Hemophagocytic lymphohistiocytosis as a diagnostic consideration of fever of unknown origin with pancytopenia and chronic liver disease

Abstract

Hemophagocytic lymphohistiocytosis (HLH) is a severe disorder of systemic immune dysregulation which can be primary or secondary to autoimmune disorders, malignancy, or infections. We hereby describe a case of a 23-year-old male with severe hepatitis along with pancytopenia and prolonged fever of unknown origin that developed HLH triggered by staphylococcal urinary tract infection. This is a discussion of this unusual disease and its presentation and the diagnostic difficulties which may be encountered in general clinical practice.

Introduction

Viral infection especially Epstein—Barr virus (EBV) is the most common trigger for secondary hemophagocytic lymphohistiocytosis (HLH) (29%). Studies have suggested infections, such as Mycobacterium, Plasmodium, hepatitis E, kala azar, malaria, Leptospira, etc., associated with HLH in tropics.[1] But staphylococcal urinary tract infection (UTI) as a cause has been less commonly described.[1]

This case illustrates the importance of thorough history taking, early diagnosis, and complete workup of a case of fever along with awareness among healthcare professionals about the pitfalls of commonly used diagnostic tests.

Case Presentation

One year back, at the age of 22 years (23 years old now), our patient had his first episode of high-grade fever associated with chills and rigors. He was evaluated in a local hospital where a diagnosis of typhoid was made based on a single Widal titer report (1:80 for both anti-H and anti-O which was below the epidemiologic cut-off for the area- 1:80 for anti-O and 1:160 for anti-H)[2]). No blood culture was done. He was treated with several courses of antibiotics for over 3 months, but his fever continued to occur on and off since then.

He was then evaluated in a tertiary care center where his liver enzymes were found to be elevated. The patient and his family refused further investigations and took a leave against medical advice, resorting to faith healing. The patient's condition worsened, and he developed progressively increasing abdominal distension, bilateral pedal edema, and jaundice over the next 6 months. This is when the patient was brought to us.

There was no history of jaundice before the onset of the current illness, no prior blood transfusions, no history of intravenous (IV) drug abuse or multiple sexual partners, joint pains, rash, recurrent oral ulcers, chest complaints, palpitations, any unusual bleeding, and neuropsychiatric complaints. The patient is a nonsmoker and nonalcoholic with no history of similar illness in other family members.

On general survey, the patient was very ill [World Health Organization (WHO) performance status-Grade 3] and wasted with body mass index of 16.8 kg/m², febrile (102 F), with a blood pressure 128/80 mm of Hg. He had mild pallor, icterus, bilateral pitting pedal edema, and features of atrophic glossitis. On abdominal examination, nontender hepatomegaly (6 cm below right costal margin) and massive splenomegaly (8 cm below left costal margin) with shifting dullness was present. There was stony hard dullness, reduced air entry, and vocal resonance in b/l lung bases. Other organ systems were unremarkable.

On admission, routine laboratory investigations showed – pancytopenia with hyperferritinemia (3354 ng/ml, normal 20–250 ng/ml), deranged liver enzymes, and lipid profile [Table 1]. Urine routine/microscopy showed pus cells 5–6/hpf and urobilinogen. On day 3, urine and blood culture showed growth of methicillin-sensitive Staphylococcus after 2 days of incubation. During a repeat in-depth history, the patient admitted to having complaints of a burning micturition and an urge to void multiple times a day in the past which he refused on previous instances. He explained that he was hesitant to disclose this in front of others due to prevailing social stigma.

Table 1.

A table summarizing the lipid profile findings of the patient

Lipid profile	Patient's value	Reference values
T. Cholesterol	292 mg/dl	<200 mg/dl
S. Triglycerides	852 mg/dl	<150 mg/dl
High-density lipoprotein	25 mg/dl	40-60 mg/dl
Low-density lipoprotein	168 mg/dl	100-129 mg/dl

Patient was then treated with parenteral amoxicillin–clavulanate and gentamicin based on sensitivity pattern, but his condition deteriorated [Table 2]. Extensive workup was done to evaluate for cause of pancytopenia and transaminitis [Table 3]. Ultimately, all the negative investigations into the suspected causes prompted bone marrow biopsy which showed hypocellular marrow for age with erythroid hyperplasia, partial maturation arrest in myeloid series, increased iron stores (Grade 4), and mildly elevated plasma cells. Morphologically examination revealed large histiocytes containing multiple concave nuclei of myeloid series cells, suggestive of hemophagocytosis. Bone marrow culture grew methicillin-sensitive Staphylococcus aureus.

Table 2.

A table summarizing the laboratory parameters during the course of hospital stay

Hemogram	Day 1	Day 13	Day 18	Day 23	Reference values
Hb (g/dl)	12.1	8.5	8.9	9.2	13-18 g/dl
TLC (cells/cumm)	2200	2230	4800	4952	4000-11 000/cumm
DLC (N/L/M)	78/10/0	87.5/9/3.1	76/21/2	81.5/9.7/8.2	N40-80/L20-40/M2-10
Platelet count (platelets/ml)	62000	90000	110000	120000	15 0000-40 0000/cumm
Liver function tests	Day 1	Day 13	Day 18	Day 23	Reference Range
Bilirubin (T) mg/dl	14.71	11.32	9.9	8.7	0.1-1.2 mg/dl
Bilirubin (D) mg/dl	8.61	6.72	4.5	4	0.0-0.3 mg/dl
SGPT U/L	137	128	184	178	<50 U/L
SGOT U/L	186	217	311	219	<50 U/L
ALP IU/ml	707	532	861	639	30-120 U/L
GGT U/L	136	555	279	144	0-30 U/L
S. Protein g/dl	3.8	4.4	4.9	5.0	6.5-8.5 g/dl
S. Albumin g/dl	1.61	1.6	2.0	2.0	3.5-5.5 g/dl
S. Globulin g/dl	2.2	2.8	2.8	2.9
A: G Ratio	0.7	0.59	0.70	0.70	0.8-2

Table 3.

A table summarizing the investigations done and the disease entities excluded while evaluating this patient with pancytopenia and chronic liver disease

Diseases excluded	Presentation and investigations for exclusion
Visceral leishmaniasis	Negative LD bodies, rk-39 ELISA
Dengue	Negative NS1 IgG and IgM antibodies
Hepatitis B and C	Negative anti-HBC Ab and HbsAg
HIV	Negative anti-HIV Ab
Celiac disease	No GI complaints and negative anti-TTG Ab
Systemic lupus erythematosus and rheumatoid arthritis	No skin rash, joint complaints, and negative ANA
EBV	Negative anti-EBV VCA
HSV	Negative serology and no skin lesions
Malaria	Negative peripheral blood smears
Tuberculosis	Negative sputum, urine and blood AFB, negative tuberculin skin test, and no chest X-ray findings
CMV	Negative PP65
Malignancy	Negative blood smear, ascitic tap transudative with normal cytology, negative lymph node biopsy, CT abdomen revealed no signs of malignancy
Scrub typhus	Negative Weil-Felix and immunochromatographic tests, no rash
Leptospirosis	Negative IgM serology or macro agglutination test
Parvo virus B19	Negative IgM serology
Wilson disease	Normal serum ceruloplasmin, negative liver biopsy, and normal slit lamp examination
Hemochromatosis	Negative liver biopsy, no other endocrine, or skin symptoms
Autoimmune hepatitis	Negative antismooth muscle antibodies, negative liver biopsy
Glycogen storage disorders	Negative liver biopsy

A diagnosis of HLH secondary to staphylococcal UTI and bacteremia was made (6 out of 8 criteria fulfilled).[3] Patient's family refused the treatment with HLH 2004 protocol drugs owing to financial restraints. He was then started on oral methylprednisolone 2 mg/kg/day along with antibiotic coverage, packed cells, and platelets transfusion. Over the next 1 week blood counts showed improvement and general condition improved. He was discharged on day 24 on oral steroids. Patient attended the outpatient department after 1 week of discharge. His general condition was found to be better. In spite of all the efforts, the patient was lost to follow-up.

Discussion

HLH-94 defined five diagnostic criteria [fever, splenomegaly, bicytopenia, hypertriglyceridemia (fasting triglycerides >2 mmol/l) and/or hypofibrinogenemia (fibrinogen <1.5 g/l), and hemophagocytosis]. In HLH-2004 update, three additional criteria were introduced; low/absent NK-cell-activity, hyperferritinemia (>500 μg/l), and high soluble interleukin-2-receptor levels (>2400 U/ml). For diagnosis, five of these eight criteria must be fulfilled.[3] HLH-2004 guidelines recommend etoposide, dexamethasone, cyclosporine A as first-line chemoimmunotherapy agents. But in any case, if this protocol cannot be followed owing to any restraints (financial in ours), initiating corticosteroids alone can prove to be life-saving (as in our case). There has been another similar case report, where corticosteroids alone were given in a 20-year-old female but HLH was secondary to primary EBV infection and not Staphylococcus.[4]

Studies have concluded that the Widal test has low sensitivity, specificity, and positive predictive value, but has a good negative predictive value.[5] A positive result should always be confirmed with a blood culture. For practical purposes, a treatment decision must be made on the basis of the results obtained with single acute phase sample (interpreted in accordance with the locally prevalent cut-offs) as the disease management cannot be delayed awaiting convalescent phase sample for paired sera study.[6]

A Belgian study on 220 women by Dr Stefan Heytens showed that almost all the patients with symptoms of lower UTI but negative urine culture actually had an infection.[7] Particularly in men, low bacterial counts are significant clinically, because contamination is uncommon.[8] Hence, studies support the use of empiric antibiotics guided by symptoms (dysuria, frequency, and urgency being highly predictive) even if urine dipstick and culture is negative.[7]

Conclusion

Primary care physicians should keep a high index of suspicion for HLH as early diagnosis and treatment reduces morbidity and mortality. HLH should be routinely suspected in a patient with unexplained fever with pancytopenia and multiorgan failure.

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

Nil.

Conflicts of interest

There are no conflicts of interest.