Autistic spectrum disorders: When traditional medicine is not enough

Autistic spectrum disorders: When traditional medicine is not enough By Susan L. Hyman, MD, and Susan E. Levy, MD Vitamins and supplements, special diets, and other nontraditional treatments for ASD are becoming increasingly popular. To advise families, pediatricians need to be knowledgeable about the efficacy and side effects of these therapies. Autistic spectrum disorders (ASD) are thought to occur in at least one in every 1,000 children. Thus, each pediatrician typically will follow one affected child, if not more. Practitioners and the families of these patients must confront the classic dilemmas of ASD: What is the cause of the disorder? Is there a cure? What are the most effective and safe treatments? ASD traditionally are managed with intense, early, and generalized educational, language, and behavioral interventions. Several disorder-specific strategies are effective in enhancing skill development and diminishing undesired behaviors. Although improvements may be dramatic, complete resolution of symptoms is rare. Practitioners may also use medications to manage hyperactivity, perseverative behaviors, obsessions, or anxiety, but these treatments should not replace the educational and behavioral interventions. Complementary and alternative therapies offer other options. To best guide the families of children with ASD, pediatricians should be well-informed about the efficacy and side effects of these various treatments and know what therapies are available in the community. What we know about ASD The Diagnostic and Statistical Manual IV categorizes ASD as pervasive developmental disorders.1 Children demonstrate a range of deficits in three core areas: social interaction, communication, and patterns of behavior, interests, and activities, which tend to be repetitive and stereotyped. Symptoms usually are evident by the time the child is 3 years old. Children may have various combinations of symptoms and a wide range of cognitive skills. ASD are chronic and poorly understood. Research points to a strong genetic predisposition and a prenatal origin.2 Although no single brain anomaly has been associated with ASD, several different neuropathologic abnormalities of prenatal origin have been reported.2 Animal data and observation of medication responses have implicated neurochemical abnormalities and provided the basis for pharmacotherapeutic trials.3 Many untested theories have emerged that purport to explain the core symptoms as well as associated medical conditions. These theories have spawned a wealth of treatments based on complementary and alternative medicine (CAM). The classic approach to assessing the efficacy of a particular therapy is to apply specific criteria to research reports (Table 1),4 as has been done in studies of many treatments for ASD that are discussed below. It is useful to compare a treatment with placebo since time and the perception of treatment efficacy may affect reported outcome. Multiple simultaneous interventions are likely to confound the impact of additional treatments. Parents interested in a CAM therapy can obtain important information from other parents, the media, and the Internet. The scientific validity of such information may be questionable, but families searching for treatment options often do not discriminate between anecdotes or personal experience and scientific evidence. In trying one of these therapies, the physician and parent need to establish and maintain a dialogue so that the practitioner can evaluate treatment efficacy and monitor potential side effects (Table 2). Pediatricians can learn more about ASD and treatment options from the sources in Table 3. Table 1 Evaluating studies of psychopharmacologic therapies Table 2 Guidelines for using CAM Evaluate treatment by selecting a response that can be measured Identify target behaviors for treatment Establish a baseline of skill or function Monitor response to therapy prospectively Consider each child as his or her own control Use valid and objective outcome measures when possible Use raters who could be blind to treatment, such as a teacher Keep lines of communication between families and physician open, even if both parties do not agree with the course of treatment Families should keep the doctor informed of all treatments so the patient can be monitored for potential side effects. If the physician is not comfortable prescribing the requested treatment, the family should find out why Physicians should avoid being judgmental Table 3 Learning more about ASD: Resources for pediatricians General information Autism Society of America (www.autism-society.org) Filipek PA, Accardo PJ, Baranek GT, et al: The Screening and diagnosis of autism spectrum disorders. J Autism Dev Disord 1999;29:439 Campbell N, Schopler E, Cueva JE, et al: Treatment of autistic disorder. J Am Acad Child Adolesc Psychiatry 1996;35:134 CAM and ASD Biological Treatments for Autism and PDD, by William Shaw, PhD, Sunflower Publications DAN! (Defeat Autism Now) Guidelines, by Autism Research Institute (order from institute at 4182 Adams Ave., San Diego, CA 92116 or 619-563-6840 [fax]) Unraveling the Mystery of Autism and Pervasive Developmental Disorder: A Mothers Story of Research and Recovery, by Karyn Serroussi and Bernard Rimland, PhD, Simon and Schuster The Yeast Connection: A Medical Breakthrough, by William Crook, MD, Vintage Books Web sites for CAM information Autism Network for Dietary Intervention (www.AutismNDI.com) Autism Research Institute, including DAN! (www.autism.com/ari) Autism Society of America (www.autism-society.org) Cure Autism Now Foundation (www.canfoundation.org) Developmental Delay Resources (www.devdelay.org/issues.html#aut) Forum on Alternative and Innovative Therapies for Children with Developmental Disorders and Brain Injury (www.healingarts.org/children) Gluten-free, casein-free diet (www.gfcfdiet.com) Web sites for other treatment options American Academy of Pediatricsalternative medicine link (www.aap.org/bpi/Alternative.html) Autism Resources (www.autism-resources.com) Health Education Services of the New York State Department of Healthfor state guidelines (www.hes.org) National Center for Complementary and Alternative Medicine (http://nccam.nih.gov/nccam) What's hot and what's not The popularity of specific CAM therapies changes rapidly. In addition, what was once CAM sometimes becomes conventional therapy. The technique of reinforcing desired behaviors and ignoring undesired behaviors, for example, is now basic to educational interventions for children with ASD. These techniques of applied behavioral analysis have been used to enhance the language and cognitive development of very young children with autism. In a controversial study by Lovaas and colleagues, 19 children under 3 years of age with autism participated in an individual, home-based program for 40 hours a week. At follow-up at 7 and 11 years of age, nine of the 19 children were in regular education classes and had typical cognitive skills for their age.5 Although other experts in the education of children with autism have expressed concern that subjects were not randomly assigned in this study, additional studies present strong evidence that a behavioral approach effectively teaches young children with autism. Indeed, the Guideline Technical Report: Autism/Pervasive Developmental Disorders, an evidence-based literature review, advocates behaviorally based interventions.6 Costs of such one-on-one interventions can range from $20,000 to $70,000 a year. CAM treatments that currently are popular can be categorized in several groups: vitamins and supplements, dietary manipulations, alternative uses of biologic agents, immune therapy, and nonpharmacologic therapies. Vitamins and supplements Vitamins and supplements are the most commonly used CAM for treatment of ASD. Advocates of this type of therapy suggest that vitamins enhance neurotransmitter actions by increasing availability of substrate or cofactors. Vitamin C (ascorbic acid). One study examined vitamin C because of scientific evidence that it inhibits central dopamine action, which may be abnormally increased in ASD.7 Although the study used only a small sample, the data it generated, using an objective outcome measure, suggest that high-dose vitamin C therapy may diminish stereotypic behaviors. Side effects are unusual, with renal stones possible at very high doses. This potentially promising intervention needs replication with a larger sample. Vitamin B6 and magnesium. Some observers have advocated supplementation with B6 (pyridoxine) because of its role in the generation of several neurotransmitters (serotonin, dopamine, gamma amino butyric acid, norepinephrine, and epinephrine). Advocates suggest administration of B6 in combination with magnesium to increase clinical response. Results of efficacy studies have varied. A double-blind, placebo-controlled study using standard psychiatric outcome measures did not identify a measurable change in behavior between groups.8 A review of 12 studies noted that about half the subjects improved within a few weeks, according to a variety of behavioral indices. Most of those whose behavior improved relapsed within one to six weeks after discontinuing therapy.9 In addition, the review found that most of these studies did not use control groups, report objective outcome data, or randomize subjects. An evidence-based review yielded insufficient objective evidence to recommend treatment with B6/magnesium.6 High-dose pyridoxine treatment (200 to 500 mg/d) may cause peripheral sensory neuropathy, indicated by a decline in fine motor skills or an increase in hand shaking or other repetitive movement. Proponents of B6 supplementation note that side effects have not been reported in children with ASD. Long-term studies are lacking. B6 is often administered in a proprietary megavitamin mixture. Families may not realize that different supplements may contain the same ingredients or that vitamins may be toxic at high doses. The potential toxicity of magnesium should also be noted. A fatal overdose of magnesium was reported in a child with cerebral palsy and seizures.10 Families should be counseled not to exceed recommended doses and to be sensitive to potential side effects. A proprietary mixture of magavitamins can cost as little as $10 a month Dimethylglycine (DMG) is a nutritional supplement that may have excitatory central neuroactive effects similar to those of the excitatory neurotransmitter, glycine.11 In a double-blind, placebo-controlled crossover study, one family reported that their child was irritable.11 No overall improvement was noted during treatment. The authors acknowledge that their sample was small, the dose was low, and the outcome measures may not have been sensitive to change. In addition, investigators did not use standardized outcome measures or control for confounding factors. Other than irritability and increased activity, no side effects are reported for this compound. Vitamin A. It has been hypothesized that administration of high-dose vitamin A as cod liver oil might improve immune function. This treatment is reportedly under study, but no data are available at present. Hypervitaminosis A may cause anorexia, dry skin, liver toxicity and hepatosplenomegaly, increased intracranial pressure, irritability, and carotenemia. Birth defects may occur with excessive maternal intake of vitamin A during pregnancy. Table 4 summarizes existing studies of vitamin and supplement therapies in ASD. Table 4 Studies of vitamin and supplement therapies in ASD Are you what you eat? The role of restrictive diets in managing ASD is unclear. Sugar, preservatives, and aspartame. Evidence is lacking that dietary exposure to sugar, aspartame, or preservatives plays a part in the etiology of hyperactivity.15 The role of food coloring is also unclear. No studies have addressed how or if these dietary factors contribute to the hyperactivity that can accompany ASD. Milk and wheat. Families that wish to treat their children's ASD with dietary manipulation usually have in mind a casein-free and gluten-free regimen. This approach rests on the hypothesis that children with autism have a "leaky gut." This is thought to lead to increased absorption of peptides with the bioactive properties of endogenous opioids. Advocates of this theory suggest that this phenomenon may be caused by a variant of celiac disease, a side effect of yeast overgrowth, a primary immunologic abnormality, or enterocolitis as a sequela of immunization. Numerous anecdotal reports on the Internet and in the print media suggest the efficacy of diet therapy for ASD. Supporters of the diet approach note that about one third of children with autism lose language and social milestones in the second year of life, a period that often coincides with the introduction of cow's milk and solid foods. These events may be coincidental rather than causal, however. Elimination of milk or wheat may result in a decrease in loose stools. This change might reduce irritability but be unrelated to amelioration of autistic symptoms. Until appropriate studies are performed, observations that dietary limitations improve autism remain speculative. Some families of children with autism report gastrointestinal symptoms that include, in addition to loose stools, food selectivity and behavioral intolerance of a variety of foods. Several investigators have proposed that increased levels of urinary peptides in children with autism are related to opioid-like compounds found in dairy and gluten products. These peptides reportedly diminish with dietary intervention.16,17 Other investigators, using a different assay, found no elevation of urinary peptides, however.18 This set of observations requires further study. Interpretation of the data in studies of this nature depends on the criteria used to select subjects and on controlling for other supplements, diet, and medications study subjects have taken. Outcome measures for clinical studies must be carefully selected. A clinical case series in Norway compared children with autism and other psychiatric diagnoses treated with diet with similar children who could not maintain the diet. They used subjective behavioral ratings scales and objective measures to determine each child's response to diet during a one-year period. 17 Confounding variables included a heterogeneous and poorly defined set of subjects, the effect over time of maturation and educational intervention, and concurrent drug treatment. The authors report a positive effect of dietary limitation, however. Very few children with ASD have been identified with antiendomysial or antireticular antibodies or biopsy-proven celiac disease. Since the clinical significance of abnormalities in the urinary peptides attributable to casein and gluten metabolites is unknown, routine urinary assay is not recommended. A subset of children with autism and diarrhea were found to have ileonodular hyperplasia on colonoscopy.19 How this relates to the hypotheses underlying the casein-free/ gluten-free diet is unclear. The scientific foundation for a milk- and wheat-free diet as a treatment for ASD currently is weak. Children treated with such a diet require medical surveillance. Calcium and protein intake and iron status need to be monitored. The child on a restricted diet may be at increased risk for rickets, and supplemental calcium and vitamin D are indicated. A licensed pediatric nutritionist can help the family meet the child's nutritional needs while restricting exposure to casein or gluten. Casein/gluten-free foods are available at health food stores or can be purchased by mail. They cost more than foods in the typical diet. Although performing studies related to diet and behavior is difficult, careful investigation of this area is needed. Alternative use of biologic agents Biologic agents used to treat ASD include medications to treat purported yeast overgrowth, secretin, and famotidine. Antiyeast therapy is based on the supposition that children with ASD have an intestinal overgrowth of yeast because of an immunodeficiency or a history of excessive antibiotic use. Advocates of this therapy propose that these risk factors may lead to the "leaky gut" syndrome described above or that the yeast itself may somehow be a toxin. This notion is based on a case report describing two boys with abnormal urine organic acid findings who had loss of language, autistic features, weakness, and ketosis.20 The authors of this report concluded that the urinary compounds must be fungal in origin. Treatment of the purported yeast overgrowth with diet supplements and medications may lead to a variety of side effects, depending on the therapy (Table 5). Some observers interpret diarrhea following administration of an antifungal medication as a "die off" reaction of the yeast. There are no peer-reviewed studies to guide treatment. Table 5 Common antiyeast therapies Carbohydrate intake reduced Sugar intake lowered Fruit consumption restricted Nutritional deficiencies, which can be avoided with appropriate dietary choices Nystatin Fluconazole Terbinafine Anemia, diarrhea Liver function abnormalities Rash, GI distress/diarrhea Proponents of antiyeast therapy advocate measuring and monitoring urine organic acids as well as performing bacteriologic culture of stool samples to document eradication of the purported candidal overgrowth. The clinical significance of the levels of these compounds, with or without antifungal treatment, is not clear because of insufficient data, however. Urine organic acid measurement is appropriate for the initial etiologic evaluation of a child with loss of milestones or developmental disability and behavioral variability. Candidal overgrowth or pathologic intestinal bacteria were not identified in a series of children with ASD who underwent endoscopy, including 12 patients in whom urine organic acid testing suggested fungal overgrowth.21 Secretin, a pancreatic hormone that mediates digestion, recently was proposed as a possible "cure" for autism. A report on national television in October 1998 described a 3-year-old child whose gastrointestinal symptoms, language, and behavior improved after receiving porcine secretin during a diagnostic gastrointestinal endoscopy. Subsequent media coverage triggered a frenzy of requests for the untested treatment. At that time, the only published information was a case report of three children.22 The brain does contain secretin receptors. High-dose infusions of this hormone may have a central action. Rigorous studies that followed the initial flurry of enthusiasm have not supported the initial claims of dramatic improvement, however. Two double-blind, placebo- controlled studies of secretin did not document a therapeutic effect. Other studies are in progress.23,24 Thousands of children with ASD have received intravenous, endoscopic, or transdermal infusions of secretin outside of clinical trials. Outcome measures usually are parental estimates of change as reflected in nonstandard questionnaires. Generalizations about efficacy may be difficult to interpret because of a lack of standard dosing, subject heterogeneity, and the potential for placebo effect in the absence of controls or double-blind assessment. Safety issues include concern about the immunologic response elicited by multiple administrations of a foreign protein. Trials of human synthetic secretin, which could be less immunologically challenging than the porcine product, are being conducted. It is not known whether sublingual or transcutaneous secretin is systemically absorbed or reaches concentrations that could be clinically significant. Because of the popular demand for secretin therapy, this has emerged as one of the most carefully examined pharmacologic interventions for children with ASD. Although 75 international units of secretin retail for about $240, many families pay several times that amount for IV infusions. Famotidine (Pepcid), a histamine-2 receptor antagonist in wide clinical use for symptoms of gastroesophageal reflux and heartburn, has been investigated as a potential treatment for some of the symptoms of schizophrenia. Because H2 receptors in the brain are related to exploratory behavior and activity in animals, it has been hypothesized that famotidine might ameliorate some of the symptoms of autism.25 Treatment of gastrointestinal symptoms in children with 1 mg/kg/d has been well tolerated. The more common side effects of this over-the-counter drug include headache, dizziness, constipation, and diarrhea. We could locate no peer-reviewed reports on the efficacy of famotidine in ASD. Alkaline salts. Antacids and alkaline salts increasingly are being used to alter stomach pH to stimulate natural secretion of secretin and other gastrointestinal peptides. The medical literature has no data on the safety or efficacy of this intervention in children with ASD. Immunologic therapies Although children with ASD are generally not considered to have clinical immune dysfunction, investigators have measured many different immune parameters for clues that might point to an infectious or autoimmune etiology. Reported abnormalities have included low immunoglobulin A or immunoglobulin G, abnormal levels of T and B lymphocytes, decreased numbers of natural killer cells, and antibodies to myelin basic protein. Based on these concerns, observers have proposed therapy with intravenous immunoglobulin (IVIG) and three case series have been published with varying results (Table 6).26-28 At this time IVIG is considered an investigational therapy for autism. Serious potential side effects include renal failure, rare infection, and acute neurologic complications. Table 6 IVIG:What the studies show Standardized objective measure Ritvo-Freeman Real-Life Rating Scale Yale-Brown Compulsive Disorder Scale Clinical Global Impression Scale for Autistic Disorders Autism Scale 10/10 improved Descriptive statistics used Nonbiologic therapies Auditory integration therapy is one of several nonbiologic therapies that have been applied to ASD. Hyperacusis, or undue sensitivity to sound, is a common symptom of ASD. The goal of auditory integration therapy is to reduce sensitivity to sound by systematic exposure to altered music via earphones. Proponents of this therapy report generalized behavioral improvement, though the physiologic mechanism is unclear. A well-designed study using a sham audiologic procedure in the controls could not demonstrate an effect, either in the short term or in the year after treatment.29 A 1998 American Academy of Pediatrics policy statement concurs with New York State Clinical Practice Guidelines that the literature currently contains no scientifically credible evidence for recommending this treatment.6,30 In some areas, auditory integration therapy is provided at public expense. Administered privately, costs range upward from $1,500 for a standard two-week course of therapy. Some interventions have been abandoned as convincing data are collected. Facilitated communication, introduced in 1989 to assist communications in individuals who lacked verbal skills, is one such intervention. Unlike augmentative communication strategies where the patient has independent access to an appropriate device or medium, facilitated communication requires another person, a "facilitator," to guide the patient's hand to a keyboard to spell out responses to questions. Studies document that facilitators usually influence what is typed, however.6,28 Keeping an open mind Families that take an active role in treating their child's ASD often turn to CAM or novel uses of pharmacologic agents. To be evaluated critically, research studies require adequate sample size, appropriately homogeneous and diagnosed populations, standardized dosages, controls for confounding factors, valid outcome measures, and placebos. These criteria apply to assessments of both conventional medical interventions and CAM treatments. It may be difficult to generalize the studies and case series we review here to clinical practice. Nutritional products are not regulated by the Food and Drug Administration for purity or concentration of active ingredients so potency may vary significantly from bottle to bottle. Nonetheless, subjective reports of parents who note improvement with various interventions cannot be ignored. The intervention itself may work, or the reported improvement may be due to the placebo effect, confounding or coincidental events, maturation, or other factors. CAM therapies will continue to proliferate until the cause and specific treatments for ASD are determined. In the meantime, pediatricians should continue to work closely with families, keep an open mind about new treatments, and review them for safety and efficacy. REFERENCES 1. Diagnostic and Statistical Manual, ed 4. Washington, DC, American Psychiatric Association, 1994 2. Piven, J: The biological basis of autism. Current Opin in Neurobiol 1997;7:708 3. Tsai LY: Psychopharmacology in autism. Psychosom Med 1999;61:651 4. Sprague RL, Werry JS: Methodology of psychopharmacological studies with the retarded. International Review of Research in Mental Retardation 1971;5:148 5. McEachin JJ, Smith T, Lovaas OI: Long term outcome for children with autism who received early intensive behavioral treatment. American Journal of Mental Retardation 1993;97:359 6. New York State Department of Health Early Intervention Program: Clinical Practice Guideline; The Guideline Technical Report; Autism/Pervasive Developmental Disorders. Assessment and Intervention for Young Children (Age 0-3 Years). Albany, NY, New York State Department of Health, 1999 7. Dolske MC, Spollen J, McKay S, et al: A preliminary trial of ascorbic acid as supplemental therapy for autism. Prog Neuropsychopharmacol Biol Psychiatry 1993;17:765 8. Findling RL, Maxwell K, Scotese-Wojtila L, et al: High-dose pyridoxine and magnesium administration in children with autistic disorder: An absence of salutary effects in a double-blind, placebo controlled study. J Autism Dev Disord 1997;27:467 9. Pfeiffer SI, Norton J, Nelson L, et al: Efficacy of vitamin B6 and magnesium in the treatment of autism: A methodology review and summary of outcomes. J Autism Dev Disord 1995;25:481 10. McGuire JK, Kulkarni MS, Baden JP: Fatal hypermagnesemia in a child treated with megavitamin/megamineral therapy. Pediatrics 2000;105: e18 11. Bolman WM, Richmond JA: A double-blind, placebo-controlled crossover pilot trial of low dose dimethylglycine in patients with autistic disorder. J Autism Dev Disord 1999;29:191 12. Tolbert L, Haigler T, Waits MM, et al: Brief report: Lack of response in an autistic population to a low-dose clinical trial of pyridoxine plus magnesium. J Autism Dev Disord 1993;23;193 13. Martineau J, Barthelemy C, Cheliakine C, et al: Brief report: An open middle-term study of combined vitamin B6-magnesium in a subgroup of autistic children selected on their sensitivity to treatment. J Autism Dev Disord 1988;18:435 14. Martineau J, Barthelemy C, Roux S, et al: Electrophysiological effects of fenfluramine or combined vitamin B6 and magnesium on children with autistic behavior. Dev Med Child Neurol 1989;31:721 15. Wolraich, ML: Diet and behavior: What the research shows. Contemporary Pediatrics 1996;13(12):29 16. Shattock P, Kennedy A, Rowell, F, et al: Role of neuropeptides in autism and their relationships with classical neurotransmitters. Brain Dysfunction 1990;3:328 17. Reichelt KL, Knivsberg AM, Lind G, et al: Probable etiology and possible treatment of childhood autism. Brain Dysfunction 1991;4:308 18. Williams, KM, Marshall T: Urinary proteins in autism as revealed by high resolution two-dimensional electrophoresis. Biochem Soc Trans 1992;20:189S 19. Wakefield AJ, Murch SH, Anthony A, et al: Ileal-lymphoid-nodular hyperplasia, non-specific colitis, and pervasive developmental disorder in children. Lancet 1998;351:637 20. Shaw W, Kassen E, Chaves E: Increased urinary excretion of analogs of Krebs cycle metabolites and arabinose in two brothers with autistic features. Clin Chem 1995;41(8):1094 21. Horvath K, Papadimitirou JC, Rabsztyn A, et al: Gastrointestinal abnormalities in children with autistic disorder. J Pediatr 1999;135(5):559 22. Horvath K, Stefanatos G, Sokolski KN, et al: Improved social and language skills after secretin administration in patients with autistic spectrum disorders. J Assoc Acad Minor Phys 1998;9:9 23. Sandler AD, Sutton KA, DeWeese J, et al: Lack of benefit of a single dose of synthetic human secretin in the treatment of autism and pervasive developmental disorder. N Engl J Med 1999;341:1801 24. Owley T, Steele E, Corsello C, et al: A double-blind, placebo-controlled trial of secretin for the treatment of autistic disorder. Medscape General Medicine, 10/6/99, Medscape Inc, www.medscape.com 25. Linday LA: Oral famotidine: A potential treatment for children with autism. Med Hypotheses 1997;48:381 26. Gupta S, Aggarwal S, Heads C: Brief reports: Dysregulated immune system in children with autism: Beneficial effects of intravenous immune globulin on autistic characteristics. J Autism Dev Disord 1996; 26:439 27. Pliopys AV: Intravenous immunoglobulin treatment of children with autism. J Child Neurol 1998;13:79 28. DelGiudice-Asch G, Simon L, Schmeidler J, et al: Brief report: A pilot open clinical trial of intravenous immunoglobulin in childhood autism. J Autism Dev Disord 1999;29:157 29. American Academy of Pediatrics: Position Paper: Auditory integration training and facilitated communication for autism, RE9752. Pediatrics 1998;102:431 30. Bettison S: The long-term effects of auditory training on children with autism. J Autism Dev Disord 1996;26:361 DR. HYMAN is Assistant Professor of Pediatrics, University of Rochester School of Medicine, Rochester, NY. DR. LEVY is Associate Professor of Pediatrics, Children's Seashore House of the Children's Hospital of Philadelphia, University of Pennsylvania, Philadelphia, PA. Susan Hyman, Susan Levy. Autistic spectrum disorders: When traditional medicine is not enough. Contemporary Pediatrics 2000;10:101.