Interrelationships Among Circulating Interleukin-6, C-Reactive Protein, and Traditional Cardiovascular Risk Factors in Women
Objective— Interleukin-6 (IL-6) and C-reactive protein (CRP) are markers of systemic vascular inflammation that herald atherothrombosis and may have important interrelationships with traditional cardiovascular risk factors.
Methods and Results— We conducted a cross-sectional analysis among 340 apparently healthy women enrolled in the Women’s Health Study. In unadjusted analyses, higher levels of IL-6 and CRP were seen with increasing body mass index (BMI), systolic and diastolic blood pressure, and smoking exposure. IL-6 levels were related to the frequency of alcohol intake (P=0.002) and showed an inverse relationship with exercise frequency and hormone replacement therapy (P<0.0001 for both). CRP levels increased with hormone replacement therapy (P=0.0002). Associations among IL-6, CRP, and lipid levels were minimal. Overall, mean levels of IL-6 and CRP increased with increasing numbers of clinical risk factors (P<0.0001). In multivariate analyses, independent relationships were seen between levels of IL-6 and age, BMI, smoking, systolic blood pressure, alcohol use, presence of diabetes, and frequency of exercise. CRP was associated with age, BMI, systolic blood pressure, high density lipoprotein, smoking, and hormone replacement therapy in adjusted analyses.
Conclusions— Plasma levels of IL-6 and CRP are independently related to several clinical cardiovascular risk factors in women.
Accumulating evidence suggests that inflammatory processes, in part, mediate the development and progression of atherosclerosis.1,2⇓ In this regard, several circulating proinflammatory molecules have been associated with thrombotic cardiovascular events. Included among these are acute-phase proteins, cellular adhesion molecules, selectins, and cytokines.3 The stimuli activating or inciting the thrombotic process are protean but may be associated with various clinical cardiovascular risk factors.4
See page 1514
The plasma cytokine interleukin-6 (IL-6) plays an important role in mediating inflammation and is a central stimulus for the acute-phase response.5 In particular, IL-6 induces the hepatic synthesis of C-reactive protein (CRP), a known proinflammatory marker of atherothrombotic vascular disease. Although circulating levels of IL-6 and CRP are therefore physiologically linked, it remains unclear whether these markers of systemic vascular inflammation track with one another in respect to various traditional risk factors in healthy individuals. However, IL-6 and CRP have independently demonstrated associations with cardiovascular events and mortality among apparently healthy men and women.6–8⇓⇓
The interplay between the inflammatory process, cardiovascular risk factors, and atherothrombosis is complex.9 Despite evolving theories regarding IL-6 and CRP in the central mechanisms of inflammation and atherogensis,10 there exists a paucity of data coherently relating clinical cardiovascular risk factors with circulating levels of systemic vascular inflammatory markers in healthy individuals.
Given this, we sought to determine the relationship among circulating baseline IL-6, CRP, and clinical cardiovascular risk factors in a cohort of apparently healthy women. We further sought to determine any additive effects of these clinical risk factors on circulating concentrations of these inflammatory markers.
A cross-sectional analysis was performed in apparently healthy women enrolled in the prospective Women’s Health Study, a randomized primary prevention trial of aspirin and vitamin E, the details of which are described elsewhere.11 Detailed self-reported questionnaires containing information on the history of hypertension, smoking status, diabetes, parental history of myocardial infarction, use of hormone replacement therapy (HRT), alcohol use, frequency of exercise, and history of hypercholesterolemia were collected at baseline and used for analysis.
Blood samples were collected at baseline in EDTA tubes and stored on liquid nitrogen until analysis. Samples were assayed for IL-6 with the use of ELISA (R&D Systems). Samples were assayed for high-sensitivity-CRP on a BN II analyzer (Dade Behring). Total cholesterol and HDL cholesterol were assayed on a Hitachi 911 analyzer (Roche Diagnostics). Samples were randomly assayed by using a blinded technique.
Samples and clinical information were available at baseline for 340 women who were previously selected in a nested case-control study within the Women’s Health Study cohort and constituted the study population for this analysis.
Baseline clinical characteristics of the study population were calculated as means or proportions. Concentrations of IL-6 and high-sensitivity-CRP were skewed rightward, such that baseline levels were logarithmically normalized to allow for parametric analysis with the use of 1-way ANOVA. Geometric means were then calculated and reported. Linear regression models to account for multiple variables were created to assess for independent relationships between clinical variables and logarithmically transformed plasma inflammatory variables. All probability values were calculated by assuming a 2-tailed α value of 0.05 with CIs at the 95% level.
Baseline clinical characteristics of the study participants are shown in Table 1. Women with an average age of 60 years had a BMI of 26. Approximately 20% of the women had systolic hypertension (systolic blood pressure ≥140 mm Hg), 4% had diabetes, and 28% were current smokers. Mean levels for total and HDL cholesterol were 223 and 49 mg/dL, respectively.
Mean levels of IL-6 and CRP were significantly higher in women with several clinical cardiovascular risk factors. As shown in Table 2, compared with women without a history of hypertension, women with a history of hypertension had significantly higher levels of IL-6 and CRP. Similarly, IL-6 and CRP significantly increased with an increase in smoking exposure, and increasing IL-6 levels were associated with increasing exercise and the presence of diabetes. CRP levels were somewhat higher in women with a history of hypercholesterolemia. Last, the frequency of alcohol use was associated with a U-shaped relationship with plasma IL-6 levels (not shown).
IL-6 and CRP levels diverged in their association with HRT (Figure 1). IL-6 levels significantly increased among women who currently used, previously used, and never used HRT (P=0.01). In contrast, CRP levels significantly decreased among women who currently used, previously used, and never used HRT (P=0.0002).
Circulating levels of plasma inflammatory markers were analyzed in association with lipid parameters. Logarithmically normalized IL-6 and CRP minimally decreased with the HDL cholesterol level (parameter estimate −0.01, P<0.0001 for both analyses). IL-6 levels were not associated with total cholesterol levels (P=0.86), and CRP levels were only minimally associated with total cholesterol levels (P=0.03) in unadjusted analyses. However, the magnitude of these effects was quite small. Results were similar when examined among women not on HRT, with the exception that CRP was no longer associated with total cholesterol levels (P=0.21).
Circulating levels of IL-6 and CRP were evaluated in association with age and obesity. Levels of IL-6 linearly increased in association with age and BMI (P=0.0002 and P<0.0001, respectively). CRP, in a similar fashion, increased with increasing BMI (P<0.0001) but was not associated with age in univariate analyses (P=0.12).
Positive relationships were demonstrated between parameters of blood pressure, IL-6, and CRP levels (Figure 2). Systolic blood pressure was categorized into seven 10 mm Hg increments. Diastolic blood pressure was divided into the following categories: <65, 65 to 74, 75 to 84, 85 to 89, and ≥90 mm Hg. Increasing systolic blood pressure was positively associated with increasing levels of IL-6 and CRP (P<0.0001 and P=0.0004, respectively). Similarly, increasing diastolic blood pressure was associated with increasing IL-6 and CRP levels (P=0.005 for both). Results were virtually identical when analyses were performed among women without hypertension and not currently on HRT.
In an effort to examine the incremental relationships between these risk factors and vascular proinflammatory markers, concentrations of IL-6 and CRP were analyzed with increasing numbers of clinical factors (Figure 3). Circulating levels of IL-6 significantly increased with an increasing number of clinical cardiovascular risk factors (P<0.0001). In a similar fashion, CRP levels increased with an increasing number of clinical cardiovascular risk factors (P<0.0001).
Finally, to examine independent correlates of plasma inflammatory markers, adjusted analyses were performed by using linear regression models to account for the contributions of multiple cardiovascular risk factors (Table 3). Age and BMI remained positively associated with increasing IL-6 (P<0.0001 for both analyses). In addition, other independent relationships were seen between IL-6 and smoking status, systolic blood pressure, alcohol use, diabetes, and exercise frequency. Significantly higher plasma concentrations of IL-6 were found in women who currently smoked and in women with diabetes compared with women who never smoked and with nondiabetic women (P=0.008 and P=0.04, respectively). Lower levels of IL-6 were found in women who exercised or used alcohol at least weekly compared with women who never exercised or did not use alcohol (P=0.0008 and P=0.02, respectively). These analyses were adjusted for all other covariates, including lipid levels. Among this cohort of women, the regression model accounted for 34% of the variance in logarithmically normalized IL-6, with BMI alone accounting for 15% of the variance in concentrations of IL-6.
Similarly, CRP was independently associated with several traditional cardiovascular risk factors in adjusted analyses. BMI, systolic blood pressure, and smoking remained positively associated with CRP (P<0.0001, P=0.04, and P=0.01, respectively). In contrast to IL-6, CRP remained associated with HRT such that women currently on HRT had higher levels of CRP compared with other women after adjustment for other covariates (P<0.0001). Furthermore, HDL cholesterol remained inversely related to CRP levels (P=0.004). Finally, in contrast to the univariate analyses, age was now significantly associated with CRP levels in a manner similar to IL-6, which was likely the result of negative confounding of other covariates. The frequency of alcohol use and exercise, total cholesterol level, and the presence of diabetes were no longer associated with CRP levels after multivariate adjustment. The full regression model accounted for 34% of the variance in logarithmically normalized CRP, with BMI alone accounting for 14% of the variance in concentrations of CRP.
The present study demonstrates that several clinical cardiovascular risk factors are related to circulating plasma levels of IL-6 and CRP in women. Specifically, age, BMI, smoking status, systolic blood pressure, alcohol use, presence of diabetes, and exercise frequency are all independent correlates of circulating plasma concentrations of IL-6. We observed that age, HRT, BMI, systolic blood pressure, smoking status, and, to a lesser extent, HDL cholesterol are independently associated with circulating CRP levels. When examined in combination, these factors account for incremental levels of plasma IL-6 and CRP among women free of symptomatic cardiovascular disease.
These results are consistent with current evidence supporting the importance of inflammation in atherogensis.12 Similarly, these data suggest important relationships between inflammation and traditional cardiovascular risk factors; the data also suggest that plasma vascular inflammatory molecules may play an integral role in either the biological mechanisms of these cardiovascular risk factors or provide a marker of smoldering systemic vascular disease. Similar data from healthy males have previously suggested that CRP levels increase in proportion to the prevalence of risk factors.13
Circulating IL-6 concentrations may be the result of a variety of stimuli, including various clinical risk factors, from which several cell types may be induced to release IL-6, including smooth muscle cells and macrophage foam cells found in atheromatous plaques.14 For example, cigarette smoking may be related to elevated levels of IL-6, CRP, and other systemic markers of vascular inflammation.15,16⇓ In theory, circulating IL-6 levels may mediate a biological interplay between various immune responses and exert autocrine, paracrine, and endocrine responses to link clinical risk factors with plasma-based markers of inflammation, such as CRP, in a pathway toward atherothrombosis.17 For example, IL-6 can be released from adipocytes and may stimulate a response from the hypothalamic-pituitary-adrenal axis, resulting in hypertension, obesity, and/or an insulin resistance.18 Recent data from our group have demonstrated associations between parameters of blood pressure and IL-6 levels in healthy males.19 Furthermore, laboratory data suggest that IL-6 may mediate essential pathways in type 2 diabetes.20 This concept is echoed in our data by the significant contribution of BMI to plasma concentrations of IL-6 and CRP. Moreover, we have recently shown that elevated levels of IL-6 and CRP are associated with an increased risk of developing type 2 diabetes.21
IL-6 is a primary stimulant for the hepatic acute-phase response, with which other known proinflammatory risk factors (primarily CRP and fibrinogen) are associated.22 In fact, IL-6 is the only known cytokine capable of inducing all acute-phase proteins involved in the inflammatory response.23 As such, the induction of CRP by IL-6 may be one step in the pathobiology of atherothrombosis. Given this, it is plausible that other proinflammatory molecules are actually surrogates for circulating IL-6 activity.24
In fact, elevations in IL-6 itself have independent associations with cardiovascular events, even after control for CRP. For example, in the Physician’s Health Study, elevated levels of IL-6 in apparently healthy males predicted up to a 2.3-fold increase in myocardial infarction over 6 years. Among these men, the number of risk factors at baseline was correlated with the plasma concentration of IL-6.6 Baseline IL-6 levels have also been found to predict cardiovascular events in the Women’s Health Study.3 Furthermore, in another study, IL-6 levels had been shown to be a better predictor of mortality than CRP.8,25⇓
Importantly, however, our data suggest that circulating concentrations of IL-6 and CRP may not always track one another. Our results echo the results from previous analyses suggesting that CRP is elevated in women on HRT irrespective of HRT preparation and other cardiovascular risk factors.26 However, we observed diverging concentrations of IL-6 and CRP in association with HRT, suggesting that IL-6-independent pathways may exist to upregulate circulating concentrations of CRP. Laboratory data suggest that estrogens do not influence IL-6 production in rat vascular smooth muscle cells,27 further supporting alternative pathways for CRP production. In fact, recent clinical data have shown the discordant effects of HRT on IL-6, CRP, and tumor necrosis factor-α and further suggest the possibility of direct hepatic stimulation of CRP by HRT.28 Furthermore, despite data indicating that circulating IL-6 levels are similar to CRP levels,8 we also observed that several other clinical risk factors were discordantly related with either IL-6 or CRP, such as alcohol use and exercise. Therefore, although CRP and IL-6 are pathobiologically linked, our data indicate that divergent levels may be seen under certain conditions among apparently healthy women and suggest that IL-6-independent pathways may exist in the regulation of circulating levels of CRP.
Taken together, IL-6 and CRP herald systemic inflammation associated atherothrombosis. Previous data have demonstrated the importance of vascular inflammation in determining vascular risk over and above that of clinical hyperlipidemia.29 In the present study, IL-6 and CRP were associated with several traditional risk factors, independent of lipid levels, supporting a separate or concurrent biological pathway toward atherothrombosis. This nonlipid association with traditional risk factors provides support that IL-6 and CRP may have direct vascular effects on plaque stability.30 Therefore, it is plausible that therapies aimed at modulating IL-6 and CRP activity, such as peroxisome proliferator-activated receptor-γ agonists,31 ACE inhibitors,14 or 3-hydroxy-3-methylglutaryl coenzyme A32 reductase inhibitors, may afford cardioprotection in the future.
Several limitations of these data should be taken into account. Because of the cross-sectional design, relationships between IL-6, CRP, and risk factors cannot be deemed causal in nature. Additionally, it is possible that unmeasured variables may account for the relationships observed. Despite these limitations, inherent with any cross-sectional analysis, our observations are consistent with known previous reports and are supported by biological plausibility. Last, because our analysis consisted of a relatively older volunteer population of women, many of whom were postmenopausal, caution should be exercised when generalizing these data to other populations.
In summary, IL-6 and CRP are independently related to several clinical cardiovascular risk factors among apparently healthy women. Furthermore, although physiologically related, changes in circulating levels of CRP and IL-6 may not always track with one another regarding several traditional cardiovascular risk factors. Finally, although the relationship between inflammation, cardiovascular disease, and traditional risk factors is complex, IL-6 and CRP may be important mediators in the pathway toward atherothrombotic events.
This study was supported by grants HL-07575-15 and HL-08755 from the National Heart, Lung, and Blood Institute, Bethesda, Md. Dr Ridker was supported by a grant from the Doris Duke Foundation and an Established Investigator Award from the American Heart Association, Dallas, Tex.
Dr Ridker is a coinventor on patents filed by the Brigham and Women’s Hospital that relate to inflammatory biomarkers in cardiovascular disease.
Received May 15, 2002; revision accepted July 2, 2002.
- ↵Danesh J, Whincup P, Walker M, Lennon L, Thomson A, Appleby P, Gallimore JR, Pepys MB. Low grade inflammation and coronary heart disease: prospective study and updated meta-analyses. BMJ. 2000; 321: 199–204.
- ↵Ridker PM, Rifai N, Stampfer MJ, Hennekens CH. Plasma concentration of interleukin-6 and the risk of future myocardial infarction among apparently healthy men. Circulation. 2000; 101: 1767–1772.
- ↵Volpato S, Guralnik JM, Ferrucci L, Balfour J, Chaves P, Fried LP, Harris TB. Cardiovascular disease, interleukin-6, and risk of mortality in older women: the women’s health and aging study. Circulation. 2001; 103: 947–953.
- ↵Yudkin JS, Stehouwer CD, Emeis JJ, Coppack SW. C-reactive protein in healthy subjects: associations with obesity, insulin resistance, and endothelial dysfunction: a potential role for cytokines originating from adipose tissue? Arterioscler Thromb Vasc Biol. 1999; 19: 972–978.
- ↵Buring JE, Hennekens CH. The Women’s Health Study: summary of the study design. J Myocard Ischemia. 1992; 4: 27–29.
- ↵Woods A, Brull DJ, Humphries SE, Montgomery HE. Genetics of inflammation and risk of coronary artery disease: the central role of interleukin-6. Eur Heart J. 2000; 21: 1574–1583.
- ↵Schieffer B, Schieffer E, Hilfiker-Kleiner D, Hilfiker A, Kovanen PT, Kaartinen M, Nussberger J, Harringer W, Drexler H. Expression of angiotensin II and interleukin 6 in human coronary atherosclerotic plaques: potential implications for inflammation and plaque instability. Circulation. 2000; 101: 1372–1378.
- ↵Tracy RP PB, Macy E, Bovill EG, Cushman M, Cornell ES, Kuller LH. Lifetime smoking exposure affects the association of C-reactive protein with cardiovascular disease risk factors and subclinical disease in healthy elderly subjects. Arterioscler Thromb Vasc Biol. 1997; 17: 2167–2176.
- ↵Chae C, Lee R, Rifai N, Ridker PM. Blood pressure and inflammation in apparently healthy men. Hypertension. 2001; 38: 399–403.
- ↵Ridker PM, Hennekens CH, Rifai N, Buring JE, Manson JE. Hormone replacement therapy and increased plasma concentration of C-reactive protein. Circulation. 1999; 100: 713–716.
- ↵Ridker PM, Glynn RJ, Hennekens CH. C-reactive protein adds to the predictive value of total and HDL cholesterol in determining risk of first myocardial infarction. Circulation. 1998; 97: 2007–2011.
- ↵van der Wal AC, Becker AE, van der Loos CM, Das PK. Site of intimal rupture or erosion of thrombosed coronary atherosclerotic plaques is characterized by an inflammatory process irrespective of the dominant plaque morphology. Circulation. 1994; 89: 36–44.