Objective This study aimed to evaluate the influential factors on the morphological changes of upper airway caused by mouth opening (MO). Methods One hundred and thirty‐eight obstructive sleep apnea–hypopnea syndrome (OSAHS) patients were enrolled. Anthropometric and demographic data, Friedman tongue position (FTP), and tonsil scores were recorded. Overnight polysomnography and upper airway computed tomography scans under two conditions (mouth closed [MC] and MO) were acquired. Morphological parameters of upper airway were compared between MC and MO. Stepwise multiple linear regression analyses were performed with the variation ratio of upper airway parameters (Para‐VRs) from MC to MO as the dependent variable, with age, gender, body mass index, neck circumference, waist circumference, four mandibular indexes, net angle or amount of MO, FTP, and tonsil scores as the independent variables. Results Overall analysis and subgroup analyses based on OSAHS severity revealed that the minimal cross‐sectional area of oropharyngeal lumen (OXmin) significantly decreased ( P < 0.05) with MO, whereas the minimal cross‐sectional area of velopharyngeal lumen (VXmin) did not significantly change with MO ( P > 0.05). The net angle of MO or amount of MO combined with tonsil scores were identified to have significant positive correlation with EXP (OXmin‐VR), [OXmin‐VR was logarithmically transformed with an exponential function, EXP(n) = e n ]; FTP appeared to be more related to EXP (VXmin‐VR). Mouth opening induced a significant increase VXmin for patient subgroup with FTP grading I and a significant decrease VXmin for patient subgroup with FTP grading IV ( P < 0.05). Conclusion Wider MO combined with larger tonsils lead to narrower oropharyngeal airway. The relative position of tongue to soft palate is the main factor influencing the changes of velopharyngeal lumen with MO. Level of Evidence 4 Laryngoscope , 128:2902–2909, 2018