Autistic spectrum disorders (ASD) are a heterogeneous group of disorders that range in severity and have in common underlying deficits in communication, social interaction, and behavior.1 The spectrum includes autism, childhood disintegrative disorder, Rett's disorder, Asperger syndrome, and pervasive developmental disorder - not otherwise specified (PDD-NOS), the latter considered to be the milder range. Prevalence estimates show 34 of 10,000 children are affected by autism2 and 62 of 10,000 children by autism spectrum disorders.3
Because of these rates, pediatricians are confronted with the task of caring for children with autistic spectrum disorders more frequently. Conventional treatment focuses on intensive educational, developmental, and behavioral therapies. For many children, progress is slow. While there are many strategies to enhance communication, promote developmental progress, and manage behavioral difficulties, none are curative.4
Suggested Guidelines for Treatment of ASD
In most children with ASD, the spec-ific etiology is not known, but a strong genetic predis-position has been demonstrated.1 Many novel theories of causation have been suggested, resulting in treatment suggestions that may be considered outside conventional educational or medical strategies. Some of these complementary and alternative medical (CAM) treatments are widely publicized in the lay press and implemented by families before scientific validation of safety and efficacy. These therapies may be chosen by families in addition to the standard therapies or as alternative treatments.5 They may include treatments such as those found in health food stores or standard medications used in an offlabel manner.
This article focuses on the rationale for use of CAM, the theoretical basis of different treatments, available scientific evidence that supports or contraindicates their use, and recommend-ations for clinicians, researchers and families. The scientific investigation of secretin, a gastrointestinal hormone proposed as a specific treatment for ASD, is used to demonstrate factors accounting for change in children with ASD and discusses placebo effect.
EVIDENCE-BASED PRINCIPLES FOR INVESTIGATION
All treatments of children must be judged based on standards of scientific research. Suggested guidelines are listed in Table I.6 There is a hierarchy of evidence supporting treatment provided by different study designs. At the peak are randomized, controlled, clinical trials, which provide the best opportunity to test for confirmation that the outcome is truly related to the treatment.4 Unfortunately, even some commonly used medical treatments have not met these standards.
Many CAM interventions are supported by case reports of improvement or cure. Such anecdotes, however, provide weak evidence of causality.1 Given the heterogeneity of autism and its variable course, studies examining treatment in autism must carefully control for many factors. The effects of other treatments, such as intensive educational, developmental, and behavioral therapies, as well as the process of neurologic maturation, must be considered. The proposed treatment mechanism should be biologically feasible.
SECRETIN: THEORY AND REALITY
Secretin is a good example of how a nonstandard treatment became widely used before it received adequate scientific scrutiny. Secretin is a vasoactive gut peptide used by gastroenterologists in the course of endoscopy to examine pancreatic secretion. Horvath et al.8 reported a case series of three children with autism who underwent diagnostic endoscopy and intravenous secretin infusion. Print and television media provided wide coverage of the purported resolution of autistic symptoms in these children. Subsequently, thousands of children with autism received intravenous secretin treatments, resulting in a secretin shortage. Subsequent studies, however, have failed to confirm the described treatment effect.
In the 5 years since this flurry of public excitement, more than 500 children have been subjects of doubleblind, placebo-controlled, single- or multiple-dose trials of secretin. Close to a dozen studies have been published in peer-reviewed journals, with great attention paid to study design, including factors such as standardized methods of diagnosis, well-established outcome measures (eg, symptoms of autism related to language, attention, maladaptive behaviors, and physiologic responses), prospective randomized trials with placebo controls, and adequate numbers of patients for statistical analysis.9,10 To date, there have been no reports of group or individual effects to support the efficacy of secretin in treatment of symptoms of autism. Phase 3 drug trials are under way.
Several reports have described the effect of the placebo response among children with autism who are treated with secretin. Sandler and Bodfish" observed that families chose to continue secretin even after they were notified that their child did not have a treatment response in the blinded study. The authors noted that the effect of attention on reinforcing communication and positive behaviors after an anticipated intervention such as secretin infusion may have an independent effect on children's behavior. They also discuss the power of positive language and the potential utility of placebo in modern practice.
In another report, families participating in a double-blind, placebo, crossover trial of human synthetic secretin were unable to accurately guess the sequence of treatment,12 despite significant changes in function in a few children in each group.
The experience with secretin is an example of the popularity of CAM for treatment of chronic disease. A recent review of patients evaluated in a regional autism center reported use of some type of CAM in 30% of patients referred for initial evaluation or confirmation of the diagnosis of autism (Levy, Mandell, Merhar, Ittenbach, Pinto-Martin, unpublished data, 2003). Further research into how families make decisions regarding treatment choices, how they analyze information available to them, and their reasons and specific expectations of CAM will be necessary to understand this seeming paradox between data and choice.
CATEGORIES OF CAM TREATMENTS
CAM treatments can be categorized according to their proposed mechanism of action. The categories are not mutually exclusive; treatments may fit in several categories. Available and proposed CAM treatments are listed in Table 2 according to the primary hypothesized mechanisms proposed by advocates of the treatments. Treatment effects may, however, be explained by other reasons after scientific investigations of response and causation.
Types of CAM
Vitamin B6 and magnesium. Supplementation with high doses of vitamin B6 with magnesium is a common biologic intervention for symptoms of inattention and to enhance language in people with autism. Fourteen studies, including a few case series, of people with autism treated with B6 reported success.13 A recent Cochrane review concluded existing studies were not adequate to comment on efficacy.14 Many of these studies had significant design flaws, however, including small and heterogeneous samples, and outcome measures that were not validated. Doses most frequently used were 30 mg/kg per day of vitamin B6 and 10 mg/kg per day of magnesium. Two double blind, placebo-controlled studies did not report positive effects. One of these studies used doses at the Recommended Daily Allowance (RDA) because of concerns regarding the known side effect of parasthesias of the hands.15
Vitamin C. Vitamin C has been suggested as a potential treatment for symptoms of autism because it may be a weak dopamine blocker. One small study examined the behavioral responses of students with autism in a residential setting who were fed a consistent diet and then supplemented with 8 g/70 kg per day of vitamin C. Significant decreases in stereotyped behavior were reported on a standardized observation instrument,16 but this study has not been replicated. At higher doses, vitamin C could lead to kidney stones.
DMG/TMG. Pantethenic acid, what was once marketed as vitamin B 15, is available as a food supplement called dimethylglycine (DMG). References to an older report from Russia suggest children with developmental disabilities increased expressive language skills when their diets were supplemented with DMG. One small, double-blind study reported negative findings in boys with autism treated at modest doses.17 One child in the experimental group experienced hyperactivity.
Omega-3, fish oils, or essential fatty acids. Intake of essential fatty acids and omega-3 fatty acids is hypothesized to affect neurotransmission and production of prostaglandins.18 A randomized, controlled study of children with attention-deficit/hyperactivity disorder (ADHD) did not confirm clinical changes with supplementation.19 No data, only theory, have been reported for children with ASD.
Changes in GI Function
Gluten-free/casein-free diet. A milk- and wheat-free diet has been suggested as a primary treatment for symptoms of autism. This is based on the hypothesis that increased intestinal permeability in children with autism allows for increased absorption of morphine-like compounds found in casein and gluten.10 These exogenous opiates then cause or increase social withdrawal, stereotyped behaviors, and other symptoms suggested to be common to both opiate intoxication and autism. Gastrointestinal problems such as gastroesophageal reflux, altered permeability, and increased pancreatic secretin have been reported in children with autism.20 However, a review using the United Kingdom General Practice database, did not demonstrate a higher incidence of gastrointestinal problems in children with autism than in the general population.21
Advocates of the diet suggest maintenance for at least 8 months before considering it ineffective. A small case series reported that children with autism on this diet significantly improved over time.22 Two randomized trials have been conducted; in both, parents were aware of the group in which their children were placed. One reported subjective improvement,23 while the other found no improvement.24 Maturation and other interventions were not addressed as confounding factors in any of these reports.
Dietary interventions such as those described above are popular. Parents of young children must be counseled regarding necessary calcium and vitamin D intake and appropriate sources. Substitution of soy, rice, or potato beverages for cow's milk may decrease protein intake. Families should be counseled regarding overall dietary sufficiency, especially when food aversions further limit the child's diet. Consultation with a pediatric registered dietician would be advised for families who pursue this diet, although most families obtain extensive information from Internet Web sites, such as http://www.gfcfdiet.com, and from lay publications.
Pepcid The H-2 antagoinist famotidine (Pepcid©) was suggested as a potential treatment for ASD because of proposed effect in patients with schizophrenia.10
Immune Mechanism Modulation
Antibiotic or antifungal treatment. It has been hypothesized that, either because of an inherent immunologic alteration or because of prior antibiotic treatments, children with autism have altered gut flora and treatment with antibiotics will change the course of symptoms of autism.25 Proponents of this theory suggest that this alteration results in production of toxic metabolites as a bacterial byproduct.
A report of vancomycin treatment in a small number of children with clinically defined autism was associated with subjective shortterm improvement in behaviors.26 This study, however, did not use standardized outcome measures or control for potentially confounding factors. Also, the use of antibiotics typically reserved for resistant organisms is not advocated in the absence of documented resistant infections, because of concern regarding emergence of drug-resistant organisms. Stool cultures for bacteria and fungi are difficult to interpret with out evidence of intestinal infection.
It has been hypothesized that many neurodevelopmental and medical disorders are caused or aggravated by yeast overgrowth in the colon. Shaw27 expanded this theory with the case report of two brothers who had putative Krebs cycle intermediates identified on urine organic acid measurement. These studies were obtained to investigate loss of milestones, intermittent motor findings, and onset of symptoms of autism. This case report describes a course not typically seen in idiopathic autism. Based on this observation, however, the low-sugar diet and the use of probiotic agents that change gut flora, such as acidophilus and lactobacillus, have gained in popularity. Neither the use of these treatments nor of nystatin or fluconazole to diminish symptoms of autism have been subject to controlled trials.
According to Shaw et al.,27 the increased urine excretion of putative yeast metabolites suggests an overgrowth in the gut without overt infection. This may be secondary to an inherent alteration in immunity, overuse of antibiotics in early childhood, abnormal yeast flora, or an idiopathic response. Organic acidurias are rare causes of developmental disorder, and testing for urine organic acids has low yield in the absence of suggestive symptoms.1
Intravenous immunoglobulin (IVIG). Many reports describe circumstantial and indirect evidence for dysfunction of the immune system in children with ASD. Intravenous immune globulin is a plasma-derived product used to treat some severe neurologic disorders of presumed immune origin, such as GuillianBarré syndrome and multiple sclerosis.28 Because disorders such as multiple sclerosis result in regression in neurologic function, the regression in milestones experienced by up to one-third of children with autism in the second year of life has caused investigators to consider the possible etiologic role of immune factors in autism.
Other studies have provided laboratory evidence of specific immune deficits in small samples of children,29 but these have not accounted for the confounding factors of stress, other biologic treatments, concurrent illness, or underlying etiology of disability. Although increased prevalence of infections has not been documented among children with ASD, it is plausible that an undetermined immunologic deficit is present in at least a subset of patients.
An open trial of treatment with intravenous immunoglobulin reported subjective improvement and improvement across a range of measures.10 The study design was not standardized, however, and did not include a control population.
IV-IG administration carries a small risk for complications such as aseptic meningitis, renal failure, or infection.28 Insurers and pharmacy oversight committees increasingly have regulated administration of this product because of its short supply and cost.
Vitamin A. The administration of supplemental vitamin A to children in the Third World with diarrheal illness has become standard of care to decrease the length of illness. Vitamin A facilitates the immune response by modulation of G-protein function in cell membranes. An abnormality in this system has been suggested in response to MMR vaccine administration, in extension of the documented response to measles infection. It is proposed that the abnormality may be overcome by the administration of cod liver oil as a natural source of vitamin A.30
No experimental data have been published to support this theory. Known toxicity of hypervitaminosis A includes pseudotumor cerebri, skin rash, and hepatomegaly. Families must be cautioned to read the labels of supplements they give their children so as not to exceed a safe dosage. Ingestion of vitamin A at rates of 6,000 pg retinol equivalents (RE) per day or more for extended periods (1 to 2 months) carries a high risk of toxicity.31 Unfortunately, many suppliers of cod liver oil do not indicate the concentration of vitamin A in their preparations. Some of the preparations have 5,000 RE per 5 mL so dosage of more than 1 teaspoon would be in the toxic range.
Withholding immunizations. Several hypotheses support the practice of either withholding immunizations or splitting the doses into separate aliquots. These hypotheses include the possibility that the immune system is overwhelmed by exposure to multiple antigens (eg, measles, mumps, and rubella), with the possibility of subsequent colonization of the gastrointestinal tract with live virus.32 Multiple epidemiologic studies have not confirmed a link between MMR vaccination and onset of autism.33 Although this remains controversial, studies have not confirmed that live viruses in gastrointestinal tracts of children with ASD are causative factors.
Other hypotheses include the possibility of vitamin A deficiency postimmunization, which has an effect on the immune system. The issue has been further clouded by the controversy of possible mercury poisoning as a result of thimerosal preservative, but thimerosal was never included in live virus vaccines such as MMR and has been removed from such vaccines as DPT, HIB, and hepatitis.
Chelation for heavy-metal poisoning. The similarities between the symptoms of mercury toxicity and ASD have led to the hypothesis that mercury and other heavy metals are causative agents in genetically vulnerable populations.34 Sources of mercury intake could include fish in the diet or thimerosal use in vaccines. Much of the evidence has been circumstantial, in the forms of comparable descriptions of symptoms or presence of mercury in relatively inert organs of the body (eg, nails, hair). Thimerosal has been eliminated from vaccines since 1998, but there has not been a documented decrease in prevalence of the disorder.
Detoxification. It has been proposed that some children with autism do not fully process compounds such as acetaminophen or chemical compounds found in certain fruits and vegetables. Administration of supplements such as Ndimethylglycine (DMG), also known as pangamic acid, or Epsom salt baths is suggested to aid in the detoxification of the metabolites, which include sulfur and cysteine. Families following this strategy avoid certain dietary substances to minimize accumulation of these toxins. No clinical trials are of this type of intervention are reported in the medical literature. Also, while absorption of magnesium after ingestion of Epsom salts is minimal, this may be a source of toxicity.35
Interventions considered to be CAM are in constant flux. New treatments emerge, older treatments become less popular, and the cycle recurs. Data supporting new treatments should be scrutinized for scientific study design, clinical safety, and scientific validity.
Many families approach the clinician armed with brochures, handouts, and printouts from Web sites that are dedicated to the care and support of parents and children with ASD. A recent web search using "autism and detoxification" resulted in almost 8,000 sites. The Defeat Autism Now! (DAN!) Project arose in 1995 from collaboration of members of the Autism Research Institute. The DAN! Project advocates a specific and extensive protocol for diagnosis and treatment and can be viewed at http://www.autism.eom/ari/#dan.
The scientific validation and support for many interventions is incomplete and disparate from the recommendation in the American Academy of Pediatrics Policy Statement.1 Families should be encouraged to discuss all proposed investigations or treatments they wish to try with their primary care provider so the practitioner can serve as the medical home38 (Sidebar, page 688). The clinician should communicate and collaborate with the family and educational professionals to encourage objective identification of what works.
With increasing access to health information and societal pressure for families to actively participate in their health management, continued growth of interest in CAM can be anticipated. Clinicians must remember that parents may have different beliefs regarding the effectiveness of treatment and different tolerance for treatment risks.25 Practitioners must keep avenues of communication open, remain open-minded, and not assume a "don't ask, don't tell" posture in the context of providing a medical home to the increasing number of children diagnosed with autism.5
1. Committee on Children with Disabilities, American Academy of Pediatrics. The pediatrician's role in the diagnosis and management of autistic spectrum disorder in children. Pediatrics. 2001;107(5):1221-6.
2. Yeargin-Allsopp M, Rice C, Karapurkar T, et al. Prevalence of autism in a US metropolitan area. JAMA. 2002;289(1):49-55.
3. Chakrabarti S, Fombonne E. Pervasive developmental disorders in preschool children. JAMA. 2002;285(24):3093-3099.
4. National Research Council. Educating Children with Autism. Committee on Educational Interventions for Children with Autism, Division of Behavioral and Social Sciences and Education. Washington, DC: National Academies Press; 2001.
5. Committee on Children with Disabilities, American Academy of Pediatrics. Counseling families who choose complementary and alternative medicine for their child with chronic illness or disability. Pediatrics. 2001;107(3):598-601.
6. Sprague RL, Werry JS. Methodology of psychopharmacological studies with the retarded. Int Rev Res Ment Retard. 1971;5:148.
7. Halsey NA, Hyman SL. Measles-mumpsrubella vaccine and autistic spectrum disorder: Repon from the New Challenges in Childhood Immunizations Conference convened in Oak Brook, Illinois, June 12-13, 2000. Pediatrics. 2001;107(5):E84.
8. 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(1):9-15.
9. Levy SE, Souders MC, Wray J, et al. Comparison of placebo and single dose of human synthetic secretin in children with autistic spectrum disorders. Archives of Disease in Children. 2003;88(8):73 1-736.
10. Hyman SL, Levy SE. Autistic spectrum disorders: when traditional medicine is not enough. Contemp Pediatr. 2000;17:101-116.
11. Sandler AD, Bodfish, JW. Placebo effects in autism: lessons from secretin. J Dev Behav Pediatr. 2000;21(5):347-50.
12. Copian J, Souders MC, Mulberg AE, et al. Parents of children with autistic spectrum disorders are unable to distinguish secretin from placebo under double-blind conditions. Archives of Disease in Children. 2003;88(8):737-739.
13. Pfeiffer SI, Norton J, Nelson L, Shott S. Efficacy of vitamin B6 and magnesium in the treatment of autism: a methodology review and summary of outcomes. Journal for Autism and Developmental Disability. 1995;25:481-493.
14. Nye C, Brice A. Combined vitamin Bo-magnesium treatment in autism spectrum disorder. Cochrane Database Syst Rev. 2000;(4):CD003497.
15. 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. Journal for Autism and Developmental Disability. 1997;27(4):467-478.
16. Dolske MC, S pollen J, McKay S, Lancashire E, Tolbert L. A preliminary trial of ascorbic acid as supplemental therapy for autism. Prog Neuropsychopharmacol Biol Psychiatry. 1993;17(5):765-774.
17. Bolman WM, Richmond JA. A double-blind, placebo-controlled, crossover pilot trial of low dose dimethylglycine in patients with autistic disorder. Journal for Autism and Developmental Disability. 1999;29(3):191194.
18. Haag M. Essential fatty acids and the brain. Can J Psychiatry. 2003;48(3): 195-203.
19. Voigt rg, Liorente AM, Jensen CL, et al. A randomized double-blind, placebo-controlled trial of docosahexaenoic acid supplementation in children with attention-deficit/hyperactivity disorder. J Pediatr. 2001; 139(2): 189-196.
20. Horvath K, Perman JA. Autistic disorder and gastrointestinal disease. Curr Opin Pediatr. 2O02;14(5):583-587.
21. UK General Practice Research Database. Gastrointestinal symptoms not linked to later autism. BMJ. 2002;325: 419-421.
22. Whitely P, Rodgers J, Avery D, Shattock P. A gluten free diet as an intervention of autism and associated spectrum disorders: preliminary findings. Autism. 1999;3:45-65.
23. Knivsberg AM, Reichelt KL, Hoien T, Nodland M. A randomized, controlled study of dietary intervention in autistic syndromes. Nutr Neurosci. 2002;5(4):251-261.
24. Sponheim E. Gluten-free diet in infantile autism: a therapeutic trial. Tidsskr Nor Laegeforen. 1991;111(6):704-707.
25. Finegold SM, Molitoris D, Song Y, et al. Gastrointestinal microflora studies in late-onset autism. Clin Infect Dis. 2002;35(Suppl 1):S6S16.
26. Sandler RH, Finegold SM, Bolte ER, et al. Short-term benefit from oral vancomycin treatment of regressive-onset autism. J Child Neurol. 2000;15(7):429-435.
27. Shaw W, Kassen E, Chaves E. Increased urinary excretion of analogs of Kerbs cycle metabolites and arabinose in two bromers with autistic features. Clinical Chem. 1995;41(8Pt. 1):1094-1104.
28. Bril V, Allenby K, Midroni G, O'Connor PW, Vajsar J. IVIG in neurology - evidence and recommendations. Can J Neurol Sci. 1999;26(2); 139-152.
29. Croonenberghs J, Wauters A, Devreese K, et al. Increased serum albumin, gamma globulin, immunoglobulin IgG, and IgG2 and IgG4 in autism. Psychol Med. 2002;32(8): 1457-1463.
30. Megson M. Is autism a G-alpha protein defect reversible with natural vitamin A? Med Hypotheses. 2000;54(6):979-983.
31. American Academy of Pediatrics. Pediatric Nutrition Handboot 4th ed. Elk Grove Village, Ill: American Academy of Pediatrics; 1998.
32. Wakefield AJ, Murch SH, Anthony A. et al. rieal-lymphoid-nodular hyperplasia, non-specific colitis, and pervasive developmental disorders in children. Lancet. 1998;351(9103):637-641.
33. Stratton K, Gable A, McCormick M, eds. Immunization safety review: thimerosal-containing vaccines and neurodevelopmental disorders. Washington, DC; Institute of Medicine, National Academies Press: 2000.
34. Morris ME, LeRoy S, Sutton SC. Absorption of magnesium from orally administered magnesium sulfate in man. J Toxicol Clin Toxicol. 1987;25(5):371-382.
35. American Academy of Pediatrics. The medical home. American Academy of Pediatrics, Policy Statement Pediatrics. 2002;110(1):184-186.
Suggested Guidelines for Treatment of ASD