ADVERTISEMENT

Issue Navigator

Volume 11 No. 07
Earn CME
Accepted Papers





Scientific Investigations

Obstructive Sleep Apnea in Obese Hospitalized Patients: A Single Center Experience

Sunil Sharma, MD1; Paul J. Mather, MD2; Jimmy T. Efird, PhD, MSc3,4; Daron Kahn, MD1; Kristin Y. Shiue, MPH3,4; Mohammed Cheema, MD1; Raymond Malloy, RRT1; Stuart F. Quan, MD5,6
1Pulmonary and Critical Care Medicine, Jefferson Sleep Disorders Center, Jefferson Medical College of Thomas Jefferson University, Philadelphia, PA; 2Advanced Heart Failure and Cardiac Transplant Center, Jefferson Medical College of Thomas Jefferson University, Philadelphia, PA; 3East Carolina Heart Institute, Department of Cardiovascular Sciences, Brody School of Medicine, East Carolina University, Greenville, NC; 4Center for Health Disparities, Brody School of Medicine, East Carolina University, Greenville, NC; 5Division of Sleep Medicine, Harvard Medical School, Boston, MA; 6Arizona Respiratory Center, University of Arizona College of Medicine, Tucson, AZ

ABSTRACT

Study Objectives:

Obstructive sleep apnea (OSA) is an important health problem associated with significant morbidity and mortality. This condition often is underrecognized in hospitalized patients. The aim of this study was to conduct a clinical pathway evaluation (CPE) among obese patients admitted to a tertiary care hospital. We also assessed oxygen desaturation index (ODI, measured by overnight pulse oximetry) as a potential low-cost screening tool for identifying OSA.

Methods:

This was a prospective study of 754 patients admitted to an academic medical center between February 2013 and February 2014. Consecutive obese patients (body mass index ≥ 30) admitted to the hospital (medical services) were screened and evaluated for OSA with the snoring, tiredness during daytime, observed apnea, high blood pressure (STOP) questionnaire. The admitting team was advised to perform follow-up evaluation, including polysomnography, if the test was positive.

Results:

A total of 636 patients were classified as high risk and 118 as low risk for OSA. Within 4 w of discharge, 149 patients underwent polysomnography, and of these, 87% (129) were shown to have OSA. An optimal screening cutoff point for OSA (apnea-hypopnea index ≥ 10/h) was determined to be ODI ≥ 10/h [Matthews correlation coefficient = 0.36, 95% confidence interval = 0.24–0.47]. Significantly more hospitalized patients were identified and underwent polysomnography compared with the year prior to introduction of the CPE.

Conclusions:

Our results indicate that the CPE increased the identification of OSA in this population. Furthermore, ODI derived from overnight pulse oximetry may be a cost-effective strategy to screen for OSA in hospitalized patients.

Citation:

Sharma S, Mather PJ, Efird JT, Kahn D, Shiue KY, Cheema M, Malloy R, Quan SF. Obstructive sleep apnea in obese hospitalized patients: a single center experience. J Clin Sleep Med 2015;11(7):717–723.


Obstructive sleep apnea (OSA) is a common sleep disorder, present in 10–20% of the general population.1,2 The condition is especially prevalent (greater than 50%) in patients with congestive heart failure (CHF), atrial fibrillation (AF), and diabetes mellitus (DM).35 OSA also has been associated with significant cardiovascular complications including hypertension (HTN), CHF, AF, coronary artery disease (CAD), and stroke.3,4,69 Although patients with chronic comorbid conditions associated with OSA are frequently hospitalized and have higher overall healthcare utilization/costs,10 the coexistence and synergistic influence of OSA often are undetected. The identification of OSA in these patients is important because risk may be mitigated by treatment with continuous positive airway pressure.11

BRIEF SUMMARY

Current Knowledge/Study Rationale: OSA in hospitalized patients has not been systematically examined. Hospitalized patients have high co-morbidities and may be at high risk for cardiovascular complications. We hypothesized that there is high prevalence and under-diagnosis of OSA in obese hospitalized patients. A clinical pathway evaluation (CPE) may be an effective tool for early recognition of this condition.

Study Impact: The study shows significant under-recognition of OSA in obese hospitalized patients. Furthermore, overnight pulse-oximetry is a simple, cost effective tool for detecting sleep apnea in hospitalized patients.

In a recent chart review study of patients admitted with acute myocardial infarction (MI), OSA was found to be significantly underdiagnosed when prospectively evaluated by overnight polysomnography (PSG).11 Similarly, patients admitted to the hospital with chronic obstructive pulmonary disease (COPD) exacerbations have been found to have increased prevalence of OSA compared with stable outpatients.12 Recent data suggest that OSA increases long-term morbidity and mortality, and may also be implicated in sudden nocturnal death.13,14 Another study revealed that the presence of OSA in patients admitted for pneumonia was associated with increased risk for mechanical ventilation, clinical deterioration, and resource utilization.15 Limited data on OSA in hospitalized patients suggest a high prevalence based on screening but without any confirmatory testing.16 Systematic evaluation of this disorder in hospitalized patients is urgently required to fill the knowledge gap in this area. The problem is compounded by low awareness of this disorder among admitting services.

We hypothesized that a clinical pathway evaluation (CPE) may help identify and treat obese hospitalized patients at high risk for OSA. We also hypothesized that the prevalence of OSA in this population is high and that overnight pulse oximetry may be a cost-effective tool in identifying OSA in hospitalized patients.

METHODS

This was a prospective study of patients admitted to an academic tertiary care hospital from March 2013 to February 2014. The Thomas Jefferson University Institutional Review Board approved the study. Patients presenting to the cardiology, internal medicine, and family practice services with a body mass index (BMI, weight [kg]/height, [m2]) ≥ 30 were screened with the STOP (snoring, tiredness during daytime, observed apnea, high blood pressure) questionnaire by a respiratory therapist.17 The STOP questionnaire was chosen because of its brevity, previous validation as a screening instrument, and ease of use. This made it an ideal instrument that would not increase the work burden of the respiratory therapist or participant burden of the patient. The admitting team was notified if the STOP questionnaire was positive (answering yes to two or more items) who then contacted a board-certified pulmonary sleep medicine physician (inpatient services) to determine whether a follow-up consultation was necessary (Figure 1). Further consultation involved the patient undergoing a comprehensive sleep history and physical examination and nocturnal pulse oximetry, unless the latter was contraindicated (e.g., oxygen requirement > 30%, severe pain, insomnia, altered mental status, anticipated disruption during sleep (imaging/tests or surgeries). If clinical suspicion of OSA was high (defined according to the Adult OSA Task Force of the American Academy of Sleep Medicine [AASM]),18 patients were advised to undergo a postdischarge confirmatory PSG study. Patients admitted during the weekend were not included in the study.

Flow chart of patient undergoing in-hospital evaluation.

jcsm.11.7.717a.jpg

jcsm.11.7.717a.jpg
Figure 1

Flow chart of patient undergoing in-hospital evaluation.

(more ...)

PSG evaluations were conducted at the Jefferson Sleep Disorder Center, an AASM accredited facility, and included an electrocardiograph, electroencephalograph, continuous oronasal airflow recording (with a thermistor and a pressure transducer), recording of chest wall and abdomen movement (using respiratory inductive plethysmography belts), transcutaneous oximetry, and chin electromyography (Comet AS 40 PSG, Grass Technologies; Warwick, RI, USA). Sleep was staged manually according to AASM scoring guidelines by a registered PSG technologist.19 The apnea-hypopnea index (AHI) was calculated as the number of apneas plus hypopneas per hour of total sleep time, with hypopneas defined using a desaturation criterion of 4%. A single, board-certified sleep physician interpreted the polysomnograms.

Overnight pulse oximetry was performed using a Masimo RAD-57 machine (Irvine, CA) standardized to an average time of 3 sec and oxygen desaturation index cutoff of 4%, which is consistent with the AASM's recommendations.19

Statistical Analysis

Categorical variables were presented as frequency and percentage whereas continuous variables were presented as a mean (plus or minus 1 standard deviation (SD), median, and inter-quartile range (IQR). Variables not previously categorized were divided into quartiles prior to statistical analysis. Quartile categorization is beneficial because it limits the influence of outliers and allows for the assessment of trends across categories.

Statistical significance for categorical variables was tested using a chi-square test for multiway comparisons. The nonparametric Deuchler-Wilcoxon method was used to assess continuous variables. Trend for increasing/decreasing levels of a variable was assessed using either a likelihood ratio test (univariable model) or score test (multivariable model). An iterative expectation-maximization (EM) algorithm was used to account for missing values. A confusion matrix was generated to summarize standard classification measures. The Matthews correlation coefficient (MCC), based on the geometric mean of the regression beta terms, was used to determine the optimal cutoff point for ODI.

A Tukey mean-difference (Bland-Altman) plot was used to assess the agreement between ODI and AHI values. In this exploratory plot, the x-axis represents the maximum-likelihood average estimate of the true result, whereas the y-axis is the computed difference between the two recordings.20 Measures are considered to have poor agreement if greater than half of the plot points fall within the discordance region outside the 95% confidence limits for the mean difference line.

Statistical significance was defined as p < 0.05. SAS Version 9.3 (Cary, NC, USA) was used for all analyses.

RESULTS

A total of 754 patients with body mass index (BMI) ≥ 30 kg/m2 underwent screening from February 2013 to February 2014. Of these, 636 were classified as high risk and 118 as low risk for OSA. The mean age was 59 ± 14 y for high-risk patients and 59 ± 18 y for low-risk patients. The mean BMI in the high-risk cohort was 37 ± 9.7 kg/m2 compared with 33 ± 4.9 kg/m2 in the low risk cohort (p < 0.001). Of the 636 patients who were found to be high risk, 410 received further consultation from inpatient services (Figure 1). As noted in Figure 1, 226 were not consulted for comprehensive sleep evaluation. The most common reason for declining was acuteness of the admitting illness. We also noted that despite the primary team approaching the patient and suggesting further evaluation, a significant number of patients refused consultation, suggesting poor awareness of potential effect of sleep disordered breathing at the community level. Interestingly, the pressures of keeping the length of stay low was evident by the number of patients not getting evaluated in hospital due to the patient getting discharged within 24 h of the screening.

Of the 410 patients who had a consultation, 149 patients underwent in-laboratory PSG at the Jefferson Sleep Disorder Center within 4 w of discharge. One hundred twenty nine patients (87%) were positive for OSA and 65% had moderate to severe OSA, with AHI ≥ 15/h. Compared with the previous year, when only one patient was directly referred to the sleep center from inpatient services, there was a significant increase in referrals following the introduction of the program (Figure 2). Furthermore, during the study period, 30% of house staff admission notes had no reference to prior history of OSA. More than 70% of admission notes had no reference to sleep history.

Comparison of number of patients who underwent overnight polysomnography within 4 w of discharge from the hospital before and after the introduction of the service.

Bars on right side indicate percentage of patients with moderate-severe and mild sleep apnea in this group. OSA, obstructive sleep apnea.

jcsm.11.7.717b.jpg

jcsm.11.7.717b.jpg
Figure 2

Comparison of number of patients who underwent overnight polysomnography within 4 w of discharge from the hospital before and after the introduction of the service.Bars on right side indicate percentage of patients with moderate-severe and mild sleep apnea in this group. OSA, obstructive sleep apnea.

(more ...)

Multivariable analysis revealed that BMI, male sex, CAD, CHF, HTN, and chronic obstructive lung disease (COPD/ asthma) were statistically significant independent risk factors for high risk of OSA (Table 1). Overnight pulse oximetry was performed on 232 patients as part of clinical care. Patients were generally maintained on room air. However, those who could not be weaned (< 1%) remained on minimal oxygen (FiO2 < 30%). A few pulse oximetry (N = 5) results were not interpretable because of technical failures (e.g., probe falling off finger, device malfunction). Baseline characteristics of patients undergoing PSG were recorded and compared with patients who did not have a PSG (Table 2). Although patients on the cardiology service were screened, there were few patients with central apneas and none met criteria for the diagnosis of central sleep apnea. There were no significant differences between those patients who had a PSG and those who did not. PSG was performed in the outpatient setting within 4 w of discharge on 85 of the patients that received overnight pulse oximetry, mean ODI was 17 ± 15/h, whereas mean AHI was 31 ± 31/h. ODI underestimated AHI in 58% of the patients (50/85).

Baseline characteristics and risk factors.

jcsm.11.7.717.t01.jpg

table icon
Table 1

Baseline characteristics and risk factors.

(more ...)

Comparison of baseline characteristics of patient undergoing polysomnography to those who did not have polysomnography.

jcsm.11.7.717.t02.jpg

table icon
Table 2

Comparison of baseline characteristics of patient undergoing polysomnography to those who did not have polysomnography.

(more ...)

The Bland-Altman plot (Figure 3) revealed that there was no major bias when using ODI versus AHI (PSG derived) to define OSA, although it appears that ODI slightly underestimates AHI. To determine an optimal cutoff point for clinical application, we generated a confusion matrix (Tables 3 and 4). Examining this matrix, an optimal cutoff point (based on MCC) was determined to be ODI ≥ 10/h when AHI ≥ 10/h. The corresponding likelihood ratio for a positive test was 3.0 (95% CI = 1.8–5.1, p < 0.0001), indicating a 200% increased posttest odds. Furthermore, we observed that an ODI < 15/h resulted in few false-positive diagnoses, whereas an ODI < 5 limited false negative diagnoses (Table 5).

Confusion matrix for oxygen desaturation index by apnea-hypopnea index ≥ 5.

jcsm.11.7.717.t03.jpg

table icon
Table 3

Confusion matrix for oxygen desaturation index by apnea-hypopnea index ≥ 5.

(more ...)

Confusion matrix for oxygen desaturation index by apnea-hypopnea index ≥ 10/h.

jcsm.11.7.717.t04.jpg

table icon
Table 4

Confusion matrix for oxygen desaturation index by apnea-hypopnea index ≥ 10/h.

(more ...)

Diagnostic value of hospital nocturnal pulse oximetry in identifying obstructive sleep apnea.

jcsm.11.7.717.t05.jpg

table icon
Table 5

Diagnostic value of hospital nocturnal pulse oximetry in identifying obstructive sleep apnea.

(more ...)

Bland-Altman plot illustrating the agreement between oxygen desaturation index (ODI) and apneahypopnea index (AHI) values.

Solid line indicates mean value, dotted lines indicate 95% confidence bounds.

jcsm.11.7.717c.jpg

jcsm.11.7.717c.jpg
Figure 3

Bland-Altman plot illustrating the agreement between oxygen desaturation index (ODI) and apneahypopnea index (AHI) values.Solid line indicates mean value, dotted lines indicate 95% confidence bounds.

(more ...)

DISCUSSION

Few currently published studies systematically examine the issue of OSA in hospitalized patients. Our prospectively collected data from a CPE suggest that OSA is underrecognized in obese hospitalized patients. Furthermore, the established screening process yields a high rate of OSA (87%), with 65% of patients having moderate to severe OSA based on AHI. Our findings also show that ODI derived from overnight pulse oximetry in the hospital has high correlation with AHI derived from postdischarge in-laboratory PSG, and ODI ≥ 10/h signified an increased likelihood of the patient having an AHI ≥ 10/h. Risk factors that were independently associated with high risk for OSA in hospitalized patients included a high BMI, male sex, CAD, CHF, and COPD. Age, ethnicity, and renal dysfunction did not appear to be associated with high risk in this cohort.

OSA has significant cardiovascular implications21 and is associated with sudden death.13 Although our finding of a high prevalence of OSA in this obese cohort with a substantial number of comorbidities is not necessarily surprising, our results highlight the underdiagnosis of OSA in a hospital setting. This may be partially explained by the lack of awareness of OSA in hospitalized settings. In our study, this is illustrated by the minimal documentation of prior OSA history by resident physicians and the finding of only one sleep referral from the hospital in the previous year. Recognizing this condition in a timely and efficient manner may lead to fewer readmissions and ultimately reduce costs for this at-risk population.22

The STOP questionnaire was used because of its recognized clinical validity among pre-operative surgical patients, simple format, interpretability, low cost, and ability to be self-administered.23 Our preliminary results indicate significantly improved yield at a lower cost by combining the STOP questionnaire (high sensitivity) with overnight pulse oximetry (high specificity) testing. Recent articles have shown high sensitivity of the STOP-BANG questionnaire, an extension of the STOP, in obese patients25,26 Our data derived from an obese population is consistent with these previous observations. We have noted an increasing number of direct consultations (not initiated by positive STOP questionnaire) with time, suggesting increasing awareness among admitting physicians and house staff.

The key to appropriate use of pulse oximetry testing in the hospital setting is to identify and exclude patients most likely to have an uninterpretable test (e.g., high oxygen requirement, severe pain, severe insomnia, altered mental status, frequent anticipated disturbances during the night, and severe restless legs syndrome). In this regard, the standardization of pulse oximetry (averaging time of 3 sec with ODI of 4% using same model of device) is essential to facilitate agreement with PSG. In contrast to other reports in the literature, the overestimation of AHI was minimized in our analysis, primarily because of clinical correlation and standardization of pulse oximeter parameters.24 Our data also suggest that an ODI ≥ 15/h essentially rules out false- positive diagnoses (Table 5) and can be a cost-effective tool.

The strengths of this study were its large sample, systematic collection of data, and use of validated measures for ODI and AHI. However, a few limitations should be noted. Only patients with BMI ≥ 30 kg/m2 were screened with the STOP questionnaire and patients with low BMIs and OSA may have been missed. Screening was conducted on only three services (cardiology, internal medicine, and family practice). Thus, our results may not generalize to other hospital-based services such as surgery or to patients who are not obese. Furthermore, we recognize that as with any screening procedure, some of the 118 patients who were negative on the STOP questionnaire may have been falsely so. However, there is a low probability that they had severe OSA.23 Although screening and consultation were not conducted over weekends, we do not believe that patients admitted over weekends would have differed significantly compared with those admitted during the week. Some patients may have undergone sleep studies at facilities closer to home or may have opted not to undergo the study, which potentially affected the true prevalence of the disorder. However, the overall demographic and clinical characteristics of this group did not substantively differ from the patients tested at our center (Table 2). The mean ODI in the non-PSG group (ODI = 14/h) was not significantly different than that in the PSG group, suggesting a high prevalence of OSA in this population. In addition, overnight pulse oximetry was not performed on patients with an oxygen requirement > 30%. However, potential bias was minimized through a restriction on tests for patients with a significant oxygen requirement, insomnia, pain, poor sleep hygiene, or severe restless leg symptoms.

CONCLUSION

This large, systematic prospective CPE shows that there is a significant burden of unrecognized OSA in obese hospitalized patients. Furthermore, we observed that overnight pulse oximetry correlates well with PSG derived AHI, and potentially can be used as a low-cost screening device in this population. However, expanded testing in hospitalized patient populations, across institutions, and over a multiyear period will provide objective statistical data to validate the comparative effectiveness, reliability, and long-term cost savings of this methodology.

DISCLOSURE STATEMENT

This was not an industry supported study. Dr. Sharma has received research support from ResMed. The other authors have indicated no financial conflicts of interest.

ABBREVIATIONS

AF

atrial fibrillation

AHI

apnea hypopnea index

BMI

body mass index

CAD

coronary artery disease

CHF

congestive heart failure

COPD

chronic obstructive pulmonary disease

CPE

clinical pathway evaluation

DM

diabetes mellitus

EM

expectation-maximization

HTN

hypertension

IQR

interquartile range

MCC

Matthew's correlation coefficient

MI

myocardial infarction

ODI

oxygen desaturation index

OSA

obstructive sleep apnea

PSG

polysomnography

SD

standard deviation

REFERENCES

1 

Peppard PE, Young T, Barnet JH, Palta M, Hagen EW, Hla KM, authors. Increased prevalence of sleep-disordered breathing in adults. Am J Epidemiol. 2013;177:1006–14. [PubMed Central][PubMed]

2 

Young T, Palta M, Dempsey J, Peppard PE, Nieto FJ, Hla KM, authors. Burden of sleep apnea: rationale, design, and major findings of the Wisconsin Sleep Cohort study. WMJ. 2009;108:246–9. [PubMed Central][PubMed]

3 

Sin DD, Fitzgerald F, Parker JD, Newton G, Floras J, Bradley TD, authors. Risk factors for central and obstructive sleep apnea in 450 men and women with congestive heart failure. Am J Respir Crit Care Med. 1999;160:1101–6. [PubMed]

4 

Gami AS, Hodge DO, Herges RM, et al., authors. Obstructive sleep apnea, obesity, and the risk of incidence atrial fibrillation. J Am Coll Cardiol. 2007;49:565–71. [PubMed]

5 

Resnick HE, Redline S, Shahar E, et al., authors. Diabetes and sleep disturbances: findings from the Sleep Heart Health Study. Diabetes Care. 2003;26:702–9. [PubMed]

6 

Javaheri S, Parker TJ, Liming JD, et al., authors. Sleep apnea in 81 ambulatory male patients with stable heart failure. Circulation. 1998;97:2154–9. [PubMed]

7 

Peppard PE, Young T, Palta M, Skatrud J, authors. Prospective study of the association between sleep-disordered breathing and hypertension. N Engl J Med. 2000;342:1378–84. [PubMed]

8 

Gottlieb DJ, Yenokyan G, Newman AB, et al., authors. Prospective study of obstructive sleep apnea and incident coronary heart disease and heart failure: the Sleep Heart Health Study. Circulation. 2010;122:352–60. [PubMed Central][PubMed]

9 

Arzt M, Young T, Finn L, Skatrud J, Bradley TD, authors. Association of sleep-disordered breathing and the occurrence of stroke. Am J Respir Crit Care Med. 2005;172:1447–51. [PubMed Central][PubMed]

10 

Diaz K, Faverio P, Hospenthal A, Restrepo MI, Amuan ME, Pugh MV, authors. Obstructive sleep apnea is associated with higher healthcare utilization in elderly patients. Ann Thorac Surg. 2014;9:92–8.

11 

Konecny T, Sert-Kuniyoshi FH, Orban M, et al., authors. Under-diagnosis of sleep apnea in patients after acute myocardial infarction. J Am Coll Cardiol. 2010;56:742–3. [PubMed Central][PubMed]

12 

Turcani P, Skrickova J, Pavlik T, Janousova E, Orban M, authors. The prevalence of obstructive sleep apnea in patients hospitalized for COPD exacerbation. Biomed Pap Med Fac Univ Palacky Olomouc Czech. Repub 2014 Feb 25. [Epub ahead of print].

13 

Gami AS, Olson EJ, Shen WK, et al., authors. Obstructive sleep apnea and the risk of sudden cardiac death: a longitudinal study of 10,701 adults. J Am Coll Cardiol. 2013;62:610–6. [PubMed]

14 

Garcia-Touchard A, Somers VK, Kara T, et al., authors. Ventricular ectopy during REM sleep: implications for nocturnal sudden cardiac death. Nat Clin Pract Cardiol. 2007;4:284–8.

15 

Lindenauer PK, Stefan MS, Johnson KG, Priya A, Pekow PS, Rothberg MB, authors. Prevalence, treatment, and outcomes associated with OSA among patients hospitalized with pneumonia. Chest. 2014;145:1032–8. [PubMed Central][PubMed]

16 

Auckley D, Ramsammy V, Shalhoub GJ, Khanna G, authors. Suspected OSA in hospitalized patients: prevalence and potential for adverse events. J Sleep Sleep Disord Res. 2010;33.

17 

Chung F, Yegneswaran B, Liao P, et al., authors. STOP questionnaire: a tool to screen patients for obstructive sleep apnea. Anesthesiology. 2008;108:812–21. [PubMed]

18 

Epstein LJ, Kristo D, Strollo PJ, et al., authors. Clinical guideline for the evaluation, management and long-term care of obstructive sleep apnea in adults. J Clin Sleep Med. 2009;5:263–76. [PubMed Central][PubMed]

19 

Berry RB, Brooks R, Gamaldo CE, et al., authors. The AASM manual for the scoring of sleep and associated events: rules, terminology and technical specifications. Version 2.1. www.aasmnet.org. Darien, IL: American Academy of Sleep Medicine, 2013.

20 

Bland JM, Altman DG, authors. Statistical methods for assessing agreement between two methods of clinical measure. Lancet. 1986;327:307–10.

21 

Mehra R, Benjamin EJ, Shahar E, et al., authors. Association of nocturnal arrhythmias with sleep-disordered breathing: the Sleep Heart Health Study. Am J Respir Crit Care Med. 2006;173:910–6. [PubMed Central][PubMed]

22 

Kauta SR, Keenan BT, Goldberg L, Schwab R, authors. Diagnosis and treatment of sleep disordered breathing in hospitalized cardiac patients-a reduction in 30-day hospital readmission rates. J Clin Sleep Med. 2014;10:1051–9. [PubMed Central][PubMed]

23 

Boynton G, Vahabzadeh A, Hammoud S, Ruzicka DL, Chervin RD, authors. Validation of the STOP-BANG Questionnaire among patients referred for suspected obstructive sleep apnea. J Sleep Disord Treat Care. 2013;2(4). [PubMed Central][PubMed]

24 

Skinner MA, Choudhury MS, Homan SD, Cowan JO, Wilkins GT, Taylor DR, authors. Accuracy of monitoring for sleep-related breathing disorders in the coronary care unit. Chest. 2005;127:66–71. [PubMed]

25 

Liu A, Kushida CA, Reaven GM, authors. Risk for obstructive sleep apnea in obese, nondiabetic adults varies with insulin resistance status. Sleep Breath. 2013;17:333–8. [PubMed Central][PubMed]

26 

Chung F, Yang Y, Liao P, authors. Predictive performance of the STOP-Bang score for identifying obstructive sleep apnea in obese patients. Obes Surg. 2013;23:2050–7. [PubMed]