Abstract
Hemophagocytic lymphohistiocytosis is a life-threatening disorder characterized by persistent pathologic activation of cytotoxic T lymphocytes, natural killer cells, and macrophages. We present details of a young patient who presented with high-grade fever, jaundice, and breathlessness. On investigations, he had hepatitis, anemia, neutropenia, and coagulopathy. He also had hypertriglyceridemia, hypofibrinogenemia, and hyperferritinemia. Bone marrow aspiration revealed histiocytosis, and transjugular liver biopsy revealed necrotizing granulomas positive for Mycobacterium tuberculosis on acid-fast bacilli staining. He was successfully managed with a combination of immunosuppressants and antitubercular therapy. Tuberculosis associated hemophagocytosis syndrome is rare and should be considered in patients with unexplained hemophagocytosis syndrome, especially in tuberculosis-endemic regions. Prompt recognition and treatment with antitubercular treatment and immunosuppressants are associated with good outcomes.
Keywords: hemophagocytic lymphohistiocytosis, tuberculosis, HLH, lymph node, PUO
Abbreviations: HLH, hemophagocytic lymphohistiocytosis; CTL, cytotoxic T lymphocyte; EBV, Epstein-Barr virus; CMV, cytomegalovirus; HSV, herpes simplex virus; ATT, antitubercular therapy; CAM, complementary and alternative medicine; ALT, alanine aminotransferase; AST, aspartate aminotransferase; ALP, alkaline phosphatase; INR, international normalized ratio; CRP, C-reactive protein; LDH, lactate dehydrogenase; HIV, human immunodeficiency virus; SAAG, serum-ascites albumin gradient; ADA, adenosine deaminase; AFB, acid-fast bacilli; TJLB, transjugular liver biopsy; PTH, parathyroid hormone; NK cell, natural killer cell; TB-HLH, tuberculosis-associated hemophagocytic lymphohistiocytosis
Hemophagocytic syndrome or hemophagocytic lymphohistiocytosis (HLH) is an immunodysregulatory syndrome characterized by widespread systemic activation of natural killer cells (NK cells) and cytotoxic T lymphocytes (CTLs), leading to systemic infiltration of these cells and macrophages and a clinical presentation as a multisystemic inflammatory syndrome due to hypercytokinemia.1 It is characterized by a spectrum of clinicopathologic findings such as fever, hepatosplenomegaly, cytopenias, hypertriglyceridemia, hypofibrinogenemia, hyperferritinemia, and demonstration of hemophagocytosis. It is either primary (genetic) or secondary (acquired). Primary HLH commonly presents in early childhood and arises as a result of a genetic predisposition due to mutation in the genes involved in NK cell and CTL function, most commonly perforin.1 Secondary HLH presents in adulthood and is often precipitated by infections, autoimmune conditions, immunosuppressive therapy, and malignancies. All types of pathogens have been associated with acquired HLH, most commonly viruses (including Epstein-Barr virus [EBV], cytomegalovirus [CMV], and herpes simplex virus), bacteria (including tuberculosis), fungi, and parasites. Secondary HLH due to tuberculosis in immunocompetent persons is very rare and has close to 50% mortality.2 Here, we describe a case of granulomatous tubercular hepatitis associated with secondary HLH managed successfully with a combination of immunosuppressants and antitubercular therapy (ATT).
Case presentation
A 25-year-old gentleman presented to All India Institute of Medical Sciences, New Delhi, with complaints of high-grade intermittent fever for 4 weeks, jaundice for 3 weeks, and shortness of breath for 3 days. He denied any jaundice outbreak in the vicinity or among close contacts, alcohol or complementary and alternative medicine intake, blood transfusion, surgery, tattooing, or high-risk behavior. He had no history of hematemesis, melena, decreased urine output, or altered sensorium. He had no history of localizing symptoms in the form of cough, expectoration, burning micturition, or loose stools. He had no chest pain, palpitations, or orthopnea. He denied any history of tuberculosis in close contacts. At presentation, on general examination, he had tachycardia and tachypnea and was febrile, anemic, and icteric. On systemic examination, he had evidence of decreased breath sounds in the right infrascapular region and had nontender hepatomegaly and mild ascites. There was no peripheral lymphadenopathy, and examination of the central nervous system and cardiovascular system was unremarkable. In view of his condition, he was admitted to the intensive care unit.
On investigation, his complete blood counts showed a hemoglobin level of 8.9 g/dL, total leukocyte count of 2040/mm3 with a differential of 41% neutrophils and 25% lymphocytes, and a platelet count of 1.72 × 105/mm3. His liver functions revealed a total bilirubin level of 17.5 mg/dL with a direct fraction of 15.9 g/dL, alanine aminotransferase (ALT) level of 600 IU/L (normal <40 IU/L), aspartate aminotransferase (AST) level of 282 U/L (normal <40 IU/L), and alkaline phosphatase level of 111 U/L (normal <280 IU/L). His kidney function tests were normal. The international normalized ratio was 5.2. The C-reactive protein (CRP) level was 45.95 mg/L (normal = 0–10 mg/L), and the lactate dehydrogenase level was 1317 U/L (normal = 140–280 IU/L). A chest roentgenogram revealed bilateral mild pleural effusion and right lower zone infiltrates. Ultrasound of the abdomen revealed hepatomegaly with ascites. His viral markers for hepatitis A, B, C, and E were negative. Malarial parasites were not demonstrated on peripheral smear or rapid diagnostic tests. Dengue, EBV and CMV, and HIV-1 and 2 serology was negative. His tropical fever workup for scrub typhus and leptospirosis was negative. His ascitic fluid workup revealed an acellular fluid with a low serum-ascites albumin gradient of 0.6 and adenosine deaminase (ADA) level of 11 IU/L (normal <36 IU/L). Pleural fluid had a cell count of 250/mm3 with 80% lymphocytes and exudative nature with the protein level of 2.6 g/dL (serum protein of 4.1 g/dL) and ADA level of 23 IU/L. Gram staining and acid-fast bacilli (AFB) staining on fluid samples were negative. The serum triglyceride level was 654 mg/dL, ferritin level was 34,547 ng/mL, and fibrinogen level was 15 mg/dL. Computed tomography of the chest and abdomen revealed bilateral pleural effusion (right > left), right lower zone collapse-consolidation, and no mediastinal or abdominal lymph nodes. There was mild to moderate ascites with an enlarged spleen of 14.6 cm. Bone marrow aspiration and biopsy revealed a cellular bone marrow with adequate myeloid, erythroid, and megakaryocytic elements and an increased number of histiocytes. However, no hemophagocytosis, granuloma, lymphomatous deposit, or pathogenic organism was demonstrable. In view of liver dysfunction, coagulopathy, and ascites, he underwent a transjugular liver biopsy which revealed active hepatitis with multiple large necrotizing epitheloid cell granulomas, which stained positive for AFB, consistent with tuberculosis; there was no associated hemophagocytosis demonstrable (Figure 1). NK cell testing carried out from the immunogenetics laboratory revealed a below-normal (82.5%) activity consistent with HLH. In discussion with a hematologist, the patient was started on injectable dexamethasone (10 mg/m2/day) and intravenous immunoglobulin at a dose of 400 mg/kg/day for 5 days along with modified ATT in the form of amikacin (15 mg/kg/d), levofloxacin (500 mg/d), and ethambutol (15 mg/kg/d) in view of severe hepatic dysfunction. During the course of treatment, he also developed severe hypercalcemia, which was parathyroid hormone independent and was a consequence of increased levels of vitamin D3 as a result of the granulomatous disease, and he was successfully managed with saline diuresis and zoledronic acid. The patient's fever responded to treatment, his cytopenias resolved, and his liver function tests also showed an improving trend. Within 1 week of initiating therapy, his ferritin level decreased to 8142 ng/mL and CRP level decreased to 23.02 mg/L. Once his liver functions improved, rifampicin (10 mg/kg/d) and isoniazid (5 mg/kg/d) were introduced, and pyrazinamide was avoided in view of hepatic synthetic dysfunction; he was discharged subsequently. On his last follow-up 1 month after discharge, he was doing well with no fever and breathlessness and a marked decrease in jaundice; his liver functions had improved with a total bilirubin level of 2.9 mg/dL and direct fraction of 1.5 mg/dL; AST and ALT levels were 91 U/L and 61 U/L, respectively.
Figure 1.
A) Photomicrographs of liver core biopsy specimen (x 40) shows presence of large, confluent lobular granulomas (black arrow) with irregular macrovesicular steatosis ( blue arrows). B) and C) A magnified view (x 100) of the large granulomas (black arrows) and macrovesicular steatosis (blue arrow) and ballooning of hepatocytes. D) A further magnified view ( x 200) shows necrosis in some of the granulomas (black arrow).
Discussion
HLH is a life-threatening immunodysregulatory disorder characterized by persistent and pathological activation of CTLs and NK cells, leading to a cytokine storm and activation of macrophages, which leads to multisystem organ dysfunction.3 It typically presents as a clinicopathologic constellation of findings such as fever, cytopenias, organomegaly, hepatitis, hypertriglyceridemia, hypofibrinogenemia, hyperferritinemia, and demonstration of hemophagocytosis. It may be either primary or secondary. Secondary or acquired HLH occurs owing to a number of triggers such as malignancies, autoimmune conditions, and infections, most commonly EBV and other infections, that evoke a strong Th1 cell response. The basic defect is uncontrolled activation of the CTLs and NK cells due to the inefficient tackling of their antigenic stimulus and failure to clear it, leading to their persistent and uncontrolled activation. This persistent activation leads to an immune storm, which further triggers macrophage activation. Hemophagocytosis occurs owing to macrophages engulfing hematopoietic cells and their precursors. There is no gold standard test to diagnose HLH, and diagnosis is typically made by clinical suspicion and fulfillment of five of the eight following criteria as suggested by the Histiocyte Society: (1) fever, (2) splenomegaly, (3) cytopenias affecting 2 cell lines (hemoglobin level <9 g/dL, absolute neutrophil count <1000/mm3, and platelet count <105/mm3), (4) hypertriglyceridemia (>265 mg/dL) and/or hypofibrinogenemia (<150 mg/dL), (5) hyperferritinemia (>500 μg/L), (6) hemophagocytosis in the bone marrow, spleen, or lymph nodes, (7) absence of or low NK cell activity, and (8) soluble CD25 level >2400 IU/mL.4 Our case fulfilled six of the eight criteria, although hemophagocytosis was per se not demonstrable. The demonstration of hemophagocytosis is neither specific nor necessary for the diagnosis of HLH as it might be cyclical.1 Our patient had a ferritin value of 34,547 ng/mL; such high ferritin values are rarely seen with other conditions; a retrospective analysis reported that ferritin levels >10,000 ng/mL are 96% specific for HLH and that addition of fever to support the diagnosis increases the specificity to 98%.5
Tuberculosis-associated HLH (TB-HLH) is rare. Brastianos et al.6 in 2006 had published a case of disseminated TB-HLH and reviewed the literature published in the English language for 36 other cases published until then and found that the mortality rate was approximately 50% and that all patients had demonstrable hemophagocytosis on bone marrow aspiration and biopsy. Of the 37 patients, 29 had been treated with either ATT alone (9 cases) or a combination of ATT with immunosuppression (20 cases); 7 of 9 of those on ATT alone survived, as compared with 12 of 20 of those on combination therapy; of those who did not receive therapy, none survived. Another review of the literature was undertaken by Padhi et al2 in 2015 who reviewed the literature published in the English language from 1975 to 2014 and found 63 cases (including those reviewed by Brastianos et al.6). They demonstrated a high mortality rate of 49% and, on univariate analysis, demonstrated that age >30 years, the presence of comorbidities, hemophagocytosis on the bone marrow, and delayed use of ATT were associated with poor prognosis. A majority (65%) of patients had associated comorbidities in the form of end-stage renal disease, requiring either hemodialysis or renal transplant, malignancies, and other autoimmune diseases. Luckily, our patient had none of the poor prognostic markers. Antitubercular treatment was initiated on day 7 of admission, and he survived. The hepatitis in our case was likely to be due to a combination of tuberculosis-associated granulomatous hepatitis and HLH per se. Liver dysfunction due to HLH is very common.7 The majority of cases of TB-HLH described in the literature have extrapulmonary tuberculosis and disseminated tuberculosis. Our case had evidence of granulomatous tubercular hepatitis and no evidence of tuberculosis elsewhere, as demonstrated by a low ADA level, negative microbiology for AFB staining in ascitic and pleural fluid, and lack of any features suggestive of tuberculosis on the chest and abdominal imaging. We further analyzed 18 more cases published since 2014, including our case, along with the previously published cases.8, 9, 10, 11, 12, 13, 14, 15 Our analysis showed that of 82 patients, 58.5% were men, and the median age of presentation was 40 years. The patients presented with a median duration of 35 days, and all had fever (100%), while majority had hepatomegaly (70%), splenomegaly (77.9%), and lymphadenopathy (62.2%) (Table 1). Majority of patients received a combination of ATT and immunosuppressants (69.5%), whereas 19.5% and 1.2% received ATT or immunosuppressants alone, respectively. Only a minority received hepatic modified ATT (15%). The outcome was poor with a mortality rate of 45.1% (Table 2). HLH was traditionally treated with an 8-week combination of immunosuppressive drugs such as cyclosporine A, steroids, and etoposide.4 Hematopoietic stem cell transplant is indicated in select cases. In secondary HLH, treatment of the triggering agent is recommended. In our case, the trigger was successfully controlled with a combination of ATT and immunosuppressants.
Table 1.
Clinicopathologic Characteristics of Tuberculosis in Patients With Hemophagocytic Lymphohistiocytosis.
| Parameter | Value (n = 82) |
|---|---|
| Age (years) | 40 (23–60) |
| Male | 48 (58.5%) |
| Duration of symptoms (days) | 35 (15–60) |
| Fever | 82/82 (100%) |
| Hepatomegaly | 49/70 (70%) |
| Splenomegaly | 60/77 (77.9%) |
| Lymphadenopathy | 28/45 (62.2%) |
| Hemoglobin (g/dL) (n = 65), mean ± SD | 8.1 ± 2.2 |
| ANC (per mm3) (n = 60) | 1217 (782.5–3030) |
| Platelet count (105/mm3) (n = 67) | 60 (37–113) |
| Peak bilirubin (mg/dl) (n = 18) | 3.4 (1.37–10.6) |
| Ferritin (ng/mL) (n = 52) | 2469 (1447.5–6431) |
| Triglycerides (mg/dL) (n = 44) | 282.5 (229.7–388.5) |
SD = standard deviation; IQR = interquartile range; ANC, absolute neutrophil count.
All values are expressed as n (%) or median (IQR), unless otherwise specified.
Table 2.
Treatment and Outcomes of Patients With Tuberculosis and Hemophagocytic Lymphohistiocytosis.
| Treatment | n = 82 |
|---|---|
| No treatment | 8/82 (9.8%) |
| Antitubercular therapy alone | 16/82 (19.5%) |
| Antitubercular with immunosuppressants | 57/82 (69.5%) |
| Immunosuppressants alone | 1/82 (1.2%) |
| Modified antitubercular therapy | 11/73 (15%) |
| Antitubercular therapy initiated first | 11/44 (40.9%) |
| Immunosuppressant initiated first | 11/44 (25%) |
| Antitubercular and immunosuppressant started together | 15/44 (34.1%) |
| Death | 37/82 (45.1%) |
HLH should be considered as a differential diagnosis in patients presenting with fever and jaundice. In cases of secondary HLH, tuberculosis should be considered as a potential trigger because giving immunosuppressive therapy alone in these patients entails a risk of a flare of tuberculosis. These patients have deranged liver function tests owing to HLH per se and tubercular granulomatous hepatitis, which makes initiation of ATT challenging. Prompt initiation of ATT can lead to rapid improvement and recovery.
Authors' contributions
A.E. contributed to data curation and draft writing. A.A. contributed to draft writing. A.G. contributed to conceptualization and draft writing. M.A. contributed to draft writing. P.D. contributed to draft writing. S. contributed to conceptualization, draft revision, and supervision.
Conflicts of interest
The authors have none to declare.
Funding
None.
References
- 1.Rosado F.G.N., Kim A.S. Hemophagocytic lymphohistiocytosis: an update on diagnosis and pathogenesis. Am J Clin Pathol. 2013 Jun;139:713–727. doi: 10.1309/AJCP4ZDKJ4ICOUAT. [DOI] [PubMed] [Google Scholar]
- 2.Padhi S., Ravichandran K., Sahoo J., Varghese R.G., Basheer A. Hemophagocytic lymphohistiocytosis: an unusual complication in disseminated Mycobacterium tuberculosis. Lung India. 2015;32:593–601. doi: 10.4103/0970-2113.168100. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Al-Samkari H., Berliner N. Hemophagocytic lymphohistiocytosis. Annual review of pathology: mechanisms of disease. 2018;13:27–49. doi: 10.1146/annurev-pathol-020117-043625. [DOI] [PubMed] [Google Scholar]
- 4.Henter J.-I., Horne A., Aricó M. HLH-2004: diagnostic and therapeutic guidelines for hemophagocytic lymphohistiocytosis. Pediatr Blood Canc. 2007 Feb;48:124–131. doi: 10.1002/pbc.21039. [DOI] [PubMed] [Google Scholar]
- 5.Allen C.E., Yu X., Kozinetz C.A., McClain K.L. Highly elevated ferritin levels and the diagnosis of hemophagocytic lymphohistiocytosis. Pediatr Blood Canc. 2008 Jun;50:1227–1235. doi: 10.1002/pbc.21423. [DOI] [PubMed] [Google Scholar]
- 6.Brastianos P.K., Swanson J.W., Torbenson M., Sperati J., Karakousis P.C. Tuberculosis-associated haemophagocytic syndrome. Lancet Infect Dis. 2006 Jul;6:447–454. doi: 10.1016/S1473-3099(06)70524-2. [DOI] [PubMed] [Google Scholar]
- 7.Dong J., Xie F., Jia L. Clinical characteristics of liver failure with hemophagocytic lymphohistiocytosis. Sci Rep. 2019 May 31;9:1–6. doi: 10.1038/s41598-019-43909-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Zhang Y., Liang G., Qin H., Li Y., Zeng X. Tuberculosis-associated hemophagocytic lymphohistiocytosis with initial presentation of fever of unknown origin in a general hospital: an analysis of 8 clinical cases. Medicine (Baltim) 2017 Apr;96 doi: 10.1097/MD.0000000000006575. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Rathnayake P.V.T.M., Kularathne W.K.S., De Silva G.C.V. Disseminated tuberculosis presenting as hemophagocytic lymphohistiocytosis in an immunocompetent adult patient: a case report. J Med Case Rep. 2015 Dec 29;9:294. doi: 10.1186/s13256-015-0772-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Kodan P., Chakrapani M., Shetty M., Pavan R., Bhat P. Hemophagocytic lymphohistiocytosis secondary to infections: a tropical experience! J Postgrad Med. 2015 Apr 1;61:112. doi: 10.4103/0022-3859.150904. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Haque W.M.D.M.U., Shuvo MdER., Rahim M.A., Mitra P., Samad T., Haque J.A. Haemophagocytic syndrome in an adult suffering from pyrexia of unknown origin: an uncommon presentation of tuberculosis: a case report. BMC Res Notes. 2017 Feb 27;10:110. doi: 10.1186/s13104-017-2434-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Seo J.-H., Lee J.A., Kim D.H., Cho J., Lim J.S. Tuberculosis-associated hemophagocytic lymphohistiocytosis in adolescent diagnosed by polymerase chain reaction. Korean J Pediatr. 2016 Jan;59:43–46. doi: 10.3345/kjp.2016.59.1.43. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Asaji M., Tobino K., Murakami K. Miliary tuberculosis in a young woman with hemophagocytic syndrome: a case report and literature review. Intern Med. 2017 Jun 15;56:1591–1596. doi: 10.2169/internalmedicine.56.8025. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Hui Y.M.T., Pillinger T., Luqmani A., Cooper N. Haemophagocytic lymphohistiocytosis associated with Mycobacterium tuberculosis infection. BMJ Case Rep. 2015 Apr 13:2015. doi: 10.1136/bcr-2014-208220. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Long B., Cheng L., Lai S. Tuberculosis-associated hemophagocytic lymphohistiocytosis in an umbilical cord blood transplant recipient. Clin Chim Acta. 2017 May;468:111–113. doi: 10.1016/j.cca.2016.12.028. [DOI] [PubMed] [Google Scholar]

