- Email: [email protected]
Sleep Disordered Breathing as a Risk of Cardiac Events in Subjects With Diabetes Mellitus and Normal Exercise Echocardiographic Findings
Sleep Disordered Breathing as a Risk of Cardiac Events in Subjects With Diabetes Mellitus and Normal Exercise Echocardiographic Findings Sinziana Seicean, MD, MPH, PhDa,b, Kingman P. Strohl, MDb,c, Andreea Seicean, MPHb, Conrad Gibby, BSb, and Thomas H. Marwick, MD, PhD, MPHa,d,* Sleep disordered breathing (SDB) is associated with type 2 diabetes mellitus (T2DM) and cardiovascular disease; however, the contribution of SDB to incident heart failure (HF), coronary artery disease (CAD), and atrial fibrillation (AF) in patients with T2DM is unknown. We followed up 834 consecutive asymptomatic patients with T2DM (age 56 – 11 years, 369 women) with normal exercise echocardiographic findings for £8 years using electronic health records. The demographics, cardiac risk factors, symptoms, diagnoses, and medications were collected at the echocardiography and validated from the electronic health records. SDB was confirmed by a comprehensive sleep evaluation and/or polysomnography before echocardiography. SDB was diagnosed in 188 patients (21%) at baseline; 116 were untreated. During a median follow-up of 4.9 years (interquartile range 3.9 to 6.1), 22 congestive HF, 72 CAD, and 40 AF incident events were observed. In the Cox proportional hazards models, SDB was associated with incident CAD (hazard ratio 1.8, 95% confidence interval 1.1 to 3.0, p [ 0.01; adjusted hazard ratio 1.9, 95% confidence interval 1.2 to 3.2, p <0.01) and AF (hazard ratio 2.6, 95% confidence interval 1.4 to 4.7, p [ 0.01; adjusted hazard ratio 2.9, 95% confidence interval 1.5 to 5.9, p <0.01). Limiting SDB to only those patients diagnosed using polysomnography (n [ 132), SDB was associated with incident CAD (hazard ratio 1.9, 95% confidence interval 1.1 to 3.3, p [ 0.03; adjusted hazard ratio 2.2, 95% confidence interval 1.2 to 3.9, p [ 0.01) and HF (hazard ratio 2.7, 95% confidence interval 1.1 to 7.0, p [ 0.03; adjusted hazard ratio 3.5, 95% confidence interval 1.4 to 9.0, p <0.01). Female gender, age, elevated blood pressure, and left ventricular mass were additional correlates of CAD in those with asymptomatic T2DM. In conclusion, the association of SDB with incident CAD, AF, and HF in patients with T2DM justifies more liberal screening for SDB in patients with T2DM, realizing that SDB is a potentially modifiable risk factor. Ó 2013 Elsevier Inc. All rights reserved. (Am J Cardiol 2013;111:1214e1220) The prevalence of patients with type 2 diabetes mellitus (T2DM) has reached epidemic proportions in the United States, involving 8.3% of the general population and 26.9% of patients aged 65 years. Adults with T2DM have a 2 to 4 times greater risk of heart disease-related mortality compared to their nondiabetic counterparts.1 Sleep disordered breathing (SDB) has been associated with T2DM and impaired glucose tolerance, independent of other comorbidities, including obesity.2,3 Although a number of studies have assessed the association of SDB with cardiovascular events,4e13 none has focused on the effect of SDB
a Heart and Vascular Institute, Cleveland Clinic, Cleveland, Ohio; bCase Western Reserve University, Cleveland, Ohio; cLouis Stokes Veterans Affairs Medical Center, Cleveland, Ohio; and dMenzies Research Institute of Tasmania, Hobart, Australia. Manuscript received September 20, 2012; revised manuscript received and accepted December 23, 2012. This work was partially supported through grants HL007913 and HS00059-14 from the National Institutes of Health (Bethesda, Maryland). See page 1220 for disclosure information. *Corresponding author: Tel: (þ61) 36-226-7703; fax: (þ61) 36-2267704. E-mail address: [email protected] (T.H. Marwick).
0002-9149/13/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.amjcard.2012.12.053
on incident cardiovascular events in patients with asymptomatic T2DM without a history of cardiac disease. The aim of the present study was to define the role of SDB in incident heart failure (HF), coronary artery disease (CAD), and atrial fibrillation (AF) in patients with T2DM without these diagnoses and without evidence of subclinical CAD at baseline. Methods The present study was a longitudinal observational study of adult patients with T2DM without a known cardiovascular history, who had been referred for stress echocardiography to rule out CAD, in the absence of cardiac symptoms. The most common cause for testing was a cardiac risk evaluation, mainly before noncardiac surgery. The day of the echocardiogram was the entry date, and patients were followed up through October 31, 2011. Baseline data available from the Cleveland Clinic echocardiogram registry were supplemented by abstraction from the electronic medical records and manual verification of the administrative and polysomnography reports. The patient data were de-identified before statistical analyses. The institutional review board of Cleveland Clinic approved the study. www.ajconline.org
Miscellaneous/Cardiac Consequences of Sleep Disordered Breathing
Figure 1. Flow diagram for study inclusion.
Exercise echocardiography was performed in 1,371 patients with T2DM aged 18 years, from January 1 2004 to December 31 2007. The exclusion criteria were a history of CAD, AF or atrial flutter, congestive HF, hypertrophic cardiomyopathy, valvular heart disease, aortic aneurysm, congenital heart disease, moderate or severe valvular heart disease, cardiac surgery, chronic pulmonary disease, neoplasia, human immunodeficiency virus, cirrhosis, kidney/ heart transplantation and/or dialysis, positive exercise echocardiographic findings, ejection fraction of <50%, and a diagnosis of congestive HF, CAD, and/or AF within 30 days of the echocardiogram (Figure 1). In addition, patients with a follow-up period of <60 days, those residing outside of Ohio, and patients with <1 comprehensive sleep evaluation within a 1-year period before the echocardiogram were excluded. Our final sample consisted of 834 patients. The baseline variables consisted of age, gender, race/ ethnicity, socioeconomic characteristics, cardiac risk factors, co-morbidities, current medications, laboratory values, weight, height, body mass index, heart rate, ejection fraction, and systolic and diastolic blood pressure. The socioeconomic characteristics were marital status, postal code, and county of residence; the latter 2 were used to estimate a “neighborhood” socioeconomic unit of analysis.14 The cardiac risk factors included age (men 50 years, women 60 years), hypertension, smoking, and a family history of CAD or low-density lipoprotein level of 100 mg/dl.15 The exercise echocardiograms were performed in the usual fashion.16 The Duke treadmill score was calculated using the symptom, electrocardiographic, and exercise responses to stress testing. Regional wall motion was assessed at rest and during stress. Only patients with normal left ventricular function at rest, no deterioration induced by stress, and a normal response in all segments were included in the present study. SDB was considered present at baseline if identified before or within a maximum of 30 days after the baseline echocardiogram. Using the electronic medical record, we identified patients with SDB if the diagnosis was made using polysomnography in the presence of >5 apnea episodes, hypopnea episodes, or respiratory event-related arousals per hour in a symptomatic patient and >15/hour in an asymptomatic patient17 or if a comprehensive sleep
1215
evaluation identified the patient to be at high risk of SDB but the patient did not undergo polysomnography. Although polysomnography is the reference standard for the diagnosis of sleep disorders,18 assuming that patients without polysomnogram-confirmed SDB were negative for the disease would have underestimated SDB in our study population owing to patient nonadherence with testing. A subanalysis was conducted limiting patients diagnosed with SDB to those with abnormal polysomnographic findings. Patients were considered to not have SDB if they were found to be at low or no risk of having SDB using the comprehensive sleep evaluation or the sleep study findings were negative within 1 year of the echocardiogram. Treatment of SDB was assessed for all patients, and adherence to treatment was measured by self-report. Patients who were prescribed continuous positive airway pressure or bilevel positive airway pressure but who reported therapy discontinuation, an inability to use the therapy, or poor or inadequate use were categorized as “untreated.” Because of the lack of objective data on the use of continuous positive airway pressure and bilevel positive airway pressure, no analyses were performed to compare patients with SDB who were treated versus untreated. Patients were followed up for 7.8 years, and incident cardiac events were abstracted using the “International Classification of Diseases, 9th revision,” diagnostic codes. The patients were censored at HF or death. Mortality was determined using the Social Security Death Index obtained from the Ohio Department of Health; 13 patients were lost because of noncardiac-related death during follow-up, and death was not identified as an event. The baseline differences were assessed according to SDB status using a chi-square test for categorical variables and a 2-sample t test for continuous variables. Differences between the Kaplan-Meier survival curves were compared using the log-rank test. Nested Cox proportional hazards models were used to assess the effect of SDB on the outcomes of interest, independent of other risk factors. Standard statistical software (SAS, version 9.2, SAS Institute, Cary, North Carolina) was used to perform the analyses, and p <0.05 was the cutoff for statistical significance. Results Our sample consisted of 834 patients with T2DM, without clinical evidence of CAD and negative baseline exercise echocardiographic findings, with a median followup of 4.9 years (interquartile range 3.9 to 6.1). The mean age was 56.2 11.3 years, 19.5% of the patients were black, and 7.9% resided in areas with high poverty rates (Table 1). The risk factors for cardiac complications were highly prevalent, and 79% had blood pressure levels greater than the 2011 American Diabetes Association recommended values.15 SDB was identified in 188 patients at baseline, 70% of whom were diagnosed using polysomnography, and 38% of whom used continuous positive airway pressure. As expected, patients with SDB had a significantly greater body mass index than patients without SDB (Table 1). During the follow-up period, 22 patients experienced incident congestive HF and 72 CAD, with patients with SDB having twice the event rate of those without SDB
1216
The American Journal of Cardiology (www.ajconline.org)
Table 1 Baseline characteristics stratified by sleep disordered breathing (SDB) Total (n ¼ 834)
Baseline* Characteristics
Women Black race High poverty postal code Single Body mass index 30 kg/m2 Body mass index 25 kg/m2 Hypercholesterolemia Hypertension Cerebrovascular disease Depression or anxiety Age (yrs) Age 75 yrs Family history of coronary artery disease Ever smoked Coronary artery disease risk factorsz Suboptimal blood pressure controlx Suboptimal glycemic control{ Creatinine 1.5 mg/dl Low-density lipoprotein 100 mg/dl Hemoglobin A1c (%) Glucose (mg/dl) Creatinine (mg/dl) Low-density lipoprotein (mg/dl) High-density lipoprotein (mg/dl) Triglycerides (mg/dl) Body mass index (kg/m2) Body surface area (m2) Maximum heart rate (beats/min) Systolic blood pressure (mm Hg) Diastolic blood pressure (mm Hg) Ejection fraction (%) Left ventricular mass (g) Diastolic dysfunction stage 1 2 3 End-diastolic volume (ml) Indexed end-diastolic volume (ml) Median left atrial area (cm2) Abnormal Duke score Insulin b Blockers Angiotensin-converting enzyme inhibitors Diuretics Aspirin Statins Nonsteroidal anti-inflammatory drug Continuous positive airway pressure or bilevel positive airway pressure at baseline
SDB
p Value
Yes (n ¼ 188)
No (n ¼ 646)
369 (44.2%) 163 (19.5%) 66 (7.9%) 122 (14.6%) 475 (57.0%) 731 (87.7%) 584 (70.0%) 508 (60.9%) 37 (4.4%) 201 (24.1%) 56.2 11.3 48 (5.8%) 317 (45.1%) 376 (45.1%) 409 (49%) 658 (78.9%) 396 (47.5%) 20 (2.4%) 387 (46.4%) 7.4 1.7 144.2 63.4 0.88 0.39 104.2 34.3 52.0 14.7 166.4 113.8 32.1 6.8 2.0 0.3 154.8 15.7 134.0 18.1 81.6 9.8 63.0 7.8 177.6 56.5
95 (50.5%) 38 (20.2%) 8 (4.3%) 29 (15.4%) 148 (78.7%) 182 (96.8%) 137 (72.9%) 131 (69.7%) 6 (3.2%) 57 (30.3%) 54.6 10.1 5 (2.7%) 83 (44.2%) 90 (47.9%) 92 (48.9%) 153 (81.4%) 99 (52.7%) 4 (2.1%) 92 (48.9%) 7.4 1.5 143.4 52.9 0.94 0.69 105.9 33.7 49.9 12.4 185.1 125.9 36.0 7.3 2.1 0.2 156 14 135 18 83 10 62 8 189.2 55.4
274 (42.4%) 125 (19.4%) 58 (9.0%) 93 (14.4%) 327 (50.6%) 549 (85.0%) 447 (69.2%) 377 (58.4%) 31 (4.8%) 144 (22.3%) 56.7 11.6 43 (6.7%) 234 (36.2%) 286 (44.3%) 317 (49.1%) 505 (78.2%) 297 (46.0%) 16 (2.5%) 295 (45.7%) 7.4 1.8 144.4 66.3 0.86 0.24 103.7 34.5 52.6 15.3 160.8 109.5 30.9 6.2 2.0 0.3 154 16 134 18 81 10 63 8 177.9 56.4
362 (43.4%) 35 (4.2%) 5 (0.6%) 85.8 25.0 42.3 10.7 16.2 4.0 103 (12.4%) 152 (18.2%) 133 (16.0%) 311 (37.3%) 285 (34.2%) 315 (37.8%) 465 (55.8%) 208 (24.9%) 72 (8.6%)
74 (39.4%) 4 (2.1%) 1 (0.5%) 90.9 26.6 42.4 11.4 16.7 3.9 19 (10%) 42 (22.3%) 33 (17.0%) 80 (42.6%) 79 (42.0%) 68 (36.2%) 108 (57.5%) 63 (33.5%) 72 (38.3%)
288 (46.6%) 31 (4.8%) 4 (0.6%) 84.3 24.3 42.3 10.4 16.0 4.0 84 (13%) 110 (17.0%) 101 (15.6%) 231 (35.8%) 206 (31.9 %) 247 (38.2%) 357 (55.3%) 145 (22.5%) NA
0.05† 0.08 0.03† 0.73 <0.0001† <0.0001† 0.43 0.005† 0.35 0.02† 0.024† 0.04† 0.14 0.38 0.97 0.34 0.12 NA 0.43 0.97 0.84 0.014† 0.45 0.031† 0.012† <0.0001† <0.0001† 0.24 0.19 0.06 0.04† 0.005† 0.16
<0.001† 0.93 0.05† 0.29 0.10 0.65 0.09 0.01† 0.61 0.60 0.002† NA
Data are presented as n (%) or mean SD. NA ¼ not available. * At stress echocardiography. † Statistically significant. z Defined as men aged 50 years or women aged 60 years, with any 1 of hypertension, smoking, family history of CAD, low-density lipoprotein 100 mg/dl. x Diastolic blood pressure control at rest of 80 mm Hg and/or systolic blood pressure control at rest of 130 mm Hg. { Postprandial glucose 180 mg/dl or hemoglobin A1c 7.
(Table 2). Patients with SDB had nearly 3 times the rate of AF compared to patients without SDB. After censoring for new congestive HF and/or death, the age-adjusted hazard
ratios showed a strong association between SDB and both incident CAD (p <0.001) and AF (p ¼ 0.002; Table 3). A marginal association was found between suspected or
Miscellaneous/Cardiac Consequences of Sleep Disordered Breathing
1217
Table 2 Incident cardiac events stratified by sleep disordered breathing (SDB) Cardiac Events
Heart failure Coronary artery disease Atrial fibrillation
Total (n ¼ 834)
22 (2.6%) 72 (8.6%) 40 (4.8%)
SDB Yes (n ¼ 188)
No (n ¼ 646)
8 (4.3%) 25 (13.3%) 17 (9.0%)
14 (2.2%) 47 (7.3%) 23 (3.6%)
Relation of SDB to Incident Cardiac Events*
p Value†
1.9 (0.8e4.6) 1.8 (1.1e3.0) 2.6 (1.4e4.7)
0.14 0.01 <0.01
Cardiac incident events included acute myocardial infarction, acute coronary syndrome, and new onset of chronic ischemic heart disease. * Data are presented as unadjusted hazard ratio (95% confidence interval). † p Value for overall effect of SDB modeled as a categorical variable.
proven SDB and incident HF (p ¼ 0.07). Elevated highdensity lipoprotein was found to be protective against CAD (p ¼ 0.03). Female gender, cerebrovascular disease, poor glycemic control (elevated hemoglobin A1c and postprandial glucose), elevated systolic blood pressure, and increased left ventricular mass were associated with incident CAD in the present cohort. The body mass index, systolic blood pressure at rest, elevated end-diastolic volume, increased left atrial area, and the use of b blockers and diuretics were also significant age-adjusted associations of AF. SDB was a significant correlate of CAD and AF in both unadjusted and adjusted models (Table 4). Age was associated with incident HF and CAD and was marginally associated with AF, and the systolic blood pressure at rest correlated with all incident cardiac events. In a subanalysis of patients diagnosed with SDB using polysomnography, SDB was independently associated with incident HF and CAD, but not with AF (Table 5). Discussion The present study has linked a greater risk of new onset cardiac disease in those with T2DM and with SDB than in those without SDB. SDB was associated with a greater risk of incident CAD after controlling for previously identified risk factors, including age, gender, hypertension, smoking, lipid profile, family history of cerebrovascular disease, medical treatment, standard of care-related risk factors, or overweight/obesity. Whether considering the SDB group as those with a high index of clinical suspicion and those diagnosed using polysomnography or limiting that group to only those patients diagnosed using polysomnography, associations were identified with AF and HF. Studies have suggested an association between T2DM and SDB, independent of obesity.2,3 Although T2DM and SDB have been individually linked to an increased risk of cardiac disorders, the present study is the first to show an association of SDB with an increased risk of incident cardiac events in patients with T2DM. The baseline prevalence of SDB in our study was similar to that recently reported by a large retrospective, population-based, multicenter study.3 It is important to note that although 188 patients were diagnosed with SDB at baseline, an additional 25 were diagnosed with SDB during the follow-up period. A number of studies in other patient populations have identified a relation between SDB and CAD,4e7 including the Sleep Heart Health Study (SHHS),7 which reported an
increased risk of incident CAD in 40- to 70-year-old men. Our work differs from previous studies by offering greater female representation, sample size, separation of end points, and exploration of the differences between SDB and nonSDB with respect to CAD risk factors, healthcare access and quality, and treatment differences for existent comorbidities. We also found that SDB was associated with more than twice the risk of incident AF in patients with T2DM, Our findings are consistent with studies that found arrhythmias to be more prevalent in patients with SDB than in patients without SBD,8,9 including another SHHS study, reporting a prevalence of 4.8% and 0.9%, respectively.10 As expected, the rates of AF were much greater in our patient population because of the co-morbidity with T2DM. In addition, SDB diagnosed using polysomnography was associated with HF. Several studies have found a high prevalence of SDB in patients with HF, ranging from 10% to 35%.11,12 One previous study done as part of the SHHS found that SDB was associated with incident HF in men, but not in women.7 SDB has been linked to vascular endothelial changes through increased proinflammatory mediators,19,20 and studies have confirmed elevated inflammatory markers in persons with SDB, independent of obesity.21 A number of physiologic mechanisms have been proposed for the association of SDB with cardiac events, 1 of which is hypertension. Although we found elevated systolic blood pressure to increase the risk of all incident HF, CAD, and AF in our final models, the baseline systolic blood pressure of patients with and without SDB was similar, suggesting that hypertension might not be a principal causal mechanism for increased incident cardiac events in patients with asymptomatic T2DM and SDB. Although a number of studies have shown that patients with SDB are more resistant to catheter ablation as treatment of AF,22,23 it is still unclear whether this relates to a different pathophysiology of AF in SDB- compared to noneSDB-associated AF.13 The finding that female gender is associated with incident CAD was unexpected and might be related to the selection of patients with T2DM. Women have been found to have more pronounced vasoconstriction in response to both hypoxia and stress compared to men.24 In addition, women with SDB have been reported to have larger brachial artery diameters than women without SDB, and an increased brachial diameter has been identified as a risk factor for CAD in women.25 There is a strong association between the apnea-hypopnea index and the flow-mediated diameter in female patients with SDB,26 suggesting a link with
1218
The American Journal of Cardiology (www.ajconline.org)
Table 3 Age*-adjusted hazard ratios and 95% confidence intervals for predictors of incident cardiac events Variable
HF HR (95% CI)
Sleep disordered breathing Women Black race High poverty postal code Single Body mass index 30 kg/m2 Hypercholesterolemia Hypertension Cerebrovascular disease Depression or anxiety Family history of coronary artery disease Ever smoke Coronary artery disease risk factorsz Suboptimal blood pressure controlx Suboptimal glucose control{ Creatinine 1.5 mg/dl Low-density lipoprotein 100 mg/dl Hemoglobin A1c Glucose (mg/dl) Creatinine (mg/dl) Low-density lipoprotein (mg/dl) High-density lipoprotein (mg/dl) Triglycerides (mg/dl) Body mass index (kg/m2) Maximum heart rate (beats/min) Systolic blood pressure at rest (mm Hg) Diastolic blood pressure at rest (mm Hg) Ejection fraction Left ventricular mass (g) Diastolic dysfunction (stage) End-diastolic volume (ml) End-diastolic volume/body surface area (ml/m2) Left atrial area (cm2) Abnormal Duke score Insulin b Blockers Angiotensin-converting enzyme inhibitors Diuretics Aspirin Statins Nonsteroidal anti-inflammatory drugs
2.2 1.2 1.3 1.2 0.6 1.7 0.9 1.4 0.9 1.4 0.7 1.5 1.2 1.6 1.7 3.1 0.6 1.2 1.0 1.1 1.0 1.0 1.0 1.0 0.96 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.1 1.9 1.2 1.3 0.9 1.7 0.8 2.2 1.2
(0.9e5.4) (0.5e2.7) (0.5e3.4) (0.3e5.0) (0.1e2.7) (0.7e4.2) (0.6e1.4) (0.5e3.6) (0.1e6.4) (0.6e3.4) (0.3e1.4) (0.6e3.4) (0.4e3.5) (0.5e5.2) (0.7e3.9) (0.7e13.8) (0.3e1.5) (1.0e1.5) (1.0e1.01) (0.4e3.0) (1.0e1.01) (1.0e1.04) (0.9e1.0) (0.9e1.1) (0.9e1.0) (1.0e1.1) (0.9e1.1) (0.9e1.1) (1.0e1.01) (0.4e2.3) (0.99e1.0) (0.98e1.0) (1.0e1.2) (0.7e5.0) (0.4e3.7) (0.5e3.5) (0.4e2.2) (0.7e3.9) (0.3e1.9) (0.8e6.1) (0.5e3.1)
CAD p Value 0.07 0.72 0.66 0.83 0.53 0.24 0.70 0.49 0.88 0.49 0.30 0.37 0.80 0.48 0.25 0.13 0.29 0.11 0.04† 0.87 0.72 0.66 0.77 0.18 0.03† <0.01† 0.22 0.99 0.07 0.93 0.78 0.82 0.11 0.23 0.70 0.66 0.88 0.23 0.63 0.11 0.68
HR (95% CI) 2.6 2.2 1.0 0.7 0.4 1.3 0.9 1.5 2.4 0.6 0.8 1.2 1.9 1.3 1.6 1.0 1.2 1.1 1.0 1.2 1.0 0.98 1.0 1.0 1.0 1.0 1.0 0.98 1.0 1.3 1.0 1.0 1.0 1.4 1.4 1.3 1.1 1.1 1.0 1.2 1.3
(1.5e4.5) (1.4e3.6) (0.6e1.9) (0.3e1.9) (0.1e1.0) (0.8e2.1) (0.6e1.1) (0.9e2.6) (1.1e5.1) (0.3e1.1) (0.6e1.1) (0.8e2.0) (1.0e3.6) (0.7e2.3) (1.0e2.6) (0.2e4.0) (0.7e1.9) (0.98e1.3) (1.0e1.01) (0.9e1.7) (1.0e1.01) (0.96e1.0) (0.99e1.0) (1.0e1.1) (0.98e1.0) (1.0e1.03) (0.98e1.0) (0.95e1.0) (1.0e1.01) (0.8e2.1) (0.99e1.0) (0.98e1.0) (1.0e1.1) (0.7e2.5) (0.8e2.5) (0.7e2.2) (0.6e1.8) (0.7e1.8) (0.6e1.6) (0.7e1.9) (0.7e2.1)
AF p Value <0.001† <0.001† 0.90 0.48 0.06 0.28 0.26 0.13 0.02† 0.09 0.18 0.36 0.05† 0.48 0.04† 0.98 0.48 0.09 <0.001† 0.21 0.24 0.03† 0.17 0.37 0.74 <0.001† 0.92 0.30 <0.001† 0.34 0.16 0.82 0.43 0.34 0.22 0.41 0.70 0.61 0.90 0.51 0.39
HR (95% CI) 2.9 1.0 0.4 0.4 1.3 0.9 1.0 1.0 2.2 0.4 1.0 1.4 0.9 4.0 1.1 1.1 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.1 1.0 1.0 1.0 1.0 1.0 0.7 1.0 1.0 1.1 0.4 0.5 2.3 0.7 2.1 1.8 1.0 1.2
(1.5e5.8) (0.5e1.9) (0.1e1.4) (0.1e2.8) (0.5e3.2) (0.4e1.7) (0.8e1.3) (0.5e1.9) (0.7e7.4) (0.1e1.1) (0.8e1.3) (0.7e2.7) (0.4e2.1) (1.0e16.7) (0.5e2.1) (0.1e8.0) (0.5e1.9) (0.8e1.3) (1.0e1.0) (0.3e2.9) (1.0e1.01) (0.98e1.03) (0.99e1.0) (1.0e1.1) (0.99e1.0) (1.0e1.04) (1.0e1.1) (0.96e1.1) (0.99e1.0) (0.4e1.4) (1.0e1.03) (0.99e1.1) (1.0e1.2) (0.1e1.7) (0.2e1.7) (1.3e5.6) (0.3e1.5) (1.1e4.2) (0.9e3.6) (0.5e2.0) (0.5e2.5)
p Value 0.002† 0.92 0.16 0.34 0.53 0.66 0.76 0.88 0.20 0.07 0.96 0.34 0.74 0.06 0.84 0.94 0.96 0.95 0.69 0.99 0.59 0.69 0.87 0.03† 0.31 0.03† 0.16 0.83 0.66 0.34 0.05† 0.06 0.04† 0.21 0.27 <0.01† 0.36 0.03† 0.08 0.98 0.67
CI ¼ confidence interval; HR ¼ hazard ratio. * Age variable was centered. † Statistically significant. z Defined as men aged 50 years or women aged 60 years, with any 1 of hypertension, smoking, family history of CAD, low-density lipoprotein 100 mg/dl. x Diastolic blood pressure control at rest of 80 mm Hg and/or systolic blood pressure control at rest of 130 mm Hg. { Postprandial glucose 180 mg/dl or hemoglobin A1c 7.0%.
endothelial dysfunction. Another study reported female gender was associated with greater mortality in a sample of patients with SDB treated with continuous positive airway pressure.27 The investigators proposed either that gender might be an independent risk factor for mortality in their sample or that their finding could be related to delayed diagnoses and treatment of SDB in female compared to male patients. Recently, a study limited to female patients with SDB confirmed the association of SDB with an increased risk of cardiovascular mortality in women and found SDB
severity to be associated with a greater risk of mortality.28 However, we found no interaction between gender and SDB in the present study. Compliance with continuous positive airway pressure or bilevel positive airway pressure treatment in our sample was similar to the rates observed in previous studies.29 Although only 38% of patients with SDB were compliant with treatment at baseline, this proportion had increased to 46% through the end of the follow-up period. However, the present study could not assess the effect of continuous
Miscellaneous/Cardiac Consequences of Sleep Disordered Breathing
1219
Table 4 Adjusted models* showing relation of sleep disordered breathing (SDB) to incident cardiac events Model
Predictors
HF HR (95% CI)
1 2
3
4
SDB Agez SDB Agez Female gender SDB Agez Female gender CAD risk factorsx SDB Agez Female gender CAD risk factorsx Systolic blood pressure
2.2 1.1 2.2 1.1 1.1 2.2 1.1 1.1 1.1 2.2 1.1 1.2 0.8 1.03
CAD p Value
(0.9e5.4) (1.0e1.1) (0.9e5.4) (1.0e1.1) (0.5e2.6) (0.9e5.4) (1.0e1.1) (0.5e2.7) (0.3e3.3) (0.9e5.4) (1.0e1.1) (0.5e2.9) (0.2e2.6) (1.0e1.04)
0.07 <0.01† 0.07 <0.01† 0.79 0.08 0.04† 0.81 0.97 0.08 0.04† 0.69 0.65 <0.01†
HR (95% CI) 2.3 1.05 2.0 1.1 2.1 1.9 1.1 2.0 1.3 1.9 1.04 2.1 1.1 1.02
AF p Value
(1.3e3.8) (1.0e1.1) (1.2e3.2) (1.0e1.1) (1.3e3.4) (1.2e3.2) (1.0e1.1) (1.2e3.3) (0.7e2.5) (1.2e3.2) (1.0e1.1) (1.3e3.5) (0.5e2.1) (1.0e1.04)
<0.01† <0.001† <0.01† <0.001† <0.001† <0.01† <0.01† <0.01† 0.45 <0.01† 0.04† <0.01† 0.85 <0.01†
HR (95% CI) 2.9 1.03 2.9 1.04 0.9 3.0 1.04 0.97 0.7 2.9 1.04 1.01 0.7 1.02
(1.5e5.8) (1.0e1.1) (1.5e5.8) (1.0e1.1) (0.4e1.8) (1.5e5.9) (0.99e1.1) (0.5e2.1) (0.3e2.0) (1.5e5.9) (0.99e1.1) (0.5e2.1) (0.3e1.7) (1.0e1.04)
p Value <0.01† 0.06 <0.01† 0.06 0.77 <0.01† 0.08 0.93 0.62 <0.01† 0.08 0.98 0.39 0.02†
Abbreviations as in Table 3. * Each model was adjusted for factors listed with it, with reported HR for each factor taking into account all other factors included in the model. † Statistically significant. z Age variable was centered. x Defined as men aged 50 years or women aged 60 years, with any 1 of hypertension, smoking, family history of CAD, or low-density lipoprotein 100 mg/dl.
Table 5 Models showing relation of sleep disordered breathing (SDB) confirmed by polysomnography (n ¼ 132) to incident cardiac events Predictors
HF HR (95% CI)
Unadjusted Model 1 Model 2
Model 3
Model 4
SDB SDB Age† SDB Age† Female gender SDB Age† Female gender CAD risk factorsz SDB Age† Female gender CAD risk factorsz Systolic blood pressure
2.7 3.7 1.1 3.5 1.1 1.2 3.6 1.1 1.1 1.1 3.5 1.1 1.2 0.8 1.03
(1.1e7.0) (1.4e9.4) (1.0e1.1) (1.4e9.2) (1.0e1.1) (0.5e2.8) (1.4e9.2) (1.0e1.1) (0.5e2.8) (0.3e3.6) (1.4e9.0) (1.0e1.1) (0.5e3.1) (0.2e2.9) (1.0e1.05)
CAD p Value 0.03 <0.01* <0.01* <0.01* <0.01* 0.11 <0.01* 0.02* 0.77 0.91 <0.01* 0.02* 0.64 0.77 <0.01*
HR (95% CI) 1.9 2.3 1.1 2.2 1.1 2.0 2.2 1.1 1.9 1.1 2.2 1.05 1.9 1.0 1.01
(1.1e3.3) (1.3e4.2) (1.0e1.1) (1.2e3.9) (1.0e1.1) (1.2e3.2) (1.2e3.9) (1.0e1.1) (1.1e3.2) (0.6e2.2) (1.2e3.9) (1.0e1.1) (1.1e3.3) (0.5e2.0) (1.0e1.03)
AF p Value 0.03 <0.01* <0.001* <0.01* <0.0001* 0.01* <0.01* <0.001* <0.01* 0.7 0.01* <0.01* 0.01* 0.9 0.06
HR (95% CI) 1.7 1.9 1.04 1.9 1.04 1.2 2.0 1.1 1.4 0.5 1.9 1.06 1.4 0.5 1.02
(0.8e3.9) (0.9e4.6) (1.0e1.1) (0.8e4.5) (1.0e1.1) (0.6e2.4) (0.9e4.9) (1.0e1.1) (0.6e2.9) (0.1e1.3) (0.8e4.7) (0.99e1.1) (0.7e3.0) (0.2e1.1) (1.0e1.04)
p Value 0.20 0.1 0.02 0.13 0.02 0.69 0.1 <0.01 0.40 0.50 0.10* <0.01* 0.40 0.08 0.10
Abbreviations as in Table 3. * Statistically significant. † Age variable was centered. z Defined as men aged 50 years or women aged 60 years, with any 1 of hypertension, smoking, family history of CAD, or low-density lipoprotein 100 mg/dl.
positive airway pressure treatment on cardiac events, owing to the limitation of the self-report of SDB treatment in the electronic health records. Our study had several important limitations. Although polysomnography is the reference standard for the diagnosis of sleep disorders,17 we also chose to consider patients to have SDB if they were identified as having a high risk of SDB18 but did not undergo polysomnography. This allowed us to reduce the bias from noncompliance with referral. Most patients (70%) we defined as having SDB in our study
had polysomnogram-confirmed SDB, and our findings were similar when limiting our sample to these patients (Table 5). This was an observational retrospective study based on the facilitation of clinical care using electronic health records. However, the baseline variables were carefully collected in the exercise echocardiogram registry at the time of the test, and all other clinical data were also entered in the electronic health records prospectively, albeit for patient care rather than for research purposes. Because the purpose of the study was to examine the association of clinically
1220
The American Journal of Cardiology (www.ajconline.org)
manifest SDB (previously diagnosed by a physician) with the prediction of incident cardiac events, exclusions according to co-morbid cardiac disease at echocardiographic testing focused the study, but also resulted in an inability to further subdivide patients with co-morbid disease. The study sample was derived from a single, large center and included only patients residing in Ohio, so that follow-up could be performed; thus, studies in other populations are warranted. Although potentially important, but difficult to verify, we did not gather data on diet and exercise. In addition, apneahypopnea index severity and associations with hypoxic stress were not quantified, because it was outside the scope of our study to assess the relation between SDB severity and incident cardiac events. Data on compliance with continuous positive airway pressure and bilevel positive airway pressure were self-reported. However, it was outside the scope of our study to assess the value of SDB treatment on cardiac outcomes.
13. 14.
15. 16.
17.
Disclosures The authors have no conflicts of interest to disclose. 1. Centers for Disease Control and Prevention. National Diabetes Fact Sheet: National Estimates and General Information on Diabetes and Prediabetes in the United States, 2011. Atlanta: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, 2011: 1e12. Available at: http://www.cdc.gov/diabetes/pubs/ pdf/ndfs_2011.pdf. Accessed December 18, 2012. 2. Seicean S, Kirchner L, Gottlieb DJ, Punjabi N, Resnick H, Sanders M, Budhiraja R, Singer M, Redline S. Sleep-disordered breathing and impaired glucose metabolism in normal-weight and overweight/obese individuals: the Sleep Heart Health Study. Diabetes Care 2008;31:1001e1006. 3. Heffner JE, Rozenfeld Y, Kai M, Stephens EA, Brown LK. Prevalence of diagnosed sleep apnea among patients with T2DM in primary care. Chest 2012;141:1414e1421. 4. Shah NA, Yaggi HK, Concato J, Mohsenin V. Obstructive sleep apnea as a risk factor for coronary events or cardiovascular death. Sleep Breath 2010;14:131e136. 5. Martinez D, Klein C, Rahmeier L, da Silva RP, Fiori CZ, Cassol CM, Gonçalves SC, Bos AJ. Sleep apnea is a stronger predictor for coronary heart disease than traditional risk factors. Sleep Breath 2012;16:695e701. 6. Peker Y, Hedner J, Norum J, Kraiczi H, Carlson J. Increased incidence of cardiovascular disease in middle-aged men with obstructive sleep apnea: a 7-year follow-up. Am J Respir Crit Care Med 2002;166:159e165. 7. Gottlieb DJ, Yenokyan G, Newman AB, O’Connor GT, Punjabi NM, Quan SF, Redline S, Resnick HE, Tong EK, Diener-West M, Shahar E. Prospective study of obstructive sleep apnea and incident coronary heart disease and heart failure: the Sleep Heart Health Study. Circulation 2010;122:352e360. 8. Hoffstein V, Mateika S. Cardiac arrhythmias, snoring, and sleep apnea. Chest 1994;106:466e471. 9. Namtvedt SK, Randby A, Einvik G, Hrubos-Strøm H, Somers VK, Røsjø H, Omland T. Cardiac arrhythmias in obstructive sleep apnea (from the Akershus Sleep Apnea Project). Am J Cardiol 2011;108: 1141e1146. 10. Mehra R, Benjamin EJ, Shahar E, Gottlieb DJ, Nawabit R, Kirchner HL, Sahadevan J, Redline S; Sleep Heart Health Study. Association of nocturnal arrhythmias with sleep-disordered breathing: the Sleep Heart Health Study. Am J Respir Crit Care Med 2006;173:910e916. 11. Sin DD, Fitzgerald F, Parker JD, Newton G, Floras JS, Bradley TD. Risk factors for central and obstructive sleep apnea in 450 men and women with congestive heart failure. Am J Respir Crit Care Med 1999;160:1101e1106. 12. Wang H, Parker JD, Newton GE, Floras JS, Mak S, Chiu KL, Ruttanaumpawan P, Tomlinson G, Bradley TD. Influence of obstructive
18.
19. 20.
21.
22.
23. 24. 25.
26. 27. 28.
29.
sleep apnea on mortality in patients with heart failure. J Am Coll Cardiol 2007;49:1625e1631. Monahan K, Redline S. Role of obstructive sleep apnea in cardiovascular disease. Curr Opin Cardiol 2011;26:541e547. Coulton C, Colabianchi N, Cook T. An Analysis of Differences in Health Status Measures Among Neighborhoods within Metro Counties of Ohio, 2001. Available at: http://www2.odh.ohio.gov/Data/OFHSurv/ ofhsrfq2sup.pdf. Accessed February 12, 2012. American Diabetes Association. Standard of medical care in diabetes2012. Diabetes Care 2012;35:511e563. Douglas PS, Geller U, Khandheria B, Stainback RF, Weissman NJ, Peterson ED, Hendel RC, Stainback RF, Blaivas M, Des Prez RD, Gillam LD, Golash T, Hiratzka LF, Kussmaul WG, Labovitz AJ, Lindenfeld J, Masoudi FA, Mayo PH, Miller TD, Porembka D, Spertus JA, Wann S, Wiegers SE, Brindis RG, Douglas PS, Hendel RC, Patel MR, Peterson ED, Wolk MJ, Allen JM, Lewin JC, Allen JM, Caruth KC, Barrett EA. ACCF/ASE/ACEP/AHA/ASNC/SCAI/SCCT/SCMR 2008 appropriateness criteria for stress echocardiography. J Am Coll Cardiol 2008;51:1127e1147. Epstein LJ, Kristo D, Strollo PJ Jr, Friedman N, Malhotra A, Patil SP, Ramar K, Rogers R, Schwab RJ, Weaver EM, Weinstein MD; Adult Obstructive Sleep Apnea Task Force of the American Academy of Sleep Medicine. Clinical guideline for the evaluation, management and long-term care of obstructive sleep apnea in adults. J Clin Sleep Med 2009;5:263e276. Kushida CA, Littner MR, Morgenthaler T, Alessi CA, Bailey D, Coleman J Jr, Friedman L, Hirshkowitz M, Kapen S, Kramer M, LeeChiong T, Loube DL, Owens J, Pancer JP, Wise M. Practice parameters for the indications for polysomnography and related procedures: an update for 2005. Sleep 2005;28:499e521. Ip MS, Tse HF, Lam B, Tsang KW, Lam WK. Endothelial function in obstructive sleep apnea and response to treatment. Am J Respir Crit Care Med 2004;169:348e353. Jelic S, Padeletti M, Kawut SM, Higgins C, Canfield SM, Onat D, Colombo PC, Basner RC, Factor P, LeJemtel TH. Inflammation, oxidative stress, and repair capacity of the vascular endothelium in obstructive sleep apnea. Circulation 2008;117:2270e2278. Larkin EK, Rosen CL, Kirchner HL, Storfer-Isser A, Emancipator JL, Johnson NL, Zambito AM, Tracy RP, Jenny NS, Redline S. Variation of C-reactive protein levels in adolescents: association with sleepdisordered breathing and sleep duration. Circulation 2005;111: 1978e1984. Hoyer FF, Lickfett LM, Mittmann-Braun E, Ruland C, Kreuz J, Pabst S, Schrickel J, Juergens U, Tasci S, Nickenig G, Skowasch D. High prevalence of obstructive sleep apnea in patients with resistant paroxysmal atrial fibrillation after pulmonary vein isolation. J Interv Card Electrophysiol 2010;29:37e41. Ng CY, Liu T, Shehata M, Stevens S, Chugh SS, Wang X. Metaanalysis of obstructive sleep apnea as predictor of atrial fibrillation recurrence after catheter ablation. Am J Cardiol 2011;108:47e51. Levenson J, Pessana F, Gariepy J, Armentano R, Simon A. Gender differences in wall shear-mediated brachial artery vasoconstriction and vasodilation. J Am Coll Cardiol 2001;38:1668e1674. Holubkov R, Karas RH, Pepine CJ, Rickens CR, Reichek N, Rogers WJ, Sharaf BL, Sopko G, Merz CN, Kelsey SF, McGorray SP, Reis SE. Large brachial artery diameter is associated with angiographic coronary artery disease in women. Am Heart J 2002;143:802e807. Faulx MD, Larkin EK, Hoit BD, Aylor JE, Wright AT, Redline S. Sex influences endothelial function in sleep-disordered breathing. Sleep 2004;27:1113e1120. Morrish E, Shneerson JM, Smith IE. Why does gender influence survival in obstructive sleep apnoea? Respir Med 2008;102: 1231e1236. Campos-Rodriguez F, Martinez-Garcia MA, de la Cruz-Moron I, Almeida-Gonzalez C, Catalan-Serra P, Montserrat JM. Cardiovascular mortality in women with obstructive sleep apnea with or without continuous positive airway pressure treatment: a cohort study. Ann Intern Med 2012;156:115e122. Weaver TE, Grunstein RR. Adherence to continuous positive airway pressure therapy: the challenge to effective treatment. Proc Am Thorac Soc 2008;5:173e178.
a Heart and Vascular Institute, Cleveland Clinic, Cleveland, Ohio; bCase Western Reserve University, Cleveland, Ohio; cLouis Stokes Veterans Affairs Medical Center, Cleveland, Ohio; and dMenzies Research Institute of Tasmania, Hobart, Australia. Manuscript received September 20, 2012; revised manuscript received and accepted December 23, 2012. This work was partially supported through grants HL007913 and HS00059-14 from the National Institutes of Health (Bethesda, Maryland). See page 1220 for disclosure information. *Corresponding author: Tel: (þ61) 36-226-7703; fax: (þ61) 36-2267704. E-mail address: [email protected] (T.H. Marwick).
0002-9149/13/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.amjcard.2012.12.053
on incident cardiovascular events in patients with asymptomatic T2DM without a history of cardiac disease. The aim of the present study was to define the role of SDB in incident heart failure (HF), coronary artery disease (CAD), and atrial fibrillation (AF) in patients with T2DM without these diagnoses and without evidence of subclinical CAD at baseline. Methods The present study was a longitudinal observational study of adult patients with T2DM without a known cardiovascular history, who had been referred for stress echocardiography to rule out CAD, in the absence of cardiac symptoms. The most common cause for testing was a cardiac risk evaluation, mainly before noncardiac surgery. The day of the echocardiogram was the entry date, and patients were followed up through October 31, 2011. Baseline data available from the Cleveland Clinic echocardiogram registry were supplemented by abstraction from the electronic medical records and manual verification of the administrative and polysomnography reports. The patient data were de-identified before statistical analyses. The institutional review board of Cleveland Clinic approved the study. www.ajconline.org
Miscellaneous/Cardiac Consequences of Sleep Disordered Breathing
Figure 1. Flow diagram for study inclusion.
Exercise echocardiography was performed in 1,371 patients with T2DM aged 18 years, from January 1 2004 to December 31 2007. The exclusion criteria were a history of CAD, AF or atrial flutter, congestive HF, hypertrophic cardiomyopathy, valvular heart disease, aortic aneurysm, congenital heart disease, moderate or severe valvular heart disease, cardiac surgery, chronic pulmonary disease, neoplasia, human immunodeficiency virus, cirrhosis, kidney/ heart transplantation and/or dialysis, positive exercise echocardiographic findings, ejection fraction of <50%, and a diagnosis of congestive HF, CAD, and/or AF within 30 days of the echocardiogram (Figure 1). In addition, patients with a follow-up period of <60 days, those residing outside of Ohio, and patients with <1 comprehensive sleep evaluation within a 1-year period before the echocardiogram were excluded. Our final sample consisted of 834 patients. The baseline variables consisted of age, gender, race/ ethnicity, socioeconomic characteristics, cardiac risk factors, co-morbidities, current medications, laboratory values, weight, height, body mass index, heart rate, ejection fraction, and systolic and diastolic blood pressure. The socioeconomic characteristics were marital status, postal code, and county of residence; the latter 2 were used to estimate a “neighborhood” socioeconomic unit of analysis.14 The cardiac risk factors included age (men 50 years, women 60 years), hypertension, smoking, and a family history of CAD or low-density lipoprotein level of 100 mg/dl.15 The exercise echocardiograms were performed in the usual fashion.16 The Duke treadmill score was calculated using the symptom, electrocardiographic, and exercise responses to stress testing. Regional wall motion was assessed at rest and during stress. Only patients with normal left ventricular function at rest, no deterioration induced by stress, and a normal response in all segments were included in the present study. SDB was considered present at baseline if identified before or within a maximum of 30 days after the baseline echocardiogram. Using the electronic medical record, we identified patients with SDB if the diagnosis was made using polysomnography in the presence of >5 apnea episodes, hypopnea episodes, or respiratory event-related arousals per hour in a symptomatic patient and >15/hour in an asymptomatic patient17 or if a comprehensive sleep
1215
evaluation identified the patient to be at high risk of SDB but the patient did not undergo polysomnography. Although polysomnography is the reference standard for the diagnosis of sleep disorders,18 assuming that patients without polysomnogram-confirmed SDB were negative for the disease would have underestimated SDB in our study population owing to patient nonadherence with testing. A subanalysis was conducted limiting patients diagnosed with SDB to those with abnormal polysomnographic findings. Patients were considered to not have SDB if they were found to be at low or no risk of having SDB using the comprehensive sleep evaluation or the sleep study findings were negative within 1 year of the echocardiogram. Treatment of SDB was assessed for all patients, and adherence to treatment was measured by self-report. Patients who were prescribed continuous positive airway pressure or bilevel positive airway pressure but who reported therapy discontinuation, an inability to use the therapy, or poor or inadequate use were categorized as “untreated.” Because of the lack of objective data on the use of continuous positive airway pressure and bilevel positive airway pressure, no analyses were performed to compare patients with SDB who were treated versus untreated. Patients were followed up for 7.8 years, and incident cardiac events were abstracted using the “International Classification of Diseases, 9th revision,” diagnostic codes. The patients were censored at HF or death. Mortality was determined using the Social Security Death Index obtained from the Ohio Department of Health; 13 patients were lost because of noncardiac-related death during follow-up, and death was not identified as an event. The baseline differences were assessed according to SDB status using a chi-square test for categorical variables and a 2-sample t test for continuous variables. Differences between the Kaplan-Meier survival curves were compared using the log-rank test. Nested Cox proportional hazards models were used to assess the effect of SDB on the outcomes of interest, independent of other risk factors. Standard statistical software (SAS, version 9.2, SAS Institute, Cary, North Carolina) was used to perform the analyses, and p <0.05 was the cutoff for statistical significance. Results Our sample consisted of 834 patients with T2DM, without clinical evidence of CAD and negative baseline exercise echocardiographic findings, with a median followup of 4.9 years (interquartile range 3.9 to 6.1). The mean age was 56.2 11.3 years, 19.5% of the patients were black, and 7.9% resided in areas with high poverty rates (Table 1). The risk factors for cardiac complications were highly prevalent, and 79% had blood pressure levels greater than the 2011 American Diabetes Association recommended values.15 SDB was identified in 188 patients at baseline, 70% of whom were diagnosed using polysomnography, and 38% of whom used continuous positive airway pressure. As expected, patients with SDB had a significantly greater body mass index than patients without SDB (Table 1). During the follow-up period, 22 patients experienced incident congestive HF and 72 CAD, with patients with SDB having twice the event rate of those without SDB
1216
The American Journal of Cardiology (www.ajconline.org)
Table 1 Baseline characteristics stratified by sleep disordered breathing (SDB) Total (n ¼ 834)
Baseline* Characteristics
Women Black race High poverty postal code Single Body mass index 30 kg/m2 Body mass index 25 kg/m2 Hypercholesterolemia Hypertension Cerebrovascular disease Depression or anxiety Age (yrs) Age 75 yrs Family history of coronary artery disease Ever smoked Coronary artery disease risk factorsz Suboptimal blood pressure controlx Suboptimal glycemic control{ Creatinine 1.5 mg/dl Low-density lipoprotein 100 mg/dl Hemoglobin A1c (%) Glucose (mg/dl) Creatinine (mg/dl) Low-density lipoprotein (mg/dl) High-density lipoprotein (mg/dl) Triglycerides (mg/dl) Body mass index (kg/m2) Body surface area (m2) Maximum heart rate (beats/min) Systolic blood pressure (mm Hg) Diastolic blood pressure (mm Hg) Ejection fraction (%) Left ventricular mass (g) Diastolic dysfunction stage 1 2 3 End-diastolic volume (ml) Indexed end-diastolic volume (ml) Median left atrial area (cm2) Abnormal Duke score Insulin b Blockers Angiotensin-converting enzyme inhibitors Diuretics Aspirin Statins Nonsteroidal anti-inflammatory drug Continuous positive airway pressure or bilevel positive airway pressure at baseline
SDB
p Value
Yes (n ¼ 188)
No (n ¼ 646)
369 (44.2%) 163 (19.5%) 66 (7.9%) 122 (14.6%) 475 (57.0%) 731 (87.7%) 584 (70.0%) 508 (60.9%) 37 (4.4%) 201 (24.1%) 56.2 11.3 48 (5.8%) 317 (45.1%) 376 (45.1%) 409 (49%) 658 (78.9%) 396 (47.5%) 20 (2.4%) 387 (46.4%) 7.4 1.7 144.2 63.4 0.88 0.39 104.2 34.3 52.0 14.7 166.4 113.8 32.1 6.8 2.0 0.3 154.8 15.7 134.0 18.1 81.6 9.8 63.0 7.8 177.6 56.5
95 (50.5%) 38 (20.2%) 8 (4.3%) 29 (15.4%) 148 (78.7%) 182 (96.8%) 137 (72.9%) 131 (69.7%) 6 (3.2%) 57 (30.3%) 54.6 10.1 5 (2.7%) 83 (44.2%) 90 (47.9%) 92 (48.9%) 153 (81.4%) 99 (52.7%) 4 (2.1%) 92 (48.9%) 7.4 1.5 143.4 52.9 0.94 0.69 105.9 33.7 49.9 12.4 185.1 125.9 36.0 7.3 2.1 0.2 156 14 135 18 83 10 62 8 189.2 55.4
274 (42.4%) 125 (19.4%) 58 (9.0%) 93 (14.4%) 327 (50.6%) 549 (85.0%) 447 (69.2%) 377 (58.4%) 31 (4.8%) 144 (22.3%) 56.7 11.6 43 (6.7%) 234 (36.2%) 286 (44.3%) 317 (49.1%) 505 (78.2%) 297 (46.0%) 16 (2.5%) 295 (45.7%) 7.4 1.8 144.4 66.3 0.86 0.24 103.7 34.5 52.6 15.3 160.8 109.5 30.9 6.2 2.0 0.3 154 16 134 18 81 10 63 8 177.9 56.4
362 (43.4%) 35 (4.2%) 5 (0.6%) 85.8 25.0 42.3 10.7 16.2 4.0 103 (12.4%) 152 (18.2%) 133 (16.0%) 311 (37.3%) 285 (34.2%) 315 (37.8%) 465 (55.8%) 208 (24.9%) 72 (8.6%)
74 (39.4%) 4 (2.1%) 1 (0.5%) 90.9 26.6 42.4 11.4 16.7 3.9 19 (10%) 42 (22.3%) 33 (17.0%) 80 (42.6%) 79 (42.0%) 68 (36.2%) 108 (57.5%) 63 (33.5%) 72 (38.3%)
288 (46.6%) 31 (4.8%) 4 (0.6%) 84.3 24.3 42.3 10.4 16.0 4.0 84 (13%) 110 (17.0%) 101 (15.6%) 231 (35.8%) 206 (31.9 %) 247 (38.2%) 357 (55.3%) 145 (22.5%) NA
0.05† 0.08 0.03† 0.73 <0.0001† <0.0001† 0.43 0.005† 0.35 0.02† 0.024† 0.04† 0.14 0.38 0.97 0.34 0.12 NA 0.43 0.97 0.84 0.014† 0.45 0.031† 0.012† <0.0001† <0.0001† 0.24 0.19 0.06 0.04† 0.005† 0.16
<0.001† 0.93 0.05† 0.29 0.10 0.65 0.09 0.01† 0.61 0.60 0.002† NA
Data are presented as n (%) or mean SD. NA ¼ not available. * At stress echocardiography. † Statistically significant. z Defined as men aged 50 years or women aged 60 years, with any 1 of hypertension, smoking, family history of CAD, low-density lipoprotein 100 mg/dl. x Diastolic blood pressure control at rest of 80 mm Hg and/or systolic blood pressure control at rest of 130 mm Hg. { Postprandial glucose 180 mg/dl or hemoglobin A1c 7.
(Table 2). Patients with SDB had nearly 3 times the rate of AF compared to patients without SDB. After censoring for new congestive HF and/or death, the age-adjusted hazard
ratios showed a strong association between SDB and both incident CAD (p <0.001) and AF (p ¼ 0.002; Table 3). A marginal association was found between suspected or
Miscellaneous/Cardiac Consequences of Sleep Disordered Breathing
1217
Table 2 Incident cardiac events stratified by sleep disordered breathing (SDB) Cardiac Events
Heart failure Coronary artery disease Atrial fibrillation
Total (n ¼ 834)
22 (2.6%) 72 (8.6%) 40 (4.8%)
SDB Yes (n ¼ 188)
No (n ¼ 646)
8 (4.3%) 25 (13.3%) 17 (9.0%)
14 (2.2%) 47 (7.3%) 23 (3.6%)
Relation of SDB to Incident Cardiac Events*
p Value†
1.9 (0.8e4.6) 1.8 (1.1e3.0) 2.6 (1.4e4.7)
0.14 0.01 <0.01
Cardiac incident events included acute myocardial infarction, acute coronary syndrome, and new onset of chronic ischemic heart disease. * Data are presented as unadjusted hazard ratio (95% confidence interval). † p Value for overall effect of SDB modeled as a categorical variable.
proven SDB and incident HF (p ¼ 0.07). Elevated highdensity lipoprotein was found to be protective against CAD (p ¼ 0.03). Female gender, cerebrovascular disease, poor glycemic control (elevated hemoglobin A1c and postprandial glucose), elevated systolic blood pressure, and increased left ventricular mass were associated with incident CAD in the present cohort. The body mass index, systolic blood pressure at rest, elevated end-diastolic volume, increased left atrial area, and the use of b blockers and diuretics were also significant age-adjusted associations of AF. SDB was a significant correlate of CAD and AF in both unadjusted and adjusted models (Table 4). Age was associated with incident HF and CAD and was marginally associated with AF, and the systolic blood pressure at rest correlated with all incident cardiac events. In a subanalysis of patients diagnosed with SDB using polysomnography, SDB was independently associated with incident HF and CAD, but not with AF (Table 5). Discussion The present study has linked a greater risk of new onset cardiac disease in those with T2DM and with SDB than in those without SDB. SDB was associated with a greater risk of incident CAD after controlling for previously identified risk factors, including age, gender, hypertension, smoking, lipid profile, family history of cerebrovascular disease, medical treatment, standard of care-related risk factors, or overweight/obesity. Whether considering the SDB group as those with a high index of clinical suspicion and those diagnosed using polysomnography or limiting that group to only those patients diagnosed using polysomnography, associations were identified with AF and HF. Studies have suggested an association between T2DM and SDB, independent of obesity.2,3 Although T2DM and SDB have been individually linked to an increased risk of cardiac disorders, the present study is the first to show an association of SDB with an increased risk of incident cardiac events in patients with T2DM. The baseline prevalence of SDB in our study was similar to that recently reported by a large retrospective, population-based, multicenter study.3 It is important to note that although 188 patients were diagnosed with SDB at baseline, an additional 25 were diagnosed with SDB during the follow-up period. A number of studies in other patient populations have identified a relation between SDB and CAD,4e7 including the Sleep Heart Health Study (SHHS),7 which reported an
increased risk of incident CAD in 40- to 70-year-old men. Our work differs from previous studies by offering greater female representation, sample size, separation of end points, and exploration of the differences between SDB and nonSDB with respect to CAD risk factors, healthcare access and quality, and treatment differences for existent comorbidities. We also found that SDB was associated with more than twice the risk of incident AF in patients with T2DM, Our findings are consistent with studies that found arrhythmias to be more prevalent in patients with SDB than in patients without SBD,8,9 including another SHHS study, reporting a prevalence of 4.8% and 0.9%, respectively.10 As expected, the rates of AF were much greater in our patient population because of the co-morbidity with T2DM. In addition, SDB diagnosed using polysomnography was associated with HF. Several studies have found a high prevalence of SDB in patients with HF, ranging from 10% to 35%.11,12 One previous study done as part of the SHHS found that SDB was associated with incident HF in men, but not in women.7 SDB has been linked to vascular endothelial changes through increased proinflammatory mediators,19,20 and studies have confirmed elevated inflammatory markers in persons with SDB, independent of obesity.21 A number of physiologic mechanisms have been proposed for the association of SDB with cardiac events, 1 of which is hypertension. Although we found elevated systolic blood pressure to increase the risk of all incident HF, CAD, and AF in our final models, the baseline systolic blood pressure of patients with and without SDB was similar, suggesting that hypertension might not be a principal causal mechanism for increased incident cardiac events in patients with asymptomatic T2DM and SDB. Although a number of studies have shown that patients with SDB are more resistant to catheter ablation as treatment of AF,22,23 it is still unclear whether this relates to a different pathophysiology of AF in SDB- compared to noneSDB-associated AF.13 The finding that female gender is associated with incident CAD was unexpected and might be related to the selection of patients with T2DM. Women have been found to have more pronounced vasoconstriction in response to both hypoxia and stress compared to men.24 In addition, women with SDB have been reported to have larger brachial artery diameters than women without SDB, and an increased brachial diameter has been identified as a risk factor for CAD in women.25 There is a strong association between the apnea-hypopnea index and the flow-mediated diameter in female patients with SDB,26 suggesting a link with
1218
The American Journal of Cardiology (www.ajconline.org)
Table 3 Age*-adjusted hazard ratios and 95% confidence intervals for predictors of incident cardiac events Variable
HF HR (95% CI)
Sleep disordered breathing Women Black race High poverty postal code Single Body mass index 30 kg/m2 Hypercholesterolemia Hypertension Cerebrovascular disease Depression or anxiety Family history of coronary artery disease Ever smoke Coronary artery disease risk factorsz Suboptimal blood pressure controlx Suboptimal glucose control{ Creatinine 1.5 mg/dl Low-density lipoprotein 100 mg/dl Hemoglobin A1c Glucose (mg/dl) Creatinine (mg/dl) Low-density lipoprotein (mg/dl) High-density lipoprotein (mg/dl) Triglycerides (mg/dl) Body mass index (kg/m2) Maximum heart rate (beats/min) Systolic blood pressure at rest (mm Hg) Diastolic blood pressure at rest (mm Hg) Ejection fraction Left ventricular mass (g) Diastolic dysfunction (stage) End-diastolic volume (ml) End-diastolic volume/body surface area (ml/m2) Left atrial area (cm2) Abnormal Duke score Insulin b Blockers Angiotensin-converting enzyme inhibitors Diuretics Aspirin Statins Nonsteroidal anti-inflammatory drugs
2.2 1.2 1.3 1.2 0.6 1.7 0.9 1.4 0.9 1.4 0.7 1.5 1.2 1.6 1.7 3.1 0.6 1.2 1.0 1.1 1.0 1.0 1.0 1.0 0.96 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.1 1.9 1.2 1.3 0.9 1.7 0.8 2.2 1.2
(0.9e5.4) (0.5e2.7) (0.5e3.4) (0.3e5.0) (0.1e2.7) (0.7e4.2) (0.6e1.4) (0.5e3.6) (0.1e6.4) (0.6e3.4) (0.3e1.4) (0.6e3.4) (0.4e3.5) (0.5e5.2) (0.7e3.9) (0.7e13.8) (0.3e1.5) (1.0e1.5) (1.0e1.01) (0.4e3.0) (1.0e1.01) (1.0e1.04) (0.9e1.0) (0.9e1.1) (0.9e1.0) (1.0e1.1) (0.9e1.1) (0.9e1.1) (1.0e1.01) (0.4e2.3) (0.99e1.0) (0.98e1.0) (1.0e1.2) (0.7e5.0) (0.4e3.7) (0.5e3.5) (0.4e2.2) (0.7e3.9) (0.3e1.9) (0.8e6.1) (0.5e3.1)
CAD p Value 0.07 0.72 0.66 0.83 0.53 0.24 0.70 0.49 0.88 0.49 0.30 0.37 0.80 0.48 0.25 0.13 0.29 0.11 0.04† 0.87 0.72 0.66 0.77 0.18 0.03† <0.01† 0.22 0.99 0.07 0.93 0.78 0.82 0.11 0.23 0.70 0.66 0.88 0.23 0.63 0.11 0.68
HR (95% CI) 2.6 2.2 1.0 0.7 0.4 1.3 0.9 1.5 2.4 0.6 0.8 1.2 1.9 1.3 1.6 1.0 1.2 1.1 1.0 1.2 1.0 0.98 1.0 1.0 1.0 1.0 1.0 0.98 1.0 1.3 1.0 1.0 1.0 1.4 1.4 1.3 1.1 1.1 1.0 1.2 1.3
(1.5e4.5) (1.4e3.6) (0.6e1.9) (0.3e1.9) (0.1e1.0) (0.8e2.1) (0.6e1.1) (0.9e2.6) (1.1e5.1) (0.3e1.1) (0.6e1.1) (0.8e2.0) (1.0e3.6) (0.7e2.3) (1.0e2.6) (0.2e4.0) (0.7e1.9) (0.98e1.3) (1.0e1.01) (0.9e1.7) (1.0e1.01) (0.96e1.0) (0.99e1.0) (1.0e1.1) (0.98e1.0) (1.0e1.03) (0.98e1.0) (0.95e1.0) (1.0e1.01) (0.8e2.1) (0.99e1.0) (0.98e1.0) (1.0e1.1) (0.7e2.5) (0.8e2.5) (0.7e2.2) (0.6e1.8) (0.7e1.8) (0.6e1.6) (0.7e1.9) (0.7e2.1)
AF p Value <0.001† <0.001† 0.90 0.48 0.06 0.28 0.26 0.13 0.02† 0.09 0.18 0.36 0.05† 0.48 0.04† 0.98 0.48 0.09 <0.001† 0.21 0.24 0.03† 0.17 0.37 0.74 <0.001† 0.92 0.30 <0.001† 0.34 0.16 0.82 0.43 0.34 0.22 0.41 0.70 0.61 0.90 0.51 0.39
HR (95% CI) 2.9 1.0 0.4 0.4 1.3 0.9 1.0 1.0 2.2 0.4 1.0 1.4 0.9 4.0 1.1 1.1 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.1 1.0 1.0 1.0 1.0 1.0 0.7 1.0 1.0 1.1 0.4 0.5 2.3 0.7 2.1 1.8 1.0 1.2
(1.5e5.8) (0.5e1.9) (0.1e1.4) (0.1e2.8) (0.5e3.2) (0.4e1.7) (0.8e1.3) (0.5e1.9) (0.7e7.4) (0.1e1.1) (0.8e1.3) (0.7e2.7) (0.4e2.1) (1.0e16.7) (0.5e2.1) (0.1e8.0) (0.5e1.9) (0.8e1.3) (1.0e1.0) (0.3e2.9) (1.0e1.01) (0.98e1.03) (0.99e1.0) (1.0e1.1) (0.99e1.0) (1.0e1.04) (1.0e1.1) (0.96e1.1) (0.99e1.0) (0.4e1.4) (1.0e1.03) (0.99e1.1) (1.0e1.2) (0.1e1.7) (0.2e1.7) (1.3e5.6) (0.3e1.5) (1.1e4.2) (0.9e3.6) (0.5e2.0) (0.5e2.5)
p Value 0.002† 0.92 0.16 0.34 0.53 0.66 0.76 0.88 0.20 0.07 0.96 0.34 0.74 0.06 0.84 0.94 0.96 0.95 0.69 0.99 0.59 0.69 0.87 0.03† 0.31 0.03† 0.16 0.83 0.66 0.34 0.05† 0.06 0.04† 0.21 0.27 <0.01† 0.36 0.03† 0.08 0.98 0.67
CI ¼ confidence interval; HR ¼ hazard ratio. * Age variable was centered. † Statistically significant. z Defined as men aged 50 years or women aged 60 years, with any 1 of hypertension, smoking, family history of CAD, low-density lipoprotein 100 mg/dl. x Diastolic blood pressure control at rest of 80 mm Hg and/or systolic blood pressure control at rest of 130 mm Hg. { Postprandial glucose 180 mg/dl or hemoglobin A1c 7.0%.
endothelial dysfunction. Another study reported female gender was associated with greater mortality in a sample of patients with SDB treated with continuous positive airway pressure.27 The investigators proposed either that gender might be an independent risk factor for mortality in their sample or that their finding could be related to delayed diagnoses and treatment of SDB in female compared to male patients. Recently, a study limited to female patients with SDB confirmed the association of SDB with an increased risk of cardiovascular mortality in women and found SDB
severity to be associated with a greater risk of mortality.28 However, we found no interaction between gender and SDB in the present study. Compliance with continuous positive airway pressure or bilevel positive airway pressure treatment in our sample was similar to the rates observed in previous studies.29 Although only 38% of patients with SDB were compliant with treatment at baseline, this proportion had increased to 46% through the end of the follow-up period. However, the present study could not assess the effect of continuous
Miscellaneous/Cardiac Consequences of Sleep Disordered Breathing
1219
Table 4 Adjusted models* showing relation of sleep disordered breathing (SDB) to incident cardiac events Model
Predictors
HF HR (95% CI)
1 2
3
4
SDB Agez SDB Agez Female gender SDB Agez Female gender CAD risk factorsx SDB Agez Female gender CAD risk factorsx Systolic blood pressure
2.2 1.1 2.2 1.1 1.1 2.2 1.1 1.1 1.1 2.2 1.1 1.2 0.8 1.03
CAD p Value
(0.9e5.4) (1.0e1.1) (0.9e5.4) (1.0e1.1) (0.5e2.6) (0.9e5.4) (1.0e1.1) (0.5e2.7) (0.3e3.3) (0.9e5.4) (1.0e1.1) (0.5e2.9) (0.2e2.6) (1.0e1.04)
0.07 <0.01† 0.07 <0.01† 0.79 0.08 0.04† 0.81 0.97 0.08 0.04† 0.69 0.65 <0.01†
HR (95% CI) 2.3 1.05 2.0 1.1 2.1 1.9 1.1 2.0 1.3 1.9 1.04 2.1 1.1 1.02
AF p Value
(1.3e3.8) (1.0e1.1) (1.2e3.2) (1.0e1.1) (1.3e3.4) (1.2e3.2) (1.0e1.1) (1.2e3.3) (0.7e2.5) (1.2e3.2) (1.0e1.1) (1.3e3.5) (0.5e2.1) (1.0e1.04)
<0.01† <0.001† <0.01† <0.001† <0.001† <0.01† <0.01† <0.01† 0.45 <0.01† 0.04† <0.01† 0.85 <0.01†
HR (95% CI) 2.9 1.03 2.9 1.04 0.9 3.0 1.04 0.97 0.7 2.9 1.04 1.01 0.7 1.02
(1.5e5.8) (1.0e1.1) (1.5e5.8) (1.0e1.1) (0.4e1.8) (1.5e5.9) (0.99e1.1) (0.5e2.1) (0.3e2.0) (1.5e5.9) (0.99e1.1) (0.5e2.1) (0.3e1.7) (1.0e1.04)
p Value <0.01† 0.06 <0.01† 0.06 0.77 <0.01† 0.08 0.93 0.62 <0.01† 0.08 0.98 0.39 0.02†
Abbreviations as in Table 3. * Each model was adjusted for factors listed with it, with reported HR for each factor taking into account all other factors included in the model. † Statistically significant. z Age variable was centered. x Defined as men aged 50 years or women aged 60 years, with any 1 of hypertension, smoking, family history of CAD, or low-density lipoprotein 100 mg/dl.
Table 5 Models showing relation of sleep disordered breathing (SDB) confirmed by polysomnography (n ¼ 132) to incident cardiac events Predictors
HF HR (95% CI)
Unadjusted Model 1 Model 2
Model 3
Model 4
SDB SDB Age† SDB Age† Female gender SDB Age† Female gender CAD risk factorsz SDB Age† Female gender CAD risk factorsz Systolic blood pressure
2.7 3.7 1.1 3.5 1.1 1.2 3.6 1.1 1.1 1.1 3.5 1.1 1.2 0.8 1.03
(1.1e7.0) (1.4e9.4) (1.0e1.1) (1.4e9.2) (1.0e1.1) (0.5e2.8) (1.4e9.2) (1.0e1.1) (0.5e2.8) (0.3e3.6) (1.4e9.0) (1.0e1.1) (0.5e3.1) (0.2e2.9) (1.0e1.05)
CAD p Value 0.03 <0.01* <0.01* <0.01* <0.01* 0.11 <0.01* 0.02* 0.77 0.91 <0.01* 0.02* 0.64 0.77 <0.01*
HR (95% CI) 1.9 2.3 1.1 2.2 1.1 2.0 2.2 1.1 1.9 1.1 2.2 1.05 1.9 1.0 1.01
(1.1e3.3) (1.3e4.2) (1.0e1.1) (1.2e3.9) (1.0e1.1) (1.2e3.2) (1.2e3.9) (1.0e1.1) (1.1e3.2) (0.6e2.2) (1.2e3.9) (1.0e1.1) (1.1e3.3) (0.5e2.0) (1.0e1.03)
AF p Value 0.03 <0.01* <0.001* <0.01* <0.0001* 0.01* <0.01* <0.001* <0.01* 0.7 0.01* <0.01* 0.01* 0.9 0.06
HR (95% CI) 1.7 1.9 1.04 1.9 1.04 1.2 2.0 1.1 1.4 0.5 1.9 1.06 1.4 0.5 1.02
(0.8e3.9) (0.9e4.6) (1.0e1.1) (0.8e4.5) (1.0e1.1) (0.6e2.4) (0.9e4.9) (1.0e1.1) (0.6e2.9) (0.1e1.3) (0.8e4.7) (0.99e1.1) (0.7e3.0) (0.2e1.1) (1.0e1.04)
p Value 0.20 0.1 0.02 0.13 0.02 0.69 0.1 <0.01 0.40 0.50 0.10* <0.01* 0.40 0.08 0.10
Abbreviations as in Table 3. * Statistically significant. † Age variable was centered. z Defined as men aged 50 years or women aged 60 years, with any 1 of hypertension, smoking, family history of CAD, or low-density lipoprotein 100 mg/dl.
positive airway pressure treatment on cardiac events, owing to the limitation of the self-report of SDB treatment in the electronic health records. Our study had several important limitations. Although polysomnography is the reference standard for the diagnosis of sleep disorders,17 we also chose to consider patients to have SDB if they were identified as having a high risk of SDB18 but did not undergo polysomnography. This allowed us to reduce the bias from noncompliance with referral. Most patients (70%) we defined as having SDB in our study
had polysomnogram-confirmed SDB, and our findings were similar when limiting our sample to these patients (Table 5). This was an observational retrospective study based on the facilitation of clinical care using electronic health records. However, the baseline variables were carefully collected in the exercise echocardiogram registry at the time of the test, and all other clinical data were also entered in the electronic health records prospectively, albeit for patient care rather than for research purposes. Because the purpose of the study was to examine the association of clinically
1220
The American Journal of Cardiology (www.ajconline.org)
manifest SDB (previously diagnosed by a physician) with the prediction of incident cardiac events, exclusions according to co-morbid cardiac disease at echocardiographic testing focused the study, but also resulted in an inability to further subdivide patients with co-morbid disease. The study sample was derived from a single, large center and included only patients residing in Ohio, so that follow-up could be performed; thus, studies in other populations are warranted. Although potentially important, but difficult to verify, we did not gather data on diet and exercise. In addition, apneahypopnea index severity and associations with hypoxic stress were not quantified, because it was outside the scope of our study to assess the relation between SDB severity and incident cardiac events. Data on compliance with continuous positive airway pressure and bilevel positive airway pressure were self-reported. However, it was outside the scope of our study to assess the value of SDB treatment on cardiac outcomes.
13. 14.
15. 16.
17.
Disclosures The authors have no conflicts of interest to disclose. 1. Centers for Disease Control and Prevention. National Diabetes Fact Sheet: National Estimates and General Information on Diabetes and Prediabetes in the United States, 2011. Atlanta: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, 2011: 1e12. Available at: http://www.cdc.gov/diabetes/pubs/ pdf/ndfs_2011.pdf. Accessed December 18, 2012. 2. Seicean S, Kirchner L, Gottlieb DJ, Punjabi N, Resnick H, Sanders M, Budhiraja R, Singer M, Redline S. Sleep-disordered breathing and impaired glucose metabolism in normal-weight and overweight/obese individuals: the Sleep Heart Health Study. Diabetes Care 2008;31:1001e1006. 3. Heffner JE, Rozenfeld Y, Kai M, Stephens EA, Brown LK. Prevalence of diagnosed sleep apnea among patients with T2DM in primary care. Chest 2012;141:1414e1421. 4. Shah NA, Yaggi HK, Concato J, Mohsenin V. Obstructive sleep apnea as a risk factor for coronary events or cardiovascular death. Sleep Breath 2010;14:131e136. 5. Martinez D, Klein C, Rahmeier L, da Silva RP, Fiori CZ, Cassol CM, Gonçalves SC, Bos AJ. Sleep apnea is a stronger predictor for coronary heart disease than traditional risk factors. Sleep Breath 2012;16:695e701. 6. Peker Y, Hedner J, Norum J, Kraiczi H, Carlson J. Increased incidence of cardiovascular disease in middle-aged men with obstructive sleep apnea: a 7-year follow-up. Am J Respir Crit Care Med 2002;166:159e165. 7. Gottlieb DJ, Yenokyan G, Newman AB, O’Connor GT, Punjabi NM, Quan SF, Redline S, Resnick HE, Tong EK, Diener-West M, Shahar E. Prospective study of obstructive sleep apnea and incident coronary heart disease and heart failure: the Sleep Heart Health Study. Circulation 2010;122:352e360. 8. Hoffstein V, Mateika S. Cardiac arrhythmias, snoring, and sleep apnea. Chest 1994;106:466e471. 9. Namtvedt SK, Randby A, Einvik G, Hrubos-Strøm H, Somers VK, Røsjø H, Omland T. Cardiac arrhythmias in obstructive sleep apnea (from the Akershus Sleep Apnea Project). Am J Cardiol 2011;108: 1141e1146. 10. Mehra R, Benjamin EJ, Shahar E, Gottlieb DJ, Nawabit R, Kirchner HL, Sahadevan J, Redline S; Sleep Heart Health Study. Association of nocturnal arrhythmias with sleep-disordered breathing: the Sleep Heart Health Study. Am J Respir Crit Care Med 2006;173:910e916. 11. Sin DD, Fitzgerald F, Parker JD, Newton G, Floras JS, Bradley TD. Risk factors for central and obstructive sleep apnea in 450 men and women with congestive heart failure. Am J Respir Crit Care Med 1999;160:1101e1106. 12. Wang H, Parker JD, Newton GE, Floras JS, Mak S, Chiu KL, Ruttanaumpawan P, Tomlinson G, Bradley TD. Influence of obstructive
18.
19. 20.
21.
22.
23. 24. 25.
26. 27. 28.
29.
sleep apnea on mortality in patients with heart failure. J Am Coll Cardiol 2007;49:1625e1631. Monahan K, Redline S. Role of obstructive sleep apnea in cardiovascular disease. Curr Opin Cardiol 2011;26:541e547. Coulton C, Colabianchi N, Cook T. An Analysis of Differences in Health Status Measures Among Neighborhoods within Metro Counties of Ohio, 2001. Available at: http://www2.odh.ohio.gov/Data/OFHSurv/ ofhsrfq2sup.pdf. Accessed February 12, 2012. American Diabetes Association. Standard of medical care in diabetes2012. Diabetes Care 2012;35:511e563. Douglas PS, Geller U, Khandheria B, Stainback RF, Weissman NJ, Peterson ED, Hendel RC, Stainback RF, Blaivas M, Des Prez RD, Gillam LD, Golash T, Hiratzka LF, Kussmaul WG, Labovitz AJ, Lindenfeld J, Masoudi FA, Mayo PH, Miller TD, Porembka D, Spertus JA, Wann S, Wiegers SE, Brindis RG, Douglas PS, Hendel RC, Patel MR, Peterson ED, Wolk MJ, Allen JM, Lewin JC, Allen JM, Caruth KC, Barrett EA. ACCF/ASE/ACEP/AHA/ASNC/SCAI/SCCT/SCMR 2008 appropriateness criteria for stress echocardiography. J Am Coll Cardiol 2008;51:1127e1147. Epstein LJ, Kristo D, Strollo PJ Jr, Friedman N, Malhotra A, Patil SP, Ramar K, Rogers R, Schwab RJ, Weaver EM, Weinstein MD; Adult Obstructive Sleep Apnea Task Force of the American Academy of Sleep Medicine. Clinical guideline for the evaluation, management and long-term care of obstructive sleep apnea in adults. J Clin Sleep Med 2009;5:263e276. Kushida CA, Littner MR, Morgenthaler T, Alessi CA, Bailey D, Coleman J Jr, Friedman L, Hirshkowitz M, Kapen S, Kramer M, LeeChiong T, Loube DL, Owens J, Pancer JP, Wise M. Practice parameters for the indications for polysomnography and related procedures: an update for 2005. Sleep 2005;28:499e521. Ip MS, Tse HF, Lam B, Tsang KW, Lam WK. Endothelial function in obstructive sleep apnea and response to treatment. Am J Respir Crit Care Med 2004;169:348e353. Jelic S, Padeletti M, Kawut SM, Higgins C, Canfield SM, Onat D, Colombo PC, Basner RC, Factor P, LeJemtel TH. Inflammation, oxidative stress, and repair capacity of the vascular endothelium in obstructive sleep apnea. Circulation 2008;117:2270e2278. Larkin EK, Rosen CL, Kirchner HL, Storfer-Isser A, Emancipator JL, Johnson NL, Zambito AM, Tracy RP, Jenny NS, Redline S. Variation of C-reactive protein levels in adolescents: association with sleepdisordered breathing and sleep duration. Circulation 2005;111: 1978e1984. Hoyer FF, Lickfett LM, Mittmann-Braun E, Ruland C, Kreuz J, Pabst S, Schrickel J, Juergens U, Tasci S, Nickenig G, Skowasch D. High prevalence of obstructive sleep apnea in patients with resistant paroxysmal atrial fibrillation after pulmonary vein isolation. J Interv Card Electrophysiol 2010;29:37e41. Ng CY, Liu T, Shehata M, Stevens S, Chugh SS, Wang X. Metaanalysis of obstructive sleep apnea as predictor of atrial fibrillation recurrence after catheter ablation. Am J Cardiol 2011;108:47e51. Levenson J, Pessana F, Gariepy J, Armentano R, Simon A. Gender differences in wall shear-mediated brachial artery vasoconstriction and vasodilation. J Am Coll Cardiol 2001;38:1668e1674. Holubkov R, Karas RH, Pepine CJ, Rickens CR, Reichek N, Rogers WJ, Sharaf BL, Sopko G, Merz CN, Kelsey SF, McGorray SP, Reis SE. Large brachial artery diameter is associated with angiographic coronary artery disease in women. Am Heart J 2002;143:802e807. Faulx MD, Larkin EK, Hoit BD, Aylor JE, Wright AT, Redline S. Sex influences endothelial function in sleep-disordered breathing. Sleep 2004;27:1113e1120. Morrish E, Shneerson JM, Smith IE. Why does gender influence survival in obstructive sleep apnoea? Respir Med 2008;102: 1231e1236. Campos-Rodriguez F, Martinez-Garcia MA, de la Cruz-Moron I, Almeida-Gonzalez C, Catalan-Serra P, Montserrat JM. Cardiovascular mortality in women with obstructive sleep apnea with or without continuous positive airway pressure treatment: a cohort study. Ann Intern Med 2012;156:115e122. Weaver TE, Grunstein RR. Adherence to continuous positive airway pressure therapy: the challenge to effective treatment. Proc Am Thorac Soc 2008;5:173e178.