Indian Journal of Medical Specialities

: 2019  |  Volume : 10  |  Issue : 4  |  Page : 225--228

Idiopathic hepatic granulomatosis: A challenging case of fever of unknown origin

Ashutosh Garg, Vivek Pal Singh, Jai Khullar, Afreen Khan 
 Department of Internal Medicine, BLK Super Speciality Hospital, New Delhi, India

Correspondence Address:
Dr. Vivek Pal Singh
Department of Internal Medicine, BLK Super Speciality Hospital, Pusa Road, New Delhi - 110 005


For any physician, fever of unknown origin (FUO) is an enigmatic problem, while being equally frustrating. It tests the limits of patience of both the physician and the patient. While the majority of such cases end up being of infective etiology, the rest are attributable to noninfectious inflammatory disorders (NIIDs) and neoplasms. Among the NIIDs, most turn out to be connective tissue disorders (CTDs). CTDs are commonly known to have granulomas on histopathology. However, hepatic granulomas are rarely encountered. We present a case of FUO in an elderly urban North Indian businessman who after thorough workup was diagnosed with idiopathic hepatic granulomatous disease. Response to disease-modifying therapy was rewarding.

How to cite this article:
Garg A, Singh VP, Khullar J, Khan A. Idiopathic hepatic granulomatosis: A challenging case of fever of unknown origin.Indian J Med Spec 2019;10:225-228

How to cite this URL:
Garg A, Singh VP, Khullar J, Khan A. Idiopathic hepatic granulomatosis: A challenging case of fever of unknown origin. Indian J Med Spec [serial online] 2019 [cited 2021 Mar 9 ];10:225-228
Available from:

Full Text


Classic fever of unknown origin (FUO) refers to a fever of at least 101 °F present on at least 2 occasions for at least 3 weeks.[1],[2] The patient should be immunocompetent, and even after taking extensive history, thorough physical examination, and obligatory investigations, the etiological diagnosis remains uncertain. Etiologies of classic FUO can be grouped under the following: infectious, noninfectious inflammatory disorders (NIIDs), and neoplastic disorders. 48% of Indian FUO patients are ultimately diagnosed with an infection.[3] Tuberculosis (TB) is the leading infectious cause (61%).[4] Neoplasms and NIIDs account for 21.6% and 20.6% of such cases. In some NIID patients with granulomas on hepatic histopathological specimens, no etiology can be found. These patients are labeled to have idiopathic hepatic granulomatous disease (IHG). Various international studies from 1950s to 1990s, including the archetypal study on FUO by Petersdorf and Beeson in 1961, have quoted IHG to be responsible for 5%–50% of cases.[3],[5],[6],[7],[8] We herein present an interesting case of IHG.

 Case Report

A 65-year-old North Indian businessman (building construction) developed intermittent fever associated with chills and sore throat in April 2018. The sore throat subsided but fever persisted. He had been investigated for persistent fever [Table 1]. Empirical antibiotics and Vitamin D supplementation were given. Although fever intensity reduced, he felt distressed. In July 2018, he had occult hematuria with dysmorphic red blood cells (RBCs) in urine. He received empirical antibiotics which provided temporary relief. He lost 2 kg in the next 4 weeks despite normal appetite and no significant fatigue. Contract-enhanced computed tomography (CECT) of the chest and abdomen revealed few prominent nonnecrotic lymph nodes in mediastinum (largest 17 mm × 12 mm) and periportal (subcentimetric) regions. He lost another 5 kg by October 2018. Review history at our center was taken which revealed past history of alcohol consumption, a positive sexual history with a commercial sexual worker (about 10–15 years back), and smoking cessation since April 2018 with 15 pack-years exposure. There was past history of a road traffic crash in 1988 when he sustained a fracture in his left proximal femur for which intramedullary nailing was done. There was no past history of TB, diabetes, jaundice, blood transfusion, animal contact, relevant travel history, other illicit substance use, or family history of connective tissue disorders (CTDs). On physical examination, he had no pallor, icterus, palpable lymphadenopathy, or clubbing. His vital signs and systemic examination were noncontributory. All his previous medications – which included antibiotics – were stopped and he was advised oral paracetamol on as-needed basis. Investigations [Table 2], [Table 3], [Table 4] revealed raised inflammatory markers, atypical p-antineutrophil cytoplasmic antibodies (p-ANCA) weak positivity, prediabetes, and a positron emission tomography (PET)/CECT whole body scan revealing mediastinal and abdominal lymphadenopathy. The patient was advised to get a liver biopsy and a bone marrow aspirate cum biopsy done but the patient was unwilling to do so at this point in time. A provisional diagnosis of FUO with significant unintentional weight loss, generalized lymphadenopathy, weak p-ANCA positivity, and raised inflammatory markers was made. Endobronchial ultrasound-guided fine needle aspiration and cytology was done from paratracheal nodes which did not reveal evidence of TB, malignancy, or granuloma. He was started on World Health Organization (WHO) category 1 antituberculosis treatment (ATT) and was asked to follow up after 3 weeks. In view of no significant clinical improvement, oral methylprednisolone (MP) 16 mg was added.{Table 1}{Table 2}{Table 3}{Table 4}

The patient also showed at another hospital wherein Epstein–Barr virus (EBV) reverse transcriptase polymerase chain reaction (RT-PCR) was found to be positive. No changes were made to his treatment.

In November 2018, his fever pattern changed from quotidian to weekly. A rheumatological consultation was taken. His ATT was switched to the continuation phase of WHO category 1, and oral MP was uptitrated to 16 mg twice daily.

In December 2018, the patient started complaining of painful swelling in the left knee joint which was associated with fever. Aspirate from the left knee joint did not show any significant leukocytes or crystals. Along with ATT and oral MP (16 mg in morning and 12 mg at night), oral hydroxychloroquine (HCQS) 200 mg twice a day was added. In January 2019, oral MP dose was changed to 16 mg in morning and 8 mg at night. A CECT chest and abdomen done in February 2019 showed no difference from previous findings. Oral MP was downtitrated to 8 mg once daily.

Patient's complaints did not subside, and he was distressed. Finally, he agreed for liver biopsy in March 2019 which revealed ill-defined occasional nonnecrotizing parenchymal granulomas [Figure 1], white arrow]. EBV RT-PCR and p-ANCA during the same time were found to be negative. In April 2019, oral MP was further downtitrated to 4 mg once a day. 6-month ATT course was over. HCQS was withdrawn. In view of continued fever and to assess disease activity, whole body PET/CECT was redone which showed no difference from previous study. Colonoscopy and fundoscopy were noncontributory. He was started on MP i.v., 500 mg once a day, oral mycophenolate mofetil (MMF) 500 mg twice a day, oral HCQS 200 mg twice a day, following which he became afebrile 3rd day onward. He was discharged on oral MMF 750 mg twice a day, oral HCQS 200 mg twice a day, oral MP 48 mg once daily, and oral colchicine 0.5 mg twice a day. He continued to be afebrile at 4 weeks, with subjective feeling of being in good health, and the inflammatory markers sharply declined (C-reactive protein was 0.63 mg/dL).{Figure 1}


Making an etiological diagnosis in a patient presenting as FUO can be a very tedious exercise for the treating physician as well as the suffering patient. FUO in itself is not a complex entity. In fact, most people have common diseases presenting with atypical manifestations.[9] These include, among others, endocarditis, diverticulitis, extra pulmonary TB, and vertebral osteomyelitis. In India, a trial of ATT is often prescribed in FUO cases, but in the present case was not helpful, except for the fact that it ruled out TB. PET scans have revolutionized the way a case of FUO is approached in the first few steps of diagnostic failure.[10] This case is a testament to its reliability. Routine culture in a seemingly noninfectious illness may be misleading. In this case also, a urine culture showing Escherichia coli is indicative of colonization rather than infection since history and urinalysis done at the same time were not supportive of the same. Similarly, the presence of dysmorphic RBCs at a later date could have blindsided an unassuming physician.

Two other tests – atypical p-ANCA and EBV RT-PCR – which were positive once and negative at a later date are also problematic and could have serious implications if not viewed with skepticism. Atypical p-ANCA is a useful marker to differentiate ulcerative colitis from Crohn's disease, but its utility in diagnosing inflammatory bowel disease is manifestations.[11] EBV is much more common in immunocompromised individuals. On top of it, the current RT-PCR assays can detect as low as 2 copies of EBV DNA in serum. Pertaining to this particular case, two different laboratories had contradictory results (although spread apart in time). It has been seen that there is a considerable variability in proficiency testing when it comes to precision-oriented tests like RT-PCR and such results should be taken with a pinch of salt.[12]

The key investigation which clinched the diagnosis in this case was liver biopsy, and as we have pointed out earlier, it was offered to the patient at first presentation in October 2018. Due to lack of consent initially, the patient could only be persuaded to do so in March 2019. The diagnosis of idiopathic hepatic granuloma was made after excluding relevant causes.

Just like the broad categories for FUO, hepatic granulomas can be infectious, noninfectious, and neoplastic in origin. The underlying relevant causes in Indian context for each category are:

Infectious: TB, other mycobacteria, actinomycosis, brucellosis, syphilis, toxoplasma, hepatitis C, CytomegalovirusNoninfectious: Drugs such as allopurinol, quinidine, and sulfonamides; primary biliary cirrhosis; CTDsNeoplastic: Hodgkin's lymphoma.[13]

On histopathological basis, there are different kinds of granulomas such as fibrin-ring granuloma, microgranuloma, lipogranuloma, stellate microabscesses containing granuloma, foamy macrophage aggregates, and predominantly suppurative granulomas. Different etiologies are present as one or the other kind of these varieties. A full discussion of these pathological variants is not within the scope of this article. Thus, with a germane history directed toward the possible pathogen or entity coupled with the histological specimen, it should not be difficult to make a diagnosis with adequate precision.

Usually, all these disorders present with deranged liver functions and systemic manifestations. Our patient did not have any localizing clues on history or examination. Liver biopsy was negative for acid-fast bacilli, and hence, the granulomas on histopathology were labeled as idiopathic.

Finally, we would like to draw attention to the heart of the problem in most cases of FUO. Mostly, these patients have a common disease with an uncommon presentation. The person in this case report is an exception. To evaluate a case of FUO, all that we need to keep in mind is to be thorough at every stage of assessment, be patient, reassuring, and be unpretentious lest we may complicate a simple case. This is not to say that a patient labeled with FUO will be a straightforward case. However, most are likely to be. For the rest, suffice it to restate what Albert Einstein is quoted to have said when asked about his rigor at unifying scientific theories: “Everything should be as simple as it can be, but not simpler.” After all where is the fun in medicine if not in the intricacies of diagnosis making!

Declaration of patient consent

The authors certify that they have obtained all appropriate patient consent forms. In the form the patient(s) has/have given his/her/their consent for his/her/their images and other clinical information to be reported in the journal. The patients understand that their names and initials will not be published and due efforts will be made to conceal their identity, but anonymity cannot be guaranteed.

Financial support and sponsorship


Conflicts of interest

There are no conflicts of interest.


1Wright W, Mackowiak P. Fever of unknown origin. In: Bennett JE, Dolin R, Blaser MJ, editors. Mandell, Douglas, and Bennett's Principles and Practice of Infectious Diseases. 8th ed. Elsevier; 2014. p. 721-31.
2Petersdorf RG, Beeson PB. Fever of unexplained origin: Report on 100 cases. Medicine (Baltimore) 1961;40:1-30.
3Rupali P, Garg D, Viggweswarupu S, Sudarsanam TD, Jeyaseelan V, Abraham OC, et al. Etiology of classic fever of unknown origin (FUO) among immunocompetent Indian adults. J Assoc Physicians India 2019;67:21-6.
4Sharma BK, Kumari S, Varma SC, Sagar S, Singh S. Prolonged undiagnosed fever in northern india. Trop Geogr Med 1992;44:32-6.
5Choe KW, Park SG, Choi SJ. Clinical study on fever of unknown origin. Korean J Infect Dis 1984;16:1-7.
6Simon HB, Wolff SM. Granulomatous hepatitis and prolonged fever of unknown origin: A study of 13 patients. Medicine (Baltimore) 1973;52:1-21.
7Sartin JS, Walker RC. Granulomatous hepatitis: A retrospective review of 88 cases at the mayo clinic. Mayo Clin Proc 1991;66:914-8.
8Klatskin G, Yesner R. Hepatic manifestations of sarcoidosis and other granulomatous diseases; A study based on histological examination of tissue obtained by needle biopsy of the liver. Yale J Biol Med 1950;23:207-48.
9Bleeker-Rovers C, van der Meer J. Fever of unknown origin. In: Jameson JL, Fauci AS, Kasper DL, Hauser SL, Longo DL, Loscalzo J, editors. Harrison's Principles of Internal Medicine. 20th ed. Vol. 17. New York: McGraw-Hill Education; 2018. p. 114-22.
10Sioka C, Assimakopoulos A, Fotopoulos A. The diagnostic role of (18) F fluorodeoxyglucose positron emission tomography in patients with fever of unknown origin. Eur J Clin Invest 2015;45:601-8.
11Mokhtarifar A, Ganji A, Sadrneshin M, Bahari A, Esmaeilzadeh A, Ghafarzadegan K, et al. Diagnostic value of ASCA and atypical p-ANCA in differential diagnosis of inflammatory bowel disease. Middle East J Dig Dis 2013;5:93-7.
12Hayden RT, Yan X, Wick MT, Rodriguez AB, Xiong X, Ginocchio CC, et al. Factors contributing to variability of quantitative viral PCR results in proficiency testing samples: A multivariate analysis. J Clin Microbiol 2012;50:337-45.
13Lamps LW. Hepatic granulomas: A Review with emphasis on infectious causes. Arch Pathol Lab Med 2015;139:867-75.