1239
Brain Connectivity and Autistic Traits Moderated by Olfactory Perception among Children and Adolescents with Congenital Heart Defects1University of Pittsburgh, Pittsburgh, PA, United States, 2IBM T.J. Watson Research Center, Ossining, NY, United States, 3Child Mind Institute, New York, NY, United States
Synopsis
We investigate the relation between congenital heart disease (CHD), autism spectrum disorder (ASD), and olfaction in a cohort of children and young adults using fcMRI (Functional Connectivity Strength; FCS, and amygdala seed-based connectivity) and DTI (RD). ASD severity was positively correlated with FCS in widespread regions, with amygdala connectivity to the DMN, and with RD in posterior parietal white matter; however, these relations were stronger in individuals with impaired olfaction scores and individuals with CHD. The relation between impaired olfaction, CHD, and ASD may be the result of individuals with impaired olfaction and/or CHD having different brain structure-function relationships.
Introduction
Recent research has shown increased risk of autism spectrum disorder (ASD) in individuals with congenital heart disease (CHD)1. Additionally, individuals with ASD display impaired olfactory processing, indicating that olfaction may be an important clinical marker for ASD2. Yet, the neuronal basis for these associations remains poorly understood. Here we explore the relation between CHD, ASD, and olfaction in a cohort of children and young adults using resting-state functional connectivity MRI (fcMRI) and diffusion tensor imaging (DTI).Materials and Methods
Participants: Participants with and without CHD were recruited prospectively at UPMC Children’s Hospital of Pittsburgh (CHP). CHD lesion types were heterogenous, including atrial/ventral septal defects, transposition of the great arteries, hypoplastic left heart syndrome, and heterotaxy. fcMRI: Data was successfully collected from 96 participants (age = 14.7 +/- 4.1 yrs, 46M, 50F, 38 CHD patients, 58 controls). DTI: Data was successfully collected from 130 participants (age = 13.9 +/- 4.1 yrs, 70M, 60F, 53 CHD patients, 77 controls).MRI Scans: All data was acquired on a 3T Siemens Skyra scanner at CHP. fcMRI: TR = 2s, TE = 30 ms, two scan runs of 5 minutes each, voxel size = 4 mm isotropic. DTI: 42 diffusion-weighted acquisitions at b = 1000 s/mm2, 7 b=0 acquisitions, voxel size = 2 mm isotropic. Standard preprocessing methods were used for fcMRI3 with volume scrubbing and for DTI (FSL) with slice dropout and eddy current correction. All successful fcMRI acquisitions contained at least 5 minutes of data combining the two runs. Both functional connectivity strength (FCS) maps and seed-based connectivity maps from the amygdalae were computed. (FCS for a gray matter voxel is the average Pearson correlation between its voxel time course and all other gray matter voxels, negative values thresholded to zero.)
Autism Rating Scale: All participants were administered the ASRS tests measuring Autistic Trait measures (Self-Regulation, Attention, Sensory Sensitivity).
Olfaction: All participants were administered the NIH Toolbox Odor ID. The fully corrected score (correcting for age and SES) was used.
Analyses: Voxelwise GLM analyses were conducted for the fcMRI and the DTI maps, testing for main effect of fcMRI/DTI metric, metric-X-olfaction interaction, and metric-X-olfaction-X-CHD interaction, with participant sex and age included as covariates of no interest. Standard methods were used to correct for multiple comparison; all maps are significant at FWE-corrected p < 0.05.
Results
Autism/Olfaction: ASRS scores were significantly different between CHD patients and controls (p ≤ 0.001); however olfaction scores were not significantly different between groups or correlated with ASRS measures.FCS: Results show a widespread positive correlation between FCS measures and autism severity (Figure 1, left), supporting previous studies4 showing that ASD is, at least in part, a hyperconnectivity disorder. However, regionally specific interactions with olfaction were found (Figure 1, middle) indicating this effect is more severe for individuals with impaired olfactory processing. Significant three-way CHD-X-olfaction interactions were also found (Figure 1, right), indicating this effect is yet more severe for CHD patients as compared to controls.
Seed-based connectivity: Results show positive correlation with autism severity and functional connectivity between the amygdala and the default mode network (DMN) but negative correlation with autism severity and functional connectivity between the amygdala and the salience and executive control networks (Figure 2, left). Interactions with olfaction were found (Figure 2, middle) in the DMN, indicating this effect is more severe for individuals with impaired olfactory processing. Three-way CHD-X-olfaction interactions were also found (Figure 2, right) indicating this effect is more severe for CHD patients as compared to controls.
DTI: Results show positive correlation between autism severity and radial diffusivity (RD) mainly in posterior parietal white matter (Figure 3, left). Interactions with olfaction were found (Figure 3, middle) in the same regions and three-way CHD-X-olfaction interactions (Figure 3, right) were also found indicating a more severe effect in CHD patients and individuals with impaired olfaction.
Discussion
Our results suggest that the relation between impaired olfaction and ASD is, at least partly, the result of individuals with impaired olfaction having different brain structure-function relationships, making them more susceptible to ASD. This phenomenon is even more prevalent in individuals with CHD. This phenomenon has likely strong genetic influences5 and research is ongoing regarding ASD phenotype rescue via restoring correct expression of autism candidate genes and use of neuro-hormones such as oxytocin. This research is likely to be of special benefit to CHD patients at increased risk for ASD.Conclusion
FcMRI and DTI metrics that correlate with ASD interact with olfaction and CHD such that the effect is stronger in individuals with impaired olfaction and who are CHD patients. Similar brain neural networks may underly autistic behavior and olfaction perception in youths with CHD. As such, measures of olfaction may aid clinicians identifying autistic traits in this population by stratifying patients into high- and low-risk groups. Results show the importance of amelioration of impaired social olfactory processes as a therapeutic target.Acknowledgements
No acknowledgement found.References
1. Sigmon ER, Kelleman M, Susi A, Nylund CM, Oster ME. Congenital Heart Disease and Autism: A Case-Control Study. Pediatrics. 2019 Nov;144(5):e20184114. doi: 10.1542/peds.2018-4114. Epub 2019 Oct 10. PMID: 31601611.
2. Alessandro Tonacci, Lucia Billeci, Gennaro Tartarisco, Liliana Ruta, Filippo Muratori, Giovanni Pioggia & Sebastiano Gangemi (2017). Olfaction in autism spectrum disorders: A systematic review, Child Neuropsychology, 23:1, 1-25, DOI: 10.1080/09297049.2015.1081678
3. Power JD, Mitra A, Laumann TO, Snyder AZ, Schlaggar BL, Petersen SE. Methods to detect, characterize, and remove motion artifact in resting state fMRI. Neuroimage. 2014;84:320-41.
4. Supekar K, Uddin LQ, Khouzam A, Phillips J, Gaillard WD, Kenworthy LE, Yerys BE, Vaidya CJ, Menon V. Brain hyperconnectivity in children with autism and its links to social deficits. Cell Rep. 2013 Nov 14;5(3):738-47. doi: 10.1016/j.celrep.2013.10.001. Epub 2013 Nov 7. PMID: 24210821; PMCID: PMC3894787.
5. Hartig R, Wolf D, Schmeisser MJ, Kelsch W. Genetic influences of autism candidate genes on circuit wiring and olfactory decoding. Cell Tissue Res. 2021;383(1):581-595. doi:10.1007/s00441-020-03390-8
Figures
