Efficacy of bright light therapy for nonseasonal major depressive disorder: a meta-analysis of randomized controlled trials

Objective To explore the efficacy of (LT) for nonseasonal major depressive disorder with systematic review and meta-analysis. Methods Literature were systematically searched from the Pubmed, Embase, Web of science, cochrane library, China national knowledge infrastructure (CNKI), Wanfang data (Wanfang), China biology medicine disc (CBMdisc) till 28/02/2016 with keywords depress* mood affective light therapy light-therapy light treatment and phototherapy. Papers were also searched from the references of relevant published systematic reviews and meta-analyses. Studies were selected by screening with predefined inclusion and exclusion criteria. Change from baseline of depression rating scales was chosen as primary outcome, and response rate and remission rate were chosen as secondary outcomes. The strength of evidence was graded according to the criteria of grading of recommendations assessment development and evaluation (GRADE). Meta-analysis and subgroup analysis were performed in Revman 5.3. Results Twelve studies with 332 patients in LT group and 325 patients in control group were finally included in the meta-analysis. The heterogeneity between the included studies was relatively large. LT was superior to control in change from baseline scores of depression rating scales (SMD: 0.58, 95% CI 0.35-0.81, P<0.01), while no significant difference between the two groups was detected in the subgroup analysis of perinatal depression and peri-menopausal depression (SMD: 0.59, 95%CI -0.18-1.36, P=0.130). LT group achieved high response rate than control group in the overall meta-analysis (OR=2.51, 95%CI 1.65-3.81, P<0.01) and subgroup analyses (LT combination with antidepressants, OR=2.62, 95%CI 1.23-5.57, P=0.01, I2=50%; LT alone, OR=2.37, 95% CI 1.11-5.07, P=0.03, I2=0%). Nevertheless, no significant difference was found between LT and control in the subgroup analyses of remission rate (LT combination with antidepressants, OR=2.46, 95% CI 0.64-9.39, P=0.19; LT alone, OR=2.49, 95% CI 0.91-6.82, P=0.08). GRADE assessment of the evidence indicated that the evidence was moderate for response rate, and low for change from baseline scores of depression scales and remission rate. Conclusions LT is effective for nonseasonal major depressive disorder as a whole, the null results of some subgroup analyses may be due to the small sample size of these subgroups and insensitivity of remission rate in detecting difference between groups. Key words: Depressive disorder; Systems analysis; Light