An experimental light mask, the Noctura 400 Sleep Mask (PolyPhotonix Medical) did not help control non-central diabetic macular oedema, in a recent trial.
In CLEOPATRA, which investigators believe to be the first phase 3 clinical trial of this approach, the light mask was no more effective than a sham mask in reducing retinal thickness.
"The analysis of compliance highlighted that wearing these light masks over 24 months might also not be a sustainable option, as compliance decreased over time," wrote Sobha Sivaprasad of Moorfields Eye Hospital, London, UK, and colleagues, in Lancet Diabetes & Endocrinology.
Diabetes can compromise the capacity of ocular blood vessels to supply oxygen to the retina. This stimulates the development of new blood vessels, which often leak, causing retinopathy and – when the central macula is affected – macular oedema, impairing vision.
Diabetic macular oedema outside the centre of the macula can progress to the centre. Current treatments, which include intravitreal injections of anti-vascular endothelial growth factor (VEGF) agents, steroids and laser treatments, are invasive.
Demand for oxygen in the retina is greatest during dark adaptation, when rod photoreceptors consume nearly all the oxygen available. This observation has led to the hypothesis that macular oedema could be alleviated by preventing dark adaptation, providing a non-invasive supplement to standard treatments.
The light mask is designed to provide 500-505 nm light during sleep without interfering with sleep. A proof-of-concept study using a chemoluminescent source in 12 patients for 3 months suggested the treatment was safe, acceptable to patients and improved both colour vision and microaneurysm count.
A follow-up study using light-emitting diodes to illuminate 1 eye in 40 patients with bilateral diabetic macular oedema showed an improvement in retinal function and a decrease in retinal thickness at 6 months.
Based on these findings, the Noctura 400 Sleep Mask received a CE mark for sale in the European Union. However, as far as Sivaprasad and colleagues could determine, it had yet to undergo a phase 3 randomized controlled trial.
So they recruited 308 patients with with type 1 or 2 diabetes mellitus and clinical and optical coherence tomography (OCT) evidence of retinal thickening in at least one noncentral ETDRS zone due to diabetic macular oedema with best-corrected visual acuity of more than 55 ETDRS letters, equivalent to 6/18 Snellen.
They assigned 144 to wear the light mast during sleep and 133 to wear a sham mask. The patients received no macular laser therapy, intravitreal steroids, or anti-VEGF agents for at least 4 months before randomization, but received standard treatment if their macular oedema worsened.
After 24 months, the maximum retinal thickness of the group wearing the light mask decreased by 9.2 Î¼m while for the sham mask it decreased 12.9 Î¼m. This difference was not statistically significant (P=0.84). At this point, the median compliance with wearing the mask was 19.5%.
There was also no statistically significant difference in the reduction in total retinal thickness, macular volume, progression of central subfield thickness to 300 Î¼m or more, or in the proportions of patients requiring treatment for new onset centre-involving diabetic macular oedema and of those treated with standard therapy during the trial due to worsening of diabetic macular oedema.
Discomfort on the eyes was the most frequent adverse event reported, affecting 14 with the light mask vs 7 with the sham mask. That was followed by painful, sticky, or watery eyes, reported by 14 with the light mask and 6 with the sham mask, and sleep disturbance, reported by 7 with the light mask vs 1 with the sham mask.
It is possible that other techniques of rod suppression in diabetic retinopathy and macular oedema might still succeed, the researchers wrote, pointing to a "growing body of scientific evidence that supports the role of photoreceptos in retinal vascular permeability and angiogenesis."