A prospective evaluation of C-reactive protein in the progression of carotid artery stenosis

Article Outline

• Abstract
• Methods
• Results
• Discussion
• Conclusion
• Author contributions
• Acknowledgment
• References
• Copyright

Objective

Our institution previously reported an association between elevated C-reactive protein (CRP) and carotid artery stenosis. Based on this finding, we sought to further evaluate the association of CRP levels with ultrasound progression of carotid artery stenosis, and/or clinical events.

Methods

A prospective observational study of patients evaluated for carotid artery stenosis was performed at a tertiary medical center from 2003-2007. Patients underwent serial lab draws for serum CRP, as well as serial duplex ultrasounds of their carotid bifurcations. Examinations were performed at 6-month intervals. Initial risk factors and CRP levels were evaluated with univariate statistics. Ultrasound progression of disease was evaluated with Kaplan-Meier curves and Cox regression analysis.

Results

During the study period, 271 patients completed study requirements with a mean follow-up of 37 (±6) months. Initial duplex examination revealed 114 (41%) of patients had 0% to 15%, 94 (35%) had 16% to 49%, and 63 (23%) had 50% to 79% stenosis of the carotid bifurcation. Sixty-three patients (23%) demonstrated progression of disease by ultrasound examination, 27 (10%) progressed to carotid endarterectomy, and three (1%) experienced a stroke during follow-up. Mean CRP levels were higher among patients that progressed on duplex examination (6.7 ± 1.28 vs 4.6 ± 0.4 mg/dl, P < .05). Kaplan-Meier analysis revealed a significant difference in freedom from progression of carotid artery disease for patients with 1^st and 3^rd quartile CRP levels (log-rank test P < .05). Adjusting for diabetes, hyperlipidemia, hypertension, coronary artery disease, aspirin or other anti-inflammatory uses, and statin therapy, 4^th quartile CRP was independently associated with disease progression (OR 1.8, 95% CI; 1.03-2.99, P < .05).

Conclusions

High CRP levels predict ultrasound progression of disease in patients with carotid artery stenosis. In addition, CRP levels may provide additional information to help guide ultimate therapy for evaluation and follow-up of patients with borderline lesions identified by duplex exam.

C-reactive protein (CRP) is a nonspecific serologic protein that is a consequence of inflammation and has become increasingly evaluated in the context of atherosclerotic coronary artery disease.¹, ², ³, ⁴ C-reactive protein has been deemed a marker of disease “activity,” and in response, the Center for Disease Control and the American Heart Association has added CRP as an adjunct to traditional cardiac risk factors for global coronary risk assessment.⁵ This recommendation was based on research that CRP levels independently identify subclinical coronary disease that will progress to coronary artery stenosis.² Furthermore, in patients with existing coronary artery disease, CRP is predictive of patients who will develop acute coronary syndromes and of those with vulnerable plaque characteristics identified during angiography.⁶

Likewise, studies have elucidated the role of inflammation in carotid artery stenosis, unstable carotid plaques, and ischemic strokes.⁷, ⁸, ⁹, ¹⁰ Inflammatory markers, to include CRP, correlate with intimal media thickness and are directly proportional to number of inflammatory cells in plaque specimens. The degree of inflammation as measured by serum biomarkers likely represents the activity of disease and may predict patients at risk for progression of atherosclerosis. Because a substantial proportion of carotid artery progression and incident ischemic strokes are largely unexplained by traditional risk factors, unique determinants are needed to determine those patients at high risk.¹¹ By extrapolating the relationship between CRP and coronary artery disease, it is foreseeable that CRP has the potential to identify patients with subclinical carotid artery stenosis most likely to progress over time. We sought to prospectively evaluate the temporal relationship of CRP levels and ultrasound progression of carotid artery stenosis, and/or clinical events.

Back to Article Outline

Methods 

This is an institutional review board-approved, prospective cohort study designed to (1) evaluate the potential relationship between CRP level and the progression of carotid artery stenosis and (2) to compare the strength of this relationship with traditional risk factors such as age, hypertension, hyperlipidemia, smoking, and diabetes. The initial phase of this protocol was to evaluate CRP levels with baseline carotid artery stenosis, and these results have been previously reported.¹², ¹³ This is the continuation of that protocol to further evaluate the association of CRP with progression of disease.

Participation and enrollment into this study was made available to all male and female patients, greater than 40 years of age, who were referred to the vascular surgery service for possible or known unilateral, bilateral, symptomatic, or asymptomatic carotid artery stenosis. In addition, patients were offered enrollment into the study as part of the annual health-risk screening fair for military retirees held at our institution. These patients were included in order to ensure there was a representative cohort of patients with minimal disease in order to follow their respective CRP levels. Generally, patients were evaluated in an outpatient setting for asymptomatic bruit, amaurosis fugax, syncope, transient ischemic attacks, or recent stroke. Patients with known disease were offered enrollment at their regularly scheduled visit or during evaluation for the development of symptoms. Patients were excluded for age less than 40 years of age, chronic infection, untreated malignant neoplasms, or steroid use. Patients were excluded from analysis that did not complete three follow-up visits to include the clinical examination, carotid duplex, and laboratory blood sampling. Patients who were deemed operative candidates for carotid endarterectomy on initial visit were also excluded since they were not able to undergo longitudinal evaluation.¹³

All eligible patients provided informed consent and were enrolled by the study research nurse. A standardized demographic and clinical history questionnaire was completed, and 6-month study visits were scheduled. Medication use was recorded by our research nurse. Patients in our medical system receive their medications through one central Department of Defense pharmacy. From our pharmacy, our clinic can cross-reference medications that have been dispensed to the patient to ensure compliance. Of the patients taking statin therapy, 94% of the patients were taking atorvastatin, either 20 mg or 40 mg.

All patients then underwent a vascular clinical evaluation by a staff vascular surgeon. Fasting high-sensitivity CRP and low density lipoprotein (LDL) levels were evaluated at each 6-month study visit. In addition, at each study visit, formal bilateral carotid duplex ultrasonography was performed on all patients by a registered vascular technologist in an Intersocietal Commission for the Accreditation of Vascular Laboratories certified vascular laboratory. The degree of internal carotid artery stenosis was determined on the basis of velocity criteria that has been validated yielding stenosis classes of none (0%-15%), mild (16%-49%), moderate (50%-79%), severe (80%-99%), near occlusive, and occluded.¹⁴, ¹⁵

Progression (A) of carotid artery stenosis was defined as an increase in classification of carotid artery stenosis by duplex ultrasound criteria. This liberal definition was utilized because our initial goal was to evaluate CRP as a potential screening tool to detect patients with early active disease. In addition, we excluded patients with less than three examinations. A modified analysis was performed defining progression (B) as only those patients that developed a lesion with ≥50% stenosis on duplex examination, or progression to a higher class if their baseline lesion was ≥50%. This second analysis was performed because a 50% stenosis has been deemed clinically significant and has previously determined the level of disease that is most likely to progress over time.¹⁶, ¹⁷

Individual CRP and LDL levels were calculated as means over the study period. Because mean CRP levels were not normally distributed secondary to outliers, they were categorized into quartiles. Age, tobacco use in pack-years, and mean LDL levels, highest CRP level, and CRP at time of progression fit a normal distribution and were analyzed as continuous variables for univariate and multivariate analysis. Clinical events (such as myocardial infarction, development of symptoms attributable to carotid stenosis, stroke, and death) were determined by patient follow-up visits and electronic medical records.

Progression of carotid artery stenosis was considered an outcome event and modeled utilizing the Kaplan-Meier method. Kaplan-Meier plots were carried out to the point in time at which standard error exceeded 10% of the value of the survival distribution (freedom from progression of carotid artery stenosis). Curves were analyzed utilizing the log-rank test, and significance was set at .05.

Cox regression proportional hazards modeling was utilized to determine the univariate impact of traditional atherosclerotic risk factors, baseline carotid artery stenosis, and CRP quartiles. Variables that demonstrated P < .10 were entered into multivariate analysis. Potential confounders such as aspirin use, 3-hydroxy-3-methylgutaryl coenzyme A reductase inhibitor (statin) use, and other anti-inflammatories (nonsteroidal anti-inflammatory drugs) were included in the multivariate model to minimize this potential source of bias that has been shown to reduce CRP levels. During modeling, variables were removed for P > .10, and were entered into the model if P < .05 for a maximum of 20 iterations. Values are reported as hazard ratios (HR) ±95% confidence interval with respective P values. Univariate and multivariate modeling were performed for dichotomous endpoints, progression (A) and progression (B). All statistical analysis was performed utilizing SPSS 15.0 (2003) (Chicago, Ill).

Back to Article Outline

Results 

Over the 3-year period of enrollment, 421 patients were enrolled into the study and completed the initial examination. Eighteen patients voluntarily withdrew from the study. Of the remaining 403 patients, 271 patients completed three or more study visits to include clinical visits, lab draws, and carotid duplex examination; this group comprised our study population for analysis. In regards to the patients that did not meet inclusion criteria, 109 (83%) were patients who were enrolled during our annual health screening fair. The remainder was excluded secondary to inadequate follow-up for analysis. Evaluating the demographics of the population not captured in this study, the average age was 66 (±7) years, 67% reported a smoking history, 62% were hyperlipidemic, 18% diabetic, and 68% hypertensive. Twenty-eight percent reported coronary artery disease, 6% transient ischemic attacks, and 3% infrainguinal occlusive disease. Eighty-six percent of patients not captured for longitudinal evaluation had initial carotid stenosis (CS) of 0% to 49%.

The baseline demographics and clinical risk factors for the cohort studied are listed in Table I. The population was predominately over the age of 60 years, and the majority of patients had a history of smoking, hypertension, and hyperlipidemia. Diabetes and coronary artery disease were identified in 27% and 35%, respectively. Forty-two percent of patients had no carotid artery stenosis at the beginning of the study; the remainder had either mild or moderate disease. There were no patients in the severe category. The mean follow-up period was 37 (±6) months, and the median number of follow-up visits per patient was six (interquartile range: 5-7). Sixty-three patients (23%) met criteria for progression (A), whereas 46 (17%) met criteria for progression (B).

Table I. Cohort demographic features, clinical risk factors, and baseline carotid artery stenosis

n = 271 patients
Mean age(y)	70(±8)
Male sex	134(49%)
Race, nonwhite	26(10%)
Risk factors:
Current smoking	26(10%)
Any smoking history	191(68%)
Hyperlipidemia	181(67%)
Diabetes	74(27%)
Hypertension	199(74%)
Coronary artery disease	95(35%)
Stroke or transient ischemic attacks	67(25%)
Infrainguinal occlusive disease	49(18%)
Medications:
Statin therapy	158(58%)
Aspirin use	190(70%)
Other anti-inflammatory medication	61(23%)
Baseline classification internal carotid artery stenosis
None(0%-15%)	114(42%)
Mild (15%-49%)	94(35%)
Moderate(50%-79%)	63(23%)
Severe(80%-99%)	—
Preocclusive or occluded	—

Because cohort mean CRP levels did not fit a normal distribution, we first evaluated mean levels by making a parametric transformation. Excluding values ±3 standard deviations from the mean, mean ±SD CRP levels were significantly higher among patients that progressed compared to those with stable carotid artery stenosis (6.7 ± 1.28 vs 4.6 ± 0.4 mg/dl, P < .05). In order to include patients with high mean CRP levels, we analyzed individual mean CRP as a categorical variable stratified into the following quartiles: 1^st quartile (0.26-1.59 mg/dl), 2^nd quartile (1.60-3.17 mg/dl), 3^rd quartile (3.19-6.04 mg/dl), and 4^th quartile (6.12-22.07 mg/dl). Utilizing Kaplan-Meier analysis, CRP quartiles were compared for progression A and for progression B. For both endpoints, there was no significant difference between quartiles 1to 3; however, there was a significant freedom from progression for both (A) and (B), Fig 1, Fig 2, respectively. For the remainder of the analysis, individual mean CRP levels were grouped as 1^st to 3^rd quartiles and 4^th quartile.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 1. 

  Kaplan-Meier method curves show probability of freedom from progression (A) as a function of time. Curves are the result of stratification of mean Hs-CRP levels into two groups: 1st-3rd quartiles and 4th quartile. Numbers below the figure denote the number of at risk patients for each subgroup. Progression (A) of carotid artery stenosis was defined as an increase in classification of carotid artery stenosis by duplex ultrasound criteria.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 2. 

  Kaplan-Meier method curves show probability of freedom from progression (B) as a function of time. Curves are the result of stratification of mean Hs-CRP levels into two groups: 1st-3rd quartiles and 4th quartile. Numbers below the figure denote the number of at risk patients for each subgroup. Progression (B) defined as only those patients that developed a carotid lesion with ≥50% stenosis on duplex examination, or progression to a higher class if their baseline lesion was ≥50%.

Our analytic strategy was to use univariate hazards modeling to identify variables within our cohort associated with progression of carotid artery stenosis (Table II). Of the traditional risk factors subjected to unadjusted analysis for progression (A), only history of hypertension (hazard ratio (HR), 2.22; 95% confidence interval (CI), 1.12-4.63; P < .05) and history of coronary artery disease (HR, 1.82; 95% CI, 1.11-2.98; P < .05) reached statistical significance, whereas diabetes and hyperlipidemia met threshold for inclusion in the multivariate model. Degree of baseline carotid artery stenosis did not correlate with progression and was not significant. Fourth quartile CRP (HR, 1.54; 95% CI, 1.06-2.60; P < .05) was significantly associated with progression (A).

Table II. Cox regression analysis utilizing a univariate model followed by multivariate modeling.

Variables	Endpoint of progression(A)a				Endpoint of progression(B)a
	Univariate model		Multivariate modelb		Univariate model		Multivariate modelb
	HRc	P	HRc	P	HRc	P	HRc	P
Age(y)d	1.01(0.98-1.04)	.26			1.03(0.49-1.1)	.06	1.03(0.99-1.07)	.06
Sex
Female	1.0e				1.0e
Male	1355(0-105)	.89			1435(0-106)	.92
Race
White	1.0e				1.0e
Non-white	1543(0-108)	.92			1112(0-107)	.95
Smoking	1.44(0.81-2.55)	.20			1.51(0.89-2.52)	.21
Tobacco (pck-y)d	1.0(0.99-1.02)	.18			1.0(.99-1.02)	.20
Diabetes	1.58(.94-2.66)	.08	1.24(0.73-2.12)	.42	1.9(1.08-3.53)	<.05	1.9(1.03-3.47)	<.05
HLP	1.69(0.95-3.01)	.08	0.99(0.98-1.0)	.38	2.47(1.15-5.29)	<.05	1.92(0.87-4.23)	.10
LDL(mg/dL)d	0.99(0.98-1.00)	.17			0.99(0.98-1.00)	.18
HTN	2.22(1.12-4.63)	<.05	2.0(0.98-4.12)	.06	3.13(1.23-7.93)	<.05	2.62(1.03-6.68)	<.05
CAD	1.82(1.11-2.98)	<.05	1.53(0.93-2.51)	.09	1.95(1.09-3.49)	<.05	1.45(0.8-2.64)	.22
PVD	1.32(0.72-2.40)	.35			1.31(0.65-2.62)	.45
Stroke or TIAs	1.41(0.68-2.93)	.36			1.48(0.85-3.24)	.27
Baseline CS
0%	1.0e				1.0e
1%-15%	103(0-106)	.90			102(0-1012)	.91
16%-49%	104(0-107)	.91			103(0-1014)	.89
50%-79%	103(0-106)	.90			103(0-1014)	.88
CRP quartilesf
1st-3rd	1.0e		1.0e		1.0e		1.0e
4th	1.54(1.06-2.60)	<.05	1.8(1.03-2.99)	<.05	1.61(1.01-2.95)	<.05	1.44(0.87-2.66)	.23

CS, Carotid stenosis; HR, hazard ratio; HLP, hyperlipidemia; LDL, low-density lipoprotein; HTN, hypertension; CAD, coronary artery disease; PVD, peripheral vascular disease(infrainguinal occlusive disease); TIA, transient ischemic attacks; Hs-CRP, high sensitivity C-reactive protein.

Because very little is known about population normal values for CRP levels, CRP was analyzed in multiple variations. Since patients had a median of six values, we were able to evaluate several temporal aspects of CRP that were not significant on univariate analysis (statistical values not presented in Table II). The absolute CRP level drawn at the time of progression was not associated with progression of disease by duplex examination. The percentage of CRP increase, peak CRP level during follow-up, and CRP variance were not associated with progression of disease.

In the multivariate Cox regression analysis adjusting for hypertension, coronary artery disease, diabetes, hyperlipidemia, aspirin or other anti-inflammatory agent, and statin use, 4^th quartile mean CRP level was independently associated with progression (A) (HR, 1.8; 95% CI, 1.03-2.99; P < .05). For progression (A), hypertension and coronary artery disease approached statistical significance, but none of the traditional risk factors imparted independent predictive value to explain the variance of progression (A).

In addition to evaluating the impact of statin therapy, we further evaluated the impact of inadequate statin therapy (defined as a mean LDL level of >100 mg/dl and on statin therapy) and the impact of potentially high-risk patients who were not appropriately on statin therapy (defined as a individual mean LDL level of >100 mg/dl and not taking a statin medication). The mean LDL for the entire cohort was 98 (±28) mg/dl. Evaluating those patients taking statin therapy, 111(71%) had mean LDL levels <100 mg/dl (effective statin therapy), and 46 (29%) had mean LDL levels >100 mg/dl (ineffective statin therapy). Ineffective statin therapy was tested as a potential covariate for Progression A (HR = 1.40, 95% CI; 0.77-2.52, P = .263) and was not significantly associated. In addition, there were 65 (60%) patients not taking a statin that had mean LDL levels of >100 mg/dl. This potentially high-risk cohort of patients was also tested as a potential covariate for Progression A (HR = 0.89, 95% CI; 0.64 to 1.80, P <.05) but was not significant.

A modified analysis was performed changing the dichotomous endpoint to progression (B) and repeating both the univariate and multivariate analysis (Table II). In the unadjusted analysis, diabetes (HR, 1.9; 95% CI, 1.08-3.53; P < .05), hyperlipidemia (HR, 2.47; 95% CI, 1.15-5.29; P < .05), hypertension (HR, 3.13; 95% CI, 1.23-7.93; P < .05), coronary artery disease (HR, 1.95; 95% CI, 1.09-3.49; P < .05), and 4^th quartile CRP (HR, 1.61; 95% CI, 1.01-2.95; P < .05) were all associated with progression (B). Paralleling the results of univariate analysis for progression (A), degree of baseline carotid artery stenosis was not associated with progression (B).

Performing the multivariate analysis adjusting for age, diabetes, hyperlipidemia, hypertension, coronary artery disease, aspirin or other anti-inflammatory agent, and statin use, 4^th quartile CRP level was not independently predictive of progression (B). Traditional risk factors accounted for all the variance in progression B. History of diabetes (HR, 1.9; 95% CI, 1.03-3.47; P < .05) and hypertension (HR, 2.62; 95% CI, 1.03-6.68; P < .05) were independently predictive of progression (B); age and coronary artery disease approached statistical significance.

In order to evaluate for unmeasured covariates that could potentially confound our results, we repeated the analysis with a 75% and 50% random sample of our cohort; the results of the univariate and multivariate analysis were unchanged. Although baseline carotid artery stenosis classification was not associated with progression in either analysis, we also repeated the analysis stratifying patients by those with baseline disease (defined as CS >50%) and those without disease (defined as CS <50%). Presence of baseline disease did not have a significant impact on disease progression in either analysis progression (A) or (B).

During the follow-up period, 27 (10%) carotid endarterectomies were performed, 2 (1%) developed symptoms attributable to CS, 3 (1%) experienced ischemic strokes, 18 (9%) experienced myocardial infarctions, and 10 (5%) deaths (all cause mortality) occurred. In regard to the two patients that developed symptoms, both patients progressed from 15% to 49% CS to a lesion greater than 70%, and one of the patients had a CRP level in the 4^th quartile. Of the patients that progressed to carotid endarterectomy, 25 (93%) were patients with asymptomatic disease that progressed to lesion of 80%. Eleven (41%) had a baseline CS of 15% to 49% with only two patients exhibiting 4^th quartile CRP levels, and 16 (59%) began with a baseline CS of 50% to 79% with 6 (38%) exhibiting 4^th quartile CRP levels. Because there were very few clinical events in this population, a composite endpoint (including stroke, myocardial infarction, and death) was evaluated, and there was no association of individual mean CRP level quartiles with these combined events (Fig 3). Analyzing the impact of 4^th quartile mean CRP on this composite outcome (HR, 0.46; 95% CI, 0.16-1.35; P = .16), there was not a significant association.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 3. 

  Kaplan-Meier method curves show probability of freedom from composite endpoint (myocardial infarction, stroke, and death) as a function of time. Four curves are the result of stratification of mean Hs-CRP levels into quartiles. Numbers below the figure denote the number of at risk patients for each subgroup.

Back to Article Outline

Discussion 

The first objective of this study was to examine the prospective relationship of CRP and progression in classification of carotid artery stenosis documented by duplex examination. To this endpoint, we found that there was no association with CRP at the time of duplex progression, percentage CRP increase, peak CRP value, or variance in CRP over the study period. Only 4^th quartile mean CRP levels (6.12-22.07 mg/dl) were significantly associated with carotid artery stenosis progression, irrespective of definition, progression (A) or (B).

Our second objective was to evaluate the strength of this relationship compared with traditional risk factors for carotid disease and degree of baseline carotid artery stenosis. In regard to progression (A), 4^th quartile means CRP exhibited a clear impact on progression with a HR of 1.8, and most notably, CRP was the only significant factor in the multivariate analysis. While traditional risk factors had HRs of similar magnitude, they did not reach statistical significance; only hypertension and coronary artery disease approached statistical significance in this cohort. In the evaluation of progression (B), 4^th quartile was again associated with progression in the univariate analysis, but utilizing this modified definition for progression to clinical disease >50%, more traditional risk factors (hypertension, coronary artery disease, diabetes, hyperlipidemia) were also significantly associated with progression. The independent association of 4^th quartile CRP was no longer apparent when adjusting for traditional risk factors; only two covariates, diabetes (HR = 1.9) and hypertension (HR = 2.6), were independently predictive of progression (B).

In this prospective cohort of patients followed over the course of 37(±6) months, persistently elevated individual mean CRP levels (4^th quartile mean CRP levels) had a compelling association with carotid artery stenosis progression. Interestingly, CRP increases at the specific time of duplex progression, highest CRP value recorded, change in CRP, and CRP variance were not predictive; however, persistently elevated mean CRP values where associated with progression. This suggests that there is a chronic systemic inflammatory process associated with progression, and acute fluctuations in CRP have little value in predicting progression of disease.

The independent impact of this progression was profound for our initial outcome measure, progression (A), and CRP was more predictive than traditional risk factors (hypertension, diabetes, hyperlipidemia, coronary artery disease, and smoking) that have been established for carotid disease progression. The chosen definition of progression (A) was utilized to detect patients with mild to moderate disease who were at risk for progression. This outcome would be suitable for screening programs and would potentially detect vulnerable patients at an early stage, which would then have implications for early medical management.

Our results reinforce published research linking CRP levels to carotid artery disease. CRP levels have been associated with baseline carotid artery stenosis and to be associated with varying plaque morphologic features.¹⁸ The degree of inflammatory cells (T lymphocytes, activated T lymphocytes, and macrophages) within plaque specimens is proportional to CRP levels;¹⁰ therefore, a systemic marker, CRP, is a plausible surrogate for local plaque activity.

Studies have addressed the utility of CRP in patients with asymptomatic carotid artery stenosis but with varying populations and contrasting results.¹⁹, ²⁰ In a large population based trial, Sander et al found CRP to directly correlate with progression in intima-media thickness (IMT) but also found age, hemoglobin A_1c, LDL level, hypertension, and body-mass index to predict progression defined as an increase in IMT.¹⁹ These differences were due to the larger population cohort and increased power to determine small statistical differences. In addition, the definition of progression was significantly different than that used in this study; they defined progression as an increase in IMT. Despite these differences, the magnitude of the HR for 4^th quartile CRP (HR, 1.97; 95% CI, 1.44-3.0; P < .05) for nondiabetic patients was not clinically different than that found in this study (HR, 1.8; 95% CI, 1.03-2.99; P < .05).

Schillinger et al²⁰ followed 1268 patients over a median follow-up of 7.5 months and found CRP at baseline and at time of follow-up to independently correlate with progression as defined as the increase of duplex criteria by at least one category (analogous to progression (A) in this study). They found age, smoking history, baseline degree of stenosis ≥70%, and CRP 4^th and 5^th quintiles to be independently associated with progression. Again, their cohort was much larger than in this study, and the population had approximately 10% of patients with baseline stenosis ≥70%, whereas our cohort had no patient with this level at disease at baseline. In addition, they found smoking to impact progression. Despite evaluating any smoking history and the number of pack-years, we were unable to confirm this relationship. Similar to our study, the highest CRP levels at baseline and at follow-up predict those patients likely to progress; this reinforces our conclusion that persistently elevated CRP levels are related to carotid stenosis progression.

There are several limitations that deserve mention. The patient population is comprised of both vascular surgery referrals and patients enrolled as part of a health fair screening process. Although we were unable to demonstrate any differences based on baseline carotid artery stenosis, there may be innate unmeasured differences in these two patient groups that could not be determined from this study. While the number of patients was adequate to statistically detect a difference in CRP levels for progression of disease, the potential for a type I and type II error exists. Traditional cardiac risk factors that reached univariate significance for progression (A) may have achieved statistical significance in the multivariate model with a larger sample size. Also in this regard, 4^th quartile CRP may have achieved statistical significance in the multivariate model for progression (B). The definitions of carotid artery progression rely on the ultrasound determination of carotid artery stenosis. There is a large amount of unmeasured variance in the interpretation of the duplex examination, the ultrasound equipment, local laboratory criteria, and the given examiner. We attempted to minimize this error by excluding patients that did not have more than three examinations, and all examinations were performed by two ultrasound technicians. However, in the absence of more precise imaging, some of these patients may have had carotid plaques too small to be detectable by duplex ultrasound; this represents uncontrolled variability.

In the present study, the results suggest the potential for future management strategies. The impetus of the study was to identify patients with asymptomatic disease that may be vulnerable to rapid progression, and thus, guide either follow-up strategies or medical management strategies. Duplex-imaging has been the gold standard for screening patients and following progression of disease in the vascular laboratory. Patients with persistently elevated mean CRP levels progress at a more rapid rate and could potentially benefit from a closer follow-up interval. Another area of developing research is the utility of magnetic resonance imaging (MRI) in determining carotid plaque morphology, and determining risk associated with morphologic changes.²¹ Zhao et al has demonstrated the modification of carotid plaque morphology measured by MRI utilizing lipid lowering agents.²² Future analyses would be needed to determine if CRP was an appropriate screening test to identify patients with asymptomatic carotid lesions that could benefit from MRI imaging and directed therapy.

Anti-inflammatory therapies have already been devised for coronary artery disease, and these agents are primarily aspirin and statin medications.²³ Both agents have been identified as having a lowering effect on CRP and their impact extends to a reduction in stroke rates.²⁴, ²⁵ The benefits are irrespective of the lipid-lowering effect and the antiplatelet impact of these agents. Blake et al has devised CRP screening regimens for patients without hyperlipidemia and treating patients with targeted statin therapy to an endpoint of CRP reduction and potential cardiac risk reduction.²⁵ It is foreseeable that the same benefits would be applicable to patients with asymptomatic carotid artery disease that have persistently elevated CRP levels. However, the benefits of pharmacologic CRP modification would need to be demonstrated in a prospective analysis to demonstrate an impact on carotid artery progression and clinical event rates.

Back to Article Outline

Conclusion 

This study confirms that there is a temporal relationship with persistently elevated C-reactive protein levels and the propensity for progression of carotid artery stenosis. The consequence of 4^th quartile mean CRP levels over time has an independent impact on carotid artery progression greater than the impact of traditional demographic and clinical risk factors in this cohort of patients with subclinical disease. In addition, the impact of CRP was independent of baseline carotid artery stenosis. These results suggest that serologic CRP levels are a marker for progression of disease and may have utility for the management of carotid artery disease and global carotid risk assessment. This may provide additional diagnostic value for evaluation and follow-up of patients with subclinical carotid artery stenosis identified by duplex examination.

Back to Article Outline

Author contributions 

Back to Article Outline

The authors thank Billi Tatum, RN, CRC, Leslie Schoneman, PA-C, Beverly Ciesinski, RVT, and John Dunsmoor, RVT for their expertise and assistance in helping complete this study.

Back to Article Outline

References 

1. Ridker PM. Inflammation, atherosclerosis, and cardiovascular risk: an epidemiologic view. Blood Coagul Fibrinolysis. 1999;10(Suppl 1):S9–S12
   • View In Article
2. Koenig W, Sund M, Frohlich M, Fischer HG, Lowel H, Doring A, 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
   • View In Article
3. Blake GJ, Ridker PM. Novel clinical markers of vascular wall inflammation. Circ Res. 2001;89:763–771
   • View In Article
   • CrossRef
4. Blake GJ, Ridker PM. C-reactive protein and other inflammatory risk markers in acute coronary syndromes. J Am Coll Cardiol. 2003;41(4 Suppl S):37S–42S
   • View In Article
   • MEDLINE
5. Pearson TA, Mensah GA, Alexander RW, Anderson JL, Cannon RO, Criqui M, 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
   • View In Article
   • CrossRef
6. Arroyo-Espliguero R, Avanzas P, Cosin-Sales J, Aldama G, Pizzi C, Kaski JC. C-reactive protein elevation and disease activity in patients with coronary artery disease. Eur Heart J. 2004;25:401–408
   • View In Article
   • MEDLINE
   • CrossRef
7. Lombardo A, Biasucci LM, Lanza GA, Coli S, Silvestri P, Cianflone D, et al. Inflammation as a possible link between coronary and carotid plaque instability. Circulation. 2004;109:3158–3163
   • View In Article
   • CrossRef
8. Cao JJ, Thach C, Manolio TA, Psaty BM, Kuller LH, Chaves PH, et al. C-reactive protein, carotid intima-media thickness, and incidence of ischemic stroke in the elderly: the Cardiovascular Health Study. Circulation. 2003;108:166–170
   • View In Article
   • CrossRef
9. Tanne D, Benderly M, Goldbourt U, Haim M, Tenenbaum A, Fisman EZ, et al. C-reactive protein as a predictor of incident ischemic stroke among patients with preexisting cardiovascular disease. Stroke. 2006;37:1720–1724
   • View In Article
   • CrossRef
10. Alvarez Garcia B, Ruiz C, Chacon P, Sabin JA, Matas M. High-sensitivity C-reactive protein in high-grade carotid stenosis: risk marker for unstable carotid plaque. J Vasc Surg. 2003;38:1018–1024
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (106 KB)
    • CrossRef
11. Ridker PM. High-sensitivity C-reactive protein, inflammation, and cardiovascular risk: from concept to clinical practice to clinical benefit. Am Heart J. 2004;148(1 Suppl):S19–S26
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (207 KB)
    • CrossRef
12. Mullenix PS, Steele SR, Martin MJ, Starnes BW, Andersen CA. C-reactive protein level and traditional vascular risk factors in the prediction of carotid stenosis. Arch Surg. 2007;(Accepted, in press)
    • View In Article
13. Mullenix P, Steele S, Martin M, Starnes B, Andersen C. Comparison of C-reactive protein and low-density lipoprotein cholesterol levels in the prediction of carotid stenosis. J Am Coll Surg. 2004;194(3S):S105
    • View In Article
14. Moneta GL, Edwards JM, Papanicolaou G, Hatsukami T, Taylor LM, Strandness DE, et al. Screening for asymptomatic internal carotid artery stenosis: duplex criteria for discriminating 60% to 99% stenosis. J Vasc Surg. 1995;21:989–994
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (695 KB)
    • CrossRef
15. Carpenter JP, Lexa FJ, Davis JT. Determination of sixty percent or greater carotid artery stenosis by duplex Doppler ultrasonography. J Vasc Surg. 1995;22:697–703discussion 703-5
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (788 KB)
    • CrossRef
16. Mansour MA, Littooy FN, Watson WC, Blumofe KA, Heilizer TJ, Steffen GF, et al. Outcome of moderate carotid artery stenosis in patients who are asymptomatic. J Vasc Surg. 1999;29:217–225discussion 225-7
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (145 KB)
    • CrossRef
17. Muluk SC, Muluk VS, Sugimoto H, Rhee RY, Trachtenberg J, Steed DL, et al. Progression of asymptomatic carotid stenosis: a natural history study in 1004 patients. J Vasc Surg. 1999;29:208–214discussion 214-6
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (100 KB)
    • CrossRef
18. Blackburn R, Giral P, Bruckert E, Andre JM, Gonbert S, Bernard M, et al. Elevated C-reactive protein constitutes an independent predictor of advanced carotid plaques in dyslipidemic subjects. Arterioscler Thromb Vasc Biol. 2001;21:1962–1968
    • View In Article
    • CrossRef
19. Sander D, Schulze-Horn C, Bickel H, Gnahn H, Bartels E, Conrad B. Combined effects of hemoglobin A1c and C-reactive protein on the progression of subclinical carotid atherosclerosis: the INVADE study. Stroke. 2006;37:351–357
    • View In Article
    • CrossRef
20. Schillinger M, Exner M, Mlekusch W, Sabeti S, Amighi J, Nikowitsch R, et al. Inflammation and Carotid Artery--Risk for Atherosclerosis Study (ICARAS). Circulation. 2005;111:2203–2209
    • View In Article
    • CrossRef
21. Phan BA, Chu B, Polissar N, Hatsukami TS, Yuan C, Zhao XQ. Association of high-density lipoprotein levels and carotid atherosclerotic plaque characteristics by magnetic resonance imaging. Int J Cardiovasc Imaging. 2007;23:337–342
    • View In Article
    • MEDLINE
    • CrossRef
22. Zhao XQ, Yuan C, Hatsukami TS, Frechette EH, Kang XJ, Maravilla KR, et al. Effects of prolonged intensive lipid-lowering therapy on the characteristics of carotid atherosclerotic plaques in vivo by MRI: a case-control study. Arterioscler Thromb Vasc Biol. 2001;21:1623–1629
    • View In Article
    • CrossRef
23. Albert MA, Danielson E, Rifai N, Ridker PM. Effect of statin therapy on C-reactive protein levels: the pravastatin inflammation/CRP evaluation (PRINCE): a randomized trial and cohort study. JAMA. 2001;286:64–70
    • View In Article
    • MEDLINE
    • CrossRef
24. Blake GJ, Ridker PM, Kuntz KM. Projected life-expectancy gains with statin therapy for individuals with elevated C-reactive protein levels. J Am Coll Cardiol. 2002;40:49–55
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (124 KB)
    • CrossRef
25. Blake GJ, Ridker PM, Kuntz KM. Potential cost-effectiveness of C-reactive protein screening followed by targeted statin therapy for the primary prevention of cardiovascular disease among patients without overt hyperlipidemia. Am J Med. 2003;114:485–494
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (292 KB)
    • CrossRef