Does drug-induced sleep endoscopy change the surgical decision in surgically naïve non-syndromic children with snoring/sleep disordered breathing from the standard adenotonsillectomy? A retrospective cohort study.

backgroundAdenotonsillectomy is the most commonly performed operation for pediatric snoring/sleep disordered breathing (S/SDB). However, 20-40% of patients will fail to improve. Drug-induced sleep endoscopy (DISE) may provide a more individualized surgical plan and limit unsuccessful surgeries. The aim of this study was to assess the impact of DISE on surgical decision-making in surgically naïve children with S/SDB.

methodsA retrospective observational cohort study was undertaken at the Stollery Children's Hospital. Patients 3-17 years of age who underwent DISE-directed surgery for S/SDB between January 2009 and December 2015 were eligible. We excluded other indications for tonsillectomy and syndromic children. The primary outcome was the level of agreement between a DISE-based surgical decision and the reference standard based on the American Academy of Pediatrics (AAP) guidelines via un-weighted Cohen's kappa. Secondary outcomes included the frequency and type of alternate surgical targets identified by DISE. The agreement on tonsil size between in-office physical assessment and DISE was also calculated. The effectiveness of DISE-directed surgery on postoperative S/SDB was not investigated.

resultsFive hundred fifty-eight patients were included. DISE changed the surgical plan in 35% of patients. Agreement between DISE-based and AAP clinical practice guidelines-based management was low (κ = 0.354 +/- 0.021 [95% CI 0.312-0.395]). An alternate diagnosis or surgical target was identified by DISE in 54% of patients. There was moderate agreement on tonsil size (κ = 0.44 [0.33-0.55]) between DISE and in-office clinical assessment.

conclusionsThis is a first phase diagnostic study, which demonstrates that DISE affects decision-making in surgically naïve children with S/SDB in up to 35% of patients. It has utility in individualizing first stage surgical treatments as well as identifying alternate targets for further surgical or medical therapy, while potentially limiting unsuccessful surgeries. Further studies to examine the effect of DISE on surgical outcomes are required.