Lamina Cribrosa Pore Convexity Predicts Neural Tissue Mechanical Insult

Abstract

Purpose : It is widely considered that glaucomatous vision loss is due to retinal ganglion cell (RGC) axon damage triggered by high levels of IOP-induced mechanical insult within the tissues of the lamina cribrosa (LC). Our recent work (Voorhees AP, et al. IOVS 2016; 57:ARVO E-Abstract 4709) has suggested that the mechanical stretch within pores of the LC may vastly exceed the stretch of the LC as a whole (Fig 1). We hypothesized that the shape and size of the individual pores may predict both the magnitude and type of deformation in the pore.

Methods : We created microstructure-based models of 7 regions of sheep LC from 3 eyes across various locations, peripheral or central LC. We simulated an IOP increase of 30 mmHg. In total, deformation was predicted in 128 whole pores. We calculated maximum (volume-weighted 90^th percentile) stretch and compression and total area change for each pore. 4 shape parameters were studied as potential predictors: area, perimeter, aspect ratio and convexity. The best fitting linear fixed effect model, controlling for the region, for each deformation measure was determined by comparing the Akaike information criterion.

Results : A summary of the best fitting models is presented in Fig 2A. All slopes including those for interactions, were significantly different from zero (p<0.05). Maximum stretch was best predicted by convexity alone (Fig 2B). Maximum compression and area change were best predicted by the combination of aspect ratio, convexity and either area or perimeter.

Conclusions : There were strong associations between pore shape and the deformation within the pore. Surprisingly, pore convexity appears to be the best determinant of the maximum stretch. The aspect ratio of the pore also appears important in predicting the maximum compression and area change. Pore shape may be a structural marker of glaucoma sensitivity.

This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.

 

Fig 1. Non-linear, anisotropic, finite element models based on histology were used to predict stretch in the ONH at both the (A) mesoscale level and (B) microscale level. Mesoscale models predict maximum stretch of 1-4%, while microscale models predict stretch within LC pores up to 20%.

 

Fig 2. A) Shape parameters correlated with deformation. B) Convexity was strongly associated with maximum stretch across all seven regions. Example pores are shown at equal scale. Areas of pores near concave edges are often seen to stretch more than the rest of the pore. C) Definition of convexity.