Abstract
Purpose: :
Central retinal artery obstruction (CRAO) causes persisting retinal damage. No treatment is currently available to prevent visual loss.Hyperbaric oxygenization is safe and increases the quantity of O2 physically dissolved in the blood. Under these conditions, the choroidal vessels can provide O2 to the inner retina via diffusion. This is the rationale behind the use of hyperbaric oxygen therapy (HBOT) in CRAO. Encouraged by case reports indicating a favorable effect on visual acuity (VA), HBOT was routinely performed in CRAO patients at our center. Here we examine the outcome of this series.
Methods: :
We analyzed data from all patients diagnosed with CRAO who presented within 12 hours of onset of symptoms. All patients received HBOT and hemodilution therapy, unless they had systemic contra-indications to HBOT or the treatment facility was unavailable. If this was the case, patients received hemodilution therapy only and were regarded as controls. Data from patients with VA of 20/200 or below was included in the analysis. However, if cilioretinal vessels or arteritic occlusion were found or choroidal perfusion was absent, patients were excluded. Our HBOT protocol entails breathing 100% O2 over 110 minutes at 2,35 bar. Hemodilution targets a hematocrite <0,4. Snellen VA was converted to logMAR scale for statistical analysis.
Results: :
We could include 51 patients into the treatment arm; 29 patients served as controls. Baseline mean VA was counting fingers for HBOT-patients and 20/1000 for controls (p=0,108). Other variables (e.g. age, duration of symptoms) also did not differ significantly at baseline. Over the course of treatment, VA in the HBOT group improved by a mean of 3 lines (p<0,0001). In the control group, mean improvement was 1 line (p=0,232). The 3 month follow-up visit yielded similar data. On discharge, mean VA had improved to 20/260 (HBOT) and 20/800 (controls; p=0,504). Again, at 3 months similar results were obtained.
Conclusions: :
To our knowledge, no larger group of CRAO patients treated with HBOT has been reported in the literature. Data from our sample are consistent with earlier reports in that we saw a significant improvement of VA following HBOT, slightly higher than in standard treatment only. A randomized prospective trial would be required to confirm and refine these results. From our sample, the number of patients required per group to detect a mean difference in VA of at least 3 lines following treatment can be estimated to 60. In addition, given the high number of non-responders to HBOT, more understanding of the metabolic insults that result from CRAO (other than those triggered by hypoxia) is crucial.
Keywords: clinical (human) or epidemiologic studies: outcomes/complications • retina • oxygen