Correlation between drug-induced sleep endoscopy criteria for patient selection and transoral robotic surgery outcomes in obstructive sleep apnea treatment

Abstract

Background: To correlate drug-induced sleep endoscopy (DISE) criteria adopted for patient selection to transoral robotic surgery (TORS) with the surgical outcomes in obstructive sleep apnea (OSA) treatment. Methods: Retrospective analysis of patients submitted to DISE as candidates to TORS for tongue base reduction (TBR) to treat OSA, from January 2016 to December 2016. Surgical treatment was proposed based on a DISE criteria to patients with apnea-hypopnea index (AHI) >15 events/hour, non-compliant to CPAP. The exam was performed prior to surgery, excluding patients with retroglossal collapse due to lack of muscular support, thus selecting exclusively patients with soft tissue hypertrophy. Surgical success was defined as postoperative AHI <15 events/hour with at least 50% AHI reduction. The Wilcoxon signed-rank test was used to compare preoperative and postoperative parameters. Results: One hundred and forty-six patients underwent DISE where 18 patients (12%) met the inclusion criteria. The average age was 38±9.3 years, BMI was 29.8±4.4 kg/m2 and baseline AHI was 32±15.4 events/hour. The postoperative AHI was 7.3±5.1 events/hour, achieving surgical success in 88.9% of cases (P=0.0001). The minimum oxygen saturation increased from 78±9.6 to 87±5.9 (P=0.0001). Patients with palatal AP collapse had greater improvement, AHI reduction from 33.5±15.8 to 5.9±3.1 (P=0.0001). Patients with complete circumferential collapse (CCC) at soft palate did not have a significant reduction in AHI, as baseline AHI was 20.7±0.5 and postoperative AHI was 18.4±3.9 (P=0.15). There was statistical significance comparing the postoperative AIH between patients with velopharyngeal anteroposterior (AP) collapse to those with palatal circumferential collapse (CC) (P=0.001). Conclusions: The DISE criteria adopted was able to identify the tongue base collapse due to lack of tongue support, using this information to contraindicate the surgical procedure. The better selection of patients with retroglossal obstruction exclusively by soft tissue hypertrophy can explain the good surgical outcome achieved.