Abstract
Purpose :
To evaluate retinal structure in patients with autism spectrum disorder (ASD) and its correlation with cognitive impairment and brain volume.
Methods :
This case-control study utilized data from the UK Biobank, including patients diagnosed with ASD and matched neurotypical controls. Exclusion criteria included history of neurodegenerative diseases, visual pathway pathology, retinal disorders, glaucoma surgery, high refractive error, or intraocular pressure <6 or >21 mmHg. Employing the Topcon Advanced Boundary Segmentation (TABS) algorithm, rapid automated retinal boundaries and sublayers segmentation was performed. Nine layers, including retinal nerve fiber layer, ganglionic cell layer, inner plexiform layer, inner nuclear layer, outer plexiform layer plus outer nuclear layer, photoreceptor inner segment, photoreceptor outer segment, retinal pigment epithelium, and choroidoscleral interface, were segmented. Cognitive impairment was assessed using four tests: pairs-matching, prospective memory, numerical/verbal reasoning, and reaction time. Brain imaging-derived phenotypes from the UK Biobank were leveraged for total brain, grey matter, white matter, and ventricular cerebrospinal fluid volumes derived from T1-weighted MRI. Subcortical volumes were obtained using Freesurfer ASEG. Generalized Estimating Equation models were employed.
Results :
We analyzed OCT images from 240 eyes (80 ASD patients,160 controls). ASD patients exhibited greater thickness in the inner retina (adjusted mean differences (aMD) 5.3 µm, 95%CI [2.1- 8.4], p=0.001) compared to controls. No correlation was found between inner retinal thickness and cognitive impairment. Brain MRI data showed positive association between inner retinal thickness and various brain volumes: grey matter (β=0.49, 95%CI [0.1- 0.8], p=0.004), lateral ventricles (β=0.53, 95%CI [0.2- 0.8], p=0.001), 3rd ventricle (β=0.53, 95%CI [0.2- 0.8], p=0.002), and 4th ventricle (β=0.46, 95%CI [0.1- 0.8], p=0.004). Ventricular cerebrospinal fluid volume positively correlated with inner retinal thickness (β=0.49, 95%CI [0.2- 0.8], p=0.001). The anterior part of corpus callosum showed the strongest associations with inner retinal thickness (β=0.73, 95%CI [0.3- 1.1], p<0.001).
Conclusions :
The inner retina may offer valuable insights into neurodevelopmental features in ASD. Future research should investigate the retinal structure and ASD pathology to enhance diagnosis and improve outcomes.
This abstract was presented at the 2024 ARVO Imaging in the Eye Conference, held in Seattle, WA, May 4, 2024.