Abstract
BACKGROUND AND OBJECTIVES: To investigate the long-term effect of permanent demyelination on axonal attrition by examining an association between intereye asymmetry of the multifocal visual evoked potential (mfVEP) latency delay and subsequent thinning of retinal ganglion cell axons in patients with a long-standing history of unilateral optic neuritis (ON). METHODS: Only patients with a significant degree of chronic demyelination (intereye latency asymmetry >5 ms) were included in this study. The level of optic nerve demyelination was estimated at baseline by the latency delay of mfVEP, while the degree of axonal loss was assessed by thinning of the retinal nerve fiber layer (RNFL) thickness between baseline and follow-up visits. Low-contrast visual acuity (LCVA) was also evaluated at baseline and follow-up. Patients were examined twice with an average interval of 6.1 ± 1.4 years. RESULTS: From 85 examined patients with multiple sclerosis, 28 satisfied inclusion criteria. Latency of the mfVEP was delayed, and RNFL thickness was reduced in ON eyes compared with fellow eyes at both visits. There was significant correlation between latency asymmetry and baseline or follow-up intereye RNFL thickness asymmetry. Intereye asymmetry of LCVA at baseline correlated with baseline latency asymmetry of mfVEP and baseline asymmetry of RNFL thickness. Latency of the mfVEP in ON eyes improved slightly during the follow-up period, whereas latency of the fellow eye remained stable. By contrast, RNFL thickness significantly declined in both ON and fellow eyes during the follow-up period. The rate of RNFL thinning in ON eyes, however, was more than 2 times faster compared with the fellow eyes (p < 0.001). Furthermore, baseline latency asymmetry significantly correlated with the rate of RNFL thinning in ON eyes during the follow-up (p < 0.001), explaining almost half of the variability of temporal RNFL progression. For each millisecond of latency delay (i.e., ∼0.5 mm of demyelination along the optic nerve), temporal RNFL thickness was annually reduced by 0.05%. DISCUSSION: Our study provides clear in vivo evidence that chronic demyelination significantly accelerates axonal loss. However, because this process is slow and its effect is mild, long-term monitoring is required to establish and confidently measure the neurodegenerative consequences of demyelination.