Abstract
Purpose :
To detect traumatic brain injury by testing oculomotor function using a VR system.
Methods :
The protocol was approved by the ethics committee (University of Erlangen) and according to the Helsinki Declaration and GCP. The prospective clinical study included 30 healthy subjects and 30 patients with mild or moderate traumatic brain injury (TBI). The Glasgow Coma Scale (GCS) characterized the severity of TBI. We examined 11 patients with moderate TBI (GCS 10-12) and 19 patients with mild TBI (GCS 13-15). The mean age was in healthy subjects 24+-2, in the patient group 43+-22 years. The healthy subjects were measured once, patients with mTBI were examined immediately after hospitalization, after 1 week and after 2 months.
VR-system: The VR-Oculomotor Tester VR-OMT consisted of a MSI laptop (G series) and HTC-Vive VR glasses. VR-OMT showed four rotating balls as stereoscopic stimuli, as one of the four balls was presented by a different horizontal disparity relatively to the other three balls. This ball appears closer to the patient, when the patient was able to perform fusion. The four balls were presented in nine different positions of the visual field. The task was to select the closer ball by a controller. The horizontal disparity difference of the ball to see by the patient varied from 330 arcsec to 990 arcsec. The disparity differences of 330, 660, 990 arcsec were tested 5 times in 9 positions, leading to 135 measurements in total. The response time (RT), the percentage of correct answers were measured. A bivariate correlation between RT and the Glasgow Coma Scale was calculated by Spearman's rho and Kendall's tau coefficient.
Results :
The correlation between RT and the GCS was highly significant (p ≤ 0.001). RTs of moderate and mild TBI were significantly (p ≤ 0.001) longer than the RTs of healthy subjects. The RTs of moderate TBI were significantly (p ≤ 0.001) higher than the RT of mild TBI patients. Response times improved with time after the accident. RTs one week after accident were significantly (p ≤ 0.001) faster than immediately after hospitalization, and RT two months after accident were significantly (p ≤ 0.001) faster than one week after accident.
Conclusions :
Patients with traumatic brain injury showed impaired stereopsis. Measuring stereopsis in different positions of the visual field by using a VR glass seems to be an effective tool for a fast concussion assessment.
This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.