Secondary Logo

Journal Logo

Original Research Articles: Original Clinical Research Report

Association of Obstructive Sleep Apnea With Difficult Intubation: Prospective Multicenter Observational Cohort Study

Seet, Edwin MMed*,†; Chung, Frances FRCPC; Wang, Chew Yin FRCA§; Tam, Stanley FRCPC; Kumar, Chandra M. FRCA; Ubeynarayana, Chalani U. MS; Yim, Carolyn C. MAnaes§; Chew, Eleanor F. F. MAnaes#; Lam, Carmen K. M. FHKCA**; Cheng, Benny C. P. FHKCA**; Chan, Matthew T. V. PhD††

Author Information
doi: 10.1213/ANE.0000000000005479
  • Free

Abstract

KEY POINTS

  • Question: Is the diagnosis or suspicion of obstructive sleep apnea associated with difficult intubation in the surgical population?
  • Findings: Moderate obstructive sleep apnea (odds ratio = 3.3); severe obstructive sleep apnea (odds ratio = 4.1); Snoring, Tiredness, Observed apnea, high blood Pressure, Body mass index, Age, Neck circumference, and Gender (STOP-Bang) score of 3–4 (odds ratio = 3.0); and STOP-Bang score of 5–8 (odds ratio = 4.4) were independently associated with an increased risk of difficult intubation.
  • Meaning: Anesthesiologists should be vigilant for difficult intubation when managing patients diagnosed with or suspected of obstructive sleep apnea.

The difficult airway remains as a major cause of perioperative morbidity and mortality1,2 and is defined as the clinical situation in which a conventionally trained anesthesiologist experiences difficulty with mask ventilation of the upper airway, difficulty with tracheal intubation, or both.3 Being able to anticipate and manage the difficult airway may reduce adverse outcomes.1,2 Unfortunately, contemporary difficult airway prediction is imperfect and unanticipated difficult intubation still occurs too frequently.4,5

Obstructive sleep apnea (OSA) is characterized by intermittent complete or partial upper airway obstruction and desaturation during sleep.6 The perioperative physician should be concerned about the possible association of OSA with the difficult airway as both conditions can contribute to increased perioperative complications, such as anoxic brain injury and subsequent litigation.2,7,8 Practice guidelines from the major international societies on the perioperative management of OSA have stated that patients with suspected or known OSA may have difficult airways and therefore should be managed as such.6,9 Sixty-six percent of patients with difficult intubation were found to have OSA.10 Features common to both relate to the restrictive skeletal changes around the upper airway with pharyngeal anatomical imbalance,11–14 exacerbated by sleep and anesthesia.

An earlier meta-analysis by Nagappa et al15 of small observational studies13,16,17 and larger database studies18–20 showed that patients with OSA had a higher risk of difficult airway. In the Postoperative Vascular Complications in Unrecognized Obstructive Sleep Apnea (POSA) study, we performed preoperative portable sleep studies on all patients undergoing surgery to quantify the apnea-hypopnea index (AHI), with data on airway predictors collected.21,22 The aim of this planned secondary study was to examine the independent association between varying thresholds of OSA severity with difficult intubation or difficult mask ventilation. We also explored the utility of the Snoring, Tiredness, Observed apnea, high blood Pressure, Body mass index, Age, Neck circumference, and Gender (STOP-Bang) score for difficult airway prediction. We hypothesized that patients with severe OSA and higher STOP-Bang scores have a greater risk for difficult intubation.

METHODS

Study Design, Population, and Airway Outcomes

The present study consists of secondary analyses of patients who had general anesthesia and tracheal intubation for major elective noncardiac surgery recruited from 8 hospitals in 5 countries. Details regarding the study methodology have been previously published.21 The POSA study was registered with the ClinicalTrials.gov (NCT01494181). Ethics approval and written informed consent were obtained for all participating sites for the POSA study, including approval for these secondary analyses (Joint Chinese University of Hong Kong Clinical Research Ethics Committee, reference number: CRE-2010.509).

Adult patients ≥45 years, with no prior diagnosis of OSA and 1 or more cardiac risk factor (history of coronary artery disease, history of stroke or transient ischemic attack, history of congestive heart failure, diabetes requiring insulin therapy or serum creatinine >175 μmol/L), scheduled for major elective noncardiac surgery requiring an anticipated hospital stay of 3 nights or more were included. We excluded patients with prior diagnosis of sleep-related breathing disorder who were scheduled for corrective surgery for OSA.

The presence and severity of OSA were determined with a type 3 portable sleep monitoring device (ApneaLink Plus; ResMed, San Diego, CA).23 Apnea and hypopnea events were scored according to the American Academy of Sleep Medicine criteria.24 Mild OSA was defined when the AHI was 5–14.9 events/h, moderate OSA with the AHI 15–30 events/h, and severe OSA when the AHI was >30 events/h. The patients, surgeons, anesthesiologists, and the research staff were blinded to the results of the preoperative portable sleep study.

All patients received routine airway management and anesthetic care at each study site according to local practice. Airway management, including mask ventilation and tracheal intubation, were performed by experienced airway managers with specialist anesthesiologists on-site. Macintosh laryngoscopes were used for the initial direct laryngoscopy and tracheal intubation attempts. Adjunct airway equipment was used thereafter at the discretion of the attending anesthesiologist. These included gum elastic bougie, stylet introducer, video laryngoscopes, video stylets, lighted-stylets, and flexible bronchoscopes. The number of attempts was recorded. Laryngeal view was scored according to Cormack and Lehane classification based on direct laryngoscopic views.25 We excluded patients from this study in situations where the intraoperative anesthesia management was performed exclusively with supraglottic airway devices or regional anesthesia.

We collected data on patient demographics and known bedside examination tests for difficult airway prediction,4 including age, gender, ethnicity, body mass index (BMI), waistline, STOP-Bang risk score, Mallampati classification, interincisor distance, thyromental distance, neck circumference, severity of OSA, AHI, lowest oxygen saturation, and American Society of Anesthesiologists (ASA) physical status. Using the STOP-Bang screening tool, we classified patients as low, intermediate, and high risk for OSA if the STOP-Bang were scored 0–2, 3–4, and 5–8, respectively.26

The primary outcome of difficult intubation and the secondary outcome of difficult mask ventilation were prospectively determined by the attending anesthesiologist according to stipulated criteria during the study. We defined difficult intubation when tracheal intubation required multiple attempts, in the presence or absence of tracheal pathology, and/or it was not possible to visualize any portion of the vocal cords after multiple attempts at conventional laryngoscopy.3,27 Notably, the traditional definition of the difficult intubation has been challenged; nonetheless, the afore-mentioned scenarios leave patients at risk of potential harm.27 Difficult mask ventilation was defined as when it was not possible for the anesthesiologist to provide adequate ventilation because of one or more of the following problems: inadequate mask seal, excessive gas leak, or excessive resistance to the ingress or egress of gas, where inadequate ventilation would be associated with signs of absent or inadequate chest movement/breath sounds/capnography/spirometric changes, cyanosis, gastric air entry, decreasing oxygen saturation, or hemodynamic changes accompanying hypoxia or hypercarbia.3

Statistical Analysis

Continuous variables were summarized by mean and standard deviations. Categorical or ordinal variables were reported by counts and percentages. The primary and secondary outcomes of interest were difficult intubation and difficult mask ventilation, respectively. Baseline characteristics were compared between patients with or without difficult intubation or difficult mask ventilation using Student t test, χ2 test, or Fisher exact test, as appropriate. Crude odds ratio (OR) for the outcomes was calculated for each variable.

The sample size justification of 1200 patients for the original POSA study was previously described.21 A published meta-analysis has shown that the expected difficult intubation rate in patients with OSA versus those without OSA were 13.7% and 2.5%, respectively, where the average rate was 8.1%.15 A sample of 869 patients, 30% with moderate or severe OSA and an overall event rate of 8.1% for difficult intubation provides 90% power to detect an OR of 2.3 for the association between moderate or severe OSA versus none or mild OSA. The association between OSA and the difficult airway was performed using 2 multivariable logistic regression models. In the first model, the dependent variable was difficult intubation. The independent variables were OSA severity thresholds determined by the portable sleep study or STOP-Bang scores, neck circumference, BMI, interincisor distance, Mallampati score, thyromental distance and the presence of crowded dentition. The second model was constructed to determine the association between difficult mask ventilation and the same independent variables. Multicollinearity was tested using the variance inflation factor. A P value of <.05 was treated as statistically significant. A complete case analysis was performed with respect to missing data, excluding 23 (2.5%) subjects from the sample. All analyses were done by R 3.6 A Language and Environment for Statistical Computing, R Foundation for Statistical Computing, Vienna, Austria.

RESULTS

The POSA study recruited 1364 patients from January 2012 to July 2017. We excluded 146 patients because the surgery did not proceed or sleep recordings were unsatisfactory. Of the remaining 1218 patients, 869 patients underwent general anesthesia with tracheal intubation and were included in the current analyses (Figure).

F1
Figure.:
Patient flow diagram showing the number of patients included and excluded from the multivariable analyses.

Patient demographics, STOP-Bang scores, sleep study results, airway morphological predictors, and difficult airway outcomes are summarized in Table 1 and Supplemental Digital Content 1, Table 1, https://links.lww.com/AA/D420. Sixty-four percent of the patients were men; the mean (±standard deviations) age and BMI were 66 ± 9 years and 26 ± 8 kg/m2. A total of 33%, 37%, 20%, and 10% of patients had no, mild, moderate, and severe OSA, respectively. A total of 19.8% of patients had a STOP-Bang score of 0–2 (low risk), 55.6% had a STOP-Bang score of 3–4 (intermediate risk), and 24.6% had a STOP-Bang score of 5–8 (high risk) (Table 1). The overall incidence of difficult intubation was 6.7% (58 of 869) and difficult mask ventilation 3.7% (32 of 869). The incidence of difficult intubation increased with the diagnosis and severity of OSA: no OSA (2.8%), mild OSA (6.2%), moderate OSA (10.1%), and severe OSA (14.6%), P value <.001. The incidence of difficult mask ventilation in patients with no OSA, mild OSA, moderate OSA, and severe OSA were 3.5%, 2.5%, 4.1%, and 7.9%, respectively (P = .115). Based on univariate analysis, STOP-Bang scores ≥3, AHI, OSA severity, and interincisor distance were associated with difficult intubation; unadjusted P values <.05. Increasing neck circumference and AHI were associated with difficult mask ventilation; unadjusted P values <.05 (Supplemental Digital Content 1 and 2, Tables 1 and 2, https://links.lww.com/AA/D420, https://links.lww.com/AA/D421).

Table 1. - Patient Demographics, Sleep Study Findings, Predictors of Difficult Airway, and Difficult Airway Outcomes
Variable Number Proportion (%) Mean (SD)
Age (y) 66.1 (9.0)
Gender
 Male/female 559/310 64.3/35.7
Ethnicity
 Chinese/Malay 532/153 61.2/17.6
 White/Indian/others 73/91/20 8.4/10.5/2.3
American Society of Anesthesiologists physical status
 I/II/III/IV 0/510/337/22 0/58.7/38.8/2.5
OSA status based on Sleep Study
 No OSA 287 33.0
 Diagnosed OSA 582 67.0
  Mild/moderate/severe 324/169/89 37.3/19.4/10.2
Body mass index (kg/m−2) 26.0 (8.4)
STOP-Bang Score
 0–2/3–4/5–8 172/483/214 19.8/55.6/24.6
Crowded dentition 29 3.3
Mallampati Score
 1/2/3/4 219/467/167/16 25.2/53.7/19.2/1.8
Interincisor distance (cm) 5.3 (1.0)
Thyromental distance (cm) 6.7 (1.5)
Neck circumference (cm) 38.6 (3.3)
Difficult airway outcomes
Difficult intubation 58/869 6.7
 No OSA 8/287 2.8
 Mild OSA 20/324 6.2
 Moderate OSA 17/169 10.1
 Severe OSA 13/89 14.6
Difficult mask ventilation 32/869 3.7
 No OSA 10/287 3.5
 Mild OSA 8/324 2.5
 Moderate OSA 7/169 4.1
 Severe OSA 7/89 7.9
Cormack and Lehane Grading
 1/2/3/4 517/291/59/2 59.5/33.5/6.8/0.2
Use of airway adjunct
 With adjunct/no adjunct 290/579 33.4/66.6
Abbreviations: OSA, obstructive sleep apnea; SD, standard deviation; STOP-Bang, Snoring, Tiredness, Observed apnea, high blood Pressure, Body mass index, Age, Neck circumference, and Gender.

We assessed the independent association of 8 variables with difficult intubation (primary outcome) and difficult mask ventilation (secondary outcome) in 2 multivariable logistic regression models. Moderate OSA (OR = 3.26 [95% confidence interval {CI}, 1.37-8.38], adjusted P = .010) and severe OSA (OR = 4.05 [95% CI, 1.51-11.36], adjusted P = .006) were associated with higher odds of difficult intubation compared to patients without OSA; but not mild OSA (OR = 2.17 [95% CI, 0.96-5.36], adjusted P = .075; Table 2). Increasing neck circumference was found to be associated with difficult mask ventilation (OR = 1.25 [95% CI, 1.09-1.46], adjusted P = .002; Table 3).

Table 2. - Multivariable Logistic Regression Model for Difficult Intubation
Variable Difficult intubation No difficult intubation OR (95% CI) Adjusted P value
BMI, kg/m−2 29.9 (26.7) 25.7 (5.0) 1.02 (1.00-1.06) .118
Interincisor distance, cm 5.1 (0.8) 5.4 (1.0) 0.81 (0.60-1.08) .142
Crowded dentition, N [%]
 Yes 4 [0.5] 25 [2.9] 2.3 (0.63-6.71) .159
 No 54 [6.2] 786 [90.5] Reference
Severity of OSA, N [%]
 No OSA 8 [0.9] 279 [32.1] Reference
 Mild OSA 20 [2.3] 304 [35.0] 2.17 (0.96-5.36) .075
 Moderate OSA 17 [2.0] 152 [17.5] 3.26 (1.37-8.38) .010
 Severe OSA 13 [1.5] 76 [8.7] 4.05 (1.51-11.36) .006
STOP-Bang, N [%]
 0–2 4 [0.5] 168 [19.3] Reference
 3–4 32 [3.7] 451 [51.9] 2.72 (1.02-9.46) .072
 5–8 22 [2.5] 192 [22.1] 3.28 (1.07-12.42) .053
Mallampati Score, N [%]
 1 12 [1.4] 207 [23.8] Reference
 2 27 [3.1] 440 [50.6] 0.9 (0.45-1.92) .783
 3 17 [2.0] 150 [17.3] 1.36 (0.28-7.25) .458
 4 2 [0.2] 14 [1.6] 1.62 (0.28-1.17) .568
Thyromental distance, cm 6.7 (1.8) 6.7 (1.4) 0.97 (0.8-1.17) .761
Neck circumference, cm 39.4 (3.3) 38.5 (3.3) 0.96 (0.88-1.06) .441
Values are presented as mean (standard deviation), N [%].
Abbreviations: BMI, body mass index; CI, confidence interval; OR, odds ratio; OSA, obstructive sleep apnea; STOP-Bang, Snoring, Tiredness, Observed apnea, high blood Pressure, Body mass index, Age, Neck circumference, and Gender.

Table 3. - Multivariable Logistic Regression Model for Difficult Mask Ventilation
Variable Difficult mask ventilation No difficult mask ventilation OR (95% CI) Adjusted P value
BMI, kg/m−2 26.3 (5.2) 26.0 (8.5) 0.98 (0.89-1.02) .504
Interincisor distance, cm 5.6 (0.9) 5.3 (1.0) 1.40 (0.93-2.15) .116
Crowded dentition, N [%]
 Yes 2 [0.2] 27 [3.1] 2.09 (0.32-7.96) .346
 No 30 [3.5] 810 [93.2] Reference
Severity of OSA, N [%]
 No OSA 10 [1.2] 277 [31.9] Reference
 Mild OSA 8 [0.9] 316 [36.4] 0.61 (0.23-1.60) .313
 Moderate OSA 7 [0.8] 162 [18.6] 0.99 (0.33-2.82) .989
 Severe OSA 7 [0.8] 82 [9.4] 1.84 (0.57-5.68) .292
STOP-Bang, N [%]
 0–2 4 [0.5] 168 [19.3] Reference
 3–4 19 [2.2] 464 [53.4] 1.12 (0.39-4.05) .851
 5–8 9 [1.0] 205 [23.6] 0.58 (0.14-2.61) .458
Mallampati Score, N [%]
 1 8 [0.9] 211 [24.3] Reference
 2 19 [2.2] 448 [51.6] 1.03 (0.45-2.61) .939
 3 3 [0.3] 164 [18.9] 0.33 (0.07-1.24) .122
 4 2 [0.2] 14 [1.6] 2.34 (0.31-11.95) .344
Thyromental distance, cm 6.5 (1.2) 6.7 (1.5) 0.89 (0.66-1.16) .413
Neck circumference, cm 40.4 (3.2) 38.5 (3.3) 1.25 (1.09-1.46) .002
Values are presented as mean (standard deviation), N [%].
Abbreviations: BMI, body mass index; CI, confidence interval; OR, odds ratio; OSA, obstructive sleep apnea; STOP-Bang, Snoring, Tiredness, Observed apnea, high blood Pressure, Body mass index, Age, Neck circumference, and Gender.

Table 4. - Multivariable Logistic Regression Model Exploring the Utility of STOP-Bang in Predicting Difficult Intubation
Variable Difficult intubation No difficult intubation OR (95% CI) Adjusted P value
BMI, kg/m−2 29.9 (26.7) 25.7 (5.0) 1.02 (1.00-1.06) .083
Interincisor distance, cm 5.1 (0.8) 5.4 (1.0) 0.77 (0.58-1.03) .071
Crowded dentition, N [%]
 Yes 4 [0.5] 25 [2.9] 2.16 (0.59-1.03) .188
 No 54 [6.2] 786 [90.4] Reference
STOP-Bang, N [%]
 0–2 4 [0.5] 168 [19.3] Reference
 3–4 32 [6.7] 451 [51.9] 3.01 (1.13-10.40) .046
 5–8 22 [2.5] 192 [22.1] 4.38 (1.46-16.36) .014
Mallampati Score, N [%]
 1 12 [1.4] 207 [23.8] Reference
 2 27 [3.1] 440 [50.6] 0.93 (0.47-1.96) .840
 3 17 [2.0] 150 [17.3] 1.42 (0.64-3.23) .395
 4 2 [0.2] 14 [1.6] 1.76 (0.25-7.80) .499
Thyromental distance, cm 6.7 (1.8) 6.7 (1.4) 0.97 (0.80-1.17) .734
Neck circumference, cm 39.4 (3.3) 38.5 (3.3) 0.99 (0.90-1.08) .815
Values are presented as mean (standard deviation), N [%].
Abbreviations: BMI, body mass index; CI, confidence interval; OR, odds ratio; STOP-Bang, Snoring, Tiredness, Observed apnea, high blood Pressure, Body mass index, Age, Neck circumference, and Gender.

To explore the utility of the STOP-Bang score and its association with difficult intubation and difficult mask ventilation, we created additional multivariable logistic regression models by excluding OSA severity in the analysis, noting the low to moderate correlation between OSA severity and the STOP-bang score (Table 4, Supplemental Digital Content 3, Table 3, https://links.lww.com/AA/D422). STOP-Bang score of 3–4 was significantly associated with an increased odds of 3.01 (95% CI, 1.13-10.40, adjusted P = .046) and STOP-Bang score of 5–8 an increased odds of 4.38 (95% CI, 1.46-16.36, adjusted P = .014) for difficult intubation versus a STOP-Bang score of 0–2 (Table 4).

DISCUSSION

In 869 patients, we found that moderate OSA, severe OSA, STOP-Bang score of 3–4, and STOP-Bang score of 5–8 were associated with difficult intubation with an increased OR of 3.3-, 4.1-, 3.0-, and 4.4-fold, respectively.

Postoperative morbidity such as failed reintubations, anoxic brain injury, and even mortality has been shown to occur in OSA patients.7,28 Anesthesiologists are concerned about possible associations of OSA with difficult intubation and difficult mask ventilation contributing to the potential for harm.6,8,9,27 OSA and the difficult airway share morphological commonalities, including pharyngeal anatomical imbalance, with increased oropharyngeal soft tissue,29,30 and/or restrictive cervico-maxillo-mandibular enclosure anatomy.11–14 Supine positioning and abolition of neural compensation to maintain pharyngeal airway opening with the administration of anesthetics exacerbate upper airway collapse.31 In the Asian subgroup, the restrictive craniofacial phenotypes (ie, smaller upper airway bony dimensions) contribute to OSA causality, whereas for the White population, greater oropharyngeal soft tissue may play a greater role.32–34

Small case-control studies supported the association of OSA with difficult intubation.13,16,17 OSA was also found to be associated with difficult laryngoscopy and/or difficult mask ventilation in database studies.18–20 In these, the outcomes of interest were global predictors of difficult airway, not specifically relating to OSA diagnosis and severity.18–20 This may have led to the under-diagnosis of OSA in the non-OSA group; that is, patients in the control group may have undiagnosed or unrecognized OSA. In a meta-analysis of 16 studies comparing difficult airway outcomes between OSA versus non-OSA, difficult intubation was found to be 3.5-fold higher (OR = 3.46 [95% CI, 2.32-5.16], P < .00001) and difficult mask ventilation 3.4-fold higher (OR = 3.39 [95% CI, 2.74-4.18], P < .00001) in the patients with OSA.15 This meta-analysis included individual studies with variations in methodology (both retrospective and prospective studies) and different difficult airway definitions, resulting in potential heterogeneity of the aggregated quantitative data. Diagnoses of OSA for 5 of 16 of the included studies were based on clinical characteristics and not sleep studies; this leads to possible false-positive and false-negative OSA results. Included studies did not quantify AHI and could not differentiate OSA severity. Furthermore, these included studies contained limited data on baseline confounders such as age, gender, BMI, and neck circumference.15

Compared with the meta-analysis, we found a similar magnitude of increased difficult intubation risk in patients with moderate OSA (OR = 3.26 [95% CI, 1.37-8.38], P = .010) and severe OSA (OR = 4.05 [95% CI, 1.51-11.36], P = .006) versus patients without OSA. From the matched case-control study by Kim and Lee16 involving 90 patients with OSA, a threshold AHI of ≥40 events/h was associated with difficult intubation. In contrast, we found that an AHI cutoff of 15 events/h (moderate OSA severity) was significantly associated with difficult intubation but not with mild OSA (OR = 2.17 [95% CI, 0.96-5.36], P = .075). In addition, we report a novel “dose-effect relationship” for the difficult intubation outcome. The incidence of difficult intubation in patients without OSA in our prospective cohort was 2.8% compared with 6.2% in mild OSA, 10.1% in moderate OSA, and 14.6% (1-in-7) in severe OSA patients (P < .001). Correspondingly, the OR for difficult intubation for mild OSA, moderate OSA, and severe OSA versus no OSA were 2.17, 3.26, and 4.05, respectively.

From multivariable analyses of database studies involving the American population (White majority), Kheterpal et al18,20 found an association between snoring or OSA with difficult mask ventilation. Correspondingly, the quantitative meta-analysis by Nagappa et al15 found that difficult mask ventilation was 3.4-fold higher in patients with OSA. Of note, the included studies in the meta-analysis based OSA diagnosis on STOP-Bang scores or electronic database records, without quantification of the AHI.15 In contrast, our study demonstrated that increasing neck circumference was independently associated with difficult mask ventilation but not OSA. A possible reason for our findings is that >60% of our cohort were from the Chinese (Asian) ethnicity, where the mechanistic contribution of OSA would more likely be due to the craniofacial phenotype as opposed to obesity and soft tissue excess.32–34 We postulate that increased neck circumference—being an indicator of pharyngeal soft tissue excess—is associated with difficult mask ventilation rather than OSA per se. Similarly, studies involving surgical cohorts have identified neck circumferences as a predictor of difficult mask ventilation.35,36

Strength of the Study

This is a large prospective international study investigating OSA severity thresholds and its association with difficult intubation. The diagnosis and exclusion of OSA were based on portable sleep study findings accompanied by standardized difficult airway definitions. As the entire cohort underwent a preoperative sleep study, OSA was excluded with certainty in the control group. Known predictors of difficult airway were included in the multivariable analyses; however, there is the potential of unmeasured confounding. The clinicians and research staff were masked to the sleep study results.

Furthermore, we collected STOP-Bang scores and explored its utility and its association with difficult intubation. In the absence of a preoperative sleep study, a STOP-Bang score 3–4 (intermediate risk) and 5–8 (high risk) were associated with 3- and 4.4-fold increased odds of difficult intubation, respectively. Smaller studies have similarly postulated an association between higher STOP-Bang scores with difficult intubation.37–39 The STOP-Bang may be considered as a composite airway risk score, with a combination of several physical characteristics, and may be a better predictor of difficult airway compared to a single airway predictor in isolation.40

Limitations

Some limitations exist for this prospective cohort study. First, the primary objective of the POSA study was to investigate cardiovascular outcomes associated with unrecognized OSA. Difficult airway outcomes were based on secondary analyses, and the original POSA sample size was not intentionally powered for these difficult airway measures. Second, the study design adopted a pragmatic approach and did not control for intubationist and airway management techniques, although the Macintosh direct laryngoscope was used for primary intubation attempts. Third, the POSA study included patients ≥45 years with at least 1 cardiac risk factor so the difficult airway–related findings are not directly translatable to the general surgical population. Fourth, we note that the STOP-Bang score of 3 or more consisted of 80.2% of our cohort, suggesting that the lack of specificity may limit the predictive capacity of STOP-Bang for difficult intubation. Finally, patients came from 4 major ethnicities: Chinese (61%), Malays (18%), Indian (11%), and White (8%). Due to morphological differences in the upper airway between the races, the results may not be applicable to a single ethnic population but generalizable to this study demographics.

CONCLUSIONS

We found that the incidence of difficult intubation was 6.7%. Moderate OSA and severe OSA patients have a 3- to 4-fold increased odds of difficult intubation, where 1-in-7 (14.5%) of patients with severe OSA had difficult intubation. In the absence of a preoperative sleep study, a higher STOP-Bang score of 3 or more may be associated with difficult intubation versus STOP-Bang score of 0–2. Anesthesiologists should have a high index of suspicion for difficult intubation when managing patients known or suspected of OSA.

ACKNOWLEDGMENTS

The authors thank the following for their contributions in patient recruitment, data collection, and analysis in this study. Singapore: Khoo Teck Puat Hospital: Clarabella Liew, MBBS, MMed, Ms Audris Chia. BSc. Malaysia: University of Malaya Medical Centre: Hou Yee Lai, MBBS, Alvin S. B. Tan, MBBS, Ching Yen Chong, BA, Jason H. Kueh, BSc, Xue Lin Chan, MBChB. Hospital Kuala Lumpur: Su Yin Loo, MD, Simon M. T. Hui, MBBS.

DISCLOSURES

Name: Edwin Seet, MMed.

Contribution: This author helped with the concept, design, literature search, patient recruitment and conduct of the study, acquisition of data, and analysis and drafting of the article.

Conflicts of Interest: None.

Name: Frances Chung, FRCPC.

Contribution: This author helped with the concept, design, literature search, patient recruitment and conduct of the study, acquisition of data, and analysis and drafting of the article.

Conflicts of Interest: F. Chung reported receiving grants from the Ontario Ministry of Health and Long-term Care, Acacia Pharma, Medtronics, and she is holding a patent pending for the STOP-Bang screening tool.

Name: Chew Yin Wang, FRCA.

Contribution: This author helped with the concept, design, literature search, patient recruitment and conduct of the study, acquisition of data, and analysis and drafting of the article.

Conflicts of Interest: None.

Name: Stanley Tam, FRCPC.

Contribution: This author helped with the patient recruitment, acquisition of data, and analysis and drafting of the article.

Conflicts of Interest: None.

Name: Chandra M. Kumar, FRCA.

Contribution: This author helped with the conception and design of the study, data analysis, and drafting of the article.

Conflicts of Interest: None.

Name: Chalani U. Ubeynarayana, MS.

Contribution: This author helped with the conception and design of the study, data analysis, and drafting of the article.

Conflicts of Interest: None.

Name: Carolyn C. Yim, MAnaes.

Contribution: This author helped with the patient recruitment, acquisition of data, and analysis and drafting of the article.

Conflicts of Interest: None.

Name: Eleanor F. F. Chew, MAnaes.

Contribution: This author helped with the patient recruitment, acquisition of data, and analysis and drafting of the article.

Conflicts of Interest: None.

Name: Carmen K. M. Lam, FHKCA.

Contribution: This author helped with the patient recruitment, acquisition of data, and analysis and drafting of the article.

Conflicts of Interest: None.

Name: Benny C. P. Cheng, FHKCA.

Contribution: This author helped with patient recruitment, acquisition of data, and analysis and drafting of the article.

Conflicts of Interest: None.

Name: Matthew T. V. Chan, PhD.

Contribution: This author helped with the concept, design, literature search, patient recruitment and conduct of the study, acquisition of data, and analysis and drafting of the article.

Conflicts of Interest: None.

This manuscript was handled by: Toby N. Weingarten, MD.

    REFERENCES

    1. Cook TM, MacDougall-Davis SR. Complications and failure of airway management. Br J Anaesth. 2012; 109(suppl 1):i68–i85.
    2. Joffe AM, Aziz MF, Posner KL, Duggan LV, Mincer SL, Domino KB. Management of difficult tracheal intubation: a closed claims analysis. Anesthesiology. 2019; 131:818–829.
    3. Apfelbaum JL, Hagberg CA, Caplan RA, et al.; American Society of Anesthesiologists Task Force on Management of the Difficult Airway. Practice guidelines for management of the difficult airway: an updated report by the American Society of Anesthesiologists Task Force on Management of the Difficult Airway. Anesthesiology. 2013; 118:251–270.
    4. Roth D, Pace NL, Lee A, et al. Bedside tests for predicting difficult airways: an abridged Cochrane diagnostic test accuracy systematic review. Anaesthesia. 2019; 74:915–928.
    5. Nørskov AK, Rosenstock CV, Wetterslev J, Astrup G, Afshari A, Lundstrøm LH. Diagnostic accuracy of anaesthesiologists’ prediction of difficult airway management in daily clinical practice: a cohort study of 188 064 patients registered in the Danish Anaesthesia Database. Anaesthesia. 2015; 70:272–281.
    6. American Society of Anesthesiologists Task Force on Perioperative Management of patients with obstructive sleep apnea. Practice guidelines for the perioperative management of patients with obstructive sleep apnea: an updated report by the American Society of Anesthesiologists Task Force on Perioperative Management of patients with obstructive sleep apnea. Anesthesiology. 2014; 120:268–286.
    7. Fouladpour N, Jesudoss R, Bolden N, Shaman Z, Auckley D. Perioperative complications in obstructive sleep apnea patients undergoing surgery: a review of the legal literature. Anesth Analg. 2016; 122:145–151.
    8. Opperer M, Cozowicz C, Bugada D, et al. Does obstructive sleep apnea influence perioperative outcome? A qualitative systematic review for the Society of Anesthesia and Sleep Medicine Task Force on preoperative preparation of patients with sleep-disordered breathing. Anesth Analg. 2016; 122:1321–1334.
    9. Memtsoudis SG, Cozowicz C, Nagappa M, et al. Society of anesthesia and sleep medicine guideline on intraoperative management of adult patients with obstructive sleep apnea. Anesth Analg. 2018; 127:967–987.
    10. Chung F, Yegneswaran B, Herrera F, Shenderey A, Shapiro CM. Patients with difficult intubation may need referral to sleep clinics. Anesth Analg. 2008; 107:915–920.
    11. Chou HC, Wu TL. Large hypopharyngeal tongue: a shared anatomic abnormality for difficult mask ventilation, difficult intubation, and obstructive sleep apnea? Anesthesiology. 2001; 94:936–937.
    12. Solow B, Skov S, Ovesen J, Norup PW, Wildschiødtz G. Airway dimensions and head posture in obstructive sleep apnoea. Eur J Orthod. 1996; 18:571–579.
    13. Hiremath AS, Hillman DR, James AL, Noffsinger WJ, Platt PR, Singer SL. Relationship between difficult tracheal intubation and obstructive sleep apnoea. Br J Anaesth. 1998; 80:606–611.
    14. Haponik EF, Smith PL, Bohlman ME, Allen RP, Goldman SM, Bleecker ER. Computerized tomography in obstructive sleep apnea. Correlation of airway size with physiology during sleep and wakefulness. Am Rev Respir Dis. 1983; 127:221–226.
    15. Nagappa M, Wong DT, Cozowicz C, Ramachandran SK, Memtsoudis SG, Chung F. Is obstructive sleep apnea associated with difficult airway? Evidence from a systematic review and meta-analysis of prospective and retrospective cohort studies. PLoS One. 2018; 13:e0204904.
    16. Kim JA, Lee JJ. Preoperative predictors of difficult intubation in patients with obstructive sleep apnea syndrome. Can J Anaesth. 2006; 53:393–397.
    17. Siyam MA, Benhamou D. Difficult endotracheal intubation in patients with sleep apnea syndrome. Anesth Analg. 2002; 95:1098–1102.
    18. Kheterpal S, Han R, Tremper KK, et al. Incidence and predictors of difficult and impossible mask ventilation. Anesthesiology. 2006; 105:885–891.
    19. Kheterpal S, Martin L, Shanks AM, Tremper KK. Prediction and outcomes of impossible mask ventilation: a review of 50,000 anesthetics. Anesthesiology. 2009; 110:891–897.
    20. Kheterpal S, Healy D, Aziz MF, et al.; Multicenter Perioperative Outcomes Group (MPOG) Perioperative Clinical Research Committee. Incidence, predictors, and outcome of difficult mask ventilation combined with difficult laryngoscopy: a report from the multicenter perioperative outcomes group. Anesthesiology. 2013; 119:1360–1369.
    21. Chan MT, Wang CY, Seet E, et al.; POSA Investigators. Postoperative vascular complications in unrecognised Obstructive Sleep apnoea (POSA) study protocol: an observational cohort study in moderate-to-high risk patients undergoing non-cardiac surgery. BMJ Open. 2014; 4:e004097.
    22. Chan MTV, Wang CY, Seet E, et al.; Postoperative Vascular Complications in Unrecognized Obstructive Sleep Apnea (POSA) Study Investigators. Association of unrecognized obstructive sleep apnea with postoperative cardiovascular events in patients undergoing major noncardiac surgery. JAMA. 2019; 321:1788–1798.
    23. Ng SS, Chan TO, To KW, et al. Validation of a portable recording device (ApneaLink) for identifying patients with suspected obstructive sleep apnoea syndrome. Intern Med J. 2009; 39:757–762.
    24. Collop NA, Tracy SL, Kapur V, et al. Obstructive sleep apnea devices for out-of-center (OOC) testing: technology evaluation. J Clin Sleep Med. 2011; 7:531–548.
    25. Cormack RS, Lehane J. Difficult tracheal intubation in obstetrics. Anaesthesia. 1984; 39:1105–1111.
    26. Chung F, Yegneswaran B, Liao P, et al. STOP questionnaire: a tool to screen patients for obstructive sleep apnea. Anesthesiology. 2008; 108:812–821.
    27. Bradley JA, Urman RD, Yao D. Challenging the traditional definition of a difficult intubation: what is difficult? Anesth Analg. 2019; 128:584–586.
    28. Subramani Y, Nagappa M, Wong J, Patra J, Chung F. Death or near-death in patients with obstructive sleep apnoea: a compendium of case reports of critical complications. Br J Anaesth. 2017; 119:885–899.
    29. Isono S. Obstructive sleep apnea of obese adults: pathophysiology and perioperative airway management. Anesthesiology. 2009; 110:908–921.
    30. Singh M, Tuteja A, Wong DT, et al. Point-of-care ultrasound for obstructive sleep apnea screening: are we there yet? A systematic review and meta-analysis. Anesth Analg. 2019; 129:1673–1691.
    31. Subramani Y, Singh M, Wong J, Kushida CA, Malhotra A, Chung F. Understanding phenotypes of obstructive sleep apnea: applications in anesthesia, surgery, and perioperative medicine. Anesth Analg. 2017; 124:179–191.
    32. Lee RW, Vasudavan S, Hui DS, et al. Differences in craniofacial structures and obesity in Caucasian and Chinese patients with obstructive sleep apnea. Sleep. 2010; 33:1075–1080.
    33. Schorr F, Kayamori F, Hirata RP, et al. Different craniofacial characteristics predict upper airway collapsibility in Japanese-Brazilian and White men. Chest. 2016; 149:737–746.
    34. Xu L, Keenan BT, Wiemken AS, et al. Differences in three-dimensional upper airway anatomy between Asian and European patients with obstructive sleep apnea. Sleep. 2020:43:zsz273.
    35. Cattano D, Killoran PV, Cai C, Katsiampoura AD, Corso RM, Hagberg CA. Difficult mask ventilation in general surgical population: observation of risk factors and predictors. F1000Res. 2014; 3:204.
    36. Riad W, Vaez MN, Raveendran R, et al. Neck circumference as a predictor of difficult intubation and difficult mask ventilation in morbidly obese patients: a prospective observational study. Eur J Anaesthesiol. 2016; 33:244–249.
    37. Acar HV, Yarkan Uysal H, Kaya A, Ceyhan A, Dikmen B. Does the STOP-Bang, an obstructive sleep apnea screening tool, predict difficult intubation? Eur Rev Med Pharmacol Sci. 2014; 18:1869–1874.
    38. Toshniwal G, McKelvey GM, Wang H. STOP-Bang and prediction of difficult airway in obese patients. J Clin Anesth. 2014; 26:360–367.
    39. Corso RM, Petrini F, Buccioli M, et al. Clinical utility of preoperative screening with STOP-Bang questionnaire in elective surgery. Minerva Anestesiol. 2014; 80:877–884.
    40. Crawley SM, Dalton AJ. Predicting the difficult airway. Contin Educ Anaesth Crit Care Pain. 2015; 15:253–7.

    Supplemental Digital Content

    Copyright © 2021 International Anesthesia Research Society