Abstract
Purpose :
Large ocular adduction tethers the ON to deform the optic disc and peripapillary region. However, local strains in the ON have not heretofore been measured. We developed an automated method to enhance MRI resolution and parameterize ON path for in vivo measurement of local ON strains.
Methods :
MRI was taken using surface coils in 2mm thick axial (resolution 390mm) and quasi-coronal (213mm) planes in 24 eyes of 12 normal adults (mean age 59±16 years) during target-controlled central gaze, large (~28°), and small (~24°) ductions. A convolutional neural network identified the ON, globe, extraocular muscles, and orbital wall in the images. Identified pixels in image stacks were converted to 3D point sets that were registered across gaze positions. Parameterization made ON geometry continuously well-defined, permitting computation of displacements and strains along the ON in 5 contiguous segments of equal length. Statistical analysis employed general estimating equations.
Results :
In both adduction and abduction, the anterior ON experienced the greatest displacement. However, ON strain depended fundamentally on duction direction. For both large and small adduction, strain was 4-5% and did not vary significantly with anteroposterior location along the entire ON length (P>0.9). In contrast, strain during large and small abduction was <1% adjacent the globe and increased monotonically with distance posteriorly(P<0.05), but did not vary significantly with duction size. In context of solid mechanics, this difference suggests that the ON behaves like a stretching truss in adduction and a bending beam in abduction.
Conclusions :
In abduction, ON strain is due to bending and is least near the optic disc. Since ON strain during adduction is due to stretching and is uniform along its length adduction that tethers the ON transmits much larger loading to disc and peripapillary region than does abduction. This finding supports the putative association between adduction tethering and optic neuropathy.
This abstract was presented at the 2023 ARVO Annual Meeting, held in New Orleans, LA, April 23-27, 2023.