Can drug-induced sleep endoscopy improve the success rates of tongue base surgery?

Background: The purpose of this study was to determine the therapeutic value of drug-induced sleep endoscopy (DISE) by comparing the outcomes of tongue base surgery based on Muller's maneuver (MM) and those based on DISE in obstructive sleep apnea (OSA) patients. Methods: Ninety-five patients who underwent the tongue base surgery in combination with palatal surgery for OSA at a tertiary referral hospital between March 2012 and March 2019 were enrolled in this retrospective comparative study. Forty-seven patients underwent MM for surgical decision and 48 patients underwent DISE in addition to MM for surgical decision. Surgical success was defined according to the Sher criteria (postoperative apnea-hypopnea index [AHI] < 20/h and ≥ 50% reduction in preoperative AHI), and AHI improvement (%) was defined as (preoperative AHI-postoperative AHI) × 100/preoperative AHI. For comparison between the MM and DISE groups, p-values were calculated using independent or paired t-tests for continuous variables and using chi-square test for categorical variables. Results: By comparing the results of MM and DISE, consensus on the tongue base level showed insignificant concordance (kappa = 0.017, p = 0.865), whereas that on the oropharynx level showed fair agreement (kappa =0.241, p = 0.005). AHI, supine AHI, rapid eyeball movement (REM) AHI, non-REM AHI, and nadir oxygen saturation were all significantly improved after the tongue base surgery in both groups. The MM group showed a significant improvement in the Epworth sleepiness scale after the tongue base surgery (p = 0.014), whereas the DISE group did not (p = 0.165). However, there was no significant difference in the AHI improvement (MM group = 47.0 ± 32.0, DISE group = 48.3 ± 35.4, p = 0.852) and surgical success (MM group = 42.6%, DISE group = 45.8%, p = 0.748) between the groups. Tonsil grade (p < 0.05) and occlusion at the oropharynx lateral wall (p = 0.031) were significantly related to surgical success in the MM group. Conclusions: In the judgment of the tongue base surgery, MM and DISE findings showed poor agreement. DISE might affect the surgical decision on the tongue base surgery in OSA patients; however, there was a lack of evidence regarding the superiority of DISE over MM with respect to the surgical outcomes.