Purpose: This study was intended to perform a cephalometric comparison between the patients with and without obstructive sleep apnea (OSA). The factors influencing the OSA in the lateral cephalogram were also investigated. Methods: Fifty patients who had visited the Sleep Disorder Clinic at the Ajou University Hospital and evaluated with the polysomnograph (PSG) and cephalogram, were included in the study. The patients had the apnea-hypopnea episode over 5 times per hour (apnea-hypopnea index <TEX>$[AHI</TEX><TEX>]</TEX><TEX>{\geq}5$</TEX>) were diagnosed as OSA after the overnight PSG. To evaluate the hard and soft tissue profiles, the cephalometric radiograms were taken at the maximal intercuspation. The correlation between the patient's age, height, weight, body mass index (BMI) and AHI was inspected in the OSA and control group. The difference between the OSA and control group was evaluated (Mann-Whitney U Test). The cephalometric influencing factors to OSA were analyzed (Pearson's correlation coefficient) statistically using SPSS statistics. Results: The OSA Group had a significantly higher BMI than the control group. The mean lower facial height (ANS-Me) was longer in the OSA group; however, statistically significant difference was not detected in the anteroposterior craniofacial measurements. The distance between mandibular plane and hyoid bone of the OSA group was significantly longer than that of the control group. The hyoid position (MP-Hyoid) had a positive correlation between AHI (P<0.001). However, the measurements of oropharyngeal airway were not different between the two groups. The hypothesis, that the antero-posteriorly narrow oropharyngeal airway may aggravate the airway resistance and give rise to a higher AHI, was rejected in the study. Conclusion: We suggest that the lateral cephalogram may be utilized as a useful method to evaluate OSA. The patients with a lower hyoid position can be expected to have higher risks of OSA. However, a comprehensive intraoral inspection, including the soft palate and tonsilar hypertrophy, is emphasized, as the lateral cepahlogram cannot visualize the oropharyngeal status completely.