Back grey_arrow_rt.gif
Largest-ever meta-analysis finds CRP is unlikely to be causal for CVD
DECEMBER 21, 2009 | Lisa Nainggolan
Cambridge, UK - In the largest and most comprehensive meta-analysis to date looking at C-reactive-protein (CRP) levels and risk of coronary heart disease (CHD) and stroke, British researchers conclude that CRP is unlikely to be a causal factor for cardiovascular disease [1].
Although CRP concentration was linearly associated with CHD, stroke, and vascular mortality, as well as nonvascular mortality, statistical adjustment for conventional cardiovascular risk factors "resulted in considerable weakening of associations," note the scientists of the Cambridge-based Emerging Risk Factors Collaboration (ERFC), who report their findings online December 21, 2009 in the Lancet.
In an editorial accompanying the paper [2], Drs S Matthijs Boekholdt and John JP Kastelein (Academic Medical Center, Amsterdam, the Netherlands) say the UK authors "are to be commended for this impressive data set." Although the findings "add weight to the evidence of noncausality" for a role of CRP in the development of cardiovascular disease, "the debate can be resolved only by randomized trials with agents that specifically target CRP, and such compounds are currently under development," say the Dutch doctors.
Commenting on the new meta-analysis for heartwire, Dr Paul Ridker (Brigham and Women's Hospital, Boston, MA), a long-time advocate of CRP and the lead investigator of the JUPITER trial, said: "Whether or not CRP is 'causal' for heart disease is neither the crucial issue at hand nor relevant for public health. What is crucial is getting international agreement that CRP identifies higher-risk individuals who would not otherwise qualify for a life-saving therapy, and then showing that such individuals clearly benefit from treatment. The new meta-analysis demonstrates the former, and JUPITER demonstrates the latter."
In this "well-done" meta-analysis, the hazard ratio associated with CRP after adjustment for traditional risk factors is "actually larger than the hazard associated with non-HDL cholesterol or blood pressure," Ridker observes. So these new data "provide strong international confirmation that the impact of inflammation on heart disease is at least as important as that of other widely accepted risk markers."
Adjusting for risk factors weakens CRP association with CVD
The ERFC researchers say their meta-analysis differs from previous reports in several important ways. As well as being the largest and most comprehensive, the study excluded individuals with a history of CHD or stroke, thereby reducing any effects of clinically evident disease on CRP concentration. They were also able to use individual participant data, which "enabled a consistent approach to adjustment for potential confounders and exploration of subgroups." And they corrected for some other statistical variations, too.
The individual records of 160 309 people were included from 54 long-term prospective studies, representing 1.31 million person-years at risk. Risk ratios (RRs) for CHD per one-standard-deviation (SD)-higher log CRP concentration (threefold higher) were 1.63 for when adjusted for age and sex only and 1.37 when adjusted further for conventional risk factors. Equivalent RRs for ischemic stroke were 1.44 and 1.27; 1.71 and 1.55 for vascular mortality; and 1.55 and 1.54 for nonvascular mortality (including deaths from several cancers and lung disease).
After further adjustment for fibrinogen, the corresponding RRs were 1.23 for CHD, 1.32 for ischemic stroke, 1.34 for vascular mortality, and 1.34 for nonvascular mortality.
"Although our results support the idea that some process related to persistent inflammation is associated with vascular disease and other chronic disorders, most of the association with ischemic vascular disease depends on conventional risk factors and fibrinogen concentration," state the researchers.
They go on to discuss previous genetic studies that have found null associations between CRP-related genotypes and fibrinogen concentrations, conventional risk factors, and CHD. "Hence our findings—taken with available genetic data—reduce the likelihood of causality for CRP in CHD."
They also note that although rosuvastatin reduced the risk of first-ever vascular disease in those with lower-than-average cholesterol concentrations and higher-than-average CRP in JUPITER—which was the basis of last week's decision by an FDA advisory committee to vote in favor of expanding the indication for this statin—studies of statins "cannot provide specific causal inferences about CRP concentration," they point out.
More analyses to come
In their editorial, Boekholdt and Kastelein say the meta-analysis "will fuel the ongoing debate about the relevance of CRP in clinical practice," including whether CRP can help guide decision-making for the primary prevention of cardiovascular disease.
The "wealth of data collected by the ERFC will be an excellent source for future analyses," they write.
The ERFC confirm that they plan to investigate this in subsequent studies. "A second topic of debate," say the Dutch doctors, "is whether CRP has a causal role in the development of cardiovascular disease." Because the associations of CRP concentrations and various cardiovascular outcomes were attenuated following adjustments for known risk factors, "the investigators interpret these findings as supportive of noncausality."
The editorialists agree that the evidence from the new meta-analysis tips the scales away from a causal role for CRP in CVD development but note that randomized trials with agents targeting CRP will be needed to fully inform this discussion.
"We need to be moving forward to improve health and we can do that now, even if CRP is 'only a marker.' "
But even if CRP is not involved in atherogenesis and plaque rupture, "it may still be a valuable tool in cardiovascular medicine," Boekholdt and Kastelein state.
Ridker said: "It could take years to precisely sort out the causal pathways that do or do not link inflammation to heart disease, but that argument is secondary to what we already know to be true—that individuals with increased CRP are at increased cardiac risk and that we can dramatically lower that risk with statin therapy.
"This is about improving patient care, which is the bottom line. We need to be moving forward to improve health and we can do that now, even if CRP is 'only a marker,' " he concludes.
The meta-analysis was funded by the British Heart Foundation, UK Medical Research Council, BUPA Foundation, and GlaxoSmithKline. Disclosures for the authors are listed in the paper. Boekholdt and Kastelein declare they have no conflicts of interest. Ridker is listed as a coinventor on patents held by the Brigham and Women's Hospital that relate to the use of inflammatory biomarkers in cardiovascular disease, including CRP.
The Emerging Risk Factors Collaboration. C-reactive protein concentration and risk of coronary heart disease, stroke and mortality: an individual participant meta-analysis. Lancet 2009; DOI:10.1016/S0140-6736(09)61717-7. Available at:
Boekholdt SM and Kastelein JJP. C-reactive protein and cardiovascular risk: more fuel to the fire. . Lancet 2009; DOI:10.1016/S0140-6736(09)62098-5. Available at:
Related links
FDA advisory panel votes in favor of broadened rosuvastatin indication [Lipid/Metabolic > Lipid/Metabolic; Dec 15, 2009]
JUPITER: Primary-prevention statin therapy in women cuts cardiovascular risk in half [Lipid/Metabolic > Lipid/Metabolic; Nov 25, 2009]
New Canadian lipid guidelines will recommend CRP testing for certain patients [Prevention > Prevention; Sep 24, 2009]
CRP, other biomarkers, not clinically useful to predict CVD risk [Clinical cardiology > Clinical cardiology; Jun 30, 2009]
JUPITER hits New Orleans: Landmark study shows statins benefit healthy individuals with high CRP levels [Clinical cardiology > Clinical cardiology; Nov 09, 2008]
  icon paper stack View Older Articles   Back to Top