Date Published: September 8, 2009
Publisher: Public Library of Science
Author(s): Leonard Kritharides
Abstract: In a commentary on two new research studies in PLoS Medicine, Leonard Kritharides discusses the role of inflammatory markers in predicting cardiovascular outcomes and patients’ responses to treatment.
Partial Text: Inflammatory processes are clearly implicated in the aetiology of vascular disease and in its sequelae . Atherosclerosis causes ischaemia and infarction by the chronic or acute occlusion of arteries, and inflammatory cells have been identified in atherosclerotic lesions. Systemic markers of inflammation such as interleukin-6 (IL-6), C-reactive protein (CRP), fibrinogen, and white cell count have previously been shown to be associated with increased risk of cardiovascular events in apparently healthy populations and patients with established vascular disease (e.g., ,), and in predicting the risk of, and outcome after, stroke (reviewed in ). Although inflammatory markers do improve prediction of cardiovascular events, their predictive value may be modest . Others have argued that the use of receiver operating characteristic (ROC) curves and C-statistics to decide the value of biomarkers may underestimate their value in certain populations . Recently, prediction of mortality was shown to be improved when combinations of biomarkers are used . Because healthy populations with CRP over 2 mg/dl show reduced rates of fatal and nonfatal cardiovascular events in response to treatment with statin , it is clinically important to clarify the role of inflammatory markers in predicting outcomes and the response to treatment.
Two studies published in PLoS Medicine—one previously  and the other in this issue —that investigate the predictive value of inflammatory markers in cardiovascular disease arrive at apparently conflicting conclusions. Sattar et al. investigated over 5,000 men and women who were aged 70–82 y and at high risk of suffering cardiovascular events as part of the PROSPER study . In these individuals, who were randomised to placebo or treatment with pravastatin, IL-6, CRP, and fibrinogen all predicted cardiovascular death, and IL-6 in particular added significantly to conventional risk factors in predicting those suffering fatal myocardial infarction or fatal stroke. In contrast, inflammatory markers were weak predictors of nonfatal cardiovascular disease. These conclusions were unaffected by treatment with pravastatin. The implications are that the many studies pooling fatal and nonfatal cardiovascular outcomes may have inadvertently underestimated the true predictive power of inflammatory markers to predict fatal AU cardiovascular events.
If the authors of the two studies have reached different conclusions on the utility of inflammatory markers, can, or should, the results of the two studies be reconciled? Sattar et al. used inflammatory markers to predict cardiovascular mortality in a large population without current infarctions, whereas Whitely et al. used inflammatory markers to predict functional outcomes in patients presenting with stroke. In the former, inflammatory markers provided insights into inflammation as a predictor or cause of life-threatening plaque rupture and new arterial occlusions, or the survival after such an event. In the latter, the relationship of inflammatory markers to the extent of neurological injury  and to comorbidity, rather than propensity to new arterial occlusions, was assessed. Thus the pathologies in the two studies are quite distinct. In addition, there are important differences in the populations being studied and the questions being posed.
The study by Sattar et al. will prompt reanalysis of the predictive value of IL-6, CRP, and other inflammatory markers in various patient populations. Clinicians may need to consider cardiovascular mortality, rather than total cardiovascular events, when combining inflammatory biomarkers with conventional Framingham criteria to calculate absolute risk. The study by Whiteley et al. will reassure clinicians that clinically meaningful prediction of outcomes after stroke is feasible using a simple clinical score. Further studies that include patients with more severe strokes will be needed to resolve the question of whether inflammatory markers have predictive value in this population.