Introduction

Hemophagocytic syndrome (HPS) is a life-threatening syndrome involving excessive immune activation. It is often accompanied by renal involvement known as acute kidney injury (AKI) [1]. The prognosis of HPS with AKI is poor [2]. Thus, early diagnosis and treatment are very important. HPS is rarely identified in renal biopsy specimens, and major manifestation includes acute tubular necrosis [3]. A previous study investigated the histologic patterns of HPS in nine children [3], which were found to include ischemic/necrotic changes (44%), interstitial nephritis (33%), lymphohistiocytic infiltration (22%), and erythrophagocytosis (11%). Cases of HPS with glomerulopathy are rare. Until now, a report by Santoriello et al. has described a case of HPS with histiocytic glomerulopathy and intraglomerular hemophagocytosis [4]. Furthermore, erythrophagocytic macrophages were observed in the urinary sediments in that case. In addition, there is only one case report of the presence of urinary phagocytic macrophages in HPS [5]. In that case report, renal biopsy revealed macrophage infiltration around the tubules and macrophage casts with typical features of acute postinfection glomerulonephritis. In our study, we report a rare case of erythrophagocytic macrophage presence in the tubular lumen of a patient with HPS-associated AKI, which was successfully treated.

Case report

A 38-year-old woman was referred to our department because of rapid deterioration of renal function. She had a medical history of mixed connective tissue disease (MCTD) and HPS 4 years prior. She was taking antibiotics for the treatment of pneumococcal pneumonia 2 weeks before presentation at our department. Although her fever and laboratory data improved upon antibiotic administration, her fever and fatigue subsequently recurred. Her blood pressure was 118/70 mmHg and body temperature was 39.7 °C. She had systemic edema, with an 11-kg increase in body weight (87.6 kg) and peripheral lymphadenopathy. Investigations (Table 1) revealed anemia (hemoglobin, 7.9 g/dL), thrombocytopenia (platelet count, 5.6 × 104/µL), leukopenia (white blood cell (WBC) 1080/µL), and AKI with a serum creatinine (Scr) level of 4.1 mg/dL. Her C-reactive protein was 23.7 mg/dL, erythrocyte sedimentation rate was 51 mm/h, ferritin level was 2913 ng/mL, sIL-2Rα level was 7920 U/mL, and triglyceride level was 211 mg/dL. Urinalysis showed proteinuria (0.63 g/day) and 10–19 red blood cells and granular casts per high- and low-power fields, respectively. Her urine β2 microglobulin level was 85,500 µg/L and serum antinuclear antibody titer was 1:1280, with a speckled pattern. Computed tomography revealed cervical and axillary lymphadenopathy and splenomegaly. Bone marrow biopsy revealed erythrophagocytic macrophages. The present case met five out of the eight diagnostic criteria of HPS (HLH-2004), namely fever, splenomegaly, cytopenia, hyperferritinemia, and elevated IL-2Rα level. Subsequently, she was diagnosed with HPS. We commenced the administration of intravenous pulse methylprednisolone (1000 mg three times daily), followed by 60 mg of oral prednisolone. After 2 weeks of treatment, her thrombocytopenia improved, but the high serum creatinine level persisted. Therefore, we performed a kidney biopsy on day 35 after admission to evaluate the cause of renal involvement.

Table 1 Laboratory findings
Full size table

The kidney biopsy sample had 11 glomeruli, which were examined by light microscopy. There were no global sclerotic glomeruli. All glomeruli were intact (Fig. 1a, b), but typical features of acute tubular necrosis and tubulointerstitial nephritis were observed. Erythrophagocytic macrophage casts were observed on Elastica–Masson staining (Fig. 1c). Immunohistochemical staining for CD68 revealed massive macrophage infiltration in the tubular lumen (Fig. 1d). However, there were no intraglomerular macrophages. Although there was a marked interstitial macrophage infiltration in CD68 staining, the cells only showed a slight interstitial erythrophagocytic macrophage infiltration in hematoxylin and eosin staining (Fig. 1e-f). Some arteriolar vascular smooth muscle cells were vacuolated because they were positive for α-smooth muscle actin (Fig. 1f, g).

Fig. 1
figure 1

Light microscopic features. a Mesangial proliferation or endocapillary hypercellularity is not observed. b No intraglomerular macrophage with immunohistochemical staining for CD68 is observed. c Erythrophagocytic macrophages are observed in the tubular lumen. d Immunohistochemical staining for CD68 reveals macrophages in the tubular lumen. e There was only a slight interstitial erythrophagocytic macrophage infiltration in hematoxylin and eosin staining. f Immunohistochemical staining for CD68 highlights interstitial macrophage infiltration. g Vacuolization of the arteriole is observed in Elastica–Masson staining. h Vacuolization of the smooth muscle cells in the arteriole is observed in α-smooth muscle actin staining

Full size image

After 8 weeks of treatment with prednisolone, the Scr level improved to 0.76 mg/dL (eGFR, 68 mL/min/1.73 m2), and the hemoglobin, platelet, and WBC levels returned to normal.

Discussion

Hyperactivation of cytotoxic T cells and natural killer (NK) cells causes HPS. Fever, cachexia, elevated serum triglyceride levels, high serum ferritin level, AKI, and tubular necrosis observed in HPS are all attributed to the high cytokine levels [1, 6].

HPS is often triggered by infectious diseases, mainly viral infections, of which Epstein–Barr virus infection is the most frequent trigger [7]. Bacterial infections also cause HPS, and the association of Staphylococcus spp. with HPS has also been reported [9, 10]. Therefore, we think that Staphylococcus pneumonia infection is the main trigger of HPS in the present case. HPS is also caused by rheumatologic disorders, such as systemic lupus erythematous (SLE), adult-onset Still’s disease, systemic vasculitis, and inflammatory bowel disease [7, 11]. Our patient had a history of MCTD; however, MCTD activity was not observed at the onset of the present episode of HPS.

Renal involvement in HPS is most commonly denoted by AKI [1, 12], followed by the nephrotic syndrome. The prognosis of HPS with AKI is poor [2]. HPS-associated AKI may occur as part of multiorgan failure due to the increased capillary permeability and pre-renal condition. It might also be affected by the nephrotoxic effects of antimicrobial therapy.

HPS-associated glomerulopathy is rare. It is often associated with severe nephrotic syndrome with severe renal dysfunction. Collapsing glomerulopathy is the most common pathologic feature, and minimal change is the second most common finding [13, 14]. In our case, the urinary protein was not in the nephrotic level. Light microscopy and electron microscopy findings did not indicate glomerulopathy, although the kidney biopsy was performed after the steroid treatment.

There were erythrophagocytic macrophage casts in the tubular lumen in the present case. Oki et al. [5] reported urinary phagocytic macrophages in HPS. Urinary sediments might provide a diagnostic clue for detecting renal involvement in HPS. A limitation of this case report is that a detailed analysis of the urinary sediments was not performed.

Although the origin of the erythrophagocytic macrophages is not clear, it is speculated that they might have infiltrated the tubular lumen through the damaged tubuli because there were no macrophages in the glomeruli. In the present case, the interstitial macrophages were not erythrophagocytic. The activated macrophages might have phagocytosed red blood cells after infiltrating into the tubular lumen and into the blood stream. The cause of arteriole smooth muscle vacuolization is unclear. However, this finding is often observed in allograft samples obtained 1 h after kidney transplantation. This vacuolization may be due to ischemia or arteriolar spasm.

Although HPS with AKI has a poor prognosis, our patient’s renal function almost completely recovered. The optimal treatment of HPS remains undefined. Elimination of triggers of HPS is crucial. Li et al. reported that patients with viral infection-related HPS had better outcomes compared to those with malignancy-related HPS, which has high mortality rates [8]. Their study suggested that patients with older age at onset, male sex, splenomegaly, or thrombocytopenia were prone to having poor outcomes.

Our patient was young, female, and her HPS was not related to any malignancy. Moreover, we were able to treat the pneumonia, which was a trigger of HPS, and promptly started steroid treatment. These factors contributed to her good clinical course.

In conclusion, our report presents a new pathologic pattern of HPS. When erythrophagocytic macrophages infiltrate the tubular lumen in a patient with HPS-associated AKI, these cells may be observed in the urinary sediments. Careful urinalysis could provide evidence for renal involvement in HPS. Knowledge of this pathologic pattern of HPS is important to recognize the disease and treat it appropriately.