A retrospective multicenter cohort study of patients with sarcoidosis and renal involvement

Abstract

Background

Sarcoidosis is a multisystem granulomatous disease that can involve the kidneys. Renal involvement is often asymptomatic and underdiagnosed, yet timely recognition and treatment are crucial to prevent irreversible kidney damage. Despite its importance, there is a lack of established treatment guidelines for renal sarcoidosis.

Methods

We conducted a retrospective multicenter study across six Dutch hospitals, including patients with biopsy-proven renal sarcoidosis diagnosed between 2010 and 2024. Clinical, laboratory, histopathological, treatment, and outcome data were collected from electronic medical records. Descriptive statistics were used to summarize findings.

Results

Twenty-nine patients with biopsy-proven renal sarcoidosis were included (median age 56 years; 66% male). Lung involvement was present in 48%, and hypercalcemia in 45% of patients. At diagnosis, median serum creatinine was 242 µmol/L and median eGFR was 23 mL/min/1.73 m². Granulomatous inflammation was present in 76% of biopsies. All patients received corticosteroids as first-line therapy (median initial prednisone dose 0.66 mg/kg/day). Maximal improvement in kidney function was observed within one month of treatment; no significant predictors for early response were identified. Relapses occurred in 48% of patients (median time to first relapse: 7 months), often during or after tapering of corticosteroids. Additional immunosuppressive agents were used in 45% of patients, mainly after relapse. No baseline factors predicted relapse risk.

Conclusions

This study confirms the heterogeneous presentation of sarcoidosis and underscores the need for better biomarkers to guide treatment and decision making in these patients. Early corticosteroid therapy leads to rapid improvement in kidney function, but relapse is common during tapering. Steroid-sparing agents are effective for relapsing disease. Close monitoring during and after corticosteroid tapering is warranted. Our findings support a treatment protocol starting with prednisone 0.5–1 mg/kg/day, followed by gradual tapering, and the addition of steroid-sparing agents in case of relapse. Further prospective studies are needed to optimize treatment strategies and identify predictive markers for relapse and response.

Clinical trial number

Not applicable.

Background

Sarcoidosis is a multisystem granulomatous disorder of unknown etiology that can affect various organs, including the kidneys. Renal involvement in sarcoidosis is often asymptomatic and therefore frequently underdiagnosed, especially in the absence of pulmonary manifestations that trigger medical attention. Diagnostic delay, however, can lead to significant morbidity and kidney failure if the renal involvement is left untreated.

Sarcoidosis can involve the kidneys in several ways, such as effects due to hypercalcemia, (granulomatous) interstitial nephritis and rarely glomerular pathology. The precise prevalence and incidence of interstitial nephritis in sarcoidosis are not known, with reported prevalences varying between 0.7 and 23% of patients [1–5]. In an Italian cohort of 36 sarcoidosis patients with acute kidney injury undergoing a kidney biopsy, 31/36 (86%) of the patients showed granulomatous inflammation [6]. Despite its overall low incidence, its timely recognition is crucial, as untreated nephritis can rapidly progress to severe kidney failure. The presence of active interstitial nephritis necessitates treatment with immunosuppressive agents. The primary goal of therapy is to restore renal function. Treatment often requires long-term management as relapses frequently occur during tapering of immunosuppressive drugs. Despite its importance, there are currently no established guidelines for optimal treatment and randomized controlled trials are lacking.

This retrospective study aims to address this gap by developing a treatment protocol for renal sarcoidosis based on available literature and retrospective data. By synthesizing existing knowledge and clinical experience, we aim to provide a structured approach to managing this challenging condition, ultimately improving patient outcomes and reducing the risk of irreversible renal damage.

Methods

This retrospective multicenter study was conducted in six hospitals across the Netherlands. Nephrologists at participating centers were asked to share clinical data of all patients under their care with biopsy-proven renal sarcoidosis. Patient inclusion was therefore based on availability of biopsy-confirmed cases rather than consecutive inclusion from a single pathology database. The study included patients diagnosed with sarcoidosis and biopsy proven renal involvement, with the biopsy performed between 2010 and 2024. The indication for kidney biopsy was determined by the treating clinician and was not standardized. Relapse was defined based on the assessment of the treating clinician, taking into account clinical, laboratory findings leading to treatment intensification or reinitiation.

Patient data were collected retrospectively from electronic medical records at each participating hospital. Active nephritis was defined based on histopathological findings reported in the original kidney biopsy reports. All kidney biopsies were evaluated by experienced nephropathologists at the time of diagnosis, who assessed the presence of active inflammatory features and documented their findings in formal pathology reports. Ethical approval was obtained from the institutional review board. The project number is 2023–2210.

Descriptive statistics were used to summarize patient characteristics and clinical features. Categorical variables were presented as frequencies and percentages, while continuous variables were summarized as means with standard deviations or medians with interquartile ranges, as appropriate.

Results

We included 29 patients with biopsy proven renal involvement of sarcoidosis from 6 different centers in the Netherlands. Median follow up is 50 months (interquartile range IQR 16–154).

Median age of the patients at diagnosis of renal involvement was 56 years (IQR 48–69 years), with a slight male predominance (19/29, 66%). Median weight at presentation was 82 kg (IQR 74–92 kg). Many patients had lung involvement (14/29, 48%), three patients eye involvement and two patients were diagnosed with cardiac involvement. In patients already known with sarcoidosis (n = 8 (28%)), the initial sarcoidosis diagnosis preceded the renal presentation by 24 months (IQR 11–93 months). In 71% of patients, the renal presentation was the first presentation of sarcoidosis.

Renal function and laboratory findings

At the time of renal sarcoidosis diagnosis, median serum creatinine level was 242 umol/l (IQR 154–410) corresponding to an estimated glomerular filtration rate (eGFR) of 23 mL/min/1.73 m² (IQR 13–38). Urinalysis data were unavailable for three patients. Among the remaining 26 patients, leukocyturia was present in 7 patients, erythrocyturia in 10 patients, and albuminuria in 12 patients. Hypercalcemia was present in 45% of patients. Median SIL2R levels were 4277 pg/ml (IQR 1105–15787).

Histopathological findings

Granulomatous inflammation was identified in 76% of biopsies. Interstitial fibrosis was present in varying degrees. Calcium deposits were seen in 4/29 of patients. Of these patients with calcium deposits, 2/4 (50%) of patients showed hypercalcemia at presentation. Interstitial fibrosis and interstitial nephritis were not correlated with baseline SIL2R, ACE nor with serum creatinine. Granulomatous inflammation was also not correlated with baseline SIL2R, serum creatinine or hypercalcemia. Findings are summarized in Table 1.

Table 1.

Kidney biopsy findings

	All patients (N = 29)
Interstitial fibrosis
<10%	11/29 (38%)
10–50%	13/29 (45%)
>50%	5/29 (17%)
Interstitial nephritis
Minimally active / no nephritis	4/29 (14%)
Moderate active	19/29 (65%)
Highly active	6/29 (21%)
Granulomatous inflammation	22/29 (76%)
Calcium deposits	4/29 (14%)

Treatment and outcomes

All patients received immunosuppressive therapy, with corticosteroids being the first-line treatment in all cases. The median initial prednisone dose was 60 mg/day (IQR 40–60 mg/day) which corresponds to 0.66 mg/kg/day (IQR 0.50–0.79). Prednisone was thereafter tapered and/or increased by the judgement of the treating clinician (Table 2).

Table 2.

Prednisone dose in time (median values with IQR)

	Mg/day	Number of patients
Start dose	60 (40–60)	29
Dose at 2 months	30 (20–40)	27
Dose at 4 months	15 (10–25)	26
Dose at 6 months	10 (5–25)	25
Dose at 12 months	3 (0–12)	21
Dose at 18 months	8 (3–10)	11
Dose at 24 months	6 (0–8)	10

The course of serum creatinine is shown in Fig. 1. Maximal response in seen within one month, and after one months no significant further improvement in kidney function is seen.

Fig. 1.

Course of serum creatinine and eGFR (median values with standard deviation), N = 29

No significant predictive factors for the kidney response within one month were identified, although there was a trend between decrease in serum creatinine and activity of interstitial nephritis (Fig. 2). The percentual decrease in serum creatinine after one month was higher in patients with moderate or highly active interstitial nephritis compared to patients with minimal/no interstitial nephritis (40% vs. 16%, p = 0.18, Mann Whitney U test). In patients with minimally or no interstitial nephritis, the maximal response in kidney function was seen after 2 months (Fig. 2).

Fig. 2.

Course of serum creatinine by interstitial nephritis severity (median values with standard deviation), N = 29. open circle: minimally or not active interstitial nephritis. asterixis:moderate active interstitial nephritis. inverted triangle: highly active interstitial nephritis

Renal relapses occurred in 48% (n = 14) of patients during the median follow-up period of 53 months. Five patients had one relapse, eight patients two relapses and one patient had three relapses. Of the six patients presenting initially with hypercalcemia and developing a relapse, only one patient showed elevated serum calcium levels again at the time of the relapse.

The median time to the first relapse was 7 months (IQR 4–22 months) after initial treatment. The median dose of prednisone at the time of the first relapse was 5 mg/day (IQR 0–20 mg/day). Six out of 14 patients were not on prednisone anymore at the time of their first relapse.

We did not find factors at baseline or of treatment that were correlated with relapsing disease (Table 3).

Table 3.

Overview of parameters in patients experiencing a relapse

	Patients without relapse (N = 15)	Patients with ≥ 1 relapse (N = 14)	P value *
Age	59 (50–69)	50 (47–70)	0.48
Sex (F/M)	6/9	4/10	0.70
Hypercalcemia at start	7/15	6/14	1.00
SIL2R at start	1811 (1000–10342)	6521 (2636–45813)	0.18
Creatinine at start	284 (188–419)	211 (124–297)	0.29
Prednisone start dose (mg/day)	60 (40–60)	50 (35–70)	0.56

* Mann Whitney U test

Patients not experiencing a relapse did not differ from the patients who did relapse with respect to tapering dose of steroids (Fig. 3).

Fig. 3.

Prednisone dose in time until the first relapse (median values). Open circles: relapse.Squares: no relapse

Additional immunosuppressive agents were used in 45% (n = 13) of patients, including mycophenolate mofetil (n = 4), azathioprine (n = 11), methotrexate (n = 4) and infliximab (n = 1). These drugs were started after the first or second relapse. Results are shown in Table 4.

Table 4.

Overview steroid sparing drugs

	Mean dose (mg/dag)	Mean treatment period (months)	Relapses during treatment	Time until relapse (months)	Dose during relapse (mg/day)
MMF (N = 4)	1375	20	0/4
Azathioprine (N = 11)	160	12	3/11	8	150
MTX (N = 4)	12 mg/week	34	1/4	96	5 mg/week

The relapse under methotrexate occurred during tapering of the dose of methotrexate. Azathioprine was stopped because of side effects in 4 patients, and methotrexate was stopped in one patient because of side effects.

Discussion

The findings from this study provide some insights into the complex nature of sarcoidosis and its treatment, while also highlighting several challenges in managing this disorder. Overall, this study confirms the heterogeneous presentation of sarcoidosis and underscores the need for better biomarkers to guide treatment and decision making in these patients.

The first difficulty encountering these patients is making the diagnosis. There is no good test to diagnose renal involvement in sarcoidosis other than kidney biopsy. Also in patients with known sarcoidosis, hypercalcemia and renal insufficiency it is important to perform a kidney biopsy because the treatment of hypercalcemia is different from the treatment of interstitial nephritis and waiting for the renal effect of hypercalcemia correction causes valuable time loss when interstitial nephritis is also present. The clinician would be helped by more refined non-invasive diagnostic tools to guide treatment decisions, such as biomarkers or (nuclear) imaging techniques. One promising urine biomarker which needs to be further studied in this context is urinary angiotensin converting enzyme (ACE) [7].

In our studies, we observed a notably high percentage of patients with concomitant hypercalcemia, a finding that contrasts with previous research in this field [8]. This discrepancy warrants careful consideration. Some patients in our cohort could have experienced renal insufficiency due to the direct effects of hypercalcemia and not of an active interstitial nephritis. Another explanation could be a selection bias in patients being biopsied.

The unpredictability of prednisolone response, even with kidney biopsy findings, remains the next clinical challenge. Consistent with previous research, our study found the maximal response to prednisolone therapy within one to two months with no further improvement in renal function thereafter [8, 9]. Previous research showed no additional benefit for initial treatment with high dose methylprednisolone [9]. We were not able to show any significant prognostic factors for this early treatment response, but previous reports showed that serum creatinine at start and amount of fibrosis on kidney biopsy were predictive markers [10]. We did see this trend as well, but it did not reach statistical significance, probably due to the small number of patients. What we can conclude, is that although a large amount of fibrosis is seen, there is still a fair chance of improvement of kidney function in response to prednisolone. So a one to two month prednisolone trial is justified in all patients.

The starting dose of prednisolone of 0.5-1.0 mg/kg/day was congruent with earlier reports [11, 12]. It is not known whether a lower dose of 0.5 mg/kg/day would be beneficial for all patients as well. In cardiac sarcoidosis this is the case [8]. A total treatment duration of 18–24 months seems necessary [8, 11, 13]. In our study, patient were off prednisolone much earlier.

Our study shows that with a prednisolone tapering schedule of 10–20 mg per month, reaching a dose of 10 mg at 6 months, 50% of patients experienced a relapse after an average of 7 months. This is higher than previous reports who reported a relapse rate of 35% [8, 14]. This might be the effect of the faster lowering of prednisolone.

This, however does not justify a more intensive initial treatment schedule with higher doses of prednisolone for all patients as half of patient would be overtreated. As long as we lack predictive tools to select patients in need of more intense treatment, we underscore the need for close monitoring during the tapering process and the months thereafter.

Our results suggest that steroid sparing agents (when adequately dosed) are effective in most patients in preventing a relapse. As soon as the first relapse occurs, we suggest to start with a steroid sparing agent and temporary doubling the dose of prednisone or restart prednisolone when already stopped. This aligns with recent literature supporting the use of immunomodulatory drugs such as methotrexate, azathioprine and mycophenolate as steroid sparing options in sarcoidosis [11]. There are no adequate data to compare the efficacy of the different steroid sparing agents.

This study has several limitations. The lack of a control group or different treatment groups, inherent to retrospective research, is the main limitation. Future randomized trials comparing different treatment strategies are needed to establish optimal management protocols. That will be difficult given the low incidence of biopsy proven renal sarcoidosis. Additionally, as this was a retrospective study, relapses and the initial diagnosis were defined by the treating clinician. Next to that, the relative short follow up period may not capture long-term outcomes or late relapses. The findings underscore the need for personalized treatment approaches, improved predictive markers and future research including steroid sparing strategies to optimize patient outcomes and minimize treatment related complications.

In conclusion, we advise the following. In case of interstitial nephritis due to sarcoidosis, start oral prednisolone in a dose of 0.5-1 mg/kg. Continue this dose for 4 weeks after which prednisone can be lowered with 5 mg/week until 5 mg. In case of a relapse we advise to start a steroid sparing agent (azathioprine, mycophenolate mofetil or methotrexate).

Author contributions

All authors contributed to collecting data, writing the article and analyzing the data.

Funding

There was no funding.

Data availability

Data are available on request (abech@rijnstate.nl).

Declarations

Ethical approval

Ethical approval was obtained from the institutional review board (Local Feasibility Commitee Rijnstate Hospital). The need for consent was deemed unnecessary according to national regulations. Our study adhered to the Declaration of Helsinki.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.