(Circulation. 1995;92:2437-2445.)
© 1995 American Heart Association, Inc.
Articles |
Blood Pressure and Mortality Among Men With Prior Myocardial Infarction
From the Hypertension Division (J.M.F.), Bowman Gray School of Medicine, Winston-Salem, NC; Division of Biostatistics (J.N.), School of Public Health, and Division of Cardiology (R.G.), School of Medicine, University of Minnesota (Minneapolis); Division of Epidemiology and Clinical Applications (J.S., J.C.), National Institutes of Health, National Heart, Lung, and Blood Institute, Bethesda, Md; Section of General Medicine (K.E.), University of Minnesota (Minneapolis), Minneapolis VA Medical Center; and Department of Medicine (S.M.), Clinical Trials Research Unit, University of Auckland, Auckland, New Zealand.
 |
Abstract |
|---|
 Top Abstract IntroductionMethods Results Discussion References |
|---|
Background The purpose of the present study was to describe the relation between blood pressure (systolic [SBP] and diastolic [DBP]) and death from coronary heart disease (CHD) and all causes for men with a history of myocardial infarction (MI).
Methods and Results The study cohort consisted of men aged
35 to 57 years screened for the Multiple Risk Factor Intervention Trial
(MRFIT) in 1973 through 1975 and followed for survival for an average
of 16 years through 1990. There were 5362 men who reported prior
hospitalization for a heart attack of at least 2 weeks' duration at
the initial screening of MRFIT. There was a J-shaped relation between
SBP and DBP with both CHD and all-cause mortality during the first
2 years of follow-up in older (age, 45 to 57 years) men only. Risk
nadirs for SBP were 152 and 145 mm Hg, respectively, for CHD death and
all-cause mortality; corresponding DBP risk nadirs were 94 and 90
mm Hg. After the first 2 years, there was a positive association
between SBP and death from CHD and all causes. By 15 years, cumulative
CHD mortality percentages for men with screening SBP <120, 120 to 139,
140 to 159, and 
160 mm Hg were 19.7%, 21.3%, 27.5%, and 32.0%,
respectively. When deaths only after year 2 were considered, although
the linear DBP coefficient was significant, the quadratic term for DBP
was no longer significant (P>.05). However, the relation
still appeared J-shaped as cumulative mortality for those with DBP
<70, 70 to 79, 80 to 89, 90 to 99, and 100 mm Hg was 24.3%, 20.8%,
21.1%, 25.5%, and 29.7%, respectively. When the joint relation of
SBP and DBP was considered, there were no survival differences among
the four cohorts (SBP 140 and DBP <80, SBP 140 and DBP 80, SBP
140 and DBP <80, and SBP 140 and DBP 80) during the first 2
years. After 2 years, both CHD and all-cause mortality rates were
approximately 40% higher for participants with SBP 140 mm Hg versus
<140 mm Hg regardless of DBP level (<80 or 80 mm Hg).
Conclusions In this large cohort of men with prior MI, the association of SBP and DBP with CHD and all-cause mortality varied over the 16-year follow-up period. During early follow-up, in older men only, J- or U-shaped relations were evident. However, after 2 years, these same relations had become positive and graded. Given the substantial excess mortality risk in this cohort associated with high blood pressure, particularly SBP, efforts to gradually lower blood pressure should receive high priority among hypertensive men with prior MI.
Key Words: blood pressure • myocardial infarction • heart diseases • mortality
|
Introduction |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
Prospective epidemiological studies1 2 3 4 5 6 have established clearly that CHD risk relates directly to BP—both SBP and DBP—even at levels below the arbitrarily defined "hypertensive" range (
140/90 mm Hg).
For the men screened for MRFIT who reported no history of prior MI, the
relation of BP with CHD death has been shown to be graded across the
entire range of SBP and DBP and to persist over 12 years of
follow-up.7 An analysis of Framingham Heart Study data8 documented a U-shaped relation between CHD death and DBP when the BP was measured within 2 years of the coronary event. These observations conflict with another report from Framingham9 that documented a positive relation of coronary death and reinfarction risk with baseline BP level, both SBP and DBP, over longer average follow-up (9.7 years). Several hypertension studies have reported a J-shaped relation between the mean on-treatment DBP and CHD risk, although all of these studies made within-group comparisons and were therefore subject to biases attributable to the absence of a concurrent control group.10 11 12 13 14 Conversely, multiple randomized clinical trials have not confirmed the presence of a J-curve.15 16 17 18 19 20 21
Interestingly, two studies involving elderly hypertensive patients reported the existence of a J-curve in both the active drug and placebo treatment arms.22 23 The J-curve phenomenon has been the subject of a recent review10 and a meta-analysis.24
Cruickshank and colleagues11 25 suggested that lowering DBP levels below a hemodynamically critical threshold results in a paradoxical increase in risk for MI. The present report is an investigation of the relation of BP and death from CHD and all causes after 16 years of follow-up among men screened for MRFIT with a history of MI. The possibility of "reverse causality bias" was explored, ie, that low BP soon after MI is a consequence of myocardial damage that predicts a poor prognosis but is not itself the cause of mortality. Thus, the short-term (first 2 years) and long-term (3 to 16 years) mortality experience is contrasted to determine the extent to which the relation of BP and mortality during the early time periods differs from that for the majority of deaths.
|
Methods |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
The MRFIT was a randomized, multicenter primarily prevention trial designed to study the effects of lowering BP and serum cholesterol and of smoking cessation on mortality from CHD.26 Approval of the MRFIT protocol was granted by the institutional review board at each participating clinical center. Between 1973 and 1975, 361 662 men aged 35 to 57 years were screened for eligibility and entry into the trial; 5440 (1.5%) of the men reported prior hospitalization for MI of at least 2 weeks' duration. Of these, 5362 had both SBP and DBP readings, and they form the basis of this report. Details of the study design, recruitment, and screening procedures have been described elsewhere.27 All men gave informed consent.
Measurements
At the initial screening visit, SBP and DBP were
measured by
trained observers using a standard mercury sphygmomanometer according
to a standard protocol. Three readings, separated by 2-minute
intervals, were taken using the right arm with the participant seated.
The average of the second and third readings defines the BP used in
these analyses. Use of antihypertensive medication was not
determined at the initial screening visit. After the BP readings were
completed, blood was drawn for serum cholesterol
determination. Serum cholesterol concentration was measured
at 1 of 14 local laboratories with an Auto Analyzer II. These
laboratories participated in and completed the cholesterol
standardization program of the CDC. Baseline MI status was determined
by the response (yes/no) to the question of whether the patient had
been previously hospitalized for 2 or more weeks for a heart attack. A
short questionnaire was also administered that recorded the number
of cigarettes smoked per day and demographic characteristics. Only
current smoking habits were assessed; thus, former smokers cannot be
distinguished from never-smokers in this cohort. Diabetes mellitus
was defined as self-reported use of insulin or oral hypoglycemic
agents. Income was estimated by using the median income of families in
the census tract where the participant reported living. Zip codes were
matched with 1980 census data.
Vital status of men screened has been determined through December 1990 and the average of 16 years after the initial screening, with use of files from the National Death Index (1979 to 1990) and the Social Security Administration (1973 to 1986).29 Death certificates were collected and coded by a trained nosologist using the ICD9.30 Deaths from CHD were coded as acute MI (ICD9 410) and other ischemic heart disease (ICD9 411-414, 429.2).
Data Analysis
Proportional hazards regression analyses,
stratified by
clinical center and with covariates corresponding to age, serum
cholesterol level, reported cigarettes smoked per day,
ethnicity (black versus nonblack), income, and use of medication for
diabetes, were performed to assess the associations of SBP and DBP with
CHD death and all-cause mortality. To test for the presence of a J-
or U-shaped relation, linear and quadratic terms for BP were entered
into the regression model. If there was evidence of a J-shaped
relation, the BP level corresponding to the nadir of risk was estimated
using the regression coefficients (-linear/(2x quadratic
coefficient). Regression analyses using indicator variables
corresponding to specific BP categories (<120, 120 to 139, 140 to 159,
and 160 mm Hg for SBP and <70, 70 to 79, 80 to 89, 90 to 99, and
100 mm Hg for DBP) also were performed.
The proportional hazards
model assumes that the relative risk of CHD
death and all-cause mortality associated with differences in
screening BP levels are approximately constant over the 16-year
follow-up period. To examine this assumption, log(-log) plots
of the survival curves for the BP categories mentioned are given (see
Figs 2
and 3). We found departures from
parallelism in these plots,
indicating that the risk of death associated with lower BP level was
different in the early follow-up compared with the late
follow-up period. This would be expected to occur if, in a subset
of men screened, the low BP levels reflected poor health, ie, the BP
was a consequence of illness instead of vice versa (reverse causality
bias).
|
|
To compare our results with those reported from Framingham participants8 and to explore further the possibility that the relation of BP varied over follow-up, the follow-up period was divided into two parts: the first 2 years after screening and years 3 through 16. Regression analyses described above were repeated for these two time periods. In addition, age-adjusted rates using the direct method for specific BP categories are presented.
|
Results |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
Table 1
displays the baseline characteristics of
MRFIT men with a history of previous MI (SBP and DBP averaged 132/85 mm
Hg). Table 2 gives average BP levels, percent with SBP
<120 mm Hg, and percent with DBP <70 mm Hg for men who died,
according to the year of follow-up during which the death occurred.
Over an average of 16 years of follow-up, 1970 of the 5362 (36.7%)
men died. Twelve hundred sixteen men died from CHD, 207 from other CVD
causes, 300 from cancer, and 247 from non-CVD, noncancer causes. In the
first 2 years of follow-up 146 men died—106 from CHD. Average BP
levels for CHD deaths in the first 2 years of follow-up (131/84 mm
Hg) were slightly lower than for baseline levels and those deaths after
2 years (135/86 mm Hg). Approximately 25% of CHD deaths in the first 2
years had a screening SBP <120 mm Hg compared with 18.6% for deaths
after 2 years. More than 10% of CHD deaths in the first 2 years had a
DBP <70 mm Hg compared with 5.6% of CHD deaths after 2 years. Results
for all-cause mortality were similar.
|
|
SBP and DBP Mortality Risk
Fig 1a and lb give
age-adjusted CHD mortality
rates by level of SBP and DBP at screening. CHD rates were lower at DBP
levels ranging from the mid 60s to the mid 90s than at more extreme
levels of DBP in either direction. A similar but less prominent trend
was noted between SBP and CHD. Table 3 gives the
cumulative percentages of CHD and all-cause mortality at 1, 2, 5,
10, and 15 years of follow-up. For the first 2 years, little
difference in CHD mortality is evident among the SBP strata (Fig
2
and Table 3). For example, at 2 years the
cumulative
percentages of CHD deaths were 2.2%, 1.8%, 2.0%, and 2.2% for those
with SBP <120, 120 to 139, 140 to 159, and 160 mm Hg, respectively.
At 15 years, the cumulative percentages of men dying from CHD for the
four strata were 19.7%, 21.3%, 27.55, and 32.0%, ie, much higher for
the upper two strata. The pattern for all cause mortality was similar
(Table 3).
|
|
The pattern of mortality for DBP was
different. Cumulative CHD and
all-cause mortality was highest in the lowest DBP strata (70 mm Hg)
through approximately 2 years of follow-up (Fig 3
and Table 2). At 2 years, the cumulative percentages of men
dying from
CHD were 3.5%, 1.8%, 2.1%, 1.7%, and 1.5% for those with DBP <70,
70 to 79, 80 to 89, 90 to 99, and 100 mm Hg, respectively (Table
3). After approximately 2 years, curves for those in the
lowest and highest strata crossed. After that time, mortality was
highest for the highest DBP stratum (100 mm Hg); mortality was
similar for those with DBP <70 mm Hg and those with DBP 90 to 99 mm
Hg, and mortality was lowest for those with DBP 70 to 79 and 80 to 89
mm Hg. The pattern for all-cause mortality was similar to that for
CHD mortality (Table 3).
Proportional Hazards Regression Analyses
Because the relative
death rates for SBP and DBP strata differed
for the first 2 years of follow-up compared with those estimated
for subsequent years, proportional hazards regression analyses
were performed first with deaths in the first 2 years of follow-up
and then with deaths after 2 years. Table 4 (top)
summarizes the results for SBP. In these analyses, the strata
with SBP <120 mm Hg were used as the reference group. In the first 2
years, no significant differences were found among the four SBP strata
for CHD or all-cause mortality. However, consistent with
the life-table results, during the first 2 years, CHD rates for the
top three strata were lower than for the bottom stratum (relative risk
estimates from 0.85 to 0.90). For deaths after 2 years, rates in the
top strata were significantly higher than the bottom stratum for both
CHD death and all-cause mortality. Proportional hazards regression
analyses were also performed with linear and quadratic terms
for SBP. These analyses indicated that a significant U-shaped
relation was evident in the first 2 years, but not after 2 years,
between SBP and all-cause mortality. The linear coefficient for SBP
in the first 2 years was .0743 (SEM= .0350; P=.03); the
coefficient for the quadratic term was .000260 (SEM=.000121,
P=.03). The estimated nadir of risk for SBP was 143 mm Hg.
During the first 2 years, the association between SBP and CHD death was
not significant when SBP was modeled either with a linear term or with
both linear and quadratic terms. After 2 years, there was no evidence
of a J- or U-shaped relation between SBP and CHD or all-cause
mortality. For both end points, significant positive log-linear
associations were found. For CHD death, the coefficient for SBP was
.0094 (SEM=.0017); for all-cause mortality, the coefficient was
also .0094 (SEM=.0013) (P<.00001 for both).
|
Relative
risk estimates for the top four DBP strata compared with the
bottom stratum (<70 mm Hg) were close to 0.5 but not significant for
either CHD or all-cause mortality when deaths in the first 2 years
were considered. After 2 years, death rates for the top DBP stratum
(100 mm Hg) were significantly higher than the rates observed in the
bottom stratum for both CHD death and all-cause mortality.
A significant quadratic association of DBP with CHD death and all-cause mortality was found when deaths in the first 2 years were considered. For CHD, the death linear and quadratic terms were .1287 (SEM=.0537) and .000696 (SD=.000298), respectively, giving an estimated nadir of risk of 92 mm Hg. For all-cause mortality, the coefficients were .1526 (SEM=.0374) and .000851 (SEM=.000199). The estimated nadir of risk based on these coefficients was 90 mm Hg. After 2 years, there was no evidence of a quadratic relation between DBP and CHD death or death from all causes. On the contrary, for both outcomes, strong positive log-linear relations were found. For CHD, the death log-linear coefficient was .0127 (P<.00001); for all-cause mortality, the corresponding coefficient was .0112 (P<.00001).
Separate analyses were performed for the 2009 men aged 35 to 44 and the 3353 men aged 45 to 57 years. A U-shaped relation in the first 2 years of follow-up was evident only for men 45 to 57 years. For this age group, nadirs of risk for SBP were 152 and 145 mm Hg for CHD death (P=.07) and all-cause mortality (P=.05), respectively. Corresponding nadirs of risk for DBP were 94 and 90 mm Hg (P=.04 and P=.0003). For both age groups, for both SBP and DBP, strong, positive log-linear relations after 2 years were found.
Joint Effects of SBP and DBP on Mortality
Death rates during
the first 2 years of follow-up among the
four groups defined by levels of both SBP (<140 and 140 mm Hg) and
DBP (<80 and 80 mm Hg) did not differ significantly (bottom third of
Tables 3 and 4). After 2 years, CHD and
all-cause mortality rates
were approximately 40% higher for SBP 140 mm Hg regardless of level
of DBP.
|
Discussion |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
The validity of previous epidemiological observations showing a continuous and positively graded association between BP, particularly DBP, with CHD has been questioned for both persons with and without documented coronary artery disease.8 24 25
Epidemiological studies have suggested an association between low DBP and increased total mortality in the elderly,31 MI risk in hypertensive patients treated with ß-blockers,32 and CHD deaths among persons with clinical evidence of myocardial ischemia.11 Thus, our findings add new and important information to the unresolved questions regarding the relation of BP with CHD and all-cause mortality as well as the persistence of these associations over long-term follow-up in men with prior MI.
There was an agexblood pressure interaction documented for both all-cause and CHD mortality. Only among older men (45 to 57 years) were U-shaped relations between SBP and DBP with both CHD and all-cause mortality documented during early follow-up (first 2 years).
However, during the early follow-up period there was a rapid
diminution in the strength of the association of low DBP with mortality
as evidenced by a 6.8- and 4.2-fold decline in the percentage of MRFIT
men who had experienced MI and who had low DBP (<70 mm Hg) (Table
2) for CHD and all-cause mortality, respectively. A
similar but less pronounced decline was noted in the percentage of men
with a low SBP (<120 mm Hg) experiencing mortality during this same
period of observation. If low BP per se causes excess mortality, then
it should follow that the association should not be time limited, nor
should the direction of the association reverse during late
follow-up. Thus, our findings do not support the hypothesis that
low DBP increases risk of fatal MI but are more consistent with
reverse causation as the explanation of this association.
Reverse causation refers to the fact that the low DBP is a consequence of myocardial damage rather than low DBP actually causing the myocardial damage. Nevertheless, low DBP may predict subsequent coronary mortality on the basis of its strong correlation with underlying myocardial damage even though low DBP may play absolutely no role in either the genesis or precipitation of clinically manifest CHD. D'Agostino et al8 reported a quadratic, U-shaped relation between DBP and CHD death among Framingham participants with a history of MI but a direct (linear) relation for those without a history of MI. With their approach, BP levels at each biennial examination were related to CHD deaths in the next 2 years; results were then pooled over the 17 examinations (34 years of follow-up). Because only BP levels proximal (within 2 years) to CHD events were investigated, it is possible that the low BP levels were simply a marker for the extent of underlying myocardial damage. There was agreement between Framingham and MRFIT early follow-up data on risk of CHD death as the 95% confidence bounds of the Framingham linear DBP proportional hazards coefficient (.2019 to .0259) includes the MRFIT linear DBP point estimate of .1287; similarly, the 95% confidence interval for the Framingham quadratic coefficient (.000282 to .001266) contains the MRFIT quadratic point estimate of .000696.
A case-control study32 studied the association between
most recent DBP level and MI risk in hypertensive patients treated with
ß-blockers. The relative risk of MI at 60 mm Hg was 2.07 (95%
confidence interval, .86 to 5.01; P=NS), and the risk nadir
occurred at a DBP of 84 mm Hg. Conversely, treated SBP had a linear
relation with MI risk. Their data implicated pulse pressure as possibly
contributing to MI risk. An alternative to a cause-and-effect
interpretation of these findings is that their observation might be
attributable to the unexplained fall in DBP in the months before
death33 or to a fall in DBP levels secondary to
subclinical myocardial pump dysfunction before clinically recognized
MI. Langer et al35 reported the 5-year mortality
experience in a community-based cohort of men aged 75 years and
older according to their change in BP over the proceeding 11 years.
Among men (including those with ISH) with unchanged or increasing SBP
and a DBP fall 5 mm Hg (widening pulse pressure), there was a 38%
excess of mortality. It is of interest that among men with ISH, the
risk of excess mortality was approximately one half as much in men
taking antihypertensive medication than in those who were not. Thus,
elderly men with widening pulse pressure experience excess mortality,
and antihypertensive drug therapy appears to lower rather than increase
mortality risk.
Pulse pressure (SBP minus DBP) widens with advancing age, at least in part because of the age-related rise in SBP attributable to a progressive loss of arterial compliance. Older persons with disproportionate SBP elevations and wide pulse pressure are at highest risk for CVD, including CHD, compared with persons with any other BP pattern. A body of evidence exists supporting the thesis that disproportionate elevation of SBP (wide pulse pressure) is associated with a high risk of both clinical and subclinical CVDs.4 36 37 Thus, persons with the lowest pretreatment DBP are at the highest risk for CVD if their SBP is elevated, a factor that may help explain why they have high subsequent risk for CHD, even among treated individuals. Such high-risk individuals disproportionately cluster in the lowest DBP strata of hypertension treatment trials, and even after several years of successful pharmacotherapy these patients may continue to manifest a higher risk of mortality than persons with higher pretreatment and during-treatment DBP values (also narrower pulse pressures).
Previous reports of a J-curve relation between low DBP during follow-up and CHD mortality have provided some insight as to why persons with low DBP may be at increased risk for CHD. In several of these uncontrolled studies, pretreatment and/or attained SBP level appeared to have confounded the within-group association between low DBP and CHD risk.11 14
Consistent with this reasoning, several epidemiological studies have suggested that SBP better predicts subsequent CHD risk than DBP.4 38 39 We confirmed these observations in men with prior MI, as risk of CHD death during late follow-up was more closely related to SBP than DBP level.
It has been suggested that lowering DBP below a putative threshold level believed necessary to maintain adequate coronary perfusion might result in MI because since virtually all coronary blood flow occurs during diastole. Moreover, the most disadvantageous balance of coronary perfusion pressure and myocardial blood flow and oxygen demands should theoretically occur in persons with the widest pulse pressures. However, when BP is lowered by pharmacological means in these same patients, the fall in SBP is disproportionate to the reduction in DBP. Thus, antihypertensive treatment should facilitate a favorable balance between coronary perfusion and blood flow requirements. Furthermore, it is difficult to implicate pharmacological treatment of hypertension as a necessary cause of the J-curve since it has been observed in the placebo arm of two hypertension trials in the elderly22 23 as well as in the Framingham Heart Study, in both treated and untreated hypertensive men and women. More evidence against the existence of a critical DBP threshold is the 10 mm Hg lower CHD risk nadir in drug-treated compared with placebo-treated elderly hypertensive patients in the trial reported by Coope and Warrander.22
Data from SHEP support the conclusion that nonprecipitous, pharmacological BP lowering in older persons (average age, 72 years) with ISH is safe. In this study, DBP levels were lowered to the high 60s (measured in millimeters of mercury) and resulted in sizable reductions in CVD clinical events, including MI.19 Another report37 from SHEP found that pharmacological BP lowering reduced the risk of BP-related clinical events in ISH patients with noninvasively detected arterial disease, as event-free survival was 81% versus 52%, in the chlorthalisone and placebo treatment groups, respectively. These observations undermine the validity of the hypothesis that drug-treated hypertensive patients with coronary atherosclerosis achieving low DBP levels are responsible for the J curve. Dissimilar results between SHEP, a randomized, placebo-controlled clinical trial, and nonrandomized studies might be explained by the lack of concurrent control group in the latter. For example, if the relation between baseline and in-study DBP level with CHD is J-shaped in the active treatment group and the same relation is present in the placebo group, the absolute level of risk for CHD mortality may, however, be lower within each baseline DBP stratum of the active treatment group. Thus, it is likely that BP treatment does not change the J- or U-shaped risk pattern across the range of DBP within the treatment group but rather lowers the absolute level of risk at any given DBP.
There are several noteworthy limitations regarding the interpretation of our data. First, these observational data are not directly relevant to the potential effect of lowering BP with lifestyle modification and/or drug therapy. Second, there is no information available on the use of antihypertensive medication, BP change, or change in status of other potentially confounding variables such as cholesterol over the period of follow-up. Thus, we can only relate initial but not follow-up BP levels with mortality risk. Furthermore, our BP mortality regression coefficients are undoubtedly attenuated because of regression-dilution bias (BP measurement imprecision). Third, we have no information on CHD morbidity; therefore, the risk estimates reported probably underestimate the absolute magnitude of the BP-CHD relation in this cohort. Finally, the MRFIT study sample does not include women; therefore, our results may not apply to women. However, epidemiological data from the Framingham study showed that after MI, women and men have an equal reinfarction rate but women have a 22% lower risk of coronary death than men.9 Nevertheless, the risk associated with the relatively lower coronary death rates in women is quite substantial in absolute terms.
Randomized clinical trials39 40 are needed to assess prospectively the impact of BP lowering to different levels within the "normal range" (SBP <140/90 mm Hg). Clinical trials comparing aggressive with less aggressive BP lowering pose many difficult design questions.41 Findings from the recently completed Treatment of Mild Hypertension trial documented a significant clinical benefit in a cohort of men and women free of clinical CVD who were 45 to 69 years old (mostly with stage I hypertension, DBP 90 to 99 mm Hg) treated for an average of 4.4 years with lifestyle modification or lifestyle modification plus drugs from one of the five antihypertensive drug classes. BP averaged 133/82 mm Hg and 127/79 mm Hg in the lifestyle modification only (placebo) and lifestyle modification plus combined drug treatment groups, respectively. Approximately 25% of participants in the Treatment of Mild Hypertension combined drug treatment group had DBP levels of <75 mm Hg compared with only 13% of placebo participants.21 Major and other clinical events were reduced 34% (P=.03) in the lifestyle modification plus combined drug treatment group, suggesting that low normal BP conferred greater protection against BP-related clinical sequelae than obtained BP levels in the higher normal range.
Determining optimal treatment strategies and therapeutic goals for the patient with a history of previous MI is problematic because there are limited morbidity or mortality data from secondary prevention trials using hypertensive therapy as the primary intervention. Moreover, no such trials are likely to be forthcoming in the foreseeable future. Our findings do not support the thesis that low DBP actually causes CHD events. Rather, our data suggest that the association between low DBP and mortality is time limited and that the major BP-associated risk in MI survivors is attributable to high rather than low BP.
|
Selected Abbreviations and Acronyms |
|---|
|
|
Acknowledgments |
|---|
This work was supported by National Heart, Lung, and Blood Institute–National Institutes of Health contract N01-HV-487048 and grants R01-HL-46630-01A1 and R01-HL-34767.
|
Footnotes |
|---|
Reprint requests to John M. Flack, MD, MPH, Associate Professor of Surgery, Medicine, and Public Health Sciences, Bowman Gray School of Medicine, Hypertension Center, Clinical Sciences Building, Room 5180, Medical Center Blvd, Winston-Salem, NC 27157-1032.
Received November 28, 1994; revision received May 4, 1995; accepted June 4, 1995.
|
References |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
1. Dyer AR. An analysis of the relationship of systolic blood pressure, serum cholesterol, and smoking to 14-year morality in the Chicago Peoples Gas Company. J Chron Dis. 1975;28:571S-577S.
2. Reid DD, Hamilton PJS, McCartney P, Rose G. Smoking and other risk factors for coronary heart disease in British civil servants. Lancet. 1976;2:979-984. [Medline] [Order article via Infotrieve]
3. Kagan A, Harris BR, Winklestein W. Epidemiologic studies of coronary heart disease and stroke in Japanese men living in Japan, Hawaii, and California: demographic, physical, dietary, and biochemical characteristics. J Chron Dis. 1974;27:345-364. [Medline] [Order article via Infotrieve]
4. Stamler J, Neaton JD, Wentworth DN. Blood pressure (systolic and diastolic) and risk of fatal coronary heart disease. Hypertension. 1989;13(suppl I):I-2-I-12.
5. MacMahon S, Peto R, Cutler J, Collins R, Sorlie P, Neaton J, Abbott R, Godwin J, Dyer A, Stamler J. Blood pressure, stroke and coronary heart disease, part I: prolonged differences in blood pressure: prospective observational studies corrected for regression dilution bias. Lancet.. 1990;335:765-774. [Medline] [Order article via Infotrieve]
6. Flack JM, Neaton JD, Daniel B, Esunge P. Ethnicity and renal disease: lessons from Multiple Risk Factor Intervention Trial (MRFIT) and the Treatment of Mild Hypertension Study (TOMHS). Am J Kidney Dis. 1993;21(suppl I):31-40.
7.
Neaton JD, Wentworth D. Serum
cholesterol, blood pressure, cigarette smoking, and death
from coronary heart disease: overall findings and differences
by age for 316,099 white men. Arch Intern Med. 1992;152:56-64.
8. D'Agostino RB, Belanger AJ, Kannel WB, Cruickshank JM. Relation of low diastolic blood pressure to coronary heart disease death in presence of myocardial infarction: the Framingham Study. Br Med J. 1991;303:385-389.
9.
Wong ND, Cupples LA, Ostfeld AM, Levy D, Kannel
WB. Risk factors for long-term coronary prognosis
after initial myocardial infarction: the Framingham Study.
Am J Epidemiol. 1989;130:469-480.
10. Fletcher AE, Bulpitt CJ. How far should blood pressure be lowered? N Engl J Med. 1992;326:251-254. [Medline] [Order article via Infotrieve]
11. Cruickshank JM, Thorp JM, Zacharias FJ. Benefits and potential harm of lowering blood pressure. Lancet. l987;1:581-584.
12. Stewart I. Relation of reduction in pressure to first myocardial infarction in patients receiving treatment for severe hypertension. Lancet. 1979;1:861-865. [Medline] [Order article via Infotrieve]
13. Samuelsson DG, Wilhemsen LW, Pennart K, Wedel H, Berglund GL. The J-shaped relationship between coronary heart disease and achieved blood pressure in treated hypertensions: further analyses of 12 years of follow-up of treated hypertensive patients in the Primary Prevention Trial in Gothenburg, Sweden. J Hypertens. 1990;8:547-555. [Medline] [Order article via Infotrieve]
14.
Alderman MH, Ooi WL, Madhaven S, Cohen H.
Treatment-induced blood pressure: reductions and risk of myocardial
infarction. JAMA. 1989;262:920-924.
15.
Veterans Administration Cooperative Study Group on
Antihypertensive Agents. Effect of treatment on morbidity in
hypertension: results in patients with diastolic pressure
averaging 115 through 129 mm Hg. JAMA. 1967;202:1028-1034.
16.
Veterans Administration Cooperative Study on
Antihypertensive Agents. Effects of treatment on mortality in
hypertension, II: results of patients with diastolic
pressures averaging 90 through 104 mm Hg. JAMA. 1970;213:1143-1151.
17. Collins R, Peto R, MacMahon S, Cutler J, Sorlie P, Neaton J, Abbott R, Godwin J, Dyer A, Stamler J. Blood pressure, stroke, and coronary heart disease, Part 2: short-term reductions in blood pressure: overview of randomized drug trials in their epidemiologic context. Lancet. 1990;335:838.
18. Medical Research Council Working Party. Stroke and coronary heart disease in mild hypertension: risk factors and the value of treatment. Br Med J. 1988;296:1565-1570.
19.
SHEP Cooperative Research Group. Prevention of stroke
by antihypertensive drug treatment in older persons with isolated
systolic hypertension. JAMA. 1991;265:3255-3264.
20. The IPPPSH Collaborative Group. Cardiovascular risk and risk factors in a randomized trial of treatment based on the beta-blocker oxypropranolol: the International Prospective Primary Prevention Study in Hypertension (IPPPSH). J Hypertens. 1985;3:379-392.[Medline] [Order article via Infotrieve]
21.
The Treatment of Mild Hypertension Research Group. The
Treatment of Mild Hypertension Study (TOMHS): final results.
JAMA. 1993;270:713-724.
22. Coope J, Warrander TS. Randomized trials of treatment of hypertension in elderly patients in primary care. Br Med J. 1986;293:1145-1151.
23. Staessen J, Bulpitt C, Clement, De Leeuw P, Fagard R, Fletcher A, Forette F, Leonetti G, Nissinen A, O'Malley K. Relation between mortality and treated blood pressure in elderly patients with hypertension: report of the European Working Party on High Blood Pressure in the Elderly. Br Med J. 1989;298:15i2-15i6.
24.
Farnett L, Mulrow CD, Linn WD, Lucey CR, Talley
MR. The J-curve phenomenon and the treatment of
hypertension. JAMA. 1991;265:489-495.
25. Cruickshank JM. Coronary flow reserve and the J-curve relation between diastolic blood pressure and myocardial infarction. Br Med J. 1988;297:1227-1230.
26.
Multiple Risk Factor Intervention Trial Research Group.
Multiple Risk Factor Intervention Trial: risk factor changes and
mortality. JAMA. 1982;248:1465-1477.
27. Sherwin R, Kaelber CT, Kezdi P, Kjelsberg MO, Thomas HE. The Multiple Risk Factor Intervention Trial, II: the development of the protocol. Prev Med. 1981;10:402-425. [Medline] [Order article via Infotrieve]
28. Multiple Risk Factor Intervention Trial Research Group. Design considerations in the NHLBI Multiple Risk Factor Intervention Trial (MRFIT). Control Clin Trial. 1987;8:41S-53S. [Medline] [Order article via Infotrieve]
29.
Wentworth DN, Neaton JD, Rasmussen WL. An
evaluation of the Social Security Administration MBR file and the
National Death Index in the ascertainment of vital status.
Am J Public Health. 1983;73:1270-1274.
30. International Classification of Disease, 1992, 9th Revision, 4th Ed, Vols 1 and 2.
31. Langer RO, Ganiats TG, Barrett-Connor E. Paradoxical survival with higher blood pressures in older men. Br Med J.. 1989;298:1356-1357.
32.
McCloskey LW, Psaty BM, Koepsell TD, Aagaard GN.
Level of blood pressure and risk of myocardial infarction among treated
hypertensive patients. Arch Intern Med. 1992;152:513-520.
33. Conner PL, Jacobs DR, Meinert CL. Some methodology for relating serial observations to mortality in men with coronary heart disease. Am J Epidemiol. 1974;100:529. Abstract.
34. The BBB Study Group. The BBB Study: a prospective randomized study of intensified antihypertensive treatment. J Hypertens. 1988;6:693-697. [Medline] [Order article via Infotrieve]
35.
Langer RD, Criqui MH, Barrett-Connor EL, Klauber MR,
Ganiats TG. Blood pressure change and survival after age
75. Hypertension. 1993;22:551-559.
36.
Psaty BM, Furberg CD, Kuller LH, Borhani NO, Rautaharju
PM, O'Leary DH, Bild DE, Robbins J, Fried LP, Reid C. Isolated
systolic hypertension and subclinical
cardiovascular disease in the elderly: initial findings
from the Cardiovascular Health Study.
JAMA. 1992;268:1287-1291.
37. Sutton-Tyrell K, Alcorn H, Kuller L. Peripheral atherosclerosis as a predictor of cardiovascular events. Circulation. 1994;89.
38. Kannel WB, Gordon T, Schwartz MJ. Systolic versus diastolic blood pressure and risk of coronary heart disease: the Framingham Study. Am J Cardiol. 1971;27:335-346. [Medline] [Order article via Infotrieve]
39. Italian Research Group of the Seven Countries Study. Incidence and prediction of coronary heart disease in two Italian rural population samples followed-up for 20 years. Acta Cardiol. 1982;37:129-145. [Medline] [Order article via Infotrieve]
40. Modification of Diet in Renal Disease Study Group. The Modification of Diet in Renal Disease Study: design, methods, and results from the feasibility Study. Am J Kidney Dis. 1992;20:18-33. [Medline] [Order article via Infotrieve]
41.
Cooper SP, Hardy RJ, Labarthe DR, Hawkins CM, Smith EO,
Blaufox MD, Cooper CJ, Entwisle G, Maxwell MH. The relation
between degree of blood pressure reduction and fatality among
hypertensive patients in the Hypertension Detection and Follow-up
Program. Am J Epidemiol. 1988;127:387-403.
This article has been cited by other articles:
 |
|
 |
|
  C. M.M. Licht, E. J.C. de Geus, A. Seldenrijk, H. P.J. van Hout, F. G. Zitman, R. van Dyck, and B. W.J.H. Penninx Depression Is Associated With Decreased Blood Pressure, but Antidepressant Use Increases the Risk for Hypertension Hypertension, April 1, 2009; 53(4): 631 - 638. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. E. Kjeldsen, T. A. Aksnes, R. H. Fagard, and G. Mancia CHAPTER 13 Hypertension ESC Textbook of Cardiovascular Medicine, January 1, 2009; 2(1): med-9780199566990-chapter - med-9780199566990-chapter. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. J. Thune, J. Signorovitch, L. Kober, E. J. Velazquez, J. J.V. McMurray, R. M. Califf, A. P. Maggioni, J. L. Rouleau, J. Howlett, S. Zelenkofske, et al. Effect of Antecedent Hypertension and Follow-Up Blood Pressure on Outcomes After High-Risk Myocardial Infarction Hypertension, January 1, 2008; 51(1): 48 - 54. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. S. Ganotakis, I. F. Gazi, J. A. Papadakis, I. A. Jagroop, D. R. Nair, and D. P. Mikhailidis The Relationship Between Circulating Fibrinogen and Lipoprotein (a) Levels in Patients With Primary Dyslipidemia Clinical and Applied Thrombosis/Hemostasis, January 1, 2007; 13(1): 35 - 42. [Abstract] [PDF]
|
|
|
|
 |
|
 H. A. Liszka, A. G. Mainous III, D. E. King, C. J. Everett, and B. M. Egan Prehypertension and Cardiovascular Morbidity Ann. Fam. Med, July 1, 2005; 3(4): 294 - 299. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. A. Papadakis, D. P. Mikhailidis, G. E. Vrentzos, A. Kalikaki, I. Kazakou, and E. S. Ganotakis Effect of Antihypertensive Treatment on PlasmaFibrinogen and Serum HDL Levelsin Patients with Essential Hypertension Clinical and Applied Thrombosis/Hemostasis, April 1, 2005; 11(2): 139 - 146. [Abstract] [PDF]
|
|
|
|
|
|
M. Kikuya, T. Ohkubo, K. Asayama, H. Metoki, T. Obara, S. Saito, J. Hashimoto, K. Totsune, H. Hoshi, H. Satoh, et al. Ambulatory Blood Pressure and 10-Year Risk of Cardiovascular and Noncardiovascular Mortality: The Ohasama Study Hypertension, February 1, 2005; 45(2): 240 - 245. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. J. Mason, J. E. Manson, H. D. Sesso, C. M. Albert, M. J. Chown, N. R. Cook, P. Greenland, P. M Ridker, and R. J. Glynn Blood Pressure and Risk of Secondary Cardiovascular Events in Women: The Women's Antioxidant Cardiovascular Study (WACS) Circulation, April 6, 2004; 109(13): 1623 - 1629. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. M.M. Lawes, D. A. Bennett, V. L. Feigin, and A. Rodgers Blood Pressure and Stroke: An Overview of Published Reviews Stroke, April 1, 2004; 35(4): 1024 - 1033. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. M.M. Lawes, D. A. Bennett, V. L. Feigin, and A. Rodgers Blood Pressure and Stroke: An Overview of Published Reviews Stroke, March 1, 2004; 35(3): 776 - 785. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Havas, E. J. Roccella, and C. Lenfant Reducing the Public Health Burden From Elevated Blood Pressure Levels in the United States by Lowering Intake of Dietary Sodium Am J Public Health, January 1, 2004; 94(1): 19 - 22. [Abstract] [Full Text]
|
|
|
|
 |
|
 PROGRESS Collaborative Group Effects of a perindopril-based blood pressure lowering regimen on cardiac outcomes among patients with cerebrovascular disease Eur. Heart J., March 1, 2003; 24(5): 475 - 484. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Pacific Study Group and B. Neal Effects of the vasopeptidase inhibitor, omapatrilat, in 723 patients with coronary heart disease Journal of Renin-Angiotensin-Aldosterone System, December 1, 2002; 3(4): 270 - 276. [Abstract] [PDF]
|
|
|
|
 |
|
 J Amar, B Chamontin, J Ferrieres, N Danchin, O Grenier, C Cantet, and J-P Cambou Hypertension control at hospital discharge after acute coronary event: influence on cardiovascular prognosis--the PREVENIR study Heart, December 1, 2002; 88(6): 587 - 591. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Bur, H. Herkner, M. Vlcek, C. Woisetschlager, U. Derhaschnig, and M. M. Hirschl Classification of Blood Pressure Levels by Ambulatory Blood Pressure in Hypertension Hypertension, December 1, 2002; 40(6): 817 - 822. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. K. Whelton, J. He, L. J. Appel, J. A. Cutler, S. Havas, T. A. Kotchen, E. J. Roccella, R. Stout, C. Vallbona, M. C. Winston, et al. Primary Prevention of Hypertension: Clinical and Public Health Advisory From the National High Blood Pressure Education Program JAMA, October 16, 2002; 288(15): 1882 - 1888. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. S. Klassen, E. G. Lowrie, D. N. Reddan, E. R. DeLong, J. A. Coladonato, L. A. Szczech, J. M. Lazarus, and W. F. Owen Jr Association Between Pulse Pressure and Mortality in Patients Undergoing Maintenance Hemodialysis JAMA, March 27, 2002; 287(12): 1548 - 1555. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Boutitie, F. Gueyffier, S. Pocock, R. Fagard, J. P. Boissel, and for the INDANA Project Steering Committee J-Shaped Relationship between Blood Pressure and Mortality in Hypertensive Patients: New Insights from a Meta-Analysis of Individual-Patient Data Ann Intern Med, March 19, 2002; 136(6): 438 - 448. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Bobrie and J. F Potter The elderly hypertensive population: what lies ahead of us? Journal of Renin-Angiotensin-Aldosterone System, March 1, 2002; 3(1_suppl): S4 - S9. [PDF]
|
|
|
|
 |
|
 N. M. Kaplan What Is Goal Blood Pressure for the Treatment of Hypertension? Arch Intern Med, June 25, 2001; 161(12): 1480 - 1482. [Full Text] [PDF]
|
|
|
|
|
|
Lifestyle and risk factor management and use of drug therapies in coronary patients from 15 countries. Principal results from EUROASPIRE II Euro Heart Survey Programme Eur. Heart J., April 1, 2001; 22(7): 554 - 572. [Abstract] [PDF]
|
|
|
|
 |
|
 S. MacMahon, N. Sharpe, G. Gamble, A. Clague, C. Ni Mhurchu, T. Clark, H. Hart, J. Scott, H. White, and PART-2 Collaborative Research Group Randomized, placebo-controlled trial of the angiotensin-converting enzyme inhibitor, ramipril, in patients with coronary or other occlusive arterial disease J. Am. Coll. Cardiol., August 1, 2000; 36(2): 438 - 443. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. MacMahon Blood Pressure and the Risk of Cardiovascular Disease N. Engl. J. Med., January 6, 2000; 342(1): 49 - 52. [Full Text]
|
|
|
|
|
|
T. O. Obisesan, C. M. Vargas, and R. F. Gillum Geographic Variation in Stroke Risk in the United States : Region, Urbanization, and Hypertension in the Third National Health and Nutrition Examination Survey Stroke, January 1, 2000; 31(1): 19 - 25. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. L. Sievers, P. H. Bennett, J. Roumain, and R. G. Nelson Effect of Hypertension on Mortality in Pima Indians Circulation, July 6, 1999; 100(1): 33 - 40. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
The Sixth Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure Arch Intern Med, November 24, 1997; 157(21): 2413 - 2446. [Abstract] [PDF]
|
|
|
|
|
|
A. W. Haider, L. Chen, M. G. Larson, J. C. Evans, M. H. Chen, and D. Levy Antecedent Hypertension Confers Increased Risk for Adverse Outcomes After Initial Myocardial Infarction Hypertension, November 1, 1997; 30(5): 1020 - 1024. [Abstract] [Full Text]
|
|
|
|
|
|
S. MacMahon, A. Rodgers, B. Neal, and J. Chalmers Blood Pressure Lowering for the Secondary Prevention of Myocardial Infarction and Stroke Hypertension, February 1, 1997; 29(2): 537 - 538. [Full Text]
|
|
|
|
|
|
M. Nishizawa, H. Kumagai, M. Ichikawa, N. Oshima, H. Suzuki, and T. Saruta Improvement in Baroreflex Function by an Oral Angiotensin Receptor Antagonist in Rats With Myocardial Infarction Hypertension, January 1, 1997; 29(1): 458 - 463. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. R. Black and J.-Y. Yi A New Classification Scheme for Hypertension Based on Relative and Absolute Risk With Implications for Treatment and Reimbursement Hypertension, November 1, 1996; 28(5): 719 - 724. [Abstract] [Full Text]
|
|
|
|
|
|
C. Lenfant High Blood Pressure: Some Answers, New Questions, Continuing Challenges JAMA, May 22, 1996; 275(20): 1604 - 1606. [Abstract] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||