Guidelines

The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement was adhered to as much as possible. In the study section paragraph, there is the PRISMA flowchart.

Population, intervention, comparison, outcome (PICO) question

This review was designed following PICO guidelines to determine the:

• 1)

  population: children and adults diagnosed with OSAS;

• 2)

  intervention: MFT;

• 3)

  comparison: MFT versus another therapy;

• 4)

  outcomes: The primary outcome was to evaluate AHI and snoring.

The secondary outcome was to investigate the efficacy of MFT in the patients with OSAS children and adults.

Search strategy

We registered the protocol on international prospective register of systematic reviews (PROSPERO) (CRD42021237554) and followed PRISMA. Searches were run in PubMed, Scopus, Embase, Web of Science, Cochrane Library, and Lilacs from 1 Jan 2009 to 31 Dec 2020; English language; humans. The core string (adapted per database) was: (myofunctional therapy OR orofacial myotherapy OR oropharyngeal exercise OR tongue exercise OR swallowing exercise OR breathing exercise) AND (apnea OR OSA OR OSAS OR snoring OR sleep-disordered breathing). Two reviewers (SS and CV), independently, deduplicated (EndNote X9), screened titles/abstracts (Rayyan), and assessed full texts; disagreements were resolved by a third reviewer (MP). Pre-specified outcomes: primary AHI and snoring; secondary oxygen saturation, daytime symptoms/sleep quality, mouth breathing, and adherence. Risk of bias was assessed with Cochrane RoB 2.0. Owing to heterogeneity of protocols and populations, we performed a structured narrative synthesis and reported effect directions; no meta-analysis was attempted. Searches were conducted up to 31 Dec 2020 according to the original protocol; therefore, studies published after this date were not included.

Study selection 1

• A.

  The selection process for this study was made in two phases. In the first stage, studies were considered according to the following inclusion (A): MFT and SDB.

• B.

  Language restrictions: English.

• C.

  Studies after 2009.

Study selection 2

The initial electronic search strategy yielded a total of 1,532 articles. After eliminating 1,305 duplicate records, 227 studies proceeded to title and abstract screening. At this stage, 203 papers were excluded (157 reviews and 46 unrelated to the topic). Consequently, 24 articles were retrieved for full-text assessment, of which 15 were further excluded for being off-topic. In the end, 9 studies fulfilled all the inclusion criteria and were incorporated into the systematic review. The selection process is visually presented in a PRISMA flowchart (Figure 1).

Display full size

Figure 1. Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flowchart. Adapted from [17]. Accessed Mar 1, 2026. © 2024–2026 the PRISMA Executive. Distributed under a Creative Commons CC BY 4.0 license.

Data screening and extraction

Duplicate records were removed using EndNote X9 software. Two independent reviewers (SS and CV) then performed title and abstract screening in a blinded fashion using Rayyan QCRI. Discrepancies were resolved by consensus or by consulting a third reviewer (MP). Studies were selected based on methodological rigor, including clarity of outcome definitions, completeness of follow-up, and use of intention-to-treat analysis. All the authors separately did initial screenings. At this stage, the title and abstract, if available, were analyzed by all the authors. The articles were selected if considered interesting by all the reviewers. A detailed full-text analysis was carried out, and three reviewers (SS, CV, and MP) extracted data from each study. There was no dissent related to data screening and extraction.

Data analysis

This systematic review was carried out by using original studies, comparative studies, research journal articles and, once analyzed, we collected data regarding six areas (country- design- sample size, intervention, inclusion criteria, sex and age, time, outcomes) as summary in Table 1 and other five areas (intervention therapy, details, duration, frequency, control therapy) as summary in Table 2. We also extracted protocol fidelity/adherence (sessions/day, minutes/session, completion rate) when available, to contextualize AHI changes.

Eligible interventions and comparators

We considered studies evaluating MFT delivered as a stand-alone program or in combination with other therapies (e.g., CPAP, adjunctive breathing/aerobic training, nasal washing), as well as trials comparing active vs. passive MFT approaches or sham protocols [11, 19–21, 24]. For each study, we extracted the MFT protocol (exercise set, frequency, duration), the comparator(s) (e.g., usual care, CPAP, sham), and follow-up. This information was prespecified in the extraction form and did not contribute to effect estimation beyond descriptive synthesis.

Quality assessment

Three reviewers (SS, CV, and MP) independently evaluated the risk of bias. This evaluation was carried out following the Cochrane-recommended approach for assessing the risk of bias in randomized controlled clinical studies, including four quality parameters: sequence generation, consideration of incomplete outcome data, freedom of selective outcome reporting, and other sources of bias.

Level of evidence

Levels of evidence and grade of recommendation followed the grading recommendations of the GRADE Working Group [26].

Our 9 final studies, published between 2009 and 2020, in Table 3, were all level 2B since we excluded reviews and meta-analyses, which would have been level 1A.

Table 3. Summary of the risk of bias for each article included in the systematic review.

Study	Incomplete outcome data	Selective reporting	Other bias
Villa et al. (2015) [11]
Diaferia et al. (2013) [18]
Diaféria et al. (2017) [19]
Erturk et al. (2020) [20]
Guimarães et al. (2009) [21]
Villa et al. (2017) [22]
Ieto et al. (2015) [23]
Kim et al. (2020) [24]
Lin et al. (2020) [25]

Display full size

+: low risk of bias; –: high risk of bias; ?: unclear risk of bias.

Assessment of risk of bias

The studies were evaluated as having high, unclear, or low risk, based on selection bias, as can be observed in Table 3. The evaluation of the methodological quality of the included studies revealed variable levels of bias across different domains. Regarding incomplete outcome data, three studies [18, 19, 22] were judged at low risk, while the majority of the others presented a high or unclear risk. For selective reporting, consistent low risk was observed in the earlier trials by the other two studies [18, 22], whereas other studies showed either high or unclear risk, suggesting that outcome reporting was not always sufficiently transparent. With respect to other potential sources of bias, most trials were rated as unclear, reflecting insufficient information to permit a firm judgment, while a few were assessed as high risk. Overall, older and more methodologically rigorous studies tended to report lower risks of bias, while more recent investigations often suffered from limitations such as incomplete reporting, small sample sizes, or insufficient methodological details. This heterogeneity in study quality highlights the need for standardized protocols and more robust trial designs in future research on MFT for OSAS.

Characteristics of interventions across included studies

Across the included trials, MFT was delivered in several configurations:

• 1)

  stand-alone exercise programs targeting tongue, soft palate, and orofacial muscles;

• 2)

  randomized comparisons allocating patients to MFT, CPAP, combined MFT + CPAP, or sham/speech therapy;

• 3)

  head-to-head comparisons of active vs. passive MFT approaches;

• 4)

  adjunctive combinations, such as MFT with nasal washing or with aerobic training sessions.

This heterogeneity in delivery and co-interventions is summarized in Table 1 (study characteristics) and is considered in the interpretation of effects below.

Effects of interventions

The primary outcome was to evaluate AHI and snoring. The results of this systematic review indicate that with MFT, there is a reduction in the index, but the reduction in snoring is not evaluated in all articles, but Ieto et al. [23] demonstrated that oropharyngeal exercises are effective in reducing objectively measured snoring and are possible treatment of population suffering from snoring [27].

The secondary outcome was to investigate the efficacy of MFT in the patients with OSAS children and adults and the results demonstrated that use both alone and in combination with other therapies is useful in both adult and children’s patients. While AHI reduction was consistently reported and remains the main outcome, other clinically relevant endpoints such as snoring intensity, sleep quality, daytime sleepiness, and patient-reported quality of life were not systematically assessed.

Notably, several studies reported a more pronounced reduction in AHI among pediatric patients compared to adults, possibly due to greater tissue plasticity and neuromuscular responsiveness in younger subjects. Additionally, improvements in peripheral oxygen saturation and reductions in mouth breathing were consistently reported, particularly in interventions combining MFT with nasal breathing exercises.

Across 9 RCTs, MFT reduced AHI versus control or usual care; snoring reduction was demonstrated in two trials that measured it; oxygen saturation and mouth breathing improved in several pediatric cohorts. These outcomes are highly consistent with emerging evidence that early functional rehabilitation can enhance upper airway tone, reduce obstruction, and improve overall sleep quality.

Overall discussion

While the Results section objectively summarizes study outcomes, here we discuss mechanisms, patient adherence, and the clinical significance of these findings. OSAS is defined as a pathology that recognizes multiple risk factors for its development, often combined with each other [20].

The mechanism of action underlying this correlation could be based on structural or functional alterations. The therapeutic approach must also be aimed at acting on various factors. Even some craniofacial factors, such as a reduced size of the skull and jaw or the reduced width of the upper respiratory tract, or a short lingual frenulum, are currently considered risk factors in the development of OSAS, together with particular malocclusions [11, 18, 23]. In the diagnosis and therapeutic choice, we must pay attention to the age of the patient, particularly in adolescents and children; therefore, the treatment of malocclusions through orthodontic or functional therapy could represent a fundamentally important aid in the treatment of OSAS in pediatric individuals [28–30]. CPAP remains the standard treatment for the most severe forms of OSAS. MADs are also highly effective in patients with moderate or severe diseases who are unable to tolerate CPAP. However, their use can be more challenging in partially or completely edentulous patients, although in such cases, integrated osteo-implants may represent a possible solution [21].

The MFT approach: a set of exercises that involve the soft palate, tongue, and facial muscles closely involved in the apnea characteristic of the syndrome; this can, in fact, determine a notable improvement in the parameters commonly used to monitor obstructive apnea syndromes, up to reducing the AHI by 50%. The physiological rationale behind these improvements lies in the enhancement of upper airway muscle tone and coordination. By strengthening the tongue, soft palate, and oropharyngeal muscles, MFT reduces airway collapsibility during sleep and promotes more stable breathing patterns. Improved nasal breathing and swallowing function further contribute to maintaining airway patency and reducing mouth breathing, mechanisms that explain the observed clinical benefits [31].

However, regarding many of these cofactors and the effect of these therapies, considerable confusion remains in the scientific world, and it is of fundamental importance to clarify their action in order to intercept and promptly treat this pathology. The study characteristics of the selected articles demonstrate the rigorous approach taken to assess the effectiveness of MFT in the treatment of OSAS [20]. Among the 9 chosen studies, considered the gold standard in clinical research for evaluating treatment efficacy. The inclusion of a minimum sample size of 15 patients in all studies ensures that the results are based on a sufficiently representative group, enhancing the reliability and generalizability of the findings. Furthermore, the minimum follow-up period of 2 months in each study allows for the evaluation of the sustained impact of MFT on OSAS over a reasonable timeframe [25].

Qualitative characteristics

Regarding qualitative characteristics, the consistent reporting of primary outcome measurements across all studies, such as the AHI, Apnea Index, oxygen saturation levels, snoring intensity, and neck circumference, highlights the standardized and comprehensive approach taken in assessing the effectiveness of MFT. These outcome measures are crucial in capturing different aspects of OSAS severity and the treatment response, ensuring a comprehensive evaluation of the therapy’s impact [22]. A major factor influencing the efficacy of MFT is patient compliance. Unlike CPAP, which offers a passive form of treatment, MFT requires active daily participation. Studies have shown that higher adherence to exercise frequency is directly associated with greater reductions in AHI and symptom severity. However, adherence metrics were rarely quantified in the included studies, making it difficult to assess dropout reasons or long-term feasibility. Strategies such as structured follow-up, caregiver involvement for pediatric patients, and the integration of digital tools (e.g., mobile applications, telemonitoring) could improve compliance and should be considered in future trials. However, variability in exercise protocols across studies poses a challenge to reproducibility [19]. There is currently no universally accepted standardized regimen, and this heterogeneity limits the ability to draw definitive conclusions on the most effective therapeutic approach [32, 33].

Adherence to MFT represents a critical factor for its effectiveness. Unlike CPAP, which provides a passive treatment modality, MFT requires active, daily patient engagement. Reported adherence rates varied across studies and were not always explicitly quantified, but several authors noted difficulties related to motivation, time commitment, and integration of exercises into daily routines. Potential barriers included a lack of supervision, limited awareness of the long-term benefits, and the repetitive nature of the exercises. Dropouts were often associated with poor motivation or perceived limited benefits. These findings underscore the importance of structured follow-up, patient education, and possibly the use of digital tools (e.g., mobile applications or telemonitoring) to enhance compliance. Future trials should systematically measure and report adherence data, since feasibility is essential for translating MFT into routine clinical practice.

In addition to behavioral interventions such as MFT, recent advances in neurotechnology have opened novel therapeutic perspectives. Temporal interference (TI) stimulation, a non-invasive technique using intersecting high-frequency electrical fields to selectively activate deep neural targets, has shown preliminary promise in modulating hypoglossal nerve activity and oropharyngeal muscles involved in OSA pathophysiology. Early studies suggest that TI stimulation may achieve functional benefits similar to MFT while overcoming the challenge of patient adherence, since it does not require daily active participation. Although still experimental, these innovations highlight the potential future integration of neuromodulation strategies with conventional therapies to broaden the management options for OSA.

Although a formal meta-analysis could not be conducted because of the high variability in study designs, populations, and intervention protocols, the findings highlight the urgent need for standardization. A preliminary framework for future MFT protocols could include:

• 1)

  clearly defined sets of oropharyngeal and tongue exercises;

• 2)

  standardized daily frequency and session duration;

• 3)

  minimum treatment length of 8–12 weeks;

• 4)

  objective outcome measures such as AHI reduction and polysomnography;

• 5)

  systematic assessment of secondary outcomes, including snoring, oxygen saturation, and quality of life.

Such a framework may guide future clinical trials and support the development of reproducible, evidence-based guidelines. To enhance clinical applicability, we also provide in Table 4 a preliminary summary of key elements that could form the basis for future standardized MFT protocols.

Population

The inclusion of a total population of 698 patients with OSAS, derived from the nine included studies, strengthens the reliability and relevance of the findings [13]. The large number of participants allows for a more comprehensive assessment of the effectiveness of various MFT interventions in managing OSAS. The confirmation of the OSAS diagnosis through clinical examinations, particularly polysomnography, ensures the accuracy and validity of the patient selection, enhancing the overall robustness of the review [23]. As far as MFT is concerned, it still has many limitations; there are not many publications that can validate protocols that make MFT fundamental for the treatment of OSA [11].

There is a clear lack of cooperation among the various medical professionals involved in the treatment of OSAS.

Study limitations

This systematic review has several limitations that should be acknowledged. First, the number of RCTs available on MFT in OSAS is still limited, and the sample sizes of most included studies were relatively small, which may reduce the generalizability of the findings. Second, the included trials were highly heterogeneous in terms of study design, patient populations (adults versus children, varying severity of OSAS), and especially in the type, frequency, and duration of MFT exercises. The absence of a standardized protocol makes it difficult to establish optimal treatment parameters and reduces comparability across studies. Third, although AHI reduction was consistently reported and remains the main outcome, other clinically relevant endpoints such as snoring intensity, sleep quality, daytime sleepiness, and patient-reported quality of life were inconsistently measured across the trials. These outcomes were therefore not omitted from our analysis but were often unavailable due to study design limitations. Moreover, long-term follow-up data are largely lacking, since most included trials assessed outcomes only over short periods (2–6 months). As a result, the durability of MFT effects beyond the intervention phase remains uncertain. Finally, some studies presented methodological weaknesses, including unclear allocation procedures, incomplete reporting of outcomes, or potential sources of bias, as highlighted in our risk of bias assessment. Publication bias cannot be ruled out, since only English-language articles were included. Taken together, these limitations suggest that while current evidence supports a potential role for MFT in OSAS management, further large-scale, well-designed, and standardized trials with longer follow-up are required before firm conclusions can be drawn.

In conclusion, this systematic review indicates that MFT can reduce the AHI and may improve snoring, oxygen saturation, and mouth breathing, with particularly encouraging signals in pediatric populations. To establish its role in routine care, future multicenter trials should adopt standardized exercise protocols (clearly defined sets, frequency, and duration), monitor adherence systematically, include core outcome sets encompassing both objective sleep metrics and patient-relevant measures, and ensure longer follow-up to assess durability. Comparative designs against, or in combination with, CPAP and MADs will help position MFT within integrated OSA pathways.