Orthodontic risk factors for obstructive sleep apnea in childhood: a pilot study (ORFOS Project)

Abstract

Background: Orthodontic risk factors have mainly been discussed as consequences of, rather than as a reason for obstructive sleep apnea (OSA) in childhood. Although they have gained importance concerning OSA treatment options, they scarcely attract attention in screening or diagnosis. The ORFOS pilot study aims to test feasibility of an OSA screening method in the setting of an orthodontic clinic in order to confirm known risk factors and identify new orthodontic conspicuous parameters. In the long term, we plan to establish a nationwide pediatric OSA screening to see if the identified OSA risk factors are transferable to the ethnic background of Germany, with the objective to define OSA risk profiles. Methods: The study comprised 162 minor orthodontic patients between 6 and 16 years of age (median 13 years). To evaluate a comprehensive medical (sleep) history, 6 questionnaires were applied and a non-invasive, self-developed ear-nose and throat (ENT) examination was performed. Orthodontic parameters covering clinical history, intra- and extraoral examination, dental and cephalometric analysis were recorded. Using the PSQ, we divided the study sample into a sleep disordered breathing/OSA (42 children) and an unaffected control group (120 children). We tested the feasibility of examination and screening methods, as well as performed a statistical group comparison. Results: Of 188 children approached, 167 (88,8 %) participated with a drop-out rate of 0 %, 5 met exclusion criteria. Concerning time factor of our study model, probands and their parents had to consider approximately 20 minutes for questionnaires and the ENT examination in addition to their routine orthodontic appointment. 7 variables showed a missing n rate greater than 10 % concerning clinical history, 3 in ENT examination (missing data rate over 5 %). The group comparison revealed 8 intergroup differences (p < 0.1) in both, orthodontic clinical history plus examination, and dental cast models: The test group "snorers" had less often sucking habits (9.5 %) than controls "non-snorers" (24.2 %, risk ratio (RR) = 0.39). Concerning lip configuration, a lower median of lip configuration lower third of 15 mm/ 67 % was detected in the test group by contrast with a median of 18 mm/ 69 % in the reference group. Another difference was seen in asymmetries of the face: "snorers" were over three times less likely to display asymmetries of the mandible to the right (n = 2, 4.9 %) than "non-snorers" (n = 21, 17.8 %, RR = 0.28). An asymmetry of the mandible to the left was less common in "snorers", too (n = 4, 9.8 % versus (vs.) n = 15, 12.7 %, RR = 0.77). The median overbite in "snorers" (4 mm) was relevantly larger than in "non-snorers" (3 mm). A frontal crossbite was diagnosed half as frequently in "snorers" (9.8 %) than in "non-snorers" (23.7 %, RR = 0.41). An edge-to-edge occlusion was seen in 33.3 % of the test and in 24.6 % of the reference group (RR = 1.35). A lateral open bite was less frequent in "snorers" (7.5 %) compared to "non-snorers" (18.6 %, RR = 0.40). Relevant differences were also found in the transversal width maxilla anterior (median of 19.5 mm in "snorers" and 18 mm in "non-snorers") and posterior (median of 28.5 mm vs. 24 mm). Also the transversal width mandible anterior differed between the groups: a median of 31.5 mm in "snorers" and 18 mm in "non-snorers". The transversal width of the posterior mandible was increased in the test group (41 mm vs. 25 mm). Conclusion: The tested OSA screening and examination methods demonstrated practicability in the setting of an orthodontic practice regarding acceptance, adherence and feasibility. The PSQ serves as a suitable screening tool. ORFOS could not clearly identify orthodontic risk factors for OSA in childhood. We found differences relating to edge-to-edge bite, frontal facial asymmetries, overbite, crossbite, transversal widths, sucking habits and open bite between potential pediatric OSA patients and controls.