Abstract

Case Report
An African American male was born at 39 weeks via C-section to a 29-year-old gravida 3, para 3 mother with pregnancy complicated by gestational diabetes A1, hypertension, and asthma. Maternal serologies were negative except for a GBS positive status. APGAR scores were 9 and 9 at 1 and 5 minutes, respectively. Birth weight was 3.65 kg. Physical examination was unremarkable.
Hypoglycemia
The patient was found to have a glucose of 20 to 26 mg/dL in the first hour of life. Despite adequate oral feeding, hypoglycemia continued into the first few weeks of life and required titration of fluids to a glucose infusion rate of 9.1 mg/kg/min. Insulin level was checked and was found to be elevated with a peak level of 37.2 µIU/mL (normal < 28.5 µIU/mL) at 4 weeks of age, significantly increased from levels of 5.7 to 10.2 µIU/mL at 2 and 3 weeks of life, respectively. The elevated insulin corresponded with glucose < 50 mg/dL. Adrenocorticotropic hormone, cortisol, growth hormone, luteinizing hormone, follicle stimulating hormone, and thyroid studies were normal, ruling out possible panhypopituitarism. Studies for infectious etiologies including blood culture, cytomegalovirus, and toxoplasmosis were negative.
Cholestasis and Hyperammonemia
Direct hyperbilirubinemia was documented starting on the third day of life at 1.0 mg/dL (normal < 1.0 mg/dL) and gradually increased to a peak of 3.0 mg/dL at 3 to 4 weeks. Aspartate transaminase and alanine aminotransferase also trended up from initial measurements of 78 and 14 U/L on day of life (DOL) 1 to levels of 205 and 103 U/L, respectively, on DOL 19. Given the presence of an inappropriately elevated insulin level in the setting of hypoglycemia and worsening cholestasis, an ammonia level was checked and was found elevated at 144 µg/dL. The level was rechecked and was confirmed, peaking at 178 µg/dL at 3 weeks of age. The infant had no excessive irritability, seizures, or neurologic abnormalities on examination, suggesting the high ammonia was not causing signs of neurotoxicity. Albumin and international normalized ratio were normal, ruling out acute liver failure. Plasma amino acid profile was normal ruling out urea cycle defects.
Due to persistent elevation in direct bilirubin, a liver ultrasound was performed on DOL 9 and was normal with visualization of the gallbladder. A repeat ultrasound at 3 weeks of age, in the setting of acholic stools, was normal as well, but the gallbladder was not well visualized. Phenobarbital was started at 2.5 mg/kg twice daily and a hepatobiliary iminodiacetic acid scan was performed on DOL 25. It showed patent extrahepatic biliary flow, making obstructive cholestasis unlikely.
A liver biopsy was performed on DOL 27. It revealed mild lymphocytic infiltrates in the portal tracts, hepatocellular ballooning, and hepatocellular and canalicular cholestasis. There was bile ductular reaction on a CK7 immunostain. The biopsy was notably negative for iron deposition, viral inclusions, or storage products (Figure 1). This pattern was consistent with neonatal hepatitis.

Liver biopsy with mild inflammatory cell infiltrates and ductular reaction in the portal tract (arrow), hepatocellular ballooning (arrow head), and hepatocellular and canalicular cholestasis.
Familial hyperinsulinemic hypoglycemia was considered especially as hypoglycemia worsened with both formula feeds (amino acid containing) and fasting. A comprehensive congenital hyperinsulinism panel sent to the University of Chicago Genetic Services Lab did not identify a causative gene among the 18 on the panel.
Final Diagnosis
Hyperinsulinism hyperammonemia (HIHA) syndrome with neonatal hepatitis.
Hospital Course
Given the persistent hypoglycemia for 4 weeks at serum glucose <70 mg/dL despite receiving enteral formula feeding, diazoxide was started at 1.74 mg/kg, 3 times a day. At 8 weeks of life, the infant was able to maintain relatively normal glucose levels with regularly timed feedings. Transaminitis, hyperammonemia, and conjugated hyperbilirubinemia started to downtrend. Diazoxide was weaned and discontinued at 9 months of age. At 10 months, the infant was able to tolerate and pass a fasting test of cure without significant decline in glucose levels. Follow-up at 3 years of age revealed euglycemia and no neurologic deficits, with no restriction of protein in the diet.
Discussion
We present a case of neonatal hypoglycemia and hyperammonemia ultimately diagnosed with HIHA syndrome.
HIHA syndrome is the second most common form of congenital hyperinsulinemia; its prevalence is estimated at 1 in 200 000. Children with HIHA usually have normal birth weights. Presentation is typically within the first 2 years of life, but occasionally, it may not be recognized until adulthood. 1
HIHA is considered a congenital inborn error of metabolism resulting from genetic anomalies that cause a gain of function of the enzyme glutamate dehydrogenase (GDH). GDH is expressed in liver, kidney, brain, and pancreatic β-cells. Because of the genetic mutation, GDH loses an inhibitory response from guanosine triphosphate and continues to convert glutamate to metabolites that increase insulin secretion by pancreatic β-cells. 1 The uninhibited GDH also leads to impaired ammonia metabolism in the liver and kidneys (Figure 2).

Glutamate Dehydrogenase (GDH) functions in pancreatic, renal, and hepatic cells. GDH induces pancreatic β-islet cell insulin secretion. The increased ATP/ADP ratio, resulting from glutamate metabolite alpha keto-glutarate (α-KG), triggers the inhibition of ATP-dependent Potassium (K+) channels, leading to further depolarization of the cell membrane, Calcium (Ca++) influx, and insulin secretion.
Clinically, the major manifestations of GDH-activating mutations are fasting hypoglycemia and hypoglycemia induced by a high-protein meal. The development of hypoglycemia after an oral protein load is often much more rapid and dramatic than that after fasting. 1 These mutations also lead to hyperammonemia, which is usually asymptomatic and mild at 3 to 5 times the normal range.
Diagnosis of HIHA starts with exclusion of other endocrinopathies such as hypopituitarism and hypothyroidism. Molecular gene testing evaluates for common mutations in the GDH gene or mutations associated with other forms of hyperinsulinism, but may not be detected.
Treatment of HIHA is with diazoxide, a potassium channel activator that inhibits insulin secretion from the pancreas and relieves hypoglycemia. Diazoxide has been effective in patients with known and unknown gene mutations associated with hyperinsulinism.2-4 Treatment is titrated to achieve gylcemic control and avoid hypoglycemia, and frequently intravenous dextrose is required until successful titration is achieved.
In this case, we believe HIHA caused cholestasis, a hepatobiliary manifestation encountered in endocrinopathies. Hormones including androgens, estrogens, glucocorticoids, and thyroid hormone play a role in regulating bile secretion and bile salt–independent biliary flow. 5 Hypophysectomy (resection of the pituitary gland) decreases bile acid synthesis, bile flow, and biliary excretion of bile acids and bilirubin. 5 The specific mechanisms of action underlying how specific hormones cause cholestasis have been postulated, but are not clear. 5 Nonetheless, the liver dysfunction can be reversible once the hormonal dysregulation is corrected, 6 as shown in congenital pituitary deficiency, 7 and as demonstrated in this case.
Hepatic histologic changes in endocrinopathies have been described. In hypopituitarism, the features are similar to idiopathic neonatal hepatitis, though there is unique bile duct hypoplasia. 8 In this report, we demonstrate that the histologic changes in HIHA are similar to neonatal hepatitis in hypopituitarism, but without bile duct hypoplasia. To our knowledge, the histologic features of cholestasis-related HIHA were not previously reported.
Conclusion
The differential diagnosis of neonatal hypoglycemia is wide and is usually focused on endocrinopathies. Exclusion of pituitary causes, along with genetic testing for causes of congenital hyperinsulinism, can help establish the diagnosis of HIHA syndrome. Treatment with diazoxide leads to the safe resolution of both the hyperinsulinism as well as the cholestatic liver disease. In this case, we also report, for the first time, the histologic features of cholestasis in HIHA syndrome.
Footnotes
Author Contributions
All of the authors contributed to the conception and design, drafted the manuscript, critically revised the manuscript, and agree to be accountable for all aspects of work ensuring integrity and accuracy.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
