Reply to Jaffuel and Borel: If Oral Breathing Does Not Determine Mask Choice for Continuous Positive Airway Pressure Delivery, What Does?

From the Authors:

We thank the authors for their interest in and constructive comments on our study (1). Despite the widespread use of oronasal continuous positive airway pressure (CPAP) to treat obstructive sleep apnea (OSA), the initial choice of oronasal mask is based solely on subjective self-report of predominant mouth breathing. However, there is no agreement between self-report and objectively detected breathing route, both when awake and during sleep (2). We agree that nasal obstruction is a major cause of mouth breathing. Specific endotypes are suggested as potential clues to guide mask choice. This scientific approach is certainly welcomed and must be widely investigated to face the current clinical reality. Mouth opening during sleep may characterize an endotype that would benefit from an oronasal mask. We have recently shown that, in contrast to control individuals, oronasal breathing while awake and asleep was common among patients with OSA (1). Oronasal breathing was associated with OSA severity, age, body mass index, and neck circumference (1). These findings are in line with the concept pointed out by the authors that increasing effort to breathe during obstructive events leads to mouth opening in patients with OSA. However, CPAP abolishes OSA and relieves respiratory effort. In concert with this view, there is evidence that most patients with OSA who breathe through the mouth during a diagnostic sleep study switch, over the period of a few months, to nasal breathing under nasal CPAP (3).

In this context, we fear that the detection of mouth opening during the first night of CPAP titration is actually a potential misleading incentive to overuse oronasal masks. Therefore, the observation of mouth opening at the diagnostic sleep study is not necessarily an indication of nasal CPAP intolerance. Mouth opening is a moving target that changes with CPAP use. Even in a single diagnostic study, mouth opening changes along sleep stages and is more common in REM, as pointed out by the authors. To finalize, the landmark study of Sullivan and colleagues 40 years ago conceived that, to abolish OSA, CPAP must be delivered through the nose because the pressure is transmitted to the back of the pharynx to splint the airway open (4). Oronasal CPAP violates this concept because positive pressure through the mouth neutralizes the splinting of the airway promoted by nasal CPAP. According to this line, we showed that oronasal CPAP is effective only when the patient breathes predominantly through the nose (1). In simple terms, an oronasal mask is often effective to treat OSA because exclusive mouth breathing is probably rare. This observation helps to explain why oronasal CPAP is in general well tolerated. One may argue that this is not a relevant question, because several patients are well adapted and prefer an oronasal mask. However, oronasal masks are more expensive than nasal masks and less effective to treat OSA, because they are associated with higher residual events despite higher CPAP levels (5). Meanwhile, we have information that in the United States, for instance, more than 30% of CPAP mask sales are oronasal masks. These numbers are in line with the observation that 28.4% of the patients with OSA in France were using oronasal masks in the InterfaceVent-CPAP study (6). The real question is why oronasal masks are so common in clinical practice. We argue that nasal interfaces should be considered the standard treatment of OSA. In our view, oronasal masks could be considered as an alternative treatment only after well-documented failure of a prolonged trial of supervised nasal CPAP.