[Skip to Navigation]
Citations 0
Editor's Correspondence
December 13/27, 2010

Are Statins Effective in High-Risk Primary Prevention?—Reply

Author Affiliations

Author Affiliations: Cardiac and Vascular Sciences, St George's University of London, London, England (Dr Ray); Department of Public Health and Primary Care, University of Cambridge, Cambridge, England (Dr Seshasai); Department of Cardiology, Leiden University Medical Center, Leiden, the Netherlands (Dr Jukema); and BHF Glasgow Cardiovascular Research Centre, Faculty of Medicine, University of Glasgow, Glasgow, Scotland (Dr Sattar).

Arch Intern Med. 2010;170(22):2042-2044. doi:10.1001/archinternmed.2010.459

In reply

We thank Brugts and Deckers for their interest in our article.1 The issue of statins for the primary prevention of nonfatal events and cardiovascular death has been dealt with in the Cholesterol Treatment Trialists' (CTT) collaborative meta-analysis (see Web Figure 1 in the original CTT meta-analysis publication2), and the only unanswered question that remained was whether statins reduced all-cause mortality in a pure primary prevention population, as prior and meta-analyses subsequent to CTT had included mixed populations. This remaining uncertainty has now been addressed by our study, by obtaining individual tabular data from trials with mixed primary and secondary prevention populations to produce a clean data set without individuals with known prevalent disease. We believe that the reason that all-cause mortality is only modestly influenced by statin use is related to the fact that cardiovascular deaths contribute to only a small proportion of deaths over 3 to 4 years in these trials despite the presence of risk factors (ie, despite the presence of risk factors, our positive predictive power for identifying individuals at risk of dying and who may benefit from statins is low). In contrast, among those with prevalent disease, this contribution reaches 40% to 50% of deaths, and hence statins show a significant benefit with smaller trial populations and in a shorter time scale. We respectfully disagree with the suggestion that statins are more beneficial in older hypertensive patients than among those of middle age with hyperlipidemia, as shown by the lack of statistical interaction per 1-mmol/L reduction in low-density lipoprotein cholesterol level and benefit in different subgroups.2 The proportional reduction in risk remains the same but, as per expectations, the absolute benefit varies with absolute risk.

Add or change institution