Atherosclerosis development is a complex process, with inflammation, indicated by elevated high-sensitivity C-reactive protein (hs-CRP), as a potential mediator. Obesity, physical activity, and depression have all been reported to affect hs-CRP. However, these factors are interconnected, and their relative individual importance remains unclear. From a separate prospective cohort study, 289 patients were selected for the present substudy. We assessed the relation of a variety of potential predictors and hs-CRP. Obesity, physical activity, and depression, in addition to several other potential factors, were analyzed in bivariate and multivariate linear regression models, adjusting for potential confounders. In unadjusted analyses, mild-to-moderate and severe depression were associated with increased hs-CRP compared to no or minimal depression. Vigorous physical activity was associated with decreased hs-CRP compared to no physical activity. All classes of obesity were associated with increased hs-CRP. In addition, attaining a college or graduate degree was associated with decreased hs-CRP compared to high school or less educational attainment. On multivariate analysis, depression was no longer associated with increased hs-CRP. Physical activity remained associated with decreased hs-CRP but only at vigorous levels. Educational attainment also remained associated but only at the collegiate or professional education level. Ultimately, obesity remained the greatest absolute predictor of elevated hs-CRP. In conclusion, in analyses of multiple factors potentially predictive of elevated hs-CRP in a large population of patients with subclinical coronary heart disease, we found the most important predictor to be obesity.
Atherosclerosis development and progression is a multifactorial process, and inflammation has been postulated to be a component of this process. Increased inflammation, as indicated by elevated high-sensitivity C-reactive protein (hs-CRP), has been associated with worsened coronary heart disease (CHD) outcomes. In addition, several studies have demonstrated that many traditional risk factors for CHD have been associated with elevated hs-CRP. These include smoking, alcohol use, diabetes, obstructive sleep apnea, and rheumatoid arthritis. Physical inactivity and obesity have also been linked to elevated hs-CRP, although the relative importance is unclear. Several studies have shown that an increase in body mass index (BMI) is associated with an increase in hs-CRP. However, the role of physical activity in decreasing hs-CRP is less clearly understood. Additional studies have shown that depression and socioeconomic status are also associated with increased hs-CRP. These risk factors are clearly intertwined, but no previous study has considered the combined influence of physical inactivity, obesity, and depression on hs-CRP in a population of patients without a high incidence of CHD. In managing the epidemic of CHD, it is important to understand which factors are most influential. To investigate the relative influence of obesity, physical inactivity, and depression, we performed a secondary analysis of data from the Symptom Mitigation in AtRial fibrillaTion (SMART) study, a separate, single-center, prospective cohort study of atrial fibrillation (AF) symptoms and health outcomes in stable outpatients with AF.
Methods
Outpatients with documented AF were recruited through the electrophysiology clinics. Participants were eligible to participate if they had had ≥1 episode of AF not attributed to a reversible cause documented by electrocardiography or continuous looping monitoring. Patients were only excluded if they were a member of a vulnerable patient population (e.g., <18 years old, incarcerated) or were planning to move from the local area within 3 years.
From September 2008 to November 2011, a total of 400 participants were enrolled and completed a baseline questionnaire of their general health and well-being. The participants also underwent routine laboratory studies. Of the total participants in the SMART study, 289, with all predictor and outcomes variables available, were included for analysis in the present separate substudy. The appropriate institutional review boards approved the protocol, and all participants provided written informed consent. Our primary predictor variables were obesity, physical inactivity, and depression. At enrollment, each participant’s height and weight were measured, and the BMI was calculated (kg/m 2 ). The participants were grouped into 3 categories according to the BMI. Using the World Health Organization classification scheme, the participants were considered normal to overweight if their BMI was <30 kg/m 2 , class I obesity if the BMI was ≥30 to 35 kg/m 2 , or class II or III obesity, if the BMI was ≥35 kg/m 2 . Physical activity was determined by the question: “Which of the following best describes how much exercise you did last month, that is, have you exercised for 15 to 20 minutes of brisk walking, swimming, general conditioning, or recreational sports?” Possible answers included not at all active, a little active (1 to 2 times/mo), fairly active (3 to 4 times/mo), quite active (1 to 2 times/wk), very active (3 to 4 times/wk), or extremely active (≥5 times/wk). The participants were grouped into 3 categories according to the activity level. They were considered inactive if they answered “not at all active,” light or moderately active if they answered “a little active, fairly active, or quite active,” and vigorously active if they answered “very active or extremely active.” To assess depression, we administered the 9-item Patient Health Questionnaire (PHQ), which assesses the severity of depressive symptoms during the previous 2 weeks. Scores on this scale were categorized as representing no to minimal (PHQ score 0 to 3), mild to moderate (PHQ score 4 to 9), or severe (PHQ score ≥10) depressive symptoms. Our primary outcomes variable was hs-CRP, measured at enrollment in the outpatient clinics. Nonfasting blood samples were taken from the antecubital vein (Vitros 5600, Ortho-Clinical Diagnostics, Rochester, New York). An hs-CRP level >3 was considered high. Patients with an hs-CRP >15 were excluded from the analyses, because these patients were considered to have a systemic inflammatory illness. Age, gender, and medication use were assessed by electronic chart review at study enrollment. Ethnicity, medical co-morbidities, and smoking status were determined by questionnaire. All patients had a diagnosis of AF, but the pattern of AF was classified as persistent if the patients had continuous AF lasting >1 week or had ever received a cardioversion. Educational attainment was assessed by asking “What is the highest level of school that you have completed?” Possible answers included no formal schooling, fifth grade or less, sixth to eighth grade, ninth to eleventh grade, high school graduate or equivalent, some college, vocational school, or junior college, college degree (4-year), or graduate or professional degree. Those who answered no formal schooling, fifth grade or less, sixth to eighth grade, or ninth to eleventh grade were collapsed into the category less than high school. Serum creatinine was measured, and the estimated glomerular filtration rate was calculated using the Cockcroft-Gault formula.
We used univariate statistics to examine the mean, standard deviation, and shapes of distributions for the continuous variables and frequencies for categorical variables. No imputation of missing data was performed, given the low overall number of missing values. Next, we performed bivariate comparisons of the primary predictor variables, in addition to other potential predictor variables, with hs-CRP as the categorical outcome (normal vs high hs-CRP), using Student’s t test for continuous variables or chi-square analyses for categorical variables. We then analyzed the relation of our primary predictor variables, with hs-CRP as a continuous variable in the bivariate and multivariate linear regression models. Because hs-CRP had a skewed distribution, hs-CRP was log transformed to create a normal distribution before analysis. We used the partial F test to exclude potential confounders in the analyses. Our final multivariate model included all our primary predictor variables (i.e., obesity, physical activity, depression), with the addition of educational attainment. The statistical tests were 2-tailed, with p <0.05 considered significant. The analyses were performed with Stata, version 11 (StataCorp, College Station, Texas).
Results
Of the 400 participants recruited for the study, 359 (90%) completed the PHQ depression questionnaire, 372 (93%) completed the question on physical activity, 363 (91%) completed the question on educational attainment, 398 (99%) had a BMI measurement, 346 (87%) had a hs-CRP measurement, of whom, 28 (8%) had an hs-CRP >15. Therefore, 289 participants with all primary predictor and outcome variables included in the multivariate model were included in the present analysis. The participants in the study were older, most were men, and most were white ( Table 1 ). Hypertension, diabetes, and smoking were quite prevalent. However, only 14% of participants had a diagnosis of CHD.
| Characteristic | Overall (n = 289) | hs-CRP <3.0 mg/dl (n = 193) | hs-CRP ≥3.0 mg/dl (n = 96) | p Value |
|---|---|---|---|---|
| Age (yrs) | 61 | 61 | 62 | 0.340 |
| Men | 202 (70%) | 144 (75%) | 55 (57%) | 0.002 |
| Race | ||||
| White | 205 (71%) | 139 (72%) | 58 (60%) | 0.515 |
| Black | 20 (7%) | 10 (5%) | 9 (9%) | 0.113 |
| Hispanic | 46 (16%) | 35 (18%) | 14 (15%) | 0.742 |
| Hypertension | 162 (56%) | 106 (55%) | 61 (64%) | 0.150 |
| Diabetes mellitus | 52 (18%) | 27 (14%) | 27 (28%) | 0.003 |
| Coronary artery disease | 40 (14%) | 29 (15%) | 13 (14%) | 0.890 |
| Heart failure | 43 (15%) | 23 (12%) | 20 (21%) | 0.046 |
| Smoker | 26 (9%) | 17 (9%) | 11 (11%) | 0.599 |
| Medication | ||||
| β Blocker | 173 (60%) | 110 (57%) | 66 (69%) | 0.047 |
| Calcium-channel blocker | 66 (23%) | 46 (24%) | 20 (21%) | 0.495 |
| Angiotensin-converting enzyme inhibitor/angiotensin receptor blocker | 116 (40%) | 73 (38%) | 43 (45%) | 0.275 |
| Statin | 127 (44%) | 79 (41%) | 48 (50%) | 0.161 |
| Persistent atrial fibrillation | 165 (57%) | 112 (58%) | 54 (56%) | 0.832 |
| Average glomerular filtration rate (ml/min) | 112 | 111 | 115 | 0.514 |
| Education | 0.001 | |||
| Less than high school/high school graduate | 75 (26%) | 42 (22%) | 35 (36%) | |
| Some college | 95 (33%) | 56 (29%) | 35 (36%) | |
| College or graduate degree | 118 (41%) | 95 (49%) | 26 (27%) | |
| Body mass index (kg/m²) | 32 | 30 | 35 | 0 |
| Current depression (9-item Patient Health Questionnaire) | 5 | 5 | 6 | 0.012 |
| Average daily activity | 0.048 | |||
| None | 78 (27%) | 44 (23%) | 33 (34%) | |
| Mild to moderate | 136 (47%) | 89 (46%) | 44 (46%) | |
| Vigorous | 73 (25%) | 59 (30%) | 18 (19%) | |
In the bivariate analyses ( Table 1 ), men, those with diabetes, those with congestive heart failure, and those taking a β blocker were more likely to have a high hs-CRP level. Participants with a greater education level were less likely to have a high hs-CRP. Obese participants, more depressed participants, and more physically inactive participants were more likely to have high hs-CRP. No association was found between statin therapy and hs-CRP level.
In the unadjusted analyses ( Table 2 ), mild-to-moderate depression and severe depression were associated with increased hs-CRP compared to no or minimal depression (p <0.01 and p <0.001, respectively). Vigorous physical activity was associated with decreased hs-CRP compared to no physical activity (p <0.001). Both class I obesity and class II or III obesity were associated with increased hs-CRP compared to normal to overweight participants (p <0.01 and p <0.001, respectively). Attaining a college or graduate degree was associated with decreased hs-CRP (p <0.001).
| Variable | Patients (n) | Unadjusted Mean (95% CI) | Adjusted Mean (95% CI) ∗ | ||
|---|---|---|---|---|---|
| Change in hs-CRP ∗ | p Value | Change in hs-CRP ∗ | p Value | ||
| Depression | |||||
| None/minimal | 128 | Reference | Reference | ||
| Mild/moderate | 116 | 1.483 | 0.004 | 1.199 | 0.187 |
| Severe | 45 | 1.958 | 0 | 1.373 | 0.094 |
| Physical activity | |||||
| None | 83 | Reference | Reference | ||
| Light/moderate | 138 | −1.191 | 0.240 | −1.078 | 0.616 |
| Vigorous | 78 | −1.863 | 0 | −1.427 | 0.048 |
| Body mass index (kg/m 2 ) | |||||
| <30 | 156 | Reference | Reference | ||
| 30–35 | 76 | 1.519 | 0.006 | 1.27 | 0.113 |
| >35 | 86 | 2.426 | 0 | 2.20 | 0 |
| Education | |||||
| Less than high school/high school graduate | 80 | Reference | Reference | ||
| Some college | 91 | −1.249 | 0.172 | −1.189 | 0.270 |
| College or graduate degree | 121 | −1.841 | 0 | −1.441 | 0.017 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
