I. Inflammation and Coronary Artery Disease: What every physician needs to know.

C-Reactive protein, inflammation, and cardiovascular disease

Traditional cardiovascular risk factors have been the bedrock for estimation of individual risk and selective use of prevention therapies.

  • Age

  • Hypertension

    Continue Reading

  • Smoking

  • Hyperlipidemia

  • Diabetes

Although incorporating lifestyle measures, such as healthy eating habits and regular exercise, are fundamental to achieving optimal lipid levels to prevent the risk of heart attacks, more than a third of patients who develop coronary heart disease (CHD) have no more than one traditional risk factor and about 40% of individuals with “normal” or average cholesterol levels (for the western world) die from CHD.

As such, there is a renewed interest in identifying and incorporating biomarkers that may improve current models of risk assessment. The most widely studied serum biomarker is C-reactive protein (CRP), which can be measured with a high sensitivity assay (hs-CRP).

Inflammatory processes are pivotal in the initiation and progression of atherosclerosis—from the lifelong accumulation of lipids within arterial beds to a plaque rupture event resulting in myocardial ischemia, infarction, or stroke.

Elevated levels of CRP—a nonspecific marker of inflammation—is associated with:

  • Underlying atherosclerosis (subclinical atherosclerosis)

  • Increased risk of a first cardiovascular event among individuals with or without risk factors

  • Increased risk of recurrent cardiovascular events

  • Diabetes

  • Obesity

  • Metabolic syndrome

  • New-onset hypertension

  • Sedentary lifestyle

  • Heart failure

CRP is a risk marker, and not an etiologic risk factor for CHD. It can provide prognostic information independent of lipid parameters in healthy individuals.

However, its ability to reclassify patients into higher risk categories beyond traditional risk factors is still debatable and a robust area of clinical investigation. Additionally, genetic data argues against a causative role for CRP in the pathogenesis of CHD.

Interventions to reduce CRP include risk factor reduction with medications and lifestyle changes. Currently, CRP testing is not recommended for the general population but for those at intermediate risk for CHD in whom the decision for lifelong aspirin and statin therapy remains uncertain.

II. Diagnostic Confirmation: Are you sure your patient has Coronary Artery Disease?

Pathogenic role of CRP—Linking immunity to atherosclerosis

The role of inflammatory cells and mediators in atherothrombosis and myocardial ischemia and infarction is firmly established. There is experimental evidence linking the role of CRP—directly and indirectly—to plaque deposition, disruption, and the onset of cardiovascular events.

Specifically, CRP:

  • Has been identified in lipid laden plaques within atherosclerotic lesions

  • Facilitates monocyte adhesion and transmigration into the vessel wall

  • Polarizes macrophages to a M1 phenotype—a strong inflammatory trigger in plaque deposition

  • Binds to oxidized LDL and phospholipids

  • Enhances the inflammatory response within plaques via expression of adhesion molecules, IL-6 production, and monocyte chemo-attractant protein-1 (MCP-1) in endothelial cells

  • Attenuates nitric oxide production

  • Inhibits angiogenesis

  • Enhances T-cell mediated endothelial cell destruction

  • Activates the complement cascade—a marker of increased myocardial infarct size

  • Destabilizes plaque via increased metalloproteinase synthesis

  • Promotes intima-medial thickening.

1. What laboratory studies (if any) should be ordered to help establish the diagnosis? How should the results be interpreted?

How high is too high?—An interpretation of laboratory CRP results

CRP assays

Standard CRP assays—with a detectable range of 2 to 5 mg/L—may be clinically valuable in the setting of an acute infection, tissue damage, exacerbation of inflammatory diseases, or response to therapy. The detectable range of a standard assay in many patients can be too high to have any predictive accuracy in the primary prevention setting.

High-sensitivity CRP (hs-CRP)—with a detectable range as low as 0.3 mg/L—is the same protein quantified by older methods, but the assay can accurately identify low levels of baseline CRP in healthy individuals for prognostication of cardiovascular risk.

Ideal hs-CRP values—is there one?

Although the Centers for Disease Controls and American Heart Association (CDC/AHA) have provided some guidance with interpretation of hs-CRP values, there is considerable lifestyle, ethnic, and gender-related variation. The CDC/AHA defined risk based on hs-CRP values that correspond to tertiles in the general population.

  • 20 to 29 year olds—mean CRP 3.5 mg/L

  • 70 to 79 year olds—mean CRP 5.7 mg/L

Analyses from large-scale clinical trials (PROVE-IT TIMI 22 and JUPITER) have used a cut-point of 2 mg/L for defining increased cardiovascular risk. Over 50% of all adults and 41% of 20 year olds in the United States have CRP levels >2 mg/dL.

Furthermore, an analysis of 6,722 individuals from the Multi-Ethnic Study of Atherosclerosis (MESA) demonstrated a mean CRP level of 3.76 mg/L, and the mean values in those with and without events did not differ. Other studies have shown CRP’s association with cardiovascular events above levels of 3 mg/L, which generally has been considered a practical cut-point above which one may be at a higher risk for future vascular events.

Serial assessment of CRP levels for assessment of cardiovascular risk is usually not indicated. Some authorities recommend an average of two values, 2 weeks apart; however, given the relative stability of the protein, there is minimal year to year variation in the absence of an acute inflammatory response.

  • Low risk: <1 mg/L

  • Average risk: 1 to 2 mg/L

  • High risk: >3 mg/L

BMI, diabetes mellitus, metabolic syndrome, hypertension, oral contraceptive use, periodontal disease, physical exercise, moderate alcohol consumption, dietary patterns, environmental pollutant burden, and smoking cause significant baseline variation in CRP levels.

Among ethnicities, the highest levels of CRP are found in African Americans, followed by Hispanics, South Asians, Caucasians, and East Asians, respectively.

An analysis of 8,874 patients from the National Health and Nutrition Examination Survey database demonstrated CRP levels that ranged from 0.1 to 296 mg/L (mean, 4.3; median, 2.1). CRP levels are higher in women, overweight ethnic minorities, and they increase with age.

2. What imaging studies (if any) should be ordered to help establish the diagnosis? How should the results be interpreted?

In current models of cardiovascular risk assessment, risk algorithms, such as Framingham Risk Score (FRS), are used to estimate short and intermediate absolute risk, but not lifetime risk. Coronary artery calcification (CAC) is pathognomonic for atherosclerosis and is absent in normal vessel walls.

It starts to accumulate in small amounts in the second and third decades of life and progresses with age as the burden of atherosclerosis increases. It is detected and quantified by noncontrast cardiac computed tomography (CT) and can provide an estimate of total plaque burden.

Many studies have found moderately high levels of CAC to be more predictive than conventional risk factors and FRS for future adverse cardiovascular events. Calcifications inside the wall of the coronary artery are a lifelong representation of atherosclerosis and is evident long before any signs or symptoms of heart disease.

CAC score is derived from the density of the calcification as measured by Hounsfield units and then multiplied by the area of the coronary calcification. Clinically, it is reported as the “calcium score.”

Current guidelines do not support testing for CAC in patients at low risk for CHD. CAC screening is also unlikely to add to the risk assessment if an individual is considered high risk by conventional risk factors and algorithms. It has the greatest role in intensifying therapy in individuals who are at intermediate risk by FRS.

Calcium score can range from 0 to 1,000 and higher. For CAC score of 1 to 100, there may be a 2 fold increase in future cardiovascular events. This risk may be as high as 10 for CAC greater than 400 as compared to patients with a score of 0. A CAC score of 0 has been shown to have excellent long-term survival and a 1% 10 year event rate.

CAC scores can be interpreted as follows:

  • 0: Very low risk

  • 1-100: Mild risk. CAD event relative risk of 2.

  • 101-400: Moderate risk .CAD event relative risk of 3.

  • 400-1,000: High risk. CAD event relative risk of 4.6

  • >1,000: Severe risk. CAD event relative risk of 8.3

A calculator from the National Heart, Lung, and Blood Institute can be used to compare a patient’s calcium score with scores of people of similar age and ethnic background: http://www.mesa-nhlbi.org/Calcium/input.aspx

III. Management.

Management for primary prevention of Cardiovascular Disease using CRP

Is there a role of CRP in screening for Cardiovascular Disease?—Beyond the Framingham Risk Assessment

The role of CRP in improving risk prediction has undergone extensive research and study. Improvement in prediction beyond Framingham Risk Score (FRS) is challenging, because the algorithm is dominated by chronologic age. However, the FRS is limited by the following:

  • Featuring a racially and socio-economically concentrated population (e.g., long-suffering Caucasian Red Sox fans)—not reflective of the urban population in the U.S. and internationally.

  • Not incorporating family history of premature cardiovascular disease

  • Underestimation of atherosclerotic vascular disease risk in women

  • Suboptimal use by physicians

  • Misclassifying subclinical and clinical risk

  • Being limited to a 10-year risk assessment of a disease that progresses during the lifespan of an individual

Is CRP an independent risk marker for cardiovascular events?

Many prospective epidemiologic and observational studies in individuals with no history of cardiovascular disease have demonstrated that a single, nonfasting elevated CRP level is associated with increased future vascular events. There is strong correlation between baseline CRP levels and future cardiovascular events.

In the Women’s Health Study of 27,939 healthy individuals, LDL-cholesterol (LDL)—a known marker of atherothrombosis—was compared with high-sensitivity CRP. After adjusting for age and traditional risk factors, CRP was more predictive of MI, stroke, coronary revascularization, and death in comparison to LDL. Patients with high CRP (2.10 to 4.19 mg/L) and low LDL (<96 mg/dL) had a greater absolute risk of cardiovascular events than those with low CRP (<0.49 mg/L) and high LDL (132 to 154 mg/dL).

In the Framingham Offspring study—a study following over 3,000 patients for 12 years—participants with a CRP level >3 mg/L had higher rates of MI, deaths attributed to CHD, and all cardiovascular events.

Several meta-analyses have found an association between baseline CRP and subsequent risk of developing CHD. In 2009, a meta-analysis of 22 studies was performed by the U.S. Preventive Services Task Force (USPSTF), in which a CRP level >3 mg/L was independently associated with a 60% excess risk in incident CHD as compared to levels <1 mg/L, after adjusting for all Framingham risk variables.

In a subsequent meta-analysis of 160,309 individuals without vascular disease, the relative risk of CHD after adjusting for risk factors was only 1.37 for each standard deviation increase in CRP, which was comparable to that of hypertension and hyperlipidemia. Furthermore, CRP concentration was increased to a similar magnitude in a myriad of conditions, including ischemic stroke, chronic lung disease, cancers, and nonvascular mortality.

Not surprisingly, CRP levels were also linearly associated with several other inflammatory markers, including fibrinogen, sedimentation rate, and albumin. The relevance of its association with such diverse biologic processes has further divided its link with CHD.

Is there incremental value of CRP over traditional risk factors in refining risk?

Despite the robust evidence of an association between CRP and incident cardiovascular disease, data supporting improvement in predictive accuracy with the addition of CRP to traditional risk factors is far more modest. A biomarker capable of discriminating events independent of conventional risk assessment should theoretically have a robust correlation with cardiovascular disease and less of an association with Framingham risk variables. CRP’s association with CVD is a function of its strong link with traditional risk factors, such as:

  • Smoking

  • Diabetes

  • Visceral obesity

  • Age

  • Hypertension

  • Reduced exercise frequency

  • Body mass index

  • Metabolic syndrome

  • Incident type 2 diabetes mellitus

Higher levels of CRP are seen in patients with more cardiovascular risk factors. Thus CRP is best considered a risk marker and not a risk factor.

That being said, studies have shown a modest association with vascular events, after adjustment for standard risk factors. The Reynolds Risk Score—incorporating hs-CRP and parental family history of premature CHD with traditional risk factors—has outperformed the FRS in a predictive accuracy model when low risk was classified as <5% 10-year risk, rather than <10% 10-year risk.

The Reynolds Risk Score

The Reynolds Risk Score (RRS) adds an hs-CRP level and a parental family history of premature CHD to conventional parameters considered in the FRS. Though both scores are predictive of disease risk, the RRS has been shown in a large multiethnic cohort to add incremental predictive power, particularly if there is discordance in classification with the FRS. It is unclear to what extent CRP levels account for this difference. It includes all the variables of the FRS as well as:

  • Level of hs-CRP

  • Parental history of MI before the age of 60

The RRS calculator is available online: http://www/reynoldsriskscore.org/

A. Immediate management.

Which individuals may achieve the greatest benefit from CRP testing?

Reclassifying risk with CRP

Ultimately, in clinical practice, one of the most important factors in the clinical application of a novel biomarker is its ability to reclassify patients into a predefined risk category that will change treatment decisions. Reclassification is a robust metric for prediction analysis.

The current guidelines from the Third Adult Treatment Panel of the National Cholesterol Education Program (ATP III NCEP) stratifies treatment recommendations based on FRS. The 10-year predicted risk of coronary death or nonfatal MI categorizes individuals into three distinct cohorts:

  • Low risk: 10%

  • Intermediate risk: 10% to 20%

  • High risk: 20%

Of note, some guidelines have classified low risk as <5% or 6% and intermediate risk as 6% to 20% risk of MI and CHD, respectively. Current guidelines recommend only lifestyle alteration for low risk individuals.

Alternatively, an intensive treatment with lifestyle modification and statin drugs is recommended in the highest risk patients—those with a >20% predicted hard CHD risk in 10 years—and those with established CHD to achieve an LDL cholesterol ideally <70 mg/dL.

Several lines of evidence suggest that high CRP levels in patients at low risk by traditional risk factors do not significantly increase their cardiovascular risk above that predicted by risk factors alone. Likewise, a low CRP level in patients at high risk does not markedly reduce risk and therefore, does not alter clinical management.

While there is some modulation of risk in these cohorts, it does not alter clinical management significantly. The guidelines do not consistently support the use of statin therapy in those at intermediate risk by FRS. CRP may be clinically valuable if it can substantially reclassify intermediate risk patients into lower or higher risk to guide clinical decision-making.

Performance of CRP in the intermediate-risk patient

Data from the RRS study indicated that nearly 8% of patients who were initially intermediate risk were reclassified as high risk when incorporating both hs-CRP and family history.

In the WHS, of the 20% of intermediate-risk individuals, nearly 75% of them were reclassified into a lower risk cohort and only 4% (<0.5% of the total population) into a higher risk cohort.

In a German analysis of men with CRP >3.0 mg/L, individuals with an initial 10-year FRS of 15% to 19% were accurately reclassified to high risk based on their observed 10-year cardiovascular event rate. CRP did not reclassify those with a 10-year predicted risk of 10% to 14%. Thus CRP provides limited incremental information beyond that available by FRS when making treatment decisions.

Multiple lines of evidence have suggested that most reclassification results in a down-reclassification into a lower risk category—a cohort that would not require a substantial change in prevention therapies.

B. Laboratory Tests to Monitor Response To, and Adjustments in, Management

Interventions to Reduce CRP levels

A number of drugs and lifestyle measures used in the treatment of CHD also reduce CRP levels.

  • Statins: Although statins primarily affect results in LDL-cholesterol reduction by inhibition of 3-hydroxy-3-methylglutarylcoenzyme A (HMG-CoA) reductase, a nonlipid effect is a decrease in serum CRP concentration, independent of reductions in LDL levels. This may be mediated by reduced monocyte expression of cytokines or by direct inhibition of CRP gene transcription.

Statins reduce CRP levels after an ACS and in the primary prevention setting. The link between the benefits of statin drugs and the proinflammatory hypothesis of atherosclerosis may, in part, be explained by the antiinflammatory properties of statin drugs.

Several large scale clinical trials have supported the role of CRP as a potential therapeutic target for statin therapy.

  • Air Force/Texas Coronary Atherosclerosis Prevention Study (AFCAPS/TexCAPS)—Lovastatin reduced CRP levels by 15% with an associated reduction in coronary events. A posthoc analysis showed no benefit from lovastatin in patients with LDL <149 mg/dL and hs-CRP<1.6 mg/L—providing support for the lack of a low CRP arm in the subsequent JUPITER (Justification for the Use of Statins in Primary Prevention: An Intervention Trial Evaluating Rosuvastatin) trial.

  • Thiazolidinediones: Both rosiglitazone and pioglitazone reduce CRP concentrations, independent of glycemic control.

  • Dietary modification: Diets low in saturated fat, carbohydrates, and calories independently reduce CRP levels in modest amounts. In one study, 30 g of fiber supplementation daily reduced mean CRP levels from 4.4 mg/L to 3.7 mg/L.

  • Methotrexate and canakinumab (IL-1β inhibitor): Targeting the inflammatory hypothesis of atherothrombosis is the subject of two large ongoing cardiovascular prevention clinical trials.

    PROVEIT-TIMI 22 (Pravastatin or Atorvastatin Evaluation and Infection Therapy–Thrombolysis in Myocardial Infarction)—Post ACS patients treated with atorvastatin 80 mg versus pravastatin 40 mg had significant reductions in CRP at 30 days with a linear relationship between CRP levels and recurrent cardiovascular events.

    REVERSAL (Reversal of Atherosclerosis with Aggressive Lipid Lowering)—Patients with angiographic CAD treated with intensive statin therapy had reductions in CRP levels, which correlated with reductions of LDL and rate of progression of atherosclerosis by intravascular ultrasound.

    JUPITER—Patients without established CAD with low LDL (<130 mg/L) and high hs-CRP(>2.0 mg/L) were randomized to rosuvastatin 20 mg versus a placebo and showed a robust reduction in adverse cardiovascular events and mortality with rosuvastatin therapy and a 50% and 37% reduction in LDL and hs-CRP in the rosuvastatin arm, respectively. Lowering both LDL and CRP are associated with the best outcomes.

  • Beta blockers: Limited data in patients with stable CAD supports significant reductions in CRP levels with beta- blocker therapy.

C. Long-term Management

Guideline recommendations

Professional organizations and agencies have published recommendations on the use of CRP in preventive screening of asymptomatic patients.

  • USPSTF (2009): Recommend against routine testing of CRP due to insufficient evidence.

  • Canadian Cardiovascular Society (2009): Recommend hs-CRP screening in men older than 50 years and women older than 60 years who are intermediate-risk by FRS, or would not otherwise qualify for lipid lowering therapy (LDL <130 mg/dL) based on JUPITER.

  • European Society of Cardiology (2009): Recommend consideration of CRP testing in patients with hypertension categorized as high-risk by FRS—a category of patients that would be treated regardless of CRP concentration.

  • American College of Cardiology and American Heart Association (2010): A class IIa recommendation is designated for testing hs-CRP in men older than 50 and women older than 60 with LDL <130 mg/dL who do not have chronic kidney disease, diabetes, and who are not on hormone replacement therapy. A class III recommendation (no benefit) for testing is given in asymptomatic high-risk adults or younger low-risk patients.

D. Common Pitfalls and Side-Effects of Management

CRP testing

Inflammation has a pivotal role in atherosclerosis and its complications, but the incremental value of CRP above traditional risk indicators is relatively small. Reclassification into different risk categories sometimes does not change clinical management; is there really a meaningful difference between an estimated 10-year risk of 4% vs. 7%? Clinicians and patients struggle with both the interpretation and application of CRP results.

The JUPITER investigators advocate use of potent statin therapy in the absence of risk factors simply on the basis of a CRP level >2 mg/L. The trial is often misinterpreted as a “CRP trial” as opposed to a primary prevention statin study.

A lack of a low LDL/low hs-CRP arm makes it impossible to fully exclude rosuvastatin’s benefit among at risk middle-aged and elderly adults, irrespective of the CRP level. Furthermore, counter-intuitively, data from the FDA have demonstrated a slightly greater apparent benefit from rosuvastatin in individuals below the CRP median cut-point of 4.2 mg/L versus above 4.2 mg/L.

This suggests that an elevated CRP concentration does not independently predict a preferential benefit to statin therapy.

Participants in the Anglo-Scandinavian Cardiac Outcome Trial – Lipid Lowering Arm (ASCOT-LLA) who were randomized to 10 mg of atorvastatin or a placebo had a 40% reduction in LDL and a 27% reduction in median hs-CRP at 6 months. Baseline LDL and hs-CRP levels both significantly predicted cardiovascular events, although hs-CRP only minimally improved risk prediction when added to FRS.

However, using the values drawn 6 months after baseline, participants in the atorvastatin group with a LDL less than the median level of 2.1 mmol/L had nearly a 60% lower predicted risk of experiencing a cardiovascular event, while an hs-CRP less than the median level of 1.8 mg/L was not significantly associated with reduced risk.

These results demonstrate that targeting a specific hs-CRP treatment goal is not warranted, and they also strengthen the argument that hs-CRP is a risk marker, but not a risk factor for cardiovascular disease.

A meta-analysis of 28,112 patients with CHD and 100,823 controls demonstrated no increased risk for cardiovascular disease in individuals with gene variants that increased levels of CRP—arguing against CRP’s role as a mediator of cardiovascular events. It also weakens the notion that CRP is a valuable therapeutic target in the prevention of cardiovascular disease.

IV. Management with Co-Morbidities

Role of CRP in the management of established cardiovascular disease and comorbidities

Elevated CRP levels are not diagnostic of cardiovascular disease, but help refine risk prediction of future cardiovascular events in individuals without a prior CHD event. Although CRP is not routinely measured in patients with established cardiovascular disease, elevated levels have some predictive value in some conditions with respect to disease severity and prognosis.

  • Acute Coronary Syndromes (ACS)

Acute phase response is considerably enhanced in patients presenting with unstable angina and myocardial infarction (MI). Elevated levels of CRP and serum amyloid A protein have been shown to predict higher rates of recurrent ischemia, myocardial infarction, revascularization, and mortality. Additionally, CRP concentration may be associated with plaque vulnerability, and elevated CRP levels correlate with the complexity of the angiographic stenoses.

Patients with non-ST-segment elevation MI (NSTEMI) have higher levels of CRP than those with stable coronary artery disease (CAD) and are a marker of worse short-term and long-term outcomes. Data from large-scale clinical trials of ACS have been associated with increasing quartiles of CRP with mortality.

Thus, CRP may have a complimentary prognostic role with cardiac troponin. In the PROVE IT-TIMI 22 trial of stable post-ACS patients, hs-CRP levels >2 mg/dL were associated with nearly a two-fold increase in hospitalization for heart failure over the 2-year trial period.

The role of CRP in the pathogenesis of atrial fibrillation is unclear; however, in several observational analyses, elevated CRP levels are predictive of developing new-onset atrial fibrillation, failure to convert back to normal sinus rhythm, as well as the development of recurrent atrial fibrillation.

Some studies have linked elevated hs-CRP levels after an MI with higher rates of mortality, larger infarct size, and recurrent heart failure. Because CRP is an acute phase marker and an elevation may reflect the extent of myocardial injury, it is recommended to wait at least 6 weeks before measuring levels of hs-CRP.

  • Stable Coronary Artery Disease

The acute phase response in patients with stable CAD is associated with underlying atherosclerosis and is thought to reflect the risk of disruption of a vulnerable plaque. There are robust data to support a strong positive correlation between baseline CRP and future cardiovascular events.

In one analysis, baseline CRP levels in 100 patients with long-standing stable CAD were 1.8 mg/L and 4.9 mg/L in those without and with recurrent acute coronary syndromes, respectively.

Furthermore, elevated CRP levels have been predictive of coronary artery disease progression, and development of both heart failure and diabetes. They have also been shown to be correlated with the development of transplant vasculopathy and allograft failure.

  • Postcoronary Revascularization

Elevated levels of CRP are predictive of recurrent ischemia after coronary artery bypass grafting (CABG). In one study, CRP levels ≥3 mg/L were associated with a much higher incidence of ischemic events.

Elevated CRP levels are also predictive of a worse prognosis after percutaneous coronary intervention (PCI). A fourfold increase in mortality at 20 months and a 2.5-fold increase in MI and death at 30 days have been reported in those undergoing PCI with elevated preprocedural CRP levels.

An enhanced inflammatory response pre-PCI may have a role in the response to injury after endothelial damage during balloon angioplasty and stent implantation.

  • Cerebrovascular disease

The link between inflammation and atherosclerosis is not limited to the coronary vasculature. Elevated CRP levels have been independently predictive in the development and progression of carotid atherosclerosis, risk of subsequent cerebrovascular accidents, and prognosis after stroke.

  • Heart failure

In a post-hoc analysis of valsartan therapy in patients with heart failure, increasing quartiles of CRP levels were associated with greater incidences of death and adverse cardiovascular events. Additionally, the median CRP level was higher in patients with heart failure.

  • Hypertension

Hypertension is at least 1.5 times as likely to develop in patients with a CRP level >3.5 mg/L. In the Women’s Health Study of over 20,000 patients without hypertension who were followed for nearly 8 years, increasing values of CRP were associated with the rate of developing hypertension. The association may be secondary to reduced nitric oxide synthesis resulting in increased vascular resistance.

  • Atrial fibrillation

The role of CRP in the pathogenesis of atrial fibrillation is unclear; however, in several observational analyses, elevated CRP levels are predictive of developing new-onset atrial fibrillation, failure to convert back to normal sinus rhythm, as well as the development of recurrent atrial fibrillation.

V. Patient Safety and Quality Measures

A. Appropriate Prophylaxis and Other Measures to Prevent Readmission.

A Practical approach in utilizing hs-CRP in practice

An elevated hs-CRP is a marker of inflammation and independently associated with an increased risk of CHD. However, hs-CRP only modestly adds to traditional risk factors in CHD risk assessment. Screening asymptomatic patients, who are classified by the Framingham Risk Score as low risk or high risk, has not been demonstrated to change treatment recommendations, and subsequently a population level screening with hs-CRP is not advisable.

In the JUPITER trial, participants with a LDL-C of <130 mg/dL and a hs-CRP ≥2 mg/L, who were randomized to 20 mg of rosuvastatin, had a 44% reduction in major cardiovascular events. However, the lack of a low hs-CRP group makes it impossible to determine if similar benefits would have been observed in a group without evidence of vascular inflammation.

Therefore, among individuals with a LDL <130 mg/dL who meet the JUPITER study criteria, it is reasonable to screen with hs-CRP to aid in the decision of initiating statin therapy. However, among the much larger group of intermediate risk patients who do not meet the JUPITER inclusion criteria, the benefit is less clear and hs-CRP screening may be considered if other risk stratification methods are not readily available (e.g. coronary artery calcium screening).

Interpretation of hs-CRP levels:

  • >10 mg/L suggest an acute infection or active systemic inflammatory process. Repeat measurement should be postponed until at least 2 weeks after any known inflammatory process has subsided.

  • Between 3 and 10 mg/L is associated with an increased risk of CHD and it is reasonable to institute statin therapy after consideration of the patient’s overall CHD risk.

  • Between 1 and 3 mg/L are indeterminate and can be repeated 2 or more weeks later. If the average value is >2mg/L, the individual’s overall risk factor profile should be reevaluated, and it is reasonable to either initiate statin therapy or an intensive lifestyle modification regimen.

  • <1 mg/L suggest a very low risk for the development of CHD. However, lifestyle and dietary modification should still be emphasized.

There is minimal compelling evidence for the CRP measurement for risk prediction or the targeting of statins in the primary prevention setting. Lowering of CRP levels with intensive therapy also appears to have a very modest role in the prevention of cardiovascular disease.

B. What's the Evidence for specific management and treatment recommendations?

Bogaty, P, Poirier, P, Simard, S. “Biological profiles in subjects with recurrent acute coronary events compared with subjects with long-standing stable angina”. Circulation. vol. 103. 2001. pp. 3062-8. (Individuals with multiple acute coronary events had higher levels of CRP compared to individuals with stable angina and no history of acute coronary disease (5.7 mg/L vs. 3.0 mg/L). An elevated CRP level was also more strongly predictive of future acute coronary events than serum cholesterol level or smoking status. In addition, a baseline CRP >3.5 mg/L was associated with more than a 7-fold increased risk for future acute coronary events.)

Wilson, PW, Pencina, M, Jacques, P. “C-reactive protein and reclassification of cardiovascular risk in the Framingham Heart Study”. Circ Cardiovasc Qual Outcomes. vol. 1. 2008. pp. 92-7. (In this prospective cohort of the Framingham offspring study, participants with a CRP ≥3 mg/L were associated with a significantly increased risk of hard CHD (HR 1.88 and total CVD HR 1.58 after multivariate adjustment). However, when CRP was added into a multivariable risk prediction model, there was no meaningful increase in the C-statistic and only a 7% and 9% improvement in the net reclassification index for hard CHD and total cardiovascular disease (CVD), respectively. This suggests that while CRP is significantly associated with CHD and CVD, it only provides a minimal improvement in risk prediction over traditional risk factors.)

Ridker, PM, Rifai, N, Rose, L. “Comparison of C-reactive protein and low-density lipoprotein cholesterol levels in the prediction of first cardiovascular events”. N Engl J Med. vol. 347. 2002. pp. 1557-65. (This study compared LDL-C and CRP as predictors of CVD among a cohort of almost 30,000 women. Women in the highest baseline CRP quintile had a 2.3-fold greater risk of developing a first cardiovascular event compared to the lowest quintile, while women in the highest baseline LDL quintile had a 1.5-fold increased risk. However, in a multivariate risk model the addition of CRP or LDL resulted in the same C-statistic, suggesting that after adjusting for traditional risk factors, CRP and LDL provide a similar improvement in risk prediction.)

Kaptoge, S, Di Angelantonio, E. “C-reactive protein concentration and risk of coronary heart disease, stroke, and mortality: an individual participant meta-analysis”. Lancet. vol. 375. 2010. pp. 132-65. (This meta-analysis of 54 trials included over 160,000 participants and found CRP to be significantly associated with CHD and ischemic stroke. After adjustment for conventional risk factors, this relationship was attenuated, but still significant with a relative risk of 1.36 for CHD and 1.32 for ischemic stroke.)

Buckley, DI, Fu, R, Freeman, M. “C-reactive protein as a risk factor for coronary heart disease: a systematic review and meta-analyses for the U.S. Preventive Services Task Force”. Ann Intern Med. vol. 151. 2009. pp. 483-95. (This meta-analysis of 23 trials revealed that after adjusting for FRS variables, participants with CRP >3.0 mg/L had an almost 60% higher risk of CHD compared to those with CRP <1.0 mg/L. Only five of the included studies compared predictive models with and without CRP. Although pooled results were not reported, three of these studies reported that CRP improved risk stratification among intermediate risk participants.)

Ridker, PM, Buring, JE, Rifai, N, Cook, NR. “Development and validation of improved algorithms for the assessment of global cardiovascular risk in women: the Reynolds Risk Score”. JAMA. vol. 297. 2007. pp. 611-9. (The Reynolds Risk Score incorporates traditional and novel cardiovascular risk factors for prediction of CVD in women. Two models were developed—both incorporating hsCRP—that resulted in a reclassification of approximately half of women with an intermediate FRS score. hsCRP was a significant variable in the models, but much of the reclassification resulting from the Reynolds Risk Score may be attributed to other variables.)

Ridker, PM, Cannon, CP, Morrow, D. “C-reactive protein levels and outcomes after statin therapy”. N Engl J Med. vol. 352. 2005. pp. 20-8. (This trial treated individuals with atorvastatin 80 mg or pravastatin 40 mg and compared outcomes based on LDL and CRP levels after treatment. Participants with a posttreatment LDL-C of <70 mg/dL had virtually the same event rate compared to those with posttreatment CRP <2 mg/L. The lowest event rate was observed in the group with LDL [<70 mg/dL and CRP <2 mg/L].)

Ridker, PM, Danielson, E, Fonseca, FA. “Rosuvastatin to prevent vascular events in men and women with elevated C-reactive protein”. N Engl J Med. vol. 359. 2008. pp. 2195-207. (The JUPITER trial was conducted in participants with a LDL-C <130 mg/dL and a hsCRP ≥2 mg/L who were randomized to rosuvastatin or a placebo. The group treated with rosuvastatin had a reduction in LDL of 50% and hsCRP of 37%. They were also 44% less likely to experience a major cardiovasular event.)

Woloshin, S, Schwartz, LM. “Distribution of C-reactive protein values in the United States”. N Engl J Med. vol. 352. 2005. pp. 1611-3. (Using NHANES data, the authors report that half of U.S. adults have a CRP >2 mg/L and that CRP values are higher for women than men, increase with age, but are less variable by race. Therefore, they argue that further evidence is necessary before adopting CRP ≥2 mg/L as the cut-point for defining high risk.)

Koenig, W, Lowel, H, Baumert, J, Meisinger, C. “C-reactive protein modulates risk prediction based on the Framingham Score: implications for future risk assessment: results from a large cohort study in southern Germany”. Circulation. vol. 109. 2004. pp. 1349-53. (This study assesed the impact of CRP in risk prediction among a cohort of middle-aged European men. The addition of CRP significantly improved the model for intermediate risk, but not for low risk participants.)

Scirica, BM, Cannon, CP, Sabatine, MS. “Concentrations of C-reactive protein and B-type natriuretic peptide 30 days after acute coronary syndromes independently predict hospitalization for heart failure and cardiovascular death”. Clin Chem. vol. 55. 2009. pp. 265-7. (This study demonstrated that patients who developed heart failure after an acute coronary syndrome had a higher hs-CRP level compared to those who did not develop heart failure (3.7 mg/L vs. 1.9 mg/L). There was also a greater risk of developing heart failure after acute coronary syndrome for patients with a higher concentration of hs-CRP.)

Milazzo, D, Biasucci, LM, Luciani, N. “Elevated levels of C-reactive protein before coronary artery bypass grafting predict recurrence of ischemic events”. Am J Cardiol. vol. 84. 1999. pp. 459-61. (Patients in this trial had a CRP level drawn before undergoing coronary artery bypass surgery. At 6 years of follow-up, those with CRP level ≥3 mg/L were significantly more likely to have experienced an ischemic event. This suggests that preoperative CRP level could be a useful marker to risk stratify patients. On the other hand, all patients with vascular disease should be treated aggressively.)

Chew, DP, Bhatt, DL, Robbins, MA. “Incremental prognostic value of elevated baseline C-reactive protein among established markers of risk in percutaneous coronary intervention”. Circulation. vol. 104. 2001. pp. 992-7. (CRP was measured in a registry of patients before routine percutaneous coronary intervention. Patients with higher CRP values had an increased 30-day event rate for death or MI, which increased progressively with the baseline CRP level. Those in the highest CRP quartile [>1.01 mg/dL] had an almost 4-fold increased risk for 30-day MI or death compared to those in the lowest quartile [<0.16 mg/dL].)

Mueller, C, Buettner, HJ, Hodgson, JM. “Inflammation and long-term mortality after non-ST elevation acute coronary syndrome treated with a very early invasive strategy in 1,042 consecutive patients”. Circulation. vol. 105. 2002. pp. 1412-5. (Patients with acute coronary syndrome had a CRP measured at hospital admission and were followed for an average >1.5 years. Patients with a CRP level >10 mg/L had a more than 4-fold risk for in-hospital death. At the end of follow-up, this increased risk persisted and patients with CRP levels ≥3 mg/L had a more than 2-fold increased risk of death, while those with a CRP >10 had an almost 4-fold increased risk.)