Neonatal hemochromatosis (NH) is a rare disease defined by the coexistence of liver disease of antenatal onset with excess iron at extrahepatic sites. NH is not considered a single disorder but rather a syndrome of unknown etiology (1). We report a patient with histologically confirmed NH who survived the initial phase of liver failure, even though the level of lentil lectin A-reactive alpha-fetoprotein-L3 (AFP-L3) was high.
The patient is the second child of non-consanguineous Japanese parents. Routine antenatal screening tests were normal. His 5-year-old sister and his parents are healthy. Many relatives of his parents had histories of various kinds of cancer. The boy was delivered by urgent cesarean section under suspicion of acute placental hemorrhage at 38 weeks gestation. Apgar scores were 9/10/10 at 1, 5 and 10 minutes, respectively. Birth weight was 2894 g, height was 46 cm and head circumference was 33 cm.
At age 7 days, coagulopathy and hepatomegaly developed. The patient was admitted to the intensive care nursery, the trachea was intubated and ventilatory support given. The subsequent course of his disease was characterized by worsening hepatic function, thrombocytopenia and disseminated intravascular coagulation. Serum concentration of ferritin was very high (145,900 ng/mL; upper limit of normal, 250 ng/mL) before transfusion. Fresh-frozen plasma, vitamin K and packed red blood cells (120 mL) were administered without improvement of hepatic function. At age 28 days, T1-weighted magnetic resonance imaging revealed that the liver produced a signal similar to that of the spleen. T2-weighted magnetic resonance imaging at about the same level showed an abnormally low hepatic signal consistent with siderosis (Fig. 1).
At age 54 days, he was referred to our hospital with hepatic failure of unknown etiology. The physical examination revealed malnutrition (body weight: 1st centile; height: 3rd centile), jaundice, anemia and hepatomegaly.
The laboratory evaluations showed elevated alanine aminotransferase of 232 IU/L (normal upper limit, 35 IU/L), aspartate aminotransferase 464 IU/L (normal upper limit, 30 IU/L), prothrombin time activity of 41% (normal >80%), hepaplastin test of 30% (normal >70%), total/direct bilirubin of 8.6/6.2 mg/dL, and albumin of 2.9 g/dL. Serologic titers for hepatotropic viruses including hepatitis A, hepatitis B, hepatitis C, GB virus C, and TT virus, group of herpes viruses, adenovirus and enterovirus were all negative. No pathologic bacteria were cultured from blood, or body secretions. Urine and serum amino acid concentrations showed a pattern compatible with severe hepatic failure. No succinyl acetone was detected in urine. Urine bile acid concentrations showed a pattern compatible with cholestasis. Alpha 1 anti-trypsin concentration was normal. Serum AFP concentration and ratio of AFP-L3 were elevated at 205,600 ng/mL (normal upper limit, 1,000 ng/mL) and 21% (normal upper limit, 10%), respectively.
A plain computed tomographic abdominal scan showed increased density of the liver. Ultrasound findings showed high echogenicity of the liver. We assumed that his condition was compatible with NH. Serum measures of iron status at the age of 62 days showed elevated ferritin concentration (1,169.6 ng/mL), normal serum iron (175 μg/dL, normal range 40-180 μg/dL), low transferrin concentration (124 mg/dL, normal range 190-320 mg/dL) and markedly elevated transferrin (97.2%, upper normal limit 30%).
The histologic findings on needle liver biopsy at age 63 days showed fibrous septa surrounded by irregular islands of liver parenchyma (Fig. 2). Periportal hepatocytes contained extensive iron deposition, whereas there was no excess of iron in Kupffer cells (Fig. 3). The grading for iron content, performed using the method of Blisard et al (2), was +1.
Because the parents refused further treatment such as an antioxidant and chelator regimen or liver transplantation, the child's treatment consisted of transfusions of fresh-frozen plasma, supplementation with fat-soluble vitamins and a protein-reduced diet. At the age of 6 months, total bilirubin, alanine aminotransferase and hepaplastin test were all normal. At the age of 9 months, AFP was normal without detectable AFP-L3. At age 2 years he continued asymptomatic.
There have been several reports of NH occurring in siblings and half-siblings. A responsible gene has not yet been found in these families. The prognosis of NH is variable. Many infants die of hepatic failure or require liver transplantation, whereas some patients survive without any specific medication. Because NH is rare, it is important to observe the clinical course in detail in each case. Accumulated data will contribute to elucidation of the etiology of and guidelines for treatment. Recognition of this entity in living neonates like the patient in our case is necessary for a better understanding of etiology, pathogenesis and treatment. This is the fourth patient worldwide and the first patient in Japan to survive in good health (3-5).
This article reports a case of NH in which the patient survived without use of any specific medication such as an antioxidant and chelator regimen or liver transplantation and compares its clinical and pathologic features with those of two additional cases with favorable outcome (3,4). In our patient, serum ferritin level at the onset was extremely high (145,900 ng/mL). Serum alpha-fetoprotein was also very high (205,600 ng/mL), as was AFP-L3, a high level of which indicates high specificity of hepatocellular carcinoma in adults. At first, we supposed that the prognosis of this patient would be poor, and we informed his parents that his clinical course appeared to be fatal. His parents refused further medical therapy such as an antioxidant-chelation cocktail or liver transplantation because those measures were not curative. Contrary to our prediction, his clinical course was good. His development has been normal without medication.
Colletti and Clemmons (3) and Müller-Berghaus et al (4) previously reported cases with favorable outcomes. We compared the clinicopathogenic features of our patient with these other survivors (Table 1). Little difference was observed between the clinical features of our patient and the previous two patients. The serum ferritin level at onset was much higher in our case than in those of the other two survivors. Flynn et al. (6) reported that survivors receiving antioxidants had lower peak ferritin levels and a milder clinical phenotype. Serum ferritin level does not seem to have been one of the prognostic factors in our patient.
Proven or suspected infectious liver disease has been associated with a hemochromatotic phenotype (7-9). As Müller-Berghaus et al. (4) have reported, NH in some infants who survive might not be associated with metabolic disorder but rather with a single episode of submassive hepatocellular necrosis such that they are responsive to supportive care. Given that the liver is the primary site of apotransferrin synthesis, hepatocellular injury in utero may produce hypotransferrinemia (10). On the other hand, primary hypotransferrinemia is not associated with neonatal liver disease (11).
The grading for iron content in our patient was +1, a grade not associated with organ dysfunction in previous reports (2). Our patient had reduced signal intensity in the liver, suggesting siderosis, but evidence of pancreatic or myocardial siderosis was not demonstrated by magnetic resonance imaging. Oliveira et al. (12) reported a patient whose diagnosis, in the absence of iron deposits in salivary glands, was determined by abnormal ferritin levels and hepatic insufficiency and was confirmed by postmortem microscopic studies. In our case, because the specimens were obtained by needle biopsy, it is possible that insufficient liver was obtained for diagnosis.
Müller-Berghaus et al. also (4) reported that since NH is associated with a substantial risk for hepatocellular carcinoma, regular screening of α-fetoprotein is advisable. In fact, a case described by Oliveira et al. (12) revealed tumor nodule. Our patient had extremely high levels of α-fetoprotein in addition to a high ratio of L3. His condition, however, was very good. In an adult series with hepatocellular carcinoma, with the conventional cut-off level of 15% of α-fetoprotein-L3, α-fetoprotein-L3-positive hepatocellular carcinoma showed biologically malignant characteristics, especially portal vein invasion and lower tumor classification (13). In a pediatric series with hepatoblastoma, the level of α-fetoprotein-L3 was shown to decrease with the age of the patient at presentation, and did not correlate with prognosis of hepatoblastoma (14).
Siafakas et al. (15) reported that patients with abnormal bile acid metabolism may be a distinct subset of neonatal hemochromatosis with an accelerated course, no response to therapy and poor prognosis. The bile acid profile of our patient showed only cholestasis, which might partly explain his good condition.
In summary, our patient with NH-like liver failure might have had a single episode of submassive liver cell necrosis and regeneration from the perinatal to the neonatal period, which enabled him to survive without any specific medication for NH. The etiologic factors for this condition are still uncertain.
The authors are grateful to Dr. Makoto Obata, who referred the patient to us. We are grateful for the agreement with our case report of the patient's parents.
1. Boissieu DD, Knisely AS. Neonatal hemochromatosis. In: FJ Suchy, RJ Sokol, WF Balistereri, eds. Liver Disease in Children
. 2nd ed. Philadelphia: Lippincott Williams & Wilkins; 2001:641-47.
2. Blisard KS, Bartow SA. Neonatal hemochromatosis. Hum Pathol
3. Colletti RB, Clemmons JJW. Familial neonatal hemochromatosis with survival
. J Pediatr Gastroenterol Nutr
4. Müller-Berghaus J, Knisely AS, Zaum R, et al. Neonatal haemochromatosis: report of a patient with favorable outcome
. Eur J Pediatr
5. Rodrigues F, Kallas M, Nash R, et al. Neonatal hemochromatosis in eleven families: pattern of presentation and outcome
6. Flynn DM, Mohan N, McKiernan N, et al. Progress in treatment and outcome
for children with neonatal haemochromatosis. Arch Dis Child Fetal Neonatal Ed
7. Bove KE, Wong R, Kagen H, et al. Exogenous iron overload in perinatal hemochromatosis: a case report. Pediatr Pathol
8. Hoogstraten J, de Sa DJ, Knisely AS. Fetal liver disease may precede extrahepatic siderosis in neonatal hemochromatosis. Gastroenterology
9. Kershisnik MM, Knisely AS, Sun CCJ, et al. Cytomegalovirus infection, fetal liver disease, and neonatal hemochromatosis. Hum Pathol
10. Knisely AS, Grady RW, Kramer EE, et al, Cytoferrin, maternofetal iron transport, and neonatal hemochromatosis. Am J Clin Pathol
11. Craven CM, Alexander J, Eldridge M, et al, Tissue distribution and clearance kinetics of non-transferrin-bound iron in the hypotransferrinemic mouse: a rodent model for hemochromatosis. Proc Natl Acad Sci U S A
12. Oliveira MG, Fernandes A, Silva AC, et al. A case of neonatal haemochromatosis. Acta Pediatr
13. Oka H, Saito A, Ito K, et al. Multicenter prospective analysis of newly diagnosed hepatocellular carcinoma with respect to the percentage of Lens culinaris agglutinin-reactive alpha-fetoprotein
. J Gastroenterol Hepatol
14. Tsuchida Y, Honna T, Fukui M, et al. The ratio of fucosylation of alpha-fetoprotein
in hepatoblastoma. Cancer
15. Siafakas CG, Jonas MM, Perez-Atayde AR. Abnormal bile acid metabolism and neonatal hemochromatosis: a subset with poor prognosis. J Pediatr Gastroenterol Nutr
Keywords:© 2005 Lippincott Williams & Wilkins, Inc.
fetoprotein; lectin-reactive profiles; survival; outcome