Purpose The aim of this randomized clinical trial was to evaluate the effects of tooth tissue-borne (TTB), tooth-borne (TB), and bone-borne (BB) rapid maxillary expansion appliances on obstructive sleep apnea (OSA) severity. Trial design Three-arm parallel randomized controlled trial. Methods This study was designed in parallel with an allocation ratio of 1:1:1. Forty six patients with narrow maxilla and diagnosis of OSA recruited from the Department of Orthodontics, ### University were randomly assigned to three groups according to appliance used: tooth tissue-borne, tooth-borne and bone-borne expanders. The primary outcome of this study included polygraphic change in sleep parameters. Secondary outcome was the correction of posterior crossbite. Each subject underwent overnight sleep test with polygraphy at baseline and 3 month-follow-up of treatment. Randomization was performed using a computer-generated randomization program The outcome assessor was blinded to group assignment. For the statistical analysis, Kruskal-Wallis analysis and Dunn-Bonferroni tests were used for inter-group comparisons and Wilcoxon analysis was used for intra-group evalaution. p<0.05 was accepted statistically significant. Results The amount of expansion in maxillary width and upper intermolar width were similar in all goups (95% confidence interval [CI], p>0.05). The groups were similar in terms of apnea-hypopnea index (AHI) and oxygen saturation parameters at baseline (95% [CI], p>0.05). After 3 months of treatment, there was no significant decrease in AHI and oxygen desaturation index, and no increase in minimum and mean oxygen saturations (95% [CI], p>0.05). Supine AHI values were decreased by the tooth tissue-borne and tooth-borne appliances, but these changes were not significant (95% [CI], p>0.05). Harms No serious harm ocurred except mild gingivitis. Conclusions Similar skeletal and dental expansion of maxilla were observed after RME with all expanders. Although the decrease in AHI was not significant, RME can be used as an adjunct to the primary treatment in OSA patients.