Autism spectrum disorder

Lisa Campisi, Nazish Imran, Ahsan Nazeer, Norbert Skokauskas, Muhammad Waqar Azeem, Autism spectrum disorder, British Medical Bulletin, Volume 127, Issue 1, September 2018, Pages 91–100, https://doi.org/10.1093/bmb/ldy026

Abstract

This article is based on key recent published literature including international guidelines and relevant reviews and meta-analyses. Authors have also supplemented this material with their own clinical experience.

There is an agreement that autism spectrum disorder (ASD) have a strong hereditary component. There is also a consensus that the reported prevalence estimates have increased in the last 5 years. There is strong support for using the broader spectrum disorder conceptualization of the DSM-5.

Higher public awareness of ASD has generated several controversial theories of causation. We review a number of environmental risk factors receiving media attention including: vaccines, mercury, heavy metal exposure and Selective Serotonin Uptake Inhibitors (SSRIs). Popular yet controversial treatment interventions are discussed. Early diagnostic screening tools are also addressed.

There is increasing scientific interest in identifying biomarkers of autism with potential for early diagnosis, prognostic indicators and predictive treatment responses. We review evidence from genetics, neuroimaging and eye tracking as candidate biomarkers.

Family studies point to a strong hereditary component in the aetiology of autism. However these studies have not established 100% concordance rates, suggesting a role for environmental factors. The gene–environment interplay has not received enough attention in scientific research. This represents an important new avenue for research in ASD.

Introduction

Autism spectrum disorder (ASD) is one the most common neurodevelopmental disorders, 1 characterized by persistent impairment in reciprocal communication and social interactions as well as restricted repetitive pattern of behaviours, interests or activities. 2 Kanner 3 wrote the first published description of autism in 1943 and after one year, Hans Asperger, a paediatrician from University of Vienna wrote about a group of children with almost similar pattern of behaviours as Kanner’s. The disorder is evident in all racial, ethnic and socioeconomic groups. 4 The diagnostic process of ASD is complex and recent changes in diagnostic criteria and how the disorder is conceptualized has initiated a discussion among professionals, policy makers as well as patients and their families. 5 In order to ensure best standards of care for these individuals, it is important for clinicians to have a better understanding of the disorder. This review provides a broader understanding of research findings, highlights current areas of agreement and controversies, clinical implications as well as critical issues for the research.

Increase in prevalence of autism spectrum disorder

There has been a dramatic increase in the prevalence of autism spectrum disorders, with current estimates of 1 in 68 children in the United States having ASD. 4 This figure is ~30% higher than what was reported in 2012 by the Center for Disease Control as 1 in 88 children having autism. However, it is unclear whether these numbers represent a true increase in prevalence, or are the result of increased awareness, differences in study methodology, or inclusion of subthreshold cases.

Clinical presentation

Autism is currently conceptualized as a spectrum disorder with significant variations in patients’ social, communicative and intellectual abilities. Symptoms lead to significant impairment in multiple domains of adaptive functioning. Individuals suffering from ASD need varying levels of psychosocial support to achieve relative independence, and in some cases, may need continuous care.

Social communication

Depending on age and intellectual abilities, children diagnosed with autism have a varying degree of communication deficits. These deficits range from speech delays, monotonous speech, echolalia, pronoun reversal, poor comprehension to a complete lack of spoken language. Nonverbal communication is also impaired and may include poor eye contact, difficulties in understanding facial expressions and descriptive gestures, to name a few.

Another important feature of individuals with ASD is deficits in socio-emotional reciprocity. These individuals are less likely to initiate conversation, show less interest in peer interactions and overall find it difficult to adjust their behaviour according to different social situations.

Restricted repetitive behaviours

Individuals with ASD have a varying presentation of restricted and repetitive behaviours. Repetitive behaviours may include simple stereotypical motor behaviour (e.g. hand flapping, finger flicking), repetitive use of objects (e.g. lining up toys), or repetitive speech (echolalia). Restricted behaviours may present as resistance to change or highly restricted fixated interests, which are abnormal in intensity or focus. Further, some children with ASD my present hyper and hyposensitivity to various sensory stimuli, presenting an extreme response to various sounds, lights or smells, or a decreased reaction to pain.

Associated clinical features

Individuals with ASD often have comorbid intellectual disability and also are prone to emotional difficulties such as anxiety or depression.

DSM-5 changes in autism spectrum criteria

In the Diagnostic and Statistical Manual for Mental Disorders, 5th Edition (DSM-5) released in May, 20 13, 2 major changes were made in the diagnostic criteria as noted in Table 1.

Comparison of changes in DSM-5 with DSM IV-TR in the diagnostic criteria of autism spectrum disorder

Comparison of changes in DSM-5 with DSM IV-TR in the diagnostic criteria of autism spectrum disorder

Implications of changes in DSM-5 diagnostic criteria for autism spectrum disorder

Differences in DSM-5 criteria for ASD in comparison to DSM-IV 6 have led to debates regarding the impact on the prevalence of the disorder, as well as the diagnosis, and subsequently clinical practice.

The DSM-5 taxonomy offers more stringent criteria for a diagnosis of ASD. 7 The literature raises concern that individuals who used to get the diagnosis of Asperger’s disorder, or Pervasive Developmental Disorder Not Otherwise Specified (PDD-NOS) under DSM-IV, are now less likely to meet criteria for the ASD diagnosis by using DSM-5 and therefore will not be eligible for services. 8 Studies also support the notion that young children and females 9 may be at higher risk of being underdiagnosed according to the DSM-5 criteria. 10– 12 These results are of concern as children who are less impaired are more likely to benefit from early intervention, may now be least likely to qualify for such services.

It is however important to note before the publication of DSM-5, there was a growing consensus among clinicians that subcategories of pervasive developmental disorders in DSM IV cannot be reliably diagnosed. 9, 13 Thus, despite the concerns of being a stringent criteria, efforts to conceptualize autism as a broad spectrum of disorders in the DSM-5 had met with less criticism by professionals. Several groups investigating the validity of shifting from a triadic model to a two-factor model (integration of impaired social interaction and limited social communication into one category) have also yielded support for it. 9, 14

Gender related differences in autism presentation

There are no significant differences in ASD core symptomatology between males and females. However, research consistently shows that females tend to have more severe symptoms, more associated intellectual disability, as well as higher risk of comorbid psychiatric problems. 15

Aetiology of ASD

There is a consensus in the literature that autism is caused by both genetic and environmental factors. Although family studies support a strong genetic component in the aetiology of idiopathic autism, concordance rates are not 100%, indicating that environmental factors also contribute to ASD. 16

A recent 10-year meta-analysis 17 identified some environmental risk factors from pre-conception to early childhood which are potentially involved in the presentation of ASD. Table 2 summarizes the most common environmental factors cited in the literature. As the evidence supporting causality is lacking and inconclusive, 18, 19 further studies are needed.

Unproven risk factors cited in the literature

Unproven risk factors cited in the literature

Controversial risk factors

Measles, mumps and rubella

Few subjects in the field of autism have garnered greater attention and controversy than that of vaccines and their potential causal link to autism. The issue was first brought forward in 1998 after an article published in the Lancet 20 in which authors postulated a causal association between the measles, mumps and rubella (MMR) vaccine and the development of autism. The study involved a group of 12 children with a prior diagnosis of autism who had also received the MMR vaccine before being diagnosed with autism. In their investigation, the authors concluded that the children were found to have an unrecognized form of inflammatory bowel disease along with small traces of the measles virus that matched the strain of measles in the vaccine. These findings led to the interpretation that the MMR vaccine, which contains a live virus, had caused a measles infection in the gastrointestinal system which in turn led to a ‘leaky gut.’ Autism was hypothesized to be the result of toxins entering the blood stream via the gut and affecting the nervous system.

The outcome of this publication gained significant media attention, however an overwhelming body of scientific evidence has since been put forth refuting a relationship between ASD and MMR, 21, 22 yet the issue remains controversial. The initial article published in Lancet on this topic was retracted in 2010 citing that ‘several elements’ of 1998 paper ‘are incorrect, contrary to the findings of an earlier investigation’.

Mercury and thimerosal

Another issue that added fuel to the vaccination controversy was thimerosal. Thimerosal, a mercury-based vaccine preservative is frequently found in combined vaccines, including diphtheria-tetanus-pertussis (DTP). Thimerosal contains 50% ethyl mercury, which is similar to methylmercury, which at high doses can be toxic to humans. 23 The controversy with thimerosal was not related to the MMR vaccine, given this vaccine never contained ethyl mercury. However, public scepticism with vaccination led to the viewpoint that high levels of thimerosal could lead to autism in children. A vast amount of scientific data on the topic has overwhelmingly disproven this claim. 24 Despite the lack of evidence and steps taken by public health officials to counteract these claims, this controversy has had significant consequences on childhood immunity and infectious disease control.

Heavy metal exposure

One meta-analyses 25 has examined metal concentrations in children with ASD. The heavy metals most commonly studied in this population have been cadmium, lead, arsenic and aluminium. There is little evidence to support a link between aluminium and ASD, whereas the evidence concerning the others is conflicting. The authors conclude that while there is evidence suggesting heavy metal exposure as a risk factor for the development of autism, data to conclusively support these claims is still lacking.

Pollutants

Air pollution is speculated to have neurological consequences including inflammation and oxidative damage in the brain that, in turn, leads to abnormal neural development. There is also some research suggesting that residing near a freeway or highway during the latter stages of pregnancy could increase the risk of a child developing autism. 26 Again, conclusive evidence supporting this claim is lacking.

Selective serotonin reuptake inhibitors

The evidence for a link between prenatal maternal selective serotonin reuptake inhibitors (SSRI) antidepressants exposure and ASD is relatively new and controversial and mainly relies on animal and preclinical studies. 27

Controversial treatments and interventions

The challenge of developing a universal theory on the aetiology of autism, along with variations in the developmental trajectory are some of the factors that account for the wide range of interventions proposed to, and endorsed by, families of children with ASD. At times, families sometimes turn to interventions which lack empirical validity or have even been found to be harmful to the child development. 28

Nutrition

In recent years, dietary interventions including gluten-free or casein-free diets (GFCF) have received a significant amount of interest from parents and researchers alike. In Europe, according to one report, 13.5% of families of children with ASD 28 have tried this dietary restriction. The rationale for this approach appears to stem from the ‘Opioid-Excess Theory’, 20, 29 according to which some individuals do not produce sufficient gluten- and casein-related digestive enzymes. Without enough of these enzymes, gluten and casein-related peptides do not get adequately metabolized and cross the brain–blood barrier. Symptoms of ASD are believed to be the result of these peptides attaching to opioid neuro-receptors, which in turn disrupts the central nervous system. However, support for this theory, the underlying mechanism, and method of treatment is limited.

The ScanBrit study 30 examined the effects of a GFCF diet in Danish children and found positive changes in symptom presentation following this intervention. However, there were several limitations to this study, including the lack of a placebo group and a high attrition rate. In contrast, two recent systematic reviews 31, 32 have since shown no significant differences in symptom relief in children with ASD following a GFCF diet.

Dietary supplements are another popular nutritional approach considered by parents and caregivers of children with ASD. These include mineral supplementation (e.g. calcium, zinc, iron), as well as folic acid. Vitamin D supplementation has received significant attention because of its impact on neural developmental, anti-inflammatory properties and effects on detoxification pathways. However, evidence supporting dietary and vitamin D supplementation is again limited and inconsistent. 33, 34

Hyperbaric oxygen

Hyperbaric oxygen therapy (HBOT) provides a higher concentration of oxygen by delivering oxygen to a chamber with elevated atmospheric pressure. The rationale for HBOT in individuals with autism includes its potential to heighten cerebral perfusion, reduce inflammation and oxidative stress. 35 Advocates of HBOT believe that improvements in these underlying pathophysiological mechanisms will lead to improvements in autistic symptoms. Evidence supporting its effectiveness in relieving symptoms of ASD is mixed at best. The few randomized controlled studies 36 that have been conducted have shown no evidence to support the benefit of HBOT in children with ASD, with some identifying adverse events in the hyperbaric oxygen group (minor-grade ear barotrauma). The United States Food and Drug Administration (USFDA) 37 has published a warning for parents cautioning them against the use of HBOT in treating symptoms of ASD.

Chelation therapy

Chelation therapy involves the administration of several chemical substances to bind and then remove specific metals from the person’s body. It is estimated that between 7% and 8% of children with ASD in the United States 38 and 1–3% of children with ASD in Europe 28 undergo this form of therapy. However, one randomized clinical trial comparing multiple doses of chelating agents have found no evidence to suggest that oral chelating agents had any effect on ASD symptomatology. 39 Moreover, serious adverse side effects have been reported including hypocalcaemia and impaired renal function in individuals undergoing chelation therapy.

Pharmacological treatments

Medications are sometime needed to treat the comorbid symptoms of irritability, aggression, and hyperactivity in individuals of ASD. The FDA has approved two atypical antipsychotic medications, risperidone and aripiprazole for this purpose. 40, 41 Although evidence has suggested short-term improvements in the behaviour of children with ASD, long-term benefits and pros and cons of this strategy are still being debated.

SSRIs are also often prescribed to treat comorbid symptoms in ASD, but clinical trials have yet to demonstrate their effectiveness. A review of nine randomized controlled-trials assessed various SSRIs including fluoxetine and citalopram 42 but failed to show a positive result in symptoms reduction.

Early diagnosis and intervention

Autism is a clinical diagnosis, made on the presentation and history of the individual and in spite of a very active hunt for biomarkers, no laboratory test has been found. The literature on early diagnosis and treatment is in agreement that there is an inverse relationship between age at diagnosis and a positive prognosis, which makes early detection by health care providers critical. A 10-year study spanning from 2004 to 2014 examined diagnostic trends in the UK 43 and found no statistical change in improving early detection during this period, with the average age at diagnosis remaining at 55 months. This was despite parents noting their first concern about their child development much earlier than the time when the child was diagnosed with ASD.

The American Academy of Pediatrics 44, 45 has recommended that paediatricians use autism screening tools, such as the Modified Checklist for Autism in Toddlers (M-CHAT) and the Social Communication Questionnaire (SCQ) at the 18 and 24-month check-up visits. A newer, improved two-step revision of the M-CHAT with Follow-up Interview (M-CHAT/F), has been developed, and is reliable and sensitive in identifying toddlers at high, medium and low risk for ASD. 46 Data are still lacking to determine if an early detection has in fact improved following its implementation.

Early detection has a positive impact on prognosis, in large part due to the fact that those children are able to benefit from early intervention. A substantial amount of scientific literature is derived from data looking at the efficacy of Applied Behaviour Analysis (ABA). ABA is an intensive treatment programme, developed in the 1960s, based on using learning principles, such as positive reinforcement to help children develop appropriate behaviours. Over the last five decades, there has been accruing evidence documenting the effectiveness of ABA in treating ASD. 47 There is also evidence that those children who received earlier and more intensive ABA therapy, were more likely to have a positive prognosis in later childhood and adulthood. 48

Biomarkers

Biological markers or biomarkers are defined as ‘biological variables associated with the disease of interest and measurable directly in a given patient or his/her biomaterials using sensitive and reliable quantitative procedures’. 49 The identification of early biomarkers of autism is important as it can be used in providing earlier, more reliable diagnoses, and helps to predict prognosis and response to specific interventions.

Genetics

As noted earlier, there is strong support for a hereditary component to the presentation of ASD with higher concordance rates of autism in monozygotic twins than dizygotic twins. Some studies are also finding higher rates in families with two or more children diagnosed with autism, with rates approaching 50% affected. However, to date, no consistent genetic variant has been identified. Moreover, given its varied clinical and behavioural manifestations, it has been estimated that there are over 500 distinct genetic variants that may be related to ASD, 50 which makes it difficult to identify the target genes accurately.

Neuroimaging

Head circumference and grey matter thickness is under investigation as a relevant diagnostic indicator, given findings of accelerated brain growth and brain size in ASD children. Whereas newer neuroimaging techniques have helped to elucidate findings on the pathophysiology of ASD, the current scientific evidence on the subject is not adequate to establish reliable neuroimaging biomarkers. 51

Eye tracking

One promising technique is using the eye-tracking technology under the assumption that infants and toddlers with ASD prefer geometric images as compared to social images. One such study 52 examined a subset of children with ASD who fixated on the geometric images and concluded that visual preference might potentially be used as an early indicator in identifying subtypes with more severe symptoms and likely negative prognosis.

Future directions

A common feature of autism research is that studies mostly consist of small samples and do not portray the full heterogeneity that is present in ASD. Given the various pathophysiological pathways that have been proposed to account for the development of this disorder, it is doubtful that a single biomarker is responsible for ASD. Furthermore, the heterogeneous features of ASD mean that different ASD patients have different requirements when it comes to treatments and interventions.

One large scale investigation aiming to address this issue is the EU‑AIMS Longitudinal European Autism Project (LEAP). LEAP is a worldwide, multidisciplinary study 52 that will include ASD and control participants across childhood and into adulthood collecting data on genomics, prenatal environmental risk factors, magnetic resonance imaging (MRI), functional MRI (fMRI), electroencephalogram (EEG) and biochemical biomarkers. The study holds promise not only in its potential to identify markers of autism but also to provide clinicians with predictive value in determining symptom progress and treatment response, which will also open up the possibility of targeted treatments.

As noted earlier, while there is a strong hereditary component thought to be involved in the aetiology of autism, environmental factors are also believed to play a role in its development. Some of the processes thought to be involved includes metabolic processes such as oxidative stress, immune function and inflammation. These processes are believed to be derived from environmental influences such as the parent’s immune functioning, pollutants, diet and other risk factors as listed above. The gene–environment interaction has received less attention than basic genetics, but future research using a developmental framework and taking into account the interplay between genes and their environment represents yet another scientific approach.

Conclusions

ASD is a complex and heterogeneous neurodevelopmental disorder. In the last decades, there has been a steep rise in public awareness on ASD that has also coincided with a rise in prevalence. Whether the rising prevalence is a true reflection of an increase in rates of ASD or due to other confounding factors, these findings highlight the need for further research into ASD, particularly with regards to its aetiology, treatments and interventions. Evidence from biomarker studies offers promising insight into early identification and targeted treatments, but research is still in its infancy. Supporting and empowering families with children diagnosed with ASD and helping highlight their children’s strengths can also help make a difference in these families’ lives. 53

Conflict of interest statement

The authors have no potential conflicts of interest.