Objectives/Hypothesis: To explore whether the variables resulting from anatomical and physiological examinations can be combined to predict the outcomes of velopharyngeal surgery for obstructive sleep apnea (OSA). Study Design: Prospective design with a retrospective review. Methods: A total of 119 patients with OSA received velopharyngeal surgery, including revised uvulopalatopharyngoplasty with uvula preservation and transpalatal advancement pharyngoplasty. The preoperative examinations of these patients, including polysomnography (PSG), physical examination, and three-dimensional computer tomography (3-D CT), were obtained for analysis. Results: The overall success rate was 62.2%. Three factors were found to be predictive in treatment outcomes (P<0.05). These were tonsil size, the percentage of time with oxygen saturation below 90% (CT90), and the vertical distance between the lower edge of the mandible and the lower edge of the hyoid (MH). After changing CT90 and MH into level variables, another regression analysis was performed and the result suggested that all three level variables could be included. A scoring system was then created based on these three variables and their odds ratio values. The total scores of all patients were calculated by the following equation: Total score=2.7 tonsil size (score) 12.2 CT90 (score) 11.6 MH (score). The differences in success rates among patients with total scores of <14, 14 to 17, 17 to 22, and >= 22 were all significant (P<0.05). Conclusion: The anatomy of the pharynx and the physiology of OSA are both important in deciding outcomes of velopharyngeal surgery. Variables that could represent these two aspects can be combined to better guide patient selections.