Secondary Logo

Journal Logo

Featured Articles: Original Clinical Research Report

Preoperative Screening for Obstructive Sleep Apnea Using Alternative Scoring Models of the Sleep Tiredness Observed Pressure-Body Mass Index Age Neck Circumference Gender Questionnaire: An External Validation

Seguin, Ludovic MD*; Tamisier, Renaud MD, PhD; Deletombe, Baptiste MD*; Lopez, Mélanie BSc; Pepin, Jean-Louis MD, PhD; Payen, Jean-François MD, PhD*,‡

Author Information
doi: 10.1213/ANE.0000000000004909

Abstract

See Articles, p 1007, p 1012, p 1032

KEY POINTS

  • Question: Can the existing alternative scoring models of the Sleep Tiredness Observed Pressure-Body mass index Age Neck circumference Gender (STOP-Bang) questionnaire be externally validated for the preoperative screening of patients at risk for obstructive sleep apnea syndrome?
  • Findings: Stepwise stratification of the STOP-Bang questionnaire and preoperative measurement of serum bicarbonate concentrations of patients with STOP-Bang scores of 3–4 both failed to improve the preoperative screening of these patients.
  • Meaning: Although a STOP-Bang score of 5–8 can identify patients with moderate-to-severe obstructive sleep apnea, further study is needed to improve screening accuracy for patients with STOP-Bang scores of 3–4.

Obstructive sleep apnea (OSA) is the most common form of sleep-disordered breathing; it affects nearly 1 billion people worldwide, and 30 million remain undiagnosed in Europe.1,2 OSA is defined by a combination of diurnal and nocturnal symptoms and an apnea–hypopnea index (AHI) of more than 5 events per hour.3 Severe OSA is an independent factor for more perioperative complications.4–7 However, a large proportion of individuals with OSA remain undiagnosed before surgery,8 and identifying OSA preoperatively could reduce the incidence of OSA-related perioperative complications.9 In this context, large efforts have been made for years to detect individuals with OSA using preoperative clinical screening tools before confirming the diagnosis with polysomnography or home sleep apnea testing.

Among these screening tools, the Sleep Tiredness Observed Pressure-Body mass index Age Neck circumference Gender (STOP-Bang) questionnaire has become the most popular instrument for surgical and nonsurgical patients. The STOP-Bang questionnaire is an OSA screening tool consisting of 4 yes/no questions (STOP) and 4 clinical attributes (Bang).10 Patients are classified according to their risk of having OSA, from 0–2 positive items (low risk) to 3–8 positive items (intermediate-to-high risk). Although the sensitivity of STOP-Bang scores ≥3 predict OSA is higher than 90%, its specificity is lower than 50% due to high false-positive rates,3,11 which means unnecessary referrals of patients for sleep studies. This is particularly true for patients with STOP-Bang scores of 3–4 (intermediate risk), who represent the largest group of patients within the 0–8 range of STOP-Bang scores.12,13 This also indicates that items on the questionnaire do not have an equal predictive weight for OSA. To improve its specificity, stepwise stratification of the STOP-Bang questionnaire has been proposed, which uses a combination of 2 positive items from the STOP questions (OSA symptoms) plus sex, body mass index (BMI), or neck circumference from the Bang items.14 Adding assessment of serum bicarbonate concentrations to STOP-Bang has also been proposed to improve the specificity of the STOP-Bang questionnaire.12 However, these alternative scoring models are yet to receive external validation, especially for patients with STOP-Bang scores of 3–4. Our aim was to conduct an external validation of the standard STOP-Bang questionnaire and existing alternative scoring models in a cohort of preoperative patients who all had STOP-Bang scores ≥3.

METHODS

This prospective cohort study was conducted between June 2015 and January 2018 in the Grenoble Alpes University Hospital as part of a study on the evaluation of postoperative cardiac rhythm abnormalities in patients with OSA (ClinicalTrials.gov, number NCT02833662, registered on July 14, 2016, principal investigator Dr Tamisier). The present study was approved on April 8, 2015 by the Institutional Review Board of Sud-Est V (Chairperson Mr J. Grunwald, Grenoble Alpes University Hospital, 38000, Grenoble, France) (Ref. 15-Centre Hospitalier Universitaire de Grenoble [CHUG]-16). Written informed consent was obtained from all patients before enrollment. Electronic database access was granted by the French data protection authority (Commission Nationale Informatique et Libertés, CNIL) (Re#1744009v0). This manuscript adheres to the applicable Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines.

Participants

Patients scheduled for urologic, abdominal, or orthopedic surgery under general anesthesia were asked to complete the STOP-Bang questionnaire during their preoperative evaluation. Information regarding BMI, age, neck circumference, and sex (Bang items), as well as the patient’s history and clinical examination, were collected by the in-charge anesthesiologist. Inclusion criteria were patients aged ≥40 years with a STOP-Bang score ≥3. Patients were then invited to undertake a portable respiratory polygraph during the night following the consultation and have their serum bicarbonate concentration measured preoperatively. Noninclusion criteria were patients with known OSA with ongoing treatment or other sleep-disordered breathing before surgery, preoperative severe respiratory or neurological deficiencies, preoperative treatment with noninvasive ventilation, bariatric surgery, a STOP-Bang score of 0–2, or a planned postoperative stay in the intensive care unit and patients who were legally vulnerable or unable to undergo nocturnal recordings using a polygraph.

Data Sources/Measurements

According to the standard STOP-Bang scoring system, patients were categorized into 2 subgroups: patients with an intermediate (STOP-Bang 3–4) or a high risk of OSA (STOP-Bang 5–8). The primary end point was the performance of a STOP-Bang score of 5–8 to predict moderate-to-severe OSA (AHI >15). In patients with a STOP-Bang score of 3–4, we applied 2 alternative scoring models identified in previous studies: stepwise stratification of the STOP-Bang questionnaire and preoperative measurement of serum bicarbonate concentrations.12,14 Patients with 2 or more positive items from the STOP questions and 1 positive question among 3 Bang items (ie, BMI >35, neck circumference >40 cm, or male sex) were classified as having a higher risk of OSA (alternative model 1). Similarly, patients with serum bicarbonate concentrations ≥28 mmol/L were classified as being at a higher risk for OSA (alternative model 2). The secondary end point was the performance of each scoring model individually and in combination to predict moderate-to-severe OSA in patients with a STOP-Bang score of 3–4, that is, the percentage of patients correctly classified.

The portable respiratory polygraph (Vista-O2, Novacor, Rueil Malmaison, France) was used at the patient’s home to record blood oxygen saturation (Spo2), electrocardiogram Holter, movements of the chest and abdomen, and breathing events via a nasal cannula overnight. This device is a reliable alternative to standard polysomnography.15 The device was set up by a trained technician, and the polygraph recordings were scored by a sleep physician who was blinded to the results of the STOP-Bang questionnaire. In line with the 2012 recommendations by the American Academy of Sleep Medicine (AASM),16 apnea was defined as a ≥90% drop in air flow from baseline lasting ≥10 seconds, and hypopnea was defined as a ≥30% reduction of air flow lasting ≥10 seconds associated with a ≥3% decrease in arterial oxygen saturation. Apneas were classified as obstructive if respiratory efforts were present and central if not. The diagnosis of OSA was based on the calculation of AHI: no OSA if AHI ≤5, mild OSA if AHI >5, moderate OSA if AHI >15, and severe OSA if AHI >30.

Statistical Analysis

Based on previous studies,12,13 a convenience sample of more than 100 moderate-to-severe OSA patients with STOP-Bang scores of 3–8 was decided on. Continuous variables were expressed as median and interquartile range (25th–75th percentile), and categorical variables were expressed as frequencies and percentages. The relationship between STOP-Bang scores and AHI was tested using the Spearman’s rank coefficient correlation. The prevalence between STOP-Bang scores groups (3–4 vs 5–8) was compared using the χ2 test. A multivariable logistic model adjusted for age, sex, BMI, history of hypertension, diabetes, and major adverse cardiac and cerebrovascular events was used to assess the relation between STOP-Bang scores and AHI >15. To assess the performance of high score of STOP-Bang (5–8), sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were computed with their 95% confidence interval (CI) based on a normal approximation. A χ2 test was used to compare proportions of positive and negative values of the 2 alternative scoring models for different AHI classes. Sensitivity, specificity, PPV, and NPV values; their 95% CI based on a normal approximation; and positive and negative likelihood ratios (likelihood ratio [LR]+ and LR−, respectively) were calculated using 2 × 2 contingency tables for the 2 alternating scoring models. Statistical significance was declared when P ≤ .05 (SAS 9.4, SAS Institute, Cary, NC).

RESULTS

A total of 315 patients completed the STOP-Bang questionnaire during their preoperative evaluation. There were 195 nonincluded patients who had STOP-Bang scores of 0–2. Following polygraph analysis, 5 patients had nonreadable polygraph recordings due to technical issues. Therefore, the analysis included 115 patients with STOP-Bang scores of 3–8, complete polygraph data, and preoperative assessment of serum bicarbonate concentrations (Figure 1). Table 1 details their characteristics, while Table 2 shows AHI distribution and responses to the STOP-Bang questionnaire. The distribution of STOP-Bang questionnaire scores is shown in Figure 2. There was a positive correlation between STOP-Bang scores and AHI (ρ = 0.38; P < .01).

Table 1. - Characteristics of the 115 Enrolled Patients
Variables
Age, y 67 (60–73)
Male sex, n (%) 89 (77)
BMI 27 (24–31)
Neck circumference, cm 41 (38–43)
ASA physical status, n (%)
 I 11 (10)
 II 57 (49)
 III–IV 47 (41)
History, n (%)
 Smoker 63 (55)
 Hypertension 59 (51)
 Cancer 37 (32)
 Diabetes mellitus 28 (24)
 Dyslipidemia 19 (16)
 Gastric reflux 13 (11)
 Coronary artery disease 12 (10)
 Arrhythmias 12 (10)
Type of surgery, n (%)
 Urology 62 (54)
 Abdomen 52 (45)
 Orthopedic 1 (1)
Data are expressed as median (25th–75th percentile) or number (percentage).
Abbreviations: ASA, American Society of Anesthesiologists; BMI, body mass index.

Table 2. - AHI and Responses to the STOP-Bang Questionnaire
Variables Results
AHI, n/h 20 (11–34)
Severity of OSA
 No OSA 5 (4)
 Mild OSA 35 (30)
 Moderate OSA 41 (36)
 Severe OSA 34 (30)
Positive response to STOP-Bang items, n (%)
 Snoring 94 (82)
 Tiredness 81 (70)
 Observed apnea 32 (28)
 Pressure 63 (55)
 BMI >35 10 (9)
 Age >50 y 105 (91)
 Neck >40 cm 64 (56)
 Male sex 89 (77)
Data are expressed as number (percentage) or median (25th–75th percentile).
Abbreviations: AHI, apnea–hypopnea index; BMI, body mass index; OSA, obstructive sleep apnea; STOP-Bang, Sleep Tiredness Observed Pressure-Body mass index Age Neck circumference Gender.

Figure 1.
Figure 1.:
Flowchart of the cohort of 115 patients with a STOP-Bang score of 3–8. Type 3 sleep recordings identified patients with moderate-to-severe OSA, reflected by an AHI >15. AHI indicates apnea–hypopnea index; STOP-Bang, Sleep Tiredness Observed Pressure-Body mass index Age Neck circumference Gender; OSA, obstructive sleep apnea.
Figure 2.
Figure 2.:
Distribution of the 115 patients according to their preoperative STOP-Bang score. STOP-Bang indicates Sleep Tiredness Observed Pressure-Body mass index Age Neck circumference Gender.

The prevalence of moderate-to-severe OSA (AHI >15) was 65% in the whole cohort and was significantly higher in patients with STOP-Bang scores of 5–8 than in patients with STOP-Bang scores of 3–4: 45 of 58 patients (78%) versus 30 of 57 patients (53%), respectively (P < .01). This difference persisted when considering the prevalence of severe OSA (AHI >30): 22 of 58 patients (38%) versus 12 of 57 patients (21%), respectively (P = .05; Figure 1). After adjustment for age, sex, BMI, history of hypertension, diabetes, and major adverse cardiac and cerebrovascular events, patients with a STOP-Bang score of 5–8 had a significantly higher risk of AHI >15 than those with a STOP-Bang score of 3–4, with an odds ratio of 2.9 (95% CI, 1.1–7.8; P = .03).

The alternative scoring models were applied to the subgroup of 57 patients with STOP-Bang scores of 3–4, that is, patients with a STOP of ≥2 and 1 positive question among Bang items (alternative model 1, n = 22 of 57 patients), patients with a serum bicarbonate concentration of ≥28 mmol/L (alternative model 2, n = 15 of 57 patients), or patients who fit the criteria of both models (combination of the 2 models, n = 8 of 57 patients). In alternative model 1, 1 patient had a BMI of >35, 49 patients were >50 years, 17 patients had a neck circumference of >40 cm, and 37 were men. We found no significant differences between the OSA diagnoses of those patients included and those not included in the 2 alternative scoring models or in the combination of the 2 models (Table 3).

Table 3. - The Use of Alternative Scoring Models in the Subgroup of 57 Patients With a Standard STOP-Bang Score of 3–4, That Is, Patients With a STOP ≥2 and 1 Positive Question Among Bang Items (n = 22 Patients; Alternative Model 1), Patients With Preoperative Serum Bicarbonate Concentrations of ≥28 mmol/L (n = 15, Patients) (Alternative Model 2), or a Combination of These 2 Models (n = 8, Patients)
Variables Patients With Alternative Scoring Positive Patients With Alternative Scoring Negative P
STOP ≥2 + 1 Bang item
 AHI >15 14/22 (64) 16/35 (48) .19
 AHI >30 6/22 (27) 6/35 (17) .51
Serum bicarbonate ≥28 mmol/L
 AHI >15 11/15 (73) 19/42 (45) .06
 AHI >30 5/15 (33) 7/42 (17) .27
Combination of the 2 models
 AHI >15 6/8 (75) 25/49 (51) .26
 AHI >30 3/8 (38) 9/49 (18) .35
Data are expressed as number (percentage).
Abbreviations: AHI, apnea–hypopnea index; STOP-Bang, Sleep Tiredness Observed Pressure-Body mass index Age Neck circumference Gender.

Table 4. - Performance of STOP-Bang Scores 5–8, the 2 Alternative Scoring Models and Their Combination to Identify Patients With AHI >15
Models Accuracy Sensitivity Specificity PPV NPV
STOP-Bang 5–8 0.71 (0.63–0.80) 0.89 (0.84–0.95) 0.38 (0.29–0.46) 0.73 (0.65–0.81) 0.65 (0.57–0.74)
Alternative model 1 0.58 (0.45–0.71) 0.47 (0.34–0.60) 0.70 (0.59–0.82) 0.64 (0.51–0.76) 0.54 (0.41–0.67)
Alternative model 2 0.60 (0.47–0.72) 0.37 (0.24–0.49) 0.85 (0.76–0.94) 0.73 (0.62–0.85) 0.55 (0.42–0.68)
Combination 0.20 (0.10–0.30) 0.93 (0.86–0.99) 0.75 (0.64–0.86) 0.51 (0.38–0.64) 0.54 (0.42–0.67)
Data are expressed as mean and 95% confidence interval.
Abbreviations: AHI, apnea–hypopnea index; NPV, negative predictive value; PPV, positive predictive value; STOP-Bang, Sleep Tiredness Observed Pressure-Body mass index Age Neck circumference Gender.

Overall, the validity of the STOP-Bang scores, the 2 alternative models, and their combination to identify patients with AHI >15 is shown in Table 4. The 2 alternative scoring models had an LR+ of 1.6 and 3.2, and an LR− of 0.7 and 0.6, respectively, to predict AHI >15.

DISCUSSION

This prospective external validation confirms that the STOP-Bang questionnaire can preoperatively detect moderate-to-severe OSA patients when the score reaches 5–8. Most of our patients (78%) reaching this score had moderate-to-severe OSA with an AHI ≥15. However, this performance was altered in patients with STOP-Bang scores of 3–4, and alternative scoring models with specific combinations of factors failed to improve the screening of these patients.

The preoperative detection of OSA patients is challenging due to the lack of specificity of clinical symptoms and/or biomarkers. More than 50% of patients have no or nonspecific symptoms for OSA. Despite guidelines and dedicated studies, a significant proportion of patients with symptomatic undiagnosed OSA are not preoperatively identified by anesthesiologists and surgeons.8 There are numerous predictive models to screen for OSA patients.3,17 Initially, the most successful models were complex and not appropriate in the immediate preoperative period because they combined multiple clinical variables with additional morphometry and cephalometry measurements.18,19 Questionnaires were subsequently developed to simplify the process. The Berlin questionnaire consists of 11 questions (snoring, sleepiness, arterial hypertension, BMI)20 but produced a large number of false-negative results.3 The perioperative sleep apnea prediction (P-SAP) score consists of 3 demographic variables, 3 history variables, and 3 airway measures21 but also produced a large number of false-negative results. More recently, a new questionnaire, the NoSAS score—based on BMI, neck circumference, snoring, age, and sex—compared favorably to the Berlin and STOP-Bang questionnaires22; although, it should be noted that a vast majority of subjects in that particular study had a BMI of <30. The STOP-Bang questionnaire is the most widely used among these questionnaires, with the quality of evidence assessed as moderate according to a group of experts.3 Our findings confirm that patients with a preoperative STOP-Bang score of 5–8 are highly suspected to have moderate-to-severe OSA and should be referred to confirm diagnosis. The use of alternative scoring models is unnecessary for this score range.

There are uncertainties regarding the predictive performance of a STOP-Bang score of 3–4 to detect OSA in patients. This midrange score lacks specificity, which results in unnecessary referrals of patients for sleep studies. For example, a man aged 55 years with a history of hypertension has a score of 3 and may have an intermediate risk of OSA, despite having no OSA symptoms or obesity. In the present study, 21% patients with a STOP-Bang score of 3–4 had severe OSA. In a surgical population derived from pooled data, the probability of severe OSA with this score was 25%.11 The predicted probability of patients with this midrange score having moderate-to-severe OSA was 40%.14 It has been shown that male sex and BMI are more predictive of OSA than age and neck circumference.13 Screening performance was improved when sex-specific cutoffs accounting for BMI and neck circumference differences between men and women were applied.23 These findings indicated that items on the Bang questionnaire have uneven predictive weight for OSA, leading to the development of alternative scoring models for patients with a STOP-Bang score of 3–4. As previously proposed by Chung et al,14 we used a STOP score of ≥2 combined with 1 item of Bang—that is, male sex, BMI, or neck circumference. We also tested adding assessment of serum bicarbonate concentrations in patients with STOP-Bang scores of 3–4.12 This criterion was based on observations of chronic nocturnal hypercapnia in 10%–40% of patients with OSA that correlated with the severity of OSA.12,24 A serum bicarbonate concentration of 28 mmol/L was found to be the best cutoff to improve the performance of the STOP-Bang questionnaire.12 However, these 2 alternative scoring models, individually and in combination, were unable to improve the classification of our patients with STOP-Bang scores of 3–4. The proportion of patients with moderate-to-severe OSA was not significantly different whether patients were classified using these alternative scoring models or not. Our group found that increased serum bicarbonate concentrations might reflect multiple morbidities and medication rather than OSA or obesity hypoventilation syndrome.25

How can we explain our inability to externally validate the alternative scoring models? The first reason is the marked differences in the incidences of 2 Bang items between our study and the derivation cohorts12,13: 9% patients with BMI >35 and 77% male patients in the present study versus 26% and 46% in the 2 derivation cohorts, respectively. More specifically, of our patients with STOP-Bang scores of 3–4, 2% had a BMI >35, and 65% were men. Our findings indicate that the proposed alternative scoring models cannot easily be transposed to other populations. Second, imbalanced prevalence of 2 of the 4 items might have altered their informative power, with 1 item being underrepresented (BMI >35) and the other being overrepresented (male sex), giving unbalanced weight for estimating the risk of OSA. Third, we studied patients with a preoperative STOP-Bang score ≥3—that is, intermediate and high risk for OSA. The OSA prevalence in our study was indeed quite high (65%) compared to the OSA prevalence in the 2 derivation cohorts: 41% and 38%, respectively.12,13 When the pretest probability for OSA is high, a positive screen adds little.26 In other words, our high prevalence of OSA might have prevented the alternative scoring models improving the detection of patients with moderate-to-severe OSA. Fourth, the development of the 2 alternative scoring models may be questionable: they were both constructed from patients with STOP-Bang scores of 3–8, that is, they included patients with intermediate and high risk for OSA.12,13 Whether these alternative models improved the screening of patients with a STOP-Bang score of 3–4 was not tested in the derivation cohorts. Future research is required to test other combinations of items with varied weight, or to describe new items, to improve the screening of this midrange score of the STOP-Bang questionnaire. Again, patients with STOP-Bang scores of 3–4 represent the largest group of patients.

This study has several limitations. First, as mentioned above, the predominance of men, the type of surgery (abdominal and urology), and the narrow BMI range in our cohort differed from the derivation cohorts.12,13 Further studies are needed to assess these models in more diverse groups of patients with STOP-Bang scores of 3–4. Second, the size of the whole cohort was limited, although the number of patients with a STOP-Bang score of 3–4 was equally balanced with those with a score of 5–8 (57 vs 58 patients, respectively). This size might have affected the power of statistical analyses, especially within the group of 57 patients with STOP-Bang scores of 3–4. Estimates of diagnostic accuracy—that is, sensitivity, specificity, PPV, and NPV—tend to be imprecise from cohorts of limited size.27 Third, we used a type 3 device for the home sleep apnea test, and this device does not have sensors to differentiate between sleep and wake periods. Such devices estimate AHI from the duration of monitoring, which may underestimate the severity of OSA. However, a recent AASM Position Statement has validated this device as an alternative to polysomnography in uncomplicated adults presenting signs and symptoms possibly related to moderate-to-severe OSA.28 The statement also recommended that raw data recordings must be reviewed and interpreted by a certified sleep medicine physician, as was performed in our study (R.T. and J.-L.P.).

In conclusion, our findings indicate that preoperative patients with a STOP-Bang score of 5–8 can be highly suspected to have moderate-to-severe OSA and should be referred to confirm diagnosis. By contrast, the detection of OSA in patients with STOP-Bang scores of 3–4 could not be improved with the use of alternative scoring models. More research is needed to improve preoperative screening in this subgroup of patients.

ACKNOWLEDGMENTS

The authors thank Meriem Benmerad and Sebastien Bailly (statisticians at Pôle Thorax et Vaisseaux, Centre Hospitalier Universitaire Grenoble Alpes) for their help supervising the statistical analysis.

DISCLOSURES

Name: Ludovic Seguin, MD.

Contribution: This author helped conduct the study, analyze and interpret the data, and draft the manuscript, and gave final approval of the manuscript.

Name: Renaud Tamisier, MD, PhD.

Contribution: This author helped design the study, conduct the study, analyze and interpret the data, and draft the manuscript, and gave final approval of the manuscript.

Name: Baptiste Deletombe, MD.

Contribution: This author helped conduct the study and analyze and interpret the data, and gave final approval of the manuscript.

Name: Mélanie Lopez, BSc.

Contribution: This author helped conduct the study and analyze and interpret the data, and gave final approval of the manuscript.

Name: Jean-Louis Pepin, MD, PhD.

Contribution: This author helped design the study, analyze and interpret the data, and draft the manuscript, and gave final approval of the manuscript.

Name: Jean-François Payen, MD, PhD.

Contribution: This author helped design the study, analyze and interpret the data, and draft the manuscript, and gave final approval of the manuscript.

This manuscript was handled by: Toby Weingarten, MD.

FOOTNOTES

    REFERENCES

    1. Lévy P, Kohler M, McNicholas WT, et al. Obstructive sleep apnoea syndrome. Nat Rev Dis Primers. 2015;1:15015.
    2. Benjafield AV, Ayas NT, Eastwood PR, et al. Estimation of the global prevalence and burden of obstructive sleep apnoea: a literature-based analysis. Lancet Respir Med. 2019;7:687–698.
    3. Kapur VK, Auckley DH, Chowdhuri S, et al. Clinical practice guideline for diagnostic testing for adult obstructive sleep apnea: an American Academy of Sleep Medicine Clinical Practice Guideline. J Clin Sleep Med. 2017;13:479–504.
    4. 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.
    5. Tamisier R, Fabre F, O’Donoghue F, Lévy P, Payen JF, Pépin JL. Anesthesia and sleep apnea. Sleep Med Rev. 2018;40:79–92.
    6. 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.
    7. Ravesloot MJL, de Raaff CAL, van de Beek MJ, et al. Perioperative care of patients with obstructive sleep apnea undergoing upper airway surgery: a review and consensus recommendations. JAMA Otolaryngol Head Neck Surg. 2019 [Epub ahead of print].
    8. Singh M, Liao P, Kobah S, Wijeysundera DN, Shapiro C, Chung F. Proportion of surgical patients with undiagnosed obstructive sleep apnoea. Br J Anaesth. 2013;110:629–636.
    9. Mutter TC, Chateau D, Moffatt M, Ramsey C, Roos LL, Kryger M. A matched cohort study of postoperative outcomes in obstructive sleep apnea: could preoperative diagnosis and treatment prevent complications? Anesthesiology. 2014;121:707–718.
    10. Chung F, Yegneswaran B, Liao P, et al. STOP questionnaire: a tool to screen patients for obstructive sleep apnea. Anesthesiology. 2008;108:812–821.
    11. Nagappa M, Liao P, Wong J, et al. Validation of the STOP-bang questionnaire as a screening tool for obstructive sleep apnea among different populations: a systematic review and meta-analysis. PLoS One. 2015;10:e0143697.
    12. Chung F, Chau E, Yang Y, Liao P, Hall R, Mokhlesi B. Serum bicarbonate level improves specificity of STOP-Bang screening for obstructive sleep apnea. Chest. 2013;143:1284–1293.
    13. Chung F, Yang Y, Brown R, Liao P. Alternative scoring models of STOP-Bang questionnaire improve specificity to detect undiagnosed obstructive sleep apnea. J Clin Sleep Med. 2014;10:951–958.
    14. Chung F, Abdullah HR, Liao P. STOP-bang questionnaire: a practical approach to screen for obstructive sleep apnea. Chest. 2016;149:631–638.
    15. Poupard L, Mathieu M, Goldman M, Chouchou F, Roche F. Multi-modal ECG Holter system for sleep-disordered breathing screening: a validation study. Sleep Breath. 2012;16:685–693.
    16. Berry RB, Budhiraja R, Gottlieb DJ, et al.; American Academy of Sleep Medicine. Rules for scoring respiratory events in sleep: update of the 2007 AASM Manual for the Scoring of Sleep and Associated Events. Deliberations of the sleep apnea definitions task force of the American Academy of Sleep Medicine. J Clin Sleep Med. 2012;8:597–619.
    17. Ramachandran SK, Josephs LA. A meta-analysis of clinical screening tests for obstructive sleep apnea. Anesthesiology. 2009;110:928–939.
    18. Kushida CA, Efron B, Guilleminault C. A predictive morphometric model for the obstructive sleep apnea syndrome. Ann Intern Med. 1997;127:581–587.
    19. Battagel JM, L’Estrange PR. The cephalometric morphology of patients with obstructive sleep apnoea (OSA). Eur J Orthod. 1996;18:557–569.
    20. Netzer NC, Stoohs RA, Netzer CM, Clark K, Strohl KP. Using the Berlin Questionnaire to identify patients at risk for the sleep apnea syndrome. Ann Intern Med. 1999;131:485–491.
    21. Ramachandran SK, Kheterpal S, Consens F, et al. Derivation and validation of a simple perioperative sleep apnea prediction score. Anesth Analg. 2010;110:1007–1015.
    22. Marti-Soler H, Hirotsu C, Marques-Vidal P, et al. The NoSAS score for screening of sleep-disordered breathing: a derivation and validation study. Lancet Respir Med. 2016;4:742–748.
    23. Mou J, Pflugeisen BM, Crick BA, et al. The discriminative power of STOP-Bang as a screening tool for suspected obstructive sleep apnea in clinically referred patients: considering gender differences. Sleep Breath. 2019;23:65–75.
    24. Mokhlesi B. Obesity hypoventilation syndrome: a state-of-the-art review. Respir Care. 2010;55:1347–1362; .
    25. Borel JC, Guerber F, Jullian-Desayes I, et al. Prevalence of obesity hypoventilation syndrome in ambulatory obese patients attending pathology laboratories. Respirology. 2017;22:1190–1198.
    26. Bianchi M. Screening for obstructive sleep apnea: Bayes weighs in. Open Sleep J. 2009;2:56–59.
    27. Cohen JF, Korevaar DA, Altman DG, et al. STARD 2015 guidelines for reporting diagnostic accuracy studies: explanation and elaboration. BMJ Open. 2016;6:e012799.
    28. Rosen IM, Kirsch DB, Carden KA, et al.; American Academy of Sleep Medicine Board of Directors. Clinical use of a home sleep apnea test: an updated American Academy of Sleep Medicine position statement. J Clin Sleep Med. 2018;14:2075–2077.
    Copyright © 2020 International Anesthesia Research Society