- Email: [email protected]
Association between different lipid-lowering treatment strategies and blood pressure control in the Brisighella Heart Study
Prevention and Rehabilitation
Association between different lipid-lowering treatment strategies and blood pressure control in the Brisighella Heart Study Claudio Borghi, MD, Ada Dormi, MB, Maddalena Veronesi, MD, Zina Sangiorgi, MS, and Antonio Gaddi, MD, on behalf of the Brisighella Heart Study Working Party* Bologna, Italy
Background Small studies have suggested that lipid-lowering strategies, and particularly statins, could influence blood pressure (BP) control. The aim of the present study was to evaluate the effect of different lipid-lowering strategies on BP control of subjects with hypercholesterolemia who were enrolled in the prospective, population-based, longitudinal Brisighella Heart Study. Methods A total of 1356 subjects with total cholesterol levels ⱖ239 mg/dL were randomly treated for 5 years (1988 –1993) with 1 of these lipid-lowering regimens: low-fat diet, cholestyramine, gemfibrozil, or simvastatin. Participants were divided at baseline into 4 quartiles according to systolic BP level and examined for the percent change in systolic and diastolic BP during the 5 years of treatment. Results
A significant decrease in BP was observed in the 2 upper quartiles of systolic BP (ⱖ140 mm Hg) and was greater in subjects treated with cholesterol-lowering drugs who also had a greater reduction in plasma levels of low-density lipoprotein cholesterol. The BP decrease was greater in patients treated with statin drugs and, among those treated with antihypertensive drugs, in subjects in the fourth quartile.
Conclusion
The use of lipid-lowering measures could significantly improve BP control in subjects with both hypercholesterolemia and hypertension. The reduction in BP seems to be enhanced in subjects treated with statins. (Am Heart J 2004;148:285–92.)
Atherosclerosis and its complications (eg, myocardial infarction and stroke) are leading causes of adult morbidity and mortality in Europe and North America.1 The overall prevalence of atherosclerosis is strongly related to that of several cardiovascular risk factors, including high blood pressure, cigarette smoking, total plasma cholesterol level, low-density lipoprotein cholesterol level, high-density lipoprotein cholesterol level, and diabetes mellitus.2–5 Risk factor modification is an integral part of optimal care for patients with or without cardiovascular disease. An integrated approach to cardiovascular risk modification has been adopted as part of the framework of programs for the manage-
From the Department of Internal Medicine, University of Bologna, Bologna, Italy. *For a complete list of the Brisighella Heart Study Working Party see the appendix. Submitted July 10, 2003; accepted February 25, 2004. Reprint requests: Professor Claudio Borghi, MD, Divisione di Medicina Interna, Policlinico S. Orsola, Via Massarenti 9, 40138 Bologna, Italy. E-mail: [email protected] 0002-8703/$ - see front matter © 2004, Elsevier Inc. All rights reserved. doi:10.1016/j.ahj.2004.02.003
ment of hypertension (including the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure and the World Health Organization–International Society of Hypertension) and hypercholesterolemia (including the National Cholesterol Education Program) both in the United States and Europe.6 – 8 Data from large population studies have demonstrated that the long-term incidence of cardiovascular disease can be reduced significantly by following therapeutic algorithms that take into consideration the cumulative impact of treatment on several risk factors.5 This suggests that effective clinical management of cardiovascular risk should rely on treatment strategies that provide a comprehensive approach to multiple cardiovascular risk factors. The aim of this study was to evaluate the effects of different lipid-lowering strategies on the blood pressure profile of a large subcohort of patients with hypercholesterolemia who were at high risk for cardiovascular disease and were enrolled in the Brisighella Heart Study.
American Heart Journal August 2004
286 Borghi et al
Methods Study participants The Brisighella Heart Study, begun in 1972, is a prospective, population-based, longitudinal cohort study of randomly selected residents of Brisighella, Italy. The study design has been described in detail elsewhere.9,10 All study participants were between 14 and 84 years old and were free of cardiovascular disease at the time of enrollment. Subjects were examined clinically at baseline and every 4 years thereafter. During the follow-up examinations, clinical and laboratory data were obtained and morbidity and mortality rates were determined. In 1988, an interventional phase was begun that focused on a subpopulation of high-risk participants with total serum cholesterol values ⱖ239 mg/dL who had been treated with an aggressive lipid-lowering regimen for a cumulative period of 5 years (1988 –1993). The eligible study population included 1567 men and women with a total serum cholesterol level ⱖ239 mg/dL. None of the subjects had previously been treated with a lipid-lowering drug. All the subjects continued their current pharmacological treatment for the duration of the study, including those being treated for hypertension. The dosage of antihypertensive medication was maintained unchanged for the whole duration of the study period in 93% of the study population, whereas in the remaining 7% it was slightly adjusted throughout the study according to the extent of blood pressure control. The study protocol was initially approved by the Ethical Committee of the University of Bologna, and all study participants provided their informed consent before inclusion in the study. All the subjects entered a 6-week run-in phase, during which they received dietary advice designed to help them reduce their fat intake to ⬍30% of their total energy intake. Subjects with a total serum cholesterol level ⬍239 mg/dL after 6 weeks of dietary modification continued the diet, whereas subjects with a total serum cholesterol level ⬎239 mg/dL received lipid-lowering drug treatment. Subjects were randomized in a 2-to-1 ratio to receive a non-statin drug (gemfibrozil or cholestyramine) or a statin drug (simvastatin). When the total serum cholesterol level remained ⬎239 mg/dL after 6 months of single-drug therapy, subjects were treated with a combination regimen.
Study procedures At the time of the first examination, all subjects underwent a complete clinical evaluation, including a standard 12-lead electrocardiogram (ECG) and routine laboratory analysis. Resting baseline blood pressure measurements were taken, with a standard sphygmomanometer, to the nearest 2 mm Hg, and the average of 3 consecutive readings, taken at 1-minute intervals after 5 minutes of rest in the supine position, served as the accepted measurement. Resting heart rate was determined with ECG tracing. A duplicate fasting blood sample was drawn by venipuncture on 2 different occasions to determine serum total cholesterol levels. All the blood samples were centralized and analyzed in the same core laboratory located at the Center for Atherosclerosis of the University of Bologna. All patients were re-examined every 6 months with the same procedure for a cumulative follow-up period of 5 years.
For the purpose of this analysis, patients were divided at baseline into 4 quartiles of systolic blood pressure (⬍130 mm Hg, n ⫽ 392; 130 –139 mm Hg, n ⫽ 391; 140 –154 mm Hg, n ⫽ 392; ⬎154 mm Hg, n ⫽ 391) and examined for percent change in blood pressure from baseline to 5 years. The same analysis was carried out separately in a subgroup treated with antihypertensive drugs. The main comparisons were made among subjects treated with diet alone, subjects treated with statin drugs, and subjects treated with non-statin drugs.
Statistical analysis Results are expressed as means ⫾ SD. Statistical analysis was carried out with an SPSS statistical package (version 9.1 for Windows). Two-way analysis of variance was used to compare the normal distribution of baseline variables among all treatment strategies. The 2 test, with Yates correction when appropriate, was used to compare categorical variables at baseline. Two-way analysis of variance was also used to compare the changes in systolic and diastolic blood pressure observed in response to different treatment strategies in the overall population of subjects, as well as in the different quartiles of systolic blood pressure.
Results A total of 211 subjects were considered to be poor responders to single-drug therapy; they were treated with a combination of lipid-lowering drugs and excluded from the study. The final study sample consisted of 1356 subjects. Of these, 542 subjects were treated with a lipid-lowering diet, 550 were treated with non-statin drugs, and 264 were treated with statin drugs. All the patients were maintained as allocated to the same baseline quartile, thereby justifying some differences in the size of quartiles analyzed in this study. The daily dose range for gemfibrozil was 400 to 800 mg; for cholestyramine, the daily dose range was 4 to 24 g; and for simvastatin, the daily dose range was 20 to 40 mg. The baseline characteristics of subjects according to treatment group are summarized in Table I. The 3 treatment groups were comparable in age, sex distribution, body mass index, and baseline blood pressure values. The proportion of patients treated with antihypertensive drugs and the distribution of the different classes of drugs was comparable in the 3 treatment groups (data not shown). Mean serum cholesterol levels were significantly higher in patients taking lipidlowering drugs. Table II summarizes the baseline characteristics of subjects according to quartile of systolic blood pressure. Table III shows baseline data of subjects in subgroups within each quartile according to treatment group. Subjects in the upper quartiles were older and approximately 50% were women. No differences in total cholesterol or low-density lipoprotein cholesterol levels were observed. Systolic and diastolic blood pressures increased progressively from quartiles
American Heart Journal Volume 148, Number 2
1 to 4, with no differences observed among subgroups treated with different lipid-lowering strategies (Table III). Conversely, low-density lipoprotein levels were largely comparable among quartiles, but were significantly higher in patients randomized to receive lipidlowering drugs. Figure 1 depicts the percentage change in blood pressure observed in response to lipid-lowering therapy in the study population by quartile of systolic blood pressure. A significant decrease in systolic and diastolic blood pressure was observed in quartiles 3 and 4; the change in blood pressure observed in quartiles 1 and 2 (systolic blood pressure ⬍139 mm Hg) did not reach statistical significance. Among subjects in quartiles 3 and 4, the reduction in either systolic or diastolic blood pressure was greater in subjects treated with lipid-lowering drugs; these subjects also showed a greater reduction in levels of low-density lipoprotein cholesterol after 5 years of follow-up (Table IV). The effects on systolic and diastolic blood pressure of non-statin drugs were compared with those of statin drugs (Figure 2). Subjects treated with statin drugs had a significantly greater decrease in systolic and diastolic blood pressure than subjects treated with fibrates or cholestyramine, despite comparable reductions in levels of low-density lipoprotein cholesterol (Table IV). The difference among treatments occurred most frequently in the upper 2 quartiles of systolic blood pressure and was most apparent in subjects whose systolic blood pressure levels were ⬎154 mm Hg. The average body mass index values were slightly reduced in the whole study population, with no differences in patients treated with statin or non-statin strategies. To further investigate the putative role of the interactions among lipid-lowering interventions, blood pressure control, and hypertension, we analyzed separately the subset of 533 subjects treated with antihypertensive drugs. Most of these subjects were treated with diuretics (32.8%), angiotensin-converting enzyme inhibitors (28.8%), calcium channel blockers (16.5%), or a combination of these agents. The effects of lipid-lowering drugs on blood pressure control in such a population of patients were restricted to the upper 2 quartiles (Figure 3). In particular, the average reduction in blood pressure was enhanced in subjects in the fourth quartile, whose blood pressure control was poor despite receiving antihypertensive drug therapy. Conversely, no significant changes in blood pressure were observed in subjects in the lower 2 quartiles, with normal baseline blood pressure values. Subjects treated with statin drugs had a greater decrease in systolic and diastolic blood pressure, irrespective of baseline blood pressure values, which were largely comparable among the therapeutic subgroups (Table V).
Borghi et al 287
Table I. Baseline characteristics of the study participants according to treatment group
Variable
LipidLipidLipidlowering lowering lowering therapy therapy diet (nonstatin) (statin)
Patients (n) 542 550 264 Mean age (y) 53.8 ⫾ 10 56.6 ⫾ 13 56.3 ⫾ 9 Sex (%) (men/women) 57/43 58/42 56/44 24.7 ⫾ 3 25.2 ⫾ 3 24.8 ⫾ 3 BMI (kg/m2) Mean SBP (mm Hg) 140.9 ⫾ 15 141.9 ⫾ 17 139.6 ⫾ 14 Mean DBP (mm Hg) 87.2 ⫾ 9 87.5 ⫾ 10 86.5 ⫾ 9 Mean total cholesterol 264.3 ⫾ 44 283.1 ⫾ 48* 287.5 ⫾ 45* (mg/dL) Mean LDL cholesterol 163.7 ⫾ 22 197.2 ⫾ 37* 198.5 ⫾ 33* (mg/dL) Patients with antihypertensive 201 (37.8) 299 (54.3) 133 (50.3) drugs, n (%) Diuretics (%) 33.9 32.1 32.4 ACE inhibitors (%) 27.3 29.1 30.0 CCBs (%) 17.8 16.3 15.4 -Blockers (%) 11.7 10.9 11.4 Combination (%) 23.1 24.0 23.8 SBP, Systolic blood pressure; DBP, diastolic blood pressure; LDL, low-density lipoprotein. *P ⬍ .001 versus diet.
Discussion This analysis of participants in the Brisighella Heart Study has provided a number of stimulating findings that may have relevance in the clinical treatment of patients who are at high risk for cardiovascular disease. The study supports the hypothesis that the longterm use of lipid-lowering measures significantly improves blood pressure control in patients with both hypercholesterolemia and hypertension. The reduction in blood pressure was particularly evident in subjects treated with statin drugs. Furthermore, the effect of lipid-lowering drugs was restricted to subjects with higher baseline blood pressure values, poorer responses to antihypertensive treatment, or both. High blood pressure and hypercholesterolemia frequently occur concomitantly in the same patients and may account for a large proportion of the cumulative risk of cardiovascular disease. A significant increase in cholesterol levels has been observed in as much as 40% of middle-aged (35– 65 years old) patients with hypertension in the United States, whereas the prevalence of hypercholesterolemia has a linear relationship with the severity of hypertension.11 A recent review of the data collected by the National Health and Nutrition Examination (NHANES) III survey has provided support for this observation by demonstrating that the proportion of patients with hypercholesterolemia increases progressively with the degree of severity of hypertension.12 When hypertension and hypercholes-
American Heart Journal August 2004
288 Borghi et al
Table II. Baseline characteristics of the study participants according to quartile of systolic blood pressure Quartile 1
2
3
4
P value for trend
333 55.2 ⫾ 10 65/35 118.7 ⫾ 6 78.8 ⫾ 6 271.3 ⫾ 44 188.7 ⫾ 21
313 59.9 ⫾ 7 57/43 131.9 ⫾ 3 84.0 ⫾ 6 273.1 ⫾ 48 186.2 ⫾ 26
370 61.2 ⫾ 8 55/45 145.3 ⫾ 6 88.9 ⫾ 7 267.5 ⫾ 45 184.7 ⫾ 28
340 65.5 ⫾ 8 51/49 167.4 ⫾ 6 96.7 ⫾ 9 269.8 ⫾ 41 186.7 ⫾ 23
⬍.0001 ⬍.001 ⬍.00001 ⬍.0001 ns ns
Variable Patients (n) Mean age (y) Sex (%) (men/women) Mean SBP (mm Hg) Mean DBP (mm Hg) Mean total cholesterol (mg/dL) LDL cholesterol (mg/dL) ns, Not significant.
Table III. Baseline characteristics in the 4 quartiles of systolic blood pressure according to treatment group
Quartile 1 (n ⫽ 333)
2 (n ⫽ 313)
3 (n ⫽ 370)
4 (n ⫽ 340)
Variable
LipidLowering Diet
LipidLowering Therapy (Non-statin)
LipidLowering Therapy (Statin)
SBP DBP LDL-C SBP DBP LDL-C SBP DBP LDL-C SBP DBP LDL-C
119.3 ⫾ 6 79.3 ⫾ 7 169.0 ⫾ 21 132.6 ⫾ 3† 84.7 ⫾ 6† 163.0 ⫾ 26 145.3 ⫾ 4‡§ 89.5 ⫾ 7‡§ 161.5 ⫾ 32 166.8 ⫾ 12㛳¶# 95.6 ⫾ 11㛳¶# 162.3 ⫾ 29
118.9 ⫾ 6 78.6 ⫾ 7 202.0 ⫾ 33* 132.8 ⫾ 3† 85.4 ⫾ 6† 193.2 ⫾ 33* 145.2 ⫾ 4‡§ 88.8 ⫾ 7‡§ 198.3 ⫾ 36* 170.7 ⫾ 11㛳¶# 97.3 ⫾ 16㛳¶ 197.4 ⫾ 32*
118.0 ⫾ 5 78.5 ⫾ 6 195.3 ⫾ 27* 130.3 ⫾ 1† 82.0 ⫾ 8† 203.1 ⫾ 44* 145.6 ⫾ 6‡§ 88.3 ⫾ 8‡§ 195.2 ⫾ 39* 164.7 ⫾ 5㛳¶# 97.2 ⫾ 6㛳¶# 201.2 ⫾ 41*
LDL-C, Low-density lipoprotein cholesterol. *P ⬍ .001 versus diet. †P ⬍ .005 versus quartile 1. ‡P ⬍ .001 versus quartiles 1 and 2. §P ⬍ .005 versus quartiles 1 and 2. 㛳P ⬍ .0001 versus quartiles 1, 2, and 3. ¶P ⬍ .001 versus quartiles 1, 2, and 3. #P ⬍ .005 versus quartiles 1, 2, and 3.
terolemia occur in the same patient, they amplify the absolute level of risk for cardiovascular complications significantly. The results of the Multiple Risk Factor Intervention Trial demonstrated that, at any blood pressure level, mortality caused by coronary artery disease is greater in patients with higher serum cholesterol levels.3 This suggests the importance of an extensive and systematic evaluation of serum cholesterol levels in patients with hypertension and in patients with only a marginal change in blood pressure (eg, subjects with high-normal blood pressure). The possible effects of interactions between serum cholesterol and blood pressure are not limited to their contribution to the pathogenesis of atherosclerosis and its complications. Hypercholesterolemia might also be implicated, directly or indirectly, with some of the
mechanisms involved in the development of hypertension. A beneficial effect of various lipid-lowering measures on blood pressure control has been described in several retrospective clinical trials carried out in ⬎17,000 patients with hypercholesterolemia.13 In this extensive review by Goode et al, a positive linear relationship between the absolute reduction in cholesterol level and the extent of blood pressure reduction was reported in a broad range of patients with hypercholesterolemia who were receiving cholesterol-lowering treatment.13 Beneficial effects of lipid-lowering treatment has been shown in studies of patients with borderline hypertension14,15 and in selected subsets of patients with untreated hypertension, diabetes mellitus, or both who were receiving statin treatment.16 –18 A primary prevention study involving a large popula-
American Heart Journal Volume 148, Number 2
Borghi et al 289
Figure 1
Figure 2
Percent change in systolic and diastolic blood pressure after 5 years of treatment with lipid-lowering diet or lipid-lowering drugs in quartiles of baseline systolic blood pressure. *P ⬍.001; †P ⬍.005; ‡P ⬍.0001 vs quartile 1. §P ⬍.005; 㛳P ⬍.001 vs diet.
Percent change in systolic and diastolic blood pressure after 5 years of treatment with statins and non-statin lipid-lowering drugs in quartiles of baseline systolic blood pressure. *P ⬍.001; †P ⬍.005; ‡P ⬍.0001 vs quartile 1; §P ⬍.005 vs non-statin drugs.
Table IV. Mean change in LDL-cholesterol levels according to quartile of systolic blood pressure in each treatment group
Quartile 1 2 3 4
Lipidlowering diet
Lipidlowering therapy (nonstatin)
⫺3.5% ⫾ 6% ⫺3.1% ⫾ 8% ⫺3.7% ⫾ 6% ⫺3.1% ⫾ 11%
⫺13.9% ⫾ 9%* ⫺12.5% ⫾ 8%* ⫺16.1% ⫾ 8%* ⫺16.2% ⫾ 8%*
Lipidlowering therapy (statin)
Figure 3
⫺15.8% ⫾ 8%* ⫺12.8% ⫾ 8%* ⫺16.9% ⫾ 10%* ⫺18.4% ⫾ 10%*
*P ⬍ .001 versus diet.
tion of patients with hypercholesterolemia revealed a 25% reduction in the incidence of new cases of hypertension in men whose serum cholesterol levels were reduced with clofibrate.19 Glorioso et al20 prospectively investigated the effects of pravastatin on the blood pressure profile in patients with hypercholesterolemia and untreated hypertension and found that pravastatin decreased resting systolic, diastolic, and pulse pressures and blunted the pressor response to the cold pressor test. This effect was independent of age, sex, and baseline serum levels of lowdensity lipoprotein and high-density lipoprotein cholesterol, which suggests that statin drugs might interfere with some of the basic mechanisms responsible for blood pressure control. More recently, a blood pressure-lowering effect was demonstrated for atorvastatin in patients with isolated systolic hypertension.21 In such patients, systolic and diastolic blood pressure reduction has been associated with an increase in the peripheral vasodilatory capacity that largely supports the capacity of statin drugs to reduce arterial stiffness and improve endothelium-dependent vasodilatation.
Percent change in systolic and diastolic blood pressure after 5 years of treatment with lipid-lowering measures in quartiles of baseline systolic blood pressure and in subjects treated with antihypertensive drugs. *P ⬍.001; †P ⬍.005 vs quartile 1; ‡P ⬍.005 vs non-statin drugs; §P ⬍.005 vs lipid-lowering diet.
The results of our study support the hypothesis that the effects of statin drugs on blood pressure control are independent of the extent of cholesterol reduction. The reduction in systolic and diastolic blood pressure associated with the use of statin drugs was significantly greater than that observed with non-statin medications, despite a comparable cholesterol level reduction. These results are broadly consistent with those of a study directly comparing simvastatin and cholestyramine in patients with hypertension and type 2 diabetes mellitus,22 which demonstrated that the 2 drugs
American Heart Journal August 2004
290 Borghi et al
Table V. Baseline characteristics of patients treated with antihypertensive drugs and included in the 4 quartiles of systolic blood pressure allocated to different lipid-lowering strategies
Quartile
Variable
1 (n ⫽ 17) 2 (n ⫽ 106) 3 (n ⫽ 162) 4 (n ⫽ 348)
SBP DBP SBP DBP SBP DBP SBP DBP
Lipidlowering diet
Lipidlowering therapy (nonstatin)
Lipidlowering therapy (statin)
121.0 ⫾ 4 79.6 ⫾ 9 133.9 ⫾ 2* 86.0 ⫾ 5* 147.1 ⫾ 4†‡ 89.1 ⫾ 7†‡ 170.9 ⫾ 14§㛳¶ 97.3 ⫾ 10§㛳¶
123.6 ⫾ 6 79.6 ⫾ 5 133.5 ⫾ 3* 89.3 ⫾ 8* 146.4 ⫾ 4†‡ 85.8 ⫾ 11†‡ 170.1 ⫾ 14§㛳¶ 96.5 ⫾ 13§㛳¶
124.2 ⫾ 4 79.3 ⫾ 6 130.0 ⫾ 1* 87.1 ⫾ 8* 151.2 ⫾ 1†‡ 85.7 ⫾ 5†‡ 169.3 ⫾ 5§㛳¶ 96.0 ⫾ 7§㛳¶
*P ⬍ .001 versus diet. †P ⬍ .005 versus quartile 1. ‡P ⬍ .001 versus quartiles 1 and 2. §P ⬍ .005 versus quartiles 1 and 2. 㛳P ⬍ .0001 versus quartiles 1, 2, and 3. ¶P ⬍ .001 versus quartiles 1, 2, and 3. #P ⬍ .005 versus quartiles 1, 2, and 3.
were equally effective in reducing total and low-density lipoprotein cholesterol levels, but only statin therapy led to a significant reduction in diastolic blood pressure. Terzoli et al23 reported that the blood pressure-lowering effect of statin drugs is more pronounced in patients who had a reduction in cholesterol level ⬎50 mg/dL (1.3 mmol/L), which suggests that a function based on a cutoff point can better explain the onset of the blood pressure lowering effect associated with statin treatment. However, most of the studies emphasizing the importance of cholesterol reduction in blood pressure control have been conducted by comparing the effect of lipid-lowering drugs with placebo, thereby preventing any possibility of distinguishing the primary effects of the drug from the consequences of serum cholesterol level reduction. For all these reasons, the issue of the relative contribution of serum cholesterol level reduction to the blood pressure-lowering effect of statin drugs is still a matter of debate and deserves further exploration with the results of ongoing clinical trials (eg, the Antihypertensive and Lipid Lowering Treatment to Prevent Heart Attack Trial). One of the most interesting observations arising from this study is that lipid-lowering drugs improve blood pressure control only in patients whose blood pressure is in the hypertensive range. This selective effect of statin drugs was demonstrated in the overall study population and in the subgroups of patients who were treated with antihypertensive drugs. These findings are in agreement with most of the data reported in the literature demonstrating the lack of any blood pressure-lowering effect for statin drugs in subjects
who are normotensive24 –29 or in patients with hypertension treated with adequate blood pressure control.30 –33 These studies suggest that the effects of statin drugs on blood pressure are probably restricted to patients with concomitant hypercholesterolemia and uncontrolled hypertension in whom modulation of peripheral vascular tone and vascular resistance is significantly impaired. Taken together, these data clearly support the possibility that the extensive use of lipid-lowering drugs, particularly the use of statin drugs, could significantly improve blood pressure control (and its clinical outcome) in patients with concomitant hypertension and hypercholesterolemia. Despite such encouraging conclusions, this study has some important methodological limitations. First, the Brisighella Heart Study was not specifically designed to test the hypothesis of any effect of cholesterol-lowering drugs on blood pressure control in patients with hypercholesterolemia. Second, within the group of patients treated with non-statin cholesterol-lowering drugs, the administration of fibrates and cholestyramine was not randomized. Third, in accordance with the observational nature of the Brisighella Heart Study, which was designed to assess the role of lipid abnormalities on the long-term incidence of cardiovascular disease, the demographic description of this population is lacking some important information (eg, prevalence of family history of hypertension, renal function, concomitant diseases, and concomitant pharmacological treatments). Finally, in the treated hypertensive population, the use of the various antihypertensive drugs was not previously randomized. This prevented the possibility of obtaining statistically reli-
American Heart Journal Volume 148, Number 2
able information on the antihypertensive effect of the various combinations of lipid-lowering drugs and antihypertensive medications in this study.
Conclusions Whether such limitations are relevant to the interpretation of the results of this study, the evidence presented strongly suggests that the interaction between high blood pressure and hypercholesterolemia is probably not solely limited to their reciprocal roles as risk factors for atherosclerosis. In particular, the blood pressure-lowering effect of statin drugs observed in patients with hypertension and hypercholesterolemia in this study suggests that a broader preventive approach to cardiovascular disease should consider the future possibility of a “therapeutic crossover” of drugs that are designed to treat different risk factors. We thank Elizabeth Melby for the revision of the manuscript and Ettore Ambrosioni for continuous support for research.
References 1. American Heart Association. Heart and stroke statistical update. Dallas (Tex): American Heart Association; 2000. 2. Kannel WB. Blood pressure as a cardiovascular risk factor: prevention and treatment. JAMA 1996;275:1571– 6. 3. Neaton JD, Blackburn H, Jacobs D, et al. Serum cholesterol level and mortality findings for men screened in the Multiple Risk Factor Intervention Trial. Multiple Risk Factor Intervention Trial Research Group. Arch Intern Med 1992;152:1490 –500. 4. Alderman MH, Cohen H, Madhavan S. Diabetes and cardiovascular events in hypertensive patients. Hypertension 1999;33:1130 – 4. 5. Wilson PWF, D’Agostino RB, Levy D, et al. Prediction of coronary heart disease using risk factor categories. Circulation 1998;97: 1837– 47. 6. The JNC-VI Sub-Committee. The sixth report of the Joint National Committee on prevention, detection, evaluation, and treatment of high blood pressure. Arch Intern Med 1997;157:2413– 46. 7. Chalmers J, MacMahon S, Mancia G, et al. 1999 World Health Organization–International Society of Hypertension Guidelines for the management of hypertension. Guidelines subcommittee of the World Health Organization. Clin Exp Hypertens 1999;21:1009 – 60. 8. Executive 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). JAMA 2001;285:2486 –97. 9. Descovich GC. The Brisighella Heart Study: an interim report. Eur Heart J 1990;11(Suppl H):32–7. 10. Borghi C, Dormi A, Ambrosioni E, et al. Relative role of systolic, diastolic and pulse pressure as risk factors for cardiovascular events in the Brisighella Heart Study. J Hypertens 2002;20:1737– 42. 11. Laurenzi M, Cirillo M, Trevisan M. The Gubbio data: epidemiology and pathophysiology. Gubbio Study Research Group. Clin Exp Hypertens 1992;14:261–9.
Borghi et al 291
12. Muntner P, He J, Roccella EJ, et al. The impact of JNC-VI guidelines on treatment recommendations in the US population. Hypertension 2002;39:897–902. 13. Goode GK, Miller JP, Heagerty AM. Hyperlipidaemia, hypertension, and coronary heart disease. Lancet 1995;345:362– 4. 14. Julius S, Jamerson K, Mejia A, et al. The association of borderline hypertension with target organ changes and higher coronary risk: Tecumseh Blood Pressure study. JAMA 1990;264:354 – 8. 15. Borghi C, Bacchelli S, Degli Esposti D, et al. Pressor and metabolic correlates to long-term development of stable hypertension in borderline hypertensive patients [abstract]. J Hypertens 1997;15: S111. 16. Borghi C, Veronesi M, Prandin MG, et al. Statins and blood pressure regulation. Curr Hypertens Rep 2001;3:281– 8. 17. Borghi C, Dormi A, Veronesi M, et al. Use of lipid-lowering drugs and blood pressure control in patients with arterial hypertension. J Clin Hypertens 2002;4:277– 85. 18. Borghi C. Interactions between hypercholesterolemia and hypertension: implications for therapy. Curr Opin Nephrol Hypertens 2002;11:489 –96. 19. Committee of Principal Investigators. A co-operative trial in the primary prevention of ischemic heart disease using clofibrate. Br Heart J 1978;40:1069 –118. 20. Glorioso N, Troffa C, Filigheddu F, et al. Effect of the HMG-CoA reductase inhibitors on blood pressure in patients with essential hypertension and primary hypercholesterolemia. Hypertension 1999;34:1281– 6. 21. Ferrier KE, Muhlmann MH, Baguet JP, et al. Intensive cholesterol reduction lowers blood pressure and large artery stiffness in isolated systolic hypertension. J Am Coll Cardiol 2002;39:1020 –5. 22. Tonolo G, Melis MG, Formato M, et al. Additive effects of Simvastatin beyond its effects on LDL cholesterol in hypertensive type 2 diabetic patients. Eur J Clin Invest 2000;30:980 –7. 23. Terzoli L, Raco R, Mircoli L, et al. Lowering of ambulatory and daytime (but not clinical and nighttime) blood pressure by statin treatment [abstract]. J Hypertens 2002;20(Suppl 4):S4. 24. Muramatsu J, Kobayashi A, Hasegawa N, et al. Hemodynamic changes associated with reduction in total cholesterol by treatment with the HMG-CoA reductase inhibitor pravastatin. Atherosclerosis 1997;130:179 – 82. 25. Antonicelli R, Onorato G, Pagelli P, et al. Simvastatin in the treatment of hypercholesterolemia in elderly patients. Clin Ther 1990; 12:165–71. 26. Kool M, Lustermans F, Kragten H, et al. Does lowering of cholesterol levels influence functional properties of large arteries? Eur J Clin Pharmacol 1995;48:217–23. 27. Sartor G, Katzman P, Eizyk E, et al. Simvastatin treatment of hypercholesterolemia in patients with insulin dependent diabetes mellitus. Int J Clin Pharmacol Ther 1995;33:3– 6. 28. Sung BH, Izzo JL, Wilson MF. Effects of cholesterol reduction on blood pressure response to mental stress in patients with high cholesterol. Am J Hypertens 1997;10:592–9. 29. van Etten RW, de Koning EJ, Honing ML, et al. Intensive lipid lowering by statin therapy does not improve vasoreactivity in patients with type 2 diabetes. Arterioscler Thromb Vasc Biol 2002;22: 799 – 804. 30. D’Agostino RB, Kannel WB, Stepanians MN, et al. Efficacy and tolerability of lovastatin in hypercholesterolemia in patients with systemic hypertension. Am J Cardiol 1993;71:82–7.
American Heart Journal August 2004
292 Borghi et al
31. O’Callaghan CJ, Krum H, Conway EL, et al. Short term effects of pravastatin on blood pressure in hypercholesterolaemic hypertensive patients. Blood Press 1994;3:404 – 6. 32. Arteaga J, Sorbet M, Anda E, et al. Atorvastatin associated to valsartan does not improve blood pressure control or proteinuria in patients with diabetic nephropathy [abstract]. Am J Hypertens 2002;15(4 part 2):57A. 33. Mancia G, Crepaldi G, Gallus G, et al. Long-term statin administration and ambulatory blood pressure in mild hypertensive and hypercholesterolemic patients [abstract]. The PHYLLIS study. J Hypertens 2002;20:S4.
Appendix Brisighella Heart Study Working Party
Professor Gian Carlo Descovich, Brisighella Study founder; Professor A. Gaddi, MD, PhD, scientific coordinator; E. Pelliconi, Brisighella coordinator, past mayor; C. Sangiorgi, current mayor. Bologna Medical Staff: C. Borghi, MD; S. D’Addato, MD, PhD; A. Fiorito; C. Galetti, MD; M.C. Grippo, MD; V. Immordino, MD; M. L. Malkowski, MD; C. Mussoni, MD, PhD, Brisighella team project coordinator; A. Moretti, MD; S. Nascetti, MD; A.R. Reggiani, MD; S. Rimondi, MD. Data handling and biometrics: A. Dormi, MathD. Laboratory staff: Z. Sangiorgi, BD; R. Mambelli. Brisighella general practitioner staff: L. Bagnaresi, MD; E. Belletti, MD; Gamberi, MD; A. Naldi, MD; C. Samore`, MD; P. Viozzi, MD. Anagraphic staff: M. Nanni. Brisighella Heart Price Committee: E. Descovich; A. Gaddi; C. Sangiorgi.
The following article is an AHJ Online Exclusive. Full text of this article is available at no charge at our Website: www.mosby.com/ahj.
Is angiotensin-converting enzyme inhibitor useful in a Japanese population for secondary prevention after acute myocardial infarction? A final report of the Japanese Acute Myocardial Infarction Prospective (JAMP) study Kenji Ueshima, MD,a Kennichi Fukami, MD,a Katsuhiko Hiramori, MD,a Saichi Hosoda, MD,b Hiroshi Kishida, MD,c Kazuzo Kato, MD,d Tsuyoshi Fujita, MD,e Kiichiro Tsutani, MD,f and Akira Sakuma, PhD,g for the Japanese Angina and Myocardial Infarction Prospective (JAMP) study group* Morioka, Tokyo, and Osaka, Japan
Background
Although angiotensin-converting enzyme (ACE) inhibi-
tors have appeared to be useful for secondary prevention after acute myocardial infarction (AMI) in Western countries, that has not been confirmed in non-western countries. We investigated whether ACE inhibitors improve survival rates in patients who have survived an AMI in Japan.
Methods
A randomized controlled trial, the first non-pharmaceutical
company-supported multicenter trial of a medication in Japan, was carried out in 48 institutions from 1993 to 2000. A total of 888 of 1163 patients with AMI were eligible for the full analysis set (FAS). The mean patient
tion because of worsening angina or congestive heart failure. The mean follow-up period was 5.8 years.
Results
There were no significant differences in the 2 groups in base-
line data. During the follow-up period, 3 patients were lost to follow-up. With Kaplan-Meier analysis, the annual rate of total cardiac events was 32% in both groups. After adjustment for clinical baseline data, ACE inhibitor administration was not revealed with Cox regression analysis to have a significant prognostic effect in our study.
age was 62 years, and 78% of patients were men. Subjects were ran-
Conclusion
domized to 2 groups; 422 received ACE inhibitors and 466 did not re-
with ACE inhibitor administration in subjects who survived after AMI in a
ceive ACE inhibitors. The primary end point was combined cardiac
Japanese study population. Further evaluations with a larger population or
events, which was defined as cardiac or non-cardiac death, recurrent non-
in subjects who are at a higher risk for AMI are necessary to confirm our
fatal myocardial infarction, coronary revascularization, and hospitaliza-
findings. (Am Heart J 2004;148:e8.)
We did not show a significant improvement in outcome
Association between different lipid-lowering treatment strategies and blood pressure control in the Brisighella Heart Study Claudio Borghi, MD, Ada Dormi, MB, Maddalena Veronesi, MD, Zina Sangiorgi, MS, and Antonio Gaddi, MD, on behalf of the Brisighella Heart Study Working Party* Bologna, Italy
Background Small studies have suggested that lipid-lowering strategies, and particularly statins, could influence blood pressure (BP) control. The aim of the present study was to evaluate the effect of different lipid-lowering strategies on BP control of subjects with hypercholesterolemia who were enrolled in the prospective, population-based, longitudinal Brisighella Heart Study. Methods A total of 1356 subjects with total cholesterol levels ⱖ239 mg/dL were randomly treated for 5 years (1988 –1993) with 1 of these lipid-lowering regimens: low-fat diet, cholestyramine, gemfibrozil, or simvastatin. Participants were divided at baseline into 4 quartiles according to systolic BP level and examined for the percent change in systolic and diastolic BP during the 5 years of treatment. Results
A significant decrease in BP was observed in the 2 upper quartiles of systolic BP (ⱖ140 mm Hg) and was greater in subjects treated with cholesterol-lowering drugs who also had a greater reduction in plasma levels of low-density lipoprotein cholesterol. The BP decrease was greater in patients treated with statin drugs and, among those treated with antihypertensive drugs, in subjects in the fourth quartile.
Conclusion
The use of lipid-lowering measures could significantly improve BP control in subjects with both hypercholesterolemia and hypertension. The reduction in BP seems to be enhanced in subjects treated with statins. (Am Heart J 2004;148:285–92.)
Atherosclerosis and its complications (eg, myocardial infarction and stroke) are leading causes of adult morbidity and mortality in Europe and North America.1 The overall prevalence of atherosclerosis is strongly related to that of several cardiovascular risk factors, including high blood pressure, cigarette smoking, total plasma cholesterol level, low-density lipoprotein cholesterol level, high-density lipoprotein cholesterol level, and diabetes mellitus.2–5 Risk factor modification is an integral part of optimal care for patients with or without cardiovascular disease. An integrated approach to cardiovascular risk modification has been adopted as part of the framework of programs for the manage-
From the Department of Internal Medicine, University of Bologna, Bologna, Italy. *For a complete list of the Brisighella Heart Study Working Party see the appendix. Submitted July 10, 2003; accepted February 25, 2004. Reprint requests: Professor Claudio Borghi, MD, Divisione di Medicina Interna, Policlinico S. Orsola, Via Massarenti 9, 40138 Bologna, Italy. E-mail: [email protected] 0002-8703/$ - see front matter © 2004, Elsevier Inc. All rights reserved. doi:10.1016/j.ahj.2004.02.003
ment of hypertension (including the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure and the World Health Organization–International Society of Hypertension) and hypercholesterolemia (including the National Cholesterol Education Program) both in the United States and Europe.6 – 8 Data from large population studies have demonstrated that the long-term incidence of cardiovascular disease can be reduced significantly by following therapeutic algorithms that take into consideration the cumulative impact of treatment on several risk factors.5 This suggests that effective clinical management of cardiovascular risk should rely on treatment strategies that provide a comprehensive approach to multiple cardiovascular risk factors. The aim of this study was to evaluate the effects of different lipid-lowering strategies on the blood pressure profile of a large subcohort of patients with hypercholesterolemia who were at high risk for cardiovascular disease and were enrolled in the Brisighella Heart Study.
American Heart Journal August 2004
286 Borghi et al
Methods Study participants The Brisighella Heart Study, begun in 1972, is a prospective, population-based, longitudinal cohort study of randomly selected residents of Brisighella, Italy. The study design has been described in detail elsewhere.9,10 All study participants were between 14 and 84 years old and were free of cardiovascular disease at the time of enrollment. Subjects were examined clinically at baseline and every 4 years thereafter. During the follow-up examinations, clinical and laboratory data were obtained and morbidity and mortality rates were determined. In 1988, an interventional phase was begun that focused on a subpopulation of high-risk participants with total serum cholesterol values ⱖ239 mg/dL who had been treated with an aggressive lipid-lowering regimen for a cumulative period of 5 years (1988 –1993). The eligible study population included 1567 men and women with a total serum cholesterol level ⱖ239 mg/dL. None of the subjects had previously been treated with a lipid-lowering drug. All the subjects continued their current pharmacological treatment for the duration of the study, including those being treated for hypertension. The dosage of antihypertensive medication was maintained unchanged for the whole duration of the study period in 93% of the study population, whereas in the remaining 7% it was slightly adjusted throughout the study according to the extent of blood pressure control. The study protocol was initially approved by the Ethical Committee of the University of Bologna, and all study participants provided their informed consent before inclusion in the study. All the subjects entered a 6-week run-in phase, during which they received dietary advice designed to help them reduce their fat intake to ⬍30% of their total energy intake. Subjects with a total serum cholesterol level ⬍239 mg/dL after 6 weeks of dietary modification continued the diet, whereas subjects with a total serum cholesterol level ⬎239 mg/dL received lipid-lowering drug treatment. Subjects were randomized in a 2-to-1 ratio to receive a non-statin drug (gemfibrozil or cholestyramine) or a statin drug (simvastatin). When the total serum cholesterol level remained ⬎239 mg/dL after 6 months of single-drug therapy, subjects were treated with a combination regimen.
Study procedures At the time of the first examination, all subjects underwent a complete clinical evaluation, including a standard 12-lead electrocardiogram (ECG) and routine laboratory analysis. Resting baseline blood pressure measurements were taken, with a standard sphygmomanometer, to the nearest 2 mm Hg, and the average of 3 consecutive readings, taken at 1-minute intervals after 5 minutes of rest in the supine position, served as the accepted measurement. Resting heart rate was determined with ECG tracing. A duplicate fasting blood sample was drawn by venipuncture on 2 different occasions to determine serum total cholesterol levels. All the blood samples were centralized and analyzed in the same core laboratory located at the Center for Atherosclerosis of the University of Bologna. All patients were re-examined every 6 months with the same procedure for a cumulative follow-up period of 5 years.
For the purpose of this analysis, patients were divided at baseline into 4 quartiles of systolic blood pressure (⬍130 mm Hg, n ⫽ 392; 130 –139 mm Hg, n ⫽ 391; 140 –154 mm Hg, n ⫽ 392; ⬎154 mm Hg, n ⫽ 391) and examined for percent change in blood pressure from baseline to 5 years. The same analysis was carried out separately in a subgroup treated with antihypertensive drugs. The main comparisons were made among subjects treated with diet alone, subjects treated with statin drugs, and subjects treated with non-statin drugs.
Statistical analysis Results are expressed as means ⫾ SD. Statistical analysis was carried out with an SPSS statistical package (version 9.1 for Windows). Two-way analysis of variance was used to compare the normal distribution of baseline variables among all treatment strategies. The 2 test, with Yates correction when appropriate, was used to compare categorical variables at baseline. Two-way analysis of variance was also used to compare the changes in systolic and diastolic blood pressure observed in response to different treatment strategies in the overall population of subjects, as well as in the different quartiles of systolic blood pressure.
Results A total of 211 subjects were considered to be poor responders to single-drug therapy; they were treated with a combination of lipid-lowering drugs and excluded from the study. The final study sample consisted of 1356 subjects. Of these, 542 subjects were treated with a lipid-lowering diet, 550 were treated with non-statin drugs, and 264 were treated with statin drugs. All the patients were maintained as allocated to the same baseline quartile, thereby justifying some differences in the size of quartiles analyzed in this study. The daily dose range for gemfibrozil was 400 to 800 mg; for cholestyramine, the daily dose range was 4 to 24 g; and for simvastatin, the daily dose range was 20 to 40 mg. The baseline characteristics of subjects according to treatment group are summarized in Table I. The 3 treatment groups were comparable in age, sex distribution, body mass index, and baseline blood pressure values. The proportion of patients treated with antihypertensive drugs and the distribution of the different classes of drugs was comparable in the 3 treatment groups (data not shown). Mean serum cholesterol levels were significantly higher in patients taking lipidlowering drugs. Table II summarizes the baseline characteristics of subjects according to quartile of systolic blood pressure. Table III shows baseline data of subjects in subgroups within each quartile according to treatment group. Subjects in the upper quartiles were older and approximately 50% were women. No differences in total cholesterol or low-density lipoprotein cholesterol levels were observed. Systolic and diastolic blood pressures increased progressively from quartiles
American Heart Journal Volume 148, Number 2
1 to 4, with no differences observed among subgroups treated with different lipid-lowering strategies (Table III). Conversely, low-density lipoprotein levels were largely comparable among quartiles, but were significantly higher in patients randomized to receive lipidlowering drugs. Figure 1 depicts the percentage change in blood pressure observed in response to lipid-lowering therapy in the study population by quartile of systolic blood pressure. A significant decrease in systolic and diastolic blood pressure was observed in quartiles 3 and 4; the change in blood pressure observed in quartiles 1 and 2 (systolic blood pressure ⬍139 mm Hg) did not reach statistical significance. Among subjects in quartiles 3 and 4, the reduction in either systolic or diastolic blood pressure was greater in subjects treated with lipid-lowering drugs; these subjects also showed a greater reduction in levels of low-density lipoprotein cholesterol after 5 years of follow-up (Table IV). The effects on systolic and diastolic blood pressure of non-statin drugs were compared with those of statin drugs (Figure 2). Subjects treated with statin drugs had a significantly greater decrease in systolic and diastolic blood pressure than subjects treated with fibrates or cholestyramine, despite comparable reductions in levels of low-density lipoprotein cholesterol (Table IV). The difference among treatments occurred most frequently in the upper 2 quartiles of systolic blood pressure and was most apparent in subjects whose systolic blood pressure levels were ⬎154 mm Hg. The average body mass index values were slightly reduced in the whole study population, with no differences in patients treated with statin or non-statin strategies. To further investigate the putative role of the interactions among lipid-lowering interventions, blood pressure control, and hypertension, we analyzed separately the subset of 533 subjects treated with antihypertensive drugs. Most of these subjects were treated with diuretics (32.8%), angiotensin-converting enzyme inhibitors (28.8%), calcium channel blockers (16.5%), or a combination of these agents. The effects of lipid-lowering drugs on blood pressure control in such a population of patients were restricted to the upper 2 quartiles (Figure 3). In particular, the average reduction in blood pressure was enhanced in subjects in the fourth quartile, whose blood pressure control was poor despite receiving antihypertensive drug therapy. Conversely, no significant changes in blood pressure were observed in subjects in the lower 2 quartiles, with normal baseline blood pressure values. Subjects treated with statin drugs had a greater decrease in systolic and diastolic blood pressure, irrespective of baseline blood pressure values, which were largely comparable among the therapeutic subgroups (Table V).
Borghi et al 287
Table I. Baseline characteristics of the study participants according to treatment group
Variable
LipidLipidLipidlowering lowering lowering therapy therapy diet (nonstatin) (statin)
Patients (n) 542 550 264 Mean age (y) 53.8 ⫾ 10 56.6 ⫾ 13 56.3 ⫾ 9 Sex (%) (men/women) 57/43 58/42 56/44 24.7 ⫾ 3 25.2 ⫾ 3 24.8 ⫾ 3 BMI (kg/m2) Mean SBP (mm Hg) 140.9 ⫾ 15 141.9 ⫾ 17 139.6 ⫾ 14 Mean DBP (mm Hg) 87.2 ⫾ 9 87.5 ⫾ 10 86.5 ⫾ 9 Mean total cholesterol 264.3 ⫾ 44 283.1 ⫾ 48* 287.5 ⫾ 45* (mg/dL) Mean LDL cholesterol 163.7 ⫾ 22 197.2 ⫾ 37* 198.5 ⫾ 33* (mg/dL) Patients with antihypertensive 201 (37.8) 299 (54.3) 133 (50.3) drugs, n (%) Diuretics (%) 33.9 32.1 32.4 ACE inhibitors (%) 27.3 29.1 30.0 CCBs (%) 17.8 16.3 15.4 -Blockers (%) 11.7 10.9 11.4 Combination (%) 23.1 24.0 23.8 SBP, Systolic blood pressure; DBP, diastolic blood pressure; LDL, low-density lipoprotein. *P ⬍ .001 versus diet.
Discussion This analysis of participants in the Brisighella Heart Study has provided a number of stimulating findings that may have relevance in the clinical treatment of patients who are at high risk for cardiovascular disease. The study supports the hypothesis that the longterm use of lipid-lowering measures significantly improves blood pressure control in patients with both hypercholesterolemia and hypertension. The reduction in blood pressure was particularly evident in subjects treated with statin drugs. Furthermore, the effect of lipid-lowering drugs was restricted to subjects with higher baseline blood pressure values, poorer responses to antihypertensive treatment, or both. High blood pressure and hypercholesterolemia frequently occur concomitantly in the same patients and may account for a large proportion of the cumulative risk of cardiovascular disease. A significant increase in cholesterol levels has been observed in as much as 40% of middle-aged (35– 65 years old) patients with hypertension in the United States, whereas the prevalence of hypercholesterolemia has a linear relationship with the severity of hypertension.11 A recent review of the data collected by the National Health and Nutrition Examination (NHANES) III survey has provided support for this observation by demonstrating that the proportion of patients with hypercholesterolemia increases progressively with the degree of severity of hypertension.12 When hypertension and hypercholes-
American Heart Journal August 2004
288 Borghi et al
Table II. Baseline characteristics of the study participants according to quartile of systolic blood pressure Quartile 1
2
3
4
P value for trend
333 55.2 ⫾ 10 65/35 118.7 ⫾ 6 78.8 ⫾ 6 271.3 ⫾ 44 188.7 ⫾ 21
313 59.9 ⫾ 7 57/43 131.9 ⫾ 3 84.0 ⫾ 6 273.1 ⫾ 48 186.2 ⫾ 26
370 61.2 ⫾ 8 55/45 145.3 ⫾ 6 88.9 ⫾ 7 267.5 ⫾ 45 184.7 ⫾ 28
340 65.5 ⫾ 8 51/49 167.4 ⫾ 6 96.7 ⫾ 9 269.8 ⫾ 41 186.7 ⫾ 23
⬍.0001 ⬍.001 ⬍.00001 ⬍.0001 ns ns
Variable Patients (n) Mean age (y) Sex (%) (men/women) Mean SBP (mm Hg) Mean DBP (mm Hg) Mean total cholesterol (mg/dL) LDL cholesterol (mg/dL) ns, Not significant.
Table III. Baseline characteristics in the 4 quartiles of systolic blood pressure according to treatment group
Quartile 1 (n ⫽ 333)
2 (n ⫽ 313)
3 (n ⫽ 370)
4 (n ⫽ 340)
Variable
LipidLowering Diet
LipidLowering Therapy (Non-statin)
LipidLowering Therapy (Statin)
SBP DBP LDL-C SBP DBP LDL-C SBP DBP LDL-C SBP DBP LDL-C
119.3 ⫾ 6 79.3 ⫾ 7 169.0 ⫾ 21 132.6 ⫾ 3† 84.7 ⫾ 6† 163.0 ⫾ 26 145.3 ⫾ 4‡§ 89.5 ⫾ 7‡§ 161.5 ⫾ 32 166.8 ⫾ 12㛳¶# 95.6 ⫾ 11㛳¶# 162.3 ⫾ 29
118.9 ⫾ 6 78.6 ⫾ 7 202.0 ⫾ 33* 132.8 ⫾ 3† 85.4 ⫾ 6† 193.2 ⫾ 33* 145.2 ⫾ 4‡§ 88.8 ⫾ 7‡§ 198.3 ⫾ 36* 170.7 ⫾ 11㛳¶# 97.3 ⫾ 16㛳¶ 197.4 ⫾ 32*
118.0 ⫾ 5 78.5 ⫾ 6 195.3 ⫾ 27* 130.3 ⫾ 1† 82.0 ⫾ 8† 203.1 ⫾ 44* 145.6 ⫾ 6‡§ 88.3 ⫾ 8‡§ 195.2 ⫾ 39* 164.7 ⫾ 5㛳¶# 97.2 ⫾ 6㛳¶# 201.2 ⫾ 41*
LDL-C, Low-density lipoprotein cholesterol. *P ⬍ .001 versus diet. †P ⬍ .005 versus quartile 1. ‡P ⬍ .001 versus quartiles 1 and 2. §P ⬍ .005 versus quartiles 1 and 2. 㛳P ⬍ .0001 versus quartiles 1, 2, and 3. ¶P ⬍ .001 versus quartiles 1, 2, and 3. #P ⬍ .005 versus quartiles 1, 2, and 3.
terolemia occur in the same patient, they amplify the absolute level of risk for cardiovascular complications significantly. The results of the Multiple Risk Factor Intervention Trial demonstrated that, at any blood pressure level, mortality caused by coronary artery disease is greater in patients with higher serum cholesterol levels.3 This suggests the importance of an extensive and systematic evaluation of serum cholesterol levels in patients with hypertension and in patients with only a marginal change in blood pressure (eg, subjects with high-normal blood pressure). The possible effects of interactions between serum cholesterol and blood pressure are not limited to their contribution to the pathogenesis of atherosclerosis and its complications. Hypercholesterolemia might also be implicated, directly or indirectly, with some of the
mechanisms involved in the development of hypertension. A beneficial effect of various lipid-lowering measures on blood pressure control has been described in several retrospective clinical trials carried out in ⬎17,000 patients with hypercholesterolemia.13 In this extensive review by Goode et al, a positive linear relationship between the absolute reduction in cholesterol level and the extent of blood pressure reduction was reported in a broad range of patients with hypercholesterolemia who were receiving cholesterol-lowering treatment.13 Beneficial effects of lipid-lowering treatment has been shown in studies of patients with borderline hypertension14,15 and in selected subsets of patients with untreated hypertension, diabetes mellitus, or both who were receiving statin treatment.16 –18 A primary prevention study involving a large popula-
American Heart Journal Volume 148, Number 2
Borghi et al 289
Figure 1
Figure 2
Percent change in systolic and diastolic blood pressure after 5 years of treatment with lipid-lowering diet or lipid-lowering drugs in quartiles of baseline systolic blood pressure. *P ⬍.001; †P ⬍.005; ‡P ⬍.0001 vs quartile 1. §P ⬍.005; 㛳P ⬍.001 vs diet.
Percent change in systolic and diastolic blood pressure after 5 years of treatment with statins and non-statin lipid-lowering drugs in quartiles of baseline systolic blood pressure. *P ⬍.001; †P ⬍.005; ‡P ⬍.0001 vs quartile 1; §P ⬍.005 vs non-statin drugs.
Table IV. Mean change in LDL-cholesterol levels according to quartile of systolic blood pressure in each treatment group
Quartile 1 2 3 4
Lipidlowering diet
Lipidlowering therapy (nonstatin)
⫺3.5% ⫾ 6% ⫺3.1% ⫾ 8% ⫺3.7% ⫾ 6% ⫺3.1% ⫾ 11%
⫺13.9% ⫾ 9%* ⫺12.5% ⫾ 8%* ⫺16.1% ⫾ 8%* ⫺16.2% ⫾ 8%*
Lipidlowering therapy (statin)
Figure 3
⫺15.8% ⫾ 8%* ⫺12.8% ⫾ 8%* ⫺16.9% ⫾ 10%* ⫺18.4% ⫾ 10%*
*P ⬍ .001 versus diet.
tion of patients with hypercholesterolemia revealed a 25% reduction in the incidence of new cases of hypertension in men whose serum cholesterol levels were reduced with clofibrate.19 Glorioso et al20 prospectively investigated the effects of pravastatin on the blood pressure profile in patients with hypercholesterolemia and untreated hypertension and found that pravastatin decreased resting systolic, diastolic, and pulse pressures and blunted the pressor response to the cold pressor test. This effect was independent of age, sex, and baseline serum levels of lowdensity lipoprotein and high-density lipoprotein cholesterol, which suggests that statin drugs might interfere with some of the basic mechanisms responsible for blood pressure control. More recently, a blood pressure-lowering effect was demonstrated for atorvastatin in patients with isolated systolic hypertension.21 In such patients, systolic and diastolic blood pressure reduction has been associated with an increase in the peripheral vasodilatory capacity that largely supports the capacity of statin drugs to reduce arterial stiffness and improve endothelium-dependent vasodilatation.
Percent change in systolic and diastolic blood pressure after 5 years of treatment with lipid-lowering measures in quartiles of baseline systolic blood pressure and in subjects treated with antihypertensive drugs. *P ⬍.001; †P ⬍.005 vs quartile 1; ‡P ⬍.005 vs non-statin drugs; §P ⬍.005 vs lipid-lowering diet.
The results of our study support the hypothesis that the effects of statin drugs on blood pressure control are independent of the extent of cholesterol reduction. The reduction in systolic and diastolic blood pressure associated with the use of statin drugs was significantly greater than that observed with non-statin medications, despite a comparable cholesterol level reduction. These results are broadly consistent with those of a study directly comparing simvastatin and cholestyramine in patients with hypertension and type 2 diabetes mellitus,22 which demonstrated that the 2 drugs
American Heart Journal August 2004
290 Borghi et al
Table V. Baseline characteristics of patients treated with antihypertensive drugs and included in the 4 quartiles of systolic blood pressure allocated to different lipid-lowering strategies
Quartile
Variable
1 (n ⫽ 17) 2 (n ⫽ 106) 3 (n ⫽ 162) 4 (n ⫽ 348)
SBP DBP SBP DBP SBP DBP SBP DBP
Lipidlowering diet
Lipidlowering therapy (nonstatin)
Lipidlowering therapy (statin)
121.0 ⫾ 4 79.6 ⫾ 9 133.9 ⫾ 2* 86.0 ⫾ 5* 147.1 ⫾ 4†‡ 89.1 ⫾ 7†‡ 170.9 ⫾ 14§㛳¶ 97.3 ⫾ 10§㛳¶
123.6 ⫾ 6 79.6 ⫾ 5 133.5 ⫾ 3* 89.3 ⫾ 8* 146.4 ⫾ 4†‡ 85.8 ⫾ 11†‡ 170.1 ⫾ 14§㛳¶ 96.5 ⫾ 13§㛳¶
124.2 ⫾ 4 79.3 ⫾ 6 130.0 ⫾ 1* 87.1 ⫾ 8* 151.2 ⫾ 1†‡ 85.7 ⫾ 5†‡ 169.3 ⫾ 5§㛳¶ 96.0 ⫾ 7§㛳¶
*P ⬍ .001 versus diet. †P ⬍ .005 versus quartile 1. ‡P ⬍ .001 versus quartiles 1 and 2. §P ⬍ .005 versus quartiles 1 and 2. 㛳P ⬍ .0001 versus quartiles 1, 2, and 3. ¶P ⬍ .001 versus quartiles 1, 2, and 3. #P ⬍ .005 versus quartiles 1, 2, and 3.
were equally effective in reducing total and low-density lipoprotein cholesterol levels, but only statin therapy led to a significant reduction in diastolic blood pressure. Terzoli et al23 reported that the blood pressure-lowering effect of statin drugs is more pronounced in patients who had a reduction in cholesterol level ⬎50 mg/dL (1.3 mmol/L), which suggests that a function based on a cutoff point can better explain the onset of the blood pressure lowering effect associated with statin treatment. However, most of the studies emphasizing the importance of cholesterol reduction in blood pressure control have been conducted by comparing the effect of lipid-lowering drugs with placebo, thereby preventing any possibility of distinguishing the primary effects of the drug from the consequences of serum cholesterol level reduction. For all these reasons, the issue of the relative contribution of serum cholesterol level reduction to the blood pressure-lowering effect of statin drugs is still a matter of debate and deserves further exploration with the results of ongoing clinical trials (eg, the Antihypertensive and Lipid Lowering Treatment to Prevent Heart Attack Trial). One of the most interesting observations arising from this study is that lipid-lowering drugs improve blood pressure control only in patients whose blood pressure is in the hypertensive range. This selective effect of statin drugs was demonstrated in the overall study population and in the subgroups of patients who were treated with antihypertensive drugs. These findings are in agreement with most of the data reported in the literature demonstrating the lack of any blood pressure-lowering effect for statin drugs in subjects
who are normotensive24 –29 or in patients with hypertension treated with adequate blood pressure control.30 –33 These studies suggest that the effects of statin drugs on blood pressure are probably restricted to patients with concomitant hypercholesterolemia and uncontrolled hypertension in whom modulation of peripheral vascular tone and vascular resistance is significantly impaired. Taken together, these data clearly support the possibility that the extensive use of lipid-lowering drugs, particularly the use of statin drugs, could significantly improve blood pressure control (and its clinical outcome) in patients with concomitant hypertension and hypercholesterolemia. Despite such encouraging conclusions, this study has some important methodological limitations. First, the Brisighella Heart Study was not specifically designed to test the hypothesis of any effect of cholesterol-lowering drugs on blood pressure control in patients with hypercholesterolemia. Second, within the group of patients treated with non-statin cholesterol-lowering drugs, the administration of fibrates and cholestyramine was not randomized. Third, in accordance with the observational nature of the Brisighella Heart Study, which was designed to assess the role of lipid abnormalities on the long-term incidence of cardiovascular disease, the demographic description of this population is lacking some important information (eg, prevalence of family history of hypertension, renal function, concomitant diseases, and concomitant pharmacological treatments). Finally, in the treated hypertensive population, the use of the various antihypertensive drugs was not previously randomized. This prevented the possibility of obtaining statistically reli-
American Heart Journal Volume 148, Number 2
able information on the antihypertensive effect of the various combinations of lipid-lowering drugs and antihypertensive medications in this study.
Conclusions Whether such limitations are relevant to the interpretation of the results of this study, the evidence presented strongly suggests that the interaction between high blood pressure and hypercholesterolemia is probably not solely limited to their reciprocal roles as risk factors for atherosclerosis. In particular, the blood pressure-lowering effect of statin drugs observed in patients with hypertension and hypercholesterolemia in this study suggests that a broader preventive approach to cardiovascular disease should consider the future possibility of a “therapeutic crossover” of drugs that are designed to treat different risk factors. We thank Elizabeth Melby for the revision of the manuscript and Ettore Ambrosioni for continuous support for research.
References 1. American Heart Association. Heart and stroke statistical update. Dallas (Tex): American Heart Association; 2000. 2. Kannel WB. Blood pressure as a cardiovascular risk factor: prevention and treatment. JAMA 1996;275:1571– 6. 3. Neaton JD, Blackburn H, Jacobs D, et al. Serum cholesterol level and mortality findings for men screened in the Multiple Risk Factor Intervention Trial. Multiple Risk Factor Intervention Trial Research Group. Arch Intern Med 1992;152:1490 –500. 4. Alderman MH, Cohen H, Madhavan S. Diabetes and cardiovascular events in hypertensive patients. Hypertension 1999;33:1130 – 4. 5. Wilson PWF, D’Agostino RB, Levy D, et al. Prediction of coronary heart disease using risk factor categories. Circulation 1998;97: 1837– 47. 6. The JNC-VI Sub-Committee. The sixth report of the Joint National Committee on prevention, detection, evaluation, and treatment of high blood pressure. Arch Intern Med 1997;157:2413– 46. 7. Chalmers J, MacMahon S, Mancia G, et al. 1999 World Health Organization–International Society of Hypertension Guidelines for the management of hypertension. Guidelines subcommittee of the World Health Organization. Clin Exp Hypertens 1999;21:1009 – 60. 8. Executive 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). JAMA 2001;285:2486 –97. 9. Descovich GC. The Brisighella Heart Study: an interim report. Eur Heart J 1990;11(Suppl H):32–7. 10. Borghi C, Dormi A, Ambrosioni E, et al. Relative role of systolic, diastolic and pulse pressure as risk factors for cardiovascular events in the Brisighella Heart Study. J Hypertens 2002;20:1737– 42. 11. Laurenzi M, Cirillo M, Trevisan M. The Gubbio data: epidemiology and pathophysiology. Gubbio Study Research Group. Clin Exp Hypertens 1992;14:261–9.
Borghi et al 291
12. Muntner P, He J, Roccella EJ, et al. The impact of JNC-VI guidelines on treatment recommendations in the US population. Hypertension 2002;39:897–902. 13. Goode GK, Miller JP, Heagerty AM. Hyperlipidaemia, hypertension, and coronary heart disease. Lancet 1995;345:362– 4. 14. Julius S, Jamerson K, Mejia A, et al. The association of borderline hypertension with target organ changes and higher coronary risk: Tecumseh Blood Pressure study. JAMA 1990;264:354 – 8. 15. Borghi C, Bacchelli S, Degli Esposti D, et al. Pressor and metabolic correlates to long-term development of stable hypertension in borderline hypertensive patients [abstract]. J Hypertens 1997;15: S111. 16. Borghi C, Veronesi M, Prandin MG, et al. Statins and blood pressure regulation. Curr Hypertens Rep 2001;3:281– 8. 17. Borghi C, Dormi A, Veronesi M, et al. Use of lipid-lowering drugs and blood pressure control in patients with arterial hypertension. J Clin Hypertens 2002;4:277– 85. 18. Borghi C. Interactions between hypercholesterolemia and hypertension: implications for therapy. Curr Opin Nephrol Hypertens 2002;11:489 –96. 19. Committee of Principal Investigators. A co-operative trial in the primary prevention of ischemic heart disease using clofibrate. Br Heart J 1978;40:1069 –118. 20. Glorioso N, Troffa C, Filigheddu F, et al. Effect of the HMG-CoA reductase inhibitors on blood pressure in patients with essential hypertension and primary hypercholesterolemia. Hypertension 1999;34:1281– 6. 21. Ferrier KE, Muhlmann MH, Baguet JP, et al. Intensive cholesterol reduction lowers blood pressure and large artery stiffness in isolated systolic hypertension. J Am Coll Cardiol 2002;39:1020 –5. 22. Tonolo G, Melis MG, Formato M, et al. Additive effects of Simvastatin beyond its effects on LDL cholesterol in hypertensive type 2 diabetic patients. Eur J Clin Invest 2000;30:980 –7. 23. Terzoli L, Raco R, Mircoli L, et al. Lowering of ambulatory and daytime (but not clinical and nighttime) blood pressure by statin treatment [abstract]. J Hypertens 2002;20(Suppl 4):S4. 24. Muramatsu J, Kobayashi A, Hasegawa N, et al. Hemodynamic changes associated with reduction in total cholesterol by treatment with the HMG-CoA reductase inhibitor pravastatin. Atherosclerosis 1997;130:179 – 82. 25. Antonicelli R, Onorato G, Pagelli P, et al. Simvastatin in the treatment of hypercholesterolemia in elderly patients. Clin Ther 1990; 12:165–71. 26. Kool M, Lustermans F, Kragten H, et al. Does lowering of cholesterol levels influence functional properties of large arteries? Eur J Clin Pharmacol 1995;48:217–23. 27. Sartor G, Katzman P, Eizyk E, et al. Simvastatin treatment of hypercholesterolemia in patients with insulin dependent diabetes mellitus. Int J Clin Pharmacol Ther 1995;33:3– 6. 28. Sung BH, Izzo JL, Wilson MF. Effects of cholesterol reduction on blood pressure response to mental stress in patients with high cholesterol. Am J Hypertens 1997;10:592–9. 29. van Etten RW, de Koning EJ, Honing ML, et al. Intensive lipid lowering by statin therapy does not improve vasoreactivity in patients with type 2 diabetes. Arterioscler Thromb Vasc Biol 2002;22: 799 – 804. 30. D’Agostino RB, Kannel WB, Stepanians MN, et al. Efficacy and tolerability of lovastatin in hypercholesterolemia in patients with systemic hypertension. Am J Cardiol 1993;71:82–7.
American Heart Journal August 2004
292 Borghi et al
31. O’Callaghan CJ, Krum H, Conway EL, et al. Short term effects of pravastatin on blood pressure in hypercholesterolaemic hypertensive patients. Blood Press 1994;3:404 – 6. 32. Arteaga J, Sorbet M, Anda E, et al. Atorvastatin associated to valsartan does not improve blood pressure control or proteinuria in patients with diabetic nephropathy [abstract]. Am J Hypertens 2002;15(4 part 2):57A. 33. Mancia G, Crepaldi G, Gallus G, et al. Long-term statin administration and ambulatory blood pressure in mild hypertensive and hypercholesterolemic patients [abstract]. The PHYLLIS study. J Hypertens 2002;20:S4.
Appendix Brisighella Heart Study Working Party
Professor Gian Carlo Descovich, Brisighella Study founder; Professor A. Gaddi, MD, PhD, scientific coordinator; E. Pelliconi, Brisighella coordinator, past mayor; C. Sangiorgi, current mayor. Bologna Medical Staff: C. Borghi, MD; S. D’Addato, MD, PhD; A. Fiorito; C. Galetti, MD; M.C. Grippo, MD; V. Immordino, MD; M. L. Malkowski, MD; C. Mussoni, MD, PhD, Brisighella team project coordinator; A. Moretti, MD; S. Nascetti, MD; A.R. Reggiani, MD; S. Rimondi, MD. Data handling and biometrics: A. Dormi, MathD. Laboratory staff: Z. Sangiorgi, BD; R. Mambelli. Brisighella general practitioner staff: L. Bagnaresi, MD; E. Belletti, MD; Gamberi, MD; A. Naldi, MD; C. Samore`, MD; P. Viozzi, MD. Anagraphic staff: M. Nanni. Brisighella Heart Price Committee: E. Descovich; A. Gaddi; C. Sangiorgi.
The following article is an AHJ Online Exclusive. Full text of this article is available at no charge at our Website: www.mosby.com/ahj.
Is angiotensin-converting enzyme inhibitor useful in a Japanese population for secondary prevention after acute myocardial infarction? A final report of the Japanese Acute Myocardial Infarction Prospective (JAMP) study Kenji Ueshima, MD,a Kennichi Fukami, MD,a Katsuhiko Hiramori, MD,a Saichi Hosoda, MD,b Hiroshi Kishida, MD,c Kazuzo Kato, MD,d Tsuyoshi Fujita, MD,e Kiichiro Tsutani, MD,f and Akira Sakuma, PhD,g for the Japanese Angina and Myocardial Infarction Prospective (JAMP) study group* Morioka, Tokyo, and Osaka, Japan
Background
Although angiotensin-converting enzyme (ACE) inhibi-
tors have appeared to be useful for secondary prevention after acute myocardial infarction (AMI) in Western countries, that has not been confirmed in non-western countries. We investigated whether ACE inhibitors improve survival rates in patients who have survived an AMI in Japan.
Methods
A randomized controlled trial, the first non-pharmaceutical
company-supported multicenter trial of a medication in Japan, was carried out in 48 institutions from 1993 to 2000. A total of 888 of 1163 patients with AMI were eligible for the full analysis set (FAS). The mean patient
tion because of worsening angina or congestive heart failure. The mean follow-up period was 5.8 years.
Results
There were no significant differences in the 2 groups in base-
line data. During the follow-up period, 3 patients were lost to follow-up. With Kaplan-Meier analysis, the annual rate of total cardiac events was 32% in both groups. After adjustment for clinical baseline data, ACE inhibitor administration was not revealed with Cox regression analysis to have a significant prognostic effect in our study.
age was 62 years, and 78% of patients were men. Subjects were ran-
Conclusion
domized to 2 groups; 422 received ACE inhibitors and 466 did not re-
with ACE inhibitor administration in subjects who survived after AMI in a
ceive ACE inhibitors. The primary end point was combined cardiac
Japanese study population. Further evaluations with a larger population or
events, which was defined as cardiac or non-cardiac death, recurrent non-
in subjects who are at a higher risk for AMI are necessary to confirm our
fatal myocardial infarction, coronary revascularization, and hospitaliza-
findings. (Am Heart J 2004;148:e8.)
We did not show a significant improvement in outcome