Results
Search terms “Pediatric,” “DISE,” “OSA,” “Tongue Surgery,” and their synonyms individually yielded tens of thousands of articles. There was significant filtration when combined in PubMed: ultimately a total of 89 articles were screened for relevance, and 49 were excluded. After identification of 40 potentially relevant articles, they were downloaded and reviewed. Eight articles met initial inclusion and exclusion criteria, but one paper18 was excluded due to duplicate patient data published by the same author (screening flowchart, Figure 1).
In total, seven articles were used in the systematic review (284 patients, age 9.5 ± 3.9
years, normal/overweight status) and six of the studies (271 patients, age 8.5 ± 3.8 years, majority normal weight status) were suitable for metanalysis.10–16 See Table 1 and S1 for a summary of the studies. Individual patient data tended to be reported in studies with lower population sizes (Esteller et al, Kanotra et al, Maksimoski et al, Ulualp et al); means and standard deviations were included.10,11,13,16 Demographic data was reported in most studies but not clearly linked to individual patient PSG outcomes. The majority of DISE-directed tongue surgery involved multi-level airway surgery (e.g. lingual tonsillectomy plus revision adenoidectomy).
He et al reported pre- and post- operative PSG data but did not specify which data corresponds to which type of surgery; however, the study did include responder and non-responder data directly linked to the type of surgery, so it was included in the systematic review but not primary outcome metanalysis.15 Only three of the seven studies (185 patients, 65% of pooled study population) documented post-operative PSG timeline with an average of minimum 2.1 months post-op.10,14,16