|DR. WATSON’S CLINICAL MYSTERY
|Year : 2021 | Volume
| Issue : 4 | Page : 284-287
Clinical approach to a febrile young infant with hepatosplenomegaly using the principle of Occam's razor
Sharmila Banerjee Mukherjee1, Jasleen Kaur1, Anu Maheshwari1, Sunita Sharma2
1 Department of Pediatrics, Lady Hardinge Medical College and Associated Hospitals, New Delhi, India
2 Department of Pathology, Lady Hardinge Medical College and Associated Hospitals, New Delhi, India
|Date of Submission||27-Oct-2021|
|Date of Decision||28-Oct-2021|
|Date of Acceptance||30-Oct-2021|
|Date of Web Publication||29-Nov-2021|
Dr. Sharmila Banerjee Mukherjee
Department of Pediatrics, Kalawati Saran Children's Hospital, Bangla Sahib Marg, New Delhi - 110 001
Source of Support: None, Conflict of Interest: None
|How to cite this article:|
Mukherjee SB, Kaur J, Maheshwari A, Sharma S. Clinical approach to a febrile young infant with hepatosplenomegaly using the principle of Occam's razor. Indian Pediatr Case Rep 2021;1:284-7
|How to cite this URL:|
Mukherjee SB, Kaur J, Maheshwari A, Sharma S. Clinical approach to a febrile young infant with hepatosplenomegaly using the principle of Occam's razor. Indian Pediatr Case Rep [serial online] 2021 [cited 2022 Jan 20];1:284-7. Available from: http://www.ipcares.org/text.asp?2021/1/4/284/331382
In the prepandemic era, a 2-month-old girl presented to the hospital with refusal to feed for 1 day. Her parents gave a history of fever for a few hours daily for a week, that had never been documented. She also had gradually progressive abdominal distension for the same duration. There was no history of vomiting, loose stools, jaundice, bleeding from any site, rashes, seizures, lethargy or excessive crying, cough, or difficulty in breathing. The urinary and stool patterns were unremarkable. Before her illness, the baby was predominantly being breastfed but also given diluted formula milk by bottle. Her feeding had become poor for a week and almost nil for a day. She had received some medication from a practitioner for 3 days, the nature of which was unknown.
The baby was hospital born to an unrelated couple at 36 weeks gestation by vaginal delivery with a birth weight of 2 kg and normal Apgar scores. On the second day of life, she developed jaundice for which she received phototherapy for a day and was discharged. The antenatal period had been uneventful with quickening perceived at 4 months. The 28-year-old mother had history of an unexplained stillborn at term and a 2-month spontaneous abortion. There was no other contributory family history. The baby had been vaccinated at birth and 6 weeks.
The baby was normothermic with stable vital parameters. The weight was less than 3rd percentile, but length and head circumference were normal. Pallor was present. The anterior fontanelle was at level and no icterus, rashes, petechiae, ecchymoses, edema, or unusual facies/overt congenital anomalies were noted. The upper abdomen appeared full without any visible distended veins. Firm hepatosplenomegaly was present with liver 3 cm below costal margin (span 8 cm) and spleen 5 cm below costal margin. Ascites was not appreciated and bowel sounds were normal. Although cry and spontaneous movements and neonatal reflexes were preserved, they appeared depressed. Cardiovascular and respiratory systems were normal.
Dr. Watson: There is no clinical indicator of a surgical cause. The clinical phenotype is an acute febrile illness and hepatosplenomegaly (HSM), so the etiology is probably infectious. It may be sepsis, but I must exclude malaria, urinary tract infection, and meningitis, the usual causes of fever in a young infant. Pending the baseline test reports, I will start intravenous fluids and empirical parenteral broad-spectrum antibiotics that will cover community-acquired microbes. The organomegaly is a bit too firm than usual for an infant though.
The hemogram displayed low hemoglobin (5.5g/dL), decreased absolute neutrophil count (704/mm3), thrombocytopenia (26 × 109/L), and a reticulocyte count of 2.5%. The peripheral smear was normal, apart from the decreased leukocyte and platelet populations, and there were no toxic granules, indicators of hemolysis, or abnormal cells. The C-reactive protein was elevated (32 mg/L). The only biochemical abnormalities found were in the liver function tests; SGOT 125U/L, SGPT 33U/L, PT 21.8 s, APTT 52.1s, and INR1.8 s. D-dimer levels were normal. Investigations for malaria were negative. Urine microscopy was normal, as were cerebrospinal fluid (CSF) cytology, gram staining, and biochemistry (the lumbar puncture was done after platelet transfusion).
Dr. Watson: Hmm. She has HSM, transaminitis, and pancytopenia. Can it be hypersplenism resulting from portal hypertension with secondary sepsis? Or an infantile malignancy like a neuroblastoma or an embryonal tumor? An ultrasonogram is non-invasive and I can get it done while I await the blood, urine, and CSF culture reports.
The abdominal ultrasonogram confirmed hepatosplenomegaly with normal echotexture. The portal vein diameter was normal. There were no calcifications or lesions in any organ or mesenteric lymphadenopathy. All three cultures were sterile after 48 h.
Dr. Watson: Sherlock always says I should use the principle of Occam's razor and look for a single simple explanation rather than multiple, complex assumptions. I need to think of an infantile-onset disorder that would explain the triad of fever, HSM, and pancytopenia. The first differential may still be infectious- maybe a congenital TORCH or Human Immunodeficiency Virus infection; the second an infiltrative disorder of the bone marrow by a hematological malignancy (acute lymphoblastic leukemia?), Hemophagocytic lymphohistiocytosis (HLH), or a storage disorder like Gaucher disease type 2.
The TORCH immunoglobulin M panel for toxoplasmosis, rubella, cytomegalovirus, and herpes viruses was nonreactive, as was the maternal HIV status. Bone marrow aspiration exhibited hypercellular marrow with erythroid hyperplasia, normal maturation of myeloid series, and megakaryocytes and absence of any abnormal cells. Ferritin levels were high (2000 ng/ml), triglycerides levels elevated (792 mg/dl), and fibrinogen low (93 mg/dl).
Dr. Watson: Five of the eight diagnostic criteria of HLH are satisfied in [Table 1]. I must start HLH directed treatment right away. I will order a Whole Exome Sequencing (WES) as well.
|Table 1: Application of the 2004 hemophagocytic lymphohistiocytosis clinical trial diagnostic criteria in the index patient|
Click here to view
Two weeks had elapsed by the time the diagnosis was made. The baby's condition had deteriorated considerably, by then. The parents were counseled about the grim prognosis and the necessity and risks of the HLH-2004 treatment protocol. After sending appropriate cultures, the baby was empirically started on the next line of intravenous antibiotics (meropenem and vancomycin) and an antifungal (caspofungin), to minimize the possibility of a nosocomial infections flaring up with immunosuppression. The treatment was initiated with dexamethasone and oral cyclosporine. Supportive care (cryoprecipitate and fresh frozen plasma) was provided as required. However, the baby continued to deteriorate (hence, etoposide could not be given) and she succumbed within 18 days of admission. The parents agreed to a postmortem liver biopsy that demonstrated evidence of hemophagocytosis. The WES report confirmed a PRF1(-) gene mutation with homozygous variation in exon 2 and c. 386G > C (p. Trap129Ser).
Dr. Watson: This mutation is the most common cause of Type 2 Familial Hemophagocytic lymphohistiocytosis (FHL). Even though we lost this baby, we can now offer prenatal testing for the next pregnancy. I must ensure that the family receives proper genetic counseling.
| Discussion|| |
Every pediatrician knows how difficult ascertaining the etiology of a patient with HSM can be, given the heterogeneous causes, and the multiple clinical phenotypes involving permutations and combinations of pallor, jaundice, lymphadenopathy, etc. Then, there are other dimensions that need to be factored in; age of onset, duration, presence or absence of fever, etiological yield, invasiveness, and cost of the multitude of tests, etc. One must employ an individualized, logical, and sequential clinical approach. This is a situation that requires the principle of Occam's Razor. Initially developed for mathematics, its use has extended to philosophy, science, and the field of clinical medicine. When an event has more than one explanation, consider the one that requires the simplest or fewest assumptions. This is what Dr. Watson used to arrive at a clinical diagnosis of HLH, and then establish etiology as FHL. However, in the less than Utopian Indian settings, availability of tests and limited financial resources are major challenges to this process.
HLH is an aggressive and life-threatening autosomal recessive syndrome resulting from mutations in the PRF1, STX11, UNC13D, or STXBP2 genes. It affects one in 100,000 children, with the highest incidence in those less than 3 months. The underlying pathophysiology is a hyperimmune state which renders cytotoxic T-lymphocytes and natural killer cells incapable of lysing macrophages. These in turn, produce increased cytokines such as tumor necrosis factor-α, interferon-ϒ, and interleukins that generate the fever and cause extensive multiorgan destruction.
Clinical manifestations are usually nonspecific, with fever and HSM. Investigations can provide a clue and a high index of suspicion should be kept when there is pancytopenia, which presents from the onset in the majority. This occurs due to two mechanisms; hematopoietic suppression by elevated cytokines or hemophagocytosis by overactivated macrophages. However, this may not be evident in the early phases of disease and may necessitate repeated bone marrow/tissue examination. Biomarkers include hypertriglyceridemia (cytokines suppress lipoprotein lipase), increased ferritin (secreted by macrophages), and reduced fibrinogen (increased release of plasminogen activator leading to hyperfibrinolysis).
According to the HLH-2004 trial diagnostic criterion, molecular diagnosis for children is confirmatory by homozygosity or compound heterozygosity for any of the HLH-associated mutations. In adults, the criteria include heterozygosity of one of the genes with consistent clinical findings or five of the eight/nine findings given in [Table 1]. It is important to remember; however, that these diagnostic criteria were established for the stringent settings of clinical trials. In reality, treatment should not be delayed if they are not satisfied given the high mortality of HLH, and the fact that patients may be too ill to undergo biopsies or tests may not be available. In these situations, modified criteria may be used for diagnosis: three of four clinical findings (fever, splenomegaly, pancytopenia, or hepatitis) and the presence of at least one abnormal immune marker (hemophagocytosis, increased ferritin, hypofibrinogenemia, or absent/very decreased NK cell function).
The 2004 trials advocate early initiation of chemotherapy and immunosuppression for 8 weeks with dexamethasone, cyclosporine, and etoposide.,,,, These may induce remission, as indicated by resolution of fever, decrease in spleen size, reduced ferritin levels, increased platelet counts, and normalization of fibrinogen. Hematopoietic stem cell transplantation (HSCT) is indicated in familial, persistent, or recurrent disease., Prognosis is better when treatment is initiated early and in older children.,, There are two studies of Indian children with HLH. The first included 43 children (mean age 46 months) who were treated with corticosteroids (67%), intravenous immunoglobulin (64%), cyclosporine (33%), and etoposide (15%). The overall 2-year survival rate was 76%. The second study included eight infants (mean age 7 months) who received management as per the 2004 HLH protocol. Of the five who received only induction, four died and one abandoned therapy. One of the two who underwent HSCT survived, while the last one was on the waiting list at the time of publication.
To conclude, a high index of suspicion of HLH should be kept in infants with fever, HSM, and pancytopenia and timely investigations sent. Molecular diagnosis is confirmatory, but as far as possible treatment should not be deferred for genetic or immunological testing, as the mortality rate is very high. Nonetheless, every attempt should be made to identify the underlying mutation so that prenatal testing can be offered.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Morimoto A, Nakazawa Y, Ishii E. Hemophagocytic lymphohistiocytosis: Pathogenesis, diagnosis, and management. Pediatr Int 2016;58:817-25.
Zhang Z, Wang J, Ji B, et al.
Clinical presentation of hemophagocytic lymphohistiocytosis in adults is less typical than in children. Clinics (Sao Paulo) 2016;71:205-9.
Risma KA, Marsh RA. Hemophagocytic lymphohistiocytosis: Clinical presentations and diagnosis. J Allergy Clin Immunol Pract 2019;7:824-32.
Bergsten E, Horne A, Aricó M, et al.
Confirmed efficacy of etoposide and dexamethasone in HLH treatment: Long-term results of the cooperative HLH-2004 study. Blood 2017;130:2728-38.
Canna SW, Marsh RA. Pediatric hemophagocytic lymphohistiocytosis. Blood 2020;135:1332-43.
Jia C, Wang B, Zhu G, et al.
Haploidentical hematopoietic stem cell transplantation using reduced-intensity conditioning for pediatric patients with familial hemophagocytic lymphohistiocytosis. Pediatr Investig 2018;2:216-21.
Imashuku S, Hlbi S, Todo S. Hemophagocytic lymphohistiocytosis in infancy and childhood. J Pediatr 1997;130:352-7.
Ramachandran B, Balasubramanian S, Abhishek N, et al.
Profile of hemophagocytic lymphohistiocytosis in children in a tertiary care hospital in India. Indian Pediatr 2011;48:31-5.
Ramzan M, Yadav SP, Kharya G, et al.
Hemophagocytotic lymphohistiocytosis in infants: A single centre experience from India. Pediatr Hematol Oncol 2014;31:285-92.