- Email: [email protected]
Advantages of new cardiovascular risk-assessment strategies in high-risk patients with hypertension
:9 LI; :~:i::i
Commentary Advantages of New Cardiovascular Risk-Assessment Strategies in High-Risk Patients with Hypertension Luis M. Ruilope, MD; and Julian Segura, MD
Hypertension Unit, Hospital 12 de Octubre, Madrid, Spain ABSTRACT Background: Accurate assessment of cardiovascular disease (CVD) risk in patients with hypertension is important when planning appropriate treatment of modifiable risk factors. The causes of CVD are multifactorial, and hypertension seldom exists as an isolated risk factor. Classic models of risk assessment are more accurate than a simple counting of risk factors, but they are not generalizable to all populations. In addition, the risk associated with hypertension is graded, continuous, and independent of other risk factors, and this is not reflected in classic models of risk assessment. Objective: This article is intended to review both classic and newer models of CVD risk assessment. Methods: MEDLINE was searched for articles published between 1990 and 2005 that contained the terms cardiovascular disease, hypertension, or risk assessment. Articles describing major clinical trials, new data about cardiovascular risk, or global risk stratification were selected for review. Results: Some patients at high long-term risk for CVD events (eg, patients aged <50 years with multiple risk factors) may go untreated because they do not meet the absolute risk-intervention threshold of 20% risk over 10 years with the classic model. Recognition of the limitations of classic risk-assessment models led to new guidelines, particularly those of the European Society of Hypertension-European Society of Cardiology. These guidelines view hypertension as one of many risk and disease factors that require treatment to decrease risk. These newer guidelines include a more comprehensive range of risk factors and more finely graded blood pressure ranges to stratify patients by degree of risk. Whether they accurately predict CVD risk in most populations is not known. Evidence from the Valsartan Antihypertensive Long-term Use Evaluation (VALUE) study, which stratified patients
by several risk and disease factors, highlights the predictive value of some newer CVD risk assessments. Conclusion: Modern risk assessments, which include blood pressure along with a wide array of modifiable risk factors, may be more accurate than classic models for CVD risk prediction. (Clin Ther. 2005;27: 1658-1669) Copyright © 2005 Excerpta Medica, Inc. Key words: cardiovascular disease, hypertension, risk assessment.
INTRODUCTION The central role of hypertension in the pathogenesis of cardiovascular disease (CVD) and stroke is well established. 1 During the latter half of the twentieth century, improvements in the treatment and control of hypertension in the United States, Western Europe, Australasia, and Japan were associated with progressive declines in cardiovascular mortality. 2,3 In the past decade, this encouraging trend has been tempered by a leveling off in the benefits of treatment and by evidence of a slight increase in the age-adjusted stroke rate among those with hypertension. In addition, morbidity and mortality from CVD have been rising rapidly in developing countries, such that coronary heart disease and cerebrovascular disease are expected to rank as the number 1 and number 4 causes, respectively, of the global burden of disease by 2020. 3 Thus, detection, treatment, and control of elevated blood pressure remain important public health challenges worldwide. Accurate assessment of CVD risk helps physicians treat their patients appropriately. If a risk-assessment
Acceptedfor publication April 26, 2005. doi:l 0.1016/j.clinthera.2005.10.013 0149-2918/05/$19.00 Printed in the USA.Reproductionin wholeor part is not permitted. Copyright© 2005 ExcerptaMedica,Inc. Volume 27 Number 10
L.M. Ruilope and.l:::Segura algorithm uses a relatively limited range of risk factors, it may underestimate patient risk, which would lead to undertreatment. In contrast, risk-assessment strategies that use a comprehensive list of risk factors would be expected to promote more appropriate care and help reduce cardiovascular morbidity and mortality. To aid physicians in assessing the risk of their patients, major societies and international organizations have developed guidelines for the evaluation and control of hypertension. 3-6 Stratification of patients by CVD risk has evolved with each revision of the various guidelines. Initially, elevated levels of blood pressure, blood cholesterol, and blood glucose were the principal criteria for risk stratification, and each risk factor was assessed individually.3, s However, risk factors tend to cluster in individuals, and the relationship between some CVD risk factors (eg, blood pressure, cholesterol) and the risk of CVD events is graded, continuous, and independent of other risk factors. Therefore, models were developed for determining risk scores to estimate the risk of a coronary or CVD event (including stroke) over a defined period (eg, 5 or 10 years) based on total risk burden. 7-9 These models, based on large European or US data sets (eg, Framingham Heart Study), have been used to develop increasingly complex approaches to the treatment of elevated blood pressure. In the past decade, much attention has focused on whether new risk models are needed to reflect the changes in new guideline recommendations beyond lowering blood pressure.4, 6 This article is intended to review the classic models and to characterize newer models of cardiovascular risk assessment in high-risk patients with hypertension.
METHODS MEDLINE was searched for articles published between 1990 and 2005 that contained the terms cardiovascular disease, hypertension, or risk assessment. Articles describing major clinical trials, new data about cardiovascular risk, or global risk stratification were selected for review. RISK ASSESSMENT: THE CLASSIC MODELS
Several national and international guidelines for hypertension management include an assessment of absolute risk to aid physicians in making treatment decisions. Among the widely used methods of estimating risk are the Sheffield table1°; the Joint British
Societies chart11; the New Zealand chart12; the World Health Organization-International Society of Hypertension (WHO-ISH) table3; the Sixth Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC 6) tableS; and models from the Framingham Heart Study.7,8 All of these risk-estimation methods are more accurate than those based on a simple counting of risk factors; however, the methods vary in their sensitivity and specificity.13 Other risk equations have been based on statistical methods and derived from specific patient populations. Such an equation was derived from patients in the Intervention as a Goal in Hypertension Treatment study, which was conducted in 8 western European countries and Israel. TM This equation was a better predictor of actual cardiovascular risk in the study population than was the Framingham model, which was derived from a largely white, suburban US population. 14 Similarly, shortcomings of the 1999 WHO-ISH risk-stratification scheme for hypertension were demonstrated with data from a population sample in northern Sweden in the Monitoring of Trends and Determinants in Cardiovascular Disease (MONICA) project. 15 This analysis estimated the 10-year risk of fatal and nonfatal myocardial infarction (MI) and stroke. Poor agreement was found between a risk equation based on statistical methods (Cox proportional hazards regression model) plus the Framingham model and the classification of patients into medium-risk and high-risk groups using the WHO-ISH definitions. Most of the risk-assessment tools and recommendations included in major guidelines for hypertension management have relied on Framingham data sets and models. 7,8,16 The Framingham models focus on hypertension, metabolically linked risk factors (eg, dyslipidemia, insulin resistance, glucose intolerance, diabetes mellitus, and obesity), left ventricular hypertrophy, and cigarette smoking as risk factors for CVD; and on elevated systolic blood pressure, diabetes, cigarette smoking, left ventricular hypertrophy, atrial fibrillation, and CVD as risk factors for stroke. 8 Overall, 50% of the coronary events that occur in patients with hypertension arise in the presence of multiple risk factors, and the stroke rate also increases dramatically with increasing numbers of risk factors. 8 Conversely, patients defined as low risk (without elevated blood pressure, dyslipidemia, or cigarette smoking), a minority in major study cohorts (5%-10%), were found to have age-adjusted relative risk levels signifi-
cantly <1 for coronary heart disease mortality, CVD mortality, and all-cause mortality, as well as greater life expectancy.17 Classic models of risk assessment such as the Framingham risk score do, therefore, reflect the continuous, positive correlation between risk factors and CVD events. Hypertension guidelines have used these classic models of risk assessment to develop treatment approaches. The level of risk determined the aggressiveness of treatment to achieve a target blood pressure, long defined as the level of diastolic pressure. Treatment recommendations tended to be rigid and included both lifestyle and pharmacologic measures. Although the ultimate goal of hypertension treatment was to reduce CVD morbidity and mortality, the target blood pressure level was the primary measurable goal of treatment. 3,s R A T I O N A L E FOR A N E W RISK-ASSESSMENT M O D E L
In recent years, the Framingham risk-assessment model has drawn criticism for not being applicable to all populations, in part because it was based on a homogeneous group. Populations differ in lifestyle, diet, genetic predisposition, and underlying rates of CVD, all of which affect the level of risk for cardiovascular events. In addition, the clinical management of men and women free of CVD has changed over the past 30 years--since the Framingham data were initially published--and these changes may affect the generalizability of the Framingham model. TM An analysis by the Framingham investigators found that the sex-specific risk-function models predicted 5-year rates of coronary events (coronary death and MI) in whites and blacks, but systematically overestimated 5-year event rates in other ethnic groups (eg, Japanese American men, Hispanic men, Native American women). 19 Other analyses of the Framingham data support these findings. 2° Overestimation of CVD risk by the Framingham model has also been noted in several European populations. Although some studies of northern and western European populations and the United Kingdom showed that the Framingham model predicted CVD risk successfully for persons with at least modestly elevated risk (sensitivity of 88% and specificity of 90% for a cardiovascular risk of ->20% over 10 years in mild hypertension) but without important nonFramingham risk factors (eg, left ventricular hypertro-
i
:
: ....
)
i
i i :
i
i
i
phy, ethnic status), 1°,21 others reported that the risk was overestimated from 2% (United Kingdom) to 7% (France). 14 Studies of southern European populations found that the Framingham model considerably overestimated absolute risk. 22,23 In an analysis of the MONICA Augsburg (southern Germany) cohort and the Prospective Cardiovascular Mi~nster (northwestern Germany) cohort, the predicted absolute risk of fatal and nonfatal MI based on the Framingham risk model was overestimated by at least 50% compared with the observed risk. 18 Thus, although populationspecific adjustments increase the predictive accuracy of the Framingham model, 19 new risk-estimation guidelines are needed that are more uniformly accurate in most populations. Current risk-estimation models typically predict 5or 10-year coronary heart disease event rates or mortality, and the threshold for intervention is generally set at an absolute risk level of ->200/o. 24 An important limitation of such relatively short-term absolute risk thresholds is that younger patients (ie, aged <50 years), particularly women, are unlikely to reach the intervention threshold despite being at high lifetime risk compared with their peers and despite having ->1 risk factor. 4 In contrast, older patients (ie, men aged >70 years) do reach the intervention thresholds despite having little increased risk compared with their peers. Resources are likely to be channeled to treating older patients with limited potential for increasing the life span, whereas younger patients at relatively high lifetime risk remain untreated, which could significantly shorten their life expectancy.24,2s In younger adults, relative risk imparts long-term, high absolute risk, which the older risk-assessment models fail to consider sufficiently. The association between blood pressure level and risk of a CVD event is graded and continuous. Higher elevations in blood pressure clearly increase risk. However, there is no single blood pressure level that defines the need for pharmacologic intervention, 26 and there is no defined lower limit at which the benefits of blood pressure reduction cease.27,28 The message from the Hypertension Optimal Treatment study, which assessed cardiovascular morbidity and mortality using several target blood pressure levels, was that "lower is better"--that is, the rates of major cardiovascular events and cardiovascular mortality were minimized in the groups with the lowest target blood pressure (diastolic blood pressure -<85 mm Hg). 27,28
,iM. Rui ope No evidence was found of a J-shaped curve for the relation between major cardiovascular events and cardiovascular mortality with lower blood pressures, except in the subgroup of patients who smoked. 29 The graded and continuous CVD risk with hypertension suggests that, at the steepest part of the risk-event curve, patients at the highest risk can realize significantly decreased rates of morbidity and mortality from even small reductions in blood pressure. 3,29,3° A reduction of 5 mm Hg in diastolic blood pressure is associated with at least a 21% lower rate of coronary heart disease and a 34% lower rate of stroke. 3° Moreover, the Framingham data indicate that, after adjusting for
J: S g.ra
other risk factors, patients receiving long-term antihypertensive treatment (20 years) reduce their rates of cardiovascular mortality by up to 60% compared with untreated patients. 31,32 Although the relation between hypertension and risk of CVD events is independent of other CVD risk factors, 6 hypertension is the sole risk factor in <20% of patients. 8 Therefore, risk-factor assessment and treatment in patients with hypertension must include consideration of comorbid conditions and lifestyle factors. Concomitant risk factors, such as diabetes, dyslipidemia, and smoking, markedly increase the risk of coronary heart disease (Figure 1) and stroke.7,8, 33
Men [] Women •
ed
6 "D e~
E
III
Systolic BP, mm Hg Cholesterol, mmol/L (mg/dL) H DL-C, mmol/L (mg/dL) Diabetes Cigarette smoking LVH by ECG
120 5.7 (220) 1.3 (50) . .
160 5.7 (220) 1.3 (50) . .
160 6.7 (259) 0.9 (50) . .
160 6.7 (259) 0.9 (35) . .
160 6.7 (259) 0.9 (35) +
160 6.7 (259) 0.9 (35) + + +
160 6.7 (259) 0.9 (35) + + +
Figure 1. Estimated rate o f coronary heart disease over 10 years according to various combinations o f risk-factor levels for men and women in the Framingham Heart Study. Systolic blood pressure (BP), cholesterol, and high-density lipoprotein cholesterol (HDL-C) are actual values. Current cigarette use was indicated as yes (+) or no (-) on each biennial examination. Criteria used for left ventricular hypertrophy (LVH), measured by electrocardiogram (ECG), are those o f Romhilt-EstesS; rate estimates are based on multivanate statistical modeling. Reprinted with permission. 8
:
::
::
i ::: :
661:
Even when blood pressure is controlled to a target level, other risk factors cause a residual risk of CVD events and mortality. 33,34 Nonetheless, overwhelming evidence demonstrates that the marked increase in risk of CVD events and renal disease in patients with hypertension and type 2 diabetes is attenuated with intensive blood pressure-lowering therapy 35 and that lowering blood pressure has a greater effect on risk reduction than does lowering blood glucose. 36-39 Clustering of metabolic risk factors and cigarette smoking with hypertension is associated with the most significant increases in CVD risk; some other risk factors (eg, physical inactivity), although associated with a less significant short-term risk, may have substantial long-term effects, especially in the presence of multiple risk factors.1 Classic CVD risk-estimation guidelines were useful in identifying high-risk patients with hypertension who required intensive blood pressure-lowering therapy. However, researchers recognized the need to improve these guidelines because they overestimated or underestimated risk in various populations. In 2001, a workshop convened in the United States by the National Heart, Lung, and Blood Institute made several proposals to increase the predictive power of riskestimation guidelines.2° This group recommended pooling databases from large cohort studies to improve the predictive power in white and black middleaged populations and adjusting the then-current Framingham risk models for various ethnic groups with different absolute baseline population risks. Other recommendations included the following: (1) expanding database populations to better characterize the hazards facing high-risk patients (eg, those with diabetes, established CVD, noncoronary forms of atherosclerosis, or left ventricular hypertrophy); (2) extending CVD risk-assessment end points beyond fatal and nonfatal MI; (3) integrating underlying risk factors (eg, diet, obesity, physical inactivity, psychosocial factors, and family history of premature coronary heart disease); and (4) improving methods of clinical assessment of underlying risk factors. The workshop stated that future efforts should determine the independent predictive power of emerging risk factors, as well as measures of abnormal cardiovascular function, myocardial ischemia, and subclinical atherosclerosis. Another important objective is to extend risk prediction to long-term, including lifetime, risk of cardiovascular events and mortality. 2°
FOUNDATION FOR NEW RISK-ASSESSMENT MODELS
Recognition of the limitations of classic risk-assessment models has led to the development of newer models that use an integrated approach to CVD risk. Because the causes of CVD are multifactorial, the latest riskassessment guidelines account for numerous risk factors to advance the concept of total CVD risk. 4,6 The absolute risk determines the absolute benefit of treatment. 4° This shift in thinking means moving away from targeting a threshold blood pressure for intervention; instead, patients with mild elevations in blood pressure may be considered for treatment if they have a high global CVD risk. The Systematic Coronary Risk Evaluation (SCORE) project in Europe was among the first to develop paper-based, absolute CVD risk-assessment charts for various high-risk (Figure 2) and low-risk European populations, with an emphasis on fatal CVD events. 9 The most recent guidelines of the European Society of Hypertension (ESH) and the European Society of Cardiology (ESC) assess risk in patients with hypertension as low, moderate, high, and very high added risk (Table I). 4 These categories correlate with an approximate absolute 10-year CVD risk of <15%, 15% to 20%, >20% to 30%, and >30%, respectively, by the Framingham risk model, 7 or with an approximate absolute risk of fatal CVD of <4%, 4% to 5%, >5% to 8%, and >8%, respectively, by the SCORE charts. 9 The designation of added risk in the ESH-ESC guidelines considers not only level of blood pressure, but also CVD risk factors, target organ damage, diabetes, and associated clinical conditions (Table II). 4 The ESH-ESC guidelines include recently identified risk factors (eg, C-reactive protein) and measures of renal function as well as CVD risk factors with established adverse effects on morbidity and mortality. The ESHESC risk assessment is more comprehensive than that presented in the Seventh Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC 7) because the treatment algorithm in JNC 7 is based principally on blood pressure level plus consideration of established diabetes or chronic renal disease. 6 The treatment recommendations in JNC 7 are stratified on the basis of whether hypertension is present, with or without compelling indications (eg, heart failure post-MI, high coronary disease risk, diabetes, chronic renal disease, recurrent stroke prevention). Volume 27 Number 10
[] >_15% 10%-14% [] 5%-9%
[]
[] 3%-4%
[] 2% [] 1% [ ] <1%
Men
Women Nonsmoker
Nonsmoker
Smoker
Smoker
180 160
Age 65
~!~ ,.~ i:~ ~
140 120
180t 160{-
Age 60
I
140t_i_ E E
~:
~'~
140 I- 1/ 21
1~
180 t
16o[
o
>..
N@
~
/
120p,(t
-1"/
,L J l "
180 160
C, .1~ '
140
0
1
Age
-1 x
2:
i'- tfi, )~l\-
'-1-, I"
-1
ss
2t212
2
1~ /
21
it, :1,
fl
/
t
--
2
t-9/
2
2 :~,
1" " 1 /
2 ii@@~ii
F q:l4 q 2
.
-1
-1
1(
8ot! o o
0
0
0
0
0
0 ,,1/
~/
0
AgeS0
2
0
120
I!/// I/
-
/
I
...............
160F[ 0 /
0
0
0
0
0
0
0
0
0
140 k
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
I
I
I
I
I
I
I
I
I
I
f
I
I
I
I
I
I
I
I
l
4
S
6
7
8
4
5
6
7
8
4
5
6
7
8
4
S
6
7
8
/
120~mmol
I
I
I
Age 40
1" ,i" .1".* / 1". 2V , .1 2 2 0 0 -> 1- ,1" ~f. 5 ' ' 1 "t- 4
0
I
t
I
I
I
mg/dL 150 200 250 300
mg/dL 150 200 250 300 Cholesterol (mmol/L)
Figure 2. Estimated risk of cardiovascular mortality over 10 years in high-risk populations stratified by total cholesterol, systolic blood pressure, age, sex, and smoking status. Reprinted with permission. 9
The newest guidelines, particularly the ESH-ESC guidelines, are more liberal than earlier guidelines with respect to the definition and classification of blood pressure levels and the trigger level for intervention& 6 In addition, blood pressure categories have been added that do not denote hypertension (eg, normal and high normal, prehypertension), with recommendations for treatment based on other associated risk factors or con-
ditions. These changes recognize the multifactorial causes of CVD, clustering of risk factors, and different absolute baseline population risks, and suggest that the appropriate level of blood pressure for intervention is patient specific. The newest guidelines, however, still use discrete intervals of blood pressure to define risk rather than truly depict the graded, continuous risk associated with blood pressure.
: ~il :ii'/: ii!ii~6~!i
Table I. Stratification of risk to quantify prognosis based on European Society of Hypertension-European Society of Cardiology 2003 Guidelines.4 Added Risk by Blood Pressure, mm Hg
Other Risk Factors and Disease History
Normal: SBP 120-129 or DBP 80-84
High Normal: SBP 130-139 or DBP 85-89
No other risk factors
-
-
Grade 1 HTN: SBP 140-159 orDBP90-99
Grade 2 HTN: SBP 160-179 or DBP 100-109
Grade 3 HTN: SBP->180 or DBP->110
Low
Moderate
High
1-2 Risk factors
Low
Low
Moderate
Moderate
Very high
->3 Risk factors, TOD, or diabetes
Moderate
High
High
High
Very high
Associated clinical conditions
High
Very high
Very high
Very high
Very high
SBP = systolic blood pressure; DBP = diastolic blood pressure; HTN = hypertension; TOD = target organ damage.
Table II. Factors influencing prognosis based on European Society of Hypertension-European Society of Cardiology 2003 guidelines.4 Category
Risk Factor
CVD
Levels of SBP and DBP Men aged >55 years Women aged >65 years Smoking Dyslipidemia Family history of premature CVD Abdominal obesity C-reactive protein ->1 mg/dL
Diabetes mellitus
Fasting plasma glucose ->7.0 mmol/L (126 mg/dL) Postprandial plasma glucose >11.0 mmol/L (198 mg/dL)
Target organ damage
Left ventricular hypertrophy Arterial wall thickening on ultrasound Slight increase in serum creatinine Microalbuminuria
Associated clinical conditions
Cerebrovascular disease Heart disease Renal disease Peripheral vascular disease Advanced retinopathy
CVD = cardiovascular disease;SBP = systolic blood pressure; DBP = diastolic blood pressure.
1166411
'
=
i[:
~
Volume 27 Number 10
L.M. Ruilope and j. Segura THE VALUE O F THE VALUE STUDY
The Valsartan Antihypertensive Long-term Use Evaluation (VALUE) study was a large, randomized, double-blind trial comparing the long-term effects of the angiotensin receptor blocker (ARB) valsartan with those of a calcium antagonist (amlodipine) in patients with hypertension and a high risk of CVD who were monitored for a mean of 4.2 years. 41 Inclusion criteria for the VALUE cohort paralleled many of the suggestions for measuring total CVD risk in new risk-estimation guidelines. Indeed, the results of the VALUE study might be used to verify the utility of the new guidelines. In addition, the study has provided a large new database of patients who were stratified by many of the same risk and disease factors as used in the ESH-ESC guidelines (Table II) as well as information on several biochemical parameters. 4,2°,41 At baseline, mean blood pressures were only moderately elevated because 92.7% (7088/7649) of valsartan patients and 92.0% (6989/7596) of the amlodipine group were receiving antihypertensive drugs at randomization. Of note, there was no washout period, and it is reasonable to suspect that the persistent natriuretic and diuretic properties of calcium-channel blockers may have augmented the antihypertensive efficacy of amlodipine in the first few months of the study. 42 Nearly half of the patients had coronary heart disease (45.8% [7013/15,313]), approximately one third had diabetes (31.7% [4854/15,313]) or elevated cholesterol levels (33.0% [5053/15,313]), and nearly one quarter were current smokers (24.0% [3675/15,313]) or had proteinuria (22.5% [3445/15,313]). 43 As defined in both the ESH-ESC guidelines 4 and the US National Cholesterol Education Program's Third Adult Treatment Panel (ATP III) report, 44 patients in VALUE had a markedly increased risk for a CVD event or mortality, a risk level closer to that in the Heart Outcomes Prevention Evaluation than in previous hypertension trials. 4s Despite unequal blood pressure lowering between treatment arms in VALUE that favored amlodipine (particularly in the early phase of the trial), no difference was found in the primary composite outcome of cardiac morbidity and mortality. 41 For secondary end points, MI was significantly lower in the amlodipine group (hazard ratio [HR], 1.19; 95% CI, 1.02-1.38; P = 0.02), stroke was similar between groups (HR, 1.15; 95% CI, 0.98-1.35; P = NS), and the rate of hospitalization for heart failure suggested a slight benefit from
valsartan (HR, 0.89; 95% CI, 0.77-1.03; P = NS). A prespecified analysis of new-onset diabetes mellitus-a risk factor in both older and newer guidelines-demonstrated a significant 23% risk reduction with valsartan compared with amlodipine (P < 0 . 0 0 1 ) . 41 Interestingly, after correcting for between-group differences in blood pressure by serial median matching of patients, rates of MI and stroke were no different between treatment arms, whereas heart failure was significantly lower in the valsartan arm. 46 The striking success of this trial was its ability to control hypertension in this high-risk population to <140/80 mm Hg. VALUE also highlighted the importance of prompt blood pressure control and the necessity for combination therapy in many high-risk patients. 41 The results of VALUE might appear to conflict with those of the Losartan Intervention for Endpoint Reduction in Hypertension (LIFE) trial. 47 Although the average systolic blood pressure in LIFE was 7 mm Hg higher than in VALUE, the incidence of cardiac events and death was lower in LIFE. This apparent paradox can be resolved by taking into account the global risk of each patient population. LIFE, which compared the ARB losartan with the 13-blocker atenolol, enrolled patients who had lower cardiovascular risk because they were at an earlier stage of CVD than patients in VALUE. In fact, the higher blood pressure but lower incidence of cardiovascular events and death in LIFE as compared with VALUE lends support to the premise that cardiovascular risk is driven not only by elevated blood pressure, but also by the extent of vascular damage, primarily atherosclerosis. This observation is another piece of evidence supporting the newer, more sophisticated risk-assessment models. The typical overall 10-year CVD risk of the VALUE cohort can be estimated using any of the 3 recent methods (ESH-ESC, SCORE, and ATP III). If the ESH-ESC formula is used, the systolic blood pressure (mean [SD], 154.7 [19.0] mm Hg) classifies them as having grade 1 hypertension, and most participants appear to have >_2 additional risk factors or target organ damage or diabetes, thereby placing them at moderate to very high added risk (see Tables I and II for criteria). 4,43 Thus, a rough estimate of the anticipated event rate over 10 years would be 25%, and the mortality rate might be 6% to 7% over the same period. If the highrisk SCORE chart is used, which would be the best fit for a predominantly US and European population, the mean values for systolic blood pressure and total cho-
166S
lesterol plus sex, age, and smoking status suggest an average cardiovascular mortality of 11% over 10 years. (The SCORE risk charts do not extend above the age of 65 years, so the women in VALUE, who averaged - 7 0 years, will have an actual risk that is somewhat higher.) The ATP III guidelines, based on Framingham projections, imply that the patients randomized in VALUE have a <20% risk per 10 years for a major coronary event. (The baseline data in VALUE do not include levels of high-density lipoprotein cholesterol, a component of the Framingham point score. 44) The results for the composite end point of cardiovascular morbidity and mortality in VALUE demonstrated a nearly linear progression over time, with a rate of - 1 2 % over 5 years in both treatment arms, which would suggest a 24% rate for first events at 10 years. 4~ This assumption may be somewhat conservative; the Multiple Risk Factor Intervention Trial demonstrated increasing risk across a span of 10.5 years, particularly in the usual-care group. 48 On the other hand, it is reasonable to speculate that long-term treatment would show an increasing benefit over time. Mechanisms that might become more apparent over time include slowed progression of CVD risk factors, such as the reduction in new-onset diabetes and its consequences in the valsartan arm of VALUE. Further, if we assume that the CVD death rate is linear, then the 4% rate at 4.2 years would imply a 10% rate at 10 years. 41 Thus, the finer gradations of SCORE and the ESH-ESC projections seem to capture the CVD risk in the VALUE cohort better than the Framingham-based estimates.
CONCLUSIONS Accurate assessment of risk is essential for determining appropriate therapy for patients and maximizing the benefit of risk-reduction strategies. Risk-stratification models for patients with hypertension have evolved to become increasingly complex and nuanced. Classic risk assessments, generally based on data from the Framingham Heart Study, focus on blood pressure and a limited number of other risk factors and end points. The classic models lack predictive value in many patient populations, however, so that new models are needed. Further, classic assessment is premised on 5to 10-year risk, which is a barrier to treatment for younger patients who may have a low short-term risk but a high risk over longer time periods. Modern risk assessments include blood pressure as an important element in the global CVD morbidity~=~1
mortality equation. Because blood pressure is a continuum and hypertension rarely exists in isolation, treatment should be directed toward both reaching blood pressure goals and preventing or treating a wide array of modifiable risk factors that affect target organ damage and mortality. Patients at highest risk tend to have the most to gain by even modest improvements in their risk profile. New guidelines include many recently discovered risk factors, including several important measurements of renal dysfunction, which improve the recognition and treatment of patients with CVD. The VALUE trial highlights the increased accuracy of these new guidelines and provides a large new database of patients with high-risk hypertension.
REFERENCES 1. Ishizaka N, Ishizaka Y, Toda E, et al. Hypertension is the most common component of metabolic syndrome and the greatest contributor to carotid arteriosclerosis in apparently healthy Japanese individuals. Hypertens Res. 2005;28:27-34. 2. 2005 Heart and Stroke Statistical Update. Dallas, Tex: American Heart Association; 2004. 3. Guidelines Subcommittee. 1999 World Health Organization-International Society of Hypertension Guidelines for the Management of Hypertension. J H_ypertens. 1999;17:151-183. 4. European Society of Hypertension-European Society of Cardiology Guidelines Committee. 2003 European Society of Hypertension-European Society of Cardiology Guidelines for the Management of Arterial Hypertension [published correction appears in J Hjpertens. 2003;21: 2203-2204].J H.ypertens. 2003;21:1011-1053. S. The Sixth Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure. Arch Intern Med. 1997;157:2413-2446. 6. Chobanian AV, Bakris GL, Black HR, et al, for the National High Blood Pressure Education Program Coordinating Committee. The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure: TheJNC 7 report [published correction appears inJAMA. 2003;290: 197].JAMA. 2003;289:2560-2572. 7. Anderson KM, Wilson PW, Odell PM, Kannel WB. An updated coronary risk profile: A statement for health professionals. Circulation. 1991 ;83:356-362. 8. Kannel WB. Blood pressure as a cardiovascular risk factor: Prevention and treatment.JAMA. 1996;275:1571-1576. 9. Conroy RM, Py6r~il/~K, Fitzgerald AP, et al, for the SCORE Project Group. Estimation often-year risk of fatal cardiovascular disease in Europe: The SCORE Project. Eur Heart_/. 2003;24:987-1003. Volume 27 Number 10
LM. Ruilope :andJ~:segura
: 10. Wallis EJ, Ramsay LE, UI Haq I, et al. Coronary and cardiovascular risk estimation For primary prevention: Validation of a new Sheffield table in the 1995 Scottish health survey population. BMJ. 2000;320:671-676. 11. Wood D, Durrington P, Poulter N, et al. Joint British recommendations on prevention of coronary heart disease in clinical practice. Heart. 1998;80(Suppl 2):$1-$29. 12. New Zealand Guidelines Group. Management of Mildly Raised Blood Pressure in New Zealand. Wellington, Core Services Committee, New Zealand Ministry of Health, 1995. Available at: http://www.nzgg. org.nz/index.cfm ?fuseaction=about Nfuseubaction=docs&documentlD = 81. Accessed June 25, 2005. 13. Yikona Jl, Wallis EJ, Ramsay LE, Jackson PR. Coronary and cardiovascular risk estimation in uncomplicated mild hypertension. A comparison of risk assessment methods. J H vpertens. 2002;20:2173-2182. 14. Bastuji-Garin S, DeverlyA, Moyse D, et al, for the Intervention as a Goal in Hypertension Treatment Study Group. The Framingham prediction rule is not valid in a European population of treated hypertensive patients.J H.ype~ens.2002;20:1973-1980. 15. Persson M, Carlberg B, Weinehall L, et al. Risk stratification by guidelines compared with risk assessment by risk equations applied to a MONICA sample. J Hypertens. 2003;21:10891095. 16. Wilson PW, D'Agostino RB, Levy D, et al. Prediction of coronary heart disease using risk factor categories. Circulation. 1998;97:1837-1847. 17. StamlerJ, Stamler R, Neaton JD, et al. Low risk-factor profile and longterm cardiovascular and noncardiovascular mortality and life expectancy: Findings for 5 large cohorts of young adult and middle-aged men and wornen. JAMA. 1999;282: 2012-2018. 18. Hense HW, Schulte H, L6wel H, et al. Framingham risk function overes-
19.
20.
21.
22.
23.
24.
25.
26.
27.
timates risk of coronary heart disease in men and women from Germany-results from the MONICA Augsburg and the PROCAM cohorts. Eur HeartJ. 2003;24:937-945. D'Agostino RB Sr, Grundy S, Sullivan LM, Wilson P, ffor the CHD Risk Prediction Group. Validation off the Framingham coronary heart disease prediction scores: Results of a multiple ethnic groups investigation.JAMA. 2001 ;286:180-187. Grundy SM, D'Agostino RB Sr, Mosca L, et al. Cardiovascular risk assessment based on US cohort studies: Findings from a National Heart, Lung, and Blood Institute workshop. Circulation. 2001 ;104:491496. Ramachandran S, French JM, Vanderpump MP, et al. Using the Framingham model to predict heart disease in the United Kingdom: Retrospective study. BMJ. 2000;320: 676-677. Menotti A, Lanti M, Puddu PE, Kromhout D. Coronary heart disease incidence in northern and southern European populations: A reanalysis of the seven countries study for a European coronary risk chart. Heart. 2000;84:238-244. Menotti A, Puddu PE, Lanti M. Comparison of the Framingham risk Function-based coronary chart with risk function from an Italian population study. EurHeartJ. 2000;21:365370. Simpson FO. Guidelines for antihypertensive therapy: Problems with a strategy based on absolute cardiovascular risk. J Hypertens. 1996;14: 683-689. Zanchetti A. Antihypertensive therapy: How to evaluate the benefits. AmJ Cardiol. 1997;79:3-8. Alderman MH, Furberg CD, Kostis JB, et al. Hypertension guidelines: Criteria that might make them more clinically useful. Am J Hypertens. 2002;15:917-923. Hansson L. The Hypertension Optimal Treatment study and the impor-
tance of lowering blood pressure.
] H#pertensSuppl. 1999;17:$9-$13. 28. Hansson L, Zanchetti A, Carruthers SG, et al, for the HOT Study Group. Effects of intensive blood-pressure lowering and low-dose aspirin in patients with hypertension: Principal results of the Hypertension Optimal Treatment (HOT) randomised trial. Lancet. 1998;351 :I 755-I 762.
29. Zanchetti A, Hansson L, Clement D, et al, for the HOT study group. Benefits and risks of more intensive blood pressure lowering in hypertensive patients of the HOT study with different risk profiles: Does a J-shaped curve exist in smokers? J Hypertens. 2003;21:797-804. 30. MacMahon S, Peto R, Cutler J, et al. Blood pressure, stroke, and coronary heart disease. Part 1, prolonged differences in blood pressure: Prospective observational studies corrected For the regression dilution bias. Lancet. 1990;335:765-774. 31. STckowskiPA, 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 to 1990. Circulation. 1996;93:697-703. 32. Franklin SS, Wong ND. Cardiovascular risk evaluation: An inexact science. J Hypertens. 2002;20:21272130. 33. Zanchetti A, Hansson L, Dahl6f B, et al, for the HOT Study Group. Effects of individual risk Factors on the incidence of cardiovascular events in the treated hypertensive patients of the Hypertension Optimal Treatment study. J H.ypertens. 2001;19: 1149-1159. 34. Zanchetti A, Hansson L, M4nard J, et ak Risk assessment and treatment benefit in intensively treated hypertensive patients of the Hypertension Optimal Treatment (HOT) study. J Hypertens. 2001 ;19:819-825. 35. Zanchetti A, Ruilope LM. Antihypertensive treatment in patients with type-2 diabetes mellitus: What guid-
.
.
.
.
.
.
•
::. . . . . . .
::: I D D !
36.
37.
38.
39.
ance from recent controlled randomized trials? -/ Hypertens. 2002;20: 2099-2110. UK Prospective Diabetes Study Group. Tight blood pressure control and risk of macrovascular and microvascular complications in type 2 diabetes: UKPDS 38 [published correction appears in BM-/. 1999;318: 29]. BM-/. 1998;317:703-713. UK Prospective Diabetes Study (UKPDS) Group. Intensive bloodglucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). Lancet. 1998;352:837-853. Mogensen CE. Combined high blood pressure and glucose in type 2 diabetes: Double jeopardy. British trial shows clear effects of treatment, especially blood pressure reduction. BMJ. 1998;317:693-694. Julius S, Majahalme S, Palatini P. Antihypertensive treatment of patients
Summary of the Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III).-/AMA. 2001 ;285:2486-2497. 45. YusufS, Sleight P, Pogue J, et al, for the Heart Outcomes Prevention Evaluation Study Investigators. Effects of an angiotensin-converting-enzyme inhibitor, ramipril, on cardiovascular events in high-risk patients [published corrections appear in N Engl-/ Med. 2000;342:748 and N Engl-/Med. 2000;342:1376]. N EnglJ Med. 2000; 342:145-153.
46. Weber MA, Julius S, Kjeldsen SE, et al. Blood pressure dependent and independent effects ofantihypertensive treatment on clinica[ events in
the VALUE Trial. Lancet. 2004;363: 2049-2051. 47. Kjeldsen SE, DahlofB, Devereux RB, et al. Lowering of blood pressure and predictors of response in patients with left ventricular hypertrophy: The LIFE study. Losartan Intervention For Endpoint. Am/Hypertens. 2000;13:899-906. 48. Mortality after 10 1/2 years For hypertensive participants in the Multiple Risk Factor Intervention Trial. Circulation. 1990;82:1616-1628.
with diabetes and hypertension. Am-/ Hypertens. 2001 ;14:310S-316S. 40. Ramsay LE, Wallis EJ, Yeo WW, Jackson PR. The rationale for differing national recommendations for the treatment of hypertension. Am-/ H.ypertens. 1998;11:79S-88S. 41. Julius S, Kjeldsen SE, Weber M, et al, for the VALUE trial group. Outcomes in hypertensive patients at high cardiovascular risk treated with regimens based on valsartan or amIodipine: The VALUE randomised trial. Lancet. 2004;363:2022-2031. 42. RomeroJC, Raij L, GrangerJP, et al. Multiple effects of calcium entry blockers on renal function in hypertension. Hypertension. 1987;10:140-151. 43. Kjeldsen SE, Julius S, Brunner H, et al. Characteristics of 15,314 hypertensive patients at high coronary risk: The VALUE trial. The Valsartan Antihypertensive Long-term Use Evaluation. Blood Press. 2001 ;10:8391. 44. Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults. Executive
1668
Address correspondence to: Luis M. Ruilope, MD, Hypertension Unit, Hospital 12 de Octubre, 28041 Madrid, Spain. E-maiL [email protected] Volume 27 Number 10
Commentary Advantages of New Cardiovascular Risk-Assessment Strategies in High-Risk Patients with Hypertension Luis M. Ruilope, MD; and Julian Segura, MD
Hypertension Unit, Hospital 12 de Octubre, Madrid, Spain ABSTRACT Background: Accurate assessment of cardiovascular disease (CVD) risk in patients with hypertension is important when planning appropriate treatment of modifiable risk factors. The causes of CVD are multifactorial, and hypertension seldom exists as an isolated risk factor. Classic models of risk assessment are more accurate than a simple counting of risk factors, but they are not generalizable to all populations. In addition, the risk associated with hypertension is graded, continuous, and independent of other risk factors, and this is not reflected in classic models of risk assessment. Objective: This article is intended to review both classic and newer models of CVD risk assessment. Methods: MEDLINE was searched for articles published between 1990 and 2005 that contained the terms cardiovascular disease, hypertension, or risk assessment. Articles describing major clinical trials, new data about cardiovascular risk, or global risk stratification were selected for review. Results: Some patients at high long-term risk for CVD events (eg, patients aged <50 years with multiple risk factors) may go untreated because they do not meet the absolute risk-intervention threshold of 20% risk over 10 years with the classic model. Recognition of the limitations of classic risk-assessment models led to new guidelines, particularly those of the European Society of Hypertension-European Society of Cardiology. These guidelines view hypertension as one of many risk and disease factors that require treatment to decrease risk. These newer guidelines include a more comprehensive range of risk factors and more finely graded blood pressure ranges to stratify patients by degree of risk. Whether they accurately predict CVD risk in most populations is not known. Evidence from the Valsartan Antihypertensive Long-term Use Evaluation (VALUE) study, which stratified patients
by several risk and disease factors, highlights the predictive value of some newer CVD risk assessments. Conclusion: Modern risk assessments, which include blood pressure along with a wide array of modifiable risk factors, may be more accurate than classic models for CVD risk prediction. (Clin Ther. 2005;27: 1658-1669) Copyright © 2005 Excerpta Medica, Inc. Key words: cardiovascular disease, hypertension, risk assessment.
INTRODUCTION The central role of hypertension in the pathogenesis of cardiovascular disease (CVD) and stroke is well established. 1 During the latter half of the twentieth century, improvements in the treatment and control of hypertension in the United States, Western Europe, Australasia, and Japan were associated with progressive declines in cardiovascular mortality. 2,3 In the past decade, this encouraging trend has been tempered by a leveling off in the benefits of treatment and by evidence of a slight increase in the age-adjusted stroke rate among those with hypertension. In addition, morbidity and mortality from CVD have been rising rapidly in developing countries, such that coronary heart disease and cerebrovascular disease are expected to rank as the number 1 and number 4 causes, respectively, of the global burden of disease by 2020. 3 Thus, detection, treatment, and control of elevated blood pressure remain important public health challenges worldwide. Accurate assessment of CVD risk helps physicians treat their patients appropriately. If a risk-assessment
Acceptedfor publication April 26, 2005. doi:l 0.1016/j.clinthera.2005.10.013 0149-2918/05/$19.00 Printed in the USA.Reproductionin wholeor part is not permitted. Copyright© 2005 ExcerptaMedica,Inc. Volume 27 Number 10
L.M. Ruilope and.l:::Segura algorithm uses a relatively limited range of risk factors, it may underestimate patient risk, which would lead to undertreatment. In contrast, risk-assessment strategies that use a comprehensive list of risk factors would be expected to promote more appropriate care and help reduce cardiovascular morbidity and mortality. To aid physicians in assessing the risk of their patients, major societies and international organizations have developed guidelines for the evaluation and control of hypertension. 3-6 Stratification of patients by CVD risk has evolved with each revision of the various guidelines. Initially, elevated levels of blood pressure, blood cholesterol, and blood glucose were the principal criteria for risk stratification, and each risk factor was assessed individually.3, s However, risk factors tend to cluster in individuals, and the relationship between some CVD risk factors (eg, blood pressure, cholesterol) and the risk of CVD events is graded, continuous, and independent of other risk factors. Therefore, models were developed for determining risk scores to estimate the risk of a coronary or CVD event (including stroke) over a defined period (eg, 5 or 10 years) based on total risk burden. 7-9 These models, based on large European or US data sets (eg, Framingham Heart Study), have been used to develop increasingly complex approaches to the treatment of elevated blood pressure. In the past decade, much attention has focused on whether new risk models are needed to reflect the changes in new guideline recommendations beyond lowering blood pressure.4, 6 This article is intended to review the classic models and to characterize newer models of cardiovascular risk assessment in high-risk patients with hypertension.
METHODS MEDLINE was searched for articles published between 1990 and 2005 that contained the terms cardiovascular disease, hypertension, or risk assessment. Articles describing major clinical trials, new data about cardiovascular risk, or global risk stratification were selected for review. RISK ASSESSMENT: THE CLASSIC MODELS
Several national and international guidelines for hypertension management include an assessment of absolute risk to aid physicians in making treatment decisions. Among the widely used methods of estimating risk are the Sheffield table1°; the Joint British
Societies chart11; the New Zealand chart12; the World Health Organization-International Society of Hypertension (WHO-ISH) table3; the Sixth Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC 6) tableS; and models from the Framingham Heart Study.7,8 All of these risk-estimation methods are more accurate than those based on a simple counting of risk factors; however, the methods vary in their sensitivity and specificity.13 Other risk equations have been based on statistical methods and derived from specific patient populations. Such an equation was derived from patients in the Intervention as a Goal in Hypertension Treatment study, which was conducted in 8 western European countries and Israel. TM This equation was a better predictor of actual cardiovascular risk in the study population than was the Framingham model, which was derived from a largely white, suburban US population. 14 Similarly, shortcomings of the 1999 WHO-ISH risk-stratification scheme for hypertension were demonstrated with data from a population sample in northern Sweden in the Monitoring of Trends and Determinants in Cardiovascular Disease (MONICA) project. 15 This analysis estimated the 10-year risk of fatal and nonfatal myocardial infarction (MI) and stroke. Poor agreement was found between a risk equation based on statistical methods (Cox proportional hazards regression model) plus the Framingham model and the classification of patients into medium-risk and high-risk groups using the WHO-ISH definitions. Most of the risk-assessment tools and recommendations included in major guidelines for hypertension management have relied on Framingham data sets and models. 7,8,16 The Framingham models focus on hypertension, metabolically linked risk factors (eg, dyslipidemia, insulin resistance, glucose intolerance, diabetes mellitus, and obesity), left ventricular hypertrophy, and cigarette smoking as risk factors for CVD; and on elevated systolic blood pressure, diabetes, cigarette smoking, left ventricular hypertrophy, atrial fibrillation, and CVD as risk factors for stroke. 8 Overall, 50% of the coronary events that occur in patients with hypertension arise in the presence of multiple risk factors, and the stroke rate also increases dramatically with increasing numbers of risk factors. 8 Conversely, patients defined as low risk (without elevated blood pressure, dyslipidemia, or cigarette smoking), a minority in major study cohorts (5%-10%), were found to have age-adjusted relative risk levels signifi-
cantly <1 for coronary heart disease mortality, CVD mortality, and all-cause mortality, as well as greater life expectancy.17 Classic models of risk assessment such as the Framingham risk score do, therefore, reflect the continuous, positive correlation between risk factors and CVD events. Hypertension guidelines have used these classic models of risk assessment to develop treatment approaches. The level of risk determined the aggressiveness of treatment to achieve a target blood pressure, long defined as the level of diastolic pressure. Treatment recommendations tended to be rigid and included both lifestyle and pharmacologic measures. Although the ultimate goal of hypertension treatment was to reduce CVD morbidity and mortality, the target blood pressure level was the primary measurable goal of treatment. 3,s R A T I O N A L E FOR A N E W RISK-ASSESSMENT M O D E L
In recent years, the Framingham risk-assessment model has drawn criticism for not being applicable to all populations, in part because it was based on a homogeneous group. Populations differ in lifestyle, diet, genetic predisposition, and underlying rates of CVD, all of which affect the level of risk for cardiovascular events. In addition, the clinical management of men and women free of CVD has changed over the past 30 years--since the Framingham data were initially published--and these changes may affect the generalizability of the Framingham model. TM An analysis by the Framingham investigators found that the sex-specific risk-function models predicted 5-year rates of coronary events (coronary death and MI) in whites and blacks, but systematically overestimated 5-year event rates in other ethnic groups (eg, Japanese American men, Hispanic men, Native American women). 19 Other analyses of the Framingham data support these findings. 2° Overestimation of CVD risk by the Framingham model has also been noted in several European populations. Although some studies of northern and western European populations and the United Kingdom showed that the Framingham model predicted CVD risk successfully for persons with at least modestly elevated risk (sensitivity of 88% and specificity of 90% for a cardiovascular risk of ->20% over 10 years in mild hypertension) but without important nonFramingham risk factors (eg, left ventricular hypertro-
i
:
: ....
)
i
i i :
i
i
i
phy, ethnic status), 1°,21 others reported that the risk was overestimated from 2% (United Kingdom) to 7% (France). 14 Studies of southern European populations found that the Framingham model considerably overestimated absolute risk. 22,23 In an analysis of the MONICA Augsburg (southern Germany) cohort and the Prospective Cardiovascular Mi~nster (northwestern Germany) cohort, the predicted absolute risk of fatal and nonfatal MI based on the Framingham risk model was overestimated by at least 50% compared with the observed risk. 18 Thus, although populationspecific adjustments increase the predictive accuracy of the Framingham model, 19 new risk-estimation guidelines are needed that are more uniformly accurate in most populations. Current risk-estimation models typically predict 5or 10-year coronary heart disease event rates or mortality, and the threshold for intervention is generally set at an absolute risk level of ->200/o. 24 An important limitation of such relatively short-term absolute risk thresholds is that younger patients (ie, aged <50 years), particularly women, are unlikely to reach the intervention threshold despite being at high lifetime risk compared with their peers and despite having ->1 risk factor. 4 In contrast, older patients (ie, men aged >70 years) do reach the intervention thresholds despite having little increased risk compared with their peers. Resources are likely to be channeled to treating older patients with limited potential for increasing the life span, whereas younger patients at relatively high lifetime risk remain untreated, which could significantly shorten their life expectancy.24,2s In younger adults, relative risk imparts long-term, high absolute risk, which the older risk-assessment models fail to consider sufficiently. The association between blood pressure level and risk of a CVD event is graded and continuous. Higher elevations in blood pressure clearly increase risk. However, there is no single blood pressure level that defines the need for pharmacologic intervention, 26 and there is no defined lower limit at which the benefits of blood pressure reduction cease.27,28 The message from the Hypertension Optimal Treatment study, which assessed cardiovascular morbidity and mortality using several target blood pressure levels, was that "lower is better"--that is, the rates of major cardiovascular events and cardiovascular mortality were minimized in the groups with the lowest target blood pressure (diastolic blood pressure -<85 mm Hg). 27,28
,iM. Rui ope No evidence was found of a J-shaped curve for the relation between major cardiovascular events and cardiovascular mortality with lower blood pressures, except in the subgroup of patients who smoked. 29 The graded and continuous CVD risk with hypertension suggests that, at the steepest part of the risk-event curve, patients at the highest risk can realize significantly decreased rates of morbidity and mortality from even small reductions in blood pressure. 3,29,3° A reduction of 5 mm Hg in diastolic blood pressure is associated with at least a 21% lower rate of coronary heart disease and a 34% lower rate of stroke. 3° Moreover, the Framingham data indicate that, after adjusting for
J: S g.ra
other risk factors, patients receiving long-term antihypertensive treatment (20 years) reduce their rates of cardiovascular mortality by up to 60% compared with untreated patients. 31,32 Although the relation between hypertension and risk of CVD events is independent of other CVD risk factors, 6 hypertension is the sole risk factor in <20% of patients. 8 Therefore, risk-factor assessment and treatment in patients with hypertension must include consideration of comorbid conditions and lifestyle factors. Concomitant risk factors, such as diabetes, dyslipidemia, and smoking, markedly increase the risk of coronary heart disease (Figure 1) and stroke.7,8, 33
Men [] Women •
ed
6 "D e~
E
III
Systolic BP, mm Hg Cholesterol, mmol/L (mg/dL) H DL-C, mmol/L (mg/dL) Diabetes Cigarette smoking LVH by ECG
120 5.7 (220) 1.3 (50) . .
160 5.7 (220) 1.3 (50) . .
160 6.7 (259) 0.9 (50) . .
160 6.7 (259) 0.9 (35) . .
160 6.7 (259) 0.9 (35) +
160 6.7 (259) 0.9 (35) + + +
160 6.7 (259) 0.9 (35) + + +
Figure 1. Estimated rate o f coronary heart disease over 10 years according to various combinations o f risk-factor levels for men and women in the Framingham Heart Study. Systolic blood pressure (BP), cholesterol, and high-density lipoprotein cholesterol (HDL-C) are actual values. Current cigarette use was indicated as yes (+) or no (-) on each biennial examination. Criteria used for left ventricular hypertrophy (LVH), measured by electrocardiogram (ECG), are those o f Romhilt-EstesS; rate estimates are based on multivanate statistical modeling. Reprinted with permission. 8
:
::
::
i ::: :
661:
Even when blood pressure is controlled to a target level, other risk factors cause a residual risk of CVD events and mortality. 33,34 Nonetheless, overwhelming evidence demonstrates that the marked increase in risk of CVD events and renal disease in patients with hypertension and type 2 diabetes is attenuated with intensive blood pressure-lowering therapy 35 and that lowering blood pressure has a greater effect on risk reduction than does lowering blood glucose. 36-39 Clustering of metabolic risk factors and cigarette smoking with hypertension is associated with the most significant increases in CVD risk; some other risk factors (eg, physical inactivity), although associated with a less significant short-term risk, may have substantial long-term effects, especially in the presence of multiple risk factors.1 Classic CVD risk-estimation guidelines were useful in identifying high-risk patients with hypertension who required intensive blood pressure-lowering therapy. However, researchers recognized the need to improve these guidelines because they overestimated or underestimated risk in various populations. In 2001, a workshop convened in the United States by the National Heart, Lung, and Blood Institute made several proposals to increase the predictive power of riskestimation guidelines.2° This group recommended pooling databases from large cohort studies to improve the predictive power in white and black middleaged populations and adjusting the then-current Framingham risk models for various ethnic groups with different absolute baseline population risks. Other recommendations included the following: (1) expanding database populations to better characterize the hazards facing high-risk patients (eg, those with diabetes, established CVD, noncoronary forms of atherosclerosis, or left ventricular hypertrophy); (2) extending CVD risk-assessment end points beyond fatal and nonfatal MI; (3) integrating underlying risk factors (eg, diet, obesity, physical inactivity, psychosocial factors, and family history of premature coronary heart disease); and (4) improving methods of clinical assessment of underlying risk factors. The workshop stated that future efforts should determine the independent predictive power of emerging risk factors, as well as measures of abnormal cardiovascular function, myocardial ischemia, and subclinical atherosclerosis. Another important objective is to extend risk prediction to long-term, including lifetime, risk of cardiovascular events and mortality. 2°
FOUNDATION FOR NEW RISK-ASSESSMENT MODELS
Recognition of the limitations of classic risk-assessment models has led to the development of newer models that use an integrated approach to CVD risk. Because the causes of CVD are multifactorial, the latest riskassessment guidelines account for numerous risk factors to advance the concept of total CVD risk. 4,6 The absolute risk determines the absolute benefit of treatment. 4° This shift in thinking means moving away from targeting a threshold blood pressure for intervention; instead, patients with mild elevations in blood pressure may be considered for treatment if they have a high global CVD risk. The Systematic Coronary Risk Evaluation (SCORE) project in Europe was among the first to develop paper-based, absolute CVD risk-assessment charts for various high-risk (Figure 2) and low-risk European populations, with an emphasis on fatal CVD events. 9 The most recent guidelines of the European Society of Hypertension (ESH) and the European Society of Cardiology (ESC) assess risk in patients with hypertension as low, moderate, high, and very high added risk (Table I). 4 These categories correlate with an approximate absolute 10-year CVD risk of <15%, 15% to 20%, >20% to 30%, and >30%, respectively, by the Framingham risk model, 7 or with an approximate absolute risk of fatal CVD of <4%, 4% to 5%, >5% to 8%, and >8%, respectively, by the SCORE charts. 9 The designation of added risk in the ESH-ESC guidelines considers not only level of blood pressure, but also CVD risk factors, target organ damage, diabetes, and associated clinical conditions (Table II). 4 The ESH-ESC guidelines include recently identified risk factors (eg, C-reactive protein) and measures of renal function as well as CVD risk factors with established adverse effects on morbidity and mortality. The ESHESC risk assessment is more comprehensive than that presented in the Seventh Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC 7) because the treatment algorithm in JNC 7 is based principally on blood pressure level plus consideration of established diabetes or chronic renal disease. 6 The treatment recommendations in JNC 7 are stratified on the basis of whether hypertension is present, with or without compelling indications (eg, heart failure post-MI, high coronary disease risk, diabetes, chronic renal disease, recurrent stroke prevention). Volume 27 Number 10
[] >_15% 10%-14% [] 5%-9%
[]
[] 3%-4%
[] 2% [] 1% [ ] <1%
Men
Women Nonsmoker
Nonsmoker
Smoker
Smoker
180 160
Age 65
~!~ ,.~ i:~ ~
140 120
180t 160{-
Age 60
I
140t_i_ E E
~:
~'~
140 I- 1/ 21
1~
180 t
16o[
o
>..
N@
~
/
120p,(t
-1"/
,L J l "
180 160
C, .1~ '
140
0
1
Age
-1 x
2:
i'- tfi, )~l\-
'-1-, I"
-1
ss
2t212
2
1~ /
21
it, :1,
fl
/
t
--
2
t-9/
2
2 :~,
1" " 1 /
2 ii@@~ii
F q:l4 q 2
.
-1
-1
1(
8ot! o o
0
0
0
0
0
0 ,,1/
~/
0
AgeS0
2
0
120
I!/// I/
-
/
I
...............
160F[ 0 /
0
0
0
0
0
0
0
0
0
140 k
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
I
I
I
I
I
I
I
I
I
I
f
I
I
I
I
I
I
I
I
l
4
S
6
7
8
4
5
6
7
8
4
5
6
7
8
4
S
6
7
8
/
120~mmol
I
I
I
Age 40
1" ,i" .1".* / 1". 2V , .1 2 2 0 0 -> 1- ,1" ~f. 5 ' ' 1 "t- 4
0
I
t
I
I
I
mg/dL 150 200 250 300
mg/dL 150 200 250 300 Cholesterol (mmol/L)
Figure 2. Estimated risk of cardiovascular mortality over 10 years in high-risk populations stratified by total cholesterol, systolic blood pressure, age, sex, and smoking status. Reprinted with permission. 9
The newest guidelines, particularly the ESH-ESC guidelines, are more liberal than earlier guidelines with respect to the definition and classification of blood pressure levels and the trigger level for intervention& 6 In addition, blood pressure categories have been added that do not denote hypertension (eg, normal and high normal, prehypertension), with recommendations for treatment based on other associated risk factors or con-
ditions. These changes recognize the multifactorial causes of CVD, clustering of risk factors, and different absolute baseline population risks, and suggest that the appropriate level of blood pressure for intervention is patient specific. The newest guidelines, however, still use discrete intervals of blood pressure to define risk rather than truly depict the graded, continuous risk associated with blood pressure.
: ~il :ii'/: ii!ii~6~!i
Table I. Stratification of risk to quantify prognosis based on European Society of Hypertension-European Society of Cardiology 2003 Guidelines.4 Added Risk by Blood Pressure, mm Hg
Other Risk Factors and Disease History
Normal: SBP 120-129 or DBP 80-84
High Normal: SBP 130-139 or DBP 85-89
No other risk factors
-
-
Grade 1 HTN: SBP 140-159 orDBP90-99
Grade 2 HTN: SBP 160-179 or DBP 100-109
Grade 3 HTN: SBP->180 or DBP->110
Low
Moderate
High
1-2 Risk factors
Low
Low
Moderate
Moderate
Very high
->3 Risk factors, TOD, or diabetes
Moderate
High
High
High
Very high
Associated clinical conditions
High
Very high
Very high
Very high
Very high
SBP = systolic blood pressure; DBP = diastolic blood pressure; HTN = hypertension; TOD = target organ damage.
Table II. Factors influencing prognosis based on European Society of Hypertension-European Society of Cardiology 2003 guidelines.4 Category
Risk Factor
CVD
Levels of SBP and DBP Men aged >55 years Women aged >65 years Smoking Dyslipidemia Family history of premature CVD Abdominal obesity C-reactive protein ->1 mg/dL
Diabetes mellitus
Fasting plasma glucose ->7.0 mmol/L (126 mg/dL) Postprandial plasma glucose >11.0 mmol/L (198 mg/dL)
Target organ damage
Left ventricular hypertrophy Arterial wall thickening on ultrasound Slight increase in serum creatinine Microalbuminuria
Associated clinical conditions
Cerebrovascular disease Heart disease Renal disease Peripheral vascular disease Advanced retinopathy
CVD = cardiovascular disease;SBP = systolic blood pressure; DBP = diastolic blood pressure.
1166411
'
=
i[:
~
Volume 27 Number 10
L.M. Ruilope and j. Segura THE VALUE O F THE VALUE STUDY
The Valsartan Antihypertensive Long-term Use Evaluation (VALUE) study was a large, randomized, double-blind trial comparing the long-term effects of the angiotensin receptor blocker (ARB) valsartan with those of a calcium antagonist (amlodipine) in patients with hypertension and a high risk of CVD who were monitored for a mean of 4.2 years. 41 Inclusion criteria for the VALUE cohort paralleled many of the suggestions for measuring total CVD risk in new risk-estimation guidelines. Indeed, the results of the VALUE study might be used to verify the utility of the new guidelines. In addition, the study has provided a large new database of patients who were stratified by many of the same risk and disease factors as used in the ESH-ESC guidelines (Table II) as well as information on several biochemical parameters. 4,2°,41 At baseline, mean blood pressures were only moderately elevated because 92.7% (7088/7649) of valsartan patients and 92.0% (6989/7596) of the amlodipine group were receiving antihypertensive drugs at randomization. Of note, there was no washout period, and it is reasonable to suspect that the persistent natriuretic and diuretic properties of calcium-channel blockers may have augmented the antihypertensive efficacy of amlodipine in the first few months of the study. 42 Nearly half of the patients had coronary heart disease (45.8% [7013/15,313]), approximately one third had diabetes (31.7% [4854/15,313]) or elevated cholesterol levels (33.0% [5053/15,313]), and nearly one quarter were current smokers (24.0% [3675/15,313]) or had proteinuria (22.5% [3445/15,313]). 43 As defined in both the ESH-ESC guidelines 4 and the US National Cholesterol Education Program's Third Adult Treatment Panel (ATP III) report, 44 patients in VALUE had a markedly increased risk for a CVD event or mortality, a risk level closer to that in the Heart Outcomes Prevention Evaluation than in previous hypertension trials. 4s Despite unequal blood pressure lowering between treatment arms in VALUE that favored amlodipine (particularly in the early phase of the trial), no difference was found in the primary composite outcome of cardiac morbidity and mortality. 41 For secondary end points, MI was significantly lower in the amlodipine group (hazard ratio [HR], 1.19; 95% CI, 1.02-1.38; P = 0.02), stroke was similar between groups (HR, 1.15; 95% CI, 0.98-1.35; P = NS), and the rate of hospitalization for heart failure suggested a slight benefit from
valsartan (HR, 0.89; 95% CI, 0.77-1.03; P = NS). A prespecified analysis of new-onset diabetes mellitus-a risk factor in both older and newer guidelines-demonstrated a significant 23% risk reduction with valsartan compared with amlodipine (P < 0 . 0 0 1 ) . 41 Interestingly, after correcting for between-group differences in blood pressure by serial median matching of patients, rates of MI and stroke were no different between treatment arms, whereas heart failure was significantly lower in the valsartan arm. 46 The striking success of this trial was its ability to control hypertension in this high-risk population to <140/80 mm Hg. VALUE also highlighted the importance of prompt blood pressure control and the necessity for combination therapy in many high-risk patients. 41 The results of VALUE might appear to conflict with those of the Losartan Intervention for Endpoint Reduction in Hypertension (LIFE) trial. 47 Although the average systolic blood pressure in LIFE was 7 mm Hg higher than in VALUE, the incidence of cardiac events and death was lower in LIFE. This apparent paradox can be resolved by taking into account the global risk of each patient population. LIFE, which compared the ARB losartan with the 13-blocker atenolol, enrolled patients who had lower cardiovascular risk because they were at an earlier stage of CVD than patients in VALUE. In fact, the higher blood pressure but lower incidence of cardiovascular events and death in LIFE as compared with VALUE lends support to the premise that cardiovascular risk is driven not only by elevated blood pressure, but also by the extent of vascular damage, primarily atherosclerosis. This observation is another piece of evidence supporting the newer, more sophisticated risk-assessment models. The typical overall 10-year CVD risk of the VALUE cohort can be estimated using any of the 3 recent methods (ESH-ESC, SCORE, and ATP III). If the ESH-ESC formula is used, the systolic blood pressure (mean [SD], 154.7 [19.0] mm Hg) classifies them as having grade 1 hypertension, and most participants appear to have >_2 additional risk factors or target organ damage or diabetes, thereby placing them at moderate to very high added risk (see Tables I and II for criteria). 4,43 Thus, a rough estimate of the anticipated event rate over 10 years would be 25%, and the mortality rate might be 6% to 7% over the same period. If the highrisk SCORE chart is used, which would be the best fit for a predominantly US and European population, the mean values for systolic blood pressure and total cho-
166S
lesterol plus sex, age, and smoking status suggest an average cardiovascular mortality of 11% over 10 years. (The SCORE risk charts do not extend above the age of 65 years, so the women in VALUE, who averaged - 7 0 years, will have an actual risk that is somewhat higher.) The ATP III guidelines, based on Framingham projections, imply that the patients randomized in VALUE have a <20% risk per 10 years for a major coronary event. (The baseline data in VALUE do not include levels of high-density lipoprotein cholesterol, a component of the Framingham point score. 44) The results for the composite end point of cardiovascular morbidity and mortality in VALUE demonstrated a nearly linear progression over time, with a rate of - 1 2 % over 5 years in both treatment arms, which would suggest a 24% rate for first events at 10 years. 4~ This assumption may be somewhat conservative; the Multiple Risk Factor Intervention Trial demonstrated increasing risk across a span of 10.5 years, particularly in the usual-care group. 48 On the other hand, it is reasonable to speculate that long-term treatment would show an increasing benefit over time. Mechanisms that might become more apparent over time include slowed progression of CVD risk factors, such as the reduction in new-onset diabetes and its consequences in the valsartan arm of VALUE. Further, if we assume that the CVD death rate is linear, then the 4% rate at 4.2 years would imply a 10% rate at 10 years. 41 Thus, the finer gradations of SCORE and the ESH-ESC projections seem to capture the CVD risk in the VALUE cohort better than the Framingham-based estimates.
CONCLUSIONS Accurate assessment of risk is essential for determining appropriate therapy for patients and maximizing the benefit of risk-reduction strategies. Risk-stratification models for patients with hypertension have evolved to become increasingly complex and nuanced. Classic risk assessments, generally based on data from the Framingham Heart Study, focus on blood pressure and a limited number of other risk factors and end points. The classic models lack predictive value in many patient populations, however, so that new models are needed. Further, classic assessment is premised on 5to 10-year risk, which is a barrier to treatment for younger patients who may have a low short-term risk but a high risk over longer time periods. Modern risk assessments include blood pressure as an important element in the global CVD morbidity~=~1
mortality equation. Because blood pressure is a continuum and hypertension rarely exists in isolation, treatment should be directed toward both reaching blood pressure goals and preventing or treating a wide array of modifiable risk factors that affect target organ damage and mortality. Patients at highest risk tend to have the most to gain by even modest improvements in their risk profile. New guidelines include many recently discovered risk factors, including several important measurements of renal dysfunction, which improve the recognition and treatment of patients with CVD. The VALUE trial highlights the increased accuracy of these new guidelines and provides a large new database of patients with high-risk hypertension.
REFERENCES 1. Ishizaka N, Ishizaka Y, Toda E, et al. Hypertension is the most common component of metabolic syndrome and the greatest contributor to carotid arteriosclerosis in apparently healthy Japanese individuals. Hypertens Res. 2005;28:27-34. 2. 2005 Heart and Stroke Statistical Update. Dallas, Tex: American Heart Association; 2004. 3. Guidelines Subcommittee. 1999 World Health Organization-International Society of Hypertension Guidelines for the Management of Hypertension. J H_ypertens. 1999;17:151-183. 4. European Society of Hypertension-European Society of Cardiology Guidelines Committee. 2003 European Society of Hypertension-European Society of Cardiology Guidelines for the Management of Arterial Hypertension [published correction appears in J Hjpertens. 2003;21: 2203-2204].J H.ypertens. 2003;21:1011-1053. S. The Sixth Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure. Arch Intern Med. 1997;157:2413-2446. 6. Chobanian AV, Bakris GL, Black HR, et al, for the National High Blood Pressure Education Program Coordinating Committee. The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure: TheJNC 7 report [published correction appears inJAMA. 2003;290: 197].JAMA. 2003;289:2560-2572. 7. Anderson KM, Wilson PW, Odell PM, Kannel WB. An updated coronary risk profile: A statement for health professionals. Circulation. 1991 ;83:356-362. 8. Kannel WB. Blood pressure as a cardiovascular risk factor: Prevention and treatment.JAMA. 1996;275:1571-1576. 9. Conroy RM, Py6r~il/~K, Fitzgerald AP, et al, for the SCORE Project Group. Estimation often-year risk of fatal cardiovascular disease in Europe: The SCORE Project. Eur Heart_/. 2003;24:987-1003. Volume 27 Number 10
LM. Ruilope :andJ~:segura
: 10. Wallis EJ, Ramsay LE, UI Haq I, et al. Coronary and cardiovascular risk estimation For primary prevention: Validation of a new Sheffield table in the 1995 Scottish health survey population. BMJ. 2000;320:671-676. 11. Wood D, Durrington P, Poulter N, et al. Joint British recommendations on prevention of coronary heart disease in clinical practice. Heart. 1998;80(Suppl 2):$1-$29. 12. New Zealand Guidelines Group. Management of Mildly Raised Blood Pressure in New Zealand. Wellington, Core Services Committee, New Zealand Ministry of Health, 1995. Available at: http://www.nzgg. org.nz/index.cfm ?fuseaction=about Nfuseubaction=docs&documentlD = 81. Accessed June 25, 2005. 13. Yikona Jl, Wallis EJ, Ramsay LE, Jackson PR. Coronary and cardiovascular risk estimation in uncomplicated mild hypertension. A comparison of risk assessment methods. J H vpertens. 2002;20:2173-2182. 14. Bastuji-Garin S, DeverlyA, Moyse D, et al, for the Intervention as a Goal in Hypertension Treatment Study Group. The Framingham prediction rule is not valid in a European population of treated hypertensive patients.J H.ype~ens.2002;20:1973-1980. 15. Persson M, Carlberg B, Weinehall L, et al. Risk stratification by guidelines compared with risk assessment by risk equations applied to a MONICA sample. J Hypertens. 2003;21:10891095. 16. Wilson PW, D'Agostino RB, Levy D, et al. Prediction of coronary heart disease using risk factor categories. Circulation. 1998;97:1837-1847. 17. StamlerJ, Stamler R, Neaton JD, et al. Low risk-factor profile and longterm cardiovascular and noncardiovascular mortality and life expectancy: Findings for 5 large cohorts of young adult and middle-aged men and wornen. JAMA. 1999;282: 2012-2018. 18. Hense HW, Schulte H, L6wel H, et al. Framingham risk function overes-
19.
20.
21.
22.
23.
24.
25.
26.
27.
timates risk of coronary heart disease in men and women from Germany-results from the MONICA Augsburg and the PROCAM cohorts. Eur HeartJ. 2003;24:937-945. D'Agostino RB Sr, Grundy S, Sullivan LM, Wilson P, ffor the CHD Risk Prediction Group. Validation off the Framingham coronary heart disease prediction scores: Results of a multiple ethnic groups investigation.JAMA. 2001 ;286:180-187. Grundy SM, D'Agostino RB Sr, Mosca L, et al. Cardiovascular risk assessment based on US cohort studies: Findings from a National Heart, Lung, and Blood Institute workshop. Circulation. 2001 ;104:491496. Ramachandran S, French JM, Vanderpump MP, et al. Using the Framingham model to predict heart disease in the United Kingdom: Retrospective study. BMJ. 2000;320: 676-677. Menotti A, Lanti M, Puddu PE, Kromhout D. Coronary heart disease incidence in northern and southern European populations: A reanalysis of the seven countries study for a European coronary risk chart. Heart. 2000;84:238-244. Menotti A, Puddu PE, Lanti M. Comparison of the Framingham risk Function-based coronary chart with risk function from an Italian population study. EurHeartJ. 2000;21:365370. Simpson FO. Guidelines for antihypertensive therapy: Problems with a strategy based on absolute cardiovascular risk. J Hypertens. 1996;14: 683-689. Zanchetti A. Antihypertensive therapy: How to evaluate the benefits. AmJ Cardiol. 1997;79:3-8. Alderman MH, Furberg CD, Kostis JB, et al. Hypertension guidelines: Criteria that might make them more clinically useful. Am J Hypertens. 2002;15:917-923. Hansson L. The Hypertension Optimal Treatment study and the impor-
tance of lowering blood pressure.
] H#pertensSuppl. 1999;17:$9-$13. 28. Hansson L, Zanchetti A, Carruthers SG, et al, for the HOT Study Group. Effects of intensive blood-pressure lowering and low-dose aspirin in patients with hypertension: Principal results of the Hypertension Optimal Treatment (HOT) randomised trial. Lancet. 1998;351 :I 755-I 762.
29. Zanchetti A, Hansson L, Clement D, et al, for the HOT study group. Benefits and risks of more intensive blood pressure lowering in hypertensive patients of the HOT study with different risk profiles: Does a J-shaped curve exist in smokers? J Hypertens. 2003;21:797-804. 30. MacMahon S, Peto R, Cutler J, et al. Blood pressure, stroke, and coronary heart disease. Part 1, prolonged differences in blood pressure: Prospective observational studies corrected For the regression dilution bias. Lancet. 1990;335:765-774. 31. STckowskiPA, 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 to 1990. Circulation. 1996;93:697-703. 32. Franklin SS, Wong ND. Cardiovascular risk evaluation: An inexact science. J Hypertens. 2002;20:21272130. 33. Zanchetti A, Hansson L, Dahl6f B, et al, for the HOT Study Group. Effects of individual risk Factors on the incidence of cardiovascular events in the treated hypertensive patients of the Hypertension Optimal Treatment study. J H.ypertens. 2001;19: 1149-1159. 34. Zanchetti A, Hansson L, M4nard J, et ak Risk assessment and treatment benefit in intensively treated hypertensive patients of the Hypertension Optimal Treatment (HOT) study. J Hypertens. 2001 ;19:819-825. 35. Zanchetti A, Ruilope LM. Antihypertensive treatment in patients with type-2 diabetes mellitus: What guid-
.
.
.
.
.
.
•
::. . . . . . .
::: I D D !
36.
37.
38.
39.
ance from recent controlled randomized trials? -/ Hypertens. 2002;20: 2099-2110. UK Prospective Diabetes Study Group. Tight blood pressure control and risk of macrovascular and microvascular complications in type 2 diabetes: UKPDS 38 [published correction appears in BM-/. 1999;318: 29]. BM-/. 1998;317:703-713. UK Prospective Diabetes Study (UKPDS) Group. Intensive bloodglucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). Lancet. 1998;352:837-853. Mogensen CE. Combined high blood pressure and glucose in type 2 diabetes: Double jeopardy. British trial shows clear effects of treatment, especially blood pressure reduction. BMJ. 1998;317:693-694. Julius S, Majahalme S, Palatini P. Antihypertensive treatment of patients
Summary of the Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III).-/AMA. 2001 ;285:2486-2497. 45. YusufS, Sleight P, Pogue J, et al, for the Heart Outcomes Prevention Evaluation Study Investigators. Effects of an angiotensin-converting-enzyme inhibitor, ramipril, on cardiovascular events in high-risk patients [published corrections appear in N Engl-/ Med. 2000;342:748 and N Engl-/Med. 2000;342:1376]. N EnglJ Med. 2000; 342:145-153.
46. Weber MA, Julius S, Kjeldsen SE, et al. Blood pressure dependent and independent effects ofantihypertensive treatment on clinica[ events in
the VALUE Trial. Lancet. 2004;363: 2049-2051. 47. Kjeldsen SE, DahlofB, Devereux RB, et al. Lowering of blood pressure and predictors of response in patients with left ventricular hypertrophy: The LIFE study. Losartan Intervention For Endpoint. Am/Hypertens. 2000;13:899-906. 48. Mortality after 10 1/2 years For hypertensive participants in the Multiple Risk Factor Intervention Trial. Circulation. 1990;82:1616-1628.
with diabetes and hypertension. Am-/ Hypertens. 2001 ;14:310S-316S. 40. Ramsay LE, Wallis EJ, Yeo WW, Jackson PR. The rationale for differing national recommendations for the treatment of hypertension. Am-/ H.ypertens. 1998;11:79S-88S. 41. Julius S, Kjeldsen SE, Weber M, et al, for the VALUE trial group. Outcomes in hypertensive patients at high cardiovascular risk treated with regimens based on valsartan or amIodipine: The VALUE randomised trial. Lancet. 2004;363:2022-2031. 42. RomeroJC, Raij L, GrangerJP, et al. Multiple effects of calcium entry blockers on renal function in hypertension. Hypertension. 1987;10:140-151. 43. Kjeldsen SE, Julius S, Brunner H, et al. Characteristics of 15,314 hypertensive patients at high coronary risk: The VALUE trial. The Valsartan Antihypertensive Long-term Use Evaluation. Blood Press. 2001 ;10:8391. 44. Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults. Executive
1668
Address correspondence to: Luis M. Ruilope, MD, Hypertension Unit, Hospital 12 de Octubre, 28041 Madrid, Spain. E-maiL [email protected] Volume 27 Number 10