Laser photocoagulation for treating neovascular age-related macular degeneration

Fifteen trials were included in the review (2064 participants). Three types of photocoagulation were used in the trials: direct photocoagulation of the entire CNV (11 trials), perifoveal photocoagulation (one trial) and grid photocoagulation (three trials). In 12 trials the control group was observation only. One trial compared photocoagulation to submacular surgery and two trials compared different lasers. Data on the progression of visual loss could be extracted from five of the eight trials of direct photocoagulation of the CNV versus observation. The treatment effect was in the direction of harm in all studies at three months follow up (RR 1.41, 95% confidence intervals (CI) 1.08 to 1.82). After two years the treatment effect was in the direction of benefit (RR 0.67, 95% CI 0.53 to 0.83). These studies were clinically heterogenous with participants having CNV lesions in different locations and different baseline visual acuities. There was little evidence of statistical heterogeneity at three months but substantial statistical heterogeneity at two years. However, all treatment effects in the individual trials were in the direction of benefit. One study comparing perifoveal photocoagulation or observation of subfoveal CNV found benefits that were statistically significant only at two years (RR 0.36, 95% CI 0.18 to 0.72). Other comparisons did not demonstrate differences.