Plasma Exchange for Steroid-Unresponsive Optic Neuritis

Abstract & Commentary

By Marc Dinkin, MD, Assistant Professor of Ophthalmology, Weill Cornell Medical College. Dr. Dinkin reports no financial relationships relevant to this field of study.

Synopsis: This study shows improvement in visual acuity in some patients with steroid-unresponsive optic neuritis, but it is not clear if plasma exchange accounts for this improvement.

Source: Roesner S, et al. Treatment of steroid-unresponsive optic neuritis with plasma exchange. Acta Neurol Scand 2011; DOI: 10.1111/j.1600-0404.2011.01612.x.

Neuro-ophthalmologists encounter a great many conditions where diagnosis is firm, but treatments are of questionable value. Optic neuritis, an inflammatory demyelinating condition resulting in loss of visual acuity and loss of fields and color vision, is no exception, since more than 95% of patients improve to an acuity of > 20/40 at 1 year, whether or not they are treated with intravenous steroids or placebo. Indeed, the results of the Optic Neuritis Treatment Trial (ONTT) indicated that IV steroids resulted only in a speedier recovery of vision, without a significant difference in the end result.1 Nevertheless, IV methylprednisolone is commonly prescribed for acute optic neuritis, in part because it is felt that there may be a subset of patients too small to have affected the results of the ONTT for whom treatment is a game changer. The 5% of patients who did not improve to > 20/40 in the ONTT represents a group of patients whose outcome might be improved with more aggressive treatments early on, and some of those may eventually be found to have neuromyelitis optica (NMO).

The study by Roesner and colleagues focused on "steroid-unresponsive optic neuritis," defined here to include patients who received 3-5 days of 1 g of IV methylprednisolone at onset, and a second cycle of 2 g per day for 3-5 days 2 weeks later, without recovery to above 50% acuity in at least one eye. Building on a few prior studies that have shown plasma exchange (PE) to be of some benefit in patients with multiple sclerosis (MS) or severe optic neuritis, the authors performed a retrospective analysis of 23 patients with steroid-unresponsive optic neuritis who received plasma exchange. Ten patients had relapsing-remitting MS, one had NMO, and 12 presented with a clinically isolated syndrome of optic neuritis.

The authors found that 11/23 patients (48%) showed a significant improvement, five of whom improved to 30-85% of normal, and six who improved to > 85%. Overall, among the nearly half of the group that appeared to respond, acuity improved from a mean of 16% prior to PE to 45% afterwards and to 60% at the first follow-up visit (average 50 days after PE). There were a few cases of benign hypotension, but no serious side effects were reported.


This is an important study in that it is the largest review to date of plasmapheresis for steroid-unresponsive optic neuritis, and offers some encouragement that this treatment may affect the outcome in this subset of patients. As the authors correctly conclude, however, while the improvement in their population "might be due to PE," it is difficult to draw any firm conclusions. The study is retrospective, there is no placebo arm, and the outcomes are not rigorously compared to the natural history of patients with steroid-unresponsive optic neuritis who recently received a second course of IV steroids and are at a mean of 31 days out from their acute episode. The authors argue that patients with optic neuritis typically experience a rapid recovery within 2 weeks of symptom onset, suggesting that their patient group would not have likely improved during the time period observed without the addition of PE. However, in the ONTT, some patients experienced recovery of visual acuity between 1 and 3 months, which is the precise time that this cohort was followed. In fact, if one limits their observations to the group of patients who did not recover within the first month, as was done here, then the majority of those patients will go on to show significant improvement in the coming months, since 95% of patients reach a final acuity of 20/40 by one year. Among the patients in this study who showed improvement to 85% or more, it should be noted that only two patients improved to that degree within the 7 days following plasmapheresis (two others reached 85% at a 3 week follow-up and another at 180 days). Therefore, a PE effect based on improvement during the treatment cannot be concluded.

This study could have been strengthened by an analysis of other markers of optic nerve function such as low contrast visual function, color vision, and visual field perimetry. The authors might have discovered a higher rate of improvement if these modalities were included, although any such improvement would also be limited by the lack of a control group or comparison with the natural history of similar patients.

There is ample reason to suspect that PE may affect the outcome of some patients with demyelinating events, since a subset of patients show histopathological evidence of intra-lesion antibody and complement deposition. As the consequence of poorly recovered optic neuritis may include low vision and a profound change in quality of life, a large prospective, double-blind, placebo-controlled trial of plasmapharesis in patients without early improvement is warranted. Such a study should preferably include a multi-modality comparison of visual function including visual fields and color testing, as well as a comparison of axonal loss as reflected in retinal nerve fiber layer thinning measured by optical coherence tomography.


1. Optic Neuritis Study Group. Visual function 15 years after optic neuritis: A final follow-up report from the Optic Neuritis Treatment Trial. Ophthalmology 2008;115:1079-1082.