Combining retinal neuronal and microvascular measurements improves discriminative power for multiple sclerosis patients without previous optic neuritis

Abstract

Purpose: Quantitative thinning of optic nerves and changes in retinal microvasculature have been detected in multiple sclerosis patients without optic neuritis (MSNON). However, data on how both retinal neuronal and microvascular parameters jointly influence the diagnostic performance is lacking. To investigate the diagnostic performance of neuronal and microvascular changes to detect the presence of MSNON from normal controls. Methods: This is a cross‐sectional study of 51 MS participants (n = 76 eyes) and 71 age‐ and sex‐matched healthy controls (n = 117 eyes). Retinal macular ganglion cell complex (GCC) and retinal nerve fibre layer (RNFL) thicknesses and capillary densities from the superficial (SCP) and deep capillary plexuses (DCP) were obtained from the optical coherence tomography (Cirrus AngioPlex model HD‐OCT 5000). We also compensated RNFL thickness for ocular anatomical factors. Thickness and vascular measurements and their corresponding areas under the receiver operating characteristic curves (AUCs) were compared between groups. Results: All retinal structural and vascular parameters were significantly altered in MS (p ≤ 0.012). The best neuronal parameter for detecting MS was compensated RNFL from the optic nerve head region (AUC = 0.85), followed by GCC from the macular region (AUC = 0.79) while the best microvasculature parameter was the SCP (AUC = 0.66). Combining retinal neuronal and microvascular parameters improved its diagnostic performance for MS detection (AUC = 0.90; P < 0.001). The joint model including both neuronal and microvasculature significantly improved the discrimination between MS and normal compared with each parameter separately (p = 0.027). Conclusions: Combining retinal neuronal and microvasculature parameters improved the discrimination of MSNON. Further studies may be warranted to evaluate the clinical utility of retinal parameters in risk prediction for disease progression.