Individuals suffering from mild traumatic brain injury (TBI) frequently report vision problems despite normal visual acuity and fundus tests which highlight the need for advanced diagnostic methods. The researcher raised a question about mild traumatic brain injury (TBI) which is associated visual dysfunction diagnosis using a battery of assessments across the visual pathways.
A recent study published in JAMA Ophthalmology determined whether a machine-learning-assisted battery of assessments could enhance the diagnosis of visual dysfunction in patients with mild TBI.
This case-control study is prospective in nature and observational in design and was carried out between May 2018 and November 2021 at a level 1 trauma research hospital. Adult men and women were included in the eligibility criteria if they had the best correctable vision and normal fundus exploration. The individuals in the case were with the history of mild TBI whereas the control group was without a history of TBI. Other exclusion criteria were the presence of ocular, neurological, or psychiatric disease, moderate-severe TBI, recent TBI, metal implants, younger than 18 years, and pregnancy. Sex and age matching was done for cases and controls. Data analyses were done from July 2023 through March 2024.
The measures taken in a single visit session included the neurobehavioral symptom inventory, oculomotor function assessments, optical coherence tomography, contrast sensitivity tests, visual evoked potentials, visual field testing, and MRI scans.
Analyzed were a total of 28 participants (mean [SD] age, 35.0 [12.8] years; 15 males [53.6%]) with mild TBI and 28 controls (mean [SD] age, 35.8 [8.5] years; 19 females [67.9%]). For mild TBI, the differences were in prism convergence tests breakpoint (−8.38; 95% CI, −14.14 to −2.62; P = .008) and recovery point (−8.44; 95% CI, −13.82 to −3.06; P = .004). Those with mild TBI also had reduced contrast sensitivity levels and thus this reduced measure (−0.07; 95% CI, −0.13 to −0.01; P = .04) showed increased visual evoked potential binocular summation index levels (0.32; 95% CI, 0.02-0.63; P = .02). One of the subgroups displayed low thickness of the peripapillary retinal nerve fiber layer, increased size of the optic nerve/sheath and volumes of brain cortical tissue. Machine learning exposed subtle differences through the primary visual pathway, particularly optic radiation, and occipital lobe areas, regardless of visual symptoms.
This study consists of 28 people, where 28 others are controls, poor eye motor functions and/or primary visual pathways were manifested in 78% of the subjects with mild TBI. The results procured from machine learning identified the posterior visual pathway deficit in about 70% of mild TBI cases concerning other cases, irrespective of their self-report visual dysfunction.
Results suggest that diagnosis of mild TBI could potentially be enhanced by using a combination of assessments and implementing machine-learning approaches.
The research findings for this case-control study indicate that the visual system has been compromised in mild TBI patients even if it may not be apparent in self-reports of vision disturbance. These results point to the utility of a battery of assessments or machine-learning approaches for the accurate diagnosis of that population.
Reference: Rasdall MA, Cho C, Stahl AN, et al. Primary Visual Pathway Changes in Individuals with Chronic Mild Traumatic Brain Injury. JAMA Ophthalmol. November 27, 2024. doi:10.1001/jamaophthalmol.2024.5076
