The success and failure of using mandible andvancing device in treating obstructive sleep apnea patients

Objectives:
Obstructive sleep apnea (OSA) is a common sleep disordered breathing affecting especially for adults over 40. It can result in some medical and life quality consequences, and needs to be solved. Mandibular advancing device (MAD), which can bring and maintain the lower jaw in a forward posture during sleep, is demonstrated to be effective in alleviating the severity of OSA in some cases. Due to a good compromise between efficacy and acceptance, MAD is getting popular as a treatment option of OSA. The American Academy of Sleep Medicine (AASM) suggests MAD as a treatment alternative of nasal continuous positive air pressure (nCPAP) for mild to moderate OSA patient.
There are several treatment modalities of OSA and oral appliance is a simple, effective and popular treatment option among them. The etiologic factors and pathogenesis of each patient are different, so the treatment effect of oral appliance varies from each other.
The purpose of this study was to evaluate the treatment effect of oral appliance in OSA patients in relation to their basic medical condition, craniofacial features, polysomnographic data and the improvement of self-estimated symptoms. We hope to illustrate an effective and specific method for predicting treatment response of oral appliance.
Materials and methods:
A total of 118 patients from Sleep Center in National Taiwan University had related clinical symtoms and signs, and were diagnosed as OSAS by polysomnography. Then patients were arranged for mandibular advancing oral appliance therapy in Special Clinic of Oral Appliance in Dental Department of NTU. Pre-treatment age, height and weight were recored, so the BMI were derived. Besides, baseline cephalometry was performed on each patient. Three months after treatment, patients came back to Sleep Center for follow-ups and received post-treatment polysomnography. In the meanwhile, patients were asked to answer the questionnaire of SF-36 before treatment and 3 months after treatment to assess the self-estimated responses of the treatment. The subjects were divided into two groups on the basis of the degree of changes in AHI: responders (>50% AHI reduction) and non-responders (<50% AHI reduction).
Paired t test was performed to examine the changes after treatment. Univariate and multivariate regression analyses were used for searching possible variables to predict the response of treatment.
Results:
The number of responders in this study was 72. The success rate of oral appliance therapy in this study was 61.0%. The BMI of all the samples significantly decreased after treatment. Improvement in polysomnographic findings were noted as well.
Predictors of oral appliance treatment outcome were successfully derived in clinically valid regression models. The characteristics of responders were younger，with larger SNA, longer N-Ba, lager maxilla length, longer S-Ba, short soft palate, normal overbite.
However, non-responders usually had retrognathic maxilla, short N-Ba, short maxilla length, short posterior basicranial length, longer soft palate, unnormal overbite. Besides, the pre-treatment rate of oxygen saturation <90% was higher in non-responders.
The PSG data predictors of responders were initial AHI more than 25, the 2nd sleep stage under 63%, the less of ESS score, less of mixed apnea. And the non-responders were initial AHI less than 25, the 2nd sleep stage more than 63%, the higher of ESS score, higher of mixed apnea.
From the analysis of questionnaire, all conditions were better in responders than in non-responders either before or after oral appliance treatment, though pretreatment and posttreatment didn’t reached statisical difference.
Conclusion:
The treatment effects of oral appliance in patients with OSA were related to multiple factors, including age, weight, craniofacial structures and pre-treatment polysomnography. Therefore, clinicians can predict better treatment outcome if proper indication was established for different treatment modalities.