Oral Appliances Found Non-Inferior to CPAP for Sleep Apnea in Sequential Effectiveness Study

A prospective, non-randomized sequential study published in the American Journal of Respiratory and Critical Care Medicine reports that mandibular advancement device (MAD) therapy achieves non-inferior clinical effectiveness compared with continuous positive airway pressure (CPAP) in moderate-to-severe obstructive sleep apnea (OSA). The primary outcome — mean disease alleviation (MDA), a composite of efficacy and adherence — favored MAD numerically but did not reach statistical significance for inferiority. However, the authors caution that the sequential design and unvalidated primary endpoint limit whether the findings establish clinical non-inferiority.

• Mandibular advancement devices proved non-inferior to CPAP in a combined measure of efficacy and adherence (MDA) after 3 months of each therapy in a sequential crossover-like study.

• Nightly adherence was significantly higher with MAD (6.7 vs. 5.4 hours/night), and more patients expressed preference for MAD (51% vs. 42%).

• CPAP achieved greater AHI reduction (median 4.1 vs. 8.4 events/hour), but the difference did not undermine non-inferiority on the primary endpoint.

• The study's sequential, non-randomized design and lack of validation of the MDA composite against long-term patient outcomes mean that results should be interpreted cautiously until confirmed in randomized trials.

Study Snapshot

Design

Prospective, single-group, non-randomized sequential intervention study (MAD then CPAP) with a wash-out period, using a non-inferiority analysis framework.

Population

94 adults (86% male; mean age 52±12 years; BMI 28.1±3.4 kg/m²) with moderate-to-severe OSA (baseline AHI 24.2 events/h).

Interventions

Customized mandibular advancement device for 3 months, followed by ≥1 week wash-out, then CPAP for 3 months.

Primary endpoint

Mean disease alleviation (MDA) — a composite measuring the percentage of AHI reduction multiplied by nightly adherence (% nights used × hours used). Non-inferiority margin was not explicitly stated in the abstract.

Key results

MDA: MAD 49.9±26.1% vs. CPAP 49.1±34.5% (non-inferiority p=0.4). AHI reduction: Both therapies significantly reduced AHI (both p<0.001). MAD adherence: 6.7 h/night; CPAP adherence: 5.4 h/night (p<0.05).

Funding

Not reported in the abstract.

Registration

Not reported in the abstract.

Why This Study Matters

CPAP is the established first-line therapy for moderate-to-severe OSA, but long-term adherence is often suboptimal. Mandibular advancement devices are commonly used as second-line treatment for patients who cannot tolerate CPAP. However, the comparative clinical effectiveness of MAD as a first-line option — defined by both efficacy and real-world use — has been uncertain. The present study introduces the concept of disease alleviation (MDA) to capture both dimensions and tests non-inferiority against CPAP in a single sequential cohort.

What the Study Examined

The First Line Obstructive Sleep Apnea Treatment (FLOSAT) study enrolled 94 patients at a single center (Antwerp University Hospital, Belgium). All participants received MAD therapy for 3 months, followed by a wash-out period (duration not specified) and then CPAP for 3 months. The primary outcome, mean disease alleviation, was defined as the product of AHI reduction (%) and a composite adherence index (percentage of nights used × average hours used per night). The analysis was performed in a modified intention-to-treat population.

Primary Effectiveness Result

The mean MDA was nearly identical between treatments: 49.9% (SD 26.1) with MAD and 49.1% (SD 34.5) with CPAP. The non-inferiority test yielded a p-value of 0.4, consistent with non-inferiority. Because the study used a non-inferiority framework with a sequential design, the authors emphasize that the result should be interpreted as showing that MAD effectiveness is not unacceptably worse than CPAP, rather than demonstrating equivalence or superiority.

Efficacy in Reducing AHI

Both therapies significantly reduced the apnea-hypopnea index from baseline (median 24.2 events/h). CPAP produced a greater reduction, to a median of 4.1 events/h (IQR 2.2–11.5), compared with MAD, which reduced AHI to 8.4 events/h (IQR 5.4–12.9). The difference was not tested for statistical significance directly, but the confidence intervals show some overlap. The higher AHI on MAD was offset by better adherence, contributing to the similar MDA.

Adherence and Patient Preference

Nightly adherence was significantly higher with MAD (median 6.7 h/night vs. 5.4 h/night for CPAP; p<0.05). Objective usage data were collected from device downloads. When patients were asked at the end of the study which therapy they preferred, 51% chose MAD, 42% chose CPAP, and 7% had no preference.

Safety and Tolerability

The abstract does not report any adverse events, serious adverse events, or discontinuations for either treatment. In typical clinical practice, MAD therapy is associated with side effects such as jaw discomfort, tooth pain, or excessive salivation, while CPAP may cause nasal congestion, mask leaks, or skin irritation. The absence of safety reporting in this abstract limits a full benefit-risk assessment.

Clinical Interpretation and Limitations

The study has several important limitations that affect how the results can be applied. First, the sequential, non-randomized design cannot separate the effect of the treatment from period or order effects — participants may have become accustomed to therapy or may have been influenced by their experience with the first device. Second, the primary outcome (MDA) has not been validated against hard outcomes such as cardiovascular events, quality of life, or daytime sleepiness, so its clinical significance is uncertain. Third, the comparatively small sample (94 patients) and lack of blinding (neither patients nor investigators were masked) may introduce bias. Finally, the source publication is an abstract; full details regarding the non-inferiority margin, wash-out duration, and missing data are not available.

Bottom Line

In this sequential study, MAD therapy showed non-inferior effectiveness to CPAP when efficacy and adherence were combined into the disease alleviation measure. The higher adherence and patient preference for MAD support its potential role as a first-line alternative for selected patients. However, until a randomized controlled trial with validated endpoints replicates these findings, CPAP remains the standard first-line treatment based on its larger AHI reduction and established long-term evidence. Patients who have difficulty tolerating CPAP can be counseled that MAD therapy may offer similar overall disease alleviation, but they should be informed about the limitations of the evidence.

References

1. Dieltjens M, Engelen S, Azarbarzin A, Van den Bossche K, Van Loo D, Braem MJ, Goossens R, Verbraecken J, Op de Beeck S, Vanderveken OM. Comparative effectiveness of oral appliances and continuous positive airway pressure: a prospective non-randomized study using a non-inferiority framework. Am J Respir Crit Care Med. 2026 Jul 20. PMID: 42475493.