Oral appliances (OAs) have demonstrated efficacy in treating obstructive sleep apnea (OSA), but many different OA devices are available. The Japanese Academy of Dental Sleep Medicine supported the use of OAs that advanced the mandible forward and limited mouth opening and suggested an evaluation of their effects in comparison with untreated or CPAP. A systematic search was undertaken in 16 April 2012. The outcome measures of interest were as follows: Apnea Hypopnea Index (AHI), lowest SpO2, arousal index, Epworth Sleepiness Scale (ESS), the SF-36 Health Survey. We performed this meta-analysis using the Grading of Recommendations, Assessment, Development and Evaluation (GRADE) system. Five studies remained eligible after applying the exclusion criteria. Comparing OA and control appliance, OA significantly reduced the weighted mean difference (WMD) in both AHI and the arousal index (favouring OA, AHI: −7·05 events h−1; 95% CI, −12·07 to −2·03;P= 0·006, arousal index: −6·95 events h−1; 95% CI, −11·75 to −2·15;P= 0·005). OAs were significantly less effective at reducing the WMD in AHI and improving lowest SpO2 and SF-36 than CPAP, (favouring OA, AHI: 6·11 events h−1; 95% CI, 3·24 to 8·98;P= 0·0001, lowest SpO2: −2·52%; 95% CI, −4·81 to −0·23;P= 0·03, SF-36: −1·80; 95% CI, −3·17 to −042;P= 0·01). Apnea Hypopnea Index and arousal index were significantly improved by OA relative to the untreated disease. Apnea Hypopnea Index, lowest SpO2 and SF-36 were significantly better with CPAP than with OA. The results of this study suggested that OAs improve OSA compared with untreated. CPAP appears to be more effective in improving OSA than OAs.