ADVERTISEMENT

Issue Navigator

Volume 11 No. 10
Earn CME
Accepted Papers
Classifieds





Scientific Investigations

Prevalence of Undetected Sleep Apnea in Patients Undergoing Cardiovascular Surgery and Impact on Postoperative Outcomes

Nancy Foldvary-Schaefer, DO, MS1; Roop Kaw, MD2; Nancy Collop, MD3; Noah D. Andrews, RPSGT1; James Bena, MS4; Lu Wang, MS4; Tracey Stierer, MD5; Marc Gillinov, MD6; Matt Tarler, PhD7; Hani Kayyali, MBA, MS7
1Cleveland Clinic Sleep Disorders Center, Cleveland, OH; 2Cleveland Clinic Department of Hospital Medicine and Anesthesiology Outcomes Research, Cleveland, OH; 3Emory Sleep Disorders Center, Atlanta, GA; 4Cleveland Clinic Quantitative Health Sciences, Cleveland, OH; 5Johns Hopkins Department of Anesthesiology, Baltimore, MD; 6Cleveland Clinic Department of Cardiac and Thoracic Surgery, Cleveland, OH; 7Cleveland Medical Devices Inc., Cleveland, OH

ABSTRACT

Study Objectives:

We examined the prevalence of obstructive sleep apnea (OSA) among patients undergoing cardiac surgery and its impact on postoperative outcomes.

Methods:

Subjects were recruited from inpatient cardiovascular surgery units at two tertiary care centers. Crystal Monitor 20-H recorded polysomnograms preoperatively. Regression analyses were performed to explore associations between OSA using different apnea-hypopnea index (AHI) cutoffs and postoperative outcomes adjusting for key covariates. Prevalence of postoperative outcomes was compared among groups defined by AHI and left ventricle ejection fraction (LVEF) median cutoffs.

Results:

Of 107 participants, the AHI was ≥ 5 in 79 (73.8%), ≥ 10 in 63 (58.9%), ≥ 15 in 51(47.7%), and ≥ 30 in 29 (27.1%). Patients with AHI ≥ 15 had significantly lower LVEF (p < 0.001). Logistic regression analyses with OSA cutoffs as above adjusting for age, gender, race, BMI, and LVEF found no significant increase in odds for any postoperative outcomes. No significant differences were found in %Total sleep time (TST) with SpO2 < 90% between AHI or LVEF groups, or by presence/absence of complications. Patients with any amount of TST with SpO2 < 90% had greater BMI, longer OR tube time, and greater prevalence of prolonged intubation (p = 0.007, 0.035, 0.038, respectively).

Conclusions:

OSA is highly prevalent in patients undergoing cardiovascular surgery. It could not be shown that OSA was significantly associated with adverse postoperative outcomes, but this may have been due to an insufficient number of subjects. AHI ≥ 15 was associated with lower LVEF. Larger samples are required to explore the impact of OSA on key postoperative outcomes that have clinical and economic importance in the care of cardiovascular surgery populations.

Commentary:

A commentary on this article appears in this issue on page 1081.

Citation:

Foldvary-Schaefer N, Kaw R, Collop N, Andrews ND, Bena J, Wang L, Stierer T, Gillinov M, Tarler M, Kayyali H. Prevalence of undetected sleep apnea in patients undergoing cardiovascular surgery and impact on postoperative outcomes. J Clin Sleep Med 2015;11(10):1083–1089.


Obstructive sleep apnea (OSA) is a known predictor of incident coronary heart disease and related mortality among the male population. Males with an apnea hypopnea index (AHI) ≥ 30 are 58% more likely to develop incident heart failure (HF), and those 40 to 70 years of age are 68% more likely to develop coronary heart disease compared to patients without OSA (AHI < 5).1

The reported prevalence of OSA (AHI ≥ 15) is 65.7% among patients admitted to the hospital with acute myocardial infarction.2 Patients with OSA have been reported to have higher incidence of major adverse cardiac events after percutaneous coronary intervention (PCI).3 Treatment for OSA in patients undergoing PCI has been associated with reduced mortality at 5 years (cardiac mortality: p = 0.027 and overall mortality p = 0.058) compared to untreated OSA patients, but no change was noted in major adverse cardiac or cerebrovascular events between groups.4 In another study, patients with OSA (AHI ≥ 10) were found to have a higher degree of late lumen loss (restenosis) > 6 months after PCI compared to those with AHI < 10.5

BRIEF SUMMARY

Current Knowledge/Study Rationale: Given the high prevalence of OSA among patients with cardiovascular disease we aimed to study the prevalence of OSA in patients undergoing cardiac surgery and the impact of unrecognized sleep apnea on perioperative and postoperative outcomes after cardiac surgery. We also obtained the necessary recordings to finalize upgrade of the Crystal Monitor 20-H wireless data transmission protocol for the hospital environment (Cleveland Medical Inc.).

Study Impact: While many perioperative and postoperative complications were numerically greater in patients with more severe sleep apnea, we did not find significant differences in complications between patients with and without sleep apnea. Given the high prevalence of sleep apnea observed, further studies exploring the impact of sleep apnea on surgical outcomes in cardiac surgery populations are warranted.

Literature regarding outcomes after cardiac surgery in patients with OSA is sparse, and little is known about the impact of OSA on postoperative outcomes.6,7 Results from large databases that included cardiac surgery among other surgeries have reported adverse postoperative outcomes, including emergent intubation, respiratory failure, cardiac complications, and ICU transfer.8,9 Given the high prevalence of OSA among patients with cardiovascular disease, we aimed to study (1) the prevalence of OSA among patients undergoing cardiac surgery at two major tertiary care facilities in the United States, and (2) the impact of unrecognized sleep apnea (SA) on perioperative and postoperative outcomes after cardiac surgery while obtaining necessary recordings to finalize the upgrade of the Crystal Monitor 20-H wireless data transmission protocol dedicated for the hospital environment (Cleveland Medical Devices Inc.).

METHODS

Subjects were recruited from the inpatient cardiovascular surgery services of the Cleveland Clinic and Johns Hopkins Medical Center.

Subject Selection

Subjects meeting the following criteria were invited to participate:

  • Ability to provide written informed consent and comply with study procedures

  • At least 18 years of age

  • Planned cardiovascular surgery under general anesthesia scheduled more than 1 day after but less than 60 days from the time of enrollment

Subjects treated for SA with positive airway pressure (PAP) therapy or an oral appliance in the prior 30 days and those unable to comply with study procedures were excluded.

Study Procedures

The study was approved by the institutional review boards of both sites. Subjects provided informed consent prior to initiation of study procedures.

Demographic and Medical History Data

All data were obtained through electronic medical record (EMR) review. Demographic data included age, race, gender, height, and weight. Medical history including diabetes mellitus, hypertension, hyperlipidemia, coronary artery disease, myocardial infarction, cardiomyopathy, HF, valve disease, arrhythmia, endocarditis, stroke, gastroesophageal reflux, hypothyroidism, COPD, kidney disease, and mood disorders (anxiety or depression) was noted. Cardiac surgery type was classified as coronary artery bypass graft (CABG), single valve repair/replacement, CABG and single valve repair/replacement or > 2 valve repair/replacement, or others including septal myomectomy, right atrial mass removal, cardiac catheterization, and cardioverter defibrillator placement. Left ventricular ejection fraction (LVEF) was obtained from the most recent echocardiogram within 6 months of study enrollment. Subjects completed the Epworth Sleepiness Scale (ESS), Berlin scale, Sleep Apnea Scale of the Sleep Disorders Questionnaire (SA/SDQ), and Global Sleep Assessment Questionnaire (GSAQ). The results of these assessments and their predictive value for moderate-to-severe SA will be presented elsewhere.

Polysomnography

PSGs were performed using the Crystal Monitor 20-H transmitting PSG data from the patient's room to the nursing station or sleep laboratory where a registered sleep technologist monitored the patient in real time. Recorded sensors included 2-channel electroencephalogram (C3-A2, O2-A1), left and right electro-oculogram (EOG), chin and bilateral tibialis anterior electromyogram (EMG), electrocardiogram (EKG), naso-oral thermistor, nasal pressure transducer, thoracic and abdominal effort using piezo electric belts, pulse oximetry, and body position. Snoring was derived from the airflow channel. Staging and event scoring was performed using standard criteria and the alternate hypopnea definition (≥ 30% amplitude reduction on the nasal pressure signal with either ≥ 3% oxygen desaturation or an arousal).10 Studies were interpreted by sleep medicine board-certified sleep specialists (NFS, NC) blinded to the clinical information. The cardiovascular treatment team received notification of participants with SA (AHI ≥ 5), and the investigative team was available to assist in the initiation of PAP therapy when requested.

Outcome Assessments

One month postoperatively, review of the EMR including operative, cardiac intensive care, and discharge summary notes was performed to collect perioperative data and identify postoperative complications. Perioperative variables included intensive care unit length of stay in days, (ICU LOS), use of insulin infusion in ICU, number of ICU days with insulin infusion, mean and maximum glucose on insulin infusion, ICU readmission, operating room and total tube time, and perioperative PAP use. Postoperative complications included prolonged intubation, respiratory failure, reintubation, hypoxemia, tracheostomy, myocardial infarction, arrhythmia, encephalopathy, infection (sternal mediastinitis), death, and 30-day re-admission. Postoperative hypoxemia was considered present if the patient developed postoperative respiratory failure, had oxygen desaturations < 90% and > 4% reduction from last recorded value, or if confirmed by arterial blood gas postoperatively. Postoperative respiratory failure was defined by the need for prolonged mechanical ventilation (> 24 h), endotracheal reintubation, or tracheostomy. Postoperative MI was defined as appearance of new Q waves > 0.04s wide and 1 mV in depth accompanied by elevated levels of troponin T (0.03 ng/mL) and creatine kinase-MB (> 100 IU/L). Subjects were contacted by phone to ascertain adverse effects related to the study and assess re-admission to hospitals other than the study sites. Letters and sleep study reports were sent to participants to share with their primary care providers.

Analytical Plan

Univariable analysis was performed to compare perioperative variables and 30-day readmission in subjects with and without SA, by t-test or Wilcoxon rank sum test for continuous variables, and χ2 test or Fisher exact test for categorical factors. Logistic regression with different AHI cut-points (AHI ≥ 5, ≥ 10, ≥ 15, and ≥ 30) was performed with adjustment of age, gender, race (Caucasian vs. not Caucasian), BMI, and LVEF. Chi-square tests or Fisher exact tests were used to compare prevalence of postoperative complications in subjects with and without SA adjusting for LVEF. Primary results are shown for AHI ≥ 15 representing moderate-to-severe SA. Additional analyses were performed based on the results of the univariable analysis. Chi-square tests or Fisher exact tests were used to compare prevalence of perioperative and postoperative outcomes among groups defined by AHI and LVEF median cutoff and between groups with LVEF < 45 and LVEF ≥ 45. The probability of any complications, any respiratory events, any cardiovascular events, death, and all complications combined to represent a composite endpoint post hoc were compared between AHI and LVEF groups. Comparisons in percentage of total sleep time with oxygen saturation under 90% (%TST with SpO2 < 90%) and any TST with SpO2 < 90% by t-tests were performed between AHI and LVEF groups and those with or without defined perioperative and postoperative outcomes. Spearman correlations were evaluated between %TST with SpO2 < 90% and the defined outcomes as well as AHI and LVEF. Data are presented as mean ± standard deviation or median [25th, 75th percentiles] for continuous variables and N (%) for categorical variables.

Initial power calculations assumed that SA would occur in 25% of the recruited population. Under this assumption, and assuming a baseline morbidity rate of no more than 9% among patients without SA, there would be ≥ 80% power to detect an increase in morbidity to 19% in the SA group with 400 total subjects. Due to significant unexpected hurdles regarding the development of the wireless system including data and video transfer from the patient's hospital room to the sleep laboratories through the hospitals' IT infrastructure, recruitment was delayed and terminated at 107 subjects. With this sample size, and under similar assumptions, there would be 80% power to detect increases in the morbidity of more than 29% in the SA group. Power calculations assumed use of a significance level of 0.05.

RESULTS

Of 107 participants, 79 (73.8%) had an AHI ≥ 5, 63 (58.9%) had an AHI ≥ 10, 51(47.7%) had an AHI ≥ 15, and 29 (27.1) had an AHI ≥ 30. Sample characteristics are shown in Table 1. Subjects with AHI ≥ 15 had significantly lower LVEF and higher prevalence of HF. Results were similar using an AHI ≥ 10 with the exception that the difference between groups in LVEF was no longer significant (p = 0.068) and the difference in age no longer trended toward significant (p = 0.12). Using an AHI ≥ 5, only age was different between groups with subjects with AHI < 5 being younger than those with AHI ≥ 5 (63.2 ± 15.7 vs. 68.9 ± 12.0, p = 0.046). Notably, 9 subjects did not have surgery due to change in medical status and/or patient decision. The demographic characteristics of these subjects were not significantly different from the rest of the cohort.

Demographic and medical comorbidities based on AHI ≥ 15.

jcsm.11.10.1083.t01.jpg

table icon
Table 1

Demographic and medical comorbidities based on AHI ≥ 15.

(more ...)

Polysomnographic data are shown in Table 2. Overall, sleep efficiency was lower and wake time after sleep onset higher than typically observed in the sleep laboratory. However, all sleep stages were represented and respiratory events were predominantly obstructive in type. While the PSG software did not have the capability for periodic breathing auto-detection, Cheyne-Stokes breathing pattern was not noted in any study participants.

Overall summary of polysomnographic data.

jcsm.11.10.1083.t02.jpg

table icon
Table 2

Overall summary of polysomnographic data.

(more ...)

Perioperative data and postoperative outcomes are shown in Tables 3, S1 and S2 (supplemental material). No subjects were treated with PAP therapy perioperatively. No significant differences were observed between subjects with and without SA using AHI cutoffs of 5, 10, or 15. However, while nonsignificant, subjects with AHI ≥ 15 had longer tube times and ICU LOS and were more likely to experience all postoperative complications other than mediastinitis and reintubation compared to those with AHI < 15. Logistic regression analyses with SA cutoffs of 5, 10, 15, and 30, adjusting for age, gender, race, BMI, and LVEF found no significant increase in odds for any perioperative or postoperative outcomes, although the percentage of subjects with any complication increased with AHI.

Perioperative and postoperative outcomes based on AHI ≥ 15.

jcsm.11.10.1083.t03.jpg

table icon
Table 3

Perioperative and postoperative outcomes based on AHI ≥ 15.

(more ...)

Perioperative and postoperative outcomes by LVEF group (< 45 vs. ≥ 45%) are shown in Table 4. Patients with LVEF < 45% had a higher ICU readmission rate than those with LVEF ≥ 45%. Although not significant, patients with lower LVEF had longer tube time and ICU LOS than those with LVEF ≥ 45%. Similarly, the percentage of all postoperative complications except arrhythmia and tracheotomy were greater, although not statistically so, in the lower LVEF group.

Perioperative and postoperative outcomes LVEF < 45 vs. LVEF ≥ 45.

jcsm.11.10.1083.t04.jpg

table icon
Table 4

Perioperative and postoperative outcomes LVEF

(more ...)

Figure 1 illustrates selected perioperative and postoperative outcomes in groups split by median of AHI and LVEF (AHI < 15/LVEF ≥ 45 vs. AHI ≥ 15/LVEF < 45). Ten postoperative complications were considered, including myocardial infarction and arrhythmia as cardiovascular complications, and prolonged intubation, respiratory failure, reintubation, hypoxemia, and tracheostomy as respiratory complications, together representing a composite complication endpoint. While not statistically different, ICU LOS, 30-day readmission, and total tube time were greater in the group with higher AHI and lower LVEF.

Comparisons of complications between groups split by median AHI and LVEF.

Shown are selected complications between groups split by median AHI and LVEF. Ten postoperative complications were considered, including myocardial infarction and arrhythmia as cardiovascular complications, and prolonged intubation, respiratory failure, reintubation, hypoxemia, and tracheostomy as respiratory complications. Total complications represent a composite complication score. Count numbers present as median and quartiles; 30−day readmission rate presents the percentage of readmission after 30 days, and the 95% confidence interval.

jcsm.11.10.1083a.jpg

jcsm.11.10.1083a.jpg
Figure 1

Comparisons of complications between groups split by median AHI and LVEF. Shown are selected complications between groups split by median AHI and LVEF. Ten postoperative complications were considered, including myocardial infarction and arrhythmia as cardiovascular complications, and prolonged intubation, respiratory failure, reintubation, hypoxemia, and tracheostomy as respiratory complications. Total complications represent...

(more ...)

No significant differences were found in %TST with SpO2 < 90% between AHI or LVEF groups, and those with or without complications. However, %TST with SpO2 < 90% was numerically greater in patients with AHI ≥ 5 and ≥ 10, ICU re-admission, readmission at 30 days, prolonged intubation, respiratory failure, hypoxemia, tracheostomy, and encephalopathy. For continuous outcome variables, only OR tube time was positively correlated (p = 0.029) with %TST with SpO2 < 90%. Patients with any amount of TST with SpO2 < 90% had greater BMI, longer OR tube time, and greater prevalence of prolonged intubation (p = 0.007, 0.035, 0.038, respectively). No other difference existed between groups.

DISCUSSION

We found a high prevalence of SA among patients undergoing cardiac surgery, including 73.8% of subjects meeting standard criteria (AHI ≥ 5) and 47.7% having moderate-to-severe disease that was virtually entirely obstructive in nature. While we excluded subjects with OSA treated with PAP or oral appliance therapy within 30 days of recruitment, we did not exclude subjects previously diagnosed but not currently treated as complete medical records were not available for all patients. This may have contributed to the higher than expected observed prevalence. While many perioperative and postoperative complications were numerically greater in patients with more severe SA, we could not find significant differences in complications individually or when considered as a composite endpoint post hoc between patients with and without SA despite exploring multiple cutpoints for AHI. It is very possible that sample size limited our ability to detect significant differences. Importantly, no subjects diagnosed with sleep apnea were treated with PAP therapy perioperatively, so our findings are not confounded by treatment effects. While the study investigators did not specifically explore reasons why this was the case, surgical teams generally were not in the practice of managing sleep apnea perioperatively at this time, and some raised concerns about the effect of treatment on surgical site integrity.

While no differences were found in traditional OSA predictors including age, gender, and BMI between patients with and without SA, those with lower LVEF and a diagnosis of HF were more likely to have undiagnosed moderate-to-severe SA, and lower LVEF was associated with higher ICU readmission rate and maximum perioperative glucose levels. Somewhat surprising is the virtual absence of central apneas and periodic breathing on PSG among our study participants. However, since the majority of events were classified as hypopneas and study investigators did not differentiate obstructive from central hypopnea events, it is possible that come patients had a component of central sleep apnea.

Possible reasons that can be cited for the lack of effect of unrecognized and untreated SA on perioperative and postoperative outcomes after cardiac surgery in our sample include universal postoperative monitoring. Studies among patients with OSA undergoing non-cardiac surgery have consistently shown higher rates of unanticipated transfer to ICU which can mainly be attributed to poor or no monitoring of high risk cases.9,11,12 Continuous monitoring using pulse oximetry with nurse notification of violation limits via wireless pager in postoperative patients found a reduction in ICU transfers of 25% to 50% and rapid response team activations of 45% to 60% on surgical floors, resulting in a reduction of ICU days by 135 in a 1-year period.13 Additionally, a recent study showed that reintubation in patients with OSA occurred earlier in the postoperative course and had a better prognosis than in patients without OSA.9 This suggests that many occurrences of respiratory failure in OSA patients may be related to transient respiratory compromise which can be managed successfully with early detection. Unlike most patients undergoing routine non-cardiac surgery, elective cardiac surgery patients are routinely monitored postoperatively in an ICU setting, as was the case in our sample, rendering it possible they fared better as a result of universal postoperative monitoring. Also since our sample was recruited from the hospital setting (i.e., patients were in the hospital awaiting cardiac surgery), it is possible that medical comorbidities may have been optimized preoperatively, contributing to a lack of significant differences in complications between subjects with and without SA. Lastly, it appears intuitive that a reperfused heart would be less susceptible to hypoxemia, sympathetic activation, and hemodynamic changes associated with OSA than prior to cardiac surgery.

The primary limitation of our study is our small sample size. Most patients undergoing elective cardiac surgery are admitted to the hospital on the day of surgery. However, our study was designed for the inpatient setting given that one of our aims was to finalize upgrade of the PSG recording device's wireless data transmission feature. Therefore, only patients admitted to the hospital awaiting cardiac surgery were eligible for enrollment, which led to recruitment challenges despite having secured full support of the cardiothoracic surgery services at both participating institutions. Due to the short time between the completion of study procedures and surgery, none of our subjects were introduced to PAP therapy preoperatively. While not optimal clinically, this had the effect of eliminating the potential confounder of treatment effect in our analysis. Additional study limitations include reliance on the EMR for the identification of complications that are not always explicitly stated, such as the presence of encephalopathy in the postoperative setting, short duration of follow-up after discharge, and determination of 30-day outcomes through a telephone encounter with the patient or family and EMR review without access to records from hospitals outside the healthcare systems of the participating sites.

Despite our inability to find an association between OSA and postoperative complications likely due to sample size limitations, associations between OSA and a variety of adverse health outcomes, including hypertension, cardiovascular disease, stroke, glucose intolerance, depression, and motor vehicle or occupational accidents have been increasingly reported.1418 More severe degrees of OSA have also been consistently associated with an increased risk of mortality as demonstrated in several large-scale epidemiologic cohorts.1921 In turn, treatment with PAP therapy improves blood pressure and metabolic and cardiovascular outcomes.14,2224 Larger samples are required to explore the impact of OSA on key postoperative outcomes that have clinical and economic importance in the care of cardiovascular surgery populations.

DISCLOSURE STATEMENT

This study was supported by the National Institute of Neurological Disorders and Stroke (NINDS) R44 NS042451-04 awarded to Hani Kayyali, Cleveland Medical Devices Inc. Dr. Foldvary-Schaefer has received research equipment from Cleveland Medical Devices Inc., research support from RESMED and UCB Inc. and has served on the Jazz Pharma speakers' bureau. Dr. Gillinov has consulted for Edwards Lifesciences, Atricure, Abbott Vascular, Tendyne, and On-X; is on the speakers' bureau for Medtronic, Edwards Lifesciences and Atricure; and receives grant support from St. Jude Medical. Hani Kayyali and Dr. Tarler receive salary from and has stock in Cleveland Medical Devices, Inc. Dr. Collop receives royalties from UpToDate. The other authors have indicated no financial conflicts of interest. The work was performed at Cleveland Clinic, Cleveland, OH and Johns Hopkins Medical Center, Baltimore, MD.

ABBREVIATIONS

AHI

apnea-hypopnea index

BMI

body mass index

EKG

electrocardiogram

EOG

electro-oculogram

GSAQ

Global Sleep Assessment Questionnaire

HF

heart failure

LVEF

left ventricle ejection fraction

OAS

obstructive sleep apnea

PCI

percutaneous coronary intervention

PSG

polysomnography

SA

sleep apnea

SA/SDQ

Sleep Apnea Scale of the Sleep Disorders Questionnaire

TST

total sleep time

REFERENCES

1 

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]

2 

Lee CH, Khoo SM, Tai BC, et al., authors. Obstructive sleep apnea in patients admitted for acute myocardial infarction. prevalence, predictors, and effect on microvascular perfusion. Chest. 2009;135:1488–95. [PubMed]

3 

Yumino D, Tsurumi Y, Takagi A, Suzuki K, Kasanuki H, authors. Impact of obstructive sleep apnea on clinical and angiographic outcomes following percutaneous coronary intervention in patients with acute coronary syndrome. Am J Cardiol. 2007;99:26–30. [PubMed]

4 

Cassar A, Morgenthaler TI, Lennon RJ, Rihal CS, Lerman A, authors. Treatment of obstructive sleep apnea is associated with decreased cardiac death after percutaneous coronary intervention. J Am Coll Cardiol. 2007;50:1310–4. [PubMed]

5 

Steiner S, Schueller PO, Hennersdorf MG, Behrendt D, Strauer BE, authors. Impact of obstructive sleep apnea on the occurrence of restenosis after elective percutaneous coronary intervention in ischemic heart disease. Respir Res. 2008;9:50. [PubMed Central][PubMed]

6 

Kaw R, Golish J, Ghamande S, Burgess R, Foldvary N, Walker E, authors. Incremental risk of obstructive sleep apnea on cardiac surgical outcomes. J Cardiovasc Surg (Torino). 2006;47:683–9.

7 

Sharma S, Daggubatti R, Tribble RW, Petit SJ, Gross K, authors. Prevalence of obstructive sleep apnea in patients undergoing coronary artery bypass graft (CABG) surgery: a pilot study. J Sleep Disord Treat Care. 2012;1:2.

8 

Mutter TC, Chateau D, Moffatt M, Ramsey C, Roos LL, Kryger M, authors. A matched cohort study of postoperative outcomes in obstructive sleep apnea: could preoperative diagnosis and treatment prevent complications? Anesthesiology. 2014;121:707–18. [PubMed]

9 

Mokhlesi B, Hovda MD, Vekhter B, Arora VM, Chung F, Meltzer DO, authors. Sleep-disordered breathing and postoperative outcomes after elective surgery: analysis of the nationwide inpatient sample. Chest. 2013;144:903–14. [PubMed Central][PubMed]

10 

Iber C, Ancoli-Israel S, Chesson A, Quan S, authors. The AASM manual for the scoring of sleep and associated events: rules, terminology and technical specifications. Westchester, IL: American Academy of Sleep Medicine, 2007.

11 

Memtsoudis S, Liu SS, Ma Y, et al., authors. Perioperative pulmonary outcomes in patients with sleep apnea after noncardiac surgery. Anesth Analg. 2011;112:113–21. [PubMed]

12 

Kaw R, Chung F, Pasupuleti V, Mehta J, Gay PC, Hernandez AV, authors. Meta-analysis of the association between obstructive sleep apnoea and postoperative outcome. Br J Anaesth. 2012;109:897–906. [PubMed]

13 

Taenzer AH, Pyke JB, McGrath SP, Blike GT, authors. Impact of pulse oximetry surveillance on rescue events and intensive care unit transfers: a before-and-after concurrence study. Anesthesiology. 2010;112:282–7. [PubMed]

14 

Botros N, Concato J, Mohsenin V, Selim B, Doctor K, Yaggi HK, authors. Obstructive sleep apnea as a risk factor for type 2 diabetes. Am J Med. 2009;122:1122–7. [PubMed Central][PubMed]

15 

Tregear S, Reston J, Schoelles K, Phillips B, authors. Obstructive sleep apnea and risk of motor vehicle crash: systematic review and meta-analysis. J Clin Sleep Med. 2009;5:573–81. [PubMed Central][PubMed]

16 

Dyken ME, Im KB, authors. Obstructive sleep apnea and stroke. Chest. 2009;136:1668–77. [PubMed]

17 

Nieto FJ, Young TB, Lind BK, et al., authors. Association of sleep-disordered breathing, sleep apnea, and hypertension in a large community-based study. sleep heart health study. JAMA. 2000;283:1829–36. [PubMed]

18 

Rodenstein D, author. Sleep apnea: traffic and occupational accidents--individual risks, socioeconomic and legal implications. Respiration. 2009;78:241–8. [PubMed]

19 

Punjabi NM, Caffo BS, Goodwin JL, et al., authors. Sleep-disordered breathing and mortality: a prospective cohort study. PLoS Med. 2009;6:e1000132. [PubMed Central][PubMed]

20 

Young T, Finn L, Peppard PE, et al., authors. Sleep disordered breathing and mortality: eighteen-year follow-up of the wisconsin sleep cohort. Sleep. 2008;31:1071–8. [PubMed Central][PubMed]

21 

Marshall NS, Wong KK, Liu PY, Cullen SR, Knuiman MW, Grunstein RR, authors. Sleep apnea as an independent risk factor for all-cause mortality: the Busselton Health Study. Sleep. 2008;31:1079–85. [PubMed Central][PubMed]

22 

McDaid C, Duree KH, Griffin SC, et al., authors. A systematic review of continuous positive airway pressure for obstructive sleep apnoea-hypopnoea syndrome. Sleep Med Rev. 2009;13:427–36. [PubMed]

23 

Keles T, Durmaz T, Bayram NA, et al., authors. Effect of continuous positive airway pressure therapy on aortic stiffness in patients with obstructive sleep apnea syndrome. Echocardiography. 2009;26:1217–24. [PubMed]

24 

Milleron O, Pilliere R, Foucher A, et al., authors. Benefits of obstructive sleep apnoea treatment in coronary artery disease: a long-term follow-up study. Eur Heart J. 2004;25:728–34. [PubMed]

SUPPLEMENTAL MATERIAL

Perioperative and Postoperative Outcomes AHI < 5 vs. AHI ≥ 15.

jcsm.11.10.1083.t0S1.jpg

table icon
Table S1

Perioperative and Postoperative Outcomes AHI

(more ...)

Perioperative and Postoperative Outcomes AHI < 5 vs. AHI ≥ 30.

jcsm.11.10.1083.t0S2.jpg

table icon
Table S2

Perioperative and Postoperative Outcomes AHI

(more ...)