Authors: Ahmet Burak DİRİM, Vafa SÜLEYMANOVA, Özge HÜRDOĞAN, Özgür Akın OTO, Ayşe Serra ARTAN, Savaş ÖZTÜRK, Yasemin ÖZLÜK, Işın KILIÇASLAN, Halil YAZICI
Categories: Research Article, Sarcoidosis, glomerulonephritis, granulomatous interstitial nephritis, prognosis
Source: Turkish Journal of Medical Sciences
Sarcoidosis is a multisystem disorder that affects many organs, including the kidneys. This single-center retrospective study investigated the clinical, pathological, and laboratory findings of patients with kidney sarcoidosis who were treated with immunosuppressives.
Twenty-three patients with biopsy-confirmed kidney sarcoidosis were included. Demographic, clinical, pathological, and laboratory findings, in addition to the treatments and outcomes of 20 patients with at least one month of follow-up were evaluated.
The median age of the patients at the time of biopsy was 47 years (60.9% were female). The median baseline estimated glomerular filtration rate (eGFR) and proteinuria were 21.5 mL/min and 1 g/g or g/day, respectively. Nineteen of the 23 patients were diagnosed with nonglomerular disease (four had glomerular diseases). Extrarenal sarcoidosis was present in 86.7% of the patients. Granulomatous interstitial nephritis (56.5 %) and nephrosclerosis with intratubular calcific casts (17.4 %) were the two most common diagnoses. All the patients initially received 1 mg/kg/day steroids for kidney involvement. Although no statistical difference was observed in kidney function during the follow-up, steroids improved the eGFR in the first month compared with baseline in patients with nonglomerular diseases (p = 0.049). Eventually, 45% of the patients developed end-stage kidney disease, and 45% of cohort had a treatment response. Patients with higher baseline calcium levels (p = 0.03) and lower degrees of interstitial fibrosis/tubular atrophy (p = 0.043) had better kidney outcomes. Moreover, none of the patients with sarcoidosis-related secondary glomerular disease had a treatment response (p = 0.043).
Hypercalcemia and lower interstitial fibrosis and tubular atrophy rates might be associated with better outcomes in sarcoidosis-related kidney involvement under immunosuppressive treatment. Moreover, late diagnosis, irregular follow-up, and glomerular disorders could be poor prognostic factors.
Keywords: Sarcoidosis, granulomatous interstitial nephritis, glomerulonephritis, prognosis
Sarcoidosis is a systemic inflammatory disease that is usually associated with noncaseating granulomatous infiltration of the affected organs, mostly the lungs. The exact mechanism and pathogenesis have yet to be elucidated [1]. Lung involvement ranges from asymptomatic hilar lymphadenopathy to severe interstitial lung disease. In contrast, 8.7% of patients with sarcoidosis may present without lung involvement [2]. In addition, many extrapulmonary organs, including the kidneys, may be involved [3]. Isolated kidney involvement has also been reported in patients with sarcoidosis. [4]. Granulomatous tubulointerstitial nephritis (TIN) is the most common kidney pathological finding in patients with sarcoidosis. Moreover, nephrocalcinosis and kidney tubular injury due to hypercalcemia or secondary glomerulonephritis (GNs), including membranous glomerulonephritis, focal segmental glomerulosclerosis (FSGS), AA-type amyloidosis, and IgA nephropathy have been reported in the literature [5]. Some retrospective cohorts have included a limited number of patients with sarcoidosis-related kidney involvement [4–13]. Kidney involvement in sarcoidosis can cause end-stage kidney disease (ESKD), which may lead to mortality and morbidity, and corticosteroids are commonly used in kidney sarcoidosis. The most commonly used steroid treatment regimen is 1 mg/kg/day of prednisolone or methylprednisolone. Additionally, other immunosuppressives, such as azathioprine and mycophenolate could be used in corticosteroid refractory or dependent kidney sarcoidosis cases [5–7,11,12]. Nevertheless, predictors of kidney involvement in patients with sarcoidosis, and the prognosis of these patients, besides optimal treatment modalities, could have been partially illuminated due to the relative rarity of this situation. Therefore, it was aimed herein to investigate the clinical, pathological, and laboratory findings, in addition to predictors of treatment response, in patients with kidney sarcoidosis who were treated with immunosuppressives in this single-center retrospective study.
Twenty-three patients diagnosed with sarcoidosis, who underwent kidney biopsy between 2003 and 2023, were included in this retrospective study. Demographic, clinical, laboratory, and histopathological data were also recorded. Laboratory findings, including hemogram, serum creatinine, estimated glomerular filtration rate (eGFR) using the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation [14], serum albumin, serum levels of corrected calcium with albumin, uric acid, phosphorus, C-reactive protein (CRP), angiotensin-converting enzyme (ACE) levels, spot urine protein-to-creatinine ratio (UPCR), and 24-h proteinuria at the time of kidney biopsy were evaluated. Histopathological findings, including pathological diagnosis, glomeruli count, global and segmental sclerotic glomeruli percentages, degree of interstitial fibrosis and tubular atrophy (IF/TA), interstitial inflammatory cell infiltration, intratubular calcific casts, and presence and degree of granuloma formation, were recorded. In addition, immunosuppressive treatments postbiopsy, mortality rates, and laboratory results at the last eGFR were evaluated in 20 patients with at least one month of follow-up data. In addition to the baseline data, the one month postbiopsy and last kidney function data were also noted. However, this retrospective study was based on patient files and the electronic healthcare system data of patients who had no regular follow-up. Therefore, some follow-up data, including clinical data, mortality, and the last eGFR, were extracted from electronic healthcare systems. The primary outcome was treatment response based on the last eGFR data. Complete recovery of kidney function was defined as an eGFR (CKD-EPI) >60 mL/min/1.73 m^2^ and proteinuria <0.5 g/g or g/day, and partial recovery of kidney function was defined as an eGFR >60 mg/dL and proteinuria 0.5–1 g/g or g/day, or an eGFR (CKD-EPI) >45 mL/min/1.73 m^2^ and proteinuria <0.5 g/g or g/day. Partial recovery was also defined as an increase in the eGFR by ≥25% of the baseline eGFR from the baseline to the last eGFR. Nonresponders were defined as individuals without complete or partial response. ESKD was defined as eGFR <15 mL/min/1.73 m^2^. The secondary outcomes were the development of ESKD and mortality.
In the descriptive statistics, median (interquartile range (IQR) 25%–75%) and frequency (percentage) values were used. The distribution of the variables was measured using the Kolmogorov–Smirnov test. The Mann–Whitney U test was used to analyze the quantitative independent data. The Wilcoxon and Friedman tests were used to analyze the dependent quantitative nonparametric data. The chi-squared test was used to analyze the qualitative independent data. The Fisher’s exact test was performed when the chi-squared test conditions were not met. IBM SPSS Statistics for Windows 27.0 (IBM Corp., Armonk, NY, USA) was used for the data analysis. p < 0.05 was considered statistically significant.
Fourteen of the 23 patients (60.9%) were female. The median age was 47 (min–max: 22–74) years. Moreover, 34.7% of the patients had a prior sarcoidosis diagnosis at the time of kidney biopsy. Twenty of the 23 patients had at least one month of follow-up data. Demographic, laboratory, histopathological, and clinical data of the study group are shown in Table 1. Twenty patients were divided into two groups, as the responders (complete or partial) and nonresponders. Nine of the 20 patients were responders (all of whom were partial responders).
The patients were categorized as having glomerular and nonglomerular diseases, including granulomatous and nongranulomatous TIN and nephrosclerosis with tubular calcium deposits. Nineteen of the 23 patients were diagnosed with nonglomerular disease, and four patients had glomerular diseases. The most common diagnosis was granulomatous TIN (n = 13, 56.5%). Four patients (17.4%) had nephrosclerosis with intratubular calcific casts, without TIN. Two patients had FSGS (8.7%), and two patients (8.7%) were diagnosed with nongranulomatous TIN. AA-type amyloidosis and phospholipase A2 receptor (PLA2R)-positive membranous nephropathy (MN) were present in one patient (4.3%). Patients with glomerular diseases have lower treatment response rates than those with nonglomerular diseases (p = 0.043) (Tables 2 and S1). The degrees of tubulointerstitial inflammatory cell infiltration, granuloma formation, and IF/TA in biopsies, as well as the percentage of global and segmental sclerotic glomeruli, are shown in Table S1. No difference was observed in the pathological diagnosis between the responders and nonresponders. In contrast, none of the responders had moderate or severe IF/TA. The degree of IF/TA (not significant/mild vs. moderate/severe) showed statistical significance between the responders and nonresponders (p = 0.043). The percentage of global sclerotic glomeruli was higher in the nonresponders than in the responders. However, this difference was not statistically significant (p = 0.274). Other histopathological data for the responders and nonresponders are shown in Table 2.
The baseline median creatinine level of the patients was 2.23 mg/dL at the time of biopsy. The baseline median eGFR and proteinuria levels of the patients were 26.5 mL/min/1.73 m^2^ and 1 g/g or g/day, respectively. Hypercalcemia (corrected serum calcium >10.4 mg/dL) was present in 17.4% of patients at the time of kidney biopsy. Eight patients had ACE levels at the time of biopsy, and for three of these patients, the ACE levels were elevated. Other laboratory data of the patient groups are shown in Table 1. The responders had higher corrected serum calcium levels than the nonresponders (p = 0.03). Correspondingly, the overt hypercalcemia rate was higher in the responders than in the other groups (p = 0.026). A comparison of the other laboratory data for the responders and nonresponders is provided in Table 2.
Of the patients, 69.6% had lung lymphadenopathy and 39.1% had extra-hilar lymphadenopathy (ultrasonography, CT, or histopathological examination). Two patients had histopathological nonnecrotizing granulomatous lymphadenitis; 17.4% cutaneous, 17.4% liver, 17.4% eye, 8.7% minor salivary gland, 4.3% arthritis, 4.3% bilateral parotitis, and 4.3% bone marrow involvement were present in the patient group during both the pre and postbiopsy follow-up. Isolated kidney sarcoidosis was present in three patients (13%). The diagnosis of isolated kidney sarcoidosis was based on granulomatous interstitial nephritis with noncaseating granulomas or nephrosclerosis with intratubular calcific casts and other findings such as fever and constitutional symptoms and/or increased ACE levels, and increased inflammatory markers and/or hypercalcemia, after excluding other granulomatous disorders. There was no statistically significant difference in extrarenal involvement between the patients with and without treatment response (Table 3).
Twenty of the 23 patients had at least one month of postbiopsy follow-up. All 20 patients were treated with corticosteroids (initially 1 mg/kg/day of prednisolone with gradual tapering) postbiopsy. Moreover, azathioprine at 2 mg/kg/day (20%), hydroxychloroquine at 200 mg/day (10%), cyclosporine A (C0 through level 50–150 ng/mL) (5 %), mycophenolic acid at 2000 mg/day (5%), and rituximab at 475 mg/m^2^/week for 4 weeks (5 %) were administered to patients with corticosteroid refractory sarcoidosis-related kidney involvement during follow-up (Tables 1 and 4).
None of the 20 patients with follow-up data had a complete treatment response. In addition, none of the patients with glomerular disorders exhibited a treatment response (p = 0.043 according to the chi-squared test for responsive vs. nonresponsive in glomerular vs. nonglomerular diseases). However, nine patients with nonglomerular diseases had a partial response (45%). In addition, nine of the 20 patients (45%) developed ESKD during the follow-up period. Changes in the eGFR during follow-up were not statistically significant, according to the Friedman test (p = 0.522). Wilcoxon signed rank test results for the baseline to one month postbiopsy eGFR data, one month postbiopsy to the last eGFR data, and baseline to the last eGFR data were also not statistically significant (p = 0.161, p = 0.588, and p = 0.940, respectively). However, the Wilcoxon signed rank test results for patients with nonglomerular diseases (16 patients) for baseline to one month postbiopsy were statistically significant (p = 0.049). The one month postbiopsy to month-to-last eGFR data and baseline-to-last eGFR data were not statistically significant (p = 0.453 and p = 0.140, respectively) (Figure). eGFR changes during follow-up were not statistically significant with the Friedman test in patients with nonglomerular disease (p = 0.185).
Figure eGFR levels in patients with sarcoidosis-related nonglomerular renal diseases during follow-up.
Of the 20 patients, five (25%) died during the follow-up. Two patients developed ESKD due to kidney involvement in sarcoidosis. One of the non-ESKD patients died of high-grade B-cell lymphoma five years after the diagnosis of sarcoidosis (sarcoidosis-lymphoma syndrome). Two patients with partial kidney disease died during the study period (the etiology of mortality remains unknown).
Table 4 shows the clinical and pathological data as well as the outcomes of the 23 patients. Table S2 shows the summary of publications related to sarcoidosis with biopsy-proven native kidney involvement, which included more than 15 patients.
The prognosis and the prognostic factors of patients with kidney sarcoidosis who were treated with immunosuppressives were investigated herein. Hypercalcemia and lower IF/TA rates were associated with better kidney outcomes after corticosteroid treatment, and glomerular diseases secondary to sarcoidosis had worse outcomes than nonglomerular sarcoidosis-associated kidney diseases. On the other hand, the absence of regular follow-up for some of the patients might lead to poor long-term patient and kidney outcomes.
Kidney involvement in sarcoidosis, a broad spectrum disease, is generally associated with multiorgan involvement. Increased soluble interleukin-2 receptor levels and decreased serum eGFR could be predictors of kidney involvement [12]. Granulomatous TIN was the most common kidney involvement in the current cohort, similar to previous publications [4,7–10,12]. In addition, nongranulomatous TIN, nephrosclerosis with intratubular calcific casts, and secondary glomerular diseases, including FSGS, AA-type amyloidosis, and PLA2R-positive MN, were present in the case series.
In the subgroup analysis, steroid treatment after diagnosis improved kidney function (eGFR) one month after steroid treatment in patients with nonglomerular disease. None of the patients with secondary glomerular diseases due to sarcoidosis responded to the treatment in the study (TIN and hypercalcemic nephrosclerosis were associated with better kidney function at the last follow-up than baseline data). This finding is consistent with that reported in the literature. Correspondingly, a study by Stehlé et al. [5], which included glomerular diseases due to sarcoidosis, had worse kidney and patient outcomes than the kidney sarcoidosis studies that included patients with TIN and hypercalcemic nephropathy [4,6–11] (Table S2). Glomerular diseases associated with sarcoidosis exhibit a broad spectrum of symptoms [5]. MN is the most common type of glomerular involvement in sarcoidosis. A recent study showed that PLA2R (38.9%) and neural epidermal growth factor-like-1 protein (NELL-1) (22.2%) positivity are common in sarcoidosis patients with MN. While the exact mechanism remains unknown, it was speculated that increased inflammation in sarcoidosis might cause increased glomerular expression of PLA2R and NELL-1 which might trigger antibody-mediated MN development [13]. Correspondingly, the patient herein with MN tested positive for PLA2R.
Histopathologically, none of the responders had moderate or severe IF/TA (p = 0.043). In addition, the responders had lower global sclerosis rates than the nonresponders, although this difference was not statistically significant (probably owing to the limited number of patients). In previous studies of kidney sarcoidosis, chronic lesions were associated with poor kidney outcomes [7,10]. In addition, most of the patients had multisystemic sarcoidosis, similar to the findings of previous studies (Table S2). The responders also had higher serum calcium levels at the time of the kidney biopsy. This finding was similar to the results reported by Rastelli et al. [10].
However, the patients herein had significantly poorer long-term kidney outcomes than those in the previous studies. This finding might be multifactorial. First, most of the patients had no history of sarcoidosis before the kidney biopsy (34.7%). Hence, the late diagnosis of multisystemic sarcoidosis may have been the reason for this. Accordingly, the nonresponders had more chronic histopathological lesions (IF/TA) than the responders. Second, most of the patients had irregular follow-ups, which may have resulted in inadequate treatment in the case of relapsing or persistent disease. Third, the study’s follow-up time was longer than that of some studies [8,11] regarding sarcoidosis-related kidney involvement, which could contribute to the declining eGFRs of the patients. Fourth, the kidney response criteria differed from those used in other studies. A more stringent response criterion was used than those in previous studies [4,7]. Fifth, the study group consisted of Turkish individuals, which might have caused different outcomes than in other studies.
Due to the retrospective nature of the study and the lack of regular follow-up, the extraction of data from patient files and electronic hospital systems were major limitations of this research. Relapsing disease was reported as a frequent problem in a study by Mahévas et al. [7]. Therefore, relapsing and refractory kidney diseases may have been inappropriately treated in the current cohort. In addition, the treatment compliance of the patients herein was indefinite. Detailed dosages of steroid treatments and side effects of immunosuppressive therapies could not be extracted due to limited data. Moreover, new-onset extra-kidney involvement might have been underdiagnosed because of irregular follow-up. However, this study, which included 23 patients with sarcoidosis-related kidney involvement, expands the spectrum of kidney sarcoidosis in the literature. Furthermore, it was also shown that hypercalcemia and tubulointerstitial involvement of sarcoidosis with early diagnosis (lower IF/TA rates) might be associated with better kidney outcomes under immunosuppressive treatments.
In conclusion, kidney sarcoidosis is a broad-spectrum disease generally associated with multisystem involvement. Moreover, hypercalcemia and lower IF/TA rates might be associated with better outcomes in kidney sarcoidosis, and the prognosis might be poor despite conventional immunosuppressive treatments in cases with late diagnosis, irregular follow-up, and secondary glomerular disorders. Therefore, multidisciplinary close follow-up of these patients should be an essential approach in clinical practice because it might require prolonged and long-term treatments.
None