Published online February 1, 2005
PEDIATRICS Vol. 115 No. 2 February 2005, pp. e233-e238 (doi:10.1542/peds.2004-1255)

This Article Full Text Full Text (PDF) Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services E-mail this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to My File Cabinet Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via CrossRef Citing Articles via Google Scholar Google Scholar Articles by Fattal-Valevski, A. Articles by Eshel, G. Search for Related Content PubMed PubMed Citation Articles by Fattal-Valevski, A. Articles by Eshel, G. Related Collections Nutrition & Metabolism Social Bookmarking                         What's this?

ELECTRONIC ARTICLE

Outbreak of Life-Threatening Thiamine Deficiency in Infants in Israel Caused by a Defective Soy-Based Formula

Aviva Fattal-Valevski, MD^*, Anat Kesler, MD, Ben-Ami Sela, PhD, Dorit Nitzan-Kaluski, MD, MPH, RD^||, Michael Rotstein, MD^*, Ronit Mesterman, MD^*, Hagit Toledano-Alhadef, MD^*, Chaim Stolovitch, MD^¶, Chen Hoffmann, MD^#, Omer Globus, MD^** and Gideon Eshel, MD

^* Institute for Child Development and Pediatric Neurology Unit, Sourasky Medical Center, Tel Aviv, Israel
Neuro-Ophthalmology Unit, Department of Ophthalmology, Sourasky Medical Center, Tel Aviv, Israel
Institute of Chemical Pathology, Sheba Medical Center, Tel Hashomer, Israel
^|| Food and Nutrition Services and Department of Epidemiology and Preventive Medicine, Ministry of Health, Jerusalem, Israel
^¶ Pediatric-Ophthalmology Unit, Department of Ophthalmology, Sourasky Medical Center, Tel Aviv, Israel
^# Neuroradiology Unit, Department of Radiology, Sheba Medical Center, Tel Hashomer, Israel
^** Pediatric Department, Sourasky Medical Center, Tel Aviv, Israel
Pediatric Intensive Care Unit, Assaf Harofeh Medical Center, Zrifin, Israel. All centers affiliated with the Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel

Objective. Between October and November 2003, several infants with encephalopathy were hospitalized in pediatric intensive care units in Israel. Two died of cardiomyopathy. Analysis of the accumulated data showed that all had been fed the same brand of soy-based formula (Remedia Super Soya 1), specifically manufactured for the Israeli market. The source was identified on November 6, 2003, when a 5.5-month-old infant was admitted to Sourasky Medical Center with upbeat nystagmus, ophthalmoplegia, and vomiting. Wernicke's encephalopathy was suspected, and treatment with supplementary thiamine was started. His condition improved within hours. Detailed history revealed that the infant was being fed the same formula, raising suspicions that it was deficient in thiamine. The formula was tested by the Israeli public health authorities, and the thiamine level was found to be undetectable (<0.5 µg/g). The product was pulled from the shelves, and the public was alerted. Thiamine deficiency in infants is very rare in developed countries. The aim of this study was to report the epidemiology of the outbreak and to describe the diagnosis, clinical course, and outcome of 9 affected infants in our care.

Methods. After the index case, an additional 8 infants were identified in our centers by medical history, physical examination, and laboratory testing. The group consisted of 6 male and 3 female infants aged 2 to 12 months. All were assessed with the erythrocyte transketolase activity assay, wherein the extent of thiamine deficiency is expressed in percentage stimulation compared with baseline (thiamine pyrophosphate effect [TPPE]). Normal values range from 0% to 15%; a value of 15% to 25% indicates thiamine deficiency, and >25% indicates severe deficiency. Blood lactate levels (normal: 0.5–2 mmol/L) were measured in 6 infants, cerebrospinal fluid lactate in 2 (normal: 0.5–2 mmol/L), and blood pyruvate in 4 (normal: 0.03–0.08 mmol/L). The diagnostic criteria for thiamine deficiency were abnormal transketolase activity and/or unexplained lactic acidosis. Treatment consisted of intramuscular thiamine 50 mg/day for 14 days combined with a switch to another infant formula.

Results. Early symptoms were nonspecific and included mainly vomiting (n = 8), lethargy (n = 7), irritability (n = 5), abdominal distension (n = 4), diarrhea (n = 4), respiratory symptoms (n = 4), developmental delay (n = 3), and failure to thrive (n = 2). Infection was found in all cases. Six infants were admitted with fever. One patient had clinical dysentery and group C Salmonella sepsis; the others had mild infection: acute gastroenteritis (n = 2); upper respiratory infection (n = 2); and bronchopneumonia, acute bronchitis, and viral infection (n = 1 each). Two infants were treated with antibiotics. Three infants had neurologic symptoms of ophthalmoplegia with bilateral abduction deficit with or without upbeat nystagmus. All 3 had blood lactic acidosis, and 2 had high cerebrospinal fluid lactate levels. Patient 1, our index case, was hospitalized for upbeat nystagmus and ophthalmoplegia, in addition to daily vomiting episodes since 4 months of age and weight loss of 0.5 kg. Findings on brain computed tomography were normal. Blood lactate levels were high, and TPPE was 37.8%. Brain magnetic resonance imaging (MRI) revealed no abnormalities. Patient 2, who presented at 5 months with lethargy, vomiting, grunting, and abdominal tenderness, was found to have intussusception on abdominal ultrasound and underwent 2 attempts at reduction with air enema several hours apart. However, the lethargy failed to resolve and ophthalmoplegia appeared the next day, leading to suspicions of Wernicke's encephalopathy. Laboratory tests showed severe thiamine deficiency (TPPE 31.2%). In patients 1 and 2, treatment led to complete resolution of symptoms. The third infant, a 5-month-old girl, was admitted on October 10, 2003, well before the outbreak was recognized, with vomiting, fever, and ophthalmoplegia. Her condition deteriorated to seizures, apnea, and coma. Brain MRI showed a bilateral symmetrical hyperintense signal in the basal ganglia, mamillary bodies, and periaqueductal gray matter. Suspecting a metabolic disease, vitamins were added to the intravenous solution, including thiamine 250 mg twice a day. Clinical improvement was noted 1 day later. TPPE assay performed after treatment with thiamine was started was still abnormal (17.6%). Her formula was substituted after 4 weeks, after the announcement about the thiamine deficiency. Although the MRI findings improved 5 weeks later, the infant had sequelae of ophthalmoplegia and motor abnormalities and is currently receiving physiotherapy. All 3 patients with neurologic manifestations were fed exclusively with the soy-based formula for 2 to 3.5 months, whereas the others had received solid food supplements. Longer administration of the formula (ie, chronic thiamine deficiency) was associated with failure to thrive. For example, one 12-month-old girl who received the defective formula for 8 months presented with refusal to eat, vomiting, failure to thrive (75th to <5th percentile), hypotonia, weakness, and motor delay. Extensive workup was negative for malabsorption and immunodeficiency. On admission, the patient had Salmonella gastroenteritis and sepsis and was treated with antibiotics. After thiamine deficiency was diagnosed, she received large doses of thiamine (50 mg/day) for 2 weeks. Like the other 5 infants without neurologic involvement, her clinical signs and symptoms disappeared completely within 2 to 3 weeks of treatment, and TPPE levels normalized within 1 to 7 days. There were no side effects. As part of its investigation, the Israel Ministry of Health screened 156 infants who were fed the soy-based formula for thiamine deficiency. However, by that time, most were already being fed alternative formulas and had begun oral thiamine treatment. Abnormal TPPE results (>15%) were noted in 8 infants, 3 male and 5 female, all >1 year old, who were receiving solid food supplements. Although their parents failed to notice any symptoms, irritability, lethargy, vomiting, anorexia, failure to thrive, and developmental delay were documented by the examining physicians. None had signs of neurologic involvement. Treatment consisted of oral thiamine supplements for 2 weeks.

Conclusions. Clinician awareness of the possibility of thiamine deficiency even in well-nourished infants is important for early recognition and prevention of irreversible brain damage. Therapy with large doses of thiamine should be initiated at the earliest suspicion of vitamin depletion, even before laboratory evidence is available and before neurologic or cardiologic symptoms appear.

---

Key Words: thiamine • Wernicke's encephalopathy • infant • formula • vitamin

Abbreviations: TPPE, thiamine pyrophosphate effect • CSF, cerebrospinal fluid • MRI, magnetic resonance imaging

---

Accepted Aug 5, 2004.

CiteULike    Connotea    Del.icio.us    Digg    Facebook    Reddit    Technorati    Twitter    What's this?

This article has been cited by other articles:

				A. Fattal-Valevski, A. Bloch-Mimouni, S. Kivity, E. Heyman, A. Brezner, R. Strausberg, D. Inbar, U. Kramer, and H. Goldberg-Stern Epilepsy in children with infantile thiamine deficiency Neurology, September 15, 2009; 73(11): 828 - 833. [Abstract] [Full Text] [PDF]

				G.-q. Fei, C. Zhong, L. Jin, J. Wang, Y. Zhang, X. Zheng, Y. Zhang, and Z. Hong Clinical Characteristics and MR Imaging Features of Nonalcoholic Wernicke Encephalopathy AJNR Am. J. Neuroradiol., January 1, 2008; 29(1): 164 - 169. [Abstract] [Full Text] [PDF]