Optic nerve head microcirculation in autosomal dominant optic atrophy and normal-tension glaucoma.

Abstract

Autosomal dominant optic atrophy (ADOA) is a hereditary optic neuropathy caused by mutations in the nuclear gene OPA1. Objective findings, such as from tonometry or optical coherence tomography (OCT), are very similar in ADOA and normal-tension glaucoma (NTG), and ADOA is thus usually distinguished from NTG based on subjective characteristics. Previously, we found that ocular blood flow and circumpapillary retinal nerve fibre layer thickness (cpRNFLT) are reduced in the temporal optic nerve head (ONH) of ADOA patients (Inoue et al. 2016). Thus, the present letter describes the results of our attempt to establish a new, objective, OCT- and blood flow-based method to differentiate ADOA and NTG patients. This study comprised 13 eyes of seven ADOA patients, diagnosed based on mutations in the OPA1 gene, and 14 specially selected NTG patients with temporal cpRNFLT resembling that of ADOA patients. The ADOA group included some patients studied by us in a previous paper (Inoue et al. 2016). CpRNFLT was measured with OCT (3D-OCT 2000; TOPCON Corporation, Tokyo, Japan). Mean blur rate (MBR), which represents blood flow, was measured with laser speckle flowgraphy (LSFG-NAVI, Softcare Co., Ltd, Fukutsu, Japan). The subjects were recruited at the outpatient clinic of Tohoku University Hospital. The clinical characteristics of the control, NTG and ADOA patients are summarized in Table 1. The ADOA and NTG patients differed in visual acuity and MD (p = 0.02 and p < 0.001, respectively), but no other clinical characteristics. Both groups had significantly lower cpRNFLT and tissue MBR than the controls. Temporal-quadrant tissue MBR was significantly lower in the ADOA patients than the NTG patients, despite the fact that temporal cpRNFLT was similar in the two groups (p = 0.01). Most importantly, a superior cpRNFLT ≥85 μm and temporal tissue MBR ≤6.5 AU allowed us to differentiate ADOA from NTG with high precision (sensitivity: 100%; specificity: 92.9%). Thus, we found that patients with ADOA had significantly lower MBR in the temporal ONH than patients with NTG, despite similar reductions in cpRNFLT. This may have been due to the anatomy of the optic nerve, especially the papillomacular bundle. The retinal ganglion cell (RGC) axons in this area are unmyelinated and thus require a high amount of energy to restore electrical potential and for axoplasmic transport. However, the RGCs in the papillomacular bundle have relatively few mitochondria and a low mitochondrial reserve, making them particularly susceptible to mitochondrial genetic dysfunction-induced apoptosis and subsequent reductions in blood flow (Carelli et al. 2004). Alternatively, ADOA-associated mitochondrial dysfunction may reduce blood flow via its effect on the capillary endothelium. Electron microscopy findings show that the mitochondria of the capillary endothelium become dysfunctional more frequently than those of myofibres (Sarnat et al. 2012). In infants with systemic mitochondrial disease, dysfunctional mitochondria in the capillary endothelium decrease angiogenesis, eventually leading to myofibre injury. Furthermore, some studies have suggested that the mitochondria have important actions in vascular endothelial cells in addition to ATP production (Quintero et al. 2006) and are crucial in maintaining endothelial cell viability (Goveia et al. 2014), suggesting that mitochondrial dysfunction in ADOA would affect blood flow. In conclusion, our results suggest that optic neuropathy in ADOA is associated with mitochondrial dysfunction, which impairs the vascular endothelial cells and causes decreased microcirculation. Clinically, measuring MBR in the ONH may thus be a useful complement to cpRNFLT as a differentiator of ADOA and glaucoma.