当前位置: 首页 > 期刊 > 《新英格兰医药杂志》 > 2005年第2期 > 正文
编号:11325576
Case 21-2005 — A Four-Week-Old Male Infant with Jaundice and Thrombocytopenia
http://www.100md.com 《新英格兰医药杂志》
     Presentation of Case

    A four-week-old male infant was admitted to this hospital because of jaundice, hyperbilirubinemia, thrombocytopenia, and abdominal distention.

    The patient was born at term at another hospital by spontaneous vaginal delivery to a 37-year-old woman (gravida 2, para 2) after an uncomplicated pregnancy. Prenatal screening revealed that the mother had type A Rh-positive blood. Tests for hepatitis B surface antigen and group B streptococcus were negative. The mother was immune to rubella, and a rapid plasma reagin test was nonreactive. Before the delivery, decreased variability was noted on the fetal-heart tracing. At delivery, the infant was limp, cyanotic, apneic, and without a heartbeat. Positive-pressure ventilation administered by bag and face mask was initiated, and chest compressions were performed for approximately two minutes, at which time the infant began crying and the heart rate was greater than 100 beats per minute. He was treated with blow-by oxygen. The infant's Apgar scores, recorded as 0 at 1 minute, improved to 7 at 5 minutes (–1 tone, –1 reflex, –1 color) and to 8 at 10 minutes.

    The infant was transferred to a special care nursery in the same hospital, where he was placed in a hood to receive oxygen therapy with a fraction of inspired oxygen (FiO2) of 0.3. On physical examination, the weight was 3555 g. A systolic ejection murmur was present. The serum levels of electrolytes, calcium, magnesium, and phosphorus were within normal ranges; the results of renal-function tests were normal, and the results of other laboratory tests are shown in Table 1 and Table 2. Transient signs of mild respiratory distress, including grunting, developed. Intravenous normal saline (10 ml per kilogram of body weight per day), dextrose (10 percent, 2 ml per kilogram), sodium bicarbonate, erythromycin ophthalmic ointment, vitamin K (1 mg), and hepatitis B vaccine (1 μg) were administered. A specimen of blood was obtained and sent for culture; a chest radiograph showed cardiomegaly. Later that day, the infant was transferred to a tertiary care pediatric hospital. Before the transfer, the oxygen saturation was above 95 percent with the patient in a hood receiving oxygen therapy with an FiO2 of 0.3, and the respiratory rate was approximately 80 to 85 breaths per minute.

    Table 1. Hematologic Laboratory Values.

    Table 2. Chemistry Laboratory Values.

    At the second hospital, the temperature was 37°C, the blood pressure 74/46 mm Hg, the heart rate 140 beats per minute, and the respiratory rate 40 breaths per minute. The oxygen saturation was 92 percent while the infant was receiving 150 ml of supplemental oxygen per minute. The length was 52 cm (the 90th percentile), the weight 3580 g (75th percentile), and the head circumference 34 cm (approximately the 70th percentile). The infant was a nondysmorphic male in mild respiratory distress with pink skin. Auscultation of the chest disclosed symmetric scattered crackles, and there were mild intercostal retractions. The first and second heart sounds were normal; there was a 3/6 systolic murmur heard best at the clavicular and left sternal border. The precordium was quiet, pulses were 2+ and equal, and the skin was well perfused. The abdomen was soft, the liver extended 2 cm below the right costal margin, and a palpable mass was noted in the left side of the abdomen, which was thought to represent the spleen; bowel sounds were present. The remainder of the examination was normal. The results of laboratory test are shown in Table 1 and Table 2.

    At the tertiary care hospital, during the first two days of life, phototherapy was administered. Specimens of blood were obtained for culture, and ampicillin and gentamicin were administered. An electrocardiogram revealed right atrial enlargement and right ventricular hypertrophy with a strain pattern and possible biventricular hypertrophy. Chest radiography revealed an enlarged cardiac silhouette and clear lungs. Cardiac ultrasonography showed right ventricular hypertrophy, right ventricular hypertension, mild mitral regurgitation, trace tricuspid regurgitation, septal hypertrophy, a dilated ascending aorta, a patent foramen ovale with bidirectional blood flow, and normal biventricular function. The results were interpreted to be consistent with premature closure of the ductus arteriosus. An abdominal radiograph showed no evidence of bowel obstruction. Abdominal ultrasonography showed mild heterogeneity of the liver, with no evidence of intrahepatic biliary dilatation. The spleen was mildly to moderately enlarged, at approximately 6.9 cm. The kidneys and pancreas appeared to be normal.

    On the third and fourth days of life, the patient was jaundiced and hypotonic. A hematologist was consulted; the infant had type A Rh-positive blood, and screening for antibodies to red-cell antigens was negative; a screening for a deficiency of glucose-6-phosphate dehydrogenase, a Coombs' test, and a Heinz-body preparation were all negative. The urinalysis showed trace blood, few red cells, and bilirubin; the results of urine cultures were negative. Oral feedings with formula were begun. Phototherapy was discontinued, and phenobarbital was started at 3 mg per kilogram per day. Blood cultures were negative, and antibiotics were discontinued. A magnetic resonance imaging (MRI) study of the brain revealed patches of mild T2 hyperintensity in the white matter of the centrum semiovale, which indicates increased water content; there was no sign of restricted diffusion to suggest ischemia. Magnetic resonance spectroscopy showed no abnormalities; there were no signs of hydrocephalus, fluid collections, or masses. The results of a newborn screening showed no abnormalities. One milligram of vitamin K was administered to the infant on the sixth hospital day. Screening tests for galactosemia and tyrosinemia were negative.

    On the eighth day of life, the infant was discharged back to the first hospital, still with jaundice and hypotonia. The supplemental oxygen was tapered and then discontinued over the course of the next seven days. The stools were yellow, and there were two brief episodes of arterial-blood desaturation while the infant was feeding. Phenobarbital was discontinued on the 16th day of life. On the 22nd day, the sclerae were icteric, and the jaundice of the skin was resolving. The infant was discharged to his home receiving no medications with plans to return as an outpatient to his pediatrician for further evaluation and follow-up laboratory testing.

    Four days later, the father noted that the infant's abdomen was distended. A pediatrician examined the infant three days later and noted abdominal distention; laboratory tests showed persistent elevation of the bilirubin level, and the liver-function test results were abnormal (Table 2). The infant was admitted to this hospital.

    The infant had been feeding well, had good urine output, and had not been febrile. There were no risk factors for sepsis at birth. He lived with his parents and a healthy sibling. The family had no history of liver, gastrointestinal, or congenital heart diseases.

    On examination, he was alert and in no distress. The temperature was 37.1°C, the blood pressure 62/42 mm Hg, the heart rate 168 beats per minute, and the respiratory rate 36 breaths per minute. The oxygen saturation was 99 percent while he was breathing ambient air. The weight was 4010 g, the length 54 cm, and the head circumference 36 cm. The sclerae were icteric, and the skin was jaundiced to the lower thighs. A 2/6 systolic ejection murmur was heard over the left sternal border; no gallop or rub was present. The abdomen was soft and moderately distended with active bowel sounds; there was a palpable spleen tip and no hepatomegaly. The remainder of the examination was normal.

    The results of laboratory tests on admission are shown in Table 1 and Table 2. A radiograph of the chest showed clear lungs and cardiomegaly without pulmonary edema. An abdominal radiograph showed no free air or bowel obstruction. An electrocardiogram showed a normal sinus rhythm, right-axis deviation of +135 degrees, right ventricular hypertrophy, and ST-segment depression in leads V1 to V3. Abdominal ultrasonography showed arterialized flow in the left hepatic vein that suggested an arteriovenous fistula, a patent ductus venosus, congested left portal veins, splenomegaly, ascites, and heterogeneous liver parenchyma that was thought to be due to abnormal perfusion. Transthoracic echocardiography with color-flow Doppler revealed mild right ventricular dilation, right ventricular hypertrophy, and prominent portal and hepatic veins. The right ventricular function was normal; there was a left-to-right shunt at the atrial level across a patent foramen ovale and no arteriovenous malformation. The peak pulmonary gradient was 12 mm Hg, the mean pulmonary gradient 5 mm Hg, and the estimated right ventricular systolic pressure 28 mm Hg. The left ventricular function was normal, and the ejection fraction was 65 percent.

    The urine was amber colored and clear with a pH of 6.5, specific gravity 1.008, and 2+ bilirubin. Cultures of urine and stool specimens for routine pathogens and cytomegalovirus were negative. A three-day course of vitamin K (4 mg per day) was initiated, and on the second hospital day an oral multivitamin suspension was begun.

    On the third hospital day, an MRI of the liver revealed diffuse nodularity that was consistent with cirrhosis. The liver and pancreas were relatively hypointense on T2-weighted images, except for a small area of hyperintensity in the right lobe of the liver that was thought to represent a cyst. The spleen was moderately enlarged but normal in signal intensity. The left portal vein was more prominent than the right, with numerous possible connections between the left portal vein and the enlarged left hepatic vein. There were some vessels in the superficial aspect of the liver that were thought to be abnormal arteries. The bowel was moderately edematous, the heart had slightly decreased signal intensity, and both kidneys enhanced symmetrically.

    A diagnostic procedure was performed.

    Differential Diagnosis

    Dr. Nancy C. Andrews: This patient presented at one month of life with signs of liver dysfunction. He had unexplained problems in his first days of life that included right ventricular hypertrophy, thrombocytopenia, and persistent direct hyperbilirubinemia. He did not have lactic acidosis or signs of hemolysis, and the results of tests for galactosemia and tyrosinemia were normal. He was discharged to his home at age three weeks, but he returned to the hospital several days later with abdominal distention, jaundice, and abnormal results on tests of liver function.

    May we review the imaging studies?

    Dr. Sudha Anupindi: An MRI obtained after the administration of intravenous contrast material revealed a nodular liver with heterogeneous architecture, severe ascites (Figure 1A and Figure 1B), and decreased signal intensity within the liver and pancreas, but normal signal in the spleen (Figure 1C). The left lobe of the liver appeared enlarged, and the right lobe was atrophied. This finding is not unusual in chronic liver disease; the caudate lobe of the liver tends to be enlarged in patients with cirrhosis.

    Figure 1. Magnetic Resonance Images of the Liver.

    On a coronal T1-weighted image (Panel A), the liver has a nodular contour (arrows). On a coronal T2-weighted image (Panel B), the liver has uniformly decreased signal intensity (white arrow). There is a large amount of ascites (black arrow). On an axial T2-weighted image (Panel C), there is decreased signal intensity within the parenchyma of the pancreas (black arrow); there is markedly decreased signal intensity within the liver (white arrow), but the enlarged spleen maintains normal signal intensity (arrowhead).

    A magnetic resonance angiogram showed abnormal vasculature within the liver. The main portal vein was normal in caliber, with an unusually large left portal vein and a hypoplastic right portal vein. Aberrant vessels were also seen in the left lobe. The aberrant vessels and the discrepancy in size between the left lobe and the right are probably a reflection of altered blood-flow patterns.

    In summary, the imaging findings were consistent with chronic liver disease and portal hypertension. The decreased signal within the liver and pancreas suggested iron overload.

    Dr. Andrews: Iron overload in a newborn is called neonatal hemochromatosis. This rare disorder is prominent in the differential diagnosis of hepatic failure in young infants. There are several causes. Some cases can be attributed to abnormalities in maternofetal iron transfer in the third trimester. In others, tissue iron deposition may be secondary to liver injury.

    Neonatal hemochromatosis is typically characterized by fulminant hepatic failure in the first few days of life, with hypoglycemia, coagulopathy, thrombocytopenia, anemia, and renal failure. Circulating alpha-fetoprotein levels, which were not measured in this pregnancy, are usually high, and serum transferrin is usually hypersaturated in an infant who has neonatal hemochromatosis. The ductus venosus is frequently patent, as it was in this patient. Neonatal hemochromatosis is not usually a diagnostic problem. However, the clinical picture can be variable, and patients who present late with the symptoms, as in this case, have been reported.1 The MRI scan was helpful in deciding on the diagnosis, because iron deposits were detected in the liver, heart, and pancreas but not in the spleen; these findings are typical of neonatal hemochromatosis. Biochemical abnormalities that are revealed in tests of liver function are relatively nonspecific and could result from other neonatal liver disorders. This patient's elevated serum ferritin level was suggestive but not diagnostic; hepatocyte destruction of any cause leads to the release of intracellular ferritin into the circulation. Also, levels of proteins produced in the liver (e.g., transferrin, ceruloplasmin, and alpha1-antitrypsin) may be depressed in the setting of severe liver damage.

    Metabolic Causes of Neonatal Hemochromatosis

    Several metabolic disorders have been reported to be associated with neonatal hemochromatosis (Table 3). This pattern of iron deposition has been seen in infants with defects in bile acid biosynthesis, particularly deficiency of 4-3-oxosteroid 5-reductase.3,4 As in this case, infants with a presumed deficiency of 4-3-oxosteroid 5-reductase have direct hyperbilirubinemia and elevated levels of transaminase but relatively normal levels of -glutamyltransferase and normal alkaline phosphatase levels. Infants with these signs tend to present at several weeks of age, although jaundice is noted at birth. The diagnosis is suggested by the very high levels of 3-oxo-4 bile acids and allo-(5-H) bile acids detected in the urine. However, levels of 4-3-oxosteroid 5-reductase may be decreased when there is liver damage from other causes. Recently, mutations in SRD5B1, the gene encoding the reductase, were described in a small series of patients.10 However, none of the patients were reported to have neonatal hemochromatosis. Thus, the question of whether inherited 5-reductase deficiency truly causes neonatal hemochromatosis remains open.

    Table 3. Disorders Associated with Neonatal Iron Overload.

    Neonatal hemochromatosis has been described as a feature of several complex syndromes. Infants with the trichohepatoenteric syndrome have iron loading in the liver in association with abnormal hair structure and intractable diarrhea.6 Renal-tubule dysgenesis has also been associated with neonatal hemochromatosis.11 Pearson's marrow–pancreas syndrome can be associated with hepatic siderosis in the newborn period.12 However, this infant did not have any of the findings associated with these disorders.

    A distinct iron-overload disorder of growth retardation, aminoaciduria, cholestasis, iron overload, lactic acidosis, and early death, termed the GRACILE syndrome, has been described in several Finnish families.9 It results from a mutation in a gene (BCS1L) encoding a mitochondrial chaperone protein that is important for assembly of the respiratory chain. However, patients with the GRACILE syndrome do not have extrahepatic siderosis, as this patient did. Evidence of other defects in the respiratory chain might be sought in this patient, but they are probably ruled out by the absence of a prominent acidosis.

    Several forms of anemia, including congenital dyserythropoietic anemias, are associated with iron overload that is due to increased intestinal iron absorption. However, the iron overload associated with these anemias is not apparent until later in childhood, and there was no evidence of a primary erythropoietic defect in this patient.

    Maternal Conditions Associated with Neonatal Hemochromatosis

    Neonatal hemochromatosis need not be due to a defect in the developing fetus. Although some family pedigrees suggest autosomal recessive inheritance, others include affected half-siblings who have the mother in common but not the father, suggesting the possibility of a maternal factor. This could be a mitochondrial DNA abnormality, but none has been reported to date. Alternatively, it could be an acquired maternal factor, such as an autoantibody. In this regard, neonatal hemochromatosis has been reported in infants of mothers with Sj?gren's syndrome who had autoantibodies against Ro/SS-A and La/SS-B antigens.5 There was no report of maternal autoimmune disease in the case under discussion, though it is still possible that iron loading was the result of an alloimmune process that took place during gestation. One trial13 suggested that high-dose intravenous immune globulin therapy improved outcomes in subsequent pregnancies of mothers of infants with neonatal hemochromatosis.

    Primary Abnormalities of Iron Homeostasis

    Mutations in five genes cause iron overload (Table 4). Hemochromatosis in adults is typically a disease of the liver, with less severe manifestations in the heart and pancreas than are seen in infants. In contrast, cardiomyopathy, arrhythmias, and endocrine dysfunction predominate in early-onset hemochromatosis. Classic hemochromatosis is a result of mutations in HFE.14 A similar disease is caused by mutations in transferrin receptor 2 (TFR2).15 The penetrance of TFR2-associated hemochromatosis is not known, but hemochromatosis develops in only a fraction of patients who are homozygous for HFE mutations. HFE mutations have not been identified in patients such as this infant with neonatal hemochromatosis.2 Most patients with juvenile hemochromatosis are homozygous for mutations in the gene encoding hemojuvelin, a novel protein of unknown function.16 A subgroup are homozygous for mutations in the gene encoding hepcidin, a peptide hormone that attenuates intestinal iron absorption and cellular iron release.17 There are no reports of mutations in either of these genes in patients with neonatal hemochromatosis.

    Table 4. Mutated Genes in Hereditary Iron-Overload Disorders.

    In most patients with neonatal hemochromatosis, no cause is identified. The characteristic features include iron deposition in the liver, heart, and exocrine pancreas. The liver shows hepatocellular siderosis and necrosis with giant-cell or pseudoacinar transformation, but no iron accumulation in macrophages of the liver, spleen, or bone marrow. This pattern of iron deposition is similar to that observed in forms of hemochromatosis that have later onset.

    The diagnostic procedure performed in this case was probably a biopsy to document tissue iron overload. It could have been a liver biopsy, if blood coagulation could be controlled adequately to make it safe. Alternatively, the biopsy could have been of the oral mucosa to document iron overload in minor salivary glands. This type of biopsy is helpful only if excess iron is detected.

    Dr. Nancy Lee Harris (Pathology): Dr. Buie, you cared for this patient in the hospital; could you give your impressions and tell us about the diagnostic procedure?

    Dr. Timothy M. Buie (Pediatric Gastroenterology): Our impression was that this patient had neonatal hemochromatosis. We had test results that ruled out other causes, and the extent of the iron deposition in the organs led us to believe that neonatal hemochromatosis was the most likely diagnosis. We attempted a biopsy of a salivary gland, but we were not able to obtain sufficient tissue and therefore performed a liver biopsy.

    Clinical Diagnosis

    Neonatal hemochromatosis.

    Dr. Nancy Andrews's Diagnosis

    Neonatal hemochromatosis of uncertain cause.

    Pathological Discussion

    Dr. Kamran Badizadegan: The diagnostic procedure was a liver biopsy, and a core of hepatic parenchyma was obtained. At low magnification (Figure 2A) the parenchyma shows lobular disarray with prominent bands of fibrosis. The ends of the core are somewhat rounded, which suggests the development of regenerative nodules. Examination after trichrome staining at higher magnification (Figure 2B) confirms the presence of bands of fibrosis with early regenerative nodules, which are suggestive of progression to early cirrhosis.

    Figure 2. Liver-Biopsy Specimen.

    At low magnification (Panel A, hematoxylin and eosin), the specimen from the core biopsy of hepatic parenchyma shows lobular disarray with focally prominent bands of fibrosis (arrowhead). The two ends of the core appear somewhat rounded, suggesting the development of early regenerative nodules. Trichrome staining (Panel B) confirms the presence of bridging and pericellular fibrosis (marked by blue staining) with a focally nodular pattern. A high-power examination of the parenchyma (Panel C, hematoxylin and eosin) shows diffuse hepatocellular damage with pseudogland formation (arrow) and multinucleated giant-cell transformation (top left corner). In addition, there is prominent cytoplasmic pigment deposition seen as brown staining of many of the cells. Prussian blue staining on an adjacent serial section (Panel D) confirms the identity of much of the cytoplasmic pigments as iron. Staining is not limited to the hepatocytes, and reticuloendothelial storage is also present, as seen in three prominent portal-tract macrophages (inset).

    A microscopical examination of the hepatocytes shows diffuse damage that includes the loss of normal sinusoidal architecture, pseudogland formation, and giant-cell transformation (Figure 2C) with focal intrahepatic cholestasis and spotty single-cell necrosis (not shown). There is cytoplasmic pigment in hepatocytes, Kupffer cells, and portal macrophages, most of which are positive for the presence of iron, as shown by Prussian blue staining (Figure 2D). Extramedullary hematopoiesis was prominent. No viral cytopathic changes or microorganisms were identified.

    The histologic findings are diagnostic of neonatal hepatitis with giant-cell transformation of hepatocytes. The extent and pattern of iron deposition in the liver biopsy are not specific for disorders of iron metabolism and can be seen with other causes of extensive neonatal liver damage or secondary iron overload. Neonatal hepatitis is a morphologic pattern of injury characterized by diffuse hepatocellular damage, focal hepatocellular necrosis, intrahepatic cholestasis, extramedullary hematopoiesis, and giant-cell transformation of hepatocytes.18 The diagnosis is typically reserved for infants younger than three months of age in whom extrahepatic biliary obstruction has been ruled out. The morphologic pattern of neonatal hepatitis can be seen in a variety of disorders, including infections and metabolic diseases, many of which are difficult to distinguish on the basis of histopathological findings alone. Although neonatal giant-cell hepatitis is a morphologic manifestation of neonatal hemochromatosis,19,20 liver biopsy alone is often not sufficient for a definitive diagnosis of this entity in the neonatal period.

    Dr. Andrews: There are few treatment options for patients with neonatal hemochromatosis.21,22 Defects in bile acid biosynthesis, if truly causative in this disorder, may be amenable to oral bile acid therapy. Antioxidant "cocktails" aimed at minimizing oxidative damage catalyzed by free iron appear to be of benefit in some patients. However, in some cases liver transplantation may be the only reasonable option. Counseling for the parents is an important component of management in this disorder, because neonatal hemochromatosis recurs in about 75 percent of subsequent pregnancies.

    Dr. Harris: Dr. Buie, would you tell us how you treated this patient and how he is doing?

    Dr. Buie: Antioxidant therapy consisting of liquid vitamin E, selenium, and acetylcysteine was administered. The ascites was controlled with diuretics, and we have continued his therapy with multivitamins. The levels of alanine aminotransferase and aspartate aminotransferase are still elevated to about twice the normal values, but clinically he is well. He has no ascites, his height and weight are at the 50th percentile, and he is currently taking no medications. He has persistently elevated levels of alpha-fetoprotein, which in this age group may be associated with inflammation. We are following him closely for signs of hepatocellular carcinoma and portal hypertension.

    Dr. Andrews: Not much is known about the natural history of this disorder. It has been reported to resolve in some patients. Recent reports suggest that a specific maternal antibody is involved.13 If so, the insult would be specific to the fetal period, and this patient would not be likely to have further problems with iron metabolism.

    Anatomical Diagnosis

    Neonatal giant-cell hepatitis with bridging fibrosis and early regenerative nodules, with iron deposition in hepatocytes and Kupffer cells (consistent with — but not diagnostic of — neonatal hemochromatosis).

    Source Information

    From the Department of Pediatrics, Children's Hospital (N.C.A.); the Departments of Radiology (S.A.) and Pathology (K.B.), Massachusetts General Hospital; and the Departments of Pediatrics (N.C.A.), Radiology (S.A.), and Pathology (K.B.), Harvard Medical School — all in Boston.

    References

    Knisely AS, Mieli-Vergani G, Whitington PF. Neonatal hemochromatosis. Gastroenterol Clin North Am 2003;32:877-889.

    Kelly AL, Lunt PW, Rodrigues F, et al. Classification and genetic features of neonatal haemochromatosis: a study of 27 affected pedigrees and molecular analysis of genes implicated in iron metabolism. J Med Genet 2001;38:599-610.

    Shneider BL, Setchell KD, Whitington PF, Neilson KA, Suchy FJ. Delta 4-3-oxosteroid 5 beta-reductase deficiency causing neonatal liver failure and hemochromatosis. J Pediatr 1994;124:234-238.

    Siafakas CG, Jonas MM, Perez-Atayde AR. Abnormal bile acid metabolism and neonatal hemochromatosis: a subset with poor prognosis. J Pediatr Gastroenterol Nutr 1997;25:321-326.

    Schoenlebe J, Buyon JP, Zitelli BJ, Friedman D, Greco MA, Knisely AS. Neonatal hemochromatosis associated with maternal autoantibodies against Ro/SS-A and La/SS-B ribonucleoproteins. Am J Dis Child 1993;147:1072-1075.

    Verloes A, Lombet J, Lambert Y, et al. Tricho-hepato-enteric syndrome: further delineation of a distinct syndrome with neonatal hemochromatosis phenotype, intractable diarrhea, and hair anomalies. Am J Med Genet 1997;68:391-395.

    Parizhskaya M, Reyes J, Jaffe R. Hemophagocytic syndrome presenting as acute hepatic failure in two infants: clinical overlap with neonatal hemochromatosis. Pediatr Dev Pathol 1999;2:360-366.

    Kershisnik MM, Knisely AS, Sun CC, Andrews JM, Wittwer CT. Cytomegalovirus infection, fetal liver disease, and neonatal hemochromatosis. Hum Pathol 1992;23:1075-1080.

    Fellman V. The GRACILE syndrome, a neonatal lethal metabolic disorder with iron overload. Blood Cells Mol Dis 2002;29:444-450.

    Lemonde HA, Custard EJ, Bouquet J, et al. Mutations in SRD5B1 (AKR1D1), the gene encoding delta(4)-3-oxosteroid 5beta-reductase, in hepatitis and liver failure in infancy. Gut 2003;52:1494-1499.

    Morris S, Akima S, Dahlstrom JE, Ellwood D, Kent A, Falk MC. Renal tubular dysgenesis and neonatal hemochromatosis without pulmonary hypoplasia. Pediatr Nephrol 2004;19:341-344.

    Krahenbuhl S, Kleinle S, Henz S, et al. Microvesicular steatosis, hemosiderosis and rapid development of liver cirrhosis in a patient with Pearson's syndrome. J Hepatol 1999;31:550-555.

    Whitington PF, Hibbard JU. High-dose immunoglobulin during pregnancy for recurrent neonatal haemochromatosis. Lancet 2004;364:1690-1698.

    Feder JN, Gnirke A, Thomas W, et al. A novel MHC class I-like gene is mutated in patients with hereditary haemochromatosis. Nat Genet 1996;13:399-408.

    Camaschella C, Roetto A, Cali A, et al. The gene TFR2 is mutated in a new type of haemochromatosis mapping to 7q22. Nat Genet 2000;25:14-15.

    Papanikolaou G, Samuels ME, Ludwig EH, et al. Mutations in HFE2 cause iron overload in chromosome 1q-linked juvenile hemochromatosis. Nat Genet 2004;36:77-82.

    Roetto A, Papanikolaou G, Politou M, et al. Mutant antimicrobial peptide hepcidin is associated with severe juvenile hemochromatosis. Nat Genet 2002;33:21-22.

    Rosenthal P. Neonatal hepatitis and congenital infections. In: Suchy FJ, Sokol RJ, Balistreri WF, eds. Liver disease in children. 2nd ed. Philadelphia: Lippincott Williams & Wilkins, 2001:239-52.

    Fienberg R. Perinatal idiopathic hemochromatosis: giant cell hepatitis interpreted as an inborn error of metabolism. Am J Clin Pathol 1960;33:480-491.

    Laurendeau T, Hill JE, Manning GB. Idiopathic neonatal hemochromatosis in siblings: an inborn error of metabolism. Arch Pathol 1961;72:410-423.

    Sigurdsson L, Reyes J, Kocoshis SA, Hansen TW, Rosh J, Knisely AS. Neonatal hemochromatosis: outcomes of pharmacologic and surgical therapies. J Pediatr Gastroenterol Nutr 1998;26:85-89.

    Flynn DM, Mohan N, McKiernan P, et al. Progress in treatment and outcome for children with neonatal haemochromatosis. Arch Dis Child Fetal Neonatal Ed 2003;88:F124-F127.(Nancy C. Andrews, M.D., P)