The Role of the Hyoid Bone in Obstructive Sleep Apnea Severity: A Systematic Review.

Save

ABSTRACT

Obstructive sleep apnea (OSA) has numerous complications that negatively impact patients' quality of life. Early diagnosis can significantly reduce these complications. The aim of the study was to evaluate the relationship between the hyoid bone position and the severity of apnea, and to determine if hyoid‐related cephalometric variables can be used as predictors of apnea severity.

A systematic review of the literature was conducted following PRISMA guidelines. The search was conducted across databases including Medline, ScienceDirect, BVS, and WOK. Two researchers analysed the results applying inclusion and exclusion criteria, with a third researcher resolving discrepancies.

After screening, a total of 22 articles were selected for qualitative review. The risk of bias and quality of the studies was assessed using the Downs and Black checklist. Among the 22 selected articles, 9% were classified as poor quality, 46% as fair, and 45% as good quality.

A positive correlation was observed between a lower hyoid bone position and the apnea‐hypopnea index (AHI), indicating that a lower hyoid position is associated with a higher AHI. The H‐MP variable (distance between point H and the mandibular plane) presented the strongest predictive power for OSA severity. However, insufficient evidence was found to establish an association between the anteroposterior position of the hyoid bone and OSA severity.

The severity of obstructive sleep apnea (OSA) is associated with a lower position of the hyoid bone, as reflected in the positive correlation between H‐MP and the apnea‐hypopnea index (AHI), while the sagittal position showed no significant relevance.

Background

Obstructive sleep apnea (OSA) is a respiratory disorder characterised by episodes of apneas and hypopneas during sleep, caused by obstruction of the upper airway (Chang et al. 2023; Semelka et al. 2016; Neelapu et al.

2017). It's associated with various risk factors such as age, elevated body mass index (BMI), and male sex. OSA is also linked to craniofacial features that decrease the diameter of the upper airway, such as small or retruded jaws, elongated faces, and a low position of the hyoid bone (Chang et al.

2023; Semelka et al. 2016).

Symptomatic OSA in adults has a prevalence of between 3% and 9% in men and between 2% and 5% in women. In contrast, asymptomatic OSA affects a larger percentage of the population, increasing to 28%–70% in males and 20%–56% in females (Neelapu et al. 2017; Lin and Suurna 2018).

The severity of this disease is classified according to the apnea‐hypopnea index (AHI), which records the average number of respiratory disturbances per hour of sleep. It can be classified into mild OSA (5–15 events/h), moderate OSA (15–30 events/h), and severe OSA (> 30 events/h) (Epstein et al. 2009; Lee and Sundar 2021).

According to the American Academy of Sleep Medicine (AAMS), the clinical manifestations that should raise suspicion of OSA are: witnessed apneas, snoring, gasping/nocturnal choking, excessive daytime sleepiness, non‐restorative sleep, insomnia, morning headaches, decreased concentration, memory loss, and decreased libido (Epstein et al. 2009; Chai‐Coetzer 2019).

Early diagnosis of OSA is crucial for preventing long‐term complications. While polysomnography (PSG) remains the gold standard for confirmation, complementary tools such as questionnaires, cephalometric studies, and home sleep studies can help identify high‐risk patients and refer them for more specialised testing. In this context, the hyoid bone has been the subject of research due to its involvement in OSA, as its position and movement may influence the obstruction of the upper airways.

Several cephalometric studies have shown that a lower position of the hyoid bone may be associated with OSA, making this variable a potentially useful indicator for early detection of the condition (Chang et al. 2020; Kapur et al. 2017; Cheryl et al.

2017).

OSA is a common but often underdiagnosed disorder that significantly impacts health. Early diagnosis is crucial to prevent complications, and tools like cephalometry are essential for evaluating anatomical features linked to OSA. Given that orthodontists routinely use cephalometry as a diagnostic tool and there is robust evidence linking hyoid position to the presence of OSA, we decided to delve deeper into this relationship.

The aim was to analyse the relation between the vertical and anteroposterior position of the hyoid bone and the severity of OSA measured by the apnea‐hypopnea index (AHI). The secondary aim was to identify cephalometric markers associated with OSA.

Materials and Method

Study Design

The systematic review was conducted following the PRISMA guidelines (Page et al. 2021), and being previously registered in the PROSPERO platform (Registration number: CRD42023424119, registration date 6 June 2023).

Focused Question

The PICO (population, intervention, comparison, outcome) question was: ‘Does the hyoid bone position influence the severity of OSA, as assessed by AHI, in adult patients?’ The position of the hyoid bone was assessed using multiple cephalometric measurements, with both vertical and sagittal evaluations. Studies were excluded if they included participants under the age of 18, those with syndromes, or those with a history of prior respiratory or orthognathic surgeries. Studies without a control group were also excluded.

Reviews, meta‐analyses, and conference abstracts were not considered for this review. No time restrictions were applied, but the search was limited to English, Spanish, and French languages.

Databases and Search Strategy

A comprehensive literature search was started in October 2022, encompassing Medline (PubMed), ScienceDirect, Biblioteca Virtual de Salud (BVS), and Web of Science (WOS). To identify additional pertinent studies, a manual search of the reference lists of retrieved articles was conducted. The search strategy employed a combination of keywords (‘Hyoid bone’, ‘AOS’, ‘Apnea’, ‘Severity’, ‘Cephalometry’), truncated terms, and medical subject headings (MeSH) to accommodate variations in controlled vocabulary across databases (Table 1).

Study Selection

A two‐stage screening process was employed. Initially, two independent reviewers conducted a title and abstract screen using the predefined search terms and eligibility criteria. Any disagreements were resolved through consensus with a third reviewer.

Studies deemed potentially eligible based on the title and abstract screen underwent a full‐text review to verify their adherence to the inclusion criteria. This final stage of the review process was conducted by a single reviewer.

Data Extraction and Study Outcomes

Following the selection of articles that met the inclusion criteria, the following data were systematically extracted: authors, year of publication, study design, sample size, age range and mean, gender ratio, body mass index (BMI), apnea‐hypopnea index (AHI), ethnicity, specific cephalometric variables, and study outcomes. All extracted data were tabulated in an Excel spreadsheet by a unique operator supervised by another blinded researcher.

Study Risk of Bias Assessment

To assess the risk of bias (RoB) in each of the selected articles, the Downs and Black checklist was conducted (Downs and Black 1998) by two researchers. Any disagreements were resolved through consensus with a third reviewer. This tool consists of 27 questions, divided into five subgroups: Reporting (Questions 1–10), external validity (Questions 11–13), internal validity‐bias (Questions 14–20), internal validity‐confounding (selection bias) (Questions 21–26), and study power (Question 27).

Each question is associated with a value based on the answer, ‘yes’ 1 point or ‘no’ 0 points; except for Question 5 which can reach a score of up to 2 points (0 = no, 1 = probably yes, and 2 = yes) and Question 27 which can be answered with a maximum score of 5. However, in this review all questions were answered as yes or no, making 27 the highest possible score. Based on the obtained score, the studies were classified as: poor quality (< 14), fair or regular quality (Johal et al.

2007; Costa E Sousa and Dos Santos Gil 2013; Silva et al. 2014; Piccin et al. 2016; Bayat et al.

2017), good (Valarelli et al. 2018; Bilici et al. 2018; Ahmadi et al.

2022; Tanellari et al. 2022; Guilleminault et al. 1984; Battagel et al.

2000) or excellent (Barrera et al. 2017; Ito et al. 2001).

Results

Flow Diagram

Following the previously described search strategy, a total of 645 articles were selected: 254 from Medline/PubMed, 186 from WOS, 188 from BVS, and 17 added manually. After removing duplicates, the sample was reduced to a total of 517, of which 441 were excluded after reading the titles and abstracts for not meeting the inclusion criteria of this review.

Seventy‐six studies were selected for full‐text examination. In this evaluation, five were excluded because they were written in a language unknown to the researchers, 18 due to lack of full‐text access, three for being pilot or preliminary studies; another 11 for not having a control group, four for not performing a diagnostic PSG on the sample, three that used a classification index for apnea other than IAH, two because the sample consisted of patients with previous surgeries, one for having a study sample with retrognathic mandibles compared to a control group with norm‐positioned mandibles; three for not studying the hyoid position in their cephalograms and four for not using lateral skull X‐ray as a tool to study the hyoid position. Finally, the qualitative synthesis was composed of 22 articles (Figure 1).

Qualitative Analysis

Of the 22 articles included (Tangugsorn et al. 1995; Battagel and L'Estrange 1996; Johal et al. 2007; Costa E Sousa and Dos Santos Gil 2013; Silva et al.

2014; Piccin et al. 2016; Bayat et al. 2017; Valarelli et al.

2018; Bilici et al. 2018; Ahmadi et al. 2022; Tanellari et al.

2022; Guilleminault et al. 1984; Battagel et al. 2000; Barrera et al.

2017; Ito et al. 2001; Johns et al. 1998; Rose et al.

2002; Verin et al. 2002; Hsu et al. 2005; Banhiran et al.

2013; Gulati et al. 2010; Hui et al. 2003) all of them were analytical observational case–control studies and their characteristics are summarised in Table A1.

The articles were all published between 1984 and 2022.

Participants ranged in age from 18 to 77 years. While most studies (18 out of 22) included both males and females, there was a marked preponderance of males. In fact, five studies were exclusively conducted on males (Tangugsorn et al.

1995; Battagel and L'Estrange 1996; Battagel et al. 2000; Ito et al. 2001).

Regarding the rest of the studies, even though both sexes were included, the proportion of male participants was significantly higher than that of females, except in two studies (Silva et al. 2014; Banhiran et al. 2013), where the opposite was observed, with a greater number of women in the group diagnosed with OSA.

All the included studies evaluated the body mass index (BMI) of participants, except for two of them (Tanellari et al. 2022; Guilleminault et al. 1984).

The diagnosis of OSA was established using PSG, and severity was categorised based on the apnea‐hypopnea index (AHI). While there was variation in the specific AHI cut‐offs used, most studies categorised OSA as mild (5–15 events/h), moderate (15–30 events/h), or severe (> 30 events/h).

Lateral cephalometric radiographs were used to assess hyoid bone position, and various cephalometric measurements were taken, including H‐MP, H‐C3, H‐Ph, H‐Me, H‐ENP, HB, HH1, SH, and H‐ENA (Table 2). The H‐MP distance, representing the distance from the mandibular plane to the hyoid bone, was the most often reported measurement.

Quantitative data regarding the relationship between hyoid bone position and AHI index were variably reported across the studies included in this systematic review (Table A2).

Several studies implemented statistical adjustments for demographic variables such as sex, age, and BMI to control for confounding effects. Full multivariate adjustments were explicitly reported in studies by Johal et al. (2007), Bayat et al.

(2017), Ahmadi et al. (2022), and Tanellari et al. (2022), enhancing the reliability of their findings regarding craniofacial risk factors.

Additionally, Costa E Sousa and Dos Santos Gil (2013) and Silva et al. Silva et al. 2014 incorporated BMI and age into their statistical models, although sex was not consistently adjusted for.

Partial adjustments were observed in several studies (Piccin et al. 2016; Valarelli et al. 2018; Bilici et al.

2018; Battagel et al. 2000; Barrera et al. 2017), which included one or two of the variables (typically age or BMI) in descriptive or comparative analyses without full multivariate control.

In contrast, earlier and more anatomically focused studies (Tangugsorn et al. 1995; Battagel and L'Estrange 1996; Guilleminault et al. 1984; Ito et al.

2001; Johns et al. 1998; Rose et al. 2002; Verin et al.

2002; Hsu et al. 2005; Banhiran et al. 2013; Gulati et al.

2010; Hui et al. 2003) did not report statistical adjustments for sex, age, or BMI, limiting the generalizability of their findings across diverse populations. Studies that performed full or partial statistical adjustments (Johal et al.

2007; Costa E Sousa and Dos Santos Gil 2013; Silva et al. 2014; Piccin et al. 2016; Bayat et al.

2017; Valarelli et al. 2018; Bilici et al. 2018; Ahmadi et al.

2022; Tanellari et al. 2022) demonstrated that inferior displacement of the hyoid bone was more pronounced in males and individuals with elevated BMI and was associated with increased AHI index. These studies also reported that age‐related changes in muscle tone and skeletal structure contributed to variations in hyoid position, further complicating the interpretation of anatomical risk factors.

Ethnic background appears to influence the anatomical and clinical presentation of OSA. Studies conducted in Caucasian populations consistently reported moderate to severe (Guilleminault et al. 1984; Battagel et al.

2000; Johns et al. 1998; Gulati et al. 2010) OSA with BMI values ranging from 27 to 34 and mean AHI values between 28 and 47.

In contrast, Asian cohorts showed higher AHI values despite lower BMI averages, suggesting a greater susceptibility to airway collapse due to craniofacial morphology rather than obesity (Ito et al. 2001; Hsu et al. 2005; Hui et al.

2003). Iranian studies also revealed elevated AHI values with similar BMI ranges, indicating that ethnicity‐specific anatomical traits may contribute to OSA severity (Bayat et al. 2017; Ahmadi et al.

2022). These findings highlight the importance of considering ethnic differences when evaluating anatomical risk factors and tailoring diagnostic thresholds.

Quantitative Synthesis

It was not possible to perform a meta‐analysis in this review due to the lack of homogeneous classifications in the patient samples reported across the different studies analysed. Variations in population characteristics hindered direct comparison of results, thus preventing a reliable quantitative synthesis.

Risk of Bias in Individual Studies

The methodological quality of the included studies was assessed using the Downs and Black checklist (Downs and Black 1998) (Figures 2, 3, 4). While the majority (20 out of 22) of the studies yielded statistically significant results (Figure 2A), the overall quality was variable. Only two studies were rated as low quality (Guilleminault et al.

1984; Hsu et al. 2005), while 46% were classified as fair or regular (Tangugsorn et al. 1995; Silva et al.

2014; Bayat et al. 2017; Valarelli et al. 2018; Ahmadi et al.

2022; Barrera et al. 2017; Ito et al. 2001; Rose et al.

2002; Verin et al. 2002; Gulati et al. 2010), and 45% were deemed to be of high quality (Battagel and L'Estrange 1996; Johal et al.

2007; Costa E Sousa and Dos Santos Gil 2013; Piccin et al. 2016; Bilici et al. 2018; Tanellari et al.

2022; Battagel et al. 2000; Johns et al. 1998; Banhiran et al.

2013; Hui et al. 2003) (Figure 2B).

External validity and information bias were evaluated (Figure 3). Regarding external validity (Figure 3A), the studies' generalizability is limited by the lack of detailed information about the participant populations. This makes it challenging to determine if the samples were representative.

Furthermore, only 4 studies (Tangugsorn et al. 1995; Battagel and L'Estrange 1996; Battagel et al. 2000; Ito et al.

2001) focused exclusively on male participants, potentially mitigating gender‐related biases (Figure 3B).

Internal validity and selection bias were analysed (Figure 4). In terms of internal validity, the studies demonstrated reasonable rigour. Blinding was implemented in 40% of the studies to minimise outcome assessment bias (Johal et al.

2007; Costa E Sousa and Dos Santos Gil 2013; Guilleminault et al. 1984; Johns et al. 1998; Verin et al.

2002; Hsu et al. 2005; Banhiran et al. 2013; Hui et al.

2003). All studies employed proper statistical methods and ensured simultaneous data collection in both groups. The reliability of the findings was high, with 85% of the sample showing consistent results.

Despite these strengths, the lack of random assignments to groups, which was common due to pre‐existing apnea diagnoses and AHI classifications, could be considered a limitation. Although 60% of studies accounted for confounding factors, and participant attrition was low in over 80% of cases, selection bias cannot be entirely ruled out.

Discussion

OSA is a multifactorial disorder influenced by genetic, neuromuscular, and anatomical factors, with obesity being the most prominent and well‐established risk factor (Du et al. 2024). Craniofacial abnormalities and soft tissue variations, such as reduced activity of the genioglossus and hypoglossal nerves, hypertrophy of tonsils and adenoids, and pharyngeal muscle dysfunction, contribute to airway narrowing and collapsibility (Shrikrishna et al.

2023). Among cephalometric parameters, H‐MP distance has emerged as the most consistently significant marker, correlating with apnea severity and distinguishing OSA patients from healthy controls (Bayat et al. 2017; Du et al.

2024; Jo et al. 2022; Udayakumar et al. 2024).

A lower hyoid position is associated with increased pharyngeal length and higher risk of upper airway (UAW) collapse (Jadoul et al. 2025; Ng et al. 2025; Skinner et al.

2002; Hoekema et al. 2007). Salman and Amatoury (2024) demonstrated a direct relationship between caudal hyoid displacement and increased upper airway collapsibility, quantified by elevated critical closing pressure (Bilici et al.

2018; Genta et al. 2014; Sforza et al. 2000).

These findings underscore the hyoid bone's pivotal role in maintaining airway patency and support its consideration as a diagnostic and therapeutic target in OSA management (Farhana et al. 2010). In a systematic review (Soares and Cahali 2023), analysing the modifications of the upper airways and craniofacial structures after weight loss, they concluded that by losing fat, the volume of soft tissues decreased, which increased the diameter of the airway and allowed for better airflow.

In turn, the hyoid bone moved closer to the PNS and the distance from this to the chin was also reduced. Importantly, beyond associations with obesity or craniofacial size, recent finite‐element simulations (Salman and Amatoury 2024) indicate that the natural or baseline position of the hyoid itself alters upper‐airway mechanics. Progressively lowering the hyoid, independent of other anatomic factors or any surgical manipulation, systematically increased upper‐airway closing pressure, consistent with a distinct ‘lower‐hyoid phenotype’ as a structural risk factor for OSA.

Regarding interethnic differences, several studies (Tangugsorn et al. 1995; Battagel and L'Estrange 1996; Bilici et al. 2018; Ahmadi et al.

2022; Tanellari et al. 2022; Guilleminault et al. 1984) consistently report a more inferior and posteriorly displaced hyoid bone in OSA patients compared to controls, contributing to airway narrowing and increased collapsibility.

In European cohorts (Tangugsorn et al. 1995; Battagel and L'Estrange 1996; Battagel et al. 2000), this displacement was associated with elongated pharyngeal structures and reduced retrolingual space.

Middle Eastern studies (Bayat et al. 2017; Bilici et al. 2018; Ahmadi et al.

2022) further confirmed the correlation between hyoid descent and OSA severity. In Brazilian populations (Silva et al. 2014; Valarelli et al.

2018), hyoid positioning was linked not only to UAW obstruction but also to altered muscular and swallowing dynamics. Asian studies (Ito et al. 2001; Hui et al.

2003) showed similar trends, although the degree of hyoid displacement appeared less pronounced, possibly due to differences in craniofacial morphology. Interestingly, Piccin et al. (Piccin et al.

2016) highlighted the role of craniocervical posture in modulating hyoid position, suggesting a multifactorial influence. Overall, while inferior hyoid displacement is a common finding in OSA patients, its anatomical impact and diagnostic relevance may vary across ethnic groups, reinforcing the need for population‐specific cephalometric norms. According to our results, a different craniofacial anatomy is observed in subjects with OSA compared to a healthy population due to significant differences in mostly of the included studies, according to a recent systematic review and meta‐analysis (Neelapu et al.

2017; Soares and Cahali 2023).

Vertical displacement of the hyoid bone has been extensively studied in the context of OSA, with H‐MP being the most frequently reported measurement. Across multiple studies (Ahmadi et al. 2022; Verin et al.

2002; Hui et al. 2003), H‐MP values were significantly higher in OSA patients compared to controls, indicating a downward shift of the hyoid. HH1 also showed increased values in OSA groups (Hui et al.

2003), correlating positively with AHI. Similarly, H‐Me and H‐S were elevated in severe OSA cases (Bayat et al. 2017; Bilici et al.

2018), suggesting inferior positioning. H‐N was found to be significantly associated with OSA severity (Rose et al. 2002), while HC3 showed mixed results, with some studies reporting no significant differences (Battagel et al.

2000) and others indicating increased values in OSA (Tanellari et al. 2022). Overall, these findings support the hypothesis that vertical descent of the hyoid bone is a consistent anatomical feature in OSA patients.

Recent studies have confirmed that H‐MP distance is a significant predictor of OSA severity. Jo et al. (2022) demonstrated that a longer H‐MP distance was associated with severe OSA, with a cut‐off value of 19.45 mm showing predictive power; patients above this threshold exhibited greater respiratory disturbance, lower oxygen saturation, and more fragmented sleep.

In accordance, Banhiran et al. (2013) established 18 mm as the distance between the hyoid bone and mandibular plane that indicated an increased risk of having an AHI≥ 15. Du et al.

(2024) similarly found that increased H‐MP distance correlated with higher AHI values and anatomical narrowing of the UAWs. Dong et al. (2025) reported that a lower and posterior hyoid position was significantly associated with elevated AHI in both male and female groups, with additional correlations to C3–mandibular plane distance.

Furthermore, Graizel‐Armoni et al. (2025) highlighted sex‐specific differences, noting that males had greater H‐MP values than females, and that the correlation between H‐MP and AHI was stronger in females. These findings support the use of H‐MP as a valuable anatomical marker in assessing OSA severity and tailoring diagnostic thresholds by sex and BMI.

The observed variability in vertical hyoid measurements across studies may be influenced by methodological differences, including patient positioning during imaging (supine vs. upright), BMI adjustment, and craniofacial morphology. For example, H‐MP values were consistently elevated in obese and non‐obese OSA patients, but some studies found no differences between non‐obese OSA and controls, suggesting that obesity may amplify vertical displacement. HH1 and H‐Me measurements also varied depending on gender and skeletal class, as noted in Bayat et al.

(2017) and Bilici et al. (2018). A plausible hypothesis is that the downward repositioning of the hyoid bone in OSA reflects compensatory anatomical changes aimed at maintaining UAWs patency in response to increased tongue mass and pharyngeal collapsibility.

These adaptations may differ based on individual craniofacial structure, severity of OSA, and body composition.

In relation to the sagittal position of the hyoid bone, the H‐Ph distance was significantly higher in OSA patients compared to controls and snorers in studies by Verin et al. (2002) and Costa E Sousa and Dos Santos Gil (2013), suggesting a posterior displacement of the hyoid in OSA. Regarding HC3, Battagel et al.

(2000) and Bilici et al. (2018) reported no significant differences between OSA and control groups, although Tanellari et al. (2022) found elevated HC3 values in OSA patients.

The H‐ENA measurement was assessed by Battagel and L'Estrange (1996) and Johal et al. (2007), both showing lower values in OSA groups, indicating a more posterior and inferior hyoid position. Finally, H‐B was significantly shorter in OSA patients in Battagel and L'Estrange (1996), while Tanellari et al.

(2022) reported higher values in OSA, reflecting variability across methodologies and populations. These findings collectively support the hypothesis that the hyoid bone assumes a more caudal and posterior position in OSA, potentially contributing to UAWs obstruction.

The variability observed across studies reporting sagittal measurements may be attributed to several methodological and anatomical factors. First, differences in imaging techniques (e.g., supine vs. upright cephalometry), landmark definitions, and measurement protocols can introduce inconsistencies. For instance, while Verin et al.

(2002) and Costa E Sousa and Dos Santos Gil (2013) found significant differences in H‐Ph values, Banhiran et al. (2013) reported no significant differences, possibly due to variations in pharyngeal wall delineation or patient positioning. Similarly, HC3 measurements showed conflicting results, with Battagel et al.

(2000) reporting no differences and Tanellari et al. (2022) identifying significant increases in OSA, which may reflect differences in cervical spine curvature or head posture during imaging. Regarding H‐ENA and H‐B, discrepancies may stem from skeletal class variations and craniofacial morphology.

The opposing reported values could be explained by population heterogeneity, including differences in BMI, ethnicity, and severity of OSA. A plausible hypothesis is that the hyoid bone adapts its position in response to UAWs collapsibility and tongue mass, shifting either posteriorly or anteriorly depending on compensatory mechanisms and anatomical constraints.

The anatomical positioning of the hyoid bone, both vertically and sagittally, plays a critical role in the pathophysiology of OSA. Numerous studies have demonstrated that OSA patients exhibit a significantly lower vertical position of the hyoid, as evidenced by increased H‐MP, HH1, H‐Me, H‐S, and H‐N measurements (Bayat et al. 2017; Hui et al.

2003; Jo et al. 2022; Graizel‐Armoni et al. 2025).

This inferior displacement is associated with increased UAWs collapsibility and correlates with apnea severity and oxygen desaturation. Sagittal measurements, including H‐B, H‐ENA, H‐Ph, and HC3, have shown more variable results. While some studies report a posterior shift of the hyoid in OSA patients (Battagel and L'Estrange 1996; Verin et al.

2002), others suggest an anterior repositioning (Tanellari et al. 2022), possibly reflecting compensatory mechanisms or population‐specific anatomical differences.

Surgical repositioning of the hyoid bone has demonstrated clinically significant improvements in UAWs patency and musculoskeletal dynamics in patients with OSA. Multiple studies evaluating pre‐ and postoperative changes consistently report that anterior and superior displacement of the hyoid bone following mandibular advancement correlates with increased pharyngeal UAWs volume and reduced AHI (Claudino et al. 2025; Madhan et al.

2022; Rückschloß et al. 2019; Souza Pinto et al. 2019; Riepponen et al.

2017). Salman and Amatoury's biomechanical model predicted that anterior surgical repositioning of the hyoid could reduce the critical closing pressure by up to 115%, while cranial repositioning led to a modest but progressive improvement. Madhan et al.

(2022) quantified that each 1 mm increase in the minimum hydraulic diameter was associated with a proportional increase in UAW volume, highlighting the mechanical relevance of hyoid movement. Sahoo et al. (2021) observed a reduction in H‐MP distance and a corresponding increase in UAW volume in Class II advancement surgeries, while Class III mandibular setbacks led to the opposite effect.

Kim et al. (2013) found that bimaxillary surgeries in Class III patients resulted in inferoposterior hyoid displacement and a significant reduction in UAW volume, with changes in palatal plane angle correlating with hyoid position. Emara, Elhamshary, et al.

(2022), Emara, Naser, et al. (2022) demonstrated that combining genioglossus advancement and tongue base reduction led to a postoperative decrease in H‐MP from 23.38 ± 1.14 mm to 15.17 ± 0.97 mm, and H‐Me from 39.47 ± 2.37 mm to 24.83 ± 2.43 mm, with an 81% surgical success rate defined by AHI < 20 or ≥ 50% reduction.

The heterogeneity in sagittal findings may be attributed to variations in craniofacial morphology, skeletal class, and imaging protocols. In contrast, vertical displacement appears to be a more consistent marker across studies and populations. Recent evidence supports the clinical relevance of hyoid positioning, due to that mandibular advancement devices have been shown to elevate the hyoid and improve AHI (Pae and Harper 2021).

Besides, computational models confirm that caudal hyoid positioning increases UAW collapsibility (Amatoury and Salman 2022). Taken together, these findings suggest that both vertical and sagittal hyoid positions contribute to airway obstruction in OSA, with vertical descent serving as a more robust and predictive anatomical indicator.

Recent advances in imaging and computational modelling have opened new pathways for understanding the biomechanical role of hyoid bone displacement in OSA. Salman and Amatoury (2024) developed a finite element model simulating various hyoid repositioning directions, demonstrating that anterior and anterior‐cranial displacements significantly reduce upper UAWs' collapsibility, while caudal repositioning increases it. These findings suggest that baseline hyoid position may influence surgical outcomes and should be considered in preoperative planning.

Similarly, Samaha et al. (2022) confirmed that anterior‐based surgical repositioning yields the greatest improvement in UAW patency, highlighting the need for personalised surgical strategies. While overnight PSG remains the gold standard for diagnosing OSA, its limitations in cost and accessibility have led to the exploration of alternative screening tools (Ng et al.

2025; Chung et al. 2016; Johns 1991; Chiu et al. 2017; Lyberg et al.

1989; Gungor et al. 2013; Samman et al. 2003).

Lateral cephalometric radiographs, commonly used in dental and orthodontic settings, offer a practical method for visualising UAW anatomy. However, their upright acquisition posture contrasts with the supine position during sleep, potentially affecting diagnostic accuracy. Studies have shown mixed results.

Pae et al. (1994) and Yildirim et al. (1991) reported posture‐related changes in UAW structures, while Hsu and Wu (2019) found no significant differences in healthy subjects.

These findings highlight the need to consider body posture when interpreting radiographs for OSA assessment.

In paediatric populations, Li et al. (2024) conducted a meta‐analysis showing that children with OSA exhibit distinct hyoid positioning and UAW morphology, suggesting early anatomical markers for diagnosis and intervention. Graizel‐Armoni et al.

(2025) emphasised sex‐specific differences in hyoid position and its correlation with AHI, indicating that future research should stratify analyses by gender. Moreover, Kurbanova et al. (2021) used CBCT and multi‐slice CT to assess hyoid bone morphology and volume, revealing significant differences between OSA and non‐OSA groups, which may inform new diagnostic criteria.

The strengths of this systematic review lie in its rigorous methodology, which included detailed inclusion criteria to ensure the selection of relevant, high‐quality studies, excluding those that are diagnosed with OSA using methods other than PSG, the gold standard for this condition. Additionally, by not filtering by publication year, the review captured the evolution of research on the topic, providing a more comprehensive understanding of the current state of knowledge. To minimise selection bias and ensure the generalizability of findings, studies with patients who had syndromic disorders, craniofacial anomalies, prior orthognathic surgery, or were undergoing treatment for apnea were excluded, allowing for an analysis of a more representative patient sample.

However, despite careful selection of the articles, certain limitations exist. The studies with the highest scientific evidence are randomised clinical trials, and none were found within the results of this review, as the study aim was to compare patients with apnea to healthy controls, which precludes randomization. Few studies divided their samples based on risk factors such as age, sex, and BMI, and there was no consensus on the classification of OSA severity according to AHI.

Additionally, X‐rays were taken with patients in an upright position with the head in a natural forward posture, which eases identification of craniofacial structures but does not reflect the natural sleeping posture during apnea episodes, and craniofacial differences among ethnicities were not considered. While a high number of studies of good scientific quality were seen, none reached an excellent level.

Further research with solid methodological designs is needed that considers risk factors and classifies participants according to OSA severity based on the AHI, following the classification by the AASM. Clinically, hyoid–mandibular plane distance may serve as a practical cephalometric marker of OSA severity that captures an underlying structural phenotype (lower hyoid) associated with greater collapsibility, complementing traditional risk factors and potentially informing patient selection for targeted therapies. Standardised imaging protocols and stratified analyses are needed to clarify the role of vertical and sagittal hyoid displacement in OSA pathogenesis.

Future research should integrate 3D imaging, computational modelling, and patient‐specific anatomical data to optimise therapeutic outcomes and refine surgical techniques targeting hyoid repositioning.

Conclusion

The results of this review suggest a relationship between the vertical position of the hyoid bone and the severity of OSA. The hyoid‐mandibular plane distance (H‐MP) appears as a potential biomarker, with larger values associated with more severe OSA. However, the relationship between the sagittal position of the hyoid and the severity of OSA is less clear and requires further investigation.

The H‐MP distance could be a useful cephalometric value for stratifying OSA risk, although additional studies are needed to confirm this association and explore other cephalometric markers. Future studies should explore ethnicity‐specific anatomical predictors of OSA severity to improve diagnostic accuracy. Additionally, integrating posture‐sensitive imaging and dynamic airway modelling may enhance personalised treatment planning.

Author Contributions

Carmen María Correa‐Morillo: writing – original draft, methodology, formal analysis, data curation, investigation. Paola Cardil París: writing – original draft, methodology, formal analysis, data curation, investigation. Marta Macarena Paz‐Cortés: conceptualization, writing – review and editing, supervision, validation.

Andrea Martín‐Vacas: writing – review and editing, methodology, formal analysis, data curation. Giovanni Giovannini: supervision, conceptualization, visualization.

Funding

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.

Related articles

PubMed

Machine learning prediction of the change in sleepiness following continuous positive airway pressure in obstructive sleep apnea: a multi-cohort analysis.

Abstract Study Objectives To predict the improvement in sleepiness following continuous positive airway pressure treatment for obstructive sleep apnea using baseline characteristics. Materials and Methods Data from five polysomnography cohorts (HomePAP, BestAIR, ABC, STAGES, and MESA; total n = 2332) were analyzed. Associations between polysomnographic metrics and Epworth Sleepiness Scale were assessed using multivariable linear regression. Cross-validated Least Absolute Shrinkage and Select...

PubMed

Ticagrelor-Associated Central Sleep Apnea: Mechanisms, Clinical Evidence, and Implications for Sleep Medicine.

Ticagrelor is a reversible P2Y12 receptor antagonist used for antiplatelet therapy and is well recognized to cause dyspnea. Emerging concern has focused on a possible association between ticagrelor and central sleep apnea, Cheyne-Stokes respiration, and other manifestations of ventilatory instability. This narrative review examines the clinical evidence for ticagrelor-associated central sleep apnea and the respiratory-control mechanisms proposed to underlie this association. A structured literat...