Introduction

Low-density lipoprotein (LDL) cholesterol has been the main target for correction in guidelines for people at risk of cardiovascular disease (CVD), for many years. New, more stringent, guidelines were issued by an expert panel in 2001 - the APT III.¹ However, the quite frequent occurrence of atherosclerosis in the absence of hyperlipidemia has led to increased interest in the measurement of C-reactive protein (CRP), a marker for inflammation, as a predictor of CVD. Now a 'head-to-head' study of the predictive values of these two tests has been done, and reported in the New England Journal.

Method

Between 1992 and 1995 the investigators measured CRP and LDL cholesterol levels in 27,939 participants in the US Women's Health Study. They also collected data on behavioral and lifestyle factors at this time. The women were carefully followed for a mean of 8 years, with all myocardial infarctions, stroke, revascularization procedures, and cardiovascular deaths recorded.

In the analysis, quintiles were formed for both CRP and LDL cholesterol levels, and the probabilities of event-free survival computed after adjustment for hormone replacement therapy, age, smoking status, blood pressure category, presence of diabetes, and high-density lipoprotein levels.

Results

At baseline, the mean age of subjects was 54.7, and their mean BMI was 25.9. Forty-four percent were on hormone replacement therapy, 25% had hypertension, 12% smoked, and 2.5% had diabetes.

The relative risk (RR) of a first cardiac event for the different quintiles of CRP and LDL cholesterol are shown in the tables, with the lowest quintile = 1.0:

C-reactive Protein

Quintile	1	2	3	4	5
mg/L	< 0.49	0.50-1.08	1.09-2.09	2.10-4.19	> 4.19
Rel. risk	1.0	1.4	1.6	2.0	2.3
95% CI		0.9-2.2	1.1-2.4	1.3-3.0	1.6-3.4

LDL Cholesterol

Quintile	1	2	3	4	5
mg/dL	< 97.6	97.7-115.4	115.5-132.2	132.3-153.9	> 153.9
Mmol/L	< 2.52	2.53-2.98	2.99-3.41	3.42-3.97	> 3.97
Rel. risk	1.0	0.9	1.1	1.3	1.5
95% CI		0.7-1.2	0.8-1.4	1.0-1.7	1.1-2.0

The trend across quintiles was in both instances highly significant (p<0.001).

In another analysis, the subjects were classified into four groups, according to their relative marker values - low CRP & low LDL, low CRP & high LDL, high CRP & low LDL, and high CRP & high LDL. With low CRP & low LDL as 1.0, the RRs of the other groups were 1.5, 1.5, and 2.1, respectively.

Finally, analyses were conducted on groups formed according to the Framingham Risk Scores that were used in the ATP III.1 They showed that increasing levels of CRP were associated with increased risk of cardiovascular events at all levels of estimated 10-year risk based on the Framingham risk score. Overall, 77% of all events occurred in those with LDL levels below 160 mg/dL (4.14 mmol/L), and 46% in those with LDL levels below 130 mg/dL (3.36 mmol/L).

Comment

An accompanying editorial² evaluates the principal finding of the study - namely, that CRP determination has a greater predictive value for cardiovascular events than LDL cholesterol levels - and tries to answer the question: should CRP be added to regular screening?

Forty years ago, there was agreement that, if there is no generally accepted treatment, it's premature to start routine screening. There are plenty of treatment options for lowering elevated blood cholesterol levels. But if one wants to address an elevated marker for systemic inflammation, should one prescribe an antibiotic?

Elevated CRP levels are associated not only with inflammation but also with metabolic risk factors - central adiposity and insulin resistance. And CRP levels can be lowered by therapy directed at other cardiovascular risk factors; both exercise and statin drugs have been shown to reduce elevated CRP values. So until we know more, it seems reasonable to 'treat' elevated CRP levels by the lifestyle changes important for lowering cardiovascular risk: control weight, blood pressure, lipid levels, stop smoking, and exercise a lot. Time will tell if CRP deserves a place in the routine cardiovascular screen.