Blood pressure measures, relative and absolute risk, and international differences : Journal of Hypertension

Journal Logo

Editorial comment

Blood pressure measures, relative and absolute risk, and international differences

Alderman, Michael H.

Author Information
Journal of Hypertension 20(9):p 1689-1691, September 2002.
  • Free

The Brisighella Heart Study [1] reports a relationship of various components of blood pressure to cardiovascular outcome that is similar to those reported from the Framingham Heart Study [2]. This is reassuring as it implies that the standard strategy of reducing cardiovascular risk factors to prevent heart attack and stroke is appropriate in both settings. However, it is an incomplete basis for the unmodified use of Framingham multifactor risk estimates as a guide for public policy or personal encounter clinical practice. What is still required is an accurate knowledge of the actual incidence of the disease. This could then be used to establish absolute or multifactor cardiovascular risk estimates that are appropriate to populations similar to those studied at Brisighella and, consequently, guide therapeutic interventions for those individuals and groups.

Professor Borghi and his colleagues and predecessors are to be congratulated for their careful design, execution and analysis of data derived from a prospective study of a representative sample of this Italian community [1]. To select, recruit and maintain such a sample is not for the faint-hearted, and its execution requires ability, resourcefulness and determination.

The data presented in this issue of the journal [1] support the growing belief that the preference for diastolic over systolic pressure, as has long been the case in clinical practice, is probably misplaced. More recently, the predictive value of pulse pressure has been recognized in several population studies, as well as among hypertensive subjects. Superiority for either systolic and pulse pressure has been difficult to establish, most likely because of their high degree of correlation and the marked interaction with age. What is clear is that systolic pressure continues to rise with age, and is associated with stiffening of the large arteries and enhanced pulse wave velocity, leading to peripheral vascular damage and cardiac hypertrophy [3].

Initial antihypertensive trials have relied primarily upon diastolic pressure to establish both entry criteria and a therapeutic goal. In most of these trials, the average systolic pressure attainment was generally at or above 140 mmHg. Protection against cardiac events achieved in these trials generally was less than predicted by the degree of diastolic pressure change achieved [4]. Three recent trials demonstrated that treatment aimed at systolic pressure reduction produced cardiovascular disease prevention superior to that seen in the earlier diastolic driven trials of younger individuals [5–7]. However, it should be noted that the average systolic blood pressure in those clinical trials generally failed to reach the currently recommended target of 140 mmHg. The Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT) was designed to compare various antihypertensive agents in older subjects with either systolic or diastolic elevation, and most patients entered the trial with elevated systolic pressure. Vigorous efforts were made to attain both systolic and diastolic goals (< 140/90 mmHg), and in-study diastolic control was realized by more than 90% of subjects, but only approximately 60% met the systolic goal [8]. The results for diabetic subjects, despite the targets being lower, were even less satisfactory. This relative inability to achieve systolic goals, as characterized in clinical trials, has also been widely documented in community practice settings [9].

There is good reason to be concerned about this failure to reach treatment goals. Observational studies have shown that the relatively higher pressures of treated hypertensive subjects are associated with an enduring excess in cardiovascular morbidity and mortality [10]. Accumulating data in the USA suggest that the incidence of both strokes and heart attacks has recently leveled off, and that the continuing decline in age-adjusted mortality may reflect more on in-hospital survival, rather than the prevention of events, which remains the primary goal of risk factor intervention! [11,12].

Despite evidence that failure to reach systolic goals is widespread, there is reason to believe that the situation is reversible In the AASK trial of blood pressure reduction in patients with renal insufficiency, participants were safely able to achieve and maintain systolic levels well below 140 mmHg [13]. This occurred in a research setting where systolic control was the target, and patients received two or three drugs according to protocol. This demonstrates that blood pressure levels well below 140 mmHg can be achieved with existing tools. Impediments to their achievement include a reluctance to titrate medicines to their full therapeutic dose, concern about too great a blood pressure reduction in older patients, and a lingering focus on diastolic pressure.

Beyond demonstrating similar results to the Framingham Heart Study in terms of relative risk and the importance of systolic pressure, the Brisighella Study offers the promise of a possibly more fundamental contribution. Because of its rigorous design and execution and long duration, data derived from the Framingham Heart Study have been the basis for risk assessment throughout the world. The data obtained by Borghi et al. [1] suggest that relative risk in both the Brisighella and Framingham studies are approximately comparable. However, this is an insufficient basis upon which to transfer the Framingham experience in unmodified form to guide risk predictions for Italians. It is also essential to know whether the absolute or multifactor risk is also similar [14]. Now that the randomized clinical trial has become the gold standard of evidence-based medicine, rigorous observational study tends to be devalued. However, there is one area in which the prospective cohort study remains the gold standard. Only observational studies can establish disease incidence. Careful follow-up of a well-described representative cohort, such as in the Brisighella study, permits establishment of the actual incidence of disease based upon different constellations of characteristics. Consequently, this becomes the database for absolute or multifactor risk determination. It is increasingly recognized that therapeutic interventions to reduce risk factors should not be based upon individual risk factor levels, but rather on absolute or multifactor risk assessment.

Absolute or multifactor risk did prove to be close enough to convince clinical scientists to apply the Framingham formulae to Northern European populations [15]. This may not be the case for Southern Europe. Data from the Brisighella study suggest that the incidence of cardiovascular disease is substantially lower in Southern Europe than in the US and Northern Europe. If these two communities differ substantially in their incidence of the disease, then application of the Framingham formulae might artificially overestimate the likelihood of an event in Brisighella. The result would mislead physicians, patients and public health policy makers about the value of intervention at any given level of risk. For example, if in community A, the baseline incidence of cardiovascular disease events in patients with a blood pressure of 140/90 mmHg, was one per 100 per year or five per 100 per 5 years, and if a fall in systolic pressure from 140 mmHg to 130 mmHg reduced events by 20%, then 100 such subjects would need to be treated for 5 years to prevent one event. By contrast, if residents in community B, at the same blood pressure, had an event rate of five per 100 or 25 per 100 per 5 years, a 20% reduction would prevent five events. Thus, in community B, only 20 individuals (instead of 100) would need to be treated for 5 years to prevent one stroke or heart attack.

The point to note is that a knowledge of community disease burden is critical to a rational determination of where both population and personal therapeutic energies should be directed. The availability of local incidence data would make it possible to adjust the Framingham formulae to more accurately predict the actual gain to be achieved by a change in any particular risk factor. My opinion is that when the local and more relevant incidence data become available, and when faced with a patient whose pressure is 140 mmHg, Doctors in Brisighella, compared to their American colleagues, will less frequently find an absolute or multifactor risk level where the potential benefit of intervention is merited.

In summary, the current report from Brisighella may be a welcome indication on what is likely to emerge. A sufficient sample size and three decades of observation provide the basis not only for determinating the relative impact of various individual factors, but also for a robust estimation of the incidence of cardiovascular disease. Armed with a more accurate guide to incidence, Italian physicians will be better equipped to ensure that treatment is directed where benefit is likely to far exceed harm. Precise guidance for public health and preventive cardiology practice depends upon the accumulation of similar data in regions around the world where disease incidence is likely to differ. The task is daunting and time consuming, but the Brisighella experience shows that careful long-term community-based observational studies can be performed.


1. Borghi C, Dormi A, Ambrosioni E, Gaddi A. Relative role of systolic, diastolic and pulse pressure as risk factors for cardiovascular events in the Brisighella Heart Study. J Hypertens 2002; 20: 1737–1742.
2. Franklin SS, Khan SA, Wong ND, Larson MG, Levy D. Is pulse pressure useful in predicting risk for coronary heart disease? The Framingham Heart Study. Circulation 1999; 100: 354–360.
3. Benetos A, Rudnichi A, Safar M, Guize L. Pulse pressure and cardiovascular mortality in normotensives and hypertensive subjects. Hypertension 1998; 32: 560–564.
4. Collins R, Peto R, MacMahon S, Hebert P, Fiebach NH, Eberlein KA. et al. Blood pressure, stroke, and coronary heart disease. Part 2, short-term reductions in blood pressure: overview of randomized drug trials in their epidemiological context. Lancet 1990; 335: 827–838.
5. SHEP Cooperative Research Group. Prevention of stroke by antihypertensive drug treatment in older persons with isolated systolic hypertension: final results of the Systolic Hypertension in the Elderly Program (SHEP). JAMA 1991; 265: 3255–3264.
6. Staessen JA, Fagard R, Thijs L. et al., for the Systolic Hypertension in Europe (Syst Eur) Trial Investigators. Randomized double-blind comparison of placebo and active treatment for older patients with isolated systolic hypertension. Lancet 1997; 350: 757–764.
7. Liu L, Wang LG, Gong L, Liu G, Staessen JA, for the Systolic Hypertension in China (Syst China) Collaborative. Comparison of active treatment and placebo in old Chinese patients with isolated systolic hypertension. J Hypertens 1998; 16: 1823–1829.
8. The ALLHAT Officers and Coordinators for the ALLHAT Collaborative Research Group. Major cardiovascular events in hypertensive patients randomized to doxazosin vs chlorthalidone. The Antihypertensive and lipid-lowering treatment to prevent heart attack trial (ALLHAT). JAMA 2000; 283: 1967–75.
9. Izzo JL, Levy D, Black HR. Importance of systolic blood pressure in older Americans. Hypertension 2000; 35: 1021–1024.
10. Andersson OK, Almgren T, Persson B, Samuelsson O, Hedner T, Wilhelmsen L. Survival in treated hypertension: follow-up study after two decades. BMJ 1998; 317: 167.171.
11. Fang J, Alderman MH. Trend of stroke hospitalization, United States, 1988–1997. Stroke 2001; 32: 2221–2226.
12. Fang J, Alderman MH. Dissociation of hospitalization and mortality trends for myocardial infarction, US 1988–1997. Am J Med 2002; (in press).
13. Agodoa LY, Appel L, Bakris GL, Beck G, Bourgoignie J, Briggs JP. et al. Effect of ramipril vs amlodipine on renal outcomes in hypertensive nephrosclerosis. JAMA 2001; 285: 2719–2728.
14. Grundy SM, Pasternak R, Greenland P, Smith Jr S, Fuster V. Assessment of cardiovascular risk by use of multiple-risk-factor assessment equations. Circulation 1999; 100: 1481–1492.
15. Haq IU, Ramsay LE, Yeo WW, Jackson PR, Wallis EJ. Is the Framingham risk function valid for northern European populations? A comparison of methods for estimating absolute coronary risk in high-risk men. Heart 1999; 81: 40–46.
© 2002 Lippincott Williams & Wilkins, Inc.