Oxytocin (OT) has been implicated to play an important role in autism spectrum disorders (ASD) etiology. We aimed to find out the differences in plasma OT levels between children with autism and healthy children, the associations of OT levels with particular autism symptoms and the associations of particular parental autistic traits with their ASD children OT levels.
We included 19 boys with autism and 44 healthy age-matched boys. OT levels were analyzed by ELISA method. Children with autism were scored by Childhood Autism Rating Scale and Autism Diagnostic Interview (ADI), adjusted research version. Autism Spectrum Quotient (AQ), Systemizing Quotient (SQ) and Empathizing Quotient were completed by parents of children with autism.
Children with autism had significantly lower plasma OT levels than controls. OT levels positively correlated with ADI Reciprocal Interaction and Communication scores. AQ and SQ of fathers positively correlated with children plasma OT level.
Our results support the hypothesis of OT deficiency in autism. The "paradoxical" associations of OT levels and social skills in children with autism indicate disturbances at various levels of OT system. We first reported associations of OT levels in children with autism and behavioral measures in fathers indicating that OT abnormalities stay between parental autistic traits and autism symptoms in their children.
Autism spectrum disorders (ASD) are neurodevelopmental disorders characterized by derangement in social-communicative abilities, emotional development, integrative and executive functions relatively to the development of other psychological functions. The symptoms include delayed/abnormal verbal and nonverbal communication, social contact avoidance and restricted or stereotyped play, interests and behavior. Autism is highly heritable with the 60–96% concordance in monozygotic twins compared to 0–23% in dizygotic twins.
Oxytocin (OT) has been implicated to play an important role in autism etiology due to its effects on emotional and social behavior.
Autism spectrum disorders are considered to be a continuum from mild autistic traits, distributed through the normal population, with the same cognitive determinants as autistic disorders.
To our best knowledge, no study has evaluated the OT associations with the particular traits of BAP within the non-autistic population or parents of children with ASD. The aforementioned studies' results of decreased plasmatic OT in children with autism and improvement of social skills after OT administration in both healthy and individuals with autism suggest that OT could be the linking component between parental autistic traits and child's ASD symptoms, particularly those including social dysfunction. We aimed to find out the associations of the particular autistic traits evaluated by AQ, SQ and EQ in parents with the OT levels in their ASD children.
Our study aimed to elucidate 3 questions: 1) Do plasmatic OT levels differ between children with autism and control healthy typically developing children? 2) Are plasmatic OT levels in children with autism associated with the severity of particular autism symptoms, and if yes, which of them? 3) Are plasmatic OT levels in children with autism associated with autistic traits of their parents?
We recruited 19 boys with the diagnosis of childhood autism and 44 healthy control typically developing boys between ages of 2 and 9. Subjects with autism and control subjects were group-matched for chronological age: boys with autism mean age=56.7 months, SD=25.4 months; control boys mean age= 58.9 months, SD=23.0 months. Children with autism were recruited from the local Autism Centre for children in Bratislava, Slovakia. All boys with autism were diagnosed as meeting criteria for ICD-10 childhood autism by a clinical child psychologist with the cooperation of a child psychiatrist, who confirmed all diagnoses. Only children with IQ≥70 and without any other psychiatric disorder were included into the study. We recruited only boys with autism classified as aloof according to the Wing's classification.
Parents of children with autism were recruited at the same time and place as their children during their visit in Autistic Centre after the diagnostic procedure of their child, during the four-month period September–December 2012. Fourteen mothers and 13 fathers participated in the study. Four mothers and 4 fathers did not assent to participate in the study, 2 fathers did not live with or did not contact their families, the mother of one boy was not alive. The study was explained to parents in Autistic Centre and the questionnaires were sent by e-mail to those consenting to the study. Parents were asked to send the completed questionnaires by e-mail. All participating parents send the completed questionnaires within 2 weeks. The age of mothers was: mean=34.6 years, SD=7.7 years, the age of fathers was: mean=38.6 years, SD=5.5 years. The education of 4 mothers was graduate, 8 had high school, and 2 elementary school education. Five fathers had graduate and 8 high school education. All parents had no history of psychiatric treatment, except one mother with depressive disorder according to her reference.
Written informed consent was obtained from all parents of participating children, parents of children with autism signed the written consent both for themselves and their children. The protocol was approved by the Ethics Committee of Faculty of Medicine, Comenius University. The study conformed to the code of ethics stated in the Declaration of Helsinki.
Three mL of venous blood were taken from children with autism in the local Neurological Centre for Children and Adults in Bratislava. Blood of control children was taken by their pediatrician. The blood samples of all children were obtained during the same daily interval from 8.00–9.00 am. Immediately after samples intake, blood was centrifuged. After 10 minutes centrifugation at 3000 RPM, plasma was taken with addition of protease inhibitor aprotinin (Sigma Aldrich). Plasma samples with aprotinin were stored at -20℃ until OT measurement. Plasma OT levels were measured using ELISA method according to manufacturer's instructions (Enzo Life Sciences). All OT measurements were performed in the laboratory of Institute of Physiology, Comenius University in Bratislava.
All children with autism were tested using The Childhood Autism Rating Scale (CARS) evaluated by child clinical psychologist with the cooperation of child psychiatrist and Autism Diagnostic Interview (ADI), the adjusted research version, which was filled by child psychiatrist as per interview with parents. Parents of children with autism were asked to fulfill Autism Spectrum Quotient (AQ), Systemizing Quotient (SQ) and Empathy Quotient (EQ) which were sent to them by e-mail.
CARS
ADI
AQ is a self-administered questionnaire designated to quantify the autistic traits in adults with normal intelligence. It has been shown to have a good test-retest and inter-rater reliability.
SQ and EQ are based on the Baron-Cohen' s empathizing-systemizing theory and the extreme male brain theory in autism.
SQ contains 60 items, 40 of them are systemizing, 20 are control items. The respondent can "definitely agree", "slightly agree", "slightly disagree" and "definitely disagree". The systemizing items can be scored 0, 1 or 2 points, the maximum possible score is 80.
EQ contains 60 items, 40 empathy and 20 control items. The respondent can "definitely agree", "slightly agree", "slightly disagree" and "definitely disagree". The empathizing items can be scored 0, 1 or 2 points, the maximum possible score is 80.
AQ, EQ and SQ questionnaires were downloaded from Autism Research Centre website
We used the two-tailed unpaired t-test with Welch's correction for group differences in plasma OT levels between children with autism and healthy children. We used Pearson correlation coefficients to correlate plasma OT levels with CARS, ADI and ADI subscales scores and plasma OT levels with AQ, SQ and EQ of parents. Linear regression was used to exclude possible confounding variables in significant correlations of parental behavioral scores with plasma OT levels in their children with autism. The extreme values exceeding mean±2 SD were excluded from correlation analyses. The significance level was set at p<0.05.
We found a high statistically significant difference in plasma OT levels between groups (t=3.71, p=0.0004). Descriptive statistics and group differences for OT levels are presented in
We excluded 1 child for OT extreme values (≥ mean±2 SD) from correlation analyses. We also did not include this value into the correlation analyses with parental behavioral measures. We found the significant positive correlation of plasma OT levels and ADI reciprocal interaction and communication subscales. We did not find the significant correlations of plasma OT levels nor with ADI speech/language and restricted/repetitive behavior subscale score, neither with the CARS score or ADI total score. Descriptive statistics of scores of children with autism and correlations with plasma OT levels are presented in
We excluded 1 father for the extreme values (≥ mean±2 SD) of the AQ imagination subscore and 1 mother for the AQ total score, AQ social skill, attention switching, imagination and communication subscore from correlation analyses. We found significant positive correlations of plasma OT levels in children and fathers' AQ and SQ scores. The linear regression showed the significant relationships between these variables (FAQ=8.258, pAQ=0.021; FSQ=7.324, pSQ=0.020). We did not find any significant correlation of children plasma OT levels and any of fathers' AQ subscale score, however there was the positive trend (p<0.09) for social skill, attention to detail and communication areas. We did not find any significant correlations between plasma OT levels and fathers' EQ score. We did not find any significant correlation between plasma OT levels and mothers' SQ and EQ score, however mothers' AQ score had a trend to positive relationship. We did not find the correlation of any of mothers' AQ subscales and children OT level. Descriptive statistics of parental measures and correlations with their children plasma OT levels are presented in
We found significantly lower plasma OT levels in the group of children with autism compared to age-matched healthy controls. We confirmed the results of previous studies
We found significant positive correlations of plasma OT levels in children with autism and fathers' SQ and AQ scores with the positive trend for social skill, attention to detail and communication areas. To our best knowledge, this is the first evidence of parental behavioral measures' associations with OT level of their children with autism. These findings support our hypothesis that OT is the linking component between the level of parental and children autistic traits, predominantly those concerning social abilities. Moreover, this relationship has the same pattern as the relationship between behavioral measures of children with autism, namely social interaction and communication and their OT level. The possible explanation of these associations includes the theory of the extreme male brain in autism
Furthermore, we found the positive correlation of children OT level and AQ and SQ only in fathers, not mothers. In spite of the fact that there was a positive trend for AQ in mothers, the relationship of maternal autistic traits and children OT level was still lower than that of paternal autistic traits. Moreover, we did not find the trend to correlation for any of mothers' AQ subscale as we did in fathers. This result can be associated with the previous finding of a study in which only father's (but not mother's) BAP severity predicted the severity of autistic symptoms of affected child.
Our results are preliminary and should be confirmed in higher number of children with autism and their parents. Except the small sample sizes, our study has several limitations which must be noted. First, we included only children classified as aloof according to the Wing's typology.
Despite the limitations, our pilot study brought the novel results. We brought the further evidence of decreased plasma OT levels in children with autism which supports the OT deficiency hypothesis in autism. We reported the "paradoxical" associations of OT levels and social skills in children with autism which could indicate the disturbances at various levels in OT endocrine system. We first reported the associations of OT levels in children with autism and behavioral measures in fathers only, namely AQ and SQ, which indicates that OT is the linking component between the level of parental autistic traits and social abilities of their children with autism, moreover supports the evolutionary hypothesis of intragenomic conflict between maternally and paternally expressed genes in autism. Further research with extended number of child-parents pairs has to be done to confirm our preliminary results.
Authors thank to all parents and their children for participation in the study. The study was supported by the following grants: APVV-0254-11, APVV-0253-10, VEGA- 1/0066/12 and UK 67/2013.
*t=3.76, p=0.0004
*p<0.05. ADI: Autism Diagnostic Interview, CARS: Childhood Autism Rating Scale
*p<0.05, †p<0.09 (trend). AQ: Autism Spectrum Quotient, EQ: Empathizing Quotient, SQ: Systemizing Quotient