Abstract
Background
It is uncertain to what extent high C-reactive protein (CRP) concentrations reflect the presence of inflammatory conditions in the community.
Methods
We evaluated 3782 Framingham participants (mean age 55 years; 52% women) free of baseline cardiovascular disease. Logistic regression models examined the prevalence of common inflammatory conditions by CRP categories whereas a separate matched case-referent analysis evaluated the prevalence of uncommon inflammatory conditions. Cox models were used to assess the influence of common inflammatory conditions on relations between CRP and incident cardiovascular disease.
Results
Common inflammatory conditions were reported by nearly half of the participants; these individuals were more likely to have markedly-high CRP concentrations (>10mg/L, P for trend=0.001). In multivariable models, there were increased odds of having at least one common inflammatory condition with CRP concentrations of 1–3.0, 3.01–10, and >10mg/L, compared to the referent category (<1mg/L); the respective odds ratios with 95% confidence intervals were 1.41 (1.07–1.86), 1.45 (1.07–2.98) and 1.64 (1.09–2.47) in men, and 1.08 (0.82–1.43), 1.07 (0.80–1.44) and 1.38 (0.97–1.96) in women. In case-referent analyses, uncommon inflammatory conditions were more common in individuals with CRP >10mg/L compared to those with CRP <1mg/L (12.1% versus 6.6%; P=0.0001). In multivariable models, higher CRP categories were not associated with incident cardiovascular disease, and with additional adjustment for inflammatory conditions, results remained unchanged.
Conclusion
There is high prevalence of common and uncommon inflammatory conditions in individuals with high CRP concentrations. Higher CRP concentrations should be interpreted with caution in cardiovascular disease risk assessment.
Keywords: C-reactive protein, Inflammation, Arthritis, Cardiovascular Disease, Cohort Study
INTRODUCTION
C-reactive protein (CRP) is a non-specific marker of inflammation and a predictor of incident coronary heart disease,1 cardiovascular disease,2 and subclinical vascular disease.3,4 Available evidence suggests that moderately-high CRP concentrations (3–10mg/L) predict increased risk of myocardial infarction5 and stroke,6 even though it remains controversial if CRP provides significant prognostic insights over and above other traditional risk factors.7,8 Nevertheless, in some studies, markedly-high CRP concentrations (>10mg/L) have also been shown to predict cardiovascular risk.9
Several factors are known to influence the CRP concentrations including body mass index (BMI),10 diabetes,10 hypertension,11 cigarette smoking,12 estrogen replacement therapy,13 cholesterol concentrations,2 and genetic polymorphisms in the CRP gene.14,15 There are large intra-individual variations in circulating CRP concentrations,16 as well as differences by ethnicity and gender.17 Prior studies have documented associations of high CRP concentrations with numerous inflammatory conditions, such as rheumatoid arthritis,18,19 or chronic lung conditions20 as recently reviewed.21 CRP, an acute phase reactant, is also known to increase in acute inflammatory states,22–24 such as acute viral infections.22 However, data are sparse regarding the extent to which inflammatory conditions are associated with circulating concentrations of CRP in the community.
High-sensitivity CRP has been proposed for use in cardiac risk stratification and preventive decision making.5,25 Therefore, to address the influence of inflammatory conditions on CRP concentrations in the community, we examined the relations between higher concentrations of CRP with common and uncommon inflammatory diseases in the Framingham Offspring Study participants. Additionally, we examined the impact of common inflammatory conditions on the relation between CRP concentrations and incident cardiovascular disease.
METHODS
Study Sample
The design and selection criteria for Framingham Offspring Study have been described previously.26 For primary analyses, we selected all participants who attended the fifth examination cycle (1991–1995) and had CRP concentrations measured (n=3782; 1973 women). We excluded participants with prevalent cardiovascular disease at baseline (n=365), resulting in an available sample of 3417 (1842 women) participants. This sample was used to assess the prevalence of common inflammatory conditions (as described below) cross-sectionally and to analyze the incidence of cardiovascular disease prospectively. For assessing the prevalence of uncommon inflammatory conditions (from the same examination cycle), we conducted a case-referent analysis matching 463 participants with CRP concentrations >10mg/L by age and smoking status to a referent group of 463 participants with CRP concentrations <1mg/L.
All participants signed the informed consent and the protocol for the study was approved by the Institutional Review Board of the Boston Medical Center.
Measurements of CRP
Blood samples for CRP were drawn from antecubital vein of individuals after an overnight fast. For examination cycle 5, CRP concentrations were measured by an enzymatic immunoassay (Hemagen Diagnostics, Inc.) as described previously.6 The coefficient of variation (CV) for replicate CRP measurements on 36 subjects was 0.86. CRP was used as a categorical predictor in our analyses, and the kappa statistic was 0.75 for CRP values in the fourth compared with first through third quartiles, as previously described.6 For examination 7 cycle participants, CRP was measured by Dade Behring BN100 nephelometer (Deerfield, Illinois), and the mean CV on 139 phantom replicates was 3.2%.
Ascertainment of Cardiovascular Disease and Risk Factors
At each examination visit, participants underwent a medical history, physical examination, anthropometry, and laboratory assessment of risk factors, as previously described.26 Individuals who reported smoking one or more cigarettes/day in the year of examination were considered current smokers.
Surveillance for cardiovascular events occurred during each examination cycle and adjudication occurred by an endpoint committee consisting of three investigators, as previously described.27 A cardiovascular event was defined as coronary heart disease (stable angina, coronary insufficiency and myocardial infarction), cerebrovascular disease (stroke and transient ischemic attack), intermittent claudication, or congestive heart failure. A family history of premature cardiovascular disease was present if a validated event occurred in father before the age of 55 or in mother before age 65, as previously described.28
Ascertainment of Common and Uncommon Inflammatory Conditions
Information on common inflammatory conditions was obtained from the database collected during the clinic visit on a structured questionnaire administered at each examination cycle by physician-examiners blinded to CRP concentrations. Participants were classified as having a common inflammatory condition if they answered ‘yes’ during the interview by a physician-examiner regarding any one of the following conditions: chronic urinary tract diseases (urinary diseases); chronic gastro-duodenal ulcer (i.e. gastro-duodenal diseases); chronic cough, chronic bronchitis, asthma (i.e. pulmonary diseases); rheumatoid arthritis, gout, degenerative joint disease (i.e. musculoskeletal diseases); or any cancer (excluding non-melanoma skin cancers). These conditions were further categorized according to body system/type— urinary, gastro-duodenal, pulmonary, musculoskeletal, and cancers as displayed in Table 3.
Table 3.
Prevalence of Common Inflammatory Conditions According to Body System/Type
| Body System* | Men n=1575 | Women n=1842 |
|---|---|---|
| Urinary tract disease | 83 (4.8) | 90 (4.7) |
| Gastro-duodenal tract disease | 166 (9.5) | 117 (6.1) |
| Pulmonary disease | 248 (14.2) | 305 (15.8) |
| Musculoskeletal disease | 25 (1.4) | 23 (1.2) |
| Cancers | 242 (13.2) | 392 (19.8) |
Values are numbers or percentages (in parentheses) of participants with inflammatory conditions.
Diseases refer to common chronic inflammatory conditions and have been grouped according to body systems/type for simplicity and display. See methods for description of each body system.
Each participant can have inflammatory condition for more than one body system/type.
Information about uncommon inflammatory conditions was elicited during the participants’ interview and physical examination by the physician-examiner and noted in text form on the questionnaire. Information about these uncommon inflammatory conditions was not recorded in the computerized database and could only be obtained by retrospective review of the participants’ medical charts (heart study charts). Therefore, to assess the prevalence of uncommon inflammatory conditions, we selected a total of 926 participants by matching all participants in the markedly-high CRP category (>10mg/L; n=463) for age and smoking status to equal number of referent participants with low concentrations of CRP (<1mg/L). Before conducting the chart review, a separate pre-specified list of uncommon inflammatory conditions was made after reviewing the literature for conditions associated with high concentrations of CRP29 and by consensus of 2 cardiologists and a rheumatologist (Appendix Table 1). All charts were reviewed by a single physician blinded to CRP concentration and case-referent status. Charts were then screened for the pre-specified inflammatory conditions (e.g., connective tissue disorders, inflammatory bowel disease, as listed in Table 5) for the fifth, sixth or the seventh examination cycle visits. For intra-observer reproducibility of chart review, we re-reviewed 20 randomly selected charts and found high concordance (>95%) of results.
Appendix Table 1.
Coding Sheet Used During Chart Extraction of Uncommon Inflammatory Diagnosis
|
For ascertainment of uncommon inflammatory conditions we reviewed charts for a wide range of potential diseases but some of the diagnoses were never reported in the chart review.
Acute conditions, acute injuries or recent surgeries were considered if the same happened within one month of examination date.
Table 5.
Number of Participants with Uncommon Inflammatory Conditions in Cases and Referents
| Inflammatory Conditions | Cases (CRP>10 mg/L) N=398 | Referents (CRP <1mg/L) N=424 |
|---|---|---|
| Connective tissue disorders | 9 | 2 |
| Systemic Lupus Erythematosis | 3 | 0 |
| Polymyalgia Rheumatica | 2 | 0 |
| Psoriasis and psoriatic arthritis | 7 | 5 |
| Vasculities | 1 | 0 |
| Inflammatory bowel diseases | 8 | 4 |
| Sarcoidosis | 1 | 0 |
| Acute Infections* | 12 | 13 |
| Recent surgical procedures or injuries* | 4 | 4 |
|
| ||
| Total† | 46 | 28 |
Within previous month of the visit.
Chi square p=0.012 comparing total cases (46/398) versus total referents (28/424).
Out of 463 cases and referents, 65 cases and 39 referents were excluded because of prevalent CVD.
Values in the table are the number of participants with inflammatory conditions
Statistical Analyses
Sex-specific baseline characteristics of participants were assessed according to presence or absence of common inflammatory conditions.
Prevalence of Common Inflammatory Conditions
In sex-specific analyses prevalence of common inflammatory conditions was calculated according to the clinically recommended CRP categories {low (<1mg/L), borderline (1–3mg/L) moderately-high (3.01–10mg/L) and markedly-high concentrations (>10mg/L)}.5,9
Cross-sectional Analyses of CRP to Common Inflammatory Conditions
Sex-specific multiple logistic regression models30 were constructed to examine the relations of CRP concentrations in categories, to the presence or absence of common inflammatory conditions (dichotomous variable). The following sex-specific models were constructed hierarchically: (a) age-adjusted models; (b) adjusting for age, BMI, systolic blood pressure, antihypertensive medications, diabetes mellitus, total/HDL cholesterol, smoking status and estrogen replacement therapy (in women). Additionally, we repeated these analyses using CRP concentrations in quartiles. We also investigated for interactions of the relations of CRP categories with common inflammatory conditions by age or by smoking status.
Prospective Analyses of CRP to Cardiovascular Disease Risk
Sex-specific Cox proportional hazard regression models were constructed to analyze the risk of incident cardiovascular disease on follow up, according to CRP concentrations in continuous models (log transformed to normalize the distribution) and by CRP categories (as above). All models were adjusted for covariates as described above (models a and b) with and without adjustment for presence of common inflammatory condition.
Case-referent Analyses of CRP to Uncommon Inflammatory Conditions
In the case-referent analysis, Cochran Mantel-Haenszel31 test and Chi-square test accounting for the matched design (age and smoking status) compared the odds of uncommon inflammatory conditions in cases (CRP >10mg/L) to referents (CRP <1mg/L). Further, sex-pooled multivariable conditional-logistic regression adjusting for all the covariates (model b, as above) was used to assess the odds of having an uncommon inflammatory condition in cases compared to referents.
RESULTS
Mean age of the participants was 55±10years (52% women; Table 1). Participants with prevalent inflammatory conditions were more likely to report current cigarette smoking. Blood pressure, total/HDL cholesterol ratio and prevalence of diabetes were slightly higher among men compared to women.
Table 1.
Baseline Characteristics of Study Participants
| Characteristic | Men (n=1575) | Women (n=1842) | ||
|---|---|---|---|---|
| Common Inflammatory Conditions | ||||
| Present n = 486 | Absent n =1089 | Present n = 568 | Absent n = 1274 | |
| Age, years | 57.1 ± 9.6 | 52.9 ± 9.6 | 56.6 ± 9.8 | 53.3 ± 9.6 |
| Body mass index, kg/m2 | 28.2 ± 4.4 | 28.1 ± 4.1 | 27.3 ± 5.6 | 26.4 ± 5.4 |
| Cigarette smoking, % | 29.0* | 15.5 | 26.9* | 15.5 |
| Diabetes mellitus, % | 12.3 | 8.2 | 7.7 | 5.3 |
| Systolic blood pressure, mmHg | 131±18 | 128±17 | 126±21 | 122±19 |
| Diastolic blood pressure, mmHg | 77 ± 10 | 77 ± 9 | 73 ± 11 | 73 ± 10 |
| Hypertension drug therapy, % | 21.2 | 16.1 | 18.9 | 14.1 |
| Total cholesterol, mg/dL | 202.6±37.2 | 201.0±33.9 | 209.7±38.0 | 206.1±37.6 |
| Total/HDL cholesterol ratio | 5.1 ± 1.6 | 4.9 ± 1.5 | 4.1 ± 1.4 | 3.9 ±1.3 |
| Cholesterol drug therapy, % | 5.8 | 6.2 | 8.6 | 3.7 |
| Family history of premature cardiovascular disease†, % | 16.6 | 14.9 | 17.6 | 15.4 |
| Estrogen replacement therapy, % | NA | NA | 16.4 | 17.3 |
Chi-squared test p-value < 0.001 comparing presence versus absence of inflammatory conditions across CRP categories.
Total number of participants with available information = 1717 [men with (259) and without (582) and women with (278) and without (661) inflammatory conditions]
NA denotes not applicable
Values are mean ± 1 standard deviation.
Prevalence of Common Inflammatory Conditions
The prevalence of inflammatory conditions in participants according to CRP concentrations is presented in Table 2. Common inflammatory conditions were reported by 30.9% of men and 30.8% of women. Markedly-high CRP concentrations (>10mg/L) were present in 9.4% of men and 14.0% of women overall. Chronic inflammatory conditions were significantly less likely to be present in men and women with CRP concentrations <1 mg/L (P=0.001; Table 2). A test for linear trend showed increasing prevalence of inflammatory conditions from the lowest CRP concentration to the highest CRP concentration (P for both gender <0.001). The prevalence of common inflammatory conditions according to body system/type is displayed in Table 3. Pulmonary conditions and cancer were most common, followed by gastro-duodenal, urinary and musculoskeletal conditions.
Table 2.
Prevalence of Common Inflammatory Conditions According to CRP Category in Participants Free of Cardiovascular Disease
| CRP concentrations | Chronic Inflammatory Conditions | |||
|---|---|---|---|---|
| Men (n= 1575) | Women (n=1842) | |||
| Total (n) | Present (%) (n= 486) | Total (n) | Present (%) (n=568) | |
| 0–1mg/L | 649 | 23.4* | 719 | 26.2* |
| 1–3 mg/L | 417 | 33.6 | 422 | 30.1 |
| 3–10 mg/L | 361 | 36.3 | 444 | 33.6 |
| >10 mg/L | 148 | 42.6† | 257 | 40.5† |
| Trend P-value | 0.001 | 0.001 | ||
Chi-squared test p-value < 0.001 and
p-value = 0.005 when comparing those with presence of inflammatory conditions to those without inflammatory conditions in each CRP category (e.g. men with CRP >10mg/L and having inflammatory conditions were compared to men with CRP >10mg/L but without inflammatory conditions)
Values are numbers or percentages of individuals with presence of inflammatory conditions.
Cross-sectional Relations of CRP to Common Inflammatory Conditions
In sex-specific age-adjusted models, the probability of a reported inflammatory condition in individuals with markedly-high CRP concentrations was significantly higher in both men and women, compared to those in the lowest CRP concentration category (Table 4). After further adjustment for traditional cardiovascular disease risk factors, the association remained statistically significant in men and marginally significant in women (Table 4).
Table 4.
Odds of Inflammatory Conditions According to CRP concentrations in Men and Women Free of Cardiovascular Diseases
| CRP† | Models | |
|---|---|---|
| Age-adjusted Odds ratio (95% CI) | Multivariable-adjusted* Odds ratio (95% CI) | |
| CRP Category, mg/L | Men | |
| < 1 | Referent | Referent |
| 1 – 3.0 | 1.55 (1.17–2.04) | 1.41 (1.07–1.86) |
| 3.01 – 10 | 1.64 (1.23–2.18) | 1.45 (1.07–1.98) |
| > 10 | 2.18 (1.49–3.19) | 1.64 (1.09–2.47) |
|
|
||
| Women | ||
| < 1 | Referent | Referent |
| 1 – 3.0 | 1.12 (0.85–1.46) | 1.08 (0.82–1.43) |
| 3.01 – 10 | 1.23 (0.94–1.60) | 1.07 (0.80–1.44) |
| > 10 | 1.69 (1.24–2.29) | 1.38 (0.97–1.96) |
Multivariable models are adjusted for age, body mass index, systolic blood pressure, hypertension treatment, diabetes mellitus, total cholesterol, smoking status, and estrogen replacement therapy (in women).
CI denotes confidence interval
Compared with the lowest (first) quartile, the multivariable odds of having an inflammatory condition were 1.00 (95% CI, 0.72–1.40), 1.29 (95% CI, 0.93–1.80) and 1.60 (95% CI, 1.14–2.25) for the second (0.32 – 1.41mg/L), third (1.42 – 4.23mg/L) and fourth (≥4.24mg/L) quartiles, respectively in men. In women the odds ratios were 1.14 (95% CI, 0.85–1.55), 1.27 (95% 0.93–1.73) and 1.31 (95% CI, 0.93–1.83), for the second (0.28 – 1.70mg/L), third (1.71 – 5.70mg/L) and fourth (≥5.71mg/L) quartiles, respectively.
Similar results were noted when these analyses were conducted by CRP quartiles rather than CRP categories (see footnote Table 4). There was no evidence of effect modification by age or smoking status for the relations of CRP to inflammatory conditions in our primary cross-sectional analyses.
Prospective Relations of CRP to Cardiovascular Disease Risk
During follow-up (mean 8 years, range 4.6–10.1 years), 347 participants developed cardiovascular events. In sex-pooled Cox models with CRP modeled as a continuous variable, increase in log CRP concentration (each log CRP= 0.27mg/L) was associated with a higher incidence of cardiovascular disease on follow up (hazard ratio [HR] 1.15, 95% CI, 1.04–1.28); however, with additional adjustment for inflammatory conditions these relations became non-significant (HR 1.20, 95% CI, 0.96–1.50). In sex-specific multivariable Cox models, adjusted for traditional risk factors, the hazard ratios for incident cardiovascular disease associated with a CRP level >10 mg/L compared with the referent (<1 mg/L) were 1.31 (95% CI, 0.81–2.10) for men and 1.74 (95% CI, 0.97–3.13) for women, respectively, as previously reported.8 With additional adjustment for inflammatory conditions, there was no significant increased risk for cardiovascular disease associated with CRP (Appendix Table 2). Further, in a subset analysis of individuals with available information on family history of premature myocardial infarction (used as a covariate), these relations were not significantly altered (data not shown). We did not find evidence of effect modification by presence or absence of inflammatory condition in the prospective models relating CRP with cardiovascular disease.
Appendix Table 2.
Cox Proportional Hazards Models (Sex-Specific) Examining the Relations of C - reactive protein to the Incidence of Cardiovascular Disease With and Without Adjustment for Common Inflammatory Conditions
| CRP | Models | |||||||
|---|---|---|---|---|---|---|---|---|
| Without adjustment for inflammatory conditions | Additionally adjusted for inflammatory conditions | |||||||
| Age-adjusted | Multivariable† | Age-adjusted | Multivariable† | |||||
| CRP Categories, mg/L | OR (95% CI) | P-value | OR (95% CI) | P-value | OR (95% CI) | P-value | OR (95% CI) | P-value |
| Men | ||||||||
| < 1 | Referent | Referent | Referent | Referent | ||||
| 1 – 3.0 | 1.28 (0.89–1.84) | 0.19 | 1.12 (0.77–1.62) | 0.56 | 1.25(0.87–1.80) | 0.23 | 1.10 (0.76–1.60) | 0.61 |
| 3.01 – 10 | 1.76 (1.25–2.48) | 0.0013 | 1.37 (0.95–1.98) | 0.09 | 1.72 (1.21–2.43) | 0.002 | 1.35 (0.94–1.95) | 0.11 |
| > 10 | 1.86 (1.19–2.90) | 0.0002 | 1.31 (0.81–2.10) | 0.27 | 1.80 (1.15–2.82) | 0.01 | 1.29 (0.80–2.07) | 0.30 |
| P for trend | <0.001 | 0.08 | <0.001 | 0.09 | ||||
|
| ||||||||
| Women | ||||||||
| < 1 | Referent | Referent | Referent | Referent | ||||
| 1 – 3.0 | 1.58 (0.94–2.67) | 0.08 | 1.36 (0.80–2.31) | 0.25 | 1.56 (0.93–2.63) | 0.09 | 1.18 (0.87–1.60) | 0.27 |
| 3.01 – 10 | 2.01 (1.24–3.25) | 0.004 | 1.31 (0.78–2.20) | 0.32 | 1.99 (1.23–3.21) | 0.005 | 1.31 (0.78–2.21) | 0.31 |
| > 10 | 3.04 (1.84–5.02) | <0.0001 | 1.74 (0.97–3.13) | 0.06 | 2.92 (1.77–4.84) | <0.0001 | 1.72 (0.96–3.10) | 0.07 |
| P for trend | <0.001 | 0.09 | <0.001 | 0.10 | ||||
Multivariable models are adjusted for age, sex, BMI, diabetes mellitus, systolic blood pressure, treatment for hypertension, estrogen use (in women) and inflammatory conditions (as indicated).
High versus Low CRP Concentrations to Uncommon Inflammatory Conditions
The presence of at least one uncommon inflammatory condition was recorded by the physician examiner in the participants’ chart for 11% (52/463) of individuals with markedly-high CRP (>10mg/L) compared with 6% (29/463) of referents with CRP <1mg/L (p =0.008; Table 5). In logistic regression analyses after excluding participants with prevalent cardiovascular disease (65 individuals with a markedly-high CRP and 39 subjects with low CRP concentrations), there was a higher odds of having at least one uncommon inflammatory condition in individuals with markedly-high versus low concentrations of CRP (multivariable-adjusted OR 2.39, 95% CI, 1.26–4.54).
Secondary analyses
At a later examination using a high sensitivity CRP assay, the prevalence of common inflammatory conditions according to body system/type (Appendix Table 3) and in the markedly-high CRP category (>10.0 mg/L) (Appendix Table 4) was higher in both men and women. Overall, these results were very similar to the relations seen using the Hemagen assay in examination cycle 5, such that individuals with CRP>10 mg/L had a significantly higher prevalence of inflammatory conditions, and conversely, those with low CRP concentrations had a significantly lower prevalence.
Appendix Table 3.
Prevalence of Inflammatory Conditions According to Body System/Type Among Participants Free of Cardiovascular Disease who Attended a Later Examination Cycle 7
| Body System | Men (n=1264) | Women (n=1617) |
|---|---|---|
| Urinary Tract Disease* | NA* | NA* |
| Gastro-Duodenal Tract Disease | 140 (9.1) | 122 (6.9) |
| Pulmonary Disease | 410 (26.7) | 483 (27.3) |
| Musculoskeletal Disease | 311 (20.3) | 359 (20.3) |
| Cancer | 312 (20.3) | 362 (20.5) |
Information about urinary tract diseases was not collected during this examination
Values are numbers or percentages (in parentheses) of participants with inflammatory conditions
Diseases refer to common chronic inflammatory conditions and have been grouped according to body systems for simplicity and display. See text for description of each body system
Appendix Table 4.
Prevalence of Inflammatory Conditions According to CRP Concentrations in Participants Free of Cardiovascular Disease who Attended a Later Examination Cycle 7
| CRP concentrations | Chronic Inflammatory Conditions | |||
|---|---|---|---|---|
| Men (n= 1264) | Women (n=1617) | |||
| Total (n) | Present (%) (n= 650) | Total (n) | Present (%) (n=741) | |
| (0–1mg/L) | 351 | 46.4 | 374 | 46.5 |
| (1–3 mg/L) | 511 | 48.5 | 515 | 51.7 |
| (3–10 mg/L) | 337 | 58.5 | 570 | 56.0 |
| (>10 mg/L) | 65 | 63.1 | 158 | 70.1 |
| Trend P-value | <0.001 | <0.001 | ||
Values are numbers or percentages of individuals with presence of inflammatory conditions.
DISCUSSION
In our sample of middle-aged to elderly men and women, markedly-high CRP concentrations (>10mg/L) were not uncommon (9.4% of men and 14.0% women). At least one chronic inflammatory condition was reported by nearly half of men and women who were free of prevalent cardiovascular disease. The prevalence of common inflammatory conditions increased with increasing concentrations of CRP. Conversely, participants with low CRP concentrations (<1 mg/L) were less likely to report a common chronic inflammatory condition. In prospective multivariable-adjusted analyses for incident cardiovascular disease, further adjustment for common inflammatory conditions did not alter the magnitude or direction of the relations between CRP concentrations and incident cardiovascular disease. Finally, in case-referent analyses, we documented that uncommon inflammatory conditions occur infrequently, as expected, but tend to occur more commonly in those with markedly-high CRP concentrations. To our knowledge, this is the first comprehensive study of inflammatory conditions related to high CRP concentrations in a community-based sample.
An AHA/CDC statement (2002)5 and a more recent CDC/AHA workshop statement (2004) suggested clinically relevant CRP categories and emphasized the existence of considerable biological variability in circulating CRP concentrations.25 Intermediate to high concentrations of CRP clearly confer increased risks for myocardial infarction, stroke and other cardiovascular disease,2,6,9 though it is unclear whether and how to implement screening.32,33 At present, it remains uncertain how clinicians should interpret an elevated CRP concentration when assessing cardiovascular risk especially in presence of a chronic inflammatory condition. Considering large intra-individual variability in the concentrations of CRP,16,22 and apparently important differences by race,17 our data supports repeat testing for CRP in persons with high CRP concentrations. In addition, caution should be exercised if a chronic inflammatory condition is present or suspected. Indeed, the AHA/CDC statement (2002) recommended that persons with active infection or systemic inflammatory conditions should not be tested.5 A few studies have shown that CRP concentrations also predict the risk of death from cardiovascular disease in inflammatory arthritis patients.34,35 In this respect, some investigators even believe that CRP can be pro-inflammatory,36,37 whereas others disagree.38,39 An important objective of further research is to assess whether there is any utility, and particularly what is the risk/benefit ratio, of initiation of drug therapy in the setting of markedly-high values of CRP in persons with documented inflammatory conditions.
Our study has some potential limitations. First, inflammatory conditions were collected retrospectively from the research charts and we were not able to validate each diagnosis of inflammatory conditions reported by the participants, and some common inflammatory conditions may not have been ascertained. Thus, our data may underestimate the prevalence and acuity of inflammatory conditions, which would likely lead to a bias towards the null. We hope our study findings will motivate accurate, validated prospective collection of information regarding common and uncommon inflammatory conditions and the inter-relations with CRP in cardiovascular cohort studies. Second, due to the cross-sectional design of our study we cannot infer a causal relation between inflammatory conditions and elevated CRP concentrations. Third, majority of our participants are middle-aged to elderly whites of European descent, which may limit the generalizability of our findings to other ethnic/racial and age groups.
A final, important potential limitation is that various high-sensitivity CRP assays may differ in their reproducibility characteristics even if they may not differ in their predictive value.40 The coefficient of variation for the Hemagen assay is somewhat higher at low concentrations (CRP<1.00 mg/L), but it is within an acceptable range at higher (CRP>1.00 mg/L) concentrations.6 Moreover, the between-assay agreement as assessed by the kappa statistic was excellent for CRP categories used in our analyses.6 Reductions in reproducibility at low CRP concentrations would be expected to bias associations toward the null. To address potential limitations of the Hemagen CRP assay, we conducted a similar cross-sectional analysis of the association of CRP categories with inflammatory conditions during a later Offspring examination cycle 7 (1999–2002) using a widely used high-sensitivity CRP assay (Dade-Behring, nephelometric assay) and found very similar results (Appendix table 3 & 4). While our prospective analyses using the Hemagen assay do not suggest any significant impact of inflammatory conditions on associations between CRP and incident cardiovascular disease, these findings warrant confirmation in other prospective cohorts using the more widely used high-sensitivity CRP assay.
CONCLUSION
In our community-based sample, elevated and in particular markedly-high concentrations of CRP (>10mg/L) are associated with a significantly higher prevalence of common, as well as uncommon inflammatory conditions. Further research is warranted using comprehensive, validated ascertainment of inflammatory conditions and high sensitivity CRP assays to assess the impact of common inflammatory conditions on cardiovascular disease risk prediction. Our findings reinforce previous statements that the presence of active inflammatory conditions should be excluded before assessing an individuals’ risk of cardiovascular disease by using CRP concentrations.
Clinical Significance
We evaluated 3782 middle aged men and women free of cardiovascular disease for inflammatory conditions according to CRP categories: <1 (referent), 1–3.0, 3.01–10, and >10 mg/L.
There was a higher prevalence of common and uncommon inflammatory conditions in those with a high CRP concentration (>10mg/L) versus low CRP (<1mg/L).
High CRP levels warrant evaluation for evidence of an underlying inflammatory condition and consideration of repeat testing.
Acknowledgments
Supported by the NHLBI’s Framingham Heart Study contract NHLBI N01 HC 25195; R01 HL 073272 01, R01 HL 64753, and R01 HL 076784; and the Donald W. Reynolds Foundation Program in Clinical Cardiovascular Research.
Footnotes
Part of this submission was presented as a poster at the American Heart Association annual scientific sessions, New Orleans, LA on Nov 9, 2004
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Koenig W, Sund M, Frohlich M, et al. C-Reactive protein, a sensitive marker of inflammation, predicts future risk of coronary heart disease in initially healthy middle-aged men: results from the MONICA (Monitoring Trends and Determinants in Cardiovascular Disease) Augsburg Cohort Study, 1984 to 1992. Circulation. 1999;99:237–242. doi: 10.1161/01.cir.99.2.237. [DOI] [PubMed] [Google Scholar]
- 2.Ridker PM, Rifai N, Rose L, et al. Comparison of C-reactive protein and low-density lipoprotein cholesterol levels in the prediction of first cardiovascular events. N Engl J Med. 2002;347:1557–1565. doi: 10.1056/NEJMoa021993. [DOI] [PubMed] [Google Scholar]
- 3.Wang TJ, Nam BH, Wilson PW, et al. Association of C-reactive protein with carotid atherosclerosis in men and women: the Framingham Heart Study. Arterioscler Thromb Vasc Biol. 2002;22:1662–1667. doi: 10.1161/01.atv.0000034543.78801.69. [DOI] [PubMed] [Google Scholar]
- 4.Wang TJ, Larson MG, Levy D, et al. C-reactive protein is associated with subclinical epicardial coronary calcification in men and women: the Framingham Heart Study. Circulation. 2002;106:1189–1191. doi: 10.1161/01.cir.0000032135.98011.c4. [DOI] [PubMed] [Google Scholar]
- 5.Pearson TA, Mensah GA, Alexander RW, et al. Markers of inflammation and cardiovascular disease: application to clinical and public health practice: A statement for healthcare professionals from the Centers for Disease Control and Prevention and the American Heart Association. Circulation. 2003;107:499–511. doi: 10.1161/01.cir.0000052939.59093.45. [DOI] [PubMed] [Google Scholar]
- 6.Rost NS, Wolf PA, Kase CS, et al. Plasma concentration of C-reactive protein and risk of ischemic stroke and transient ischemic attack: the Framingham study. Stroke. 2001;32:2575–2579. doi: 10.1161/hs1101.098151. [DOI] [PubMed] [Google Scholar]
- 7.Danesh J, Wheeler JG, Hirschfield GM, et al. C-reactive protein and other circulating markers of inflammation in the prediction of coronary heart disease. N Engl J Med. 2004;350:1387–1397. doi: 10.1056/NEJMoa032804. [DOI] [PubMed] [Google Scholar]
- 8.Wilson PW, Nam BH, Pencina M, et al. C-reactive protein and risk of cardiovascular disease in men and women from the Framingham Heart Study. Arch Intern Med. 2005;165:2473–2478. doi: 10.1001/archinte.165.21.2473. [DOI] [PubMed] [Google Scholar]
- 9.Ridker PM, Cook N. Clinical usefulness of very high and very low levels of C-reactive protein across the full range of Framingham Risk Scores. Circulation. 2004;109:1955–1959. doi: 10.1161/01.CIR.0000125690.80303.A8. [DOI] [PubMed] [Google Scholar]
- 10.Ford ES. Body mass index, diabetes, and C-reactive protein among U.S. adults. Diabetes Care. 1999;22:1971–1977. doi: 10.2337/diacare.22.12.1971. [DOI] [PubMed] [Google Scholar]
- 11.Sesso HD, Buring JE, Rifai N, et al. C-reactive protein and the risk of developing hypertension. JAMA. 2003;290:2945–2951. doi: 10.1001/jama.290.22.2945. [DOI] [PubMed] [Google Scholar]
- 12.Bazzano LA, He J, Muntner P, et al. Relationship between cigarette smoking and novel risk factors for cardiovascular disease in the United States. Ann Intern Med. 2003;138:891–897. doi: 10.7326/0003-4819-138-11-200306030-00010. [DOI] [PubMed] [Google Scholar]
- 13.Ridker PM, Hennekens CH, Rifai N, et al. Hormone replacement therapy and increased plasma concentration of C-reactive protein. Circulation. 1999;100:713–716. doi: 10.1161/01.cir.100.7.713. [DOI] [PubMed] [Google Scholar]
- 14.Carlson CS, Aldred SF, Lee PK, et al. Polymorphisms within the C-reactive protein (CRP) promoter region are associated with plasma CRP levels. Am J Hum Genet. 2005;77:64–77. doi: 10.1086/431366. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Kathiresan S, Larson MG, Vasan RS, et al. Contribution of clinical correlates and 13 C-reactive protein gene polymorphisms to interindividual variability in serum C-reactive protein level. Circulation. 2006;113:1415–1423. doi: 10.1161/CIRCULATIONAHA.105.591271. [DOI] [PubMed] [Google Scholar]
- 16.Macy EM, Hayes TE, Tracy RP. Variability in the measurement of C-reactive protein in healthy subjects: implications for reference intervals and epidemiological applications. Clin Chem. 1997;43:52–58. [PubMed] [Google Scholar]
- 17.Khera A, McGuire DK, Murphy SA, et al. Race and gender differences in C-reactive protein levels. J Am Coll Cardiol. 2005;46:464–469. doi: 10.1016/j.jacc.2005.04.051. [DOI] [PubMed] [Google Scholar]
- 18.Guidelines for the management of rheumatoid arthritis: 2002 Update. Arthritis Rheum. 2002;46:328–346. doi: 10.1002/art.10148. [DOI] [PubMed] [Google Scholar]
- 19.Nielen MM, van Schaardenburg D, Reesink HW, et al. Increased levels of C-reactive protein in serum from blood donors before the onset of rheumatoid arthritis. Arthritis Rheum. 2004;50:2423–2427. doi: 10.1002/art.20431. [DOI] [PubMed] [Google Scholar]
- 20.Mannino DM, Ford ES, Redd SC. Obstructive and restrictive lung disease and markers of inflammation: data from the Third National Health and Nutrition Examination. Am J Med. 2003;114:758–762. doi: 10.1016/s0002-9343(03)00185-2. [DOI] [PubMed] [Google Scholar]
- 21.Kushner I, Rzewnicki D, Samols D. What does minor elevation of C-reactive protein signify? Am J Med. 2006;119:166–28. doi: 10.1016/j.amjmed.2005.06.057. [DOI] [PubMed] [Google Scholar]
- 22.Anderson NL, Anderson NG. The human plasma proteome: history, character, and diagnostic prospects. Mol Cell Proteomics. 2002;1:845–867. doi: 10.1074/mcp.r200007-mcp200. [DOI] [PubMed] [Google Scholar]
- 23.Pepys MB, Baltz ML. Acute phase proteins with special reference to C-reactive protein and related proteins (pentaxins) and serum amyloid A protein. Adv Immunol. 1983;34:141–212. doi: 10.1016/s0065-2776(08)60379-x. [DOI] [PubMed] [Google Scholar]
- 24.Zimmerman MA, Selzman CH, Cothren C, et al. Diagnostic implications of C-reactive protein. Arch Surg. 2003;138:220–224. doi: 10.1001/archsurg.138.2.220. [DOI] [PubMed] [Google Scholar]
- 25.Myers GL, Rifai N, Tracy RP, et al. CDC/AHA Workshop on Markers of Inflammation and Cardiovascular Disease: Application to Clinical and Public Health Practice: Report From the Laboratory Science Discussion Group. Circulation. 2004;110:e545–e549. doi: 10.1161/01.CIR.0000148980.87579.5E. [DOI] [PubMed] [Google Scholar]
- 26.Kannel WB, Feinleib M, McNamara PM, et al. An investigation of coronary heart disease in families. The Framingham offspring study. Am J Epidemiol. 1979;110:281–290. doi: 10.1093/oxfordjournals.aje.a112813. [DOI] [PubMed] [Google Scholar]
- 27.Kannel WB, Wolf PA, Garrison RJ, editors. Some risk factors related to the annual incidence of cardiovascular disease and death in pooled repeated biennial measurements: Framingham Heart Study, 30-year follow up. Bethesda, MD: National Heart, Lung and Blood Institute; Feb, 1987. The Framingham Study: an epidemiological investigation of cardiovascular disease. Section 34. NIH publication no. 87-2703. [Google Scholar]
- 28.Lloyd-Jones DM, Nam BH, D’Agostino RB, Sr, et al. Parental cardiovascular disease as a risk factor for cardiovascular disease in middle-aged adults: a prospective study of parents and offspring. JAMA. 2004;291:2204–2211. doi: 10.1001/jama.291.18.2204. [DOI] [PubMed] [Google Scholar]
- 29.Pepys MB, Hirschfield GM. C-reactive protein: a critical update. J Clin Invest. 2003;111:1805–1812. doi: 10.1172/JCI18921. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Hosmer DW, Lemeshow S. Applied Logistic Regression. New York, NY: John Wiley and Sons Inc; 1989. pp. 1–307. [Google Scholar]
- 31.Cochran WG. The Comparison of Percentages in Matched Samples. Biometrika. 1950;37:256–266. [PubMed] [Google Scholar]
- 32.Koenig W. C-Reactive Protein and Cardiovascular Risk: Has the Time Come for Screening the General Population? Clin Chem. 2001;47:9–10. [PubMed] [Google Scholar]
- 33.Mosca L. C-Reactive Protein -- To Screen or Not to Screen? N Engl J Med. 2002;347:1615–1617. doi: 10.1056/NEJMe020127. [DOI] [PubMed] [Google Scholar]
- 34.Goodson NJ, Symmons DP, Scott DG, et al. Baseline levels of C-reactive protein and prediction of death from cardiovascular disease in patients with inflammatory polyarthritis: a ten-year followup study of a primary care-based inception cohort. Arthritis Rheum. 2005;52:2293–2299. doi: 10.1002/art.21204. [DOI] [PubMed] [Google Scholar]
- 35.Pasceri V, Yeh ET. A tale of two diseases: atherosclerosis and rheumatoid arthritis. Circulation. 1999;100:2124–2126. doi: 10.1161/01.cir.100.21.2124. [DOI] [PubMed] [Google Scholar]
- 36.Cermak J, Key NS, Bach RR, et al. C-reactive protein induces human peripheral blood monocytes to synthesize tissue factor. Blood. 1993;82:513–520. [PubMed] [Google Scholar]
- 37.Nakagomi A, Freedman SB, Geczy CL. Interferon-{gamma} and Lipopolysaccharide Potentiate Monocyte Tissue Factor Induction by C-Reactive Protein : Relationship With Age, Sex, and Hormone Replacement Treatment. Circulation. 2000;101:1785–1791. doi: 10.1161/01.cir.101.15.1785. [DOI] [PubMed] [Google Scholar]
- 38.Hirschfield GM, Gallimore JR, Kahan MC, et al. Transgenic human C-reactive protein is not proatherogenic in apolipoprotein E-deficient mice. Proc Natl Acad Sci U S A. 2005;102:8309–8314. doi: 10.1073/pnas.0503202102. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Pepys MB, Hawkins PN, Kahan MC, et al. Proinflammatory effects of bacterial recombinant human C-reactive protein are caused by contamination with bacterial products, not by C-reactive protein itself. Circ Res. 2005;97:e97–103. doi: 10.1161/01.RES.0000193595.03608.08. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Clarke JL, Anderson JL, Carlquist JF, et al. Comparison of differing C-reactive protein assay methods and their impact on cardiovascular risk assessment. Am J Cardiol. 2005;95:155–158. doi: 10.1016/j.amjcard.2004.08.087. [DOI] [PubMed] [Google Scholar]