Evidence based management of hypertension: Cardiovascular risk factors and their effects on the decision to treat hypertension: evidence based review

Abstract

Hypertension and cardiovascular risk Relative risk Most population based studies confirm that hypertension increases an individual's risk of various cardiovascular consequences approximately two to three times (figure). Large population based cohort studies consistently show continuous, strong, and graded relations between blood pressure (particularly systolic pressure) and the subsequent occurrence of various atherosclerotic events. 2 3 The sizes of the relative risks reported in each study depend on the duration of follow up and the definition of hypertension in use.4 These relative risks are consistent across all settings5 and for all patient subgroups, including those with and without known atherosclerotic disease.6 View larger version: In this window In a new window Risk of atherosclerotic disease in people with hypertension Multiple high quality long term cohort studies and randomised clinical trials have shown that the risks from raised blood pressure can be partially reversed. 6 7 Two important issues, however, remain unclear: the exact reduction in pressure that will achieve the greatest reduction in cardiovascular risk, and whether the benefits of treatment are specifically related to the extent the pressure is lowered (see next paper in this series). Hypertension is implicated in 35% of all atherosclerotic cardiovascular events,2 including 49% of all cases of heart failure.8 Absolute risk As hypertension is only one of the many risk factors for cardiovascular disease, a patient's prognosis depends more on the sum of their risk factors than on their blood pressure. 2 5 Numerous methods to calculate a patient's absolute cardiovascular risk have been described (table 1). View this table: In this window In a new window Table 1 Tools for determining cardiovascular prognosis in individual patients Guidelines (for the management of both hypertension and hyperlipidaemia) generally now recommend the use of simplified versions of the Framingham risk equations for formal estimation of risk and specify absolute risk treatment thresholds. Framingham risk equations—The Framingham risk equations were developed to predict coronary disease, heart failure, or stroke,9–11 and they estimate the 10 year risk of each event and the average risk in controls matched for age and sex. Although the Framingham investigators have urged caution in extrapolating from their cohort of predominantly middle class white people, the risk equations have been shown to be reasonably accurate when applied to other populations in northern Europe and the United States (although they may overestimate risk elsewhere).12 The equations have been criticised for not including several atherosclerosis risk factors (such as family history, sedentary lifestyle, and obesity). Cardiovascular disease life expectancy model—The cardiovascular disease life expectancy model is a Markov model developed using data from the Lipid Research Clinics Follow-up Cohort, the Canadian Heart Health survey, and Canadian life tables.13 It has two key advantages over the Framingham risk equations. Firstly, it provides a single estimate for the risk of non-fatal or fatal coronary events and strokes in any one person (the Framingham equations for coronary events and for strokes are different, and there is no way of combining them). Secondly, this model was derived from a cohort of patients with and without overt coronary heart disease and thus can be used to predict the potential benefits (and cost effectiveness) of modifying risk factors both before and after the development of overt atherosclerotic disease (the Framingham equations were derived only from people without coronary disease). The major disadvantage of this model is that it requires access to the original formulas and is not yet available in a simple form. Dundee coronary risk disk—The Dundee coronary risk disk provides an estimate of a patient's relative risk for coronary mortality matched for age and sex.14 It was derived solely in men and has not been independently validated in women; there is no information on its generalisability to other populations; and its predictions correlate only moderately well with the Framingham estimates.12 PROCAM risk function—Estimates derived from the PROCAM risk function15 correlate reasonably well with those derived from the Framingham equation, but it cannot be used to predict coronary risk in women and, again, its generalisability to other populations is unknown.12 British regional heart study risk function—The British regional heart study function16 has never been validated in an independent test set, cannot be used to predict coronary risk in women, and has been found to systematically underestimate risk when compared with all other risk functions.17