INTRODUCTION
Hemophagocytic lymphohistiocytosis (HLH) is a rare but frequently fatal hyperinflammatory syndrome caused by dysregulated immune activation, resulting in excessive cytokine release and progressive multiorgan dysfunction.1 Hepatic involvement is common in HLH and may manifest with hepatocellular injury, cholestasis, coagulopathy, or acute liver failure. HLH may occur as a primary genetic disorder or arise secondary to triggers such as infection, malignancy, or autoimmune disease.2 Macrophage activation syndrome (MAS) is a form of HLH in the context of rheumatologic and autoimmune disease.2
Sarcoidosis is a multisystem granulomatous disease characterized by noncaseating granulomatous inflammation that most commonly affects the lungs and lymphatic system, though hepatic involvement is frequently identified histologically.3 Clinical manifestations of hepatic sarcoidosis range from asymptomatic cholestatic liver enzyme elevation to portal hypertension and cirrhosis. Despite the inflammatory nature of sarcoidosis, it is an uncommon trigger of secondary HLH or MAS/HLH, particularly when hepatic involvement predominates. We present a case of fever of unknown origin initially diagnosed as HLH on bone marrow examination and subsequently determined to represent hepatic sarcoidosis-associated MAS/HLH following liver biopsy.
CASE REPORT
A 72-year-old man presented to our quaternary care hospital with fever, lightheadedness, and malaise after completing a treatment course of amoxicillin-clavulanate for pneumonia 4 days prior. He endorsed having intermittent night sweats for about 3 years. He denied joint pain, blurry vision, skin rashes, weight loss, recent sick contacts, travel, or illicit drug, supplement, or alcohol use.
On admission, he had a maximum temperature of 103.4°F and was tachycardic. He had mild scleral icterus, had no stigmata of chronic liver disease, and had no hepatomegaly, ascites, or abdominal tenderness. Initial laboratory test results showed pancytopenia: white cell count 2.6 B/L, hemoglobin 11.9 g/dL, platelets 50 B/L, aspartate aminotransferase 325 IU/L, alanine aminotransferase 347 IU/L, alkaline phosphatase 764 IU/L, total bilirubin 3.8 mg/dL, direct bilirubin 2.7 mg/dL, international normalized ratio 1.35, fibrinogen 266 mg/dL, creatinine 1.2 mg/dL, triglyceride 247 mg/dL, anti-smooth muscle antibody 20, and ferritin 7,108 ng/mL.
Initial infectious workup including chest X-ray, blood cultures, urinalysis, and respiratory viral pathogen panel were negative. An interleukin-2 (IL-2) receptor level was sent and was found to be 33,500 pg/mL (reference range 532–1,891 pg/mL). Additional laboratory evaluation for infectious, metabolic, autoimmune, and toxicologic etiology was unremarkable (Table 1).
Chest, abdominal, and pelvic CT were negative for masses, abscesses, hepatomegaly, or adenopathy but showed splenomegaly and mild gallbladder thickening, with follow-up abdominal ultrasound showing similar findings. Given the hyperbilirubinemia, fever, and suspicion for cholangitis, he was started on empiric intravenous antibiotics. A magnetic resonance cholangiopancreatography was obtained which did not suggest hepatic, biliary, or pancreatic disease, although did reveal multiple hypointensities of the spleen suggestive of infarcts (Figure 1).
Transthoracic echocardiogram showed normal left ventricular systolic function and no vegetations.
Given the constellation of fever, splenomegaly, hyperferritinemia, pancytopenia, abnormal transaminases, elevated IL-2 receptor, the patient underwent a core-needle bone marrow biopsy showing hemophagocytosis (Figure 2), further supporting the diagnosis of HLH. Due to a weakly positive anti-smooth muscle antibody and rising transaminases, he was empirically treated for autoimmune hepatitis with prednisone. However, on further workup, angiotensin-converting enzyme (ACE) levels returned at 223, suggesting the presence of sarcoidosis.
As his transaminase and bilirubin levels continued to increase despite treatment with prednisone, he underwent a liver biopsy for further evaluation of secondary HLH (Table 2). Histopathology revealed hemophagocytosis expanding the sinusoids along with noncaseating granulomatous inflammation compatible with sarcoidosis (Figure 3); the Masson trichrome stain, Grocott methenamine silver stain, and acid-fast bacillus stain were all negative. He was initiated on pulse steroids followed by escalation to anakinra, an IL-1R antagonist, as laboratory test results, including ferritin, transaminases, bilirubin, as well as pancytopenia, continued to worsen (Table 2).
Moreover, his ACE level increased to 708, and C-X-C motif chemokine ligand 9 (CXCL9) was obtained at 58,211 (reference value <647). Given the lack of improvement, significantly elevated CXCL9 levels, and hypoenhancing splenic lesions on imaging, he was initiated on etoposide for a possible malignant cause and underwent a splenectomy 2 days later to rule out a splenic lymphoproliferative process. Surgical histopathology demonstrated marked red pulp congestion with histiocytic hemophagocytosis and geographic infarction without evidence of malignancy (Figure 4).
His postoperative course was notable for improvement in HLH biomarkers (Table 2). However, on postoperative day 4, he became significantly tachycardic with increasing oxygen requirements, and then subsequently unresponsive. Despite resuscitative efforts, he died, suspected to be from hemorrhagic shock, as follow-up autopsy results showed extensive bleeding in the lumen of the small and large bowel without identification of a specific source.
DISCUSSION
HLH is a life-threatening syndrome characterized by dysregulation of the immune response that requires a thorough diagnostic approach to determine its etiology to prevent a potential devastating outcome.1,2 This rare disorder may be classified into primary and secondary forms, with the former being genetic in nature, found predominantly in the pediatric population.4 Secondary HLH, commonly in older children and adults, is usually driven by an underlying inflammatory trigger, predominantly associated with infection (50%), followed by malignancy (28%) and autoimmune disease (12%).5 MAS, is considered a subtype of HLH that develops in the background of autoimmune disease usually in the setting of systemic juvenile idiopathic arthritis, the Still disease, or systemic erythematous lupus.6 Given the hyperinflammatory nature of HLH, the syndrome may mimic a presentation similarly to septic shock with fever, organomegaly, neurologic deficits or dysfunction, and pancytopenia.1 Per the American Society of Hematology, the HLH-2004 diagnostic criteria, which require fulfillment of 5 of 8 criteria (fever, splenomegaly, cytopenia affecting at least 2 lineages, hypertriglyceridemia/hypofibrinogenemia, hyperferritinemia, elevated soluble IL-2 receptor, hemophagocytosis on bone marrow, spleen, or lymph node pathology), is a useful tool to guide diagnosis of HLH while meticulously investigating secondary causes.7 Diagnosis is essential as critical illness can quickly ensue and include complications such as liver failure and shock.1 Our patient met diagnostic criteria based on persistent fever, splenomegaly, cytopenias, hyperferritinemia, elevated soluble IL-2 receptor level, and histologic evidence of hemophagocytosis on bone marrow biopsy.
Our patient's liver injury was a mixed hepatocellular and cholestatic pattern. Diagnoses such as ascending cholangitis, primary sclerosing cholangitis, drug-induced liver injury from amoxicillin-clavulanate, and congestive hepatopathy were all considered on the original differential diagnosis. While HLH-related liver injury may vary between hepatocellular or cholestatic patterns, liver biopsy remained crucial for diagnosis to rule out infiltrative disease such as amyloidosis and lymphoma.
Interestingly, this patient's liver biopsy results exhibited non-necrotizing granulomatous inflammation which, in conjunction with an elevated ACE level, strongly suggested a diagnosis of hepatic sarcoidosis that likely provoked MAS/HLH.8 Sarcoidosis frequently affects the liver on a clinical spectrum ranging from asymptomatic to clinically significant liver injury and decompensated cirrhosis.3,9 Based on current literature, systemic sarcoidosis is an uncommon cause of secondary MAS/HLH (1%), with primary hepatic sarcoidosis as an exceptionally rare etiology.10–12 Given the rarity of sarcoid-induced MAS/HLH, an underlying lymphoproliferative disorder was also suspected given little improvement in HLH with initial therapy of steroids and anakinra and significant elevation in CXCL9. Therefore, our patient underwent a splenectomy for further investigation for which pathology did not show malignancy.13 Despite these objective findings, a limitation of this case is a possible coexisting hematologic malignancy, particularly lymphoma, with sarcoidosis as noted in the literature.14 Occult lymphomas have been documented to emerge later after the initial diagnosis of HLH with appropriate treatment.2 Treatment with corticosteroids and chemotherapy before splenectomy may have also masked an underlying lymphoma in this case.15
For primary HLH, the HLH-94 and HLH-2004 treatment protocols involve a combination of dexamethasone, etoposide, and cyclosporine. In contrast, treatment of secondary HLH is usually adapted toward the underlying etiology.2 Therapy for secondary HLH should be initiated in a timely manner to avoid rapid progression to critical illness and multiorgan failure. Systemic glucocorticoids are the first-line therapy for both HLH and hepatic sarcoidosis, with at least a 60% therapeutic response in hepatic sarcoidosis.16,17 Cyclosporine or anakinra may be introduced if no immediate response in HLH from glucocorticoids is detected.7 Reduced dose etoposide may be reserved for refractory cases of HLH, however in this case it was used due to suspicion of lymphoproliferation.
Second-line therapy for hepatic sarcoidosis also includes azathioprine, methotrexate, and TNF-alpha inhibitors, although these drugs carry risk of worsening HLH via infection and should be used cautiously.17 Liver transplantation, assuming appropriate candidacy, is the final intervention for hepatic sarcoidosis if refractory to medical therapy and requires lifelong immunosuppressive therapy for prevention of allograft rejection and disease relapse.18 Overall, there is scarce literature on treating HLH and hepatic sarcoidosis simultaneously due to the rarity of this association. Therefore, multidisciplinary assessment by hematology, rheumatology, and hepatology is necessary, as the hyperinflammatory state, coagulopathy, and risk of multiorgan failure in the context of MAS/HLH pose barriers to optimal treatment of hepatic sarcoidosis. In a previous case, a patient was noted to have improved with steroid therapy and modified HLH-94 protocol, unlike in this case.11 Notably, HLH biomarkers in our patient status post splenectomy improved, though with complication of postprocedural bleeding, which has been observed in previous studies in patients with HLH of unknown cause.19 Prognosis of HLH related to hepatic sarcoidosis remains unclear, though an approximate 70% 1-year survival has been reported in patients with autoimmune-associated MAS/HLH.20
This case describes a patient with MAS/HLH ultimately found to have hepatic sarcoidosis as the likely underlying inflammatory trigger, a rarely reported association in the literature. Clinicians should maintain a broad differential including infection, malignancy, autoimmune disorders, and infiltrative disease when observing acute liver injury while suspecting secondary HLH. Early tissue diagnosis, particularly liver biopsy in cases with significant hepatic involvement, may be critical in identifying uncommon secondary causes of HLH and excluding infiltrative malignancy.
Multidisciplinary review is imperative to balance risks and benefits of treating hepatic sarcoidosis and MAS/HLH simultaneously.
DISCLOSURES
Author contributions: All persons listed have met all 4 criteria for authorship per ICMJE guidelines, including providing substantial contribution to the intellectual content and supplementary material of the case report, critically reviewing each draft, approving the final version of the report, and holding accountability for accuracy and integrity of the work. A. Mencia, W.
Jiang, and J. Qualtieri provided images and accurate descriptions of the pathology figures incorporated into this manuscript. H.
Shariff is the article guarantor.
Financial disclosure: None to report.
Informed consent was obtained for this case report.