Twenty-six patients suffering from both sarcoidosis and glomerular disease were retrospectively identified and followed between 1977 and 2012 in the Nephrology Departments of seven French hospitals (Henri Mondor Hospital, Necker Hospital, Valenciennes Hospital, Tenon Hospital, La Louvière Hospital, Amiens Hospital, Bichat Hospital). We conducted this retrospective study by sending a questionnaire to all nephrology departments involved in the management of glomerular diseases to determine if some patients exhibited glomerular involvement in the context of sarcoidosis occurrence. In each hospital, patients were identified by computing of the renal pathology and clinical diagnosis databases. All patients underwent a renal biopsy for the exploration of proteinuria and/or renal impairment. Demographic, clinical and laboratory data were assessed for each patient at the time of kidney biopsy. Glomerular kidney disease was considered as occurring simultaneously with sarcoidosis when the delay between the diagnoses of both diseases was less than three months. Follow up data (at least six months after kidney biopsy) were obtained for all patients.

Sarcoidosis was diagnosed according to the statement of the American Thoracic Society, the European Respiratory Society and the World Association of Sarcoidosis 27 . Its diagnosis was supported by a body of evidence including, clinical, biological, radiographic findings and/or histological evidence of non-caseating granulomas on tissue biopsy 1 2 5 . The number of affected organs was systematically reviewed for all patients. In patients with lung involvement, the four-stage radiographic classification of sarcoidosis was used to determine the severity of the pulmonary lesions associated with sarcoidosis: stage 1 for bilateral hilar lymphadenopathy; stage 2 for pulmonary infiltrates associated with intrathoracic adenopathies; stage 3 for isolated parenchymal infiltration without fibrosis; and stage 4 consisted of pulmonary fibrosis 28 . Sarcoidosis therapy, including steroid and/or immunosuppreseive agents use was recorded for all patients.

Chronic kidney disease was defined as a permanently (lasting at least three months) decrease of estimated glomerular filtration rate (eGFR) of less than 60 ml/min/1.73 m² according to the modification of diet in renal disease (MDRD) formula 29 . Proteinuria was defined as an albumin excretion rate of > 0.3 g/d. Nephrotic syndrome was defined by urinary protein excretion rates of more than 3 g/d and serum albumin excretion rates below 30 g/L.

Diagnosis of MCNS included the presence of minimal change glomerular lesions and the absence of immunoglobulin and/or complement deposits 30 . FSGS lesions were identified by the presence of segmentally collapsed glomerular capillaries with areas of glomerular scarring associated with focal and segmental granular deposition of IgM and C3 within the segmental glomerular scleroses 31 . MN was characterized granular subepithelial deposition of immunoglobulins and complement along the basement membrane in IF study 32 . Diagnosis of IgAN consisted of the presence of IgA-dominant or codominant immune deposits within the mesangium 33 . Patients who fulfilled the 1997 American College of Rheumatology revised criteria for systemic lupus erythematosus (SLE) with documented sarcoidois were also included in this study 34 . In these cases, renal pathological lesions of SLE were analyzed according to the 2003 International Society of Nephrology and Renal Pathology Society (ISN/RPS) classification 35 .

For patients with a diagnosis of MN, MCNS or FSGS, a proteinuria level of less than 0.3 g/d at the six-month follow-up defined a complete remission. Patients with a proteinuria level between 0.3 and 3 g/d and those whom with at least 50% reduction in the level of proteinuria, with an albumin level of > 30 g/L were considered to be in partial remission. IgAN and lupus nephritis were considered to be in remission when the proteinuria level decreased (of more than 50% of initial proteinuria level) with the stabilization or improvement of GFR. Specific treatment of glomerular disease (steroid and/or immunosuppressive treatment) was recorded for all patients.

Twenty-six patients (18 men and 8 women) were retrospectively identified. Their demographic, clinical and biological data are summarized in Table 1. The mean ages at the onset of the sarcoidosis and glomerular disease diagnosis were 37 years (range of 19–56 years) and 39 years (range of six months-59 years), respectively. In 22 patients, the diagnosis of sarcoidosis was confirmed by organ or tissue biopsy. Pathological confirmation of the sarcoidosis diagnoses was based on lung biopsy (nine cases), skin biopsy (three cases), lymph node biopsy (seven cases), muscle biopsy (one case), peritoneal biopsy (one case), and salivary glands biopsy (one case). In four patients, the diagnosis of sarcoidosis was based only on compatible clinical, biological and radiographic presentations 27 .

MN: Membranous Nephropathy, IgAN: IgA nephropathy, MCNS: minimal change nephrotic syndrome, FSGS: focal and segmental glomerulosclerosis, GD: glomerular disease, GFR: glomerular filtration rate, NS: nephrotic syndrome, GTIN: granulomatous tubulointerstitial nephritis, CR: complete remission, PR: partial remission.

* after exclusion of patients with end-stage renal disease.

Chest radiographs confirmed the stages of the sarcoidosis thoracic involvement in 22 patients: nine patients had stage 1, ten patients had stage 2 and three patients had stage 3. Extrathoracic localization (other than GTIN) was present in 22 patients. The mean number of organs affected by sarcoidosis was estimated to be 2.6.

Sarcoidosis was considered to be an indolent disease, not requiring specific curative treatment in nine patients (34%). One patient received exclusively topical steroid therapy for localized uveitis. Sixteen patients (61%) received systemic steroid therapy to control active, life-threatening sarcoidosis. In eight patients, steroid therapy led to complete remission of granulomatous disease. Among the eight remaining patients, three developed steroid-dependent sarcoidosis, which required prolonged therapy, whereas five patients displayed steroid-resistant sarcoidosis, which required immunosuppressive therapy consisting of Azathioprine (two cases) or Plaquenil (three cases).

Glomerular disease was diagnosed after sarcoidosis in 11 cases (42%) with a mean delay of 9.7 years (range ten months to 20 years) between the two diagnoses (Table 1). Sarcoidosis was considered to be in complete remission at the time of renal biopsy in six of these patients, whereas it was either: indolent and not treated in three patients; resistant to steroid therapy in one patient; and steroid-dependent in one patient. In six patients (23%), the glomerulopathy preceded the diagnosis of sarcoidosis, with an average delay of eight years (range from six months to 25 years) between the two diagnoses. Nine patients (35%) simultaneously displayed glomerular disease and sarcoidosis (Table 2). All of these patients (except patient 2 [pt2]) received exclusive steroid therapy to treat both conditions, regardless of the type of glomerular lesions. With this therapeutic management, sarcoidosis was considered to be in complete remission in four cases (pt5, pt7, pt8 and pt9). In these patients, effective treatment of sarcoidosis was associated with complete remission of glomerular disease in only one case (pt7) and with partial remission in three cases (pt5, pt8 and pt9). In three other patients (pt1, pt3 and pt6), steroid therapy led to complete remission of glomerular disease but failed to induce remission of sarcoidosis. In one patient (pt4), steroids failed to induce remission of either glomerular disease or sarcoidosis.

GD: Glomerular disease, MN: membranous nephropathy, IgAN: IgA nephropathy, MCNS: minimal change nephrotic syndrome, FSGS: focal and segmental glomerulosclerosis, GTIN: granulomatous tubulointerstitial nephritis, GFR: glomerular filtration rate, Aza: azathioprine, MMF: Mycophenolate Mofetil, CR: complete remission, PR: partial remission.

The spectrum of glomerular lesions associated with sarcoidosis consisted of MN in eleven patients (42%), IgA nephropathy in six cases (23%), FSGS in four patients (15%), MCNS for three patients (12%) and proliferative lupus nephritis in two patients (8%). As showed in Figure 1, MN seems to be more common in the context of sarcoidosis than in general population (42% of cases vs 10–15% of glomerular diseases in general population) whereas incidence of other glomerular diseases seems to be quite similar 36 37 . GTIN was observed in association with glomerular injury in six patients (23%) (Figures 2, 3A, 3B and 4). GTIN was exclusively present in renal biopsy of patients exhibiting concomitantly sarcoidosis and glomerular disease (Table 2). Mean proteinuria level at the onset of renal involvement was estimated to be 5.7 g/d (range of 0.45-20 g/d). Mean GFR at the time of glomerular disease diagnosis was 70.7 mL/min/1.73 m² (range of 16–133 ml/min per 1.73 m²). Nine patients displayed significant renal impairment (GFR of < 60 ml/min/1.73 m²) at the time of renal biopsy.

In our study, MN was the most frequent glomerular disorder associated with sarcoidosis (42% of cases). Among these cases, one of them has been described previously 14 . In this group, only three patients exhibited glomerular lesions simultaneously with sarcoidosis, whereas MN preceded the onset of sarcoidosis in five patients (mean delay of 4.5 years). In three other patients, MN was diagnosed after sarcoidosis (mean delay of 12.6 years). Mean proteinuria level was 7.4 g/d and the mean GFR at the time of MN diagnosis was estimated to be 79.2 mL/min/1.73 m². In two patients with simultaneous MN and sarcoidosis (pt1 and 9), the renal biopsy revealed a typical GTIN pattern in association with glomerular injury. In patient 1, steroid therapy induced complete remission of MN, but did not induce remission of sarcoidosis until after Plaquenil treatment was added to steroid therapy (Table 2).

Among the six patients with biopsy-proven IgAN, three of them presented with glomerular disease and sarcoidosis simultaneously. In three other patients, IgAN occurred after sarcoidosis with a mean delay of 6.7 years. The mean proteinuria level at presentation was 5 g/day. The mean GFR of this subgroup of patients was estimated at 60.9 mL/min/1.73 m². All patients with simultaneous sarcoidosis and IgAN (pt5, pt6 and pt7) received steroid treatment, which led to complete remission of sarcoidosis in two cases (pt 5 and pt7) (Table 2), whereas Azathioprin treatment was required for uncontrolled sarcoidosis in one patient (pt6). Steroid therapy induced complete remission of IgAN in two of these three cases (pt6 and pt7) and partial remission in one case (pt5).

Four patients exhibited typical FSGS lesions on renal biopsy. Renal function was significantly impaired (mean GFR of 51.2 mL/min/1.73 m²) and the mean proteinuria level was 4.5 g/d. Two patients were diagnosed with FSGS and sarcoidosis simultaneously and two patients developed FSGS after sarcoidosis with delays of 7.3 years and 19.8 years. Typical GTIN lesions were also found in the biopsy of one patient who had simultaneous FSGS and sarcoidosis (pt3). Steroid therapy for this patient (pt3) led to complete remission of proteinuria but did not significantly improve sarcoidosis. In the second patient with simultaneous FSGS and sarcoidosis (pt4), steroid treatment failed to reduce proteinuria or to improve sarcoidosis symptoms. Despite Azathioprin treatment, sarcoidosis remained uncontrolled and the patient progressed to end-stage renal disease 3.7 years after initial presentation.

Three patients exhibited MCNS in the context of sarcoidosis. In one case, both diseases developed simultaneously whereas one patient had MCNS 5 years after sarcoidosis and one acquired it 25 years before the onset of sarcoidosis. Nephrotic syndrome without significant impairment of renal function was present in all patients at the time of renal biopsy (mean GFR: 77.1 ml/min per 1.73 m²). GTIN was associated with MCNS lesions in the patient with concomitant MCNS and sarcoidosis (pt8). In this patient, steroid and mycophenolate mofetyl therapies led to partial remission of the nephrotic syndrome, whereas sarcoidosis was considered to be in complete remission after steroid therapy. The patient who acquired MCNS before sarcoidosis had multiple relapses of nephrotic syndrome and received several lines of immunosuppressive treatment for steroid-dependent MCNS. Unexpectedly, steroid and/or immunosuppressive treatments were not administered at the time sarcoidosis was diagnosed. Re-introduction of steroid therapy led to sarcoidosis remission.

Two patients with previous history of steroid-resistant sarcoidosis had biopsy-proven lupus (SLE) glomerulonephritis (class IIIA in one case and IIIA-V in the second case). These patients developed sarcoidosis ten months and seven years, respectively, before the SLE diagnosis. Renal biopsies were performed because of significant proteinuria (1 g/d and 0.45 g/d, respectively) associated with microscopic hematuria. Antinuclear antibodies were not detected at the time sarcoidosis was diagnosed but became evident when the renal manifestations began. Therapy for lupus nephritis consisted of steroid and cyclophosphamide drugs in both cases and led to complete remission in one case, but failed to induce remission of lupus nephritis in the second case.

Steroid and/or cytotoxic therapy was used to treat glomerular disease in 18 of 26 patients. Supportive treatment was initiated in eight patients (five patients with MN, two patients with IgAN and one patient with FSGS). At the end of follow-up (average of 8.4 years), complete remission of glomerular disease was confirmed in nine patients, whereas remission was considered to be partial for eight additional patients. Three other patients displayed significant persistent proteinuria. Six patients developed end-stage renal disease requiring intermittent hemodialysis and three of them underwent kidney transplantation. At the end of the follow-up, the mean GFR after excluding the six patients with end stage renal disease was estimated to be 67.8 mL/min per 1.73 m² and three patients died.