Send to

Choose Destination
Sleep Breath. 2018 May;22(2):385-392. doi: 10.1007/s11325-017-1578-2. Epub 2017 Oct 24.

Oral appliance treatment outcome can be predicted by continuous positive airway pressure in moderate to severe obstructive sleep apnea.

Author information

Department of Thoracic Medicine, Center for Sleep Medicine, Haukeland University Hospital, Bergen, Norway.
Department of Thoracic Medicine, Center for Sleep Medicine, Haukeland University Hospital, Bergen, Norway.
Department of Clinical Dentistry-Prosthodontics, Faculty of Medicine and Dentistry, University of Bergen, Bergen, Norway.
Department of Global Public Health and Primary Care, University of Bergen, Bergen, Norway.
Department of Clinical Science, University of Bergen, Bergen, Norway.



Studies show that the therapeutic CPAP pressure is associated with oral appliance (OA) treatment outcome in obstructive sleep apnea (OSA) patients. However, these studies included either CPAP adherent patients using fixed pressure devices, or partly CPAP non-adherent patients using fixed pressure or auto-adjusting (auto-CPAP) devices. In many countries, auto-CPAP is predominately used, and only those non-adherent to therapy need a change to OA. Therefore, studies examining the relationship between CPAP pressures and OA treatment outcome should focus on patients non-adherent to auto-CPAP.


The purpose of this paper is to assess if CPAP pressures predicted OA treatment outcome in patients non-adherent to auto-CPAP therapy.


The OA treatment responders and non-responders were defined by two success criteria ((1) AHI < 5; (2) 5 ≤ AHI < 10 and > 50% AHI reduction). Logistic regression analyses were performed for CPAP pressures and baseline variables. ROC curve analyses were used to identify CPAP pressure cutoff values, alone and combined with other explanatory variables, predicting the OA treatment outcome.


Eighty-seven patients with moderate or severe OSA were included. Maximum CPAP pressures (CPAPmax) were higher in non-responders by both criteria and were, together with baseline AHI, associated with the OA treatment outcome in multivariate regression analyses. ROC curves identified an optimal CPAPmax cutoff of 12 cm H2O, corresponding to a positive predictive value (PPV) of 0.85 in predicting non-response using criterion 1. A prediction model combining CPAPmax > 12 and baseline AHI ≥ 30 had a PPV of 1.0 for non-response by both criteria.


Maximum CPAP pressure was a moderate predictor of OA treatment outcome, but combined with baseline AHI, the ability to identify OA non-responders was high.


Apnea-hypopnea index; Continuous positive airway pressure; Obstructive sleep apnea; Oral appliances; Prediction; Treatment outcome

[Indexed for MEDLINE]
Free PMC Article

Supplemental Content

Full text links

Icon for Springer Icon for PubMed Central
Loading ...
Support Center