C-Reactive Protein Levels and Outcomes after Statin Therapy

Abstracts & Commentary

Sources: Nissen SE, et al. N Engl J Med. 2005;352:29-30; Ridker PM, et al. N Engl J Med. 2005;352:120-128.

Two articles have been recently published emphasizing the importance of C-reactive protein (CRP) in statin treated patients. Both are follow-up reports following the initial publication of the REVERSAL1 and PROVE-IT trials.2

These somewhat related studies provide an analysis of CRP and LDL levels and their relationship to cardiovascular outcomes and demonstrates markedly similar results. CRP, as well as LDL cholesterol levels, were predictive of atherosclerotic progression and in an acute coronary syndrome cohort, clinical outcomes. The best results in PROVE-IT were found in those patients in the lowest quartile of achieved 30-day LDL cholesterol and CRP levels. REVERSAL and PROVE-IT both compared the effects of LDL lowering with high dose atorvastatin, 80 mg daily vs pravastatin 40 mg. The REVERSAL trial showed that aggressive LDL lowering with atorva reduced progression of coronary atherosclerosis over 18 months; stabilization of the atherosclerotic process occurred when LDL levels were lowered to < 70 mg. Reversal used intravascular ultrasound in 500 individuals who were randomized to aggressive vs moderate statin. The entire cohort had a mean baseline LDL-C of 150 mg/dL, decreasing to 94 mg/dL; the atorva group reached a mean nadir of 69 mg/dL, while in the prava patients LDL decreased to 110 mg/dL. Post-hoc analyses indicated that CRP decreased with both statins. There was a poor correlation between the degree of CRP reduction and that of LDL lowering.

In REVERSAL, LDL and CRP levels were both associated with atherosclerosis progression or regression; "only patients with the largest reductions in CRP levels had regression of atheroma." Individuals who had a sub-optimal decrease in LDL or CRP (above median) had significantly higher rates of progression than those with lower 18-month of LDL and CRP levels, ie, below the median of the entire cohort. Nissen and colleagues state that the data confirm "a continuous relationship between the magnitude of reduction of either LDL or CRP levels and the rates of progression of atherosclerosis." Regression was only found in subjects with a maximum reduction of CRP, but not in patients who had the maximum reduction in LDL. Atorva treated patients had slower rates of progression partly related to the associated decline in CRP. CRP levels were 30-40% lower at 30 days in REVERSAL.

In the PROVE-IT study, 3700 individuals with acute coronary syndromes were treated with the same drugs at identical dosage for an average of 2 years. The results are parallel to the REVERSAL trail in that both LDL and CRP levels were related to clinical outcomes. While CRP was lowered more in the atorva patients, as in REVERSAL, there was not a good correlation between reduction of CRP and reduction in LDL. Results were related to the median values for LDL-C (70 mg/dL) and CRP (2 mg/L); below the median subjects were favorably impacted by statin therapy. Thus, echoing REVERSAL, PROVE-IT demonstrated a relationship between LDL lowering and the risk of recurrent MI or coronary death; there was a near linear relationship between achieved CRP and the primary endpoint. CRP levels at 30 days were used rather than CRP level at the end of the study. A cut point of 70 mg/dL for LDL and 2 mg/L for CRP correlated with outcomes, including rates of recurrent myocardial infarction or coronary death. CRP less than 2 mg/L bestowed a significantly lower death rate than those > 2 mg/L; patients with an LDL level < 70 mg/dL also had significantly fewer events than those > 70 mg/dL, irrespective of assigned statin. A low CRP (< 2 mg/L) also predicted improved outcomes in individuals irrespective of achieved LDL levels. Seventy-two percent of the atorva patients had an LDL < 70 at 30 days, compared to only 22% of prava patients. "There was little evidence that either agent led to better clinical outcomes once target levels of both LDL and CRP were achieved." Patients with the lowest LDL (< 70 mg/dL) and CRP levels (< 1 mg/L) had the lowest rates of recurrent events; 80% of such individuals were on atorva.

Cannon and colleagues concluded that achieving a target CRP of < 2 is beneficial and predicts a higher event-free survival at all levels of LDL cholesterol achieved. Furthermore, they suggest that monitoring of CRP may help guide statin therapy; Cannon et al suggest that subsequent guidelines for statin use should include measurements of CRP.

The investigators of both studies believe that their data strongly suggest that reduction of inflammation by statins, as assessed by CRP, may have a significant role in modulating atherothrombosis. Individuals in whom CRP fell to less than 2 mg/L benefited even when LDL cholesterol did not fall below 70 mg/dL. If such a low LDL was achieved in either atorva or prava patients, there was "little evidence of a differential outcome according to specific statin given." Nissen et al believe that coronary risk will be further reduced by using more aggressive therapy to achieve CRP and LDL targets. Nissen et al conclude that their data provide "strong evidence to support the hypothesis that therapies designed to reduce inflammation after acute coronary ischemia may improve cardiovascular outcomes."

Comment by Jonathan Abrams, MD

Earlier publications assessing CRP in large groups of individuals in several different populations have suggested that both the CRP and LDL have independent effects on coronary artery disease. Fortunately, REVERSAL and PROVE-IT had a similar study feature in that prava 40 was tested against atorva 80. Both papers conclude that the CRP data indicates support for non-lipid actions of statins, ie, anti-inflammatory effects that appear to quench plaque activation and decrease recurrent cardiovascular events. REVERSAL included stable individuals with single vessel disease, whereas, in PROVE-IT, individuals were enrolled 10 days or more after an acute coronary event; both demonstrated that CRP and LDL interactions influence outcomes.

These 2 studies follow a number of reports by Ridker and colleagues confirming the relative independence of CRP reduction vis-à-vis LDL reduction with the conclusion that both are optimal. What does this mean for the practicing physician? The data support an increasing role for CRP measurements to monitor patients with coronary artery disease (CAD), as well as to optimize lipid-lowering therapy. For instance, a high-risk individual with CAD and CRP of > 2 with an acceptable LDL level may benefit from further lowering of LDL. Certainly, future research will result in other therapies that decrease inflammation and lower and reduce other inflammatory markers. In that CRP levels are only weakly related to reductions in LDL, it is not at all clear that a boost in statin dosage in individuals with a high CRP would be beneficial, although this strategy might be worth trying, as well as consideration of using other lipid-lowering agents, all of which have been shown to lower CRP. These data do seem to support a separate anti-inflammatory action of statins that could be considered non-lipid or pleiotropic properties.

Nissen et al have previously suggested that atorvastatin may be more effective than prava, not only because of its more potent action in lowering LDL but because of greater pleiotropic actions. However, in the analysis by Ridker et al, once target LDL levels were achieved (< 70 mg/dL), there was no evidence that atorva is superior; this seems to refute the idea that atorva’s greater efficacy is modulated by non-lipid actions. The marked potency of 80 mg atorva dose is important in bringing LDL levels down to very low levels, as well as contributing to reductions of CRP. It seems reasonable to suggest that the published CRP guidelines take a rather conservative view on measuring CRP and using this information to modify therapy. Whether CRP is a player or a marker for atherothrombosis is still debated, although these 2 studies point to an active role for CRP, contributing to atherosclerosis and plaque activation. It is likely that subsequent guidelines relating to the role of lowering LDL cholesterol will contain more robust and positive recommendations for measuring CRP with the concept of modifying therapy according to CRP levels.

Dr. Abrams, Professor of Medicine, Division of Cardiology, University of New Mexico, Albuquerque, is on the Editorial Board of Clinical Cardiology Alert.


1. Nissen SE, et al. JAMA. 2004;291:1071-1080.

2. Cannon CP, et al. N Engl J Med. 2004;350:1495-1504.