A 2-year-old asymptomatic girl born to consanguineous (first cousins) parents was referred to our pediatric hepatology practice for incidentally found hepatomegaly. Her birth, development, and medical and family histories were unremarkable except for an adenoidectomy for obstructive sleep apnea. The only positive finding on physical examination was a large, soft, and nontender liver palpable down to her pelvis. Her body mass index was 16.4 (48th percentile), and she did not have splenomegaly, ascites, jaundice, pallor, or facial dysmorphism. The abdominal ultrasound revealed hepatomegaly (right hepatic lobe measured 14.2 cm at right midclavicular line) with increased parenchymal echogenicity and a normal spleen. Her liver chemistries, serum electrolytes, blood glucose, and lactate were normal but with mild microcytic anemia (Hb 8.1 g/dL). Labs were also drawn to investigate for glycogen storage disorders (GSD), hyperlipidemia, autoimmune hepatitis, α1-antitrypsin deficiency, and Wilson disease (Table 1). All of the values were normal except for low total cholesterol and low-density lipoprotein. Hepatitis A, B, and C were also ruled out. Urine for reducing substances, although requested, could not be obtained.
Liver biopsy revealed diffuse, marked mixed macrovesicular, and small-droplet steatosis, with the macrovesicular pattern predominating in periportal (zone 1) location (Fig. 1A). Ballooning change and Mallory-Denk hyaline were not observed. Rare small clusters of chronic inflammatory cells were present in the lobules, but not in portal areas. Trichrome stain showed focal fibrous expansion of portal tracts with septum formation (Fig. 1B). On electron microscopy (EM), there were abundant intrahepatocytic neutral lipid vacuoles that frequently compressed the remaining organelles to the periphery of the cells. Glycogen was present in usual amounts and dispersed evenly in the cytosol (that was not occupied by fat). The mitochondria and endoplasmic reticulum appeared normal. There was no evidence of abnormal lysosomal storage material. Based on liver histology, further labs were done to investigate for lipid storage diseases, fatty acid oxidation disorders, and congenital disorders of glycosylation (CDG), tyrosinemia, organic acid disorders, and abetalipoproteinemia (online-only supplemental Table 2, http://links.lww.com/MPG/A155).
DIFFERENTIAL DIAGNOSIS AND DISCUSSION
An inborn error of metabolism, particularly GSD, was high in the differential but there was no associated hypoglycemia, lactic acidosis, no facial features, and normal serum creatine phosphokinase and uric acid levels. The liver biopsy (histology and EM) showed no excessive accumulation of intracellular glycogen, and GSD was therefore ruled out. The lack of biliary changes both on histology and imaging made congenital hepatic fibrosis unlikely. The liver biopsy showed changes that could be consistent with nonalcoholic steatohepatitis (NASH), specifically with type 2 or pediatric NASH (1); however, the patient was only 2 years old, with normal body mass index, and had no associated metabolic risk factors such as insulin resistance (normal fasting insulin levels and HbA1C), dyslipidemia, or hypertension, making NASH unlikely.
Rare lipid storage diseases, including lysosomal acid lipase deficiency and the neutral lipid storages diseases, were considered on the basis of the EM findings; however, the patient's symptomatology was not consistent. She did not have elevated triglycerides, and there were no neuromuscular involvement or ichthyosis. Absence of hypoglycemia and normal acylcarnitine and urine organic acid profiles eliminated fatty acid oxidation disorders. Tyrosinemia was ruled out by a normal plasma amino acid profile and urine succinylacetone levels. The diagnosis of abetalipoproteinemia was considered briefly because of a low total cholesterol and low-density lipoprotein; however, serum apolipoprotein B levels were normal. Carbohydrate-deficient transferrin testing (CDT), which screens for CDG, was performed and found to be abnormal (mono-oligo/di-oligo ratio 0.125 (normal 0.0–0.1). Given the nonspecific nature of CDT levels, and given that phosphomannose isomerase deficiency was easily treatable, the 2 most common causes of CDG were ruled out by normal peripheral leukocyte phosphomannomutase and phosphomannose isomerase enzyme activities (supplemental Table 2, http://links.lww.com/MPG/A155). It was noted at this juncture that CDT testing can also be abnormal in cirrhosis, galactosemia, and hereditary fructose intolerance (HFI).
Owing to a family history of consanguinity, a sample for whole-genome chromosome single nucleotide polymorphism microarray was ordered in the hopes to identify candidate loci causing steatosis in our patient. Based on the microarray result, a gene search was done using the coordinates of regions of homozygosity (http://www.ccs.miami.edu/cgi-bin/ROH/ROH_analysis_tool.cgi) (2). We found 3 genes listed in Online Mendelian Inheritance of Man associated with autosomal recessive conditions with steatosis: carnitine palmitoyl transferase (CPT) II gene, tRNA 5-methylaminomethyl-2-thiouridylate methyltransferase (TRMU) gene, and the aldolase B (ALDOB) gene. CPT II disorder was already ruled out by a normal acylcarnitine profile. The TRMU gene, which is implicated in transient neonatal liver failure, did not fit clinically.
Mutations in the ALDOB gene cause HFI, an autosomal recessive condition, which can present with hepatomegaly and steatosis, thus matching the patient's clinical profile. At this point, the mother revealed, on elaborate dietary questioning, the patient's history of some aversion to sweets. There was, however, a curious lack of supportive biochemical tests with normal bilirubin, international normalized ratio, glucose, lactate, and uric acid. HFI, however, can give CDT abnormalities (3), which she had. Hence, based on these observations, testing for ALDOB gene sequencing was sent, and a homozygous mutation that has been previously reported to cause HFI (c.448G>C, p. A150P) was found (4). The hepatic histological changes in HFI, although nonspecific, appear to correlate with the age of presentation. In early infancy, giant cell hepatitis, intracanalicular cholestasis, bile duct proliferation, and a biliary pattern of cirrhosis have been reported (5). Beyond early infancy, diffuse pronounced macrovesicular steatosis and fibrosis (portal ± lobular) may predominate, as were seen in this patient (5–7). After being informed of the diagnosis of HFI, the mother further reported that her husband also avoided sweets, as did their shared grandmother. A number of cousins on this grandmother's side had a history of needing hospitalization after eating sweets. A son of a distant cousin was also recently diagnosed with HFI in Europe. The relatively occult presentation here is likely the result of low-dose but chronic exposure to fructose in the diet. The present case highlights the fact that HFI can present as isolated massive hepatomegaly with steatosis in a relatively asymptomatic patient with normal liver tests and normal glucose.
1. Schwimmer JB, Behling C, Newbury R, et al. Histopathology of pediatric nonalcoholic fatty liver disease. Hepatology
2. De Leeuw N, Dijkhuizen T, Hehir-Kwa JY, et al. Diagnostic interpretation of array data using public databases and internet sources. Hum Mutat
3. Pronicka E, Adamowicz M, Kowalik A, et al. Elevated carbohydrate-deficient transferrin (CDT) and its normalization on dietary treatment as a useful biochemical test for hereditary fructose intolerance and galactosemia. Pediatr Res
4. Cross NC, Tolan DR, Cox TM. Catalytic deficiency of human aldolase B in hereditary fructose intolerance caused by a common missense mutation. Cell
5. Black HA, Simpson K. Fructose intolerance. Br Med J
6. Levin B, Snodgrass GJAI, Oberholzer VG, et al. Fructosaemia. Observation on seven cases. Am J Med
7. Odièvre M, Gentil C, Gautier M, et al. Hereditary fructose intolerance in childhood; diagnosis, management and course in 55 patients. Am J Dis Child