In a large multiethnic sample of US adults, we found that markers of SDB, including short sleep duration (<7 hours), long sleep duration (>7 hours), occasional or frequent snoring or snorting (>2 night/week), and daytime sleepiness (>1/month) had moderately strong positive associations with the presence of CRP levels > 1
mg/dL in an initial multivariable model adjusting for age, sex, race/ethnicity, education, smoking, alcohol intake, and physical activity. However, additional adjustment for potential mediating factors such as BMI, depression, diabetes, hypertension, and total cholesterol substantially attenuated this association. Similarly, a summary SDB clustering score which counted the co-occurrence of various SDB markers showed that, when compared to those without SDB markers, those with clustering of SDB markers were associated with higher odds of having CRP levels > 1
mg/dL in the initial multivariable model. Here also the observed association was attenuated with additional adjustment for physical activity, body mass index, depression, diabetes, hypertension, and total cholesterol.
In the subgroup analysis examining the association between SDB clustering score and high CRP by gender, there was only a weak association in men, but a moderate association in women. In the subgroup analysis by race/ethnicity, the association between SDB clustering score and high CRP was strongly present in Mexican Americans, modestly present with a significant linear trend in Non-Hispanic Whites and absent in Non-Hispanic Blacks.
CRP is a nonspecific inflammatory biomarker that has been shown to be an important predictor of cardiovascular disease [14
]. Additionally, CRP is implicated in inducting adhesion molecules, and the production of monocyte chemoattractant protein-1 and sensitizing endothelial cells to the cytotoxic process by CD4+
T cells may also be an active atherosclerotic pathogenic agent, in and of itself [12
This study's main contribution to the existing literature concerning SDB is the demonstration in subgroup analyses that the SDB-CRP association is strong in Mexican Americans in the race/ethnicity subgroup analysis and that it is more strongly present in women in the gender subgroup analysis. The finding implies that SDB has a residual, independent association after the role of lifestyle and metabolic risk factors such as BMI, lack of physical activity, depression, diabetes, and total cholesterol in these subgroups.
The postulated mechanism of the association between SDB and CRP is complex. SDB may have the effects of hypoxemia, reoxygenation, hypercapnia, and arousals which activate systemic inflammation and the production of CRP [27
Similar studies have presented disparate results in analyzing the associations of CRP and SDB. Our study is consistent with a study of 69 men which indicated CRP is associated with SDB independent of adiposity [1
]. Another study of 316 men also found an association of SDB and increased CRP, especially in men who were not overweight [15
]. A third study of 935 women with type 2 diabetes also indicated SDB to be associated with increased CRP after adjusting for age, BMI, lifestyle, family history of diabetes, glycemic control, and medication use [28
]. But a study utilizing the Wisconsin Sleep Cohort Study of 907 participants reported no association between CRP and SDB after controlling for age, sex, and BMI [14
]. Utilizing the same study, but limiting the analysis to 589 women participants, SDB was 2.6 times more likely to occur postmenopausally than perimenopausally, after adjusting for age, body habitus, and certain lifestyle factors [29
], indicating a potential hormonal influence associated with SDB. Our study will be largest to date that examined the association between SDB markers and CRP levels. We found that there was a modest positive association, which, in subsequent subgroup analysis, was found more strongly present in women and in Mexican Americans. However, there is a need to examine the gender-and race/ethnicity-specific associations between SDB and markers of inflammation in larger studies and employing alternative markers of inflammation such as interleukin-6 or tumor necrosis factor-alpha to prove or disprove our findings.
Additionally, research is needed to determine if SDB may be associated with chronic fatigue syndrome (CFS). CFS results in prolonged (6 months or greater) disabling daytime sleepiness and is associated with high levels of proinflammatory markers [30
]. The mechanism of CFS appears to be an abnormal immunological response to an antigen/infection or a neuroendocrine problem resulting in a change in the CNS, particularly in young adults [30
]. Examining the role of SDB may provide greater understanding of CFS.
This study's strengths include the large and national sample size of the study population and the breadth of the national survey which included factors which we were able to use as potential confounders. Our study limitations are the cross-sectional nature of the national study (which prohibits the making of causal inferences) and the self-reported nature of the SDB reports (which has the potential of nondifferential misclassification which would bias the association to the null).
In summary, this study shows that there is a positive association of SDB markers to high CRP in a national study of US adults, particularly for Mexican Americans and moderately for women. Prospective studies are needed to support the findings. These findings may be clinically useful in determining interventions to improve sleep and maintain lower levels of CRP.