Oral Aversion, Growth Faltering, and Hepatomegaly in a 4-Month-Old Girl

A 4-month-old Afghan girl is admitted from her pediatrician’s office for growth faltering and refusal of oral intake. The infant was born at 41 weeks of gestation to a 29-year-old, gravida 2 para 2 mother after an uneventful pregnancy and delivery. She was appropriate for gestational age, with a birthweight of 3420 g (68th percentile). The infant’s newborn nursery course was uncomplicated, with normal newborn screening results.The infant initially breastfed with supplementation using standard infant formula provided by the Women, Infants, and Children (WIC) program. She demonstrated acceptable growth at the 2-week visit, but at her 1-month visit, the mother expressed concerns about decreased breastmilk production since returning to work. The infant’s weight had dropped to the 22nd percentile. Increased formula supplementation was recommended, with subsequent improvement in her growth percentiles at the 2-month visit. However, at the 4-month visit, the parents reported that the infant had been difficult to feed since switching to a partially hydrolyzed formula based on WIC formula availability 1 month prior, with arching and crying during supplemental feeding attempts, though she continued to breastfeed. On the day of the visit, she had taken only 5 ounces in the preceding 24 hours and had gained only 15 g in 2 months, with weight falling to less than the first percentile (z-score −2.58). During an attempted feeding by the pediatrician, the infant refused any formula intake, prompting urgent direct admission to the hospital for growth faltering and feeding refusal. The parents denied noting any fever, lethargy, vomiting, choking, cyanosis, diaphoresis, change in stools, or other signs of illness.Upon admission to the hospital, the infant is afebrile with normal vital signs. She is well-developed and alert but fussy and difficult to console. She has moist mucous membranes, no oropharyngeal lesions, and a normal palate. Lung and cardiac examinations are normal. Her abdomen is rounded, soft, and nontender. Bowel sounds are hypoactive. The liver edge is palpable in the lower right quadrant, approximately 8 cm below the costal margin. The spleen edge is not palpable. Her neurologic assessment demonstrates appropriate tone, strength, and reflexes. She is not jaundiced.Initial laboratory studies are notable for bicarbonate 13 mEq/L (13 mmol/L) (reference range [RR] 17–29 mEq/L [17–29 mmol/L]) and unexpectedly normal glucose, anion gap, and urinary ketone levels. Liver enzyme and cholesterol/ triglyceride levels are elevated and significantly elevated, respectively (see Table 1). Complete blood count, prothrombin time, activated partial thromboplastin time, α-fetoprotein (AFP), quantitative β human chorionic gonadotropin (β-HCG), lactate, uric acid, and urinalysis are unremarkable. Complete abdominal ultrasound confirms an enlarged liver with normal echotexture and no other abnormalities; Doppler ultrasound was not performed. Additional history findings and testing led to the diagnosis.The differential diagnosis for hepatomegaly can be considered according to mechanism1: 1) enlargement of cellular size due to abnormal storage of fats (fatty acid oxidation defects, lysosomal storage disorders, acid sphingomyelinase deficiency, excess vitamin A), glycogen (glycogen storage diseases or Beckwith-Wiedemann syndrome), or other substances (alpha-1 antitrypsin deficiency), 2) inflammation from infectious, autoimmune, or toxin-induced hepatitis, 3) infiltration by primary or metastatic tumor (such as hepatoblastoma, the most common liver tumor in infants), 4) enlargement of the vascular space due to hepatic venous outflow obstruction (Budd-Chiari syndrome), sinusoidal obstruction syndrome, or congestive heart failure, or 5) increase in the size of the biliary system, as seen in congenital hepatic fibrosis or Caroli disease.In this case, the infant’s healthy appearance, normal cardiac examination, reassuring AFP and β-HCG levels, and hepatic appearance on ultrasound made inflammation, infiltration by tumor, and enlargement of the vascular or biliary spaces less likely. Her elevated triglyceride and total cholesterol levels were suggestive of a glycogen storage disease, but the lack of hypoglycemia and normal uric acid and lactate levels were inconsistent.The patient had tolerated her feedings and demonstrated normal growth until a switch in formula was made. Review of the formula label and consultation with a metabolic expert confirmed that the new formula contained added sugar (sucrose; Table 2 lists carbohydrates commonly added to infant formulas and their primary sugars). Further history revealed that 2 other family members could not tolerate eating sweets without experiencing abdominal pain and vomiting, and that the parents were consanguineous, raising the suspicion of a metabolic disorder. The patient’s carbohydrate-deficient transferrin profile, sent at the recommendation of our metabolic/genetic specialist, was markedly abnormal with a high proportion of circulating transferrin devoid of attached carbohydrate moieties (“a-sialo transferrin”), compared with the normal isoform pattern, in which 97% of circulating transferrin is tetra-sialo or penta-sialo in nature.2 While common in congenital glycosylation defects, this pattern can also be seen in hereditary fructose intolerance.3 The prior tolerance of an infant formula containing lactose and glucose as the primary carbohydrate sources, and the onset of symptoms coinciding with the introduction of a protein hydrolysate formula containing sucrose, especially in the setting of parental consanguinity, made the diagnosis of hereditary fructose intolerance highly likely. Whole exome sequencing confirmed biallelic pathogenic mutations in the ALDOB gene (c.1013 C>T p.A338V).Hereditary fructose intolerance (HFI) is an autosomal recessive condition caused by a deficiency of aldolase B, the enzyme responsible for metabolizing dietary fructose from fructose-1-phosphate (F1P) to glyceraldehyde and dihydroxyacetone phosphate.3 The incidence is thought to be underreported at 1:20 000 births worldwide, with no specific ethnic predilection.4 In affected patients, the ingestion of fructose causes rapid accumulation of F1P, which impairs glucose production and depletes adenosine triphosphate, typically leading to increased urate metabolism, hypoglycemia, lactic acidosis, impaired protein production, and eventual renal and liver disease.3The classic presentation is an infant with nausea, vomiting, poor feeding, and lethargy following the introduction of foods containing sucrose or fructose. Presentation in early infancy may be seen with the consumption of infant formula to which sucrose or fructose has been added, substances “generally recognized as safe” by the U.S. Food and Drug Administration. Recent studies have highlighted the ubiquity of added sugars in US infant formulas, especially lower lactose-containing protein hydrolysate formulas (“gentle” or “sensitive” formulas),5 whereas in the European Union, a minimum lactose content is required, sucrose may only be added to formulas made from protein hydrolysates to a maximum of 20% of the carbohydrate total, and the addition of fructose is not permitted.6 Since our patient had significant formula refusal, which limited her sucrose intake, her primary signs were poor intake and weight faltering, rather than vomiting and lethargy. We hypothesize that she was able to maintain euglycemia via frequent breastfeeding, though this was not enough for her to sustain growth.Patients with residual enzyme activity often have later and more subtle presentations, such as growth faltering, hepatomegaly, intermittent transaminase elevations, chronic liver disease with steatohepatitis, proximal renal tubular acidosis, chronic renal insufficiency, or relapsing acute neuronal neuropathy.3 Patients with sweet avoidance and growth faltering due to HFI may be misdiagnosed with disordered eating unless a careful dietary history is undertaken in these circumstances.3The diagnosis of HFI should be suspected in patients with a clear association between exposure to dietary fructose and the onset of gastrointestinal symptoms, as well as those with an aversion to sweets or idiopathic steatohepatitis.3 Laboratory findings suggestive of HFI include positive urinary reducing substances, low serum phosphorus, and elevation of the fractional excretion of urate.3 Triglyceride levels are abnormal due to assay detection of high levels of glycerol, the triglyceride “backbone,” which is produced in excess due to secondary effects of F1P on the gluconeogenesis pathway.3,7 Carbohydrate-deficient transferrin (CDT) levels are a useful screen because high levels of F1P, as seen in HFI, inhibit phosphomannose isomerase, which is essential for the normal glycosylation of transferrin.3,8 Genetic testing has replaced liver biopsy to confirm the diagnosis of HFI.3 Most patients have biallelic mutations in the ALDOB gene on chromosome 9q22.3, as our patient demonstrated.3Treatment for HFI is the complete dietary avoidance of fructose, sucrose, and sorbitol, referred to as “FSS” (sorbitol is metabolized to fructose by sorbitol dehydrogenase).3 With strict dietary adherence, a good prognosis and normal lifespan can be achieved.3 Families of children with HFI should have the opportunity to work regularly with an experienced metabolic dietician to ensure sufficient caloric intake and prevention of vitamin C and B complex deficiencies, which are the most common complications of the FSS-free diet.8 CDT levels can be used to monitor dietary adherence.8 There is no safe exposure level of FSS for patients with HFI, and patients with persistently elevated CDT levels in this context are at risk for chronic renal insufficiency, hepatic fibrosis, metabolic dysfunction-associated steatotic liver disease (formerly called nonalcoholic fatty liver disease), and reduced height.3,4,8Our patient was initially managed with a sucrose/fructose-free formula via the nasogastric route due to her oral aversion, with almost immediate improvement in her condition. She quickly transitioned back to full oral feeds without intolerance or further weight loss.Before discharge, the patient’s family was provided with a WIC prescription for a sucrose/fructose-free formula and cautioned to seek input from a metabolic dietician before any substitution. The family was counseled regarding the patient’s need for strict FSS avoidance with the assistance of a Dari language interpreter.At a follow-up clinic visit just 3 weeks later (aged 4.5 months), all signs of oral aversion had resolved, and the infant was demonstrating catch-up growth of 38 g/d. By her 6-month visit, her weight was tracked at the 17th percentile. At her 10-month visit, her weight was at the 38th percentile for age and sex, and the family was provided additional information and resources to support a safe transition to table foods and a toddler diet.Clinicians should be aware that some infant formulas contain sucrose or fructose as an added sugar in place of lactose, which can lead to early presentation of HFI. Early diagnosis and referral can improve outcomes.Clinicians should suspect HFI in patients who have gastrointestinal symptoms with intake of sweets, have an aversion to sweets, or have unexplained steatotic liver disease.CDT levels serve as a useful screen for HFI while awaiting genetic confirmation of the diagnosis.