednisolone.
Renal biopsies that were reported as GIN by a number of specialist renal pathologists between 1990 and 2004 were retrieved from the archives of the Departments of Pathology, providing a second method of identifying patients. Histologic specimens were reviewed by one member of the group (B.Y.) without previous knowledge of the clinical history at the time of biopsy or of the final clinical diagnosis. When possible, an original hematoxylin and eosin (H&E)-stained slide was assessed along with a trichrome stain. Some cases required additional H&E stains to be prepared because the original was faded or missing. When it was not possible to assess histology on an H&E stained slide, special stains such as periodic acid-Schiff were assessed instead. After identification of granulomata, all biopsies were stained using a Ziehl-Nielsen stain.
GIN was defined as interstitial nephritis in which the inflammatory infiltrate included one or more distinct aggregates of epithelioid cells with or without multinucleate giant cells. Cases in which there was a diffuse granulomatous background but no granulomata as defined were excluded from the analysis. Biopsies were assessed for the following criteria: Presence or absence of a distinct lymphocyte cuff around the granulomata, presence or absence of necrosis within the granulomata, calcification in granulomata, and presence of asteroid bodies. Semiquantitative scores were assigned for the degree of fibrosis and intensity of background inflammation: 1, mild; 2, moderate; and 3, severe. The proportion of eosinophils in the background inflammation and number of giant cells in the granulomata also were scored semiquantitatively: 1, scanty; 2, moderate; and 3, plentiful.
|
Results
|
|---|
Eighteen patients were identified in the 15-yr period. Sixty-one percent were men, and the mean age at diagnosis was 55 yr. Data on individual patients are shown in Table 1.
Laboratory Data
Table 2 highlights data at diagnosis. Patients had advanced renal failure with minimal proteinuria. Only four patients presented with an ECC >50 ml/min, and 11 patients presented with an ECC <25 ml/min. ANCA was negative in all patients.
Causative Factors
Medication Related.
A possible drug cause was identified in two patients; the offending medicine combinations were omeprazole/diclofenac and sulfasalazine/indomethacin. The patient who was on sulfasalazine and indomethacin also had a seronegative arthropathy but normal CXR and serum ACE and calcium concentrations. Only one of these patients had an elevated eosinophil count at 0.8 x 109/L.
Sarcoidosis.
Table 3 summarizes background characteristics of five patients with sarcoidosis. Only one patient had sarcoidosis diagnosed before renal biopsy. This patient presented with lethargy and myalgia and although the CXR was normal was found to have elevated serum creatinine, calcium, and ACE concentrations. Four subsequent patients received a diagnosis of sarcoidosis after a finding of GIN on renal biopsy. Three patients had elevated serum ACE levels and normal CXR, and one of these patients also had hypercalcemia. The fifth patient had hypercalcemia, lymphadenopathy on CXR, but a normal ACE concentration.
Three of the five patients with sarcoidosis developed extrarenal disease: Granulomata were identified in a nasal biopsy in the patient with known sarcoidosis, granulomata subsequently were identified on lymph node biopsy in the patient with the abnormal CXR, and a third patient developed uveitis and a restrictive lung defect during follow-up.
TINU.
A diagnosis of TINU was made in two patients. One patient presented with uveitis and renal failure simultaneously, and the second initially was thought to have idiopathic GIN but subsequently developed uveitis.
Idiopathic.
Nine patients had no obvious contributing factors and were classified as idiopathic. Mean age was 57 yr, and 67% were men. At diagnosis, the median serum creatinine was 344 µmol/L (range 102 to 534), ECC was 25 ml/min (range 10 to 77), and albumin:creatinine ratio was 9.1 mg/mmol (range 6.7 to 29.7). No patient had hypercalcemia, elevated serum ACE, or peripheral eosinophilia. One patient had a combination of membranous nephropathy and GIN with normal CXR and serum ACE and calcium concentration.
Pathology
All 18 patients had a diagnosis of GIN made by renal biopsy. Original slides of the renal biopsy were obtained in 17 cases to allow the pathologist to review and score the tissue. Original slides were not available in one patient who had idiopathic GIN, but on review of the original biopsy report, the diagnosis of GIN was clearcut. All cases had GIN as a primary diagnosis, and cases in which GIN was believed to be secondary to a glomerulonephritis or a vasculitic illness (e.g., Wegener’s granulomatosis) were excluded. One patient who had a crescentic glomerulonephritis and granulomata was excluded. One patient had a coexisting membranous glomerulopathy that was believed to be unrelated to the GIN. No patient had histologic evidence of infection.
The number of cases was too small for statistical analysis of pathologic features between diagnostic groups. Pathologic features are summarized in Table 4, and representative photomicrographs of the main pathologic features are illustrated in Figure 1. Only one patient (drug induced) was considered to have necrosis within a granulomata, and this was very focal. Calcification and asteroid bodies were uncommon, with calcification of granulomata seen in only two cases (one drug induced and one idiopathic; Figure 1A), and a single patient showed asteroid bodies (idiopathic). Most of the patients (15 of 17) had "naked" granulomata without a discrete rim of lymphocytes. Of the two patients with a lymphocyte cuff, one had idiopathic GIN and the other had sarcoidosis.
View larger version (125K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Figure 1. Histologic features of granulomatous interstitial nephritis (GIN). Representative microphotographs of the histologic features assessed granuloma with calcification (A, arrow), granuloma with plentiful multinucleated giant cells (B, arrows), plentiful eosinophils in the background inflammation (C, arrow) and granuloma with central necrosis (D, arrow). Magnifications: x200 in A and C; x100 in B and D (hematoxylin and eosin).
|
|
There was wide variation in degree of interstitial fibrosis. Nine cases were assessed as mild fibrosis, seven as moderate fibrosis, and two as severe, including one case of TINU and one of sarcoidosis. There were no obvious trends in degree of fibrosis between the diagnostic groups.
Background inflammation ranged from mild to severe. Overall, 10 cases were scored as mild, five as moderate, and three as severe. The three severely inflamed cases included one each of idiopathic, sarcoidosis, and TINU. Again, no trends were apparent for the various diagnostic groups.
Giant cells were scanty in 12 cases, moderate in four, and plentiful in two (Figure 1B). Two cases with plentiful giant cells included one case of drug induced and one of sarcoidosis. Eosinophil infiltration was graded as scanty in nine cases, moderate in seven (Figure 1C), and plentiful in two. The two cases with severe infiltration by eosinophils included one drug induced and one TINU.
Treatment
Mean follow-up of patients was 45 mo (range 12 to 131 mo). During this period, 16 patients received treatment with prednisolone with dosages and duration of treatment being variable. The mean starting dosage of prednisolone was 0.56 mg/kg (range 0.25 to 0.96). The mean dosage of prednisolone in the first 3 mo was 0.33 mg/kg per d (range 0.1 to 0.61). The mean length of treatment was 25 mo (range 2 to 82). Six patients relapsed as the prednisolone dosage was reduced; four of these patients had a diagnosis of sarcoidosis, and these four patients now remain on prednisolone long term. One patient had TINU and was able to discontinue prednisolone after 62 mo of treatment, and the final patient had idiopathic GIN.
Four patients received additional treatment, three of whom had a diagnosis of sarcoidosis. Three of these patients received azathioprine and one patient received mercaptopurine as steroid-sparing agents. All four patients had relapsed on reduction of prednisolone dosage.
Patient and Renal Outcome
Renal function improved or stabilized in 17 patients. Median rate of change in renal function was +0.47 ml/min per mo (range +2.32 to –0.25). Median serum creatinine at last clinic visit (12 to 131 mo after presentation) was 159 µmol/L (range 95 to 340) with a median ECC of 54 ml/min (range 24 to 93). Eleven patients improved renal function to an ECC >50 ml/min, whereas only three patients recovered renal function to an ECC >80 ml/min. Only one patient had an ECC <25 ml/min at the last clinic visit. Rate of change in renal function in the first year after diagnosis was +1.9 ml/min per mo (range +4.66 to –1.39) and for the remaining period was +0.11 ml/min per mo (range +0.8 to –1.07). Median serum creatinine at 1 yr was 190 µmol/L (range 126 to 247) with an ECC of 53 ml/min. Five of the 16 patients who had >2 yr of follow-up had slowly progressive renal failure after the initial period of improved renal function in the first year. No patient had a repeat renal biopsy.
One patient required renal replacement therapy at time of presentation. This patient presented with a serum creatinine of 1363 µmol/L and was started on prednisolone and hemodialysis. Her renal function improved and she became dialysis independent 4 mo later. At her last clinic visit, 10 yr after diagnosis, she has a serum creatinine of 170 µmol/L and is no longer on prednisolone. No patient required renal replacement therapy for progressive renal failure. There was only one death in the group. This patient died from pneumonia 4 yr after diagnosis. He had responded to prednisolone treatment, and his serum creatinine fell from 412 to 160 µmol/L.
There was no relationship between the underlying diagnosis and renal outcome. Figure 2 highlights renal function during the first 2 yr according to cause. In each group, the major improvement was seen within 3 mo. The majority of relapses were in patients who had sarcoidosis or TINU. Consequently, patients with sarcoidosis received prednisolone for an average of 36 mo, whereas those with idiopathic GIN or a drug-related GIN received prednisolone for 17 mo and 7 mo, respectively.
View larger version (10K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Figure 2. Estimated creatinine clearance (ECC) according to cause. Data at month 24 exclude two patients with <24 mo of follow-up. *Range 28 to 131 mo, excludes two patients with <24 mo of follow up.
|
|
Figure 3 compares renal outcome with dosage of prednisolone used within the first 3 mo. The median serum creatinine at diagnosis for the four patients who received steroid-sparing agents was 607 µmol/L (range 299-1080), and ECC was 16 ml/min (range 4 to 25). The overall rate of change in renal function was +0.33 ml/min per mo (range +0.55 to –0.06). The rate of change in the first year after diagnosis was +1.9 ml/min per mo (range +3.1 to +1.17) and for the remaining period was –0.02 ml/min per mo (range +0.21 to –0.2). After a mean follow-up of 72 mo, the median serum creatinine was 179 µmol/L (range 155 to 340), and ECC was 37 ml/min (range 27 to 57).
View larger version (10K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Figure 3. ECC according to prednisolone dosage within the first 3 mo (mg/kg per d). Data at month 24 exclude two patients with <4 mo of follow up. *Range 28 to 131 mo, excludes two patients with <24 mo of follow-up.
|
|
There was no difference in renal outcome comparing patients with different degrees of fibrosis or inflammation on renal biopsy. Eight patients with mild fibrosis presented with an ECC of 35 ml/min and improved to 53 ml/min, whereas eight patients with moderate fibrosis presented with an ECC of 19 ml/min and improved to 50 ml/min. Nine patients with mild inflammation presented with an ECC of 35 ml/min that improved to 62 ml/min, whereas six patients who presented with moderate inflammation presented with an ECC of 18 ml/min and improved to 40 ml/min.
Figure 4 compares renal outcome according to age at time of diagnosis. There was a tendency for younger patients to have better renal function at diagnosis. From 3 mo onward, patients who were younger than 60 yr had significantly better renal function. Despite these differences, the rate of improvement in renal function did not differ between those who were younger then 60 (+0.54 ml/min per mo) compared with those who were older than 60 (+0.47 ml/min per mo).
Exclusion of the two patients with drug-induced GIN did not alter the outcome. Renal function improved or stabilized in 15 of the 16 patients. The median serum creatinine at diagnosis was 357 µmol/L (range 102-1363), and ECC was 24 ml/min (range 4 to 61). The rate of change in renal function during the mean 48 mo of follow-up was +0.33 ml/min per mo (range +1.13 to –0.25), the rate of change in the first year after diagnosis was +1.61 ml/min per mo (range +4.66 to –1.39), and the rate of change in the remaining period +0.07 ml/min per mo (range +0.8 to –1.07). At the final visit, the median serum creatinine was 159 µmol/L (range 95 to 340), and ECC was 55 ml/min (range 24 to 94).
|
Discussion
|
|---|
GIN is a rare histologic diagnosis. We report our experience with 18 patients in Glasgow during a 15-yr period. This represents <1% of our native renal biopsies. Patients presented with advanced renal failure and minimal proteinuria. Outcome is favorable with the greatest improvement in renal function seen in the first 3 mo after diagnosis. The observations reinforce the value of renal biopsies in patients with unexplained renal failure. Thirty percent of patients with >2 yr of follow-up had a gradual decline in renal function, but no patient reached end-stage renal failure.
Table 5 compares our series with four clinical series of five or more cases of GIN in the literature (1,5–7). A fifth report identified 12 patients between 1974 and 1994: Sarcoidosis, medication, and infection each accounted for three cases; there was one case of oxalosis and one case of Wegener’s granulomatosis; and the cause in the final case was unknown. The outcome was known in only 50% of patients: Two recovered, one died, and three had chronic renal failure (9). Our series, with 18 cases, is the largest to date. In the previous four clinical reports, drug-related cases were excluded, and cases were described as either idiopathic or in association with sarcoidosis or renal isolated sarcoidosis. This highlights the difficulty with nomenclature; the series of seven patients with idiopathic GIN had three patients with elevated ACE concentrations (5), whereas in the series of five patients who had a diagnosis of isolated sarcoidosis, only one had an elevated ACE (1). We included two patients in the sarcoidosis group with no extrarenal clinical manifestations, one of whom had an elevated ACE and hypercalcemia and the other of whom had an elevated ACE only. Serum ACE concentrations can be normal in sarcoidosis, and false-positive results also occur. It is likely that a spectrum of diseases is involved, ranging from idiopathic GIN to TINU and sarcoidosis, with patients at diagnosis being classified as idiopathic but subsequently developing extrarenal manifestations. This occurred in one patient with TINU and one patient with sarcoidosis in our series. There have been case reports of membranous nephropathy in association with sarcoidosis usually with extrarenal manifestations, hypercalcemia, or elevated ACE (10,11). Our patient with membranous nephropathy had a normal CXR and normal ACE and calcium concentrations and therefore was considered to have idiopathic GIN.
Renal outcome of our patients is better than in the previous reports. At the last visit, the mean ECC was 56 ml/min and median creatinine was 159 µmol/L. Removal of the two patients with drug-induced causes did not alter these results. Younger patients recovered renal function to a greater degree, perhaps reflecting the effect of age-related changes in the tubulointerstitium in older patients. Our series is the first to describe rates of change in renal function. Rate of improvement in renal function during the entire follow-up period was +0.47 ml/min per mo, with only one patient having a progressive decline in renal function at a rate of –0.25 ml/min per mo. The greatest improvement in renal function was seen in the first year, when the median rate of change in renal function was +1.9 ml/min per mo. Improvement did occur after 1 yr, albeit at a slower rate of +0.11 ml/min per mo. Five patients, who were followed up for >2 yr, had a gradual decline in renal function. One patient did require renal replacement therapy at diagnosis but recovered renal function after 4 mo of treatment, and no patient required renal replacement therapy for progressive established renal failure. This is in comparison with previous reports in which two of the 23 patients required renal replacement therapy after 3 and 15 mo of follow-up. No patient in our series had a second renal biopsy. It was shown previously that granulomata can disappear but often are replaced by scar tissue, and this is the likely reason that renal function does not return to normal in all patients.
The pathologic features that were assessed in this study were those that are likely to discriminate from among various diagnostic categories and to predict renal outcome. It is pathologic dogma that asteroid bodies, calcification, and "naked" granulomata are characteristic of sarcoidosis. In this study, only one case had asteroid bodies and calcification, and the final clinical diagnosis in that case was of a drug-induced GIN. Likewise, "naked" granulomata were the rule, with only two cases exhibiting a distinct lymphocyte cuff, one of which was a case of sarcoidosis. All of the other features assessed were distributed equally across the various diagnostic groups and were not helpful in determining the final diagnosis. In particular, the presence of eosinophils did not help with diagnosing drug-induced GIN. It is of interest that there was no difference in renal outcome when the patients with various degrees of inflammation or fibrosis were compared. Therefore, even in the presence of moderate fibrosis, renal prognosis is good.
Only a randomized, controlled study can determine efficacy and optimal dosage of corticosteroids. However, the initial response to treatment and response to relapses are persuasive of the benefit of corticosteroids in this condition. Six patients relapsed in our series, and all responded to an increase in steroid dosage; four subsequently received steroid-sparing agents. Four of the six patients with relapses had sarcoidosis as the underlying causative factor, and one had TINU. In 15 of the 16 patients who received corticosteroids, renal function stabilized or improved. We found no difference in outcome between patients who received >0.3 mg/kg per d averaged over the first 3 mo and those who received <0.3 mg/kg per d. Of the 23 patients in the four previously published series, all received corticosteroids and 20 had improvement in renal function, and there did not seem to be a benefit in using high-dosage corticosteroids. No difference in response to treatment was seen using 0.5 or 1 mg/kg in a previous report (1). No patients in our series had GIN in association with tuberculosis (TB). However, it seems reasonable to make efforts to exclude TB, particularly in areas where TB is more common, before starting prednisolone.
Many questions remain unanswered: What happens if patients with drug-induced GIN are exposed to the same drug again? When should we use steroid-sparing agents? Are other agents effective? Infliximab has been shown to be effective in one case of isolated sarcoid GIN (12). Can GIN recur in transplanted kidneys? There are no reported cases of patients who had GIN and received a renal transplant. It is encouraging that end-stage renal function did not occur in our patients after a mean follow-up of 45 mo. In the 23 patients reported from the four series, only two patients required long-term renal replacement therapy. GIN has been reported in renal transplants but usually as a consequence of infections. One important message from our analysis and the previously published reports is that better data collection systems such as biopsy registries are needed to provide information on the epidemiology and treatment of rare renal diseases.
|
Conclusion
|
|---|
GIN is a rare, treatable, histologic diagnosis in patients who present with varying degrees of renal failure and mild proteinuria. GIN may be the first manifestation of a systemic disease such as TINU or sarcoidosis, and drugs may be implicated. Histologic features do not seem to distinguish the underlying cause of GIN. In the absence of therapeutic trials, our retrospective data suggest that treatment with a moderate dosage of prednisolone is associated with a good prognosis (relative to other causes of renal failure) irrespective of the underlying cause and of the degree of interstitial fibrosis.
|
Disclosures
|
|---|
None.
|
Acknowledgments
|
|---|
This work was presented at the meetings of the European Renal Association (Istanbul, Turkey; June 2005) and the Scottish Renal Association (Inverness, UK; March 2006).
|
Footnotes
|
|---|
Published online ahead of print. Publication date available at www.cjasn.org.
Received May 24, 2006.
Accepted December 6, 2006.
|
References
|
|---|
- O’Riordan E, Willert RP, Reeve R, Kalra PA, O’Donoghue DJ, Foley RN, Wadek S: Isolated sarcoid granulomatous interstitial nephritis: Review of five cases at one center.
Clin Nephrol55
:297
–302,2001[Medline]
- Mignon F, Mery JP, Mougenot B, Ronco P, Roland J, Morel-Maroger L: Granulomatous interstitial nephritis.
Adv Nephrol Necker Hosp13
:219
–245,1984[Medline]
- Meehan SM, Josephson MA, Haas M: Granulomatous tubulointerstitial nephritis in the renal allograft.
Am J Kidney Dis36
:E27
,2000[Medline]
- Goncalves AR, Caetano MA, Paula FJ, Ianhez LE, Saldanha LB, Sabbaga E: Tuberculous interstitial granulomatous nephritis in renal transplants: Report of three cases.
Transplant Proc24
:1911
–245,1992[Medline]
- Robson MG, Banerjee D, Hopster D, Cairns HS: Seven cases of granulomatous interstitial nephritis in the absence of extrarenal sarcoid.
Nephrol Dial Transplant18
:280
–284,2003[Abstract/Free Full Text]
- Brause M, Magnusson K, Degenhardt S, Helmchen U, Grabensee B: Renal involvement in sarcoidosis: A report of 6 cases.
Clin Nephrol57
:142
–148,2002[Medline]
- Hannedouche T, Grateau G, Noel LH, Godin M, Fillastre JP, Grunfeld JP, Jungers P: Renal granulomatous sarcoidosis: Report of six cases.
Nephrol Dial Transplant5
:18
–24,1990[Medline]
- Cockcroft DW, Gault MH: Prediction of creatinine clearance from serum creatinine.
Nephron16
:31
–41,1976[Medline]
- Viero RM, Cavello T: Granulomatous interstitial nephritis.
Hum Pathol26
:1347
–1353,1995[CrossRef][Medline]
- Toda T, Kimoto S, Nishio Y, Ehara T, Sasaki S: Sarcoidosis with membranous nephropathy and granulomatous interstitial nephritis.
Intern Med38
:882
–886,1999[Medline]
- Khan IH, Simpson JG, Catto GR, McLeod AM: Membranous nephropathy and granulomatous interstitial nephritis in sarcoidosis.
Nephron66
:459
–461,1994[Medline]
- Thumfart J, Muller D, Rudolph B, Zimmering M, Querfeld U, Haffner D: Isolated sarcoid granulomatous interstitial nephritis responding to infliximab therapy.
Am J Kidney Dis45
:411
–414,2005[CrossRef][Medline]
Top
Abstract
Introduction
