| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Division of Cardiovascular Diseases (Drs. Gami, Rader, Svatikova, Wolk, Winnicki, and Somers, and Ms. Huyber), and the Sleep Disorders Center (Mr. Herold), Mayo Clinic College of Medicine, Rochester, MN.
Correspondence to: Virend K. Somers, MD, PhD, Mayo Clinic College of Medicine, 200 First St SW, Rochester, MN 55905; e-mail: somers.virend{at}mayo.edu
Abstract
Background: Obstructive sleep apnea (OSA) is linked to both coronary artery disease (CAD) and sudden death, but any causal role remains unclear. A family history of premature CAD and related mortality is an independent risk factor for the development of CAD. We hypothesized that OSA is associated with a family history of premature mortality from ischemic heart disease.
Methods: We prospectively studied 588 subjects who underwent polysomnography from May 2000 to June 2004. Demographics, comorbidities, family history of cardiovascular disease, and the ages and causes of death for 10 strata of family members were recorded for all subjects. We excluded those subjects with known causes of premature cardiac death, such as hypertrophic cardiomyopathy and long-QT syndrome. OSA was defined by American Academy of Sleep Medicine criteria (ie, apnea-hypopnea index 
5). Premature CAD mortality was defined as death due to ischemic heart disease or sudden cardiac death before 55 years of age (men) or 65 years of age (women).
Results: Polysomnography confirmed OSA in 316 subjects and excluded it in 202 subjects. The unadjusted odds ratio (OR) for OSA and a family history of premature CAD mortality was 2.11 (95% confidence interval [CI], 1.10 to 4.31; p = 0.031). After adjusting for each subject’s sex, body mass index, and history of CAD, there was a significant and independent association between OSA and family history of premature CAD mortality (OR, 2.13; 95% CI, 1.04 to 4.66; p = 0.046).
Conclusions: Regardless of their own CAD status, people with OSA are more likely than those without OSA to have a family history of premature CAD mortality.
Key Words: cardiovascular disease • family history • risk factor • sleep apnea
Family medical history provides crucial information regarding the individual risk for diseases, such as coronary artery disease (CAD), that results from poorly understood, complex interactions between genetic and environmental factors.1 A family history of premature CAD and related mortality are important independent risk factors for the development of CAD in an individual.23456789101112
Obstructive sleep apnea (OSA) is closely associated with many cardiovascular diseases13 and has been proposed as an independent risk factor for CAD morbidity and mortality.1415 Abundant data have described the association of OSA with multiple traditional and newly recognized CAD risk factors.13 However, the relationship of OSA with the powerful risk factor of family history of CAD and related mortality is unknown. The aim of the present study was to determine the association between OSA and a family history of premature CAD mortality.
Materials and Methods
Study Population and Polysomnography
We prospectively performed a cross-sectional study of 588 individuals who underwent diagnostic polysomnography between May 2000 and June 2004. Subjects included research volunteers from the community and patients who had been referred to the Mayo Clinic Sleep Disorders Center. Informed consent was obtained from all subjects, and the institutional review board approved the study. Polysomnography consisted of the overnight measurement of the EEG, electrooculogram, electromyogram, ECG, thoracoabdominal excursions, pulse oximetry, and oronasal airflow. The apnea-hypopnea index (AHI) was calculated as the sum of apneas and hypopneas occurring per hour of sleep. According to American Academy of Sleep Medicine criteria,16 an AHI of 5 established the diagnosis of OSA.
Family Medical History
Subjects were interviewed by a study physician using standardized forms to collect demographic and anthropomorphic information, personal medical history, and a detailed family medical history. For the latter, subjects recalled the presence or absence of hypertension, diabetes, smoking, obesity, CAD, myocardial infarction, stroke, congestive heart failure, and other diseases in their biological mother, father, brothers, sisters, sons, daughters, maternal grandmothers and grandfathers, and paternal grandmothers and grandfathers. They also recalled the cause of death and age at the time of death for each family member. We did not collect information about the age of onset of CAD and comorbidities in family members, because an exact time of the development of CAD and most other diseases is arbitrary. For this reason, premature CAD mortality was the risk factor assessed in this study, since the age of death of family members is definitive and offspring would be likely to recall accurately this detail about their immediate family members. Family history of premature CAD mortality was defined as death due to CAD, myocardial infarction, or sudden death (defined as death associated with an unexplained and sudden collapse) before 55 years of age (for men) and 65 years of age (for women) in any first-degree relative or grandparent. Incomplete or imprecise data for a family member were recorded as a negative family history.17 The accuracy of family cardiovascular disease history recall has been systematically validated previously.17181920
Statistical Analysis
The characteristics of the study population were described for individuals with OSA (AHI, 5) and individuals without OSA (AHI, < 5) as means (SDs) and counts (percentages), and these were compared by two-tailed t test and 
2 test. Simple logistic regression modeled the odds ratios (ORs) and 95% confidence intervals (CIs) for individual subject characteristics, including OSA status and family history of premature CAD mortality. Stepwise multiple logistic regression was performed to include statistically significant variables from the univariate analysis and other clinically relevant variables to identify the adjusted OR and 95% CI for associations with a family history of premature CAD mortality. Separate analyses were performed to assess for selection bias based on subjects’ status as community volunteers or clinic volunteers.
Results
Study Population and Sleep Diagnoses
Of the 588 individuals who underwent polysomnography during the years of the study, 15 were excluded due to incomplete information (ie, unavailable family medical history or incomplete polysomnographic data), 2 were excluded because they were adopted and could not provide information about their biological family members, and 53 were excluded because they had long-QT syndrome or hypertrophic cardiomyopathy (and thus a recognized familial risk of cardiac mortality). The final study sample included 316 individuals with OSA and 202 without OSA, which was established by performing complete overnight polysomnography studies. For the entire study sample, the mean (± SD) sleep efficiency during polysomnography was 76 ± 16%. For patients with OSA, the mean AHI was 32 ± 28, the mean nocturnal oxygen saturation was 93 ± 2%, and the mean lowest nocturnal oxygen saturation was 81 ± 10%. The results of polysomnography in patients without OSA revealed central sleep apnea (4%), behavioral sleep disorder (2%), obesity hypoventilation syndrome (1%), and primary snoring or no sleep disorder (93%). As expected, patients with OSA were more likely to be male and obese, and to have diabetes, hypertension, and CAD (Table 1
). The association between OSA and CAD was strong (OR, 2.38; 95% CI, 1.53 to 3.79; p < 0.001).
|
Family History of Premature CAD Mortality
Despite the similar proportion of subjects with and without OSA who had a personal history of CAD and a family history of CAD (at any age regardless of associated mortality), subjects with OSA were more likely than those without OSA to have a family history of premature CAD mortality (11.8% vs 6.0%, respectively; p = 0.025). The unadjusted OR for the association of OSA and a family history of premature CAD mortality was 2.11 (95% CI, 1.10 to 4.31; p = 0.031). In the multiple logistic regression model, OSA was significantly and independently associated with a family history of premature CAD mortality (adjusted OR, 2.13; 95% CI, 1.04 to 4.66; p = 0.046). The only other characteristic associated with a family history of premature CAD mortality was a subject’s personal history of CAD (unadjusted OR, 2.01; 95% CI, 1.06 to 3.72; p = 0.028), but this association was attenuated in the multivariate analysis (adjusted OR, 1.76; 95% CI, 0.92 to 3.31; p = 0.081). Separate analyses based on subjects’ status as community or clinic volunteers yielded similar results, providing no evidence for selection bias.
Discussion
Previous studies131521 have shown the association of OSA with multiple CAD risk factors, but no previous studies have evaluated a family history of CAD events in people with OSA. A family history of CAD and related mortality is a powerful cardiovascular risk factor representing yet unknown genetic and environmental factors that underlie an individual’s risk for CAD.23456789101112
The novel finding of this study is that individuals with proven OSA are more than twice as likely to have a family history of premature CAD mortality as those without OSA. This association is independent of important comorbidities, including an individual’s personal history of CAD. These findings suggest that individuals with OSA, regardless of whether they are healthy or have CAD risk factors, have an increased risk of CAD that is partly due to presently unknown familial factors acting independently of traditional CAD risk factors.
An important consideration regarding the classification of premature CAD is that the exact timing of the development of CAD is unknown, since atherosclerotic occlusion of coronary arteries may be asymptomatic and only recognized at the time that an overt cardiac event occurs. Thus, the date or age at which CAD developed in an individual is arbitrary. To overcome this limitation, most clinical studies use CAD events, such as myocardial infarction or sudden cardiac death, to signal its incidence. We followed this same strategy in our present study and assessed the association of OSA with familial premature CAD mortality. Furthermore, the term premature CAD may be arguable since it assumes that there is an age after which the development of CAD is expected. We used this term as it has been used historically, to describe the occurrence of CAD mortality before 55 years of age in women and before 65 years of age in men, since CAD mortality at these ages has been shown to identify heightened familial cardiovascular risk.12
A personal history of CAD was not strongly associated with a family history of premature CAD mortality after adjustment for other characteristics. This may be explained by the relatively young age of our study population, in whom overt CAD had not yet manifested. The proportion of people in our study population with a family history of CAD and other cardiovascular diseases (not taking into consideration age and mortality) reflects the epidemiology of heart disease in the United States,22 and people in our study with and without OSA did not differ in this regard. In our study, as in previous studies,12 the true effect of a family history of CAD is more apparent when the age of incidence and associated mortality are taken into account.
Another consideration is that the present study depended on the accurate and unbiased recollection of subjects’ family medical history. The "gold standard" approach to assessing familial risk would be to perform interviews and tests of all family members to identify all diseases and events of interest.12 This notwithstanding, multiple studies17181920232425 have validated family medical history recall as a conservative and accurate tool for assessing family history, especially for cardiovascular diseases.17181920 The National Heart Lung and Blood Institute Family Heart Study19 demonstrated that a family history of CAD can be assessed effectively based on an individual’s report (sensitivity, > 80%; specificity, > 90%). In that study, age affected an individual’s recall accuracy for family history, with deleterious effects after 60 years of age.19 Importantly, recollection was unlikely to be affected by older age in our study, since the mean age was only 45 years. The methods used in our present study are also strengthened by the fact that the primary variable of interest is related to mortality, which can be recalled more reliably than the presence or absence of a disease state. Moreover, from a practical standpoint, in a clinical setting the accuracy and usefulness of this information depends on the patients’ recollection, which was obtained in a fashion similar to our methods.
In conclusion, this cross-sectional study of > 500 individuals whose OSA status was established by complete overnight polysomnography demonstrates that a strong and independent association exists between OSA and a family history of premature CAD mortality. The association of OSA with this significant CAD risk factor has important implications for understanding cardiovascular risk, and perhaps its genetic underpinnings, in individuals with OSA.
|
Abbreviations: AHI = apnea-hypopnea index; CAD = coronary artery disease; CI = confidence interval; OR = odds ratio; OSA = obstructive sleep apnea
The authors are supported by National Institutes of Health grants No. HL61560, HL65176, HL73211, and M01-RR00585, and by the Mayo Clinic College of Medicine.
Dr. Somers is a consultant for Respironics, received an honorarium from ResMed, is a coinvestigator for a study funded with a grant from ResMed Foundation, and is involved with Mayo Health Solutions in technology development for sleep apnea and heart disease. All other authors have reported to the ACCP that no significant conflicts of interest exist with any companies/organizations whose products or services may be discussed in this article.
Received for publication June 4, 2006. Accepted for publication July 6, 2006.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
