Predictors of Visual Outcome After Acute Optic Neuritis

Abstract & Commentary

By Erik J. Kobylarz, MD, PhD, Assistant Professor of Neurology and Neuroscience, Weill Cornell Medical College, Cornell University. Dr. Kobylarz reports no financial relationship relevant to this field of study.

Synopsis: Visual afferent indices at baseline and one month following an acute optic neuritis event can predict visual dysfunction at six months.

Source: Kupersmith MJ, et al. Visual function at baseline and 1 month in acute optic neuritis: predictors of visual outcome. Neurology 2007; 69:508-514.

Kupersmith and colleagues used the optic neuritis Treatment Trial (ONTT) database to evaluate visual afferent indices at baseline and one month after an episode of acute optic neuritis, which could predict abnormal vision at six months. Specifically, they computed correlations between the degree of visual loss at baseline, one month, and the change in visual function from baseline to one month, with visual acuity, contrast sensitivity, and threshold visual field after six months.

Data from 426 ONTT patients who had baseline and six-month vision evaluations were analyzed in this study. Abnormal vision for each measure was defined using previous ONTT criteria (i.e., visual acuity less than 20/20, contrast sensitivity less than 1.6 log units, and visual field mean deviation less than -3.00 dB). A combined variable was created that indicated the number of visual measures (0 to 3 for visual acuity, contrast sensitivity, and visual field) that showed moderate-to-severe abnormalities at baseline and one month. For each of the three visual measures and the combined variable, the positive predictive value for visual abnormality at six months was computed for various cutpoints at baseline and at one month. For selected cutpoints, 95% confidence intervals were computed for the positive predictive values, and sample sizes were computed for a 50% relative reduction in the proportion of patients who would have abnormal visual function at six months.

They found that reducing the baseline and one month cutpoint (indicating worse vision) increased the proportion of patients with vision abnormalities at six months, but also reduced the number of eyes meeting the criterion. Due to substantial improvement during the time period from baseline to one month, the number of eyes for each moderate-to-severe abnormal cutpoint level was much less than at baseline. However, almost all eyes with moderate-to-severe abnormal vision at one month had abnormal vision at six months. No change in any visual measure from baseline to one month provided any more than weak predictability of having abnormal vision at 6 months. As described in their earlier reports,1 no baseline amount of loss of any single vision measure or any combination of measures was predictive of having moderately to severely abnormal vision at 6 months.


The ONTT has yielded a wealth of important data regarding the natural history, visual recovery, and treatment of optic neuritis. In this article, Kupersmith et al have provided a novel analysis of these data that can be used for counseling patients on expected visual outcome and for designing studies to test therapies that might offer neuroprotection and/or induce remyelination for patients with acute optic neuritis, including those who are at high risk of permanent optic nerve damage.

The authors note that almost all patients with moderate-to-severe vision loss after one month had abnormal contrast sensitivity at six months; slightly fewer had abnormal visual fields; and the fewest, but still a significant number, had abnormal visual acuity. However, baseline moderate-to-severely abnormal visual acuity, contrast sensitivity, and threshold perimetry were not predictive of having moderate-to-severe visual deficits at six months. They correctly point out that these results must be viewed in the context of the ONTT eligibility criteria, in particular the occurrence of a first episode of optic neuritis in the study eye and evaluation within eight days of the onset of visual symptoms.

It is possible that other more newly developed types of visual afferent testing, such as alternative methods for measuring contrast sensitivity; optical coherence tomography of the peripapillary retinal nerve fiber layer;2 visual evoked potentials, including multifocal visual evoked potentials;3 serum biomarkers of axonal degeneration; or MRI of the optic nerve, could at present or in the future more accurately determine which patients are most likely to have permanent deficits in visual function after optic neuritis. These evolving and potentially more quantifiable measures might serve as more precise indicators of permanent deficits than conventional testing of visual acuity using standard letter charts, contrast sensitivity charts, or visual field assessment using a 24-2 or 30-2 strategy.

Further studies are warranted to determine which vision indices, including those derived from more newly developed types of studies, accurately predict longer-term outcome (i.e., > 6 months) after acute optic neuritis, as well as the efficacy of emerging therapies.


1. Beck R, et al. Ophthalmology 1994; 101:1771-1778.

2. Trip SA, et al. Ann Neurol 2005; 58:383-391.

3. Hood D, et al. Invest Ophthalmol Vis Sci 2000; 41:4032-4038.