Baltimore, MD - High levels of C-reactive protein (CRP) are tightly linked with traditional coronary heart disease (CHD) risk factors [1]. That is the chief finding from a large analysis of more than 15 000 men and women in which investigators showed that the risk attributable to elevated CRP levels was primarily accounted for by the presence of at least one abnormal CHD risk factor.

In a paper published in the October 11, 2005 issue of the Archives of Internal Medicine, first author Dr Michael Miller (University of Maryland, Baltimore) and colleagues conclude that the "measurement of CRP levels may have limited clinical utility as a screening tool for CHD risk assessment unless randomized clinical trials can demonstrate that lowering CRP levels offsets CHD events beyond well-established lifestyle and pharmacologic modalities."

"If you have an agent that works purely on the inflammatory cascade without affecting lipids, blood pressure, or blood sugars, and you were able to show that the agent reduced inflammation and beneficially affected cardiovascular event rates, independent of these other risk factors, then it's a different ball game," Miller said during an interview with heartwire. "But because CRP is so tightly linked to these risk factors and treating them also lowers CRP, it makes it very difficult to separate out the impact of CRP beyond traditional risk factors."

In an editorial accompanying the paper, Drs Russell Tracy (University of Vermont, Burlington) and Lewis Kuller (University of Pittsburgh Medical Center, PA) write that many recent studies have helped emphasize the importance of inflammation in atherosclerosis, but they reiterate Miller's concerns, arguing that CRP testing offers limited clinical usefulness as a screening tool [2].

The editorialists point out, as do Miller and colleagues, that the 2003 Centers for Disease Control (CDC) and American Heart Association (AHA) guidelines state that CRP measurements should be reserved for patients at intermediate risk of CHD. The CDC/AHA guidelines state that if a person's risk is either low or high after the established risk factors have been assessed, CRP measurements are unlikely to add new information to guide clinical decision-making [3].

Despite such caution, others argue that CRP can provide additional information, essentially acting as a "tiebreaker" in these patients who present with borderline risk factors. Others see CRP measurements as an adjunct to guide a patient's progress in reducing overall CHD risk.

"CRP can often be a motivator for patients," Dr Christopher Cannon (Brigham and Women's Hospital, Boston, MA) commented to heartwire. "If they know they are overweight and that they need to stop smoking, a CRP level above 5.0 mg/L might alert some patients. As a screening tool it can be used to get patients to do all the things they know they are supposed to be doing. The second thing is that while cholesterol is relatively straightforward and easy to measure and monitor, who is to argue about how much weight to lose? We all know we need to lose weight but do you need to lose 5 lbs or 20 lbs? If you could measure CRP then you could gauge that you really need to keep pushing. It could be a useful adjunct, and certainly not replace other tools, to monitor a patient's progress in controlling all these other risk factors."

Cannon said there is still a debate as to whether CRP is a pathologic molecule capable of causing cardiovascular events or a marker, as this paper has shown, of having a lot of risk factors. The paper by Miller et al suggests that the different risk factors are probably influencing coronary disease by irritating the artery and causing inflammation, thus elevating CRP levels, he said.

Speaking with heartwire, Miller noted that inflammation has been implicated in cardiovascular disease and that CRP is one of the most actively studied biomarkers. He pointed out, however, that despite inflammation having been shown to be predictive of CHD, few data exist on the extent to which high CRP levels are attributable to traditional risk factors. If linked to these other conventional risk factors, there would be little use for measuring CRP during routine clinical visits, he said.

If you're proactive in dealing with a patient who has established coronary disease, you're going to take out all your ammunition.

"If you're proactive in dealing with a patient who has established coronary disease, you're going to take out all your ammunition," said Miller. "This is what we're advocating. Instead of focusing on one measurement, even though CRP is a great marker for inflammation, and even though inflammation is extremely important in atherosclerosis, we can't lose sight of the fact that we need to intensively treat all risk factors. CRP often serves as a diversion and instead of getting the weight down, getting blood pressure down, getting lipids down to an optimal level, and treating glucose levels, now we're looking at CRP."

In this study, the investigators examined data from the third National Health and Nutrition Examination Survey (NHANES III), conducted between 1988 and 1994. The presence of high CRP levels in NHANES III was measured along with CHD risk factors, including total cholesterol levels, fasting blood glucose levels, blood pressure, body mass index, HDL cholesterol levels, triglyceride levels, and current smoking status, designated as abnormal, borderline, or normal.

Weighted multiple logistic regression analysis adjusted for age and race identified excess weight, hypertension, female sex, diabetes, cigarette smoking, and low HDL cholesterol levels as the factors most highly associated with elevated CRP levels. The analysis also revealed that the risk attributable to elevated levels of CRP in men and women was accounted for primarily by the presence of at least one abnormal risk factor. Elevated CRP levels were reported rare in the absence of borderline or abnormal levels of conventional risk factors.

"Our feeling is that unless you can demonstrate that CRP adds information above and beyond what we know about other risk factors, and it clearly doesn't in this study, I don't really see the need for routine CRP testing," said Miller.

Also commenting on the paper for heartwire, Dr James de Lemos (University of Texas Southwestern Medical Center, Dallas) pointed out that in a real-world setting, a lot of patients with high CRP have other risk factors that might lead to aggressive care anyway. In contrast to the Physicians' Health Study and the Women's Health Study, where the role of CRP was initially uncovered, the NHANES patients, as well as other patients, including those in the Dallas Heart Study, might be a little different in that they have a larger prevalence of obesity, which appears to be an important, thriving influence for CRP levels.

"I think this paper highlights that there is a limited role for CRP right now," said de Lemos. "It demonstrates that most subjects in the general population have some other cardiovascular risk factors. That doesn't mean that CRP is not valuable. It could mean that CRP integrates some of the effects of these risk factors or that, as Paul Ridker (Brigham and Women's Hospital, Boston, MA) has shown, that CRP might be a particularly important risk factor for people with other risk factors, like obesity. What we don't know is whether these high CRP levels in obesity are always bad for the patient, or whether obesity might present to some extent a challenging situation where we might overestimate cardiac risk if we use CRP as a measurement, such as in obese women, for example."

Drs Roger Blumenthal (Johns Hopkins University Medical Center) and Steven Nissen (Cleveland Clinic, OH) both take a more optimistic view of CRP in clinical practice.

"If you look at the totality of the evidence you do see about a doubling of risk with elevated CRP levels," said Blumenthal. "But you have to figure how and when CRP should be used. I think CRP is helpful in the patient with a broad 6%-to-20% 10-year risk of coronary heart disease. If you're not sure if the patient qualifies for aspirin or aggressive lipid lowering, a high CRP can help provide additional information. I think it can also be a motivating factor, in that if a CRP level is high, then it can jump-start a patient to get exercising, quit smoking, and do other things."

Like Blumenthal, Nissen argued that CRP testing is relatively inexpensive and provides additive information above and beyond traditional risk factors, especially in patients with a borderline CHD risk profile.

"Many of us frequently see patients with borderline risk factors, leaving some question as to whether or not they should be treated, let's say with a statin," said Nissen. "If we find a high CRP, then we will edge toward treating. If they have a low CRP, then we may not treat. Paul Ridker has shown that with CRP levels below 1.0 mg/L, event rates are very low, even in patients who have modest elevations in cholesterol. I use the information in patients who have borderline cholesterol levels for [choosing] treatment. If their CRP is 0.4 mg/L, it is going to edge me in the direction of not treating. That's a cost-saving approach that is very good for patients. Why expose them to a drug they might not need?"

Miller, on the other hand, is not convinced of the prognostic capabilities of CRP, noting to heartwire that some patients present to the emergency room after an acute myocardial infarction or stroke with virtually undetectable CRP levels.

In their editorial, Tracy and Kuller note the considerable media attention surrounding the "inflammation and CHD story," pointing to a New York Times editorial suggesting CRP be measured in otherwise healthy individuals [4]. The paper by Miller et al provides support for clinicians to retain focus on established, modifiable risk factors.

Lowering CRP level alone may be equivalent to tightening your belt to reduce waist circumference, providing little real benefit because the underlying systems have not been affected.

The editorialists also express concern with the ramifications of specifically targeting CRP for therapeutic intervention. They acknowledge that monitoring inflammation might be clinically useful in certain specific settings, such as when interpreting data from clinical trials of statins, but stress that they do not believe CRP levels should be the specific focus of efforts to develop pharmaceutical agents.

"We offer the following analogy: as epidemiologists have known for many years, a simple measure of waist circumference is a reasonably good method to identify obesity-related increased risk for vascular disease and diabetes," they write. "Exercise and weight loss will reduce both your waist circumference and your risk (and your CRP level, for that matter), since many underlying systems are positively affected by these measures. However, lowering CRP level alone may be equivalent to tightening your belt to reduce waist circumference, providing little real benefit because the underlying systems have not been affected."

"Seeing a good CRP report would not take away the importance of getting blood pressure down and treating other risk factors," countered Cannon. "Hopefully, it would be additive. Waist circumference tells you more than just BMI as it provides a window into visceral fat, and so it is going to be highly correlated with weight, which in turn will be correlated with blood pressure, but it doesn't mean you shouldn't measure waist circumference. The information CRP provides might be additive, but it is a simple and inexpensive way to monitor risk that hopefully provides additional information. If this were a $1000 test, then that would be different."