Role of environmental and genetic factors in autism spectrum disorder

Abstract

The aim of this study was to find out the environmental as well as genetic factors responsible for increasing the number of autism spectrum disorder (ASD) patients in Bangladesh. A questionnaire was developed based on 12 environmental factors and genetic aspects. Sixty six patients of ASD and 66 non-ASD control were selected randomly. Among the environmental factors, the age of the mother, premature birth, air pollution, age of the father, hypoxia during childbirth and oral contraceptive came out as significant (p<0.05) factors for ASD incidence compared to the control. Association of multiple factors on an individual was found to be crucial to enhance the risk and exposure to five and six factors was statistically significant (p<0.05) for ASD development. Prospective parents should try to keep the number of risk factors as low as possible before 1-2 months of pregnancy, during pregnancy and 1-2 years after the child birth (for child only).

Introduction

Autism spectrum disorder (ASD) is a heterogeneous neurodevelopmental disorder depicted by altered communication and social inter-action and repetitive behaviors.¹ Neurological disorder shows abnormal EEG and MRI.² It can be classified into three subgroups: Autistic disorder (autism), Asperger disorder, and pervasive developmental disorder.³ But the later version DSM-5 collapsed these three into one single broad diagnosis of ASD.⁴

ASD is considered as a major public health issue. At present estimated ASD incidence rate is significantly higher than pediatric cancer, HIV and heart disease collectively.⁵ The global prevalence of autism has increased 20 to 30-fold since the earliest epidemiologic data were collected in the late 1960s and early 1970s. During that period, estimated prevalence from European studies were 1 in 2,500 children in the population and by the 2000s broader survey estimated about 1-2% of all children having ASD.⁵

The scenario of Bangladesh in this context is critical. Like most developing countries, there is the lack of data on the number of children or adults suffering from this lifelong incapacitating developmental neurological condition. As one of the densely populated countries in the world, proper epidemiological data derived information is urgent to address the multifaceted need of ASD-affected people in Bangladesh. For this reason, Bangladesh government launched Dhaka Declaration on autism and mental disorders in 2011⁶ and later the resolution of the declaration was adopted in UN general assembly (67^th session, agenda item 127). In 2013, a pilot study utilizing community health workers in Bangladesh has indicated a prevalence of 0.15% for ASD (3% in Dhaka city and 0.07% in rural area).⁷ The number of the ASD children in Bangladesh is increasing at an alarming rate and it should be treated as a concerned issue. So, it is very urgent to find out the causes and factors responsible behind ASD in Bangladesh.

Both genetic and environmental factors seem to be responsible for ASD. Twin studies have shown a high concordance among identical twins (between 60 and 92%) which is much lower in fraternal twins (0 to 10%).⁸ In the hunt of the specific gene responsible for ASD, it turned out that many genetic mutations in different chromosomes are symptomatic to ASD. However, it is very unlikely to find the specific gene for ASD manifestation as it seems to develop in combination of complex genetic factors and environment. From a number of array comparative genomic hybridization studies, even on large cohorts of patients, it was found that up to 10% of sporadic (non-inherited) ASD cases do show de novo copy number variants.⁹ However, chromodomain helicase DNA binding protein 8 (CHD8) gene is the mostly proposed candidate for ASD so far. CHD8 may serve as a master regulator of a common ASD etiology.^10-12

In addition to genetic influences on ASD etiology, environmental parameters can contribute a major part. Environmental factors can be classified based on the affecting time period: Factors affecting at the time of conception, at the prenatal period (the time between conception and birth) and at the postnatal period (the time starts immediately after birth and extending for about 6 weeks) (Figure 1). But the most important period is the perinatal period. It is the time period immediately before and after birth (time starts at 22 completed weeks of gestation or at 25^th week of the second trimester and ends after seven complete days after birth).

The age of the father and mother at the time of conception is found to be critical for their child to develop ASD.^13-18 Exposure to pesticide-insecticide-,^{1, 8, 19, 20} infection driven inflammation,^21-23 medication (paracetamol, thalidomide)^{1, 24} and stress of the mother²² at the time of conception, the prenatal and perinatal period can enhance the risk to develop ASD for the child. Hypoxia during child birth¹³ and premature birth (birth of the baby before 35 weeks of gestation)^{13, 23, 25} are considered as risk factors for ASD. Air pollution,^{13, 26-33} heavy metals,^34-37 and vitamin D deficiency^{22, 38-40} during prenatal, perinatal or postnatal period can increase the risk for developing ASD for a child. Maternal and fetal inflammation,^{22, 23} maternal diseases (diabetes mellitus, asthma),²² maternal cigarette smoking,²² use of oral contraceptive,⁴¹ organophosphorus flame retardants,⁸ prenatal and perinatal analgesic (paracetamol/acetaminophen)^{1, 24} and valproic acid exposure^{1, 22} etc. are suspected to be associated with ASD.

The main objective of the study was to shed light on the important environmental factors responsible for ASD incidence in Bangladesh and the pattern how these factors might influence ASD development. It is very important to find out the major environmental risk factors associated with ASD to further investigate their mode of action and to take preventive measures to reduce the prevalence of ASD in Bangladesh.

Materials and Methods

Participants

The participants were ASD patients (n=66) and healthy non-ASD controls (n=66), randomly selected from Khulna and Comilla city of Bangladesh in 2016. Guardian of the participants was interviewed or they directly filled the structured questionnaires. The demographic data of participants of the study is presented in Table I. The age range of both ASD and non-ASD controls were between 3 to 24 years. Their gender ratio, educational and socio-economic status was not considered as the variable in this study. Some participants of the control group (n=25) was siblings of ASD patients. This was also a random choice and was not intended to influence any outcome of the study. ASD participants were selected both from specialized schools for ASD patients and also randomly from the neighborhood who does not attend specialized schools.

Measurement of data

The questionnaire was constructed with a focus on possible environmental risk factors that might be associated with the incidence of ASD in Bangladesh. After critical analysis of published literature total 12 environmental factors were considered to be analyzed. They were a) Age of the father, b) Age of the mother, c) Hypoxia during childbirth, d) Heavy metals, e) Air pollution, f) Infection, g) Medication, h) Maternal diseases, i) Premature birth, i) Mental stress of the mother, k) Pesticides and insecticides and l) Oral contraceptive. However, for few factors, more than one question was constructed. In such cases, positive response to any one question was considered positive for that particular factor (The questionnaire is in Supplementary data. The Bengali version of the questionnaire was used for the survey).

For environmental factors, after collecting the primary data, it was further processed in two different ways. First, factor-wise analysis was done. For each factor under study, a number of positive response(s) for both ASD and control group was counted and their percent exposure was measured. Second, person-wise association of factors was measured. To perform this, a number of factors (single or multiple factors) positively responded by a single individual was assessed. This was done for both ASD and the control group. Then the exposure distribution of single or multiple factors (for example, how many individuals are exposed to 2 factors or 5 factors) was converted to percent data for analysis.

For data related to the genetic association, the question about the family history (possible inherence) and the presence of other genetic diseases was investigated. However, this was only limited to ASD patients and was not performed in control group for two reasons. First, parents of healthy individuals are not aware of other genetic diseases and the tests are not regularly done in Bangladeshi clinical settings. Second, in the control group, 25 participants were siblings of ASD patients, so their family history could not be considered and thus could not be compared with ASD group.

Statistical analysis

All statistical analysis was performed in GraphPad Prism v. 6.0. The contingency table was constructed for each factor and Fisher’s exact test was used to analyze the statistical significance. p value of ≤0.05 was considered as significant when compared with control group.

Results

Environmental factors highly associated with ASD

By analyzing the survey data, among 12 environmental factors, the age of mother (p=0.0012) and premature birth (p=0.0012) were found as significant environmental factors for ASD incidence compared to the control group (Figure 2). There were also significant exposure differences between the ASD and non-ASD groups for air pollution (p=0.015), the age of father (p=0.0246), hypoxia during childbirth (p=0.0164) and oral contraceptive (p=0.0447). Other factors e.g. heavy metals, infection, maternal diseases, medication, the mental stress of mother and pesticides and insecticides were found to be non-significant in the current study.

Genetic factors

The analysis of the genetic factors was performed only for the ASD samples. Among the ASD participants (n=66), 15.1% had a record of other ASD individuals in their family. In the case of presence of related genetic disease, a large portion (69.7%) was not aware of the related genetic diseases. Only 10.6% participants confirmed that they had related genetic diseases (Table II).

Multifactor exposure to single individual can pose risk for ASD

An exciting and interesting observation was found by analyzing the number of factors exposed by an individual. For both ASD and the control group, percent data of exposure from ‘no factor’ towards increasing number of factors were plotted (Figure 3). Up to three factors, percent incidence was much higher in non-ASD group compared to ASD group. Especially for ‘one factor’, the difference was statistically significant. This trend, however, started to change in opposite direction, when the number of factors was over three. Percent incidence of association with more than three factors was found to be higher in ASD group compared to the control. In the cases of five factors (p=0.017) and six factors (p=0.0062), the difference in incidence was statistically significant. It was also observed that for a higher number of factors, there was no incidence observed in healthy control at all. From these, it might be concluded that many ASD patients showed multiple factor association at a time (e. g. 5, 6 or more factors). This may increase the risk of ASD development which is not evident in non-ASD control.

Discussion

In our study, we aimed to know the important factors behind ASD in Bangladesh. Air pollution, the age of parents, premature birth, hypoxia during childbirth and oral contraceptive were found as significant risk factors. Physical and developmental alteration on the fetus had been reported where mothers were exposed to air pollution during pregnancy. In recent years, several researchers investigated the effect of air pollution exposure on ASD.^{13, 26-33} In California Volk et al examined the relationship between air pollution due to traffic and autism in two population based case-control studies from the Childhood Autism Risks from Genetics and the Environment (CHARGE) study. The first study reported increased risk among those living within 309 m (~1010 feet) of a major freeway; especially during the third trimester.²⁶ The second research revealed the correlation of increased autism risk with exposure to model indicators of traffic-related air pollutants and regional level of NO₂, fine particulate matter PM2.5 and PM10.²⁷

Some other researchers found that living in areas with higher levels of styrene and chromium in the air was associated with increased risk of ASD.¹³ Additionally, autism has been associated with estimated regional concentrations of hazardous air pollutants, including arsenic and nickel, and with diesel PM exposure in early childhood. Toxicological studies suggested a biologically plausible role of air pollution in disrupting brain development and function during critical time points in gestation and early life. Diesel exhaust particles present in traffic-related pollution can disrupt endocrine activity and affect sexual differentiation transplacentally and alter cognitive function in mice.²⁶ From a recent study conducted on mice, researchers suggested air pollution as a risk factor not only for ASD but also for other neurodevelopmental disorders.³² As Bangladesh is a highly populated country with numerous industries and huge traffic in the major cities, lack strict monitoring of air pollution, people residing in the city vicinity are at high-risk to develop ASD.

Advanced paternal and/or maternal age can be a risk factor for ASD^13-18 and has emerged in our present study too. In 2015, a population-based study spanning five countries (Denmark, Israel, Norway, Sweden and Western Australia) with 57,66,794 children including 30,902 cases of ASD, researchers found that autism risk was associated with parental age and increasing age difference between parents.¹⁶ In another study on 2,53,347 participants, researchers found that both maternal and paternal age were independently associated with autism.¹⁵ Croen et al. found that risk of ASD was increased significantly with each 10 years increase in maternal and paternal age.¹⁴ According to some researchers, the father aged over 35 years⁴² and mother aged over 30¹ or 35 years²¹ were suspected as a high-risk condition. This delaying childbearing trend contributes around 4.6% increase in autism over the decade.²⁰

Some researchers found autism risk across generations epigenetically. A statistically significant nonlinear monotonic association was found between grand-paternal advanced age at the time of birth of the parent and autism risk in their grandchildren. The study was conducted on 5,936 cases and 30,923 controls and it was found that men who became the father at age over 50, were 1.79 times more likely to have a grandchild with autism, compared with who became the father at age 20 to 24 years.⁴³ DNA repair genes are down-regulated as males’ ages, and oxidative DNA damage level increases three times in spermatozoa when a man is over 35 compared to that of young aged men. Thus, paternal aging was known to be linked with several adverse clinical conditions including ASD.⁴² The study of Lampi et al supports premature birth as a significant risk factor for ASD.²⁵ In Bangladesh around 13 to 25% babies are reported to born preterm,^{44, 45} which is quite high and supports our finding that neurodevelopmental disorder may arise from preterm birth leading ASD. Some researchers suggested hypoxia during pregnancy¹³ and use of oral contraceptive⁴¹ as risk factors for ASD which also turned out as significant risk factors in our study.

Heavy metal poisoning can be triggered from different sources including chemical products, fertilizers, industrial paint, building materials, fish with a high level of mercury, silver containing dental fillings, nasal sprays and mercury-containing preservatives (thiomersal) in vaccines. Mohamed et al analyzing hair minerals (mercury, lead, and aluminum) of 100 autistic and 100 healthy children found that the levels were significantly higher among cases than controls. To track the sources of the metals they concluded that mercury, lead, and aluminum levels were positively correlated with maternal fish intake, residence near gasoline stations, and use of aluminum pans, respectively.³⁴ Another study conducted on 60 children (ASD=30 and controls=30), researchers got strong association and significant difference in the lead, mercury and cadmium concentrations between ASD and control.³⁵ Yassa, 2014 conducted a research on blood and hair samples obtained from 45 children with autism and 45 children served as controls between the ages of 2 and 10 years to find out exposure variation to lead and/or mercury as heavy metals. The results showed significant difference among the two groups, kids with autism had high level of mercury and lead.³⁶ In a study on 18 ASD participants, Geier et al., 2012 found that increasing hair Hg concentrations were significantly correlated with increased ASD severity.³⁷

Children being unable to differentiate between edible and non-edible item, may develop toxic lead levels as they get the poisoning from old houses with old painted wall or plumbing with cracks and dust which can be swallowed.^{36, 46} Mercury is found in many foods, including high fructose corn syrup. Dietary zinc is important in the metabolic processes to eliminate mercury; few artificial food color additives had shown to confer zinc deficiency. Dietary deficiencies of iron, zinc, iodine, selenium, copper, manganese, fluoride, chromium and molybdenum are linked with mild to considerable alterations in brain function resulting underdeveloped health, behavior and learning.¹ However, some researchers claimed a negative association between heavy metals and ASD ^47-49 which are reflected in our findings. This could be due to the less frequent use of canned or packed food and fish product in Bangladesh and comparatively less industrial exposure compared to developed countries.

It is critical to identify the more vulnerable period of the fetus in terms of immunity. But infection driven inflammation can increase the risk of ASD.^{21, 22} In a study on 407 cases and 2,075 controls, no overall association was reported between any maternal infection during pregnancy and ASD in the child. But they found women with bacterial infections, were at increased risk of delivering a child with ASD.²³ Atladottir et al. conducted a study on children in Denmark from 1980, through 2005.⁵⁰ They found no association between maternal infection and ASDs, but viral infection and bacterial infection were found to be connected to ASD in the offspring in the first trimester and in the second trimester respectively. They also reported urinary tract infections as the most common bacterial infection during hospitalization.⁵⁰ Maternal fever during pregnancy has also been linked to increased risk of ASD.⁵¹ However, in our study we found maternal infection as a non-significant risk factor for ASD.

Although carefully designed, surveyed and analyzed, the limitation of this preliminary study was small sample size; and a larger sample from multiple cities is needed to improve the statistical power of our findings. The participants were mainly from urban or semi-urban areas, due to the fact that specialized school was situated in cities and this restricted the participation of ASD patients mainly from urban areas. This should be noted that environmental risk factors may vary in villages and can have a different impact on ASD due to comparatively less urbanization in rural areas. To answer some questions, respondent had to answer based on memory, and depending on the age of their child the information might be slightly inaccurate as they had difficulties to remember exact information after several years.

Conclusion

The age of the mother, premature birth, air pollution, the age of the father, hypoxia during childbirth and oral contraceptive were found as significant risk factors for ASD in Bangladesh compared to control group. In addition, exposure to five and six factors on an individual was found to be statistically significant risk factors for ASD development and progression.

Acknowledgement

The authors cordially thank the authority of Sarwar Autistic Children’s Welfare Organisation, Khulna and Swedish Autism Center, Comilla, Bangladesh for cooperation in data collection. The authors would like to thank Mr. Molla Mohammad Shafiqur Rahman, Associate Professor, Environmental Science Discipline, Khulna University, Bangladesh for his cordial support by sharing his experiences and thoughtful discussions. The authors also thank Mr. Devojyoti Dutta, Biotechnology and Genetic Engineering Discipline, Khulna University for his assistance in collecting data from Comilla, Bangladesh.

References

1. Ratajczak HV. Theoretical aspects of autism: Causes- A review. J Immunotoxicol. 2011; 8: 68-79.

2. Mullick M, Rahman W, Alim S, Chowdhury H. A neuroimaging study in childhood au-tism. Bangabandhu Sheikh Mujib Med Univ J. 2013; 6: 121-26.

3. Diagnostic and statistical manual of mental disorders (4th ed): American Psychiatric Association; 1994.

4. Diagnostic and statistical manual of mental disorders (DSM-5®): American Psychiatric Association; 2013.

5. Community report from the autism and develop-mental disabilities monitoring (ADDM) network. Centers for Disease Control and Prevention, 2012.

6. World autism awareness day-"Autism-courageous march of life for peace and happiness". Department of social services, Ministry of Social Welfare, Government of the People's Republic of Bangladesh, 2013.

7. Global Autism Movement and Bangladesh: Centre for research and information; 2014 [cited 2014 2 April 2016]. Available from: http://cri.org.bd/2014/09/03/global-autism-movement-and-bangladesh/.

8. Woods AG, Wormwood KL, Wetie AGN, Aslebagh R, Crimmins BS, Holsen TM, et al. Autism spectrum disorder: An omics perspective. Proteomics Clin Appl. 2015; 9: 159-68.

9. Gurrieri F, Neri G. Defective oxytocin function: A clue to understanding the cause of autism? BMC Med. 2009; 7.

10. Barnard RA, Pomaville MB, O'Roak BJ. Mutations and modeling of the chromatin remodeler CHD8 define an emerging autism etiology. Front Integr Neurosci. 2015; 9: 1-13.

11. Willsey AJ, Sanders SJ, Li M, Dong S, Tebbenkamp AT, Muhle RA, et al. Coexpression net-works implicate human midfetal deep cortical projection neurons in the pathogenesis of autism. Cell 2013; 155: 997-1007.

12. Terrone G, Cappuccio G, Genesio R, Esposito A, Fiorentino V, Riccitelli M, et al. A case of 14q11. 2 microdeletion with autistic features, severe obesity and facial dysmorphisms suggestive of Wolf–Hirschhorn syndrome. Am J Med Genet A. 2014; 164: 190-93.

13. Talbott EO, Marshall LP, Rager JR, Arena VC, Sharma RK, Stacy SL. Air toxics and the risk of autism spectrum disorder: The results of a population based case-control study in southwestern Pennsylvania. Environ Health. 2015; 14: 80.

14. Croen LA, Najjar DV, Fireman B, Grether JK. Maternal and paternal age and risk of autism spectrum disorders. Arch Pediatr Adolesc Med. 2007; 161: 334-40.

15. Durkin MS, Maenner MJ, Newschaffer CJ, Lee LC, Cunniff CM, Daniels JL, et al. Advanced parental age and the risk of autism spectrum disorder. Am J Epidemiol. 2008; 168: 1268-76.

16. Sandin S, Schendel D, Magnusson P, Hultman C, Suren P, Susser E, et al. Autism risk associated with parental age and with increasing difference in age between the parents. Mol Psychiatr. 2015; 21: 693-700.

17. Hultman C, Sandin S, Levine S, Lichtenstein P, Reichenberg A. Advancing paternal age and risk of autism: New evidence from a population-based study and a meta-analysis of epidemiological studies. Mol Psychiatr. 2011; 16: 1203-12.

18. Reichenberg A, Gross R, Weiser M, Bresnahan M, Silverman J, Harlap S, et al. Advancing paternal age and autism. Arch Gen Psychiatry. 2006; 63: 1026-32.

19. Mullen BR, Khialeeva E, Hoffman DB, Ghiani CA, Carpenter EM. Decreased reelin expression and organophosphate pesticide exposure alters mouse behaviour and brain morphology. ASN Neuro. 2013; 5: 27-42.

20. Shelton JF, Geraghty EM, Tancredi DJ, Delwiche LD, Schmidt RJ, Ritz B, et al. Neurodevelopmental disorders and prenatal residential proximity to agricultural pesticides: The CHARGE study. Environ Health Perspect. 2014; 122: 1103-09.

21. Meldrum SJ, Strunk T, Currie A, Prescott SL, Simmer K, Whitehouse AJ. Autism spectrum disorder in children born preterm-role of exposure to perinatal inflammation. Front Neurosci. 2013; 7: 123.

22. Ornoy A, Weinstein-Fudim L, Ergaz Z. Prenatal factors associated with autism spectrum disorder (ASD). Reprod Toxicol. 2015; 56: 155-69.

23. Zerbo O, Qian Y, Yoshida C, Grether JK, Van de Water J, Croen LA. Maternal infection during pregnancy and autism spectrum disorders. J Autism Dev Disord. 2015; 45: 4015-25.

24. Bauer AZ, Kriebel D. Prenatal and perinatal analgesic exposure and autism: An ecological link. Environ Health. 2013; 12: 1-13.

25. Lampi KM, Lehtonen L, Tran PL, Suominen A, Lehti V, Banerjee PN, et al. Risk of autism spectrum disorders in low birth weight and small for gestational age infants. J Pediatr. 2012; 161: 830-36.

26. Volk HE, Hertz-Picciotto I, Delwiche L, Lurmann F, McConnell R. Residential proximity to freeways and autism in the CHARGE study. Environ Health Perspect. 2011; 119: 873.

27. Volk HE, Lurmann F, Penfold B, Hertz-Picciotto I, McConnell R. Traffic-related air pollution, particu-late matter, and autism. JAMA Psychiatry. 2013; 70: 71-77.

28. Roberts AL, Lyall K, Hart JE, Laden F, Just AC, Bobb JF, et al. Perinatal air pollutant exposures and autism spectrum disorder in the children of Nurses' Health Study II participants. Environ Health Perspect. 2013; 121: 978-84.

29. Lyall K, Schmidt RJ, Hertz-Picciotto I. Maternal life-style and environmental risk factors for autism spectrum disorders. Int J Epidemiol. 2014; 43: 443-64.

30. Dickerson AS, Rahbar MH, Han I, Bakian AV, Bilder DA, Harrington RA, et al. Autism spectrum disorder prevalence and proximity to industrial facilities releasing arsenic, lead or mercury. Sci Total Environ. 2015; 536: 245-51.

31. Raz R, Roberts AL, Lyall K, Hart JE, Just AC, Laden F, et al. Autism spectrum disorder and particulate matter air pollution before, during, and after pregnancy: A nested case-control analysis within the Nurses' Health Study II Cohort. Environ Health Perspect. 2015; 123: 264-70.

32. Allen J, Oberdorster G, Morris-Schaffer K, Wong C, Klocke C, Sobolewski M, et al. Developmental neurotoxicity of inhaled ambient ultrafine particle air pollution: Parallels with neuropathological and behavioral features of autism and other neurodevelopmental disorders. Neurotoxicology 2015 (in press).

33. Windham GC, Zhang L, Gunier R, Croen LA, Grether JK. Autism spectrum disorders in relation to distribution of hazardous air pollutants in the San Francisco Bay area. Environ Health Perspect. 2006:1438-44.

34. Mohamed FEB, Zaky EA, El-Sayed AB, Elhossieny RM, Zahra SS, Salah Eldin W, et al. Assessment of hair aluminum, lead, and mercury in a sample of autistic Egyptian children: Environmental risk factors of heavy metals in autism. Behav Neurol. 2015; 2015: 1-9.

35. Tabatadze T, Zhorzholiani L, Kherkheulidze M, Kandelaki E, Ivanashvili T. Spectrumhair heavy metal and essential trace element concentration in children with autism disorder. Georgian Med. 2015; 248: 77-82.

36. Yassa HA. Autism: A form of lead and mercury toxicity. Environ Toxicol Pharmacol. 2014; 38: 1016-24.

37. Geier DA, Kern JK, King PG, Sykes LK, Geier MR. Hair toxic metal concentrations and autism spectrum disorder severity in young children. Int J Environ Res Publ Health. 2012; 9: 4486-97.

38. Grant WB, Soles CM. Epidemiologic evidence for supporting the role of maternal vitamin D deficiency as a risk factor for the development of infantile autism. Dermato-Endocrinology. 2014; 1: 223-28.

39. Duan X, Jia F, Jiang H. Relationship between vitamin D and autism spectrum disorder. Chin J Contemp Pediatr. 2013; 15: 698-702.

40. Kocovska E, Fernell E, Billstedt E, Minnis H, Gillberg C. Vitamin D and autism: Clinical review. Res Dev Disabil. 2012; 33: 1541-50.

41. Strifert K. An epigenetic basis for autism spectrum disorder risk and oral contraceptive use. Med Hypotheses. 2015; 85: 1006-11.

42. Gavriliouk D, Aitken RJ. Damage to sperm DNA mediated by reactive oxygen species: Its impact on human reproduction and the health trajectory of offspring. Adv Exp Med Biol. 2015; 868: 23-47.

43. Frans EM, Sandin S, Reichenberg A, Långström N, Lichtenstein P, McGrath JJ, et al. Autism risk across generations: A population-based study of advancing grandpaternal and paternal age. JAMA Psychiatry. 2013; 70: 516-21.

44. Baqui AH, Rosen HE, Lee ACC, Applegate JA, El Arifeen S, Rahman SM, et al. Preterm birth and neonatal mortality in a rural Bangladeshi cohort: Implications for health programs. J Perinatol. 2013; 33: 977-81.

45. Shah R, Mullany LC, Darmstadt GL, Mannan I, Rahman SM, Talukder RR, et al. Incidence and risk factors of preterm birth in a rural Bangladeshi cohort. BMC Pediatr. 2014; 14: 1-11.

46. Geier DA, Kern JK, Garver CR, Adams JB, Audhya T, Nataf R, et al. Biomarkers of environmental toxicity and susceptibility in autism. J Neurol Sci. 2009; 280: 101-08.

47. McKean SJ, Bartell SM, Hansen RL, Barfod GH, Green PG, Hertz-Picciotto I. Prenatal mercury exposure, autism, and developmental delay, using pharmacokinetic combination of newborn blood concentrations and questionnaire data: A case control study. Environ Health. 2015; 14: 1-12.

48. Yau VM, Green PG, Alaimo CP, Yoshida CK, Lutsky M, Windham GC, et al. Prenatal and neonatal peripheral blood mercury levels and autism spectrum disorders. Environ Res. 2014; 133: 294-303.

49. Rahbar MH, Samms-Vaughan M, Dickerson AS, Loveland KA, Ardjomand-Hessabi M, Bressler J, et al. Blood lead concentrations in Jamaican children with and without autism spectrum disorder. Int J Environ Res Publ Health. 2015; 12: 83-105.

50. Atladóttir HÓ, Thorsen P, Østergaard L, Schendel DE, Lemcke S, Abdallah M, et al. Maternal infection requiring hospitalization during pregnancy and autism spectrum disorders. J Autism Dev Disord. 2010; 40: 1423-30.

51. Zerbo O, Iosif A-M, Walker C, Ozonoff S, Hansen RL, Hertz-Picciotto I. Is maternal influenza or fever during pregnancy associated with autism or developmental delays? Results from the CHARGE (CHildhood Autism Risks from Genetics and Environment) study. J Autism Dev Disord. 2013; 43: 25-33.