OA_m therapy is effective in reducing the apnea severity in some but not in all patients. In general 65 % of patients respond to the treatment with a ≥50 % reduction in apnea–hypopnea index (AHI) with the OA_m in situ compared to baseline. On average 52 % of patients achieve an AHI < 10 events/h with the OA_m in situ [12]. In the past, retrospective analyses of clinical, physiological, and polysomnographic variables were performed in order to identify predictors of treatment success with OA_m therapy. There is evidence to support the findings that OA_m therapy is more likely to be successful in younger female patients [16], with lower body mass index [17], a smaller neck circumference [11], and less severe sleep apnea [16, 18, 19].

In this chapter, we will focus on studies evaluating the effect of sleep position or the presence of supine-dependent OSA (sdOSA) on OA_m efficacy. Eight [16, 20–26] studies evaluated the effect of sdOSA on the outcome of OA therapy. Six studies [16, 20–24] reported that the efficacy of OA therapy is influenced by sleep position in a way that patients with sdOSA have better treatment outcomes, where two other studies did not find a difference in success rates between non-sdOSA and sdOSA patients [25, 26].

In a study of Cartwright et al. [20], the association between the efficacy of a TRD and the factors obesity, age, supine sleep posture, and severity of sleep apnea was investigated in 16 male patients. The authors concluded that an increase in sleep apnea severity in the supine sleep position was the strongest predictor of success with a TRD.

Marklund et al. [16, 23] evaluated the effect of sdOSA on therapy outcome with success defined as AHI < 10 events/h in both the supine and non-supine sleeping positions with the OA_m in situ. They reported that sdOSA was a strong predictor of successful apnea reduction with OA_m therapy. In addition, it was suggested that a low AHI in the lateral position is important in predicting a successful apnea reduction with OA_m therapy [23]. After subdividing the patient population according to gender, sdOSA remained the strongest predictor of OA_m therapy success in men but did not relate to a successful apnea reduction in female patients [16].

Yoshida [24] assessed a significant decrease in both supine AHI and AHI in prone position under OA_m therapy. The AHI increased in the lateral position, although not significantly. A successful apnea reduction under OA_m therapy (AHI < 10 events/h) was achieved in 61.4, 84.6, and 0 % of patients with respiratory disturbances most frequently observed in supine, prone, and lateral position, respectively. Yoshida [24] concluded that the efficacy of OA_m therapy is influenced by sleep posture.

In a study of Chung et al. [22] using Cartwright’s definition [27], the decrease in both total AHI as well as supine AHI under OA_m therapy was significantly higher in sdOSA patients when compared to non-sdOSA patients. The decrease in non-supine AHI did not differ between the two groups. The complete response rate with AHI < 5 events/h under OA_m therapy was higher in sdOSA patients when compared to non-sdOSA patients. Applying a multiple linear regression model, the presence of sdOSA turned out to be the only factor associated with a decrease in overall AHI or with a complete response.

Lee et al. [21] evaluated the efficacy of OA_m therapy in 100 Korean patients in terms of supine dependency. The success rate, defined as a reduction in AHI of 50 % or more and an AHI under OA_m of <10 events/h, was significantly higher in sdOSA compared to the non-sdOSA group.

Fransson et al. [25] subdivided their total patient population in supine-dependent ODI patients if 50 % or more of the estimated sleeping time was in supine position when desaturations were registered. In contrast with the previously described results that suggest that sdOSA is associated with a better treatment response, Fransson et al. [25] did not find a supine-dependent difference in responder rate, defined as patients with a reduction in ODI of at least 50 % or with an ODI value under OA_m therapy of <5 events/h.

Sutherland et al. [26] assessed differences in treatment response among 386 patients with and without sdOSA. In this study, no difference in complete response (AHI < 5 events/h) was noted between sdOSA and non-sdOSA patients.

Several confounders must be taken into account when comparing the studies evaluating the effect of sdOSA on the outcome of OA therapy.

A first confounder is the presence of different criteria for sdOSA in the literature (Table 1). The application of three different criteria for sdOSA makes it hard to compare the prevalence of sdOSA and the effect of sdOSA among the different studies.

A second confounder is the lack of a consensus in literature regarding the definition of successful treatment outcome. Some studies defined success as a reduction in AHI under therapy of ≥50 % compared to baseline, while other studies used a posttreatment AHI of less than 5, 10, or 20 events/h as a successful treatment outcome. In addition, some studies used a combination of a reduction in AHI of ≥50 % compared to baseline combined with a posttreatment AHI of less than 5, 10, or 20 events/h as criteria for success. One study uses the ODI as the main outcome parameter with success defined as a reduction in ODI of at least 50 % or with an ODI value under OA_m therapy of <5 events/h [25].

A third confounder is the use of different types of oral appliances in the discussed studies: one study used a TRD [20] whereas the other studies used an OA_m [16, 21–26]. Furthermore, a monobloc OA_m was used [16, 21, 23–25], whereas only in two studies a titratable OA_m was used [22, 26]. In the studies using a monobloc OA_m, four out of five studies (80 %) did find an association between the efficacy of OA_m therapy and the presence of sdOSA, whereas only one out of two studies (50 %) using a titratable OA_m could confirm these results.

The prevalence of sdOSA in a general population ranges from 20 to 60 %, depending on the definition used (Table 1). This prevalence was also studied in a more restricted population of patients starting OA_m therapy. In order to do so, Marklund et al. [23] defined sdOSA as a supine AHI ≥ 10 events/h with a lateral AHI < 10 events/h. According to this definition, 46 % of patients were diagnosed as having sdOSA. Applying the same definition, Dieltjens et al. [28] found a comparable prevalence of 46 % prior to the start of the OA_m therapy. In three studies, the prevalence of sdOSA as defined by Cartwright’s criteria was assessed before starting OA_m therapy and ranged from 58 to 80 % [21, 22, 28]. In a study of Yoshida, 61 % of patients starting OA_m therapy exhibited the respiratory events most frequently in supine sleeping position [24]. Overall, the prevalence of sdOSA found in patients starting OA_m therapy ranged from 27 to 80 % and was comparable to the prevalence of sdOSA in the general population (Table 1). These results however do not reveal the evolution of sdOSA once OA_m has started, and until recently, the prevalence of sdOSA under OA_m therapy was unknown. In a recent study 183 patients with polysomnographic data before and under OA_m therapy were evaluated showing a prevalence under OA_m therapy ranging from 18 to 34 %, depending on the definition used. In addition, it was shown that up to one third of patients shift from non-sdOSA at baseline to sdOSA under OA_m therapy [28].