Complementary and alternative medicine for autism spectrum disorders: Rationale, safety and efficacy
- Conflict of interest: None declared.
Correspondence: Dr Andrew JO Whitehouse, Telethon Institute for Child Health Research, Centre for Child Health Research, University of Western Australia, 100 Roberts Road, Subiaco, WA 6872, Australia. Fax: +61 8 9489 7700; email: firstname.lastname@example.org
Complementary and alternative medicine is widely used for children with autism spectrum disorder, despite uncertainty regarding efficacy. This review describes complementary and alternative practices commonly used among this population, the rationale for the use of each practice, as well as the side-effect profile and evidence for efficacy. The existing evidence base indicates that melatonin can be recommended as a treatment for sleeping disturbances associated with autism spectrum disorder, while secretin can be rejected as an efficacious treatment for broader autistic symptoms. There is insufficient evidence to draw conclusions on the efficacy of modified diets, hyperbaric oxygen therapy, immune therapy, and vitamin and fatty acid supplementation. There is a clear need for methodologically rigorous studies to provide evidence-based guidance to families and clinicians regarding complementary and alternative practices for individuals with autism spectrum disorders.
- Complementary and alternative medicine is widely used for individuals with autism, despite uncertainty regarding efficacy.
- Existing evidence suggests that melatonin can be recommended for sleeping disturbances, while secretin can be rejected as an efficacious treatment.
- There is insufficient evidence to draw conclusions on the efficacy of modified diets, hyperbaric oxygen therapy, immune therapy, and vitamin and fatty acid supplementation.
Autism spectrum disorder (ASD) is the collective term for neurodevelopmental disorders characterised by qualitative impairments in social interaction and communication, and a restricted range of activities and interests. Many countries, including Australia, have reported a dramatic increase in the number of diagnoses, and recent estimates put the prevalence of ASD at 1 in every 110 individuals. There is now a substantial evidence base for a range of treatment strategies that target the core features of ASD. Educational and behavioural interventions in particular have been the focus of several well-designed studies that have reported benefits of intense and sustained therapy. Pharmacological interventions have also been trialled with mixed success; a recent systematic review found that risperidone and aripiprazole are the most efficacious psychotropic medications for reducing challenging and repetitive behaviours. Given the considerable time and cost associated with behavioural and educational interventions for children with ASD, in addition to the inconsistent evidence of efficacy and significant side-effect profiles for many pharmacological interventions, additional treatments have been sought for this population.
Over the past two decades, complementary and alternative medicine (CAM) – diagnosis, treatment and prevention philosophies and techniques based on traditional (often, non-Western) medicine that are used with (complementary) or in place of (alternative) conventional medicine – has become widely used by families seeking treatment for ASD. Audit studies have estimated the prevalence of CAM use in children with ASD at between 52% and 95%,[7-9] compared with approximately 30% of children in the general population. Given the prevalence of CAM usage in children with ASD, an understanding of the rationale and evidence of efficacy underpinning the use of different therapies in this population is essential for clinicians to provide informed guidance to families.
This review describes biologically based CAM practices that are most commonly used for children with ASD. The review outlines the rationale for the use of each practice, as well as the side-effect profile and evidence for efficacy. The review provides an overview of efficacy trials, with a focus on those studies that utilised a randomised placebo-controlled trial design.
Gluten-Free, Casein-Free Diet
Diet modification as a treatment for ASD is based on the as-yet scientifically unproven hypothesis that opioid peptides, formed from the incomplete breakdown of foods containing gluten (found in wheat, rye and barley) and casein (the main protein in dairy products), may enter the bloodstream due to increased intestinal permeability, cross the blood–brain barrier and impede central nervous system development and functioning. Despite the wide use of gluten-free and casein-free diets (GFCF) diets in the ASD community – estimated at 40% of children with ASD – very few well-designed trials have been conducted. The first study in this area was a single-blind trial of 20 children with ASD (age range: 5 to 10 years) randomly allocated to a GFCF or control diet for 12 months, which found a significantly greater reduction in the severity of autistic symptomatology in the GFCF group, as well as small (though non-significant) improvements in verbal and non-verbal ability. A second study, utilising a double-blind design, investigated behavioural changes in 13 children with ASD (age range: 2 to 16 years) randomised to a GFCF or control diet for 12 weeks. There were no significant differences between the treatment and control groups in parent-rated and experimenter-observed behaviours across the length of the trial. A Cochrane review of these two studies concluded that evidence for efficacy of GFCF diets for children with ASD was poor. A third randomised placebo-controlled trial has been published since the last update of the Cochrane review. Children with ASD aged between 3 and 5 years were randomly allocated to a GCFC diet (n = 8) or a low-sugar diet (n = 14) for a 3-month period using an open-label design. While improvements in a range of behavioural and developmental outcomes were observed in both the GCFC and low-sugar diet groups, there were no statistically significant differences between the groups. No major adverse effects were observed in any of the three studies. Currently, there is a lack of evidence to support the use of GFCF diets as an effective intervention for children with ASD.
Secretin is a 27-amino-acid polypeptide secreted by the duodenum that aids in digestion by stimulating the release of bicarbonate and enzymes from the pancreas and bile from the liver. Like GFCF diets, the administration of secretin for ASD is based on the unproven hypothesis that secretin may aid digestion, thus restricting the flow of harmful opioid peptides crossing the blood–brain barrier. Interest in secretin as a treatment for ASD arose from an uncontrolled study of three children, which found that intravenous administration of the hormone was followed by a marked improvement in social and communicative skills. Since this report, secretin has become one of the most intensely studied biological treatments for ASD. A Cochrane review last updated in 2010 identified 16 randomised controlled trials that met criteria for review, which included a total of over 900 children with ASD, with duration from the start of the intervention to outcome assessment ranging from 3 to 6 weeks. Meta-analysis of these studies found that there was no substantial evidence that single or multiple dose intravenous secretin is effective in improving core ASD symptomatology both within the broader ASD population and within ASD subgroups. While no serious adverse effects of secretin administration have been reported, several studies have observed milder side effects including flushing, hyperactivity and vomiting.
Melatonin is a neurohormone produced primarily in the pineal gland, which is known to have a regulatory role of the circadian rhythm. Prevalence studies estimate that up to 80% children with ASD have sleep difficulties, including longer sleep onset latency, increased frequency of night-time awakenings and reduced total sleep duration. Melatonin is commonly used for insomnia in otherwise typically developing children owing to its demonstrated efficacy, favourable side-effect profile and relatively low cost. Five randomised double-blind crossover studies of melatonin have been conducted in the ASD population[24-27] (including one study of children with Rett syndrome), with sample sizes ranging from 5 to 20 children. Dosage ranged from 2 mg to 10 mg per night, and treatment periods varied from 10 days to 3 months. Sleep behaviour was measured by parental sleep diaries,[24, 27] wrist actigraphy or both.[26, 28] Meta-analysis of these studies revealed significant improvements in sleep duration (44 min compared with placebo) and sleep onset latency (39 min), but no improvements in night-time awakenings. While positive effects of melatonin were greater for those studies that estimated sleep parameters using parental report sleep diaries, all studies reported positive effects for treatment compared with placebo.
The majority of studies of the ASD population, including four of the five double-blind randomised controlled trials,[24, 26-28] have reported no major or minor adverse effects associated with melatonin use. The largest study of melatonin and ASD, a retrospective medical chart review, found that only 3 of 107 children had mild adverse effects associated with treatment (enuresis or morning somnolence) based on parent report. This study also found no adverse effects of combining melatonin with other psychotropic medication, including no increase in seizure activity among 21 children with co-morbid epilepsy. While melatonin dosage varies considerably according to preparation, trials have established safety at doses up to 10 mg. It is hypothesised that short-term release preparations may be more efficacious for children having difficulty initiating sleep, and long-term release preparations for children having difficulty maintaining sleep, but this is yet to be established in trials with children with ASD.
Hyperbaric Oxygen Therapy (HBOT)
HBOT involves inhaling up to 100% oxygen in a pressurised chamber that maintains air pressure at greater that 1 atmosphere, but typically at pressures greater than 2 atmospheres, for a sustained period (usually 60 min per treatment session). The rationale for this therapy is based on the observation of cerebral oxidative stress and neuroinflammation in some children with ASD, and observations in animal models that HBOT may alleviate oxidative stress and decrease inflammatory responses. Rossignol et al. conducted the first double-blind placebo-controlled trial of this treatment, randomising 62 children with ASD to receive either hyperbaric treatment at 1.3 atmosphere and 24% oxygen (n = 33) or a placebo group, who were exposed to slightly pressurised room air at 1.03 atmosphere and 21% oxygen (n = 29). HBOT was administered twice a day (separated by a minimum of 4 h), 5 days per week for 4 consecutive weeks (total of 40 sessions). Compared with baseline performance, the treatment group had significant improvements in a range of behavioural domains, including receptive language, social interaction and eye contact. However, when improvements between the treatment and control groups were compared, there were few significant differences.
Since publication, this research has been further criticised for a number of methodological flaws, including ascertainment bias and loss of data. Furthermore, the peak body for hyperbaric medicine, the Undersea and Hyperbaric Medical Society (UHMS), produced a report highlighting that the oxygen and pressures used in the ‘treatment’ did not actually constitute HBOT, and that this environment can be experienced without a hyperbaric chamber. A subsequent replication trial by an independent research group, using the same oxygen and pressure parameters as the study by Rossignol et al. except for a longer duration of treatment (80 h of treatment over a 15-week period), observed no differences between HBOT and placebo groups on a wide range of social, communicative and adaptive outcome variables. While no participant in either study experienced a serious adverse effect as a result of the HBOT, there are known risks of barotrauma and exacerbation of pulmonary disease at higher atmospheric pressures. Based on the methodological flaws of the study by Rossignol et al., and the null findings of the replication study, the UHMS has declared that there is currently no evidence for that HBOT is an efficacious treatment of ASD.
Vitamin B6, Magnesium
The use of mega-vitamins as a treatment for ASD emerged in the late 1960s with a report that some children with ASD had improved speech and language following pyridoxine supplementation. This prompted a surge of interest in the efficacy of vitamin B6–magnesium (B6–Mg) combination in reducing ASD symptomatology, with the underlying rationale that these nutrients are crucial to the formation of several neurotransmitters, including serotonin, amniobutyric acid, dopamine, norepinephrine and epinephrine. Neurotransmission systems are hypothesised to be involved in the pathogenesis of ASD, although the exact link remains unclear. While there is a large literature on the use of B6–Mg as a therapeutic intervention for ASD, the majority of these studies have been plagued by a lack of rigorously controlled experimental designs, and a recent Cochrane review identified three studies only that met methodologically rigorous criteria for inclusion. Two double-blind placebo-controlled crossover trials reported no benefit of treatment of B6–Mg over a 4-, 10- or 20-week period. The third placebo-controlled trial utilised a parallel design, with four participants each in the treatment and placebo groups (age range: 8 to 12 years). Significant gains in verbal IQ were reported in the treatment versus control group after 4 weeks of B6–Mg supplementation, although the small number of participants temper the conclusions that can be drawn from these data. While no serious adverse effects were reported in these three short-term studies, long-term administration of pyridoxine has been implicated in sensory peripheral neuropathy. Based on the findings of the three methodologically rigorous studies undertaken in this area, there is currently no substantial evidence that B6–Mg conveys therapeutic benefit to children with ASD.
Essential Fatty Acids
Long-chain polyunsaturated fatty acids (LCPUFA), including docosahexaenoic acid and eicosapentaenoic acid, serve as indispensable components of cellular membranes and play a critical role in neuronal development. A number of studies have reported low levels of omega-3 fatty acids in children with ASD compared with typically developing children. Two small double-blind randomised controlled trials of n-3 LCPUFA supplementation (Total n = 37) in children with ASD have reported no statistically significant reduction in symptomatology for treatment versus control groups, even despite an overall increase in circulating levels of n-3 LCPUFA.[46, 47] No serious adverse effects were observed in either study, and the number of non-serious adverse effects (rashes, nosebleed, increased gastrointestinal symptoms) did not differ between treatment and control groups. There is insufficient evidence that LCPUFA supplementation is efficacious as a treatment for ASD.
Immunoglobulin is attracting increasing attention as a possible treatment for a range of central nervous system disorders such as multiple sclerosis, Guillain–Barré syndrome and intractable epilepsy. Preliminary evidence of a link between abnormalities in a range of immune system markers, such as antibodies and cytokines, and gastrointestinal symptoms in ASD has provided the basis for two double-blind placebo-controlled trials of immunoglobulin treatment in this population. The first study, a double-blind crossover trial, compared the behaviours of children with ASD with baseline performance, 6 weeks after the administration of intravenous immunoglobulin or placebo (n = 12). Although parent and teacher report indicated improvements in eye contact and speech and a reduction in hyperactivity for the treatment versus placebo injections, there was no difference in clinician ratings. A considerably larger randomised parallel trial (n = 125), examined the effect of three different doses of oral human immunoglobulin (140, 420 or 840 mg/day) and placebo on children with ASD after 12 weeks of treatment. There were no differences between groups in the level of reduction in gastrointestinal symptoms, nor in the magnitude of improvement in ASD symptoms. While treatment was well tolerated by children in both studies, there is insufficient evidence to recommend immune therapies for children with ASD.
This review identified a paucity of evidence for the majority of biologically based CAM practices used for treatment of ASD. Currently, only melatonin can be recommended as an efficacious treatment for the sleep disturbances that are common among children with ASD (predominantly sleep onset latency and sleep duration), although improvement in sleep may also have benefits for wake-time behaviour. In contrast, a large volume of research allows the rejection of secretin as a treatment for ASD, while there is insufficient evidence to draw conclusions on the efficacy of GFCF diets, HBOT, immune therapy, and LCPUFA and B6–Mg supplementation. Other biologically based CAM therapies not reviewed here have either not been the subject of randomised placebo-controlled trials (antifungal agents, carnitine, methyl B12), are awaiting replication studies after an initial randomised placebo-controlled trial (carnosine) or do not meet the safety standards required for an efficacy trial (chelation), and thus should be treated with caution. CAM is currently administered to the majority of children with ASD, and this review highlights the urgent need for methodologically rigorous, placebo-controlled trials to provide evidence-based guidance to families and clinicians regarding these practices. In the absence of these studies, the efficacy for the majority of CAM treatments for ASD remains unclear.
The author is supported by a Career Development Fellowship from the National Health and Medical Research Council (No. 1004065).
Multiple Choice Questions
- 1.Which complementary and alternative medicine can we recommend as efficacious in the treatment of autism spectrum disorders?
- Fatty acid supplementation
- Immune therapy
- Gluten-free and casein-free diets
ANSWER: D) Melatonin is efficacious in the treatment of sleeping problems associated with autism spectrum disorders.
- 2.Which complementary and alternative medicine can we reject as efficacious in the treatment of autism spectrum disorders?
- Fatty acid supplementation
- Immune therapy
- Gluten-free and casein-free diets
ANSWER: A) Placebo-controlled trials of over 900 children with autism spectrum disorders have found that secretin is not efficacious in this population.
- 3.The Undersea and Hyperbaric Medical Society currently recommends that hyperbaric oxygen therapy is:
- Used once a week for children with autism spectrum disorders
- Used twice a week for children with autism spectrum disorders
- Used three times a week for children with autism spectrum disorders
- Used in conjunction with behavioural therapy for children with autism spectrum disorders.
- Not used for children with autism spectrum disorders
ANSWER: E) The Undersea and Hyperbaric Medical Society has stated that there is currently no evidence that hyperbaric oxygen therapy is efficacious for children with autism spectrum disorders.