Treatment-emergent central sleep apnea in patients treated with a mandibular advancement device

Treatment-emergent central sleep apnea (TECSA) is well established in continuous positive airway pressure therapy but was barely studied in mandibular advancement device (MAD) treatment. This study aims to evaluate the prevalence of TECSA in patients treated with a MAD and to determine its risk factors and clinical relevance.

A total of 139 patients from the IRSR Pays de la Loire Sleep Cohort suffering from snores or obstructive sleep apnea syndrome (OSAS) and treated with a custom-made titratable MAD were included. Baseline and follow-up sleep recordings enabled identification of TECSA patients. Comparative analyses were carried out between TECSA and non-TECSA groups to identify potential risk factors. Clinical relevance of TECSA in both groups was assessed through baseline and follow-up Pichot's self-assessment questionnaire for depressive symptoms (QD2A), 36-item short form survey (SF-36) and Epworth Sleepiness Scale (ESS) scores.

According to the definition selected, a prevalence between 0 % and 5.04 % was found for TECSA in the present study. No statistical differences were found in terms of treatment characteristics, sleep architecture, demographic data or comorbid conditions, although there was a trend towards a higher prevalence of arterial hypertension in TECSA-1 than in non-TECSA group (42.9 % vs 25.4 % respectively, p = 0.379). Baseline ESS showed a trend towards a higher score in TECSA-1 patients compared to non-TECSA patients (13/24 vs 10/24 respectively, p = 0.074), with a high proportion of TECSA-1 patients suffering from excessive daytime sleepiness before initiation of treatment (85.7 %, vs 52.4 % in non-TECSA patients, p = 0.124). No statistical differences were found regarding delta Pichot's QD2A and ESS scores between baseline and follow-up although there was a trend towards higher ESS scores at follow-up in TECSA-1 group compared to non-TECSA patients. Median delta SF-36 score for the General health scale was significantly lower in TECSA-1 and there was a trend towards lower scores for Mental health category in TECSA-1 patients.

TECSA is a rare phenomenon that can occur in patients treated with a MAD for an OSAS. Clinical, polysomnographic and treatment-related risk factors have yet to be reassessed in larger cohorts. These findings suggest probably poorer subjective clinical outcomes in terms of sleepiness and quality of life in patients with MAD-related TECSA.