ABSTRACT
Purpose: Many studies have assessed the changes of retinal nerve fiber layer (RNFL) thickness in patients with obstructive sleep apnea syndrome (OSAS), but the results were inconsistent. Therefore, a meta-analysis was performed to evaluate the RNFL changes in OSAS measured in vivo.
Methods: Pertinent studies were identified by a comprehensive search of PubMed, Embase, Web of science, Cochrane library, Scopus, and Chinese biomedical disc databases from inception to August 2016. A fixed effects model was used to pool the weighted mean difference (WMD) and 95% confidence interval (CI) between OSAS group and control group.
Results: Seventeen studies were included in the final analysis, with 12 for descriptive analysis and 5 for meta-analysis, involving a total of 1757 eyes (1106 in the OSAS group and 651 in the control group). The RNFL in OSAS was significantly lower than control group, with pooled WMD −3.53 (95%CI: −4.80 to −2.26, P < 0.001) for average RNFL, −3.69 (95%CI:−5.49 to −1.89 P < 0.001) for superior RNFL, −4.66 (95%CI: −6.92 to −2.39, P < 0.001) for inferior RNFL, −3.15 (95%CI:−5.19 to −1.10, P = 0.003) for nasal RNFL, and −2.45 (95%CI: −4.59 to −0.31, P = 0.025) for temporal RNFL. Along with severities of OSAS, a trend of more profound reduction of average RNFL was observed in advanced OSAS, with WMD of average RNFL thickness −1.75 (95%CI:−4.47 to −0.98, P = 0.209) for mild OSAS, −3.54 (95%CI:−6.33 to −0.73, P = 0.013) for moderate OSAS, and −7.17 (95%CI:−10.00 to −4.34, P < 0.001) for severe OSAS. The majority of studies in the descriptive review demonstrated similar findings.
Conclusion: The OSAS was associated with a reduced RNFL in all quadrants compared to controls. Evaluation of RNFL may serve as a tool for grading severities of OSAS. Considering the limited evidence, the conclusions should be interpreted cautiously.
Funding
This study was supported by the National Natural Science Foundation of China (81570843). The sponsor had no role in the design or conduct of this research.
Declaration of interest
The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.