Diagnosis of Inborn Errors of Metabolism

Archives of Medical Research 31 (2000) 145–150

REVIEW ARTICLE

Diagnosis of Inborn Errors of Metabolism Antonio Velázquez,* Marcela Vela-Amieva,* Isabel Cicerón-Arellano,* Isabel Ibarra-González,* Martha Elva Pérez-Andrade,* Zazil Olivares-Sandoval* and Gerardo Jiménez-Sánchez** *Unidad de Genética de la Nutrición, Instituto de Investigaciones Biomédicas UNAM e Instituto Nacional de Pediatría, México, D.F., Mexico **Howard Hughes Medical Institute, Johns Hopkins University School of Medicine, Baltimore, MD, USA Received for publication July 19, 1999; accepted December 20, 1999 (99/113).

Systematic detection of inborn errors of metabolism (IEM) has usually encountered difficulties in developing countries. We present our experience in a high-risk population in Mexico between 1973 and 1998 with particular reference to the last 10 years, during which time infrastructure and support were considerably improved. Only disorders of intermediary metabolism were sought. The total number of patients studied is not available, but in the last 10 years, patients numbered 5,186. Routine metabolic screening was performed on all patients, with additional tests according to the clinical picture and screening results. The referral criteria have increasingly diversified, one-third being neurological conditions. Of the referrals, 33.8% were from pediatricians (31.1% of whom were at critical medicine departments) and the remainder from specialists. The number of diagnosed patients has increased to 1 per 43.9 patients studied. Amino acid defects have been the most prevalent, the proportion of organic acid and carbohydrate disorders having increased in the last 10 years, associated with improved diagnostic facilities. The most frequently diagnosed diseases were PKU, type 1a glycogen storage, and maple syrup urine disease (MSUD), their frequency apparently varying among different regions of Mexico. Other results of our program include training of specialists and technicians, development of the Latin American Metabolic Information Network, a procedure to locally prepare a special food product low in phenylalanine for the treatment of PKU patients, and extension of approaches for these disorders to the investigation metabolic derangements of infant malnutrition. This work demonstrates that inherited metabolic diseases constitute a significant load in pediatric pathology and that their study can and should be pursued in developing nations. © 2000 IMSS. Published by Elsevier Science Inc. Key Words: Inborn errors of metabolism, Inherited metabolic disorders, Mexico.

Introduction The establishment of programs for the systematic detection and study of inborn errors of metabolism (IEM) has usually been a difficult process in developing nations (1). Mexico was no exception in the existence of many limitations and obstacles to be surpassed. Since 1973, our group has pioneered these efforts in Mexico, and indeed in Latin America (2), although it has been mainly over the past 10 years that a

Address reprint requests to: Antonio Velázquez, MD, PhD, Instituto Nacional de Pediatría, Av. de la IMAN No. 1, 4⬚ piso, 04530 México, D.F., Mexico. Tel.: (⫹525) 606-3558; FAX: (⫹525) 606-3489; E-mail: [email protected]

technically competent infrastructure with adequate support has been developed. Based on a pilot study that was carried out from 1974 to 1977 (3), a national neonatal screening program for congenital hypothyroidism (and a much more limited screening program for PKU, not reported here) has been in operation since 1989 (4,5). Unfortunately, there has been no governmental support to include screening for inborn errors of metabolism, such as PKU, except at a limited scale. The main objective of this article is to report our experience in the diagnosis of these disorders during 1973 through 1998, particularly in the last 10 years of this 25year effort. Our results on patient management will be the matter of another article. Early on, we decided to restrict our studies to disorders of intermediary metabolism, particu-

0188-4409/00 $–see front matter. Copyright © 2000 IMSS. Published by Elsevier Science Inc. PII S0188-4409(00)00 0 5 3 - 9

146

Velázquez et al./ Archives of Medical Research 31 (2000) 145–150

larly of amino acid, organic acid, and carbohydrate (simple sugars and glycogen storage diseases) metabolism, in order to better focus our rather limited resources, and thus be able to devote part of the resources to research into related areas of nutritional metabolism. Other groups in Mexico have worked on other groups of inborn errors of metabolism, such as lysosomal storage disorders (6), erythrocyte glycolytic defects (7), porphyrias (8), congenital adrenal hyperplasia (9), and mitochondrial diseases (10). In spite of these limitations, ours is the only comprehensive program for diagnosis and management of IEM in the entire country. Since 1980, our endeavor has been institutionalized at the Unit of Genetics in Nutrition (henceforth called the Unit) operated conjointly by the National Autonomous University of Mexico (UNAM) and the National Institute of Pediatrics (INP) (one of the major pediatric hospitals in Mexico). This arrangement has eased the influx of patients and their appropriate collaborative study by specialized personnel of the Unit and the clinical staff of the Institute, supported by an up-todate technical infrastructure. More recently, the Reproductive Health Agency of the Ministry of Health in Mexico has joined our efforts by seeing to the logistic and financial aspects of the neonatal screening program, extending it throughout the country. Our efforts have had ample impact on pediatrics, and indeed on health care, in Mexico. The purpose of this work is to describe our experience, ordeals, and achievements. We hope that part of our experience might be of benefit to others. Materials and Methods Patients. The patients studied were clinically affected and were referred to the Unit for metabolic evaluation. They form a clinically biased sample from which no epidemiological inferences may be drawn. The criteria for clinical suspicion has varied over the years, increasingly including acutely ill infants, reflecting both an increased awareness of metabolic diagnoses by the house staff (mainly pediatric residents) and our better technical facilities. In general, patients are referred to us when they have the following: (a) clinical manifestations suggestive of a specific disorder (e.g., galactosemia in the case of a patient with mental retardation, cataracts, and liver damage); (b) non-specific symptoms resulting in a diagnostic puzzle, especially if the patient has had similarly affected siblings, siblings with sudden infant death, and/or if the parents are consanguineous, and (c) acute neonatal or early infancy episodes, particularly if they are periodic and/or suggestive of sepsis or Reye’s syndrome. Most of the referred patients were from our own hospital (the National Institute of Pediatrics), but approximately one-fifth came from other hospitals. The Unit has been appointed the national reference center for the detection and study of these disorders by the Ministry of Health of the Federal Government. Initial testing. After a clinical history was obtained and a physical examination was carried out, laboratory tests were

performed according to the patient’s clinical picture. A metabolic screen was performed on all patients, regardless of their clinical picture. The samples included random urines and plasma or blood drops collected on filter paper (Schleicher & Schuell #903, Schleicher & Schuell, Inc., Keene, NH, USA) and obtained 1–2 h after a protein-containing meal. The screening comprised measurement of urinary pH, qualitative tests for glucose, ketones, bilirubin and blood, using a commercial paper strip (Bililabstix, Bayer Diagnósticos, S.A. de C.V., Mexico City, Mexico), the Benedict test for sugars and other reducing substances, the dinitrophenylhydrazine test for ketoaciduria, the nitrosonaphthol test for tyrosine metabolites, and the cyanide-nitroprusside test for compounds with a disulfide linkage (11). Screening also included blood and urine one-dimensional amino acid thin layer chromatography in butanol-acetic acid:water (12:3:5) (12). Specialized testing. In the case of an abnormal result, further tests were performed, in accordance with the abnormal screening results, for the definitive diagnosis. Among them, plasma amino acids were quantitatively measured by highresolution liquid chromatography (13), and one-dimensional chromatography for urinary sugars (12) if the Benedict test was positive. In acutely sick patients, blood glucose, arterial gases, electrolytes including calculation of the anion gap, ketone bodies, ammonia, and lactic acid were determined by standard methods as well as urinary organic acid analysis by gas-liquid chromatography—mass spectrometry— GCMS (14) introduced into our laboratory over the last 5 years. GCMS was also performed in patients’ urine that suggested the presence of an organic acidemia (ketoacidosis, periodic episodes, neurologic manifestations, and in general, having diagnostic difficulties). When considering specific disorders, the following particular metabolites and/ or enzymes were assayed: succinyl acetone for hepatic tyrosinemia (performed by Claude Laberge, PhD, University of Laval, Quebec, Canada); urinary orotic acid (15) and the allopurinol challenge test (16) when a urea cycle defect is suspected; the Beutler test (17) for classical galactosemia; blood arginase (performed by Stephen S. Cederbaum, PhD, UCLA, Los Angeles, CA, USA) (18); cultured fibroblast mitochondrial carboxylases (19), and serum biotinidase (20). Blood lactate, pyruvate and the L/P ratio, and skin fibroblast pyruvate dehydrogenase (PDH performed by Douglas Kerr, PhD, Case Western Reserve University, Cleveland, OH, USA) were estimated (21) when an energy defect was suspected, or when the patient had a history of ataxia episodes. When considering a fatty acid oxidation disorder, blood acylcarnitines (performed by tandem mass spectrometry by Edwin Naylor, PhD, Neo Gen Screening, Pittsburgh, PA, USA) (22) and urinary organic acids (14) were determined, as well as prolonged fasting or triglyceride loading tests, measuring blood glucose, 3-hydroxybutyrate, acetoacetate and insulin levels, and urinary organic acids. In cases in which a glycogen storage defect is suspected, primary

Velázquez et al. / Archives of Medical Research 31 (2000) 145–150

screening included blood glucose and lactate after a fast that the patient could tolerate and after glucagon administration, as well as a galactose loading test; confirmation of diagnosis was through the assay of liver glycogen-metabolizing and glucogenogenic enzymes (initially performed by Robert E. Grier, PhD, Nemours Children’s Clinic, Jacksonville, FL, USA, and currently by Y.T. Chen, PhD, Duke University, Durham, NC, USA) (23). All previously mentioned studies were performed with the informed consent of the patients or their legal guardians after review and approval of the hospital’s Ethics Committee (Protocol no. 12, 1989, Instituto Nacional de Pediatría). The assessment of the final diagnosis depended on the clinical picture and the results of the previously mentioned laboratory tests.

Results During the first 15 years of the program, 57 patients with an IEM were diagnosed (Table 1). Many of the patients’ files during this period are incomplete and the total number studied is not available. During the next 10 years, we studied 5,186 patients, 45.4% of whom were females. An inherited metabolic disorder was diagnosed in 118 patients, 1 per 43.9 patients studied, 42.4% of whom were girls. The number of patients diagnosed per year has substantially increased, from 3.5 in the previous 15 years, to 11.7 in the past 10, but distribution according to groups of diseases has significantly changed. Amino acid defects were the most prevalent throughout these 25 years. However, the proportion of organic acid and carbohydrate disorders substantially increased over the last 10 years (Table 1). Figure 1 shows the cumulative frequency of new cases diagnosed with an IEM and, in particular, with an organic aciduria, during these 10 years. A substantial rise is observed, especially for the overall number of patients during the past 5 years, and for those with an organic acidemia in the last 3 years. The most frequent diseases diagnosed in the last period were phenylketonuria (PKU) and type 1a glycogen storage disorder—von Gierke’s disease (14 and 16 cases, respectively), followed by maple syrup urine disease (MSUD) (13 cases). Only two of the 13 patients with MSUD had the intermediate variety. The other 11 have the classic neonatal variety, and 8 died within the next few months despite intensive care measures. The increase in the number of some of the diseases over the past 10 years is noteworthy. The diseases whose frequency changed most significantly are MSUD, ornithine transcarbamylase (OTC) deficiency, propionic and methylmalonic acidemias, Canavan disease, classic galactosemia (galactose-1-phosphate uridil transferase deficiency), and glycogen storage diseases. Other notable observations are that most of the patients with PKU and MSUD (or their parents or grandparents) came, respectively, from two distinct geographical regions of the Mexican Republic: the Los Altos region of

147

Table 1. Number of diagnosed patients and their distribution according to group of metabolic disordersa Disorder

1973–1988

1989–1998

Amino acids Phenylketonuria Hyperphenylalaninemias Maple syrup urine disease Tyrosinemia Homocystinuria Histidinemia Nonketotic hyperglycinemia Hypermethioninemia Ornithine transcarbamylase deficiency Citrullinemia Argininemia Gyrate atrophy of choroid and retina Membrane transport systems Cystinuria Hartnup disorder Cystinosis Organic acids Multiple carboxylase deficiency Isovaleric acidemia Propionic acidemia Methylmalonic acidemia Canavan disease 3-Methylcrotonyl-CoA carboxylase deficiency Alkaptonuria Glutaric acidemia type I Pyruvate dehydrogenase deficiency Mitochondrial acetoacetyl CoA thiolase deficiency Carbohydrates Classic galactosemia Von Gierke disease Glycogen debrancher deficiency Branching enzyme deficiency Liver phosphorylase deficiency Fructose 1,6-bisphosphate deficiency Total

35 (61.40%) 21

57 (48.30%) 14 1 13 7 8

3 2 2 1 1 1 1 2 1 4 (7.02%) 1 1 2 7 (12.28%) 1 1 2

3

11 (19.30%) 1 8 1

1 57 (100%)

3 2 7 1 1 1 (0.85%)

1 29 (24.58%) 1 1 6 7 5 1 2 1 3 2 31 (26.27%) 7 16 5 1 2 118 (100%)

a

The total number of patients evaluated in the first 15 years is not available; the total number of patients from 1989 to 1998 is 5,186.

the state of Jalisco, and from the state of Sinaloa, north:northwest of Mexico City. All patients are severely handicapped or dead, despite receiving the specific treatment (manuscript in preparation). The 10 most frequent clinical criteria for referral during the last 5 years, when the cumulative frequency of new cases rose substantially, are shown in Table 2. They comprise 56.2% of all criteria; approximately one-third (34.9%) were neurologic (mental retardation, seizures, hypo- or hypertonia, spasticity, ataxia, severe lethargy, or coma). Other referral criteria were the following: kidney disease; Reye’s syndrome; hyperuricemia; peculiar smell; dyskinesia; myopathy; cardiomyopathy; recurrent infections; stroke; cataracts; corneal clouding; trichorrhexis nodosa, and alopecia. Distribution of patients according to the institution or health

148

Velázquez et al./ Archives of Medical Research 31 (2000) 145–150

Figure 1. Cumulative number of patients with a inborn error of metabolism (䊐) or with an organic aciduria (䊏) diagnosed at the Unit of Genetics in Nutrition from 1989 to 1998.

professionals who referred them for metabolic evaluation has been changing over the years. We present the data corresponding to the last 5 years, which we believe provides a reasonably accurate current picture. The number of patients studied was 3,630, 73.2% of whom were from our hospital (the National Institute of Pediatrics); 96.9% were from metropolitan Mexico City. Table 3 shows patient distribution according to our hospital’s departments that referred them. About one-third (33.8%) came from pediatricians, 31.1% of whom were at critical medicine departments (emergency room, intensive care unit, surgery), and 37.1% were referred by specialists. Geneticists contributed over one-fourth (29.1%) of the referrals. This is in sharp contrast to what occurred during the initial 15 years, when 59% were referred by geneticists in 1989, this proportion probably being higher in the previous years. One-fifth (20.6%) of the patients were referred by neurologists. Discussion We have been studying inborn errors of metabolism in a systematic and organized way for over 25 years, our program being one of the earliest in Latin America or indeed, in developing countries. In the past 10 years, there has been a

substantial increase in the number of patients referred for metabolic evaluation (particularly of acutely sick infants), in the number of diagnoses (from 3.5–11.7 per year), as well as in the number of several individual diseases that were detected. In the past year, we diagnosed a new patient for every 44.3 referrals. This article is, of course, not an epidemiTable 2. The 10 most frequent clinical criteria for referral in the last 5 years

Clinical criteria Psychomotor retardation Seizures (intractable, myoclonic) Hypotonia Failure to thrive Hepatic disease (liver failure, hepatomegaly, cholestatic jaundice) Metabolic acidosis Refusal to feed, episodic vomiting, chronic diarrhea Hyperammonemia Family history of siblings with psychomotor retardation or sudden death syndrome Hypoglycemia

Percent of 3,630 patients referred 439 (12.1%) 323 (8.9%) 370 (10.2%) 269 (7.4%) 319 (8.8%) 243 (6.7%) 83 (2.3%) 36 (1.0%)

33 (0.9%) 54 (1.5%)

Velázquez et al. / Archives of Medical Research 31 (2000) 145–150 Table 3. Distribution of 3,630 patients from the National Institute of Pediatrics referred for metabolic evaluation from 1994 to 1998, according to hospital department Pediatric services Acute (neonatology, intensive care units, emergency, surgery) Chronic (general pediatrics)

10.5% 23.3%

Specialties (neurology, gastroenterology, cardiology, infectology, nephrology, endocrinology, hematology, ophthalmology, pneumology, immunology, dermatology) Genetics (clinical genetics, nutritional genetics)

37.1% 29.1%

ological study, given that the patients studied were mainly from one tertiary-care pediatric hospital. However, it shows that inherited metabolic diseases constitute a significant load in pediatric pathology, even if some groups, such as lysosomal storage and peroxisomal disorders, were not studied in order to more efficiently focus our limited resources. The rising number of specific diagnoses (Table 1) are the result of better knowledge of IEM, particularly by the hospital’s interns and residents, as possible causes of pediatric diseases, as well as of improved infrastructure, confirmatory facilities, and organization of our Unit. For example, the larger number of detected organic acidemias in the last 4 years (Figure 1) was associated with the implementation of organic acid analysis by GCMS. We also have more and better trained personnel; in the past 6 years, we have recruited and trained a pediatrician, a nurse, a biochemist, and a nutritionist, who are committed full-time to the program. That approximately one-half of all diagnoses correspond to amino acid disorders is likely a reflection of the routine metabolic screen, which includes blood and urine amino acid chromatography performed on all patients, reinforced by our new capacity to identify and quantitate amino acids by high-resolution liquid chromatography. The amino acid disorder most frequently diagnosed was PKU; it is noteworthy that most patients (or their ancestors) with this disorder were significantly more likely to have originated from the Mexican state of Jalisco, an observation that has been further evidenced (24). The current governmental neonatal screening program of Mexico is only for congenital hypothyroidism. Therefore, almost all patients with an IEM are detected late in Mexico, when severe and irreversible damage has occurred. Mexican health authorities are now considering including PKU and other diseases, especially since the technology for newborn screening has improved greatly over the last decade and it is now possible to detect over 30 inherited metabolic disorders in time to begin effective management (25,26). Referral criteria for our metabolic unit have greatly diversified (Table 2), including an increasing number of acutely sick infants, reflecting greater interest in this type of disorder by pediatricians and by a wider range of specialists (Table 3). This increased clinical awareness likely explains

149

the increased number of some specific diagnoses, like propionic and methylmalonic acidemias, urea cycle disorders (ornithine transcarbamylase deficiency), classical galactosemia, glycogen storage disorders, and maple syrup urine disease (MSUD), the latter two having been the most frequently diagnosed conditions together with PKU. In regard to PKU, it is interesting that most of our patients have been from the northwest Mexican state of Sinaloa, a result that is likely a founder effect. MSUD has been found more frequently in Hispanic populations. It is interesting that an even higher frequency has been found in Saudi Arabia (22). Because Spain was an Islamic country for 800 years, it is tempting to speculate whether some of the MSUD mutations in the Spanish-speaking populations are of Near Eastern origin. Our Unit is the national reference center for IEM, by mandate of the Mexican Health Department. However, 97.4% of the referred patients were from metropolitan Mexico City and only 2.6% from the rest of Mexico, a result of the size of the country, of the surge of new facilities for the study of these disorders in other Mexican cities (in great part due to our training activities; see later), and of local resistance toward what is perceived of as centralism. However, 21.2% of the patients referred to our Unit have been from hospitals other than ours in metropolitan Mexico City, where nearly one-fifth of the Mexican population lives. We have also devoted part of our efforts to the training of specialists and technicians from Mexico and other Latin American countries, resulting in the establishment of new clinical and laboratory groups interested in these diseases in different parts of the country (27). In order to share our experiences, we developed the Latin American Metabolic Information Network, a website for communication of specialists in our geographical region (http://www.unam.mx/ redlaem). We still face serious problems, principally related to financial support and scarcity of products for the management of this group of diseases. To contend at least partially with these problems, we have been working on the development of a procedure to prepare locally a special food product low in phenylalanine for treatment of PKU patients (28). We have also extended the approaches used for IEM, in particular the non-invasive analysis of urinary organic acids by GCMS, to the investigation of metabolic derangements in nutritional disorders, especially severe infantile malnutrition, which unfortunately is still an important public health problem in some areas of Mexico. Our results have shown that most undernourished infants transitorily exhibit organic acidurias similar to the genetic ones (29). Having reached this state of affairs has been a slow and difficult process that involved indifference, resistance, and at times frank hostility, conditions shared by other developing countries (30). However, as shown by these results our pioneer activities for the study of inborn errors of metabolism have developed into a full-grown and successful program.

150

Velázquez et al./ Archives of Medical Research 31 (2000) 145–150

Acknowledgments We acknowledge the competent work of Catalina Martín del Campo, Elba Márquez, and Armida Báez. Despite the lack of governmental or institutional financial support for the treatment of patients diagnosed with an IEM, aid has been possible thanks to the generous help from Ross Laboratories, (Division of Abbott Laboratories, Columbus, OH, USA) and, in particular, Drs. Phylis Acosta and Steve Yannicelli. Dr. Silvestre Frenk carefully read this manuscript and contributed many valuable comments and suggestions. We wish to highlight the extraordinary and friendly support, encouragement, and indeed tutoring of Dr. Stephen Cederbaum throughout these 25 years, without which little of what is reported here would have been achieved.

References 1. Velázquez A. Neonatal screening in countries with socioeconimic development problems: results of an international inquiry. In: Farriaux J-P, Dhondt J-L editors. New horizons in neonatal screening. New York: Elsevier;1994. p. 301. 2. Velázquez A, Villarreal ML. El diagnóstico de errores congénitos del metabolismo. Neurolog Neurocir (Mex) 1973;14:7. 3. Villareal ML, Galindo LM, Velázquez A. Newborn genetic screening: the Mexican program. In: Lisker R, Armendares S, editors. Human genetics. Amsterdam: Excerpta Medica;1977. p. 214. 4. Velázquez A, Loera A, Aguirre BE, et al. Resultados del programa mexicano de tamiz neonatal para hipotiroidismo congénito y fenilcetonuria. Salud Publica Mex 1994;36:249. 5. Vela M, Gamboa S, Loera-Luna A, Aguirre BE, Pérez-Palacios G, Velázquez A. Neonatal screening for congenital hypothyroidism in Mexico: experience, obstacle, and strategies. J Med Screen 1999;6:77. 6. Zetina ME. Enfermedades hereditarias lisosomales. I. Resultados iniciales del programa para su diagnóstico en México. Rev Inv Clin 1989;41:319. 7. Vaca G, Velázquez AL, Cantú JM. Hereditary erythroenzymopathies. I. Biochemical and genetic aspects. Bol Oficina Sanit Panam 1984;97:225. 8. Sánchez-Anzaldo FJ, Ruíz-Argüelles GJ, Ortíz-López R. Erythropoietic protoporphyria. Description of the first case identified in Mexico. Sangre (Barc) 1987;32:236. 9. Tusié-Luna MT, Ramírez-Jiménez S, Ordóñez-Sánchez ML, et al. Low frequency of deletion alleles in patients with steroid 21-hydroxylase deficiency in a Mexican population. Hum Genet 1996;98:36. 10. Vázquez-Acevedo M, Coria R, González-Astiazaran A, MedinaCrespo V, Ridaura-Sánz C, González-Halphen D. Characterization of a 5025 base pair mitochondrial DNA deletion in Kearns-Sayre syndrome. Biochim Biophys Acta 1995;1271:363.

11. Thomas GH, Howell RR. Selected screening test for genetic metabolic diseases. Chicago: Year Book Medical Publishers;1973. 12. Shih VE. Laboratory techniques for the detection of hereditary metabolic disorders. Cleveland, OH, USA: CRC Press;1973. 13. Hill D, Burnoworth L, Snea W, Pfeifer R. Quantitative HPLC analysis of plasma amino acids as orthophthaldialdehyde/ethanethiol derivatives. J Liq Chromatogr 1982;5:2369. 14. Sweetman L, Hoffmann G, Aramaki S. New diagnostic techniques for the detection of organic acidemias. Enzyme 1987;38:124. 15. Harris ML, Oberholzer VG. Conditions affecting the colorimetry of orotic acid and orotidine in urine. Clin Chem 19800;26:473. 16. Burlina AB, Ferrari V, Dionisi-Vici C, Bordugo A, Zacchello F, Tuchman M. Allopurinol challenge test in children. J Inherit Metab Dis 1992;15:707. 17. Beutler E, Baluda MC. A simple spot test for galactosemia. J Lab Clin Med 1966;68:137. 18. Buga GM, Singh R, Pervin S, et al. Arginase activity in endothelial cells: inhibition by NG-hydroxy-L-arginine during high-output NO production. Am Physiol Soc 1996;40:H1988. 19. Velázquez A, Terán M, Báez A, Gutiérrez J, Rodríguez R. Biotin suplementation affects lymphocyte carboxylases and plasma biotin in severe protein-energy malnutrition. Am J Clin Nutr 1995;61:385 20. Wolf B, Grier R, Parker W. Biotinidase deficiency: the defect in lateonset multiple carboxylase deficiency. Clin Chim Acta 1983;13:273. 21. Robinso BH, MacMillan H, Petrova-Benedict R, Sherwood WG. Variable clinical presentation in patients with deficiency of the pyruvate dehydrogenase complex. A review of 30 cases with a defect in the E1 component of the complex. J Pediatr 1987;573:337. 22. Millington DS, Kodo N, Norwood DL, Roe C. Tandem mass spectrometry: a new method for acylcarnitine profiling with potential for neonatal screening for inborn errors of metabolism. J Inherit Metab Dis 1991;13:321. 23. Shin YS. Diagnosis of glycogen storage disease. J Inherit Metab Dis 1990;13:419. 24. Velázquez A, Bilbao G, González-Trujillo JL, et al. Apparent higher frequency of phenylketonuria in the Mexican State of Jalisco. Hum Genet 1996;97:99. 25. Chace DH, Hillman SL, Millington DS, Kahler SG, Roe CR, Naylor EW. Rapid diagnosis of maple syrup urine disease in blood spots from newborns by tandem mass spectrometry. Clin Chem 1995;41:62. 26. Levy HL. Newborn screening by tandem mass spectrometry: a new era. Clin Chem 1998;44:2401. 27. Vaca G, Hernández A, Ibarra B, et al. Detección de errores congénitos del metabolismo en 1117 pacientes estudiados por sosopecha de enfermedad hereditaria. Arch Invest Med (Mex) 1981;12:341. 28. López-Bajonero LJ, Lara-Calderón P, Gálvez-Mariscal A, VelázquezArellano A, López-Munguía A. Enzymatic production of low-phenylalanine product from skim milk powder and caseinate. J Food Sci 1991;56:938. 29. Terán-García M, Ibarra I, Velázquez A. Urinary organic acids in infant malnutrition. Pediatr Res 1998;44(3):386. 30. Coelho JC, Wajner M, Burin MG, Vargas CR, Giugliani R. Selective screening of 10,000 high-risk Brazilian patients for the detection of inborn errors of metabolism. Eur J Pediatr 1997;156:650.