To evaluate the effectiveness of oral appliances (OAs) for managing patients with obstructive sleep apnea (OSA).
PubMed, Embase, Web of Science, CENTRAL and SIGLE were electronically searched from January 1980 to September 2015 for randomized or nonrandomized controlled trials that assessed the effectiveness of OAs on OSAS. The processes of study search, selection, data extraction, assessment of risk of bias and evaluation of evidence quality were conducted independently by two reviewer authors. Meta-analyses were performed in Review Manager 5, Stata11.0 and StatsDirect 2.7.9.
Finally, we included 17 eligible studies which compared OAs and placebo or blank control. Six outcomes were assessed in this meta-analysis, i.e., apnea hypopnea index (AHI), respiratory arousal index (RAI), minimum oxygen saturation(MinSaO2), rapid eye movement (REM) sleep, sleep efficiency and Epworth Sleepiness Scale (ESS). Meta-analysis revealed that the pooled mean differences were -10.26 [95% CI: (-12.59, -7.93)], -9.03 [95% CI: (-11.89, -6.17)], 3.08 [95% CI: (1.97, 4.19)], 0.36 [95% CI: (-0.30, 1.02)], 1.34 [95% CI: (-0.05, 2.73)] and -1.76 [95% CI: (-2.57, -0.94)], respectively. The sensitivity analysis and subgroup analysis displayed generally robust results except for MinSaO2, REM sleep and Sleep efficiency. Furthermore, publication bias was detected in RAI and MinSaO2.
The available evidence indicates benefits in respiration and sleep quality with oral appliances as compared to placebo devices or blank control, while we cannot determine its effectiveness in sleep efficiency and sleep architecture alterations. However, due to low evidence quality as revealed by GRADE, this finding should be interpreted with caution.
Through critical meta-analyses, we found that oral appliances are effective in respiration improving and sleep quality. The existing evidence supports the employment of OAs as a recommendable treatment option for OSA. This meta-analysis helps to direct clinical practice and future research, and promises to be of great interest for both practitioners and researchers.