Perioperative Pulmonary Outcomes in Patients with Sleep Apnea After Noncardiac Surgery : Anesthesia & Analgesia

Secondary Logo

Journal Logo

Patient Safety: Research Reports

Perioperative Pulmonary Outcomes in Patients with Sleep Apnea After Noncardiac Surgery

Memtsoudis, Stavros MD, PhD*; Liu, Spencer S. MD*; Ma, Yan PhD; Chiu, Ya Lin MS; Walz, J. Matthias MD; Gaber-Baylis, Licia K. BA; Mazumdar, Madhu PhD

Author Information
Anesthesia & Analgesia 112(1):p 113-121, January 2011. | DOI: 10.1213/ANE.0b013e3182009abf

Although patients with sleep apnea (SA) are considered at increased risk for perioperative complications and guidelines for safe management of individuals with the disease have been published,1,2 evidence supporting increased risk of perioperative morbidity among this patient population is limited. Conclusions are largely derived from studies with limited sample sizes.39 In addition, published data are primarily reflective of the experience of single, academic medical centers, thus limiting external validity. No studies have addressed surgical outcomes in patients with SA on a national and more representative level.

Complications affecting the pulmonary system are of particular concern in patients with SA.7 This is especially problematic as a number of researchers have shown the negative impact that perioperative factors have on respiratory physiologic variables,1016 exposing patients with SA at additional risk for respiratory or pulmonary complications. Knowledge of the incidence and risk for the development of pulmonary complications among SA patients is important to accurately inform patients and their physicians and to target efforts toward preventing them.

Therefore, we sought to study perioperative demographics and outcomes of patients with SA after orthopedic and general surgical abdominal procedures, using data collected for the National Inpatient Sample (NIS), the largest all-payer database in the United States (US). We hypothesized that SA is an independent risk factor for perioperative pulmonary complications. If correct, results would provide a basis for an increase in the utilization of resources, including intensive monitoring and development of strategies to prevent and treat these events.


Data Source

NIS data files sponsored by the Agency for Healthcare Research and Quality (AHRQ) were commercially obtained from the Hospital Cost and Utilization Project and analyzed for this study. The NIS is the largest all-payer inpatient discharge database in the US. Detailed information on the NIS design can be accessed electronically.a,b Recognizing its utility to answer valuable clinical questions, many studies addressing various aspects across the spectrum of medical specialties have been published. The AHRQ performs validation procedures of the NIS against other national databases including the National Hospital Discharge Survey and MedPAR, thus validating estimates provided in this database.17 Because the data used in this study are sufficiently deidentified, this project was exempt from review by the IRB.

Selection of Study Sample and Statistical Methods

Our study sample consists of all data in NIS for each year between 1998 and 2007. Entries indicating the performance of a lower extremity joint arthroplasty or an open abdominal surgical procedure (hereafter referred to as orthopedic and general surgical procedures) were identified using the Clinical Classifications Software for Services and Procedures (CCSSP)c provided by the AHRQ and included in the sample. Discharges with International Classification of Diseases, ninth revision, Clinical Modification (ICD-9-CM) diagnosis codes 327.2 and 780.5 were used to define SA in this study as described previously.18,19 ICD-9 code 327.2 was used in the years 2005 to 2007, whereas all other codes were used in all years of study.

Patient characteristics were reported by SA status and procedure type (Table 1). Patient demographics included age (continuous and categorized as 0–44, 45–64, 65–74, and >75 years), comorbidity burden (according to Deyo comorbidity index categories), gender, race (White, Black, Hispanic, Other), and admission type (emergent, elective, urgent, and others). Obesity, a common comorbidity in SA patients, was added because (1) it is not one of the comorbidities accounted for in the Deyo index, and (2) we sought to analyze its independent impact on outcomes. Two-year time periods (1998–1999, 2000–2001, 2002–2003, 2004–2005, and 2006–2007) were considered in all analyses to account for any temporal changes in practice or coding. Aspiration pneumonia, respiratory insufficiency after trauma or surgery/adult respiratory distress syndrome (ARDS), pulmonary embolism (PE), and the need for respiratory intubation and mechanical ventilation were chosen as primary outcomes and defined by respective ICD-9-CM codes and the CCSSP (see Appendix Table 1 for specific codes used).

Table 1:
Demographic Data on Discharges of Orthopedic and General Surgical Procedures for Patients With and Without Sleep Apnea: Full Sample

Weighted means and percentages (±SE) were shown for continuous and categorical variables, respectively. Approximately 40% of entries in the race category were not available and were inputted as “white.” This step was based on an approach previously described and the fact that facilities with high rates of missing data for race served populations with higher than average white/black patient ratios.20,21 Of note, in this analysis, results did not change significantly when treating missing entries as a separate group. Additionally, to account for overall comorbidity burden, Deyo comorbidity index categories were created.22,23 In brief, the Deyo index is a measure of overall comorbidity burden based on the presence of a number of comorbidities that are each given individual scores depending on their propensity to be associated with adverse outcomes. Comorbidities considered in the Deyo index are listed in Appendix Table 2.

A matched sample was created using the propensity score method based on a multivariable logistic regression with outcome variable of SA (yes/no). Covariates in the model included age, gender, race, admission type, obesity, Deyo comorbidity index categories, and ordinal time periods (1998–1999, 2000–2001, 2002–2003, 2004–2005, and 2006–2007). Given the relatively low incidence of SA (2.51% in orthopedic and 1.49% in general surgical procedures, respectively) and the large number of patients without SA, we decided to match 1 patient with SA (case) with 3 non-SA patients (controls) because it has been shown that the improvement in statistical efficiency for using multiple controls is not considerable beyond 3 or 4 controls per case.24 The matched pairs were generated by comparing the predicted propensity scores between cases and controls using the SAS macro25 with 8→1 digit match without replacement. Absolute standardized percent difference, defined as the difference in proportion in units of the pooled SD, was computed for assessing imbalance in the covariates between SA and non-SA patients.26 An absolute difference >10% is used to imply significant covariate imbalance. Patient demographic data items stratified by patients with and without SA and by procedure type (orthopedic and general surgical) were then described for the matched sample (Table 2). Reduction in mean absolute standardized percent difference from the full sample to the matched sample was examined to assess the effectiveness of the matching procedure. The incidence of each outcome (aspiration pneumonia, ARDS, PE, and the need for intubation and mechanical ventilation) for the full and matched sample was described using bar graphs (Fig. 1). Based on the matched sample, the effect of SA on outcomes was tested for significance using the Cochran-Mantel-Haenszel test.27,28 Corresponding overall common odds ratio (OR) and 95% confidence intervals were estimated. It has been recommended that researchers report both a relative and an absolute measure and present these with appropriate confidence intervals.29 Therefore, in addition to OR, absolute risk reduction (ARR), defined as the absolute difference between the incidence of each complication among SA patients (experimental event rate) and non-SA patients (control event rate), is also presented (Table 3).

Table 2:
Demographic Data on Discharges of Orthopedic and General Surgical Procedures for Patients With Sleep Apnea Matched to Those Without Sleep Apnea Based on Propensity Scoring: Matched Sample
Figure 1:
Incidence of respiratory complications for patients with and without sleep apnea (SA) undergoing orthopedic (A) and general surgical (B) procedures for patients utilizing the full and matched sample. ARDS = adult respiratory distress syndrome; PE = pulmonary embolism; AE = adverse event.
Table 3:
Odds Ratio and 95% Confidence Interval, Control Event Rate in the Non–Sleep Apnea Study Subjects, Experimental Event Rate in the Sleep Apnea Study Subjects, and Absolute Risk Reduction and 95% Confidence Intervals for the Matched Sample

Additionally, logistic regression models with adjustment for the same variables used for matching procedure were fitted on the full sample to check for the sensitivity of the results to different statistical approaches.30 The split-sample approach was used for internal validation.31 Multicollinearity was assessed before fitting logistic regression models. The area under the receiver operating characteristic curve (also referred to as the c statistic or concordance index) was used for assessing the model's discriminatory power.

All statistical analyses were performed using SAS version 9.2 (SAS Institute, Cary, NC). To facilitate analysis of data and to obtain consistent estimates of mean and variance parameters considering the complex survey data setting, SAS procedures SURVEYMEANS, SURVEYFREQ, and SURVEYLOGISTIC were used for descriptive analyses and modeling efforts.d


Sample Characteristics

We identified 3,441,262 entries for general surgical and 2,610,441 for orthopedic procedures performed between 1998 and 2007 in the NIS. These numbers are a national estimate of 16,828,312 and 12,760,565 performed procedures during this time frame in the US. Of those, 51,509 (weighted estimate 250,388) (1.4% ± 0.01%) and 65,774 (weighted estimate 320,157) (2.52% ± 0.01%), respectively, carried a diagnosis of SA. The coding for SA increased steadily over time reaching a prevalence of 2.7% ± 0.03% (46,942 of 1,710,001) for general surgical and 5.5% ± 0.04% (82,491 of 1,513,137) for orthopedic procedures in 2007 (Fig. 2). Table 1 details the demographics of entries with and without SA for discharges associated with either procedure type. For orthopedic procedures, patients with SA tended to be slightly younger and more frequently male, whereas the opposite trend was found for general surgery recipients. For either procedure, however, males had a higher prevalence of SA compared with the general population of patients undergoing surgery. SA patients carried a higher comorbidity burden than the group without the disease for either procedure type. Obesity was approximately 5 times more prevalent among SA patients compared with those without the disease (4.3% ± 0.01% vs 23.8% ± 0.19% for general surgical procedures and 6% ± 0.02% vs 30.1% ± 0.18% for orthopedic procedures, respectively [P < 0.0001]).

Figure 2:
Prevalence of a diagnosis code for sleep apnea among patients undergoing orthopedic and general surgical procedures from 1998 to 2007.

Propensity Score Matching

Table 2 details information after the SA patients were matched to non-SA patients. Of the orthopedic patients, 81.4% (58,538 of 65,774), and of general surgical patients, 88.4% (45,547 of 51,509) with SA were successfully matched. Because of matching, the mean absolute standardized percentage difference for the covariates decreased from 21.46% (range, 0.14%–55.72%) to 0.77% (range, 0.01%–3.23%) for the orthopedic sample and from 18.81% (range, 0.16%–58.47%) to 0.33% (range, 0.001%–2.35%) for the general surgical sample, respectively, demonstrating that matching was effective.

Pulmonary Complications

Patients with SA developed pulmonary complications more frequently than their matched controls after both orthopedic and general surgical procedures, respectively (i.e., aspiration pneumonia: ARR of 0.34% [95% confidence interval: 0.30%, 0.39%] and 0.74% [0.66%, 0.81%]; ARDS: ARR of 0.61% [0.57%, 0.65%] and 1.35% [1.26%, 1.44%]; intubation/mechanical ventilation: ARR of 3.19% [3.12%, 3.27%] and 4.87% [4.73%, 5.01%]; Table 3). Comparatively, PE was more frequent in SA patients after orthopedic procedures (ARR of 0.09% [0.06%, 0.12%]) but somewhat reduced after general surgical procedures (ARR of 0.05% [0.01%, 0.08%]). SA was associated with a significantly higher adjusted OR of developing pulmonary complications after both orthopedic and general surgical procedures, respectively, with the exception of PE (OR for aspiration pneumonia: 1.41 [1.35, 1.47] and 1.37 [1.33, 1.41]; for ARDS: 2.39 [2.28, 2.51] and 1.58 [1.54, 1.62]; for PE: OR 1.22 [1.15, 1.29] and 0.90 [0.84, 0.97]; for intubation/mechanical ventilation: 5.20 [5.05, 5.37] and 1.95 [1.91, 1.98]; Table 3).

Sensitivity Analysis

Sensitivity analysis using a regression model (results not shown) yielded similar ORs thus confirming our results and conclusions. Furthermore, the outcomes did not change when the Deyo index, the overall indicator for comorbidity, was replaced by its individual comorbidity components.


In this analysis of a large nationally representative sample, we were able to identify SA as an independent risk factor for perioperative pulmonary complications. Most significantly, SA increased the OR for the need of perioperative tracheal intubation and mechanical ventilation by 5-fold after orthopedic surgery and doubled the odds after general surgical procedures. These findings are important because they provide evidence that patients with SA are indeed at increased risk of perioperative complications, thus supporting efforts targeted to more intensively monitor this population and develop strategies to prevent adverse events. Furthermore, the incidence of these adverse events is relevant insofar as they are frequent enough to be associated with significant demand for resources.

The true incidence of SA among patients undergoing surgery continues to be debated and reports vary widely depending on the methodology used. Fidan et al.32 tested patients who were scheduled for surgical procedures and had risk factors for SA and found that approximately one-third of those at risk actually had SA. They concluded that the prevalence of SA among the general surgical population was 3.2%. Using the Berlin questionnaire, Chung et al.33 identified 24% of patients (73 of 305) as being at high risk of SA (95% confidence interval, 19%–29%). Thirteen patients were confirmed to have obstructive SA, resulting in a frequency of 4.2%; 9 patients had a history of obstructive SA, and 5 patients were identified by polysomnography. The prevalence of SA in our study population was 1.5% in open abdominal procedures and 2.52% in those undergoing orthopedic surgery. The relatively low proportion found in our study may be attributable to the relatively recently increased awareness of SA among the general physician population.

Indeed, the prevalence of SA in our study in 1998 versus 2007 was 0.4% and 2.7% for general surgical procedures and 0.4% and 5.5% for orthopedic procedures, respectively. In addition and in contrast to the methodology used by studies that actively screen for the disease,32,33 entries in the NIS likely only include diagnosed cases of SA.

Regardless of methodology used, it is unlikely that our main objective of this study, namely, the identification of SA as a risk factor for perioperative pulmonary complications, is compromised by the limitations of our dataset and the reported prevalence of the disease. It must be mentioned, however, that the relatively low prevalence found may be a marker of the increasing but still insufficient attention SA as a disease receives in the medical community. Furthermore, the increasing prevalence of obesity among the US population has to be considered as a contributing factor.34

We identified SA to be an independent risk factor for perioperative pulmonary complications and need for mechanical ventilation. The negative impact of perioperative factors on pulmonary and pharyngeal physiology has been well described in the literature and may have especially detrimental effects in patients with SA.1012 Valipour et al.35 determined that 74% of patients with SA had acid reflux–related symptoms. In this context, it has been suggested that patients with SA are at increased risk of aspiration even during normal sleep.36 Disturbances of the pharyngeal musculature brought on by pharmacologic and mechanical factors such as neuromuscular blocking drugs and the use of endotracheal tubes intraoperatively may be aggravating contributors to the increased risk for aspiration pneumonia among patients with SA found in our study.

The higher incidence of aspiration events and ARDS estimated in this study may partially explain the increased need for postoperative intubation/ventilation among SA patients in our analysis. The increased use of intensive care units by SA patients after surgical procedures was noted by Chung et al.7 Although no specific mention for the need of intubation and mechanical ventilation was made by the authors, transfer of patients to a higher level of care may be a surrogate marker for the increased need for respiratory interventions, including airway instrumentation, in SA patients. An additional factor leading to the increased odds for postoperative intubation and ventilation may be the increased prevalence of abnormal oropharyngeal anatomy that may lead clinicians to err on the side of caution when deciding to extubate patients' tracheas immediately after surgery. However, it must be mentioned that contrary to common perception, the link between difficult intubation and SA has been disputed.37 Nevertheless, caution is advised because in one study, two-thirds of patients with difficult airways were found to have SA.38

Patients with SA were found to be at increased risk for PE after orthopedic but not abdominal surgical procedures in our study. Although PE is a known complication in patients undergoing orthopedic surgery,39 this analysis establishes SA as an independent risk factor for such an outcome. Although no mechanism can be established from our data, it is possible that right ventricular dysfunction, which is more prevalent in patients with SA,40 may promote venostasis and thus PE. The observation that intraoperative embolization of cement and bone debris during orthopedic surgery may aggravate right heart dysfunction secondary to increases in pulmonary arterial pressures41 may serve as an explanation for why arthroplasty patients but not abdominal surgical patients with SA had increased risk of PE in our analysis. In addition, there is mounting evidence that SA is associated with endothelial dysfunction, resulting in an increase in platelet aggregation and induction of a hypercoagulable state.42,43 Specifically, levels of coagulation factors XIIa, thrombin-antithrombin complex, and plasma fibrinogen activity are increased.43 Although biologically plausible, these proposed mechanisms remain speculative at this point and further studies are necessary to demonstrate a causal relationship between SA and an increased risk of perioperative PE.

We found an increase in the OR for ARDS among patients with SA in our study. Although not fully explaining a causal relationship, it is of interest that proinflammatory changes have been described in patients with SA. There are several reports in the literature demonstrating a link between SA and increased levels of C-reactive protein, leukocyte superoxide, and soluble adhesion molecules.4446 In this context, there is experimental evidence in models of repetitive hypoxemia/ regeneration that mimic SA suggesting an upregulation of the proinflammatory transcription factor NF-κB (nuclear factor κ-light chain-enhancer of activated B cells) and consequently several proinflammatory genes.47,48 In combination with an increased risk of aspiration in patients with SA, these data might help explain how untreated SA could be linked to an increased perioperative risk of ARDS as observed in our analysis, but more research is needed to prove the link and establish cause and effect.

Our study is limited by a number of factors inherent to secondary analysis of large administrative databases. As such, clinical information (i.e., intraoperative blood loss, length of procedure, etc.) available in the NIS is limited, and our analysis must be interpreted in this context. Because of the nature of the NIS, only inpatient data are available and thus complications and events after discharge are not captured. Furthermore, readmissions cannot be discerned from this database. Thus, conclusions should be limited to the acute perioperative setting with the notion that complications are likely underestimated. Although it cannot be excluded that data entry may be subject to some form of coding or reporting bias, there is no reason to believe that reporting should differ among patients within the database, thus exposing both SA and non-SA discharges to the same bias within the same data collection construct. Comparative analysis should therefore be less likely affected by such bias. Although the ICD-9 codes used in this study have been used to identify SA in previous investigations analyzing large administrative databases,18,19 it has to be mentioned that code 780.5 includes various forms of insomnia and therefore may be relatively nonspecific. With this realization, we conducted a separate analysis of cases defined by code 327.2 and found no significant difference in the main outcomes in our study.

A further limitation is the lack of information about severity of SA with the use of ICD-9 codes. It is reasonable to hypothesize that more severe cases of SA are associated with increased risk of adverse outcomes.7,8

In conclusion, using a large nationally representative database, we identified SA as a risk factor for a number of pulmonary complications after orthopedic and general surgical procedures, thus providing a basis for an increase in the utilization of resources, including intensive monitoring and development of strategies to prevent and treat these events.

Findings from this analysis may be used to design research projects to elucidate the mechanism and design interventions to reduce the risk of pulmonary complications among patients with SA undergoing surgery.

a HCUP Databases. Healthcare Cost and Utilization Project (HCUP). Agency for Healthcare Research and Quality, Rockville, MD, 2008. Available at: Accessed March 15, 2010.
Cited Here

b Introduction to the HCUP National Inpatient Sample (NIS) 2006. Agency for Healthcare Research and Quality, Healthcare Cost and Utilization Project (HCUP). Rockville, MD, 2008. Available at: Accessed March 15, 2010.
Cited Here

c HCUP CCS-Services and Procedures. Healthcare Cost and Utilization Project (HCUP). June 2009. Agency for Healthcare Research and Quality, Rockville, MD. Available at: Accessed March 15, 2010.
Cited Here

d Anthony B. SUGI 27 Statistics and Data Analysis Paper 258-27 Performing Logistic Regression on Survey Data with the New SURVEYLOGISTIC Procedure. Available at: Accessed March 15, 2010.
Cited Here


1. Meoli AL, Rosen CL, Kristo D, Kohrman M, Gooneratne N, Aguillard RN, Fayle R, Troell R, Kramer R, Casey KR, Coleman J Jr; Clinical Practice Review Committee; American Academy of Sleep Medicine. Upper airway management of the adult patient with obstructive sleep apnea in the perioperative period: avoiding complications. Sleep 2003;26:1060–5
2. Gross JB, Bachenberg KL, Benumof JL, Caplan RA, Connis RT, Coté CJ, Nickinovich DG, Prachand V, Ward DS, Weaver EM, Ydens L, Yu S; American Society of Anesthesiologists Task Force on Perioperative Management. Practice guidelines for the perioperative management of patients with obstructive sleep apnea: a report by the American Society of Anesthesiologists Task Force on Perioperative Management of patients with obstructive sleep apnea. Anesthesiology 2006;104:1081–93
3. Gupta RM, Parvizi J, Hanssen AD, Gay PC. Postoperative complications in patients with obstructive sleep apnea syndrome undergoing hip or knee replacement: a case-control study. Mayo Clin Proc 2001;76:897–905
4. Liao P, Yegneswaran B, Vairavanathan S, Zilberman P, Chung F. Postoperative complications in patients with obstructive sleep apnea: a retrospective matched cohort study. Can J Anaesth 2009;56:819–28
5. Kaw R, Golish J, Ghamande S, Burgess R, Foldvary N, Walker E. Incremental risk of obstructive sleep apnea on cardiac surgical outcomes. J Cardiovasc Surg 2006;47:683–9
6. Sabers C, Plevak DJ, Schroeder DR, Warner DO. The diagnosis of obstructive sleep apnea as a risk factor for unanticipated admissions in outpatient surgery. Anesth Analg 2003;96:1328–35
7. Chung F, Yegneswaran B, Liao P, Chung SA, Vairavanathan S, Islam S, Khajehdehi A, Shapiro CM. Validation of the Berlin questionnaire and American Society of Anesthesiologists checklist as screening tools for obstructive sleep apnea in surgical patients. Anesthesiology 2008;108:822–30
8. Hwang D, Shakir N, Limann B, Sison C, Kalra S, Shulman L, Souza Ade C, Greenberg H. Association of sleep-disordered breathing with postoperative complications. Chest 2008; 133:1128–34
9. Ahmad S, Nagle A, McCarthy RJ, Fitzgerald PC, Sullivan JT, Prystowsky J. Postoperative hypoxemia in morbidly obese patients with and without obstructive sleep apnea undergoing laparoscopic bariatric surgery. Anesth Analg 2008;107:138–43
10. Keamy MF III, Cadieux RJ, Kofke WA, Kales A. The occurrence of obstructive sleep apnea in a recovery room patient. Anesthesiology 1987;66:232–4
11. Lamarche Y, Martin R, Reiher J, Blaise G. The sleep apnoea syndrome and epidural morphine. Can Anaesth Soc J 1986;33:231–3
12. Samuels SI, Rabinov W. Difficulty reversing drug-induced coma in a patient with sleep apnea. Anesth Analg 1986;65:1222–4
13. Cullen DJ. Obstructive sleep apnea and postoperative analgesia: a potentially dangerous combination. J Clin Anesth 2001;13:83–5
14. Catley DM, Thornton C, Jordan C, Lehane JR, Royston D, Jones JG. Pronounced, episodic oxygen desaturation in the postoperative period: its association with ventilatory pattern and analgesic regimen. Anesthesiology 1985;63:20–8
15. Reeder MK, Goldman MD, Loh L, Muir AD, Casey KR, Lehane JR. Late postoperative nocturnal dips in oxygen saturation in patients undergoing major abdominal vascular surgery: predictive value of pre-operative overnight pulse oximetry. Anaesthesia 1992;47:110–5
16. Rosenberg J, Wildschiodtz G, Pedersen MH, von Jessen F, Kehlet H. Late postoperative nocturnal episodic hypoxaemia and associated sleep pattern. Br J Anaesth 1994;72:145–50
17. Whalen D, Houchens R, Elixhauser A. 2005 HCUP Nationwide Inpatient Sample (NIS) Comparison Report. HCUP Method Series Report #2008-01. US Agency for Healthcare Research and Quality, 2008. Available at: Accessed May15, 2010
18. Spurr KF, Graven MA, Gilbert RW. Prevalence of unspecified sleep apnea and the use of continuous positive airway pressure in hospitalized patients, 2004 National Hospital Discharge Survey. Sleep Breath 2008;12:229–34
19. Resnick HE, Phillips B. Documentation of sleep apnea in nursing homes: United States 2004. J Am Med Dir Assoc 2008;9:260–4
20. Memtsoudis SG, Ma Y, González Della Valle A, Mazumdar M, Gaber-Baylis LK, MacKenzie CR, Sculco TP. Perioperative outcomes after unilateral and bilateral total knee arthroplasty. Anesthesiology 2009;111:1206–16
21. Kozak LJ. Underreporting of race in the National Hospital Discharge Survey. Adv Data 1995;265:1–12
22. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 1987;40:373–83
23. Deyo RA, Cherkin DC, Ciol MA. Adapting a clinical comorbidity index for use with ICD-9-CM administrative databases. J Clin Epidemiol 1992;45:613–9
24. Ury HK. Efficiency of case-control studies with multiple controls per case: continuous or dichotomous data. Biometrics 1975;31:643–9
25. Lori SP. SUGI 29 Posters Paper 165-29. Performing a 1:N Case-Control Match on Propensity Score. Available at:–29.pdf. Accessed May 15, 2010
26. Flury BK, Riedwyl H. Standard distance in univariate and multivariate analysis. Am Stat 1986;40:249–51
27. Cochran WG. The comparison of percentages in matched samples. Biometrika 1950;37:256–66
28. Mantel N, Haenszel W. Statistical aspects of the analysis of data from retrospective studies of disease. J Natl Cancer Inst 1959;22:719–48
29. Schechtman E. Odds ratio, relative risk, absolute risk reduction, and the number needed to treat: which of these should we use? Value Health 2002;5:431–6
30. Hosmer DW, Lemeshow S. Applied Logistic Regression. 2nd ed. New York: Wiley, 2000
31. Steyerberg EW, Harrell FE Jr, Borsboom GJ, Eijkemans MJ, Vergouwe Y, Habbema JD. Internal validation of predictive models: efficiency of some procedures for logistic regression analysis. J Clin Epidemiol 2001;54:774–81
32. Fidan H, Fidan F, Unlu M, Ela Y, Ibis A, Tetik L. Prevalence of sleep apnoea in patients undergoing operation. Sleep Breath 2006;10:161–5
33. Chung F, Ward B, Ho J, Yuan H, Kayumov L, Shapiro C. Pre-operative identification of sleep apnea risk in elective surgical patients, using the Berlin questionnaire. J Clin Anesth 2007;19:130–4
34. Flegal KM, Carroll MD, Ogden CL, Curtin LR. Prevalence and trends in obesity among US adults, 1999–2008. JAMA 2010;303:235–41
35. Valipour A, Makker HK, Hardy R, Emegbo S, Toma T, Spiro SG. Symptomatic gastroesophageal reflux in subjects with a breathing sleep disorder. Chest 2002;121:1748–53
36. Beal M, Chesson A, Garcia T, Caldito G, Stucker F, Nathan CO. A pilot study of quantitative aspiration in patients with symptoms of obstructive sleep apnea: comparison to a historic control group. Laryngoscope 2004;114:965–8
37. Neligan PJ, Porter S, Max B, Malhotra G, Greenblatt EP, Ochroch EA. Obstructive sleep apnea is not a risk factor for difficult intubation in morbidly obese patients. Anesth Analg 2009;109:1182–6
38. Chung F, Yegneswaran B, Herrera F, Shenderey A, Shapiro CM. Patients with difficult intubation may need referral to sleep clinics. Anesth Analg 2008;107:915–20
39. Memtsoudis SG, Besculides MC, Gaber L, Liu S, González Della Valle A. Risk factors for pulmonary embolism after hip and knee arthroplasty: a population-based study. Int Orthop 2009;33:1739–45
40. Tugcu A, Yildirimtürk O, Tayyareci Y, Demiroglu C, Aytekin S. Evaluation of subclinical right ventricular dysfunction in obstructive sleep apnea patients using velocity vector imaging. Circ J 2010;74:312–9
41. Urban MK, Sheppard R, Gordon MA, Urquhart BL. Right ventricular function during revision total hip arthroplasty. Anesth Analg 1996;82:1225–9
42. Bokinsky G, Miller M, Ault K, Husband P, Mitchell J. Spontaneous platelet activation and aggregation during obstructive sleep apnea and its response to therapy with nasal continuous positive airway pressure: a preliminary investigation. Chest 1995;108:625–30
43. Robinson GV, Pepperell JC, Segal HC, Davies RJ, Stradling JR. Circulating cardiovascular risk factors in obstructive sleep apnoea: data from randomized controlled trials. Thorax 2004;59:777–82
44. Schulz R, Mahmoudi S, Hattar K, Sibelius U, Olschewski H, Mayer K, Seeger W, Grimminger F. Enhanced release of superoxide from polymorphonuclear neutrophils in obstructive sleep apnea. Am J Respir Crit Care Med 2000;162:566–70
45. Shamsuzzaman AS, Winnicki M, Lanfranchi P, Wolk R, Kara T, Accurso V, Somers VK. Elevated C-reactive protein in patients with obstructive sleep apnea. Circulation 2002;105:2462–4
46. Dyngovskaya L, Lavie P, Lavie L. Increased adhesion molecules expression and production of reactive oxygen species in leukocytes of sleep apnea patients. Am J Respir Crit Care Med 2001;165:859–60
47. Greenberg H, Ye X, Wilson D, Htoo AK, Hendersen T, Liu SF. Chronic intermittent hypoxia activates nuclear factor-kappaB in cardiovascular tissues in vivo. Biochem Biophys Res Commun 2006;343:591–6
48. Ryan S, Taylor CT, McNicholas WT. Selective activation of inflammatory pathways by intermittent hypoxia in obstructive sleep apnea syndrome. Circulation 2005;112:2660–8


Appendix Table 1:
Clinical Classifications Software for Services and Procedures (CCSSP) and International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM) Codes Used in This Study
Appendix Table 2:
Comorbidities Accounted for in the Deyo Comorbidity Index


Name: Stavros Memtsoudis, MD, PhD

Contribution: This author helped design the study, conduct the study, analyze the data, write the manuscript, and secured funding.

Attestation: This author has seen the original study data, reviewed the analysis of the data, approved the final manu-script, and is the author responsible for archiving the study files.

Name: Spencer S. Liu, MD

Contribution: This author helped design the study and write the manuscript.

Attestation: This author has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.

Name: Yan Ma, PhD

Contribution: This author helped design and conduct the study, and analyze the data.

Attestation: This author has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.

Name: Ya Lin Chiu, MS

Contribution: This author helped conduct the study and analyze the data.

Attestation: This author has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.

Name: J. Matthias Walz, MD

Contribution: This author helped design the study and write the manuscript.

Attestation: This author has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.

Name: Licia K. Gaber-Baylis, BA

Contribution: This author helped design the study, analyze the data, and write the manuscript.

Attestation: This author has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.

Name: Madhu Mazumdar, PhD

Contribution: This author helped design the study, analyze the data, write the manuscript, and secured funding.

Attestation: This author has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.


Spencer S. Liu is Section Editor of Pain Medicine for the Journal. This manuscript was handled by Sorin J. Brull, Section Editor of Patient Safety, and Dr. Liu was not involved in any way with the editorial process or decision.

© 2011 International Anesthesia Research Society