P100 amplitude of 15 minute checks in the NMOSD-ON group had a significant reduction at 6 months relative to the IDON group, suggesting more several axonal damage.
Findings from a Chinese-based, case-control study suggest that patients with neuromyelitis optica spectrum disorder with optic neuritis (NMOSD-ON) have more severe axonal damage than those with idiopathic demyelinating optic neuritis (IDON), as demonstrated by differences in abnormal visual evoked potential (VEP) response pattern.1
All told, normal VEP responses of 15-minute checks in the IDON group (n = 26) were significantly more frequent than in the NMOSD-ON group (n = 58) at 3 months (24.0% vs 4.5%; P = .017), while no significant group difference in a normal VEP response of 60-minute checks was found at any of the follow-ups (1, 3, and 6 months). Between the 2 groups, the frequency of previous episodes was significantly higher in those with NMOSD-ON (34.6% vs 6.9%; P = .003). Additionally, the study showed that 15-minute checks were more sensitive than 60-minute checks in demonstrating the differences in VEP abnormalities.
VEP is widely used in ON, reflecting demyelination and axonal damage in the visual pathway; however, little had been previously known about VEP changes in patients with NMOSD-ON who’ve had an acute attack. Led by senior investigator Hui Yang, MD, professor, Sun Yat-Sen University, the study included 84 adults with either NMOSD-ON or IDON who were within 30 days of an acute attack, had no relapse in at least the 90 days prior to the attack, had no brain lesions or myelitis, and had at least 1 episode of clinical ON.
All patients underwent best-corrected visual acuity (BCVA) and full-field pattern reversal VEP recordings at the onset, 1 month, 3 months, and 6 months. Full-field monocular stimulation by pattern reversal black or white checkerboards was performed at a viewing distance of 100 cm. All patients received treatment with high-dose intravenous methylprednisolone followed by oral prednisone tablets tapering off after at least 6 months. In addition, prednisone and azathioprine were prescribed for maintenance treatment.
VEP patterns were divided into 5 types: normal wave, only decreased amplitude with normal P100 latency (A), only prolonged P100 latency with normal amplitude (L), decreased amplitude with prolonged P100 latency (AL), and no wave. VEP patterns with A responses were found to be rare, occurring in 0-5.3% of those in the NMOSD-ON group and 0-4.5% of those in the IDON group. L pattern response of 15-minute checks in the IDON group was significantly more frequent than in the NMOSD-ON group at 6 months (30.0% vs 57.8%; P = .048), while no statistical group difference in the L pattern response of 60-minute checks was found at any time point.
Annualized relapse rates were elevated compared with the prepregnancy period, especially during the initial 3 months after delivery, whereas EDSS scores worsened during pregnancy and the postpartum period.
P100 amplitude of 15-minute checks in the NMOSD-ON group showed no significant improvement at each follow-up, whereas that in the IDON group increased significantly at 1 month (P = .008) and 3 months (P = .002). Despite this, investigators observed significantly lower P100 amplitude of 15-minute checks in the NMOSD-ON group than those with IDON at 6 months (P = .034). Between the 2 groups, the P100 latency of 60-minute checks showed P values of 0.776, 0.138, and 0.067, at 1, 3, and 6 months respectively, which were not significant. Similarly, nonsignificant P values of 0.074, 0.435, and 0.102, were found for P100 latency of 15-minute checks.
LogMAR BCVA in the NMOSD-ON group negatively correlated with P100 amplitudes of both the two checks (P <.001), whereas that in the IDON group only showed a negative correlation with P100 amplitudes of 60-minute checks (P <.001) and did not correlate with P100 amplitudes of 15-minute checks (P = .208). In comparison with the standardized partial regression coefficient (SPRC) of 60-minute checks (SPRC = –0.416) and 15-minute checks (SPRC = –0.317), the P100 amplitude of 60-minute checks had a relatively greater influence on logMAR BCVA than that of 15-minute checks.
