Visual outcome of mega-dose intravenous corticosteroid treatment in non-arteritic anterior ischemic optic neuropathy - retrospective analysis

Abstract

Background: To date, non arteritic anterior ischemic optic neuropathy (NAION) is still incurable. We wish to evaluate the effect of intravenous (IV) corticosteroids on the visual outcome of NAION patients.

Methods: Visual parameters were retrospectively compared between NAION patients treated with IV corticosteroids and untreated NAION patients (control). Visual acuity (VA) and Humphrey automated static perimetry visual field (VF) defects of the affected eye were compared between groups at baseline, 1, 3, 6 months, and end of follow-up visits. The VF analysis consisted of number of quadrant involvements and mean deviation (MD).

Results: Each group comprised 23 patients (24 eyes). Mean initial VA was similar in the control and treatment groups (p = 0.8). VA at end of follow-up did not improve in either groups (p = 0.8 treated group, p = 0.1 control group). No improvement and no difference in VF defects were found by either quadrant analysis (p = 0.1 treated group, p = 0.5 control group) or MD analysis (p = 0.2, treated group, p = 0.9 control group). VA and VF parameters tended to be worse in the treated group, although without statistical significance.

Conclusions: Our results suggest that IV corticosteroids may not improve the visual outcome of NAION patients. Since intravenous corticosteroids could potentially cause serious adverse effects, this treatment for NAION is questionable.

Figure 1

Visual acuity (VA) in LogMAR units in the control group (upper) and IV methylprednisolone treated group. Value above lines represents p-value of mean LogMAR VA compared to mean baseline VA at different time points. Baseline mean VA was the same in both groups: Log MAR 0.54 (equals to 20/70). The final VA was not statistically different form baseline in both groups: 0.43 (equals to 20/53) in the control group (p = 0.1 compared to baseline) and 0.61 (equals to 20/80) in the treated group (p = 0.8 compared to baseline). Note a very mild trend toward VA improvement in the control group versus worsening in the treated group, although statistically not significant.

Figure 2

Humphrey visual field (HVF) analysis in quadrants (upper graphs) and according to the mean deviation (MD) (lower graphs). Value above columns represents the p-value of mean quadrant involvement (upper graphs) and mean MD value (lower graphs) compared to baseline. Note that in both groups there was no change in the number of quadrant involvement at any time point. Analysis by MD showed no statistically significant difference in both groups between baseline and final values (p = 0.9 for controls, p = 0.2 for treated group). Note a mild trend toward improvement in the control group versus worsening in the treated group in the MD analysis. This trend was not shown in the analysis by quadrants involvement. *After bonferroni correction for multiple comparisons: p = 0.12.