Relationship between OSA pathophysiological phenotypes and treatment response to mandibular advancement devices: a pilot study.

Abstract

To assess whether critical pathophysiological phenotypes predict treatment response in patients with obstructive sleep apnea (OSA) using a mandibular advancement device (MAD). Thirty-one OSA patients were treated with a MAD. Individuals were categorized and graded into four pathophysiological phenotypes based on polysomnographic features (anatomical, ventilatory control, arousal threshold and muscle responsiveness). Morphoanthropometric data were additionally assessed. Patients were classified as responders or nonresponders. Associations between polysomnographic phenotypes and treatment response were documented, as was morphoanthropometric data and their impact on therapeutic success. There was a male predominance (64.5%), with a median age of 49 years (25p:40; 75p:55), BMI=27.4 kg/m2 (26; 28.8) and apnea-hypopnea index (AHI) of 18.2 (25p:11.7; 75p: 27.6). The majority of patients treated with a MAD (58%) were good responders (68.0% mild and moderate versus 16.7% severe). Treatment response was associated with shorter intermolar and interpremolar distances in the lower arch (p = 0.0092 and 0.0129). Rapid eye movement sleep AHI (REMAHI) and MAD-related treatment response were inversely correlated (p = 0.0013). Favorable anatomical (p = 0.0339) and low muscle response (p = 0.0447) phenotypes were correlated with outcomes. According to our results, a favorable response occurred in a better 'anatomical phenotype' and in the worse 'muscular responsiveness phenotype' according to polysomnographic data. Furthermore, other favorable predictors, such as a REMAHI <16 and a smaller distance between lower molars and premolars, were found. These findings indicate that clinical and polysomnographic aspects can discriminate phenotypes that may guide decisions on MAD treatment for OSA.