Identification of obstructive level is crucial for successful surgical outcomes in patients with obstructive sleep apnea (OSA). Unfortunately, most of the dynamic airway evaluations are performed for a short duration under drug-induced sleep; therefore, it is uncertain whether they represent airway events that occur during a whole night of sleep. This study was aimed to evaluate the correlation between obstructive levels that were identified by a short-time and a whole-night test in patients with OSA. Total 101 patients with OSA underwent drug-induced sleep fluoroscopy (DISF) and pressure manometry (PM). For DISF, the obstructive pattern was classified into one of three groups: soft palate, tongue-based, and a combined obstruction. PM was used to measure the proportion of retroglossal events out of total whole-night obstructive events in each patient. The mean age of the patients was 43.8 years. The obstructive pattern was identified as soft palate in 56 patients, combined in 38 patients, and tongue-based in 7 patients following DISF. Results from PM showed that the mean percentage of retroglossal obstructive events was 31.2 ± 30.7%. The average proportion of retroglossal obstructive events that were identified by PM in patients with soft palate, combined, and tongue-based obstruction was 27.2%, 32.1%, and 59.0%, respectively (p = 0.033). There are limitations of evaluating obstructive events that occur during a whole night with short-time tests. Surgeons should be aware the possibility of disagreement in the obstructive level between short-time and whole-night tests.