Skip to main content
NCBI home page
Laryngoscope Investigative Otolaryngology logoLink to Laryngoscope Investigative Otolaryngology
. 2025 May 14;10(3):e70154. doi: 10.1002/lio2.70154

The Obstructive Sleep Apnea Physical Exam: A Systematic Review and Meta‐Analysis

Nihal Punjabi 1,2, Roy W Qu 1, Alexandra Vacaru 3, Brianna Pandey 1, Shanalee Tamares 4, Jared C Inman 1,✉
PMCID: PMC12076602  PMID: 40370337

ABSTRACT

Introduction

This systematic review aimed to determine the clinical utility of various physical exam findings in the diagnosis of obstructive sleep apnea (OSA).

Methods

A systematic review of English articles identified from PubMed, Embase, CENTRAL, and Web of Science databases. Search terms included “sleep apnea,” “physical exam,” “polysomnography,” and all relevant synonyms. Two reviewers independently screened abstracts, reviewed full texts, and extracted data from all studies that presented associations between physical characteristics and apnea‐hypopnea index (AHI).

Results

A total of 35 studies representing 13,854 patients were included in this review. The mean difference between high AHI and low AHI groups was 4.09 kg/m2 (95% CI: 2.78–5.39) for BMI, 7.93 cm (3.59–12.28) for waist circumference, and 3.67 cm (2.64–4.71) for neck circumference. The odds ratios for having a high AHI were 2.44 (1.07–5.55) for macroglossia, 2.23 (1.68–2.96) for Mallampati > 2, 1.88 (1.67–2.11) for tonsil grade > 2, 3.99 (1.94–8.21) for pharyngeal grade > 2, and 1.57 (1.48–1.67) for enlarged uvula. Thyromental distance, retrognathia, Friedman grade, septal deviation, and enlarged turbinates were also assessed and were not found to be statistically significant between AHI groups.

Discussion

Several physical exams have strong evidence in the literature supporting their strength at differentiating patients with and without OSA. These should be used routinely among providers who treat OSA, regardless of specialty, to help guide decisions about recommending a sleep study and selecting appropriate treatment. Other physical characteristics may be better assessed through advanced exam techniques or require more research and standardization in the way they are assessed by practitioners.

Keywords: apnea‐hypopnea index, meta‐analysis, obstructive sleep apnea, physical exam, polysomnography

1. Introduction

Congenital and acquired characteristics, such as reduced size of the cranial bony structures and airway soft tissue redundancy, can make the upper airway more prone to collapse, leading to obstructive sleep apnea (OSA) [1]. While polysomnography is the gold standard for diagnosing OSA, it does not identify the anatomical sites of obstruction [2]. Furthermore, polysomnography is often difficult to access due to its complexity and cost. A thorough physical exam can aid in appropriate patient selection for sleep study referral and identify the anatomic level(s) of obstruction, guiding further advanced evaluations (e.g., cephalometric X‐rays or nasal endoscopy) and eventual treatment [3].

A clinical guideline produced by The Adult OSA Task Force of the American Academy of Sleep Medicine (AASM) suggests examining obesity, neck circumference, Friedman tongue position (FTP), Mallampati classification, macroglossia, tonsillar hypertrophy, enlarged/elongated uvula, lateral peritonsillar narrowing, high arched/narrow palate, retrognathia, overjet, nasal polyps, nasal deviation, nasal valve abnormalities, and turbinate hypertrophy when assessing a patient with potential OSA [4]. Yet, many of these physical characteristics are examined inconsistently and have conflicting evidence on their ability to accurately identify patients at risk for OSA. Several physical exam findings have been shown to be significantly correlated with OSA severity, and previous studies have attempted to combine historical and physical findings to make predictive OSA tools [5, 6, 7, 8, 9, 10]. However, these tools often only include a few physical findings and exclude others that may be important to routinely examine. A standardized, comprehensive physical exam to assess risk for OSA is still absent. Thus, while one anatomical area of obstruction may be treated, others may be overlooked, leading to inaccurate or incomplete treatment.

The purpose of this systematic review and meta‐analysis is to examine which physical exam characteristics are associated with AHI and can clinically distinguish OSA patients from healthy controls or patients with only mild OSA.

2. Methods

2.1. Search Strategy

This study follows the Preferred Reporting Items for Systematic Reviews and Meta‐Analyses [11]. A research librarian with expertise in conducting systematic reviews queried the PubMed, Embase, CENTRAL, and Web of Science libraries with the terms “sleep apnea,” “physical exam,” “polysomnography,” and all relevant synonyms. Full search terms are presented in the Supporting Information. All databases were searched from inception until August 2023. Deduplication was conducted in EndNote (Clarivate 2021) [12]. Systematic study selection was conducted using the Rayyan web tool (Rayyan Systems Inc. 2016).

2.2. Study Selection

Studies that correlated upper aerodigestive tract physical characteristics to AHI were included. Studies that examined physical characteristics that are not easily assessed in an average clinical setting and require advanced imaging or techniques (i.e., CINE MRI, cephalometric X‐rays or CT, endoscopy) were not included. The full inclusion/exclusion criteria are presented in Table 1. Two reviewers (N.P. and A.V.) independently reviewed abstracts and full texts. Conflicts were discussed between the reviewers and resolved by consensus. Quality assessment was conducted using the Quality Assessment of Diagnostic Accuracy Studies (QUADAS) tool [13].

TABLE 1.

Inclusion and exclusion criterion.

Criteria category Inclusion criteria Exclusion criteria
Participants Nonsyndrome adult subjects ≥ 18 years of age with a diagnosis of OSA by overnight PSG or overnight sleep study Children (age < 18 years) syndromic subjects.
Interventions/investigations Studies evaluating physical characteristics in adult OSA subjects using at least one clinically observable assessment of upper aerodigestive tract structures Studies evaluating physical characteristics using advanced modalities (i.e., cephalogram, endoscopy)
Comparator (control) If available, individuals without OSA or with mild OSA N/A
Outcome Studies providing physical exam parameters in a quantitative or qualitive measurement Studies not providing quantitative or qualitative measurements and focusing on treatment and surgery of OSA rather than diagnosis
Study design

Randomized controlled trials

Cohort studies

Case control studies

Cross‐sectional studies

Survey studies

Reviews articles

Systematic reviews

Meta‐analysis

Animal studies/testing

Case reports

Language other than English
Open in a new tab

Abbreviations: OSA, obstructive sleep apnea; PSG, polysomnography.

2.3. Data Extraction

Two reviewers (N.P. and B.P.) independently extracted data and compared results for accuracy. Extracted data included author, year of publication, country where the study was conducted, sample size, age, sex distribution, physical exam measurements between different AHI groups, and correlation p‐value between physical exam and AHI, if available.

2.4. Statistical Analysis

Studies that compared physical exam findings based on AHI classes (either moderate/severe OSA vs. mild OSA, or OSA vs. healthy controls) were included in a meta‐analysis. Physical exam findings with three or more studies were included in the pooled analysis—exam findings with fewer than three studies were qualitatively assessed for their associations with AHI. All statistical analysis was performed using the metafor (Version 4.0‐0) and meta (Version 6.2‐1) packages in R (Version 4.3.0; R Project for Statistical Computing). Random‐effect models using the Restricted maximum likelihood (REML) estimator were created for mean differences for continuous variables and odds ratios for binary variables. Heterogeneity was reported as low (Higgin's I 2 = 0%–25%), moderate (I 2 = 26%–50%), or high (I 2 > 50%), along with [2] and χ 2 statistics. Formal heterogeneity hypothesis testing was evaluated with Cochran's Q‐test. Outlier analysis was performed using influential case diagnostics and leave‐one‐out testing, which were reported in the body of the results but not removed from the analysis. Publication bias was assessed with funnel plots and Egger tests using a mixed‐effects meta‐regression model.

For continuous variables, averages for the high AHI and low AHI groups, mean differences, 95% confidence intervals, study weight, and pooled mean differences with reference line and confidence interval are reported. For binary variables (when continuous variables were not available), counts for the high AHI and low AHI groups and odds ratios are instead reported. A significance level of p = 0.05 was used for all analyses.

3. Results

Overall, 35 studies including 13,854 patients were included (Figure 1, Table 2) [8, 9, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46]. Of the 35 studies, 37.1% were published in otolaryngology journals, 25.7% in sleep medicine journals, and 11.4% in pulmonology journals. Other specialties represented include internal medicine, anesthesiology, and dentistry. Data was extracted for four continuous variables and 9 binary variables. These data were separated into two groups (high versus low AHI) and the mean differences or odds ratios were compared. The cutoffs for low versus high AHI ranged from 5 to 15 between the different studies. The studies included in the meta‐analysis and the physical exams they reviewed are summarized in Table 3. The remaining studies were not included in the meta‐analysis because they focused on physical exams where there were fewer than three studies evaluating them or because of heterogeneity in the reported statistics, but they are reported in Table 2.

FIGURE 1.

FIGURE 1

Open in a new tab

PRISMA diagram for search methodology.

TABLE 2.

Study characteristics.

Author, year, country n Age % Female Significant association w/AHI Nonsignificant association w/AHI
Ahn, 2015, Korea 64 43.8 (12.3) 0 BMI, Mallampati, Hyomental distance Friedman
Ashraf, 2022*, India 120 47.2 (14.1) 23.3 BMI, NC, Mallampati, Macroglossia, narrow dental arch Facial profile
Banabilh, 2010*, Malaysia 120 18–65 52.5 BMI, NC, Facial profile shape, Malocclusion class Palatal shape
Banhiran, 2013*, Thailand 283 48.7 (11.6) 41.7 BMI, NC, WC, Cricomental distance Friedman, Tonsil size, Hyomental distance, Thyromental distance
Barceló, 2011, Spain 301 18–82 28.9 BMI, NC, Friedman, Tonsil size, Uvula —
Dahlqvist, 2007, Sweden 801 50.5 (11.3) 25.6 Male: BMI, NC, Tonsil size, Uvula; Female: Retrognathia Male: Retrognathia; Female: NC, BMI, Tonsil size, Uvula, Macroglossia
de Athayde, 2023, Brazil 80 46.8 (13.0) 0 BMI, NC, WC, Mallampati —
He, 2022*, China 202 18–68 22.3 BMI, NC, Tonsil size, Pharyngeal grade, Peritonsillar narrowing, Septal deviation Hypertrophic inferior turbinates
He, 2021*, China 197 18–68 21 BMI, NC, Tonsil size, Pharyngeal grade, Peritonsillar narrowing Mallampati, Macroglossia, Deviated septum, Hypertrophic inferior turbinate
Hiremath, 1998*, Australia 30 35–76 13.3 Mallampati, NC, Decreased head extension Thyromental distance
Hukins, 2010, Australia 953 50 (6.4) 65 BMI, Mallampati —
Jara, 2019, USA 83 43 (12) 39 Tonsil size —
Laharnar, 2021, International 150 57.5 (12.3) 40 BMI, NC, WC, Mallampati —
Lai, 2014, Taiwan 51 27–62 9.8 Tonsil size —
Lam, 2005*, Canada/China 239 49 (12) 17 Mallampati, Thyromental angle BMI, NC, Thyromental distance
Lin, 2019, Taiwan 325 20–65 15.7 NC, WC, HC, BMI, Friedman, Tonsil size, Uvula, Thyromental distance Hyomental distance
Meskill, 2021, USA 1708 18–90 47.2 BMI, NC, Simmons Chin Press and Tongue Curl —
Nuckton, 2006*, USA 137 46 (12) 28 BMI, NC, Mallampati —
Oliveira, 2015*, Brazil 993 41.8 (0.9) 53.9 BMI, NC, Tonsil size, Uvula, Mallampati, Septal deviation Retrognathia, High arched palate, Dental occlusion, Inferior turbinate hypertrophy
Pang, 2016, USA 102 24–76 36.3 BMI, Mallampati, Friedman —
Park, 2021, Korea 3432 46.2 (13.5) 18.4 BMI, NC, WC, HC, Tonsil Size, Macroglossia, Uvula —
Pawar, 2018, India 193 25–82 23.3 BMI, NC, Mallampati, Tonsil size Retrognathia, Macroglossia, Uvula, High arched palate
Remya, 2016*, India 76 25–50 0 Mallampati BMI, NC, WC, HC, Thyromental distance, Thyromental angle
Schwab, 2017*, USA 860 47.4 (13.7) 52.1 BMI, Mallampati, Pharyngeal grade Tonsil size, Macroglossia
Soares, 2006*, Brazil 80 18–71 32.5 BMI, NC Retrognathia, Palatal Height
Svensson, 2005*, Sweden 132 51.3 (10.1) 100 BMI Tonsil size, Palatal height, Retrognathia, Septal deviation, Uvula, Macroglossia
Thong, 2008, Singapore 84 22–66 17.9 BMI, Mallampati Tonsil size
Ucok, 2011*, Turkey 98 48.8 (9.0) 0 Mallampati BMI, WC, HC
Weihu, 2010*, China 103 32–75 35 BMI, NC, Mallampati Tonsil size, Fauce's narrowness scale
Wu, 2010*, Taiwan 244 43.1 (12.1) 27 BMI Mallampati, Tonsil size
Wysocki, 2015, Poland 138 50.7 (11.6) 12.3 BMI, NC, WC, HC, Friedman Pirquet's Score
Yagi, 2009*, Japan 141 49 (14) 22 BMI, NC, Mallampati, Tonsil size, Fauce's narrowness scale Retroglossal space
Yilmaz, 2017*, Turkey 147 47.8 (9.0) 32 BMI, NC, Thyromental distance —
Yuceege, 2014*, Turkey 185 46.3 (10.3) 34.1 BMI, NC, Neck/thyromental ratio Thyromental distance
Zhang, 2022*, China 1002 35–55 17.5 BMI, NC, Uvula, Tonsil size, Pharyngeal stenosis, Macroglossia, Retrognathia Septal deviation, Turbinate hypertrophy, Friedman
Open in a new tab

Note: Studies marked with asterisks were included in the meta‐analysis. Age is presented as mean (SD) or range.

Abbreviations: HC, hip circumference; NC, neck circumference; WC, waist circumference.

TABLE 3.

Summary of results from meta‐analysis.

Physical exam # of studies Low AHI n High AHI n OR or mean difference (95% CI) Heterogeneity (I 2), p
Anthropometrics
BMI (kg/m2) 13 1569 1888 4.09 (2.78–5.39) 91.0%, < 0.001
Waist (cm) 3 165 292 7.93 (3.59–12.28) 0.0%, 0.525
Neck (cm) 13 1258 1434 3.67 (2.64–4.71) 98.6%, < 0.001
Craniofacial measurements
Thyromental (cm) 6 323 637 −0.78 (−1.84–0.28) 100.0%, < 0.001
Retrognathia 3 1005 1122 1.30 (0.78–2.18) 99.4%, < 0.001
Oral cavity/oropharynx
Macroglossia 5 1398 1711 2.44 (1.07–5.55) 99.7%, < 0.001

Mallampati Grade

(≤ 2 vs. > 2)

 7 1337 1286 2.23 (1.68–2.96) 86.3%, < 0.001

Friedman Grade

(≤ 2 vs. > 2)

 3 569 1014 5.94 (0.26–133.64) 91.6%, < 0.001

Tonsil Grade

(≤ 2 vs. > 2)

 7 1254 1079 1.88 (1.67–2.11) 1.2%, < 0.420

Pharyngeal Grade

(≤ 2 vs. > 2)

 3 317 505 3.99 (1.94–8.21) 97.9%, < 0.001
Enlarged Uvula 3 1005 1122 1.57 (1.48–1.67) 61.6%, < 0.001
Nasal cavity
Septal deviation 5 1132 1394 1.37 (0.87–2.16) 95.7%, < 0.001
Enlarged turbinates 7 1068 1326 1.02 (0.81–1.29) 98.3%, < 0.001
Open in a new tab

Note: Mean differences are presented for continuous variables, while odds ratios are presented for binary variables. Significant results are bolded.

3.1. Anthropometrics

3.1.1. BMI

There were 13 studies included for BMI (Figure 2). The pooled difference was 4.09 kg/m2 (2.78–5.39, 95% CI; p < 0.0001). Patients with high AHI tend to have higher BMIs than those with low AHI. There was high heterogeneity between studies (I 2 = 91.0%; τ 2 = 5.11; Q = 133.93; p < 0.0001). This suggests that the true difference in BMI between high and low AHI groups varies between studies. Case diagnostics found Banabilh et al. 2010 to potentially be an influential study (Supporting Information Figure 1). However, removing this study did not significantly change the pooled differences (Supporting Information Figure 1). The funnel plot did not suggest publication bias, nor did the Egger test (Supporting Information Figure 2, p = 0.316).

FIGURE 2.

FIGURE 2

Open in a new tab

Forest plots for anthropometric measurements. Horizontal lines depict 95% confidence intervals, square size represents the study weight, and the diamond width shows the pooled 95% confidence interval.

3.1.2. Waist Circumference

There were three studies included for waist circumference (Figure 2). The pooled difference was 7.93 cm (3.59–12.28, 95% CI; p = 0.0003). Patients with high AHI tend to have higher waist circumference than those with low AHI. There was low heterogeneity between studies (I 2 = 0.0%; τ 2 = 0; Q = 1.29; p = 0.525). This suggests all studies shared the same true difference in waist circumference. Ucok et al. 2011 was an influential study, but excluding it did not change pooled differences in waist circumference (Supporting Information Figure 1). The funnel plot did not suggest publication bias, nor did the Egger test (Supporting Information Figure 2, p = 0.994).

3.1.3. Neck Circumference

There were 13 studies included for neck circumference (Figure 2). The pooled difference was 3.67 cm (2.64–4.71, 95% CI; p < 0.0001). Patients with high AHI tend to have higher neck circumference than those with low AHI. There was high heterogeneity between studies (I 2 = 98.6%; τ 2 = 3.31; Q = 861.42; p < 0.0001). This suggests that the true difference in neck circumference between high and low AHI groups varies between studies. No influential cases were identified on case diagnostics and leave‐one‐out testing (Supporting Information Figure 1). The funnel plot did not suggest publication bias, nor did the Egger test (Supporting Information Figure 2, p = 0.641).

3.2. Craniofacial Measurements

3.2.1. Thyromental Distance

There were six studies included for thyromental distance (Figure 3). The pooled difference was −0.78 cm (−1.84 to 0.28, 95% CI; p = 0.147). There was no difference in thyromental distance between the high and low AHI groups. There was high heterogeneity between studies (I 2 = 100%; τ 2 = 1.74; Q = 10338.11; p < 0.0001). This suggests that the true difference in thyromental distance between high and low AHI groups varies between studies. Yilmaz et al. 2017 was an influential study (Supporting Information Figure 1). Excluding this study reduced the difference between groups, further supporting that there is no difference in thyromental distance (Supporting Information Figure 1). The funnel plot did not suggest publication bias, nor did the Egger test (Supporting Information Figure 2, p = 0.897). Of note, these studies represented assessments in the clinic and did not include radiographic or cephalometric exams.

FIGURE 3.

FIGURE 3

Open in a new tab

Forest plots for craniofacial measurements. Horizontal lines depict 95% confidence intervals, square size represents the study weight, and the diamond width shows the pooled 95% confidence interval.

3.2.2. Retrognathia

There were three studies included for retrognathia (Figure 3). The pooled odds ratio was 1.30 (0.78–2.18, 95% CI; p = 0.315). Patients with retrognathia are not at greater odds of high AHI than those without. There was high heterogeneity between studies (I 2 = 99.4%; τ 2 = 0.197, Q = 0.4433; p < 0.0001). Zhang et al. 2022 and Oliveira et al. 2015 were influential cases, but excluding either study did not significantly change the pooled odds ratio (Supporting Information Figure 1). The funnel plot did not suggest publication bias, nor did the Egger test (Supporting Information Figure 2, p = 0.891). Of note, these studies represented assessments in the clinic and did not include radiographic or cephalometric exams.

3.3. Oral Cavity/Oropharynx

3.3.1. Macroglossia

There were five studies included for macroglossia (Figure 4). The pooled odds ratio was 2.44 (1.07–5.55, 95% CI; p = 0.051). Patients with macroglossia are at greater odds of having high AHI than those without. There was high heterogeneity between studies (I 2 = 99.7%; τ 2 = 0.870, Q = 1375.514; p < 0.0001). Ashraf et al. 2022 was identified as an influential study, though excluding this study did not change the pooled odds ratio (Supporting Information Figure 1). The funnel plot did not suggest publication bias, nor did the Egger test (Supporting Information Figure 2, p = 0.293).

FIGURE 4.

FIGURE 4

Open in a new tab

Forest plots for oral cavity/oropharyngeal measurements. Horizontal lines depict 95% confidence intervals, square size represents the study weight, and the diamond width shows the pooled 95% confidence interval.

3.3.2. Mallampati

There were nine studies included for Mallampati classification (Figure 4). The pooled odds ratio was 3.20 (1.94–5.27, 95% CI; p < 0.0001). Again, Ashraf et al. 2022 was identified as an influential study and excluding this study lowered the pooled odds ratio by 0.8 to 2.45 (data not shown). The funnel plot suggested publication bias, specifically for Ashraf et al. 2022 and Ucok et al. 2011 (data not shown). The Egger test also suggested publication bias (p = 0.0226). If Ashraf et al. 2022 is excluded, a similarly significant publication bias is found and still identifies Ucok et al. 2011 as an influential study. Therefore, we opted to exclude both Ashraf et al. 2022 and Ucok et al. 2011. If both are excluded, the pooled odds ratio is 2.23 (1.68–2.96, 95% CI; p < 0.0001). Patients with high Mallampati class are at greater odds of having high AHI than those with low Mallampati class. There was high heterogeneity between studies (I 2 = 86.3%; τ 2 = 0.126, Q = 43.6; p < 0.0001). Oliveira et al. 2015 was identified as an influential study within these remaining seven studies, but excluding it did not significantly change the pooled outcome (Supporting Information Figure 1). Furthermore, the funnel plot did not suggest publication bias, nor did the Egger test (p = 0.240) (Supporting Information Figure 2).

3.3.3. Friedman Tongue Position

There were three studies included for Friedman classification (Figure 4). The pooled odds ratio was 5.94 (0.26–133.64, 95% CI; p = 0.262). Patients with a high FTP were not at significantly greater odds of having high AHI than those with low FTP. There was high heterogeneity between studies (I 2 = 91.6%; τ 2 = 7.23, Q = 23.85; p < 0.0001). Barcelo et al. 2011 was an influential case. Additionally, removing any single study would significantly change the pooled results (Supporting Information Figure 1). The funnel plot suggested publication bias by the Egger test (Supporting Information Figure 2, p = 0.026). However, because there were only three studies included in this analysis, this was difficult to interpret. Therefore, we included all three studies in this analysis but recommend a guarded interpretation.

3.3.4. Tonsillar Hypertrophy

There were three studies included for tonsil size (Figure 4). The pooled odds ratio was 1.88 (1.67–2.11, 95% CI; p < 0.0001). Patients with large tonsils are at greater odds of high AHI than those with small tonsils. There was low heterogeneity between studies (I 2 = 1.2%; τ 2 = 0, Q = 6.07; p < 0.0001). No studies were identified as outliers on influential case diagnostics or leave‐one‐out testing (Supporting Information Figure 1). The funnel plot did not suggest publication bias, nor did the Egger test (Supporting Information Figure 2, p = 0.356).

3.3.5. Pharyngeal Grade

There were three studies included for pharyngeal grade (Figure 4). The pooled odds ratio was 3.99 (1.94–8.21, 95% CI; p = 0.0002). Patients with high pharyngeal grade are at greater odds of having a high AHI than those with low pharyngeal grade. There was high heterogeneity between studies (I 2 = 97.9%; τ 2 = 0.392, Q = 93.14; p < 0.0001). Schwab et al. 2017 was an influential case. Excluding it significantly increased the pooled odds ratio to 5.80 (OR 2.19–9.41), but this finding was still statistically significant (Supporting Information Figure 1). The funnel plot suggested publication bias by the Egger test (Supporting Information Figure 2, p = 0.001). This was difficult to interpret as there were only three studies that reported pharyngeal grade. Therefore, we included all three studies in this analysis but recommend a guarded interpretation.

3.3.6. Enlarged Uvula

There were three studies included for enlarged uvula (Figure 4). The pooled odds ratio was 1.57 (1.48–1.67, 95% CI; p < 0.0001). Patients with an enlarged uvula were at greater odds of having high AHI than those with small uvulas. There was high heterogeneity between studies (I 2 = 61.6%; τ 2 = 0.0014), though formal heterogeneity testing failed to reach significance (Q = 5.21; p = 0.074). This suggests that the true odds ratio varies between studies. No influential cases were identified on case diagnostics and leave‐one‐out testing (Supporting Information Figure 1). The funnel plot did not suggest publication bias, nor did the Egger test (Supporting Information Figure 2, p = 0.732).

3.4. Nasal Cavity

3.4.1. Septal Deviation

There were five studies included for septal deviation (Figure 5). The pooled odds ratio was 1.37 (0.87–2.16, 95% CI; p = 0.173). Patients with septal deviation are not at greater odds of high AHI than those without. There was high heterogeneity between studies (I 2 = 95.7%; τ 2 = 0.261, Q = 93.82; p < 0.0001). Svensson et al. 2006 was an influential case. influential cases, but excluding either study did not significantly change the pooled odds ratio (Supporting Information Figure 1). The funnel plot did not suggest publication bias, nor did the Egger test (Supporting Information Figure 2, p = 0.419).

FIGURE 5.

FIGURE 5

Open in a new tab

Forest plots for nasal cavity measurements. Horizontal lines depict 95% confidence intervals, square size represents the study weight, and the diamond width shows the pooled 95% confidence interval.

3.4.2. Turbinate Hypertrophy

There were seven studies included for turbinate hypertrophy (Figure 5). The pooled odds ratio was 1.02 (0.81–1.29, 95% CI; p = 0.861). Patients with turbinate hypertrophy are not at greater odds of high AHI than those without. There was high heterogeneity between studies (I 2 = 98.3%; τ 2 = 0.0524, Q = 180.43; p < 0.0001). Zhang et al. 2022 and Oliveira et al. 2015 were influential cases, but excluding either study did not significantly change the pooled odds ratio (Supporting Information Figure 1). The funnel plot did not suggest publication bias, nor did the Egger test (Supporting Information Figure 2, p = 0.817).

4. Quality Assessment

No studies were excluded from this study through the quality assessment. The study populations were diverse, especially since the included studies were conducted in many different countries, which may introduce heterogeneity but increases the representation of the global population of patients with OSA. Variation in the studies was also found with respect to their definitions of hypopneas: the AASM defines a hypopnea as a reduction in airflow with a 3% desaturation or cortical arousal; however, a 4% desaturation without consideration of cortical activity can also be used. AHI can vary based on which definition is used [47]. The studies included in this review may differ in which definition they used, as some did not specify. Likewise, some studies did not describe their physical exam protocols in detail, as physical exams are rarely standardized, contributing to poorer quality.

5. Discussion

The anatomic source of obstruction in OSA can occur to varying degrees at multiple sites in the upper airway. Multilevel obstruction has been associated with more severe OSA [48]. Identifying only one area of obstruction in a patient with multilevel disease may lead to incomplete treatment, which may partially explain the high rates of treatment failure seen in OSA [49]. As such, it is imperative to qualitatively assess and quantitatively measure physical characteristics in all patients with suspected OSA through a detailed anatomic physical exam.

This systematic review and meta‐analysis identified several physical exam findings that have statistically significant associations with AHI. Anthropometric measures, including BMI, waist circumference, and neck circumference, were all significantly higher in patients with higher AHI. Craniofacial measurements, including retrognathia and thyromental distance, were not found to be significantly different between AHI groups. Among measurements of oral cavity/oropharyngeal crowding, Mallampati and macroglossia were found to be significant, while FTP was not. The other oropharyngeal physical characteristics included in this analysis were all significant, with patients having increased odds of having a higher AHI if they had > 2 tonsil grade, > 2 pharyngeal grade, or an enlarged uvula. Finally, nasal septal deviation and inferior turbinate hypertrophy were not significant.

BMI and neck circumference are well‐established risk factors for OSA. Obesity increases the size of the soft tissue structures surrounding the airway through fat deposition [50]. Furthermore, increased abdominal fat decreases lung volumes, which reduces pharyngeal wall tension forces and makes collapse more likely [51]. The strong association between OSA and obesity is further evidenced by the fact that a 10% increase in weight is associated with a sixfold increase in the risk of developing OSA [52]. Along with BMI, neck circumference and waist circumference have both been identified as markers of central obesity [53]. The evidence from this review suggests that these measurements should routinely be included when examining patients with suspected OSA.

The craniofacial structural characteristics examined in this study through clinical assessment—retrognathia and thyromental distance—were not found to be significantly associated with AHI. These results suggest that more advanced techniques may be better suited to accurately examine these physical characteristics than clinical soft tissue measurements. A previous meta‐analysis on cephalometric radiograph measurements found statistically significant differences in both mandibular length (retrognathia) and hyomental distance between patients with lower AHI compared to those with higher AHI [3]. Furthermore, nonsyndromic craniofacial characteristics that contribute to OSA are relatively rare compared to more common causes like obesity, making it difficult to detect statistically significant associations for these features. Similarly, in the nasal cavity, septal deviation and inferior turbinate hypertrophy were also not significant, suggesting nasal endoscopy may be needed to better evaluate this area. No studies examined the hypopharynx, which was expected, as this level is not readily observable through traditional examination techniques. Drug induced sleep endoscopy can be used to better examine these sites and has been shown to decrease failure rates for patients undergoing surgical treatment for OSA [54]. In addition to providing a better assessment of certain upper airway structures, endoscopy can also rule out more serious pathologies; a previous scoping review found that although rare, head and neck neoplasms can present as OSA. On average, patients had symptoms of OSA for 2.5 years until the correct diagnosis was made using endoscopy [55].

Of the five studies that examined macroglossia, different measurement techniques were used: some defined ridging on the tongue due to tooth pressure as evidence of an enlarged tongue, while others used the relationship of the tongue to the occlusion plane. There was significant heterogeneity in the odds ratios found in these studies, ranging from 0.94 to 10.80. This variability suggests that while macroglossia is a well‐known risk factor for OSA, using unstandardized, subjective assessments may result in inaccurate classification of patients. A previous study found high rates of inter‐observer discordance when making treatment decisions for OSA [56]. The variability in exam techniques may explain this.

When using more standardized assessments that consider tongue size, Mallampati grade was significant while FTP was not. Both these scales assess crowding of the oral cavity and oropharynx by grading the visibility of the tonsillar pillars, uvula, and soft palate on a scale of 1–4; however, in Mallampati grade, the tongue is maximally extended while it remains in a neutral position for FTP. Even when using these standardized scales, some studies have found interrater reliability to be poor. Both these scales have been shown to be associated with OSA in past studies, but the correlations have been weak. However, having a score of 1 or 2 does appear to help rule out OSA better than a score of 3 or 4 helps rule it in [57]. Consequently, these previous results in the context of the present study suggest that low Mallampati grade may be used to decrease a clinician's index of suspicion for OSA.

The specialties that treat OSA are diverse, with this review identifying studies published in the otolaryngology, sleep medicine, pulmonology, dentistry, anesthesiology, neurology, and internal medicine literature. A comprehensive physical exam may ensure that patients with OSA are appropriately referred to specialists when advanced diagnostics or differential treatment strategies are needed. Once a patient is referred to a specialist, a broad physical exam should still be employed in case further referral to another specialty is required. A previous study found that only 35% of patients who have failed treatment with continuous positive airway pressure (CPAP) are referred for further evaluation or alternative management strategies [58]. When untreated, the sequelae of OSA are significant, including worsening hypertension and ischemic heart disease [59]. Thus, a comprehensive physical exam is especially crucial in patients who have failed CPAP therapy, as identifying anatomic sites of obstruction can guide future interventions, such as surgery. Importantly, surgical success in OSA has been shown to depend on preoperative physical exam findings [60, 61]. Comprehensive treatment for patients who have failed CPAP may also be achieved through a multidisciplinary “sleep board,” where the “tumor board” model for cancer care is followed to ensure adequate communication and evaluation completeness between specialists [62].

The main limitation to this study was the high heterogeneity. This likely stems from the fact that the OSA patient population is very heterogeneous and the OSA physical exam is vulnerable to interrater variation. The physical characteristics of patients with OSA have been shown to vary widely based on ethnicity/race and sex [63, 64]. One study included in this review that presented its results separately for males and females found that different characteristics were associated with AHI between the sexes (e.g., BMI and neck circumference were significant in males but not females) [18].

Additionally, many of the physical characteristics assessed do not have a standardized protocol for measurement. This led to significant variation between studies in the exact technique that was used to measure each parameter. Examples of the different techniques used are presented in Supporting Information Table 1. The variation was especially apparent for physical exam findings that do not have a commonly used standardized rating system; for example, for features such as uvula size and jaw position, some authors had specific measurements they used to identify potentially pathologic variants, while other authors simplified subjectively identified patients without concrete measurements. Even in exams that have a standardized protocol, interrater reliability may arise. Previous studies have found varying interrater reliability for Mallampati grade (κ = 0.31–1.00), FTP (κ = 0.36–0.82), and Tonsil size (intraclass correlation coefficient = 0.763) [57, 65, 66, 67, 68]. The interrater reliability for many other physical exam findings has not been studied. This highlights the need for standardized exam techniques to ensure results can be interpreted between providers and appropriate treatment protocols are followed.

Finally, publication bias was observed in FTP and pharyngeal grade. Consequently, the interpretability of these results is limited, but no studies were excluded from these analyses since the Egger test is less reliable when used in analyses with less than 10 studies [69]. Despite these limitations, this review demonstrated that many physical exam findings correlate with AHI and may be important components of a structured, multilevel, reproducible physical exam to diagnose the underlying cause of OSA. More research is needed to optimize exam technique and create standardized exam protocols.

6. Conclusion

A comprehensive physical exam for patients with OSA is important to guide adequate treatment. While polysomnography may confirm a diagnosis of OSA, it does not preclude a careful and thorough anatomically based physical exam. Moreover, weight management and CPAP therapy have varied tolerances and adherence over time, further underscoring the need for an accurate, anatomically based exam to guide other management options. This systematic review and meta‐analysis found several significant associations between physical exam characteristics and AHI. These may be important to include in a comprehensive physical exam. Several other physical findings may also have clinical utility but may be better assessed through more advanced techniques, such as endoscopy or cephalometric radiographs. The heterogeneity found in this study also highlights the need for more standardization in the way the physical exam is executed to ensure treatment decisions are consistent and appropriate.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Data S1. Supporting Information.

LIO2-10-e70154-s003.docx (16KB, docx)

Supporting Information Figure 1 Leave one out and influence plots for all physical exam findings examined (BMI, waist circumference, neck circumference, thyromental distance, retrognathia, macroglossia, Mallampati grade, Friedman grade, tonsil grade, pharyngeal grade, enlarged uvula, septal deviation, and enlarged turbinates).

LIO2-10-e70154-s002.docx (290.1KB, docx)

Supporting Information Figure 2 Funnel plots for all physical exam findings examined (BMI, waist circumference, neck circumference, thyromental distance, retrognathia, macroglossia, Mallampati grade, Friedman grade, tonsil grade, pharyngeal grade, enlarged uvula, septal deviation, and enlarged turbinates).

LIO2-10-e70154-s001.docx (148.4KB, docx)

Supporting Information Table 1 Examples of how each parameter was measured among the different studies, demonstrating that different techniques were often used.

LIO2-10-e70154-s004.docx (16.4KB, docx)

Acknowledgments

The authors have nothing to report.

References

  • 1. Dempsey J. A., Veasey S. C., Morgan B. J., and O'Donnell C. P., “Pathophysiology of Sleep Apnea,” Physiological Reviews 90, no. 1 (2010): 47–112, 10.1152/physrev.00043.2008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2. Rundo J. V., “Obstructive Sleep Apnea Basics,” Cleveland Clinic Journal of Medicine 86, no. 9 S1 (2019): 2–9, 10.3949/ccjm.86.s1.02. [DOI] [PubMed] [Google Scholar]
  • 3. Neelapu B. C., Kharbanda O. P., Sardana H. K., et al., “Craniofacial and Upper Airway Morphology in Adult Obstructive Sleep Apnea Patients: A Systematic Review and Meta‐Analysis of Cephalometric Studies,” Sleep Medicine Reviews 31 (2017): 79–90, 10.1016/j.smrv.2016.01.007. [DOI] [PubMed] [Google Scholar]
  • 4. Epstein L. J., Kristo D., Strollo P. J., et al., “Clinical Guideline for the Evaluation, Management and Long‐Term Care of Obstructive Sleep Apnea in Adults,” Journal of Clinical Sleep Medicine 5, no. 3 (2009): 263–276. [PMC free article] [PubMed] [Google Scholar]
  • 5. Unal Y., Ozturk D. A., Tosun K., and Kutlu G., “Association Between Obstructive Sleep Apnea Syndrome and Waist‐To‐Height Ratio,” Sleep & Breathing 23, no. 2 (2019): 523–529, 10.1007/s11325-018-1725-4. [DOI] [PubMed] [Google Scholar]
  • 6. Shigeta Y., Ogawa T., Tomoko I., Clark G. T., and Enciso R., “Soft Palate Length and Upper Airway Relationship in OSA and Non‐OSA Subjects,” Texas Dental Journal 130, no. 3 (2013): 203–211. [PubMed] [Google Scholar]
  • 7. Kang K. T., Weng W. C., Lee C. H., Hsiao T. Y., Lee P. L., and Hsu W. C., “Clinical Risk Assessment Model for Pediatric Obstructive Sleep Apnea,” Laryngoscope 126, no. 10 (2016): 2403–2409, 10.1002/lary.25912. [DOI] [PubMed] [Google Scholar]
  • 8. Banhiran W., Junlapan A., Assanasen P., and Chongkolwatana C., “Physical Predictors for Moderate to Severe Obstructive Sleep Apnea in Snoring Patients,” Sleep & Breathing 18, no. 1 (2014): 151–158, 10.1007/s11325-013-0863-y. [DOI] [PubMed] [Google Scholar]
  • 9. Lin H. C., Lai C. C., Lin P. W., et al., “Clinical Prediction Model for Obstructive Sleep Apnea Among Adult Patients With Habitual Snoring,” Otolaryngology and Head and Neck Surgery 161, no. 1 (2019): 178–185, 10.1177/0194599819839999. [DOI] [PubMed] [Google Scholar]
  • 10. Deflandre E., Degey S., Brichant J. F., Poirrier R., and Bonhomme V., “Development and Validation of a Morphologic Obstructive Sleep Apnea Prediction Score: The DES‐OSA Score,” Anesthesia and Analgesia 122, no. 2 (2016): 363–372, 10.1213/ANE.0000000000001089. [DOI] [PubMed] [Google Scholar]
  • 11. Moher D., Liberati A., Tetzlaff J., and Altman D. G., “Preferred Reporting Items for Systematic Reviews and Meta‐Analyses: The PRISMA Statement,” PLoS Medicine 6, no. 7 (2009): e1000097, 10.1371/journal.pmed.1000097. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12. Bramer W. M., Giustini D., de Jonge G. B., Holland L., and Bekhuis T., “De‐Duplication of Database Search Results for Systematic Reviews in EndNote,” Journal of the Medical Library Association 104, no. 3 (2016): 240–243, 10.5195/jmla.2016.24. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13. Whiting P., Rutjes A. W., Reitsma J. B., Bossuyt P. M., and Kleijnen J., “The Development of QUADAS: A Tool for the Quality Assessment of Studies of Diagnostic Accuracy Included in Systematic Reviews,” BMC Medical Research Methodology 3 (2003): 25, 10.1186/1471-2288-3-25. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14. Ahn S. H., Kim J., Min H. J., et al., “Tongue Volume Influences Lowest Oxygen Saturation but Not Apnea‐Hypopnea Index in Obstructive Sleep Apnea,” PLoS One 10, no. 8 (2015): e0135796, 10.1371/journal.pone.0135796. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15. Ashraf A., Menon I., Gupta R., Arora V., Ahsan I., and Das D., “Oral Findings as Predictors of Obstructive Sleep Apnea‐A Case‐Control Study,” Journal of Family Medicine and Primary Care 11, no. 9 (2022): 5263–5267, 10.4103/jfmpc.jfmpc_582_21. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16. Banabilh S. M., Samsudin A. R., Suzina A. H., and Dinsuhaimi S., “Facial Profile Shape, Malocclusion and Palatal Morphology in Malay Obstructive Sleep Apnea Patients,” Angle Orthodontist 80, no. 1 (2010): 37–42, 10.2319/011509-26.1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17. Barceló X., Mirapeix R. M., Bugés J., Cobos A., and Domingo C., “Oropharyngeal Examination to Predict Sleep Apnea Severity,” Archives of Otolaryngology – Head & Neck Surgery 137, no. 10 (2011): 990–996, 10.1001/archoto.2011.176. [DOI] [PubMed] [Google Scholar]
  • 18. Dahlqvist J., Dahlqvist Å., Dahlqvist J., et al., “Physical Findings in the Upper Airways Related to Obstructive Sleep Apnea in Men and Women,” Acta Oto‐Laryngologica 127, no. 6 (2007): 623–630, 10.1080/00016480600987842. [DOI] [PubMed] [Google Scholar]
  • 19. de Athayde R. A. B., Colonna L. L. I., Schorr F., et al., “Tongue Size Matters: Revisiting the Mallampati Classification System in Patients With Obstructive Sleep Apnea,” Jornal Brasileiro de Pneumologia 49, no. 2 (2023): e20220402, 10.36416/1806-3756/e20220402. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20. He S., Li Y., Xu W., et al., “The Predictive Value of Photogrammetry for Obstructive Sleep Apnea,” Journal of Clinical Sleep Medicine 17, no. 2 (2021): 193–202, 10.5664/jcsm.8836. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21. He S., Li Y., Xu W., and Han D., “Using Clinical Data to Predict Obstructive Sleep Apnea,” Journal of Thoracic Disease 14, no. 2 (2022): 227–237, 10.21037/jtd-20-3139. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22. Hiremath A. S., Hillman D. R., James A. L., Noffsinger W. J., Platt P. R., and Singer S. L., “Relationship Between Difficult Tracheal Intubation and Obstructive Sleep Apnoea,” British Journal of Anaesthesia 80, no. 5 (1998): 606–611, 10.1093/bja/80.5.606. [DOI] [PubMed] [Google Scholar]
  • 23. Hukins C., “Mallampati Class is not Useful in the Clinical Assessment of Sleep Clinic Patients,” Journal of Clinical Sleep Medicine 6, no. 6 (2010): 545–549. [PMC free article] [PubMed] [Google Scholar]
  • 24. Jara S. M. and Weaver E. M., “Association of Palatine Tonsil Size and Obstructive Sleep Apnea in Adults,” Laryngoscope 128, no. 4 (2018): 1002–1006, 10.1002/lary.26928. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25. Laharnar N., Herberger S., Prochnow L. K., et al., “Simple and Unbiased OSA Prescreening: Introduction of a New Morphologic OSA Prediction Score,” Nature and Science of Sleep 13 (2021): 2039–2049, 10.2147/NSS.S333471. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26. Lai C. C., Friedman M., Lin H. C., et al., “Objective Versus Subjective Measurements of Palatine Tonsil Size in Adult Patients With Obstructive Sleep Apnea/Hypopnea Syndrome,” European Archives of Oto‐Rhino‐Laryngology 271, no. 8 (2014): 2305–2310, 10.1007/s00405-014-2944-3. [DOI] [PubMed] [Google Scholar]
  • 27. Lam B., Ip M., Tench E., and Ryan C., “Craniofacial Profile in Asian and White Subjects With Obstructive Sleep Apnoea,” Thorax 60, no. 6 (2005): 504–510, 10.1136/thx.2004.031591. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28. Meskill G. J., Kincheloe K., Simmons J. H., and Meskill S. D., “An Easy Maneuver to Screen for Moderate‐To‐Severe Obstructive Sleep Spnea: The Simmons Chin Press and Tongue Curl,” Sleep Science 14 (2021): 185–188, 10.5935/1984-0063.20200092. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29. Nuckton T. J., Glidden D. V., Browner W. S., and Claman D. M., “Physical Examination: Mallampati Score as an Independent Predictor of Obstructive Sleep Apnea,” Sleep 29, no. 7 (2006): 903–908, 10.1093/sleep/29.7.903. [DOI] [PubMed] [Google Scholar]
  • 30. Soares Oliveira M. C., Tufik S., Louise Martinho Haddad F., Santos‐Silva R., Gregório L. C., and Bittencourt L., “Systematic Evaluation of the Upper Airway in a Sample Population,” Otolaryngology–Head and Neck Surgery 153, no. 4 (2015): 663–670, 10.1177/0194599815577598. [DOI] [PubMed] [Google Scholar]
  • 31. Pang K. P., Terris D. J., and Podolsky R., “Severity of Obstructive Sleep Apnea: Correlation With Clinical Examination and Patient Perception,” Otolaryngology–Head and Neck Surgery 135, no. 4 (2006): 555–560, 10.1016/j.otohns.2006.03.044. [DOI] [PubMed] [Google Scholar]
  • 32. Park D. Y., Kim J. S., Park B., and Kim H. J., “Risk Factors and Clinical Prediction Formula for the Evaluation of Obstructive Sleep Apnea in Asian Adults,” PLoS One 16, no. 2 (2021): e0246399, 10.1371/journal.pone.0246399. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33. Pawar K., Goyal A., Khurana A., and Pakhare A., “Correlation of Various Anthropometric and Craniofacial Variables With Severity of Obstructive Sleep Apnea in Indian Population,” Sleep and Vigilance 2, no. 2 (2018): 127–133, 10.1007/s41782-018-0045-x. [DOI] [Google Scholar]
  • 34. Remya K. J., Mathangi K., Mathangi D. C., et al., “Predictive Value of Craniofacial and Anthropometric Measures in Obstructive Sleep Apnea (OSA),” Cranio 35, no. 3 (2017): 162–167, 10.1080/08869634.2016.1206701. [DOI] [PubMed] [Google Scholar]
  • 35. Schwab R. J., Leinwand S. E., Bearn C. B., et al., “Digital Morphometrics,” Chest 152, no. 2 (2017): 330–342, 10.1016/j.chest.2017.05.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36. Soares M. C. M., de Azeredo Bittencourt L. R., Zonato A. I., and Gregório L. C., “Application of the Kushida Morphometric Model in Patients With Sleep‐Disordered Breathing,” Brazilian Journal of Otorhinolaryngology 72, no. 4 (2015): 541–548, 10.1016/S1808-8694(15)31002-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37. Svensson M., Holmstrom M., Broman J. E., and Lindberg E., “Can Anatomical and Functional Features in the Upper Airways Predict Sleep Apnea? A Population‐Based Study in Females,” Acta Oto‐Laryngologica 126, no. 6 (2006): 613–620, 10.1080/00016480500468984. [DOI] [PubMed] [Google Scholar]
  • 38. Thong J. F. and Pang K. P., “Clinical Parameters in Obstructive Sleep Apnea: Are There Any Correlations?,” Journal of Otolaryngology ‐ Head & Neck Surgery 37, no. 6 (2008): 894–900. [PubMed] [Google Scholar]
  • 39. Ucok K., Aycicek A., Sezer M., et al., “Resting Metabolic Rate and Anthropometric Measurements in Male Sleep Apnea Patients,” Internal Medicine 50, no. 8 (2011): 833–838, 10.2169/internalmedicine.50.4779. [DOI] [PubMed] [Google Scholar]
  • 40. Weihu C., Jingying Y., Demin H., and Boxuan W., “Relationship of Body Position, Upper Airway Morphology, and Severity of Obstructive Sleep Apnea/Hypopnea Syndrome Among Chinese Patients,” Acta Oto‐Laryngologica 131, no. 2 (2011): 173–180, 10.3109/00016489.2010.518158. [DOI] [PubMed] [Google Scholar]
  • 41. Wu M. J., Ho C. Y., Tsai H. H., Huang H. M., Lee P. L., and Tan C. T., “Retropalatal Müller Grade is Associated With the Severity of Obstructive Sleep Apnea in Non‐Obese Asian Patients,” Sleep & Breathing 15, no. 4 (2011): 799–807, 10.1007/s11325-010-0441-5. [DOI] [PubMed] [Google Scholar]
  • 42. Wysocki J., Krasny M., and Prus M., “Morphological Predictors of Sleep Apnoea Severity,” Folia Morphologica 75, no. 1 (2016): 107–111, 10.5603/FM.a2015.0069. [DOI] [PubMed] [Google Scholar]
  • 43. Yagi H., Nakata S., Tsuge H., et al., “Morphological Examination of Upper Airway in Obstructive Sleep Apnea,” Auris, Nasus, Larynx 36, no. 4 (2009): 444–449, 10.1016/j.anl.2008.11.003. [DOI] [PubMed] [Google Scholar]
  • 44. Yilmaz A. and Akcaalan M., “What Can Anthropometric Measurements Tell Us About Obstructive Sleep Apnoea?,” Folia Morphologica 76, no. 2 (2017): 301–306, 10.5603/FM.a2016.0058. [DOI] [PubMed] [Google Scholar]
  • 45. Yuceege M., Firat H., Altintas N., Mutlu M., and Ardic S., “The Utility of Neck/Thyromental Ratio in Defining Low‐Risk Patients With Obstructive Sleep Apnea in Sleep Clinics,” European Archives of Oto‐Rhino‐Laryngology 271, no. 9 (2014): 2575–2581, 10.1007/s00405-014-3062-y. [DOI] [PubMed] [Google Scholar]
  • 46. Zhang C., Chen M., Shen Y., Gong Y., Ma J., and Wang G., “Impact of Upper Airway Characteristics on Disease Severity and CPAP Therapy in Chinese Patients With OSA: An Observational Retrospective Study,” Frontiers in Neurology 13 (2022): 767336, 10.3389/fneur.2022.767336. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47. Kapur V. K., Auckley D. H., Chowdhuri S., et al., “Clinical Practice Guideline for Diagnostic Testing for Adult Obstructive Sleep Apnea: An American Academy of Sleep Medicine Clinical Practice Guideline,” Journal of Clinical Sleep Medicine 13, no. 3 (2017): 479–504, 10.5664/jcsm.6506. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48. Phua C. Q., Yeo W. X., Su C., and Mok P. K. H., “Multi‐Level Obstruction in Obstructive Sleep Apnoea: Prevalence, Severity and Predictive Factors,” Journal of Laryngology & Otology 131, no. 11 (2017): 982–986, 10.1017/S0022215117001906. [DOI] [PubMed] [Google Scholar]
  • 49. Xia F. and Sawan M., “Clinical and Research Solutions to Manage Obstructive Sleep Apnea: A Review,” Sensors (Basel) 21, no. 5 (2021): 1784, 10.3390/s21051784. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50. Schwartz A. R., Patil S. P., Laffan A. M., Polotsky V., Schneider H., and Smith P. L., “Obesity and Obstructive Sleep Apnea,” Proceedings of the American Thoracic Society 5, no. 2 (2008): 185–192, 10.1513/pats.200708-137MG. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51. Drager L. F., Togeiro S. M., Polotsky V. Y., and Lorenzi‐Filho G., “Obstructive Sleep Apnea: A Cardiometabolic Risk in Obesity and the Metabolic Syndrome,” Journal of the American College of Cardiology 62, no. 7 (2013): 569–576, 10.1016/j.jacc.2013.05.045. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52. Peppard P. E., Young T., Palta M., Dempsey J., and Skatrud J., “Longitudinal Study of Moderate Weight Change and Sleep‐Disordered Breathing,” Journal of the American Medical Association 284, no. 23 (2000): 3015–3021, 10.1001/jama.284.23.3015. [DOI] [PubMed] [Google Scholar]
  • 53. Kang H. H., Kang J. Y., Ha J. H., et al., “The Associations Between Anthropometric Indices and Obstructive Sleep Apnea in a Korean Population,” PLoS One 9, no. 12 (2014): e114463, 10.1371/journal.pone.0114463. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54. Qi Y., Zhao Y., Yan Y., and Wu D., “Surgical Failure Guided by DISE in Patients With Obstructive Sleep Apnea: A Systematic Review and Meta‐Analysis,” European Archives of Oto‐Rhino‐Laryngology 281, no. 7 (2024): 3333–3343, 10.1007/s00405-024-08484-5. [DOI] [PubMed] [Google Scholar]
  • 55. Moore P., Cardemil F., Hayward N. J., and Flatman S., “Scoping Review of Head and Neck Neoplasms Presenting With Obstructive Sleep Apnoea: The Importance of Flexible Nasendoscopy,” ANZ Journal of Surgery 91, no. 1–2 (2021): 55–61, 10.1111/ans.16090. [DOI] [PubMed] [Google Scholar]
  • 56. Lugo V. M., Torres M., Garmendia O., et al., “Intra‐ and Inter‐Physician Agreement in Therapeutic Decision for Sleep Apnea Syndrome,” Archivos de Bronconeumología (English Edition) 56, no. 1 (2020): 18–22, 10.1016/j.arbr.2019.02.025. [DOI] [PubMed] [Google Scholar]
  • 57. Yu J. L. and Rosen I., “Utility of the Modified Mallampati Grade and Friedman Tongue Position in the Assessment of Obstructive Sleep Apnea,” Journal of Clinical Sleep Medicine 16, no. 2 (2020): 303–308, 10.5664/jcsm.8188. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 58. Russell J. O., Gales J., Bae C., and Kominsky A., “Referral Patterns and Positive Airway Pressure Adherence Upon Diagnosis of Obstructive Sleep Apnea,” Otolaryngology–Head and Neck Surgery 153, no. 5 (2015): 881–887, 10.1177/0194599815596169. [DOI] [PubMed] [Google Scholar]
  • 59. Fisher D., Pillar G., Malhotra A., Peled N., and Lavie P., “Long‐Term Follow‐Up of Untreated Patients With Sleep Apnoea Syndrome,” Respiratory Medicine 96, no. 5 (2002): 337–343, 10.1053/rmed.2001.1277. [DOI] [PubMed] [Google Scholar]
  • 60. Zhang J., Cao X., Yin G., et al., “The Significance of Better Utilization of Patients' Preoperative Information in Predicting Outcomes of Velopharyngeal Surgery: A Prospective Cohort Study,” Journal of Clinical Sleep Medicine 18, no. 3 (2022): 843–850, 10.5664/jcsm.9734. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 61. Zhang J., Li Y., Cao X., et al., “The Combination of Anatomy and Physiology in Predicting the Outcomes of Velopharyngeal Surgery,” Laryngoscope 124, no. 7 (2014): 1718–1723, 10.1002/lary.24510. [DOI] [PubMed] [Google Scholar]
  • 62. Ibrahim B., Guillemaud A., Sigaux N., Pouzet C., Gormand F., and Bouletreau P., “The Sleep Board: Impact of a Multidisciplinary Assessment Model in CPAP Failing OSA Patients,” Sleep Medicine 64 (2019): S165, 10.1016/j.sleep.2019.11.455. [DOI] [Google Scholar]
  • 63. Liu Y., Lowe A. A., Zeng X., Fu M., and Fleetham J. A., “Cephalometric Comparisons Between Chinese and Caucasian Patients With Obstructive Sleep Apnea,” American Journal of Orthodontics and Dentofacial Orthopedics 117, no. 4 (2000): 479–485, 10.1016/s0889-5406(00)70169-7. [DOI] [PubMed] [Google Scholar]
  • 64. Sutherland K., Keenan B. T., Bittencourt L., et al., “A Global Comparison of Anatomic Risk Factors and Their Relationship to Obstructive Sleep Apnea Severity in Clinical Samples,” Journal of Clinical Sleep Medicine 15, no. 4 (2019): 629–639, 10.5664/jcsm.7730. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 65. Rosenstock C., Gillesberg I., Gätke M. R., Levin D., Kristensen M. S., and Rasmussen L. S., “Inter‐Observer Agreement of Tests Used for Prediction of Difficult Laryngoscopy/Tracheal Intubation,” Acta Anaesthesiologica Scandinavica 49, no. 8 (2005): 1057–1062, 10.1111/j.1399-6576.2005.00792.x. [DOI] [PubMed] [Google Scholar]
  • 66. Sundman J., Bring J., and Friberg D., “Poor Interexaminer Agreement on Friedman Tongue Position,” Acta Oto‐Laryngologica 137, no. 5 (2017): 554–556, 10.1080/00016489.2016.1255776. [DOI] [PubMed] [Google Scholar]
  • 67. Friedman M., Soans R., Gurpinar B., Lin H. C., and Joseph N. J., “Interexaminer Agreement of Friedman Tongue Positions for Staging of Obstructive Sleep Apnea/Hypopnea Syndrome,” Otolaryngology and Head and Neck Surgery 139, no. 3 (2008): 372–377, 10.1016/j.otohns.2008.06.017. [DOI] [PubMed] [Google Scholar]
  • 68. Ng S. K., Lee D. L. Y., Li A. M., Wing Y. K., and Tong M. C. F., “Reproducibility of Clinical Grading of Tonsillar Size,” Archives of Otolaryngology – Head & Neck Surgery 136, no. 2 (2010): 159–162, 10.1001/archoto.2009.170. [DOI] [PubMed] [Google Scholar]
  • 69. Tang J. L. and Liu J. L., “Misleading Funnel Plot for Detection of Bias in Meta‐Analysis,” Journal of Clinical Epidemiology 53, no. 5 (2000): 477–484, 10.1016/S0895-4356(99)00204-8. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Data S1. Supporting Information.

LIO2-10-e70154-s003.docx (16KB, docx)

Supporting Information Figure 1 Leave one out and influence plots for all physical exam findings examined (BMI, waist circumference, neck circumference, thyromental distance, retrognathia, macroglossia, Mallampati grade, Friedman grade, tonsil grade, pharyngeal grade, enlarged uvula, septal deviation, and enlarged turbinates).

LIO2-10-e70154-s002.docx (290.1KB, docx)

Supporting Information Figure 2 Funnel plots for all physical exam findings examined (BMI, waist circumference, neck circumference, thyromental distance, retrognathia, macroglossia, Mallampati grade, Friedman grade, tonsil grade, pharyngeal grade, enlarged uvula, septal deviation, and enlarged turbinates).

LIO2-10-e70154-s001.docx (148.4KB, docx)

Supporting Information Table 1 Examples of how each parameter was measured among the different studies, demonstrating that different techniques were often used.

LIO2-10-e70154-s004.docx (16.4KB, docx)

Articles from Laryngoscope Investigative Otolaryngology are provided here courtesy of Wiley

RESOURCES