Optical Coherence Tomography Detected Axonal Loss but Failed to Identify Acute Optic Nerve Axonal Injury: A Diffusion MRI Study

Abstract

Purpose : Optical coherence tomography (OCT) is widely used to assess progressive retinal ganglion cell (RGC) and retinal nerve fiber layer (RNFL) thickness changes for pre-chiasm neuropathy. We have developed a diffusion basis spectrum imaging (DBSI) to assess coexisting optic nerve axonal damage and inflammation. The current study aims to determine the relationship between OCT-detected RGC/RNFL thickness changes and DBSI-derived metrics in pre-chiasm neuropathy.

Methods : C57BL/6 female mice underwent left optic nerve crush (ONC) (approximately 1-2 mm from the eyeball) for 20s. Right eyes served as control (CTL). Spectral domain OCT cross-sectional imaging was performed (around peripapillary region) prior to DBSI scans at 7 days (n=5), 1 (n=5) and 3 (n=6) months post-ONC. The 25-direction DBSI scan was performed on a 4.7-T scanner: TR = 1.5 s, TE = 35 ms, Δ = 18 ms, δ = 6 ms, maximal b-value = 2,200 s/mm², slice thickness = 1.0 mm, FOV = 22.5 × 22.5 mm², and matrix size = 192 × 192. DBSI-assessed axonal injury (λ_||), demyelination (λ⊥), axonal loss and inflammation were derived using a lab-developed software. After in vivo DBSI, tissues were kept for H&E (retina) or immunohistochemistry (IHC) for optic nerve

Results : Representative H&E images highlight injury evolution of RGC loss and RNFL thinning post-ONC (Fig. 1 A). Quantitative OCT results are consistent with H&E findings (Fig.1 B). Optic nerve IHC exhibits coexisting axonal pathologies and inflammation post-ONC (Fig. 1 C). The RGC/RNFL thickness does not correlate with DBSI λ_|| or restricted fraction suggesting that OCT fails to detect acute optic nerve axonal injury or inflammation (Fig. 2 B, D), while strongly correlating with optic nerve λ⊥ and axonal volume (Fig. 2 C, E). Results suggest that demyelination is secondary to traumatic axonal injury (Wallerian degeneration) and chronic axonal loss.

Conclusions : Our findings suggest that OCT-detected RGC/RNFL thinning may not reflect optic nerve axonal injury or inflammation. Thus, the notion that RGC/RNFL thinning may serve as a surrogate marker of pre-chiasm optic nerve injury.

This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.

 

Figure 1 H & E and OCT showed retinal injury (A, B); IHC confirmed coexisting optic nerve pathologies (C).

Black/White bar: 50 μm
Yellow bar: 10 μm

 

Figure 2 Correlations of RGC/RNFL thickness and DBSI metrics.