(Circulation. 1996;93:638-640.)
© 1996 American Heart Association, Inc.
Articles |
Have Long-term Benefits of Antihypertensive Therapy Been Underestimated?
Provocative Findings From the Framingham Heart Study
From the Boston (Mass) University School of Medicine.
Correspondence to Aram V. Chobanian, MD, Boston University School of Medicine, 80 E Concord St, Boston, MA 02118.
Key Words: Editorials • hypertension • trials
The article by Sytkowski et al1 in this issue of Circulation provides an interesting analysis of secular trends of hypertension, antihypertensive therapy, cardiovascular complications, and overall mortality in the Framingham Heart Study population. Successive cohorts of men and women between the ages of 50 and 59 years were followed for a 20-year period beginning in either 1950, 1960, or 1970, thereby spanning the 40-year period during which the most important advances in the control of hypertension have been made. The authors suggest that the long-term benefits of the control of hypertension for cardiovascular disease (CVD) and total mortality may be much greater than those observed in prior shorter-term clinical trials.
These secular trend data must be interpreted with caution because
of the inability to control for certain variables that could have
led to unknown selection bias. The treatment was not controlled, and
major changes undoubtedly occurred in drug therapy during the 40 years
of follow-up, during which several new classes of antihypertensive
drugs became available. The diagnostic criteria for
clinical end points changed as a result of new methods that provided
greater specificity and earlier recognition of CVD. Secular trends
occurred in other cardiovascular risk factors, such as
serum cholesterol and smoking, and in overall
cardiovascular mortality during the study period. The
definitions used for hypertension (systolic pressure 
160
mm Hg and/or diastolic pressure 95 mm Hg) and for its
control (blood pressure <160/95 mm Hg) changed nationally during the
period of study, and treatment was probably initiated in some of the
patients at blood pressure levels <160/95 mm Hg. In addition, those
who received antihypertensive therapy may have adopted healthier
lifestyles. Furthermore, the attention focused on Framingham Study
participants over the years may have enhanced their motivation to seek
treatment for their abnormal cardiovascular risk
factors. In fact, in the 1970 hypertensive cohort, 89% of women and
78% of men received drug treatment, figures that are substantially
greater than those reported in National Health and Nutrition
Examination Surveys (NHANES) during this period.2 3
Despite these various caveats and concerns, the remarkable differences in both total and CVD mortality observed between treated and untreated hypertensive patients cannot be readily dismissed. The investigators are well aware of the complexity of the analyses and have controlled for as many variables as possible in their calculations. When the data in all three cohorts were combined and levels of other risk factors adjusted for by Cox regression analysis, the risk of death from CVD was 60% lower in treated than in untreated individuals. In addition, all-cause mortality in treated hypertensive patients was 31% lower than in untreated counterparts. These apparent benefits of therapy are substantially greater than those reported in controlled clinical trials that have evaluated the effects of antihypertensive drug therapy on CVD. Meta-analyses of such trials have suggested an overall 35% to 40% lower stroke incidence and a 14% to 18% lower coronary heart disease (CHD) incidence in treated compared with control subjects, but the effects on total mortality have been inconsistent.4 The average treatment period in such trials was only between 2 and 3 years, and the amount of reduction in blood pressure compared with control subjects was modest, averaging 5 to 6 mm Hg of diastolic blood pressure. In most of these controlled trials, hypertensive patients at highest risk were excluded, so that the overall risk in the study participants was probably not representative of that present in the general hypertensive population.
The apparent large reductions in mortality with antihypertensive therapy in this study might be attributable in part to the age of the participants. The secular trends were assessed in subjects whose follow-up began while they were all in the 50- to 59-year age range, so that by the end of the 20-year follow-up period, participants were between 70 and 79 years of age. In recent controlled trials involving elderly hypertensive patients, such as the Systolic Hypertension in the Elderly Program, the Swedish Trial in Old Patients with Hypertension, and the Medical Research Council trial in the elderly, treatment of hypertension was associated with larger reductions in CHD incidence than had been reported earlier in trials involving younger subjects.5
In the present study, CVD end points were determined during the first 10 years of follow-up, while mortality was assessed in the second 10-year period. As the authors point out, such handling of data may have led to a lower number of CVD events and an underestimation of the effects of treatment on mortality. The lack of effect of blood pressure control on CVD incidence during the first 10 years of follow-up is somewhat surprising in light of the subsequent marked effects on mortality, although a similar apparent dissociation between cardiovascular events and CVD mortality has also been observed in other studies. Since increasingly sensitive methods for detecting CVD events were introduced during the period 1950 to 1990, the reporting of nonfatal clinical complications would be expected to rise. In addition, the new events might not have been as lethal in later years of the study because of the many advances that have occurred in the treatment of clinical complications of CVD. The possibility also exists that the major benefits of treatment occurred in patients with existing CVD, as has been reported for cholesterol-lowering approaches.6 Little information is available in the literature on this latter issue, and serious consideration needs to be given in the future to the assessment of the relative benefits of blood pressure lowering on both the primary and secondary prevention of CVD.
Few data are available regarding the effects of the duration of control
of hypertension or of any other abnormal cardiovascular
risk factor on CVD incidence. In the initial phase of the Multiple Risk
Factor Intervention Trial, which had an average follow-up period of
6.9 years, multifactor intervention was not associated with any
significant effect on mortality.7 However, when the
follow-up was extended to 10.5 years, the original special
intervention group had a 10.6% lower mortality from CHD and a 7.7%
lower total mortality than the usual-care group.8
Conversely, in the Oslo Hypertension Study, which involved a relatively
small group of 785 relatively young, mildly hypertensive patients,
treatment for 10 years failed to influence mortality.9 For
cholesterol treatment trials involving reduction of serum
cholesterol of 5 years' duration, it has been estimated
that for every 1-mg/dL decrease in serum cholesterol, an
average 2% reduction in CHD mortality would be expected, whereas in
briefer trials, CHD mortality would be decreased by 1% for each
1-mg/dL cholesterol reduction.6
The extent of the benefit of risk factor modification on the course of CVD may thus depend on several factors, including the duration of treatment, the degree of correction of the abnormality, and the presence or absence of existing vascular disease or CVD complications. Unfortunately, a major gap exists in our current knowledge of how hypertension and other risk factors affect the arterial wall and how such effects cause clinical complications. Experimental studies have demonstrated that hypertension causes injury to the arterial intima in a manner that resembles the arterial damage induced by hypercholesterolemia and that hypertension markedly accelerates the development of atherosclerosis in the presence of hypercholesterolemia.10 Clinically, coronary artery occlusion appears to be related most often to formation of a thrombus at the site of an atherosclerotic plaque, and recent experimental data suggest that such thrombi may be initiated because of plaque instability and rupture. Blood pressure reduction could influence not only the development of atherosclerosis but also the risk for plaque rupture by reducing hemodynamic stresses on the arterial wall. Plasma cholesterol reduction, on the other hand, may decrease entry of lipoproteins into the artery and thereby inhibit accumulation of macrophages. Such accumulation is now thought to occur at regions associated with plaque instability, where macrophages may release degradative enzymes that could promote plaque breakdown. A better understanding of these processes would be extremely valuable for designing new therapeutic approaches for the prevention of cardiovascular complications.
Although the data presented here can only point to guilt by association, these new findings from the Framingham Study are nevertheless worthy of serious consideration. Should the data accurately reflect the long-term benefits of blood pressure control in the general population, then a major reassessment of the value of antihypertensive treatment would need to occur. Partly on the basis of data from the relatively short-term trials of hypertension, it has generally been concluded that control of hypertension has been of only minor importance in explaining the overall reduction in CHD mortality in this same period, even though it has played a major role in reducing stroke incidence in the United States during the past 25 years. Concerns have been raised about the more than $7 billion per year currently being spent on antihypertensive drugs11 and the overall value of treating many of the patients with mild hypertension. However, if prior estimates have grossly understated the benefits of long-term therapy, the impact on future health policy could be considerable. The Framingham Study is unique, and it probably will be difficult to confirm or refute the findings presented in the current report. Nevertheless, it may prove to be of value to analyze similar secular trends using data from other studies that have provided long-term follow-up of large numbers of hypertensive individuals. Possible candidates for such evaluation might include the Honolulu Heart Study, the Tecumseh Study, the Atherosclerosis Risk in Communities Study, the NHANES data, and data from the Kaiser-Permanente healthcare system. With more than 50 million hypertensive individuals in this country and the major human and financial costs of the disease, it remains of critical importance to determine the long-term consequences of hypertension and its control.
 |
Footnotes |
|---|
The opinions expressed in this editorial are not necessarily those of the editor or of the American Heart Association.
|
References |
|---|
 TopIntroduction References |
|---|
-
Sytkowski PA, D'Agostino RB, Belanger AJ, Kannel
WB. Secular trends in long-term sustained hypertension,
long-term treatment, and cardiovascular mortality:
the Framingham Heart Study 1950-1990.
Circulation. 1996;93:697-703.
[Abstract/Free Full Text]
-
Centers for Disease Control, National Center for
Health Statistics, Third National Health and Nutrition Examination
Survey. 1988-1991. NIH publication.
-
The 1993 Report of the Joint National Committee on
Detection, Evaluation, and Treatment of High Blood Pressure. The
Fifth Report of the Joint National Committee on Detection, Evaluation,
and Treatment of High Blood Pressure. Arch Intern
Med. 1993;153:154-183. [Medline]
[Order article via Infotrieve]
-
Collins R, Peto R, MacMahon S, Hebert P, Fiebach NH,
Eberlein KA, Godwin J, Qizilbash N, Taylor JO, Hennekens CH.
Blood pressure, stroke, and coronary heart disease, II:
short-term reductions in blood pressure: overview of randomised
drug trials in their epidemiological context.
Lancet. 1990;335:827-838. [Medline]
[Order article via Infotrieve]
-
Insua JT, Sacks HS, Lau T-S, Lau J, Reitman D, Pagano
D, Chalmers T. Drug treatment of hypertension in the elderly: a
meta-analysis. Ann Intern Med. 1994;121:355-362.
[Abstract/Free Full Text]
-
Expert Panel on Detection, Evaluation, and Treatment
of High Blood Cholesterol in Adults (Adult Treatment Panel
II). Summary of the Second Report of the National
Cholesterol Education Program (NCEP).
JAMA. 1993;269:3015-3023. [Medline]
[Order article via Infotrieve]
-
Multiple Risk Factor Intervention Trial Research
Group. Multiple risk factor intervention trial: risk factor
changes and mortality results. JAMA. 1982;248:1465-1477. [Abstract]
-
The Multiple Risk Factor Intervention Trial Research
Group. Mortality rates after 10.5 years for participants in the
Multiple Risk Factor Intervention Trial: findings related to a priori
hypotheses of the trial. JAMA. 1990;263:1795-1801. [Abstract]
-
Leren P, Helgeland A. Oslo Hypertension
Study. Drugs. 1986;31 (suppl 1):41-45.
-
Chobanian AV. Adaptive and maladaptive responses
of the arterial wall to hypertension: the 1989 Corcoran
Lecture. Hypertension. 1990;15:666-674.
[Abstract/Free Full Text]
-
Manolio TA, Cutler JA, Furberg CD, Psaty BM, Whelton
PK, Applegate WB. Trends in pharmacologic management of
hypertension in the United States. Arch Intern Med. 1995;155:829-837. [Abstract]
This article has been cited by other articles:
 |
|
 |
|
  O. K Andersson, T. Almgren, B. Persson, O. Samuelsson, T. Hedner, and L. Wilhelmsen Survival in treated hypertension: follow up study after two decades BMJ, July 18, 1998; 317(7152): 167 - 171. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||