Interesting News from the Lipid Police
Abstracts & Commentary
Synopsis: Statin therapy may be clinically effective in persons without hyperlipidemia and suggest that evaluation of CRP may provide a method for appropriate targeting of statin therapy for primary prevention.
Sources: Rubins HB, et al. Circulation. 2001;103:2828-2833; Ridker PM, et al. N Engl J Med. 2001;344: 1959-1965; Munford, RS [editorial]. N Engl J Med. 2001;344:2016-2018.
I. C-Reactive Protein
The inflammatory surrogate, high sensitivity C-reactive protein (hsCRP), has received a great deal of recent attention as a potential marker of risk for patients with and without underlying coronary artery disease (CAD). The latest publication from Ridker and colleagues is an analysis of the AFCAPS/TexCAPS primary prevention study, with particular attention to cardiac events, baseline lipid levels, and CRP. This study of almost 6000 normal individuals, with average total and LDL-C and below average HDL-C, randomized subjects to lovastatin 20-40 mg/d vs. placebo for a mean of 5 years. As previously reported, lovastatin reduced the composite CAD event rate by 37% over placebo; P < 0.001.1 The actual event rate in this study was quite low compared to the other randomized statin trials, as this was a low-risk primary prevention cohort. This analysis used hsCRP levels obtained at baseline on 5742 (87%) of the entire cohort and evaluated the subsequent risk of an event related to baseline hsCRP as well as baseline LDL-C. The patients were divided into quartiles of LDL cholesterol and hsCRP. The analysis was based on cohorts with initial LDL-C and CRP levels lower or higher than the median for the entire group. The reduction in relative risk was computed for 4 groups. LDL < the median (< 149 mg/dL) and CRP < median; LDL < median, CRP > median; LDL > median and CRP < median; and both LDL and CRP > than the median. There were approximately 1400 participants in each cell. Baseline mean hsCRP was 0.31 mg/dL with a median of 0.16 mg/dL. There was not an important relationship between baseline CRP and lipid levels.
The results indicated that rates of coronary events increased with each quartile of hsCRP, with lovastatin reducing the relative risk of an event vs. placebo by 1.0-, 1.2-, 1.3-, 1.7-fold when the lowest to the highest quartiles of CRP were assessed. Lovastatin was associated with an overall decrease of 15% in the median CRP level by the end of the first year of treatment. Placebo had no such effect. Lovastatin’s effect on LDL-C was not correlated with reductions in CRP; "virtually none of the observed variance . . . could be explained by lovastatin-induced changes in lipid fractions." In an analysis of the 4 LDL-C and CRP cohorts, lovastatin reduced events in participants with high LDL (> median), irrespective of hsCRP. The statin was also effective in subjects with high CRP and lower-than-median LDL. However, in individuals with both low LDL-C and low CRP levels (< the median), there was no evidence of statin reduction in CAD risk. Using the total cholesterol/HDL ratio instead of LDL-C resulted in virtually identical findings, with the statin being effective in those individuals with a high ratio, and those with < median ratio but > median hsCRP. "In contrast, lovastatin was far less as effective among those with a ratio of total HDL cholesterol lower than the median who also had a C-reactive protein level < the median." The number needed to treat was in the 40s for 3 of the 4 cohorts but was unacceptably high for the low-LDL, low-CRP cohort (25% of individuals). In the placebo patients, a < median LDL and a high CRP was associated with high rates of events. Ridker et al conclude that in AFCAPS/TexCAPS, baseline CRP levels were an independent predictor of first coronary events, and that lovastatin was highly effective in reducing this risk in individuals with a high CRP even without hyperlipidemia. Ridker et al suggest that the study raises the possibility that statin therapy may be clinically effective in persons without hyperlipidemia and suggest that evaluation of CRP may provide a method for appropriate targeting of statin therapy for primary prevention. They hypothesize that assessment of CRP adds to risk assessment, even in the absence of hyperlipidemia, and helps detect individuals at high and low risk. The 15% reduction in CRP levels of 1 year is concordant with data from other trials with different statins. These data, in the aggregate, suggest that statins have nonlipid or pleiotropic actions, potentially leading to plaque stabilization and a reduction of vascular inflammation. They point out that > 50% of all heart attacks occur in the absence of overt hyperlipidemia, and thus the use of CRP as a marker for CAD risk seems rational. In the CARE trial, pravastatin was found to decrease CAD event rates in postMI patients who had elevated CRP. Finally, Ridker et al call for studies randomizing patients to statin therapy who do not have hyperlipidemia but who do have evidence of systemic inflammation, as assessed by hsCRP.
Comment by Jonathan Abrams, MD
Ridker et al has previously proposed that hsCRP be used as a potential marker for cardiovascular risk in healthy populations, particularly in women.2 The effect of statins on hsCRP is clearly a class effect, having been demonstrated for most of the available statins. The fascinating aspect of the AFCAP/TexCAP analysis is that individuals with a lower-than-median LDL cholesterol but a higher-than-median CRP (levels LDL < 150; CRP > 0.16 mg/dL) had a reduction of risk comparable to the higher LDL cholesterol groups. Furthermore, the finding that lower levels of LDL and CRP together resulted in lower risk and no benefit with the statin is an important observation. Thus, in primary prevention populations, measurements of hsCRP may aid in the selection of pharmacological therapy for dyslipidemia, particularly if individuals have levels > the median and/or are distinctly abnormal. What level of CRP should trigger therapy remains to be determined. Ridker et al do not make any suggestions in this regard, although they clearly are proactive in including systemic inflammation as part of CAD risk screening. It should be pointed out that Ridker is a "co-inventor on patent applications" related to hsCRP and the Dade Behring latex-based immunoassay used in this study and all of his other publications.
II. Statins in Acute Coronary Syndromes
An editorial by Fonarow and Ballantyne supports initiation of lipid-lowering therapy in all patients hospitalized for an acute coronary syndrome (ACS).3 They cite several large databases that support such an approach, including the NRMI registry, the PURSUIT trial, the MIRACL study, and the recently published Swedish data demonstrating a reduction in 1-year mortality in survivors in acute myocardial infarction throughout Sweden.4 Data from the Cleveland Clinic and elsewhere also suggest a benefit at 6-12 months follow-up in ACS individuals discharged from hospitals on a statin. However, it must be stressed that except for MIRACL, there has been no prospective, large, randomized trial assessing statin vs. placebo in ACS. Anecdotal evidence and my own observations from conversations with other cardiologists, indicate that a large number of physicians are now routinely starting statin therapy in patients with acute myocardial infarction or an unstable coronary syndrome. The MIRACL trial, which did not stratify individuals by baseline lipids, demonstrated a reduction in rehospitalization for unstable angina and need for repeat revascularization at 4 months, although no survival benefits were observed with high-dose atorvastatin with ACS. These reports do not stratify subjects by baseline lipids. Fonarow and Ballantyne provide a number of rationales supporting initiating of lipid-lowering therapy during hospitalization rather than possibly doing so later on. These arguments include the likelihood that patients and their families are concerned about subsequent cardiovascular risk (the opportune moment); initiation of such therapy may help in alleviating questions of the adverse effect of medications and need for monitoring; and the potential failure to reduce cardiovascular events by delay in outpatient statin initiation. They argue that the concerns about safety of statins in acute coronary patient cases and/or unnecessary use are not supported by the MIRACL data. Many studies confirm that drug therapy for dyslipidemia is woefully low. It thus appears that the pendulum may be shifting from inadequate use of statin therapy to a more aggressive posture. Furthermore, the recent release of the NCEP-ATPIII guidelines substantially expands the pool of patients eligible for lipid-drug therapy. The atmosphere for statin use is surely a changin’!
III. Fibrates and Stroke
More good news from the VA HDL Intervention Trial (VA-HIT). In this trial, Rubins, employed a fibrate in an older male population with established CAD, normal LDL-C, and low HDL-C, which resulted in a 22% reduction in nonfatal myocardial infarction and cardiovascular death. Other cardiovascular end points were reduced. In a new report from the VA-HIT investigators, it is demonstrated that gemfibrozil also reduced the incidence of stroke during the study. The decrease of risk was substantial, with a relative risk reduction of 31%; P = 0.036. Event curves began to separate within 6-12 months. These data suggest that individuals with lower levels of HDL enjoyed a greater benefit. The results should be viewed in concert with most of the large statin trials which also have shown a reduction in stroke and TIAs in patients with preexisting CAD, but no prior cerebrovascular disease. While it has been previously believed that cholesterol and cerebrovascular events do not correlate well, and that adverse lipid levels had little to do with ischemic stroke, it now is quite clear from the aggregate data from multiple statin trials and now with gemfibrozil in VA-HIT, that aggressive treatment of dyslipidemia in CAD patients may prevent or at least slow down cerebrovascular disease. The actions of gemfibrozil are of great interest, and may involve more than raising HDL-C and lowering triglycerides. Anti-inflammatory effects, improving endothelial function, and reduction in the prothrombotic milieu have been noted. These drugs act through the PPAR gamma transcription factor cascade and therefore may have additional cellular and metabolic effects that will be confirmed in the years ahead. It is reassuring, and perhaps surprising, that gemfibrozil was able to reduce coronary and cerebrovascular events in this high-risk population of men with dyslipidemia, all of whom had a different lipid profile than those in the classic secondary prevention trials with statins (4S, CARE, and LIPID).
References
1. Downs et al. JAMA. 1998;279:1615-1622.
2. Ridker et al. N Engl J Med. 2000;342:836-843.
3. Fonarow GC, Ballantyne CM. Circulation. 2001;103:2768-2770.
4. Stenestrand U, Wallentin L. JAMA. 2001;285:430-436.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content