Green or yellow laser treatment for diabetic macular edema: exploratory assessment within the Diabetic Retinopathy Clinical Research Network.

*Department of Ophthalmology, Wilmer Eye Institute, Johns Hopkins University School of Medicine, Baltimore, Maryland; †Jaeb Center for Health Research, Tampa, Florida; ‡Department of Ophthalmology, Beetham Eye Institute, Joslin Diabetes Center, Harvard Medical School, Boston, Massachusetts; §Division of Epidemiology and Clinical Applications, National Eye Institute and the National Institutes of Health, Bethesda, Maryland; and ¶Department of Ophthalmology, Wake Forest University Eye Center, Winston-Salem, North Carolina.

Retina (Philadelphia, Pa.). 2013;(10):2080-8

Abstract

PURPOSE Explore differences in green compared with yellow focal/grid laser treatment on functional and anatomical endpoints in eyes with diabetic macular edema. METHODS Data from two randomized clinical trials were evaluated for differences in visual acuity and optical coherence tomography parameters for eyes assigned to sham injection + prompt laser, ranibizumab + prompt laser, or prompt laser only: among subgroups of eyes treated exclusively and electively with either green or yellow laser. RESULTS In the sham injection + prompt laser group, the mean visual acuity letter score change for eyes receiving green and yellow laser treatment, respectively, was +2.4 ± 14 and +5.1 ± 13 at the 52-week visit (P = 0.06) and +2.4 ± 15 and +6.0 ± 13 at the 104-week visit (P = 0.13), with no corresponding evidence of differences in optical coherence tomography thickness. When comparing wavelength groups in the ranibizumab + prompt laser and prompt laser-only groups, meaningful differences in visual acuity and optical coherence tomography thickness were not detected at 1 year or 2 years. CONCLUSION A trend toward improved vision outcome with yellow laser observed in one trial was not corroborated by anatomical outcomes or by the other trial. In this study, without random assignment to different wavelengths controlling for bias and confounding, it is not possible to determine whether one wavelength is better than the other.

Methodological quality

Metadata