A 1670-g male infant was born by emergent cesarean section at 32 and 5/7 weeks’ gestation caused by nonreassuring fetal heart rate tracings. The patient's mother was a 25-year-old gravida 1 woman with a history of Graves disease (antithyroid peroxidase [APO] antibody positive at 208 IU/mL) and hypothyroidism secondary to radioiodine ablation performed 2 years before the delivery. Her thyroid studies were normal throughout pregnancy, with the exception of a mildly elevated thyroid-stimulating hormone level of 12 μIU/mL obtained during the third trimester. The pregnancy was further complicated by type 1 diabetes mellitus and drug, tobacco, and alcohol use. A fetal echocardiogram revealed a constricted ductus arteriosus, right ventricular hypertrophy, right atrial enlargement, and tricuspid regurgitation.
After birth, the infant was intubated for poor respiratory effort and concern for underlying heart disease. He also developed early thrombocytopenia requiring platelet transfusion. Other early laboratory values showed hypoglycemia, evidence of disseminated intravascular coagulopathy, and elevated neonatal bilirubin (Table 1). On day of life (DOL) 2, the patient was jaundiced (conjugated bilirubin level of 11.2 mg/dL). Liver enzymes and alkaline phosphatase concentrations were elevated (Table 1). Doppler ultrasound excluded anatomic anomalies of the biliary tree and gallbladder. Serum tests for toxoplasmosis, rubella, cytomegalovirus, and herpes simplex virus were negative, as were bacterial cultures of urine and blood. Metabolic studies, including those for α1-antitrypsin deficiency, were normal. The patient received fresh frozen plasma and platelet concentrate and his coagulopathy resolved, as did respiratory distress.
On DOL 4, the patient became hypertensive and tachycardic (heart rate 235 bpm), prompting evaluation of his thyroid status. On DOL 5, he was diagnosed with thyrotoxicosis (Table 1). He was started on propylthiouracil (PTU) and propranolol on DOL 6, and heart rate and blood pressure normalized within 24 hours. His conjugated bilirubin concentration decreased following initiation of therapy on DOL 6. Given the lower incidence of hepatatoxicity with methimazole (MMI) compared with PTU in children (1), PTU was discontinued after 6 doses and therapy with MMI initiated on DOL 7; however, the patient's transaminase levels increased further while on MMI therapy, peaking at an aspartate aminotransferase concentration of 601 U/L and an alanine transaminase concentration of 190 U/L on DOL 17; thus, MMI was discontinued. Hyperthyroidism did not recur and the patient became euthyroid soon after (Table 1).
The infant was diagnosed postnatally with mild pulmonary hypertension (pulmonary artery pressure two-thirds of systemic pressure), believed to be secondary to ductus arteriosus constriction in utero, requiring no postnatal intervention. Follow-up echocardiograms revealed no significant pathology.
The patient was discharged on DOL 24. Outpatient follow-up showed improving cholestasis with a conjugated bilirubin concentration normalizing by 3 months of age and liver transaminase concentrations by 4 months of age. Thyroid-stimulating hormone and free T4 concentrations have remained normal.
The differential diagnosis of neonatal cholestasis is extensive; etiologies are often divided into obstructive, infectious, and metabolic causes (2). Hypothyroidism and hypopituitarism are 2 endocrinopathies associated with neonatal cholestasis. Hyperthyroidism is not typically considered a cause of neonatal conjugated hyperbilirubinemia, although to date 2 previous reports have detailed instances in which hyperthyroid infants born to mothers with Graves disease have developed cholestasis (3,4). In addition, hepatic dysfunction with cholestatic jaundice has also been reported in adults with symptomatic hyperthyroidism (5). We present a third case of neonatal cholestasis associated with hyperthyroidism and suggest that hyperthyroidism be considered a potential etiology of cholestasis and liver dysfunction in neonates.
Neonatal hyperthyroidism caused by maternal Graves disease is a transient process because of transplacental passage of maternal antibodies, which stimulate the fetal thyroid. A thorough maternal health history is imperative, because the development of fetal and neonatal thyrotoxicosis does not depend on previous treatment of the mother or the mother's thyroid status during pregnancy. Common neonatal symptoms of hyperthyroidism include tachycardia, hypertension, goiter, growth retardation, diarrhea, flushing, and exophthalmos (6). Thrombocytopenia, which our patient exhibited, is also rarely associated with neonatal hyperthyroidism by an unclear mechanism (6,7), but may have been related to the coagulopathy. Pulmonary hypertension, also seen in our case, is occasionally associated (8).
Similar to 1 other report (4), our patient's conjugated bilirubin concentration peaked on DOL 6. Although the mechanism of hyperthyroidism-associated cholestasis in these cases is unknown, improvement correlated with initiation of antithyroid therapy, as it did in a report of adults with hyperthyroidism (5).
None of the previous reports in neonates or adults listed potential causes for this observation and we can only speculate. Lorijn et al (9) studied the effects of experimental fetal hyperthyroidism in chronically catheterized fetal sheep and noted a marked acceleration of heart rate and oxygen consumption (VO2). In other experiments using other agents, such as catecholamines, to increase VO2 (10), acidosis and hypoxemia have resulted. Furthermore, in another experimental model of fetal hyperthyroidism (11), other significant adverse effects were found, including depression of fetal hepatic weight and increase in fetal hepatic lipid content. Lastly, a study in rats (12) has demonstrated that thyroxine, perhaps by increasing VO2, increases free radical generation, particularly in liver, thus inducing liver injury. It is notable that neonates have fewer defense mechanisms against the deleterious effects of reactive oxygen species (13). We suggest that excessive demands upon oxygen delivery and free radical generation may have induced the changes noted in affected patients. The mild pulmonary hypertension and fetal ductal constriction in our patient may have been coincidental.
In our case, antithyroid therapy was quickly discontinued earlier than would be typical because of concerns regarding what was thought to be drug-induced hepatitis, a rare complication of MMI therapy. Although we cannot definitively exclude PTU toxicity as the cause of the elevation in transaminases, because the patient received multiple doses of the drug, the temporal pattern implicates MMI. Ours is the first report to suggest possible MMI-related hepatotoxicity in infancy, and supports recent reports about rare adverse events associated with pediatric MMI (14). The present report also raises the possibility that the risk of MMI-associated liver injury is higher in patients with underlying cholestasis (caused by hyperthyroidism or alternate etiologies); additional caution in these patients is thus warranted.
We report on a neonate with cholestasis associated with hyperthyroidism caused by maternal Graves disease. Hyperthyroidism is not typically considered to be a cause of cholestatic liver disease in children; however, this report and 2 previous reports support this association, as do recent reports of similar findings in adults with hyperthyroidism. This report is the first to our knowledge to suggest potential MMI-associated hepatotoxicity in an infant, a rare but potentially serious reaction in patients with underlying liver disease.
1. Rivkees SA, Mattison DR. Ending propylthiouracil-induced liver failure in children. N Engl J Med
2. De Bruyne R, Van Biervliet S, Vande Velde S, et al. Clinical practice: neonatal cholestasis. Eur J Pediatr
3. Dryden C, Simpson JH, Hunter LE, et al. An unusual cause of neonatal coagulopathy and liver disease. J Perinatol
4. Beroukhim RS, Moon TD, Felner EI. Neonatal thyrotoxicosis and conjugated hyperbilirubinemia. J Matern Fetal Neonatal Med
5. Hull K, Horenstein R, Naglieri R, et al. Two cases of thyroid storm-associated cholestatic jaundice. Endocr Pract
6. Zimmerman D. Fetal and neonatal hyperthyroidism. Thyroid
7. Adrouny A, Sandler RM, Carmel R. Variable presentation of thrombocytopenia in Graves’ disease. Arch Intern Med
8. O’Donovan D, McMahon C, Costigan C, et al. Reversible pulmonary hypertension in neonatal Graves disease. Ir Med J
9. Lorijn RH, Nelson JC, Longo LD. Induced fetal hyperthyroidism: cardiac output and oxygen consumption. Am J Physiol
10. Gournay VA, Roman C, Rudolph AM. Effect of beta-adrenergic stimulation on oxygen metabolism in the fetal lamb. Pediatr Res
11. Rosato RR, Jahn GA, Gimenez MS. Amelioration of some metabolic effects produced by hyperthyroidism in late pregnant rats and their fetuses. Effects on lipids and proteins. Horm Metab Res
12. Chandra AK, Sinha S, Choudhury SR. Thyroxine induces stress and its possible prevention by catechin. Indian J Exp Biol
13. Perrone S, Negro S, Tataranno ML, et al. Oxidative stress and antioxidant strategies in newborns. J Matern Fetal Neonatal Med
2010; 23 (suppl 3):63–65.
14. Rivkees SA, Stephenson K, Dinauer C. Adverse events associated with methimazole therapy of Graves’ disease in children. Int J Pediatr Endocrinol
2010 [Epub Mar 7].