Clinical efficacy and safety of a light mask for prevention of dark adaptation in treating and preventing progression of early diabetic macular oedema at 24 months (CLEOPATRA): a multicentre, phase 3, randomised controlled trial.

backgroundWe aimed to assess 24-month outcomes of wearing an organic light-emitting sleep mask as an intervention to treat and prevent progression of non-central diabetic macular oedema.

methodsCLEOPATRA was a phase 3, single-blind, parallel-group, randomised controlled trial undertaken at 15 ophthalmic centres in the UK. Adults with non-centre-involving diabetic macular oedema were randomly assigned (1:1) to wearing either a light mask during sleep (Noctura 400 Sleep Mask, PolyPhotonix Medical, Sedgefield, UK) or a sham (non-light) mask, for 24 months. Randomisation was by minimisation generated by a central web-based computer system. Outcome assessors were masked technicians and optometrists. The primary outcome was the change in maximum retinal thickness on optical coherence tomography (OCT) at 24 months, analysed using a linear mixed-effects model incorporating 4-monthly measurements and baseline adjustment. Analysis was done using the intention-to-treat principle in all randomised patients with OCT data. Safety was assessed in all patients. This trial is registered with Controlled-Trials.com, number ISRCTN85596558.

findingsBetween April 10, 2014, and June 15, 2015, 308 patients were randomly assigned to wearing the light mask (n=155) or a sham mask (n=153). 277 patients (144 assigned the light mask and 133 the sham mask) contributed to the mixed-effects model over time, including 246 patients with OCT data at 24 months. The change in maximum retinal thickness at 24 months did not differ between treatment groups (mean change -9·2 μm [SE 2·5] for the light mask vs -12·9 μm [SE 2·9] for the sham mask; adjusted mean difference -0·65 μm, 95% CI -6·90 to 5·59; p=0·84). Median compliance with wearing the light mask at 24 months was 19·5% (IQR 1·9-51·6). No serious adverse events were related to either mask. The most frequent adverse events related to the assigned treatment were discomfort on the eyes (14 with the light mask vs seven with the sham mask), painful, sticky, or watery eyes (14 vs six), and sleep disturbance (seven vs one).

interpretationThe light mask as used in this study did not confer long-term therapeutic benefit on non-centre-involving diabetic macular oedema and the study does not support its use for this indication.

fundingThe Efficacy and Mechanism Evaluation Programme, a Medical Research Council and National Institute for Health Research partnership.