Study On The Identification And Assessment Of Children With Autism Spectrum Disorders Based On Neurochemical Indicators

Autism Spectrum Disorder(ASD)describes a kind of neurodevelopmental disorder which affects 1/59 of people all around the world.The etiology of ASD is not clear and it has no specific therapy.Early intervention can improve behavioral outcomes of patients instead of healing,which requires accurate and early identification.Up to now,diagnosis or screening is mostly made by behavioral observation according to diagnostic or screening tools,which relies on subjective judgment of evaluators and lacks objectivity.Identification by biomarkers is one of novel approaches.Previous studies firstly pointed to genes.However,single gene could explain less than for 20% of ASD cases.Diagnosis using multiple genes was accurate but costly.Some neuroactive compounds such as monoamines,hormones,amino acids,organic acids,oxidative stress metabolites were reported as potential biomarkers.But most of these reports focused on one single biomarker or several biomarkers from a same metabolic pathway,which was possibly confused with some other diseases.Therefore,follow-up studies tend to integrate biomarkers from multiple systems associated with ASD.Recently,up to 90% of ASD patients were reported to have gastrointestinal problems.This phenomenon drive researchers to pain attention to the role of the interaction between central nervous system(CNS)and enteric nervous system(ENS)in ASD.Gut bacteria produce neuroactive compounds to affect CNS through stimulating the afferent nerve endings of ENS.In immunoreaction,glucocorticoids were released to facilitate the interaction of brain and intestine.These findings formed the concept of Microbiota-Gut-Brain Axis(MGBA).The neurochemical compounds played key roles in signal transmission.Therefore,some neurochemical compounds from MGBA were selected to test the applicability in identification and prognostic evaluation for ASD.The main work is as follows.1.Determination of urinary cortisol and cortisone in children and its application in ASD identificationFollowing fiber-packed solid phase extraction(PFSPE)using polystyrene,urinary cortisol(F)and cortisone(E)were determined by high-performance liquid chromatography tandem mass spectrometry(HPLC-MS).The linear range,limit of detection(LOD),limit of quantification(LOQ)for F were 0.6-150 ng/m L,0.20 ng/m L and 0.65 ng/m L,while 0.8-200 ng/m L,0.30 ng/m L and 0.88 ng/m L for E.Cohorts of 50 ASD children and 58 typically developing(TD)children were recruited.Their first morning urine samples were collected and determined for F and E.The results showed significantly lower F and RUCC(the ratio of F to E)in ASD group.The receiver operating characteristic curve(ROCC)obtained a sensitivity and specificity of 0.845 and 0.800 on F,while 0.862 and 1.000 on RUCC.2.Determination of urinary amino acids in children and its application in ASD identificationFiltered using surfactant-free cellulose acetate microporous membrane,urine samples were injected into HPLC to detect amino acids(tyrosine,methionine,phenylalanine,valine,leucine,isoleucine,tryptophan).The linear range,LOD and LOQ were 0.1-100 ?g/m L,0.01-0.04 ?g/m L and 0.05-0.2 ?g/m L,respectively.The levels of all amino acids were significantly lower in ASD children.ROCCs drawn on these amino acids reached the highest sensitivity and specificity in tryptophan(0.948 and 0.900).A binary logistic regression(ASD/TD)analysis was carried out with amino acids as independent variables.Methionine,phenylalanine and tryptophan entered the final equation and generated a joint indicator.The sensitivity and specificity of the ROCC on amino acids joint indicator were 0.966 and 0.940,higher than each single amino acid and conventional diagnostic/screening tools.Adding cortisol into the binary logistic regression analysis as an independent variable,a new joint indicator combining amino acids and cortisol was generated.The sensitivity and specificity of the ROCC on amino acids-cortisol joint indicator raised to1.00 and 0.96.3.Analysis of urinary p-cresol in children and its application in ASD identificationUrine samples were pretreated by PFSPE(polypyrrole)and taken into HPLC to determine p-cresol.The linear range,LOD and LOQ were 0.2-300 ?g/m L,0.1 ?g/m L and 0.35 ?g/m L,respectively.Significantly higher levels of p-cresol were observed in ASD group.The ROCC on p-cresol obtained a sensitivity and specificity of 0.78 and 0.793.Adding p-cresol as an additional independent variable in the binary logistic regression,the regression equation was unchanged.4.Application of neurochemical indicators in prognosis evaluation for ASD childrenThree ASD children were recruited into executive function training project lasting for a month.Executive function and neurochemical indicators were measured before and after training.Elevation of executive function,amino acids and RUCC,accompanied by slight alteration of urinary cortisol,cortisone and p-cresol were observed after intervention.5.Validation of the identification models based on neurochemical indicatorsIdentification models by RUCC,amino acids joint identification,and amino acidscortisol joint identification were verified in 73 healthy babies and 53 healthy preschool children respectively.All babies were recognized as ASD by RUCC model,while over80% of babies were correctly distinguished by amino acids joint model and amino acids-cortisol joint model gained accuracies over 80%.In preschool children group,the accuracies of RUCC model,amino acids joint model and amino acids-cortisol joint model increased in turn(all over 90%).It was suggested the application of ASD identification models based on neurochemical indicators was limited by age,and identification based on multiple categories of indicators was superior.