Dyslipidemia and the Risk of Incident Hypertension in Men
Abstract
Evidence suggests that hypertension may share a similar pathophysiology with cardiovascular disease (CVD). Thus, dyslipidemia, a strong predictor of CVD, may also predict incident hypertension. We analyzed 3110 men free of hypertension, CVD, and cancer from the Physicians’ Health Study, who provided baseline blood samples from which we measured total cholesterol (TC) and HDL cholesterol (HDL-C), and calculated non-HDL-C and the TC/HDL-C ratio. We categorized each lipid parameter into quintiles and considered National Cholesterol Education Project clinical cut points. Other risk factor information was provided from self-reports on the baseline questionnaire. Incident hypertension was defined as either the initiation of antihypertensive treatment, self-reported systolic blood pressure ≥140 mm Hg, or diastolic blood pressure ≥90 mm Hg. Over a mean follow-up of 14.1 years, 1019 men developed hypertension. In Cox proportional hazards models adjusted for lifestyle and clinical risk factors, men in the highest quintile of TC, non-HDL-C, and TC/HDL-C ratio had increased risks of developing hypertension of 23%, 39%, and 54%, respectively, compared with participants in the lowest quintile. Furthermore, men in the highest quintile of HDL-C had a 32% decreased risk of developing hypertension compared with those in the lowest quintile. Models using National Cholesterol Education Project cut points demonstrated similar associations with hypertension. Models excluding men with diabetes and obesity maintained an independent association between baseline lipids and hypertension. These prospective cohort data suggest that dyslipidemias may lead to the subsequent development of hypertension. Thus, plasma lipids may be useful in the identification of men at risk for hypertension.
Hypertension is a common condition that affects >50 million Americans1 and is an important risk factor for cardiovascular disease. Although several risk factors for the development of hypertension have been identified,2 its etiology is still not fully understood.1 Hypertension is commonly associated with other cardiovascular risk factors, such as obesity, diabetes, and dyslipidemia.3 The presence of these cardiovascular risk factors and the resulting endothelial dysfunction may play a role in the pathophysiology of hypertension.4
Dyslipidemia, a strong predictor of cardiovascular disease,5 causes endothelial damage,6–8 and the loss of physiological vasomotor activity that results from endothelial damage may become manifested as increased blood pressure (BP). Therefore, factors like dyslipidemia that cause endothelial dysfunction may lead to hypertension. Cross-sectional studies have suggested a link between abnormal lipids and hypertension.4,9,10 A few studies have prospectively examined the relationship between plasma lipids and the future development of hypertension, finding that there is an association between plasma lipids and development of hypertension.9,10 Small trials have looked at the effect of lipid lowering on BP.11,12 Additional prospective data on the association of lipids with hypertension will help us clarify the relationship. We, therefore, prospectively examined whether total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C), and non-high density lipoprotein cholesterol (non-HDL-C) are associated with the risk of developing hypertension in men initially free of hypertension.
Methods
Physicians’ Health Study
The Physicians’ Health Study (PHS) was a randomized, double-blind, placebo-controlled trial of aspirin and β-carotene in the primary prevention of cardiovascular disease and cancer. The study consisted of 22 071 male physicians between the ages of 40 and 84 who were free from prior myocardial infarction (MI), stroke, and cancer (except nonmelanoma skin cancer).13 All of the participants provided written informed consent, and the Institutional Review Board at Brigham and Women’s Hospital approved the study protocol.
Blood Collection, Lipid Analysis, and Baseline Covariates
Before randomization into the PHS, all 22 071 participants were asked to provide nonfasting baseline blood samples, which were collected in EDTA and processed for long-term storage at −80°C. Specimens were received from 14 916 (68%) of the randomized physicians. We used previous measurements of TC and HDL-C from a study of 4483 healthy subjects for a prior study of renal function.14 Baseline plasma samples had been thawed, and TC and HDL-C levels were measured on a Hitachi 911 analyzer (Roche Diagnostics) with reagents from Roche Diagnostics and Genzyme. At baseline, participants also completed a questionnaire with self-reported demographic, medical, and lifestyle information. We collected potential baseline risk factors that may be confounders, such as age (years), body mass index (kg/m2), smoking status (never, past, or current), alcohol consumption (rarely/never, monthly, weekly, daily), exercise to sweat at least once per week (yes or no), diabetes (yes or no), and parental history of MI <60 years (yes or no).
Among the 4483 men with available TC and HDL-C, we excluded subjects who reported a history of hypertension, systolic BP (SBP) of ≥140 mm Hg, or diastolic BP (DBP) of ≥90 mm Hg, as well as those having a history of angina, coronary artery revascularization, or transient ischemic attack. In addition, we excluded the 23 men who had either current or past treatment for high cholesterol. After these exclusions, 3110 subjects remained for analysis.
Hypertension Ascertainment
Subjects completed follow-up questionnaires at 6 months and then annually thereafter. Mean follow-up was 14.1 years, and maximum follow-up was 18.6 years. The primary outcome for these analyses is time to hypertension during follow-up, based on the definition in the Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC7).1 For a subject to develop hypertension during follow-up, a subject must have initiated treatment for hypertension, had a SBP ≥140 mm Hg, or had a DBP ≥90 mm Hg. The diagnosis of incident hypertension was based on self-reported treatment and BP provided by subjects. Self-reported BP and/or hypertension has been validated previously in physicians, female health professionals,15 and the general population.16 Morbidity and mortality follow-up in the PHS has exceeded 99% over time.17
Data Analyses
From measurements of TC and HDL-C, we calculated the TC/HDL-C ratio and non-HDL-C (defined as TC minus HDL-C). Each lipid parameter (TC, HDL-C, non-HDL-C, and TC/HDL-C ratio) was then categorized in quintiles, and baseline characteristics were compared by quintile of TC. We then fitted age- and multivariate-adjusted Cox proportional hazards models to determine the relative risks (RRs) and 95% CIs of incident hypertension by quintile of lipid parameter, using the first quintile as the reference group. The multivariate model was adjusted for baseline age, body mass index, diabetes, vigorous exercise, smoking status, alcohol use, and parental history of MI. We also considered multivariate models that additionally controlled for baseline SBP and DBP to consider the impact of lipids on the risk of hypertension through mechanisms other than BP. A linear trend across quintiles for each lipid parameter was tested with an ordinal variable for the median level within each quintile. The proportional hazards assumption was tested for the measured variables, TC and HDL-C, using time-varying covariates and was found to be valid (P>0.05).
We also considered National Cholesterol Education Project (NCEP) cut points for additional Cox proportional hazards models that used clinical values for each lipid parameter.18 To examine the role of baseline BP on the association between lipids and hypertension, we modeled the risk of incident hypertension for NCEP cut points of lipids, stratified by baseline SBP (<120 versus 120 to 139 mm Hg) and DBP (<80 versus 80 to 89 mm Hg) according to the JNC7 definition of prehypertension.1 Lastly, we examined RRs when we excluded 138 men with baseline diabetes or a body mass index ≥30 kg/m2.
Results
The cohort was followed for up to 18.6 years, during which 1019 men developed hypertension. The mean (SD) ages at baseline of the men remaining normotensive and developing hypertension were 47.9 (6.4) and 50.0 (7.0) years, respectively (P<0.0001). The mean TC levels at baseline for the men remaining normotensive and developing hypertension were 210.5 (39.6) and 217.7 (40.4) mg/dL, respectively (P<0.0001). At baseline there was no significant difference in the percentage of diabetes across quintiles, although there were very few diabetics at baseline in this cohort. At baseline, the subjects in the first quintile of TC were younger, leaner, less likely to smoke, and more likely to exercise at least once per week (Table 1). The subjects with TC in the first quintile tended to consume less alcohol; they were less likely to drink daily (16.7% compared with 24% for the fifth quintile) and more likely to be nondrinkers (18.5% compared with 10.1% for the fifth quintile). In Table 2, compared with men in the lowest TC quintile, there was an age-adjusted 33% increased risk of developing incident hypertension in the highest TC quintile (P, trend=0.0003). Adjusting for all of the other variables modestly attenuated the association of TC with incident hypertension, with men in the highest quintile having a 23% increased risk of incident hypertension compared with those in the lowest quartile (P, trend=0.0067). For HDL-C, there was an inverse association with the development of hypertension in both age- and multivariate-adjusted models. In the multivariate-adjusted model, compared with the lowest quintile of HDL-C, men in the highest quintile of HDL-C had a 32% decreased risk of developing hypertension. For non-HDL-C, those in the highest quintile had a 55% increased risk of hypertension compared with those in the lowest quintile in the age-adjusted model (P, trend <0.0001); this RR was attenuated to a significant 39% increased risk in the multivariate-adjusted model (P, trend=0.0001).TABLE 1. Baseline Lipid Parameters and Characteristics Among 3110 Men According to Quintiles of Total Cholesterol
TABLE 2. RR and 95% CI of Hypertension by Quintiles of Lipids in 3110 Men
Baseline Characteristics | Quintile of Each Lipid Parameter (mg/dL) | ||||
---|---|---|---|---|---|
1st | 2nd | 3rd | 4th | 5th | |
*Range (median). | |||||
†Mean±SD. | |||||
Total cholesterol* | ≤180 (167) | >180 to 200 (192) | >200 to 218 (209) | >218 to 243 (229) | >243 (263) |
HDL cholesterol* | ≤31 (27) | >31 to 38 (36) | >38 to 44 (41) | >44 to 53 (49) | >53 (61) |
Non-HDL cholesterol* | ≤135 (121) | >135 to 156 (147) | >156 to 176 (167) | >176 to 201 (188) | >201 (223) |
TC/HDL cholesterol ratio* | ≤3.76 (3.28) | >3.76 to 4.57 (4.19) | >4.57 to 5.49 (5.00) | >5.49 to 6.79 (6.05) | >6.79 (7.91) |
Based Upon Quintiles of Total Cholesterol | |||||
Age (y)† | 46.8±5.9 | 48.1±6.5 | 48.8±7.0 | 48.8±6.3 | 50.3±7.2 |
Body mass index (kg/m2)† | 24.0±2.7 | 24.4±2.7 | 24.5±2.6 | 24.5±2.8 | 24.6±2.7 |
Current smokers (%) | 5.1 | 6.1 | 4.7 | 7.1 | 5.3 |
History of diabetes mellitus (%) | 0.93 | 0.64 | 0.83 | 0.81 | 0.96 |
Parental history of MI <60 years (%) | 9.4 | 8.9 | 10.0 | 13.3 | 11.0 |
Exercise to sweat ≥1 time/week (%) | 79.7 | 76.3 | 76.1 | 76.5 | 76.2 |
Alcohol consumption (%) | |||||
≥1 drink/day | 16.7 | 16.9 | 20.8 | 21.1 | 24.1 |
1 to 6 drinks/week | 51.1 | 55.5 | 55.5 | 55.8 | 54.2 |
1 to 3 drinks/month | 13.7 | 13.8 | 9.2 | 11.9 | 11.6 |
Rarely/never | 18.5 | 13.8 | 14.6 | 11.2 | 10.1 |
SBP† | 119.1±8.7 | 120.0±8.7 | 120.4±8.5 | 121.3±8.0 | 121.2±8.4 |
DBP† | 75.3±6.8 | 75.8±6.6 | 75.8±6.7 | 76.4±6.5 | 76.5±6.3 |
Lipid Parameters/Models | Quintile of Each Lipid Parameter (mg/dL) | |||||
---|---|---|---|---|---|---|
1st | 2nd | 3rd | 4th | 5th | P Trend | |
*No. of hypertension cases. | ||||||
†Adjusted for age. | ||||||
‡Adjusted for age, body mass index, exercise, smoking status, alcohol intake, parental history of myocardial infarction <60 years, and history of diabetes. | ||||||
TC (mg/dL) | ≤180 | >180 to 200 | >200 to 218 | >218 to 243 | >243 | |
No. of cases* | [177] | [184] | [190] | [230] | [238] | |
Age-adjusted RR† | 1.00 | 1.03 (0.84 to 1.27) | 1.11 (0.91 to 1.37) | 1.35 (1.10 to 1.64) | 1.33 (1.09 to 1.62) | 0.0003 |
Multivariate-adjusted RR‡ | 1.00 | 1.01 (0.82 to 1.24) | 1.07 (0.87 to 1.31) | 1.26 (1.03 to 1.54) | 1.23 (1.01 to 1.50) | 0.0067 |
HDL cholesterol (mg/dL) | ≤31 | >31 to 38 | >38 to 44 | >44 to 53 | >53 | |
No. of cases* | [237] | [213] | [202] | [189] | [178] | |
Age-adjusted RR† | 1.00 | 0.84 (0.70 to 1.02) | 0.77 (0.64 to 0.93) | 0.68 (0.56 to 0.83) | 0.65 (0.53 to 0.79) | <0.0001 |
Multivariate-adjusted RR‡ | 1.00 | 0.84 (0.69 to 1.01) | 0.80 (0.66 to 0.97) | 0.72 (0.59 to 0.87) | 0.68 (0.56 to 0.84) | 0.0002 |
Non-HDL cholesterol (mg/dL) | ≤135 | >135 to 156 | >156 to 176 | >176 to 201 | >201 | |
No. of cases* | [161] | [194] | [191] | [228] | [245] | |
Age-adjusted RR† | 1.00 | 1.12 (0.91 to 1.39) | 1.18 (0.95 to 1.45) | 1.48 (1.21 to 1.81) | 1.55 (1.27 to 1.90) | <0.0001 |
Multivariate-adjusted RR‡ | 1.00 | 1.08 (0.87 to 1.34) | 1.10 (0.89 to 1.36) | 1.37 (1.12 to 1.69) | 1.39 (1.13 to 1.70) | 0.0001 |
TC/HDL cholesterol ratio | ≤3.76 | >3.76–4.57 | >4.57–5.49 | >5.49–6.79 | >6.79 | |
No. of cases* | [165] | [168] | [227] | [204] | [255] | |
Age-adjusted RR† | 1.00 | 0.98 (0.79 to 1.22) | 1.49 (1.22 to 1.83) | 1.29 (1.05 to 1.59) | 1.69 (1.39 to 2.06) | <0.0001 |
Multivariate-adjusted RR‡ | 1.00 | 0.95 (0.77 to 1.19) | 1.42 (1.16 to 1.75) | 1.21 (0.98 to 1.50) | 1.54 (1.25 to 1.90) | <0.0001 |
We then fitted age- and multivariate-adjusted Cox proportional hazards models using quintiles of the calculated TC/HDL-C ratio. In the age-adjusted model, compared with having a TC/HDL-C ratio in the lowest quintile, those in the highest quintile had an increased risk of incident hypertension of 69% (P, trend <0.0001). This was attenuated in the multivariate-adjusted model to 54% for the highest quintile of TC/HDL-C ratio (P, trend <0.0001). We also considered adjustment for baseline SBP and DBP in our overall multivariate models from Table 2 for each plasma lipid parameter to evaluate whether plasma lipids impact the risk of hypertension through a mechanism separate from BP. Adjusting for SBP and DBP attenuated the RRs of hypertension for TC, non-HDL-C, and the TC/HDL-C ratio but did not alter the RRs for HDL-C. Even with adjustment for baseline BP, a statistically significant linear trend remained for each lipid parameter (P, trend <0.05) except TC (P, trend=0.09).
When we divided the lipid parameters by the NCEP clinical cut points (Table 3), the RRs of hypertension for the 4 lipid parameters paralleled the analyses of quintiles. In a multivariate-adjusted model, having a TC ≥240 mg/dL compared with a TC <200 was associated with a 22% increased risk of hypertension (P, trend=0.015). Compared with the reference group of men with a non-HDL-C <160 mg/dL, a non-HDL-C ≥190 was associated with a 29% increased risk of hypertension (P, trend=0.0009). An HDL-C <40 was associated with a 24% decreased risk compared with those with an HDL-C ≥60 (P, trend=0.0021), and a TC/HDL-C ratio of ≥6 was associated with a 47% increased risk compared with a TC/HDL-C ratio <4.TABLE 3. Multivariate-Adjusted* RRs and 95% CIs of Hypertension for NCEP Clinical Cut Points of Each Lipid Parameter Overall and Stratified by Baseline Blood Pressure (SBP and DBP) According to JNC VII Categories of Prehypertension
Lipid Parameters | Overall | By Baseline SBP | By Baseline DBP | ||
---|---|---|---|---|---|
<120 mm Hg | 120 to 139 mm Hg | <80 mm Hg | 80 to 89 mm Hg | ||
*Adjusted for age, body mass index, exercise, smoking status, alcohol intake, parental history of myocardial infarction <60 years, and history of diabetes. | |||||
Total cholesterol | |||||
<200 mg/dL | 1.00 (ref) | 1.00 (ref) | 1.00 (ref) | 1.00 (ref) | 1.00 (ref) |
200–239 mg/dL | 1.14 (0.99 to 1.32) | 1.06 (0.76 to 1.48) | 1.13 (0.96 to 1.33) | 0.97 (0.75 to 1.24) | 1.23 (1.03 to 1.48) |
≥240 mg/dL | 1.22 (1.04 to 1.44) | 1.27 (0.87 to 1.86) | 1.13 (0.94 to 1.36) | 1.15 (0.87 to 1.52) | 1.17 (0.96 to 1.43) |
P, trend | 0.015 | 0.24 | 0.18 | 0.37 | 0.11 |
HDL cholesterol | |||||
<40 mg/dL | 1.00 (ref) | 1.00 (ref) | 1.00 (ref) | 1.00 (ref) | 1.00 (ref) |
40–59 mg/dL | 0.81 (0.70 to 0.92) | 0.74 (0.54 to 1.02) | 0.83 (0.71 to 0.97) | 0.81 (0.64 to 1.03) | 0.81 (0.69 to 0.96) |
≥60 mg/dL | 0.76 (0.61 to 0.95) | 0.68 (0.41 to 1.14) | 0.80 (0.62 to 1.02) | 0.77 (0.53 to 1.12) | 0.77 (0.58 to 1.01) |
P, trend | 0.0021 | 0.066 | 0.018 | 0.087 | 0.015 |
Non-HDL cholesterol | |||||
<160 mg/dL | 1.00 (ref) | 1.00 (ref) | 1.00 (ref) | 1.00 (ref) | 1.00 (ref) |
160–189 mg/dL | 1.20 (1.03 to 1.40) | 0.82 (0.57 to 1.19) | 1.27 (1.07 to 1.51) | 0.82 (0.62 to 1.08) | 1.48 (1.22 to 1.78) |
≥190 mg/dL | 1.29 (1.11 to 1.50) | 1.29 (0.91 to 1.83) | 1.20 (1.01 to 1.42) | 1.27 (0.99 to 1.64) | 1.21 (1.00 to 1.46) |
P, trend | 0.0009 | 0.24 | 0.03 | 0.092 | 0.035 |
TC/HDL cholesterol ratio | |||||
<4 | 1.00 (ref) | 1.00 (ref) | 1.00 (ref) | 1.00 (ref) | 1.00 (ref) |
4–<6 | 1.20 (1.01 to 1.42) | 1.18 (0.81 to 1.72) | 1.15 (0.95 to 1.40) | 1.20 (0.90 to 1.60) | 1.15 (0.93 to 1.42) |
≥6 | 1.47 (1.23 to 1.76) | 1.55 (1.03 to 2.33) | 1.39 (1.13 to 1.70) | 1.35 (0.98 to 1.84) | 1.48 (1.19 to 1.86) |
P, trend | <0.0001 | 0.03 | 0.0009 | 0.074 | 0.0002 |
We then categorized the lipids according to NCEP cut points with stratification by baseline SBP and DBP, including the JNC7 definitions for prehypertension of 120 to 139 mm Hg SBP and 80 to 89 mm Hg DBP (Table 3). In the models stratified by SBP, the RRs were slightly stronger for those subjects with baseline SBP <120 mm Hg. For the models stratified by DBP, the RRs were essentially equal for men with normal and prehypertensive DBP. Finally, the RRs in models that excluded men who at baseline had diabetes or a body mass index ≥30 kg/m2 mirrored the results for the entire cohort (Table 4).TABLE 4. Multivariate* RRs and 95% CIs of Hypertension by Quintiles of Lipids in 2972 Men Without Diabetes and With a Body Mass Index <30 kg/m2†
Lipid Parameters | Quintile of Each Lipid Parameter (mg/dL) | |||||
---|---|---|---|---|---|---|
1st | 2nd | 3rd | 4th | 5th | P trend | |
*Adjusted for age, body mass index, exercise, smoking status, alcohol intake, parental history of myocardial infarction <60 years, and history of diabetes. | ||||||
†Excludes 138 men with baseline diabetes or a body mass index ≥30 kg/m2. | ||||||
TC (mg/dL) | ≤180 | >180 to 200 | >200 to 219 | >219 to 243 | >243 | |
Multivariate-adjusted RR† | 1.00 | 0.99 (0.80 to 1.23) | 1.06 (0.86 to 1.32) | 1.28 (1.04 to 1.57) | 1.24 (1.01 to 1.53) | 0.0041 |
HDL cholesterol (mg/dL) | ≤31 | >31 to 38 | >38 to 44 | >44 to 53 | >53 | |
Multivariate-adjusted RR† | 1.00 | 0.86 (0.71 to 1.05) | 0.81 (0.66 to 0.98) | 0.73 (0.60 to 0.90) | 0.72 (0.58 to 0.89) | 0.0012 |
Non-HDL cholesterol (mg/dL) | ≤135 | >135 to 156 | >156 to 176 | >176 to 201 | >201 | |
Multivariate-adjusted RR† | 1.00 | 1.04 (0.84 to 1.30) | 1.07 (0.86 to 1.33) | 1.36 (1.10 to 1.68) | 1.36 (1.10 to 1.68) | 0.0002 |
TC/HDL cholesterol ratio | ≤3.76 | >3.76 to 4.57 | >4.57 to 5.49 | >5.49 to 6.80 | >6.80 | |
Multivariate-adjusted RR† | 1.00 | 0.89 (0.71 to 1.12) | 1.33 (1.08 to 1.65) | 1.14 (0.92 to 1.42) | 1.46 (1.18 to 1.81) | <0.0001 |
Discussion
This prospective study demonstrates that higher levels of plasma TC, non-HDL-C, and the TC/HDL-C ratio are independently associated with a subsequent increased risk of incident hypertension in apparently healthy men and that higher levels of HDL-C are associated with a decreased risk of incident hypertension. Elevated lipid levels appear to predate the onset of hypertension by years. The relationship between lipids and hypertension is preserved even after adjustment for multiple confounders and after the exclusion of men with diabetes and obesity.
Lipids and BP have been associated in several cross-sectional studies.4,19,20 Castelli and Anderson19 found that BP and serum cholesterol were strongly correlated among hypertensive patients, which led to early recommendations to treat elevated cholesterol in patients with hypertension.5,19 Gaziano et al21 also noted a potential interaction between elevated cholesterol and hypertension in the development of MI that suggested a direct relationship rather than the effect of 2 independent predictors.
A few smaller studies have looked prospectively at the relationship between plasma lipids and the future development of hypertension. A 7-year follow-up of 1039 initially nondiabetic, nonhypertensive subjects from the San Antonio Heart Study suggested that risk factors for atherosclerosis, including triglycerides, also predicted hypertension.9 A prospective study of 1482 adults in Utah followed for 7 years with 40 cases of incident hypertension reported a significant age-adjusted RR of 1.42 for a 1 SD increase in triglycerides (110 mg/dL) and a nonsignificant RR of 0.82 for HDL-C (11 mg/dL).10 We would expect that if dyslipidemia played a role in the development of hypertension, then treating dyslipidemia would have some effect on BP. Data are limited to small or secondary analyses of the effects of lipid lowering on BP. A study of intensive cholesterol reduction in 22 patients with isolated systolic hypertension demonstrated that after 3 months of intensive treatment, SBP (148 versus 154 mm Hg, P=0.03) and DBP (81 versus 83 mm Hg, P=0.04) were lower among those assigned to atorvastatin versus placebo.11 The Brisighella Heart Study looked at the effect of different strategies of lipid lowering on BP, comparing diet, cholestyramine, gemfibrozil, and simvastatin. The investigators demonstrated a significant decrease in BP in the upper 2 quartiles of SBP (≥140 mm Hg) over a 5-year period and showed that this decrease was greater in subjects treated with cholesterol-lowering medication and greatest for those treated with a statin.12 In another study, there was no decrease in BP among normotensive subjects.22 Glorioso et al23 conducted a trial of pravastatin in 30 hypertensive, hypercholesterolemic subjects and demonstrated that, compared with placebo, treatment with pravastatin significantly decreased systolic, diastolic, and pulse pressures and blunted the BP increase induced by the cold pressor test.
The biological mechanisms by which lipids may play a role in the development of hypertension remain poorly understood. Atherogenic lipid abnormalities clearly cause endothelial dysfunction.4 A dysfunctional endothelium, possibly through impaired nitric oxide production and activity, as well as alterations in endothelin-1 and endothelin A and B receptor expression,24 cannot respond to changes in intravascular conditions to constrict and dilate as needed. This vasodysregulation could lead to an inability or difficulty in vasodilatation to appropriate stimuli and eventually to increased resting BP. Because atherosclerosis can be a diffuse process, it is possible that hypertension is a manifestation of a diffuse atherosclerotic process in large conduit arteries, as well as smaller resistance vessels.4 Nickenig and Harrison7,8 have linked lipids and hypertension via a mechanism of angiotensin I overexpression. Lipid abnormalities and insulin resistance have been associated with sympathetic hyperfunction,25 which may play a role in the development of hypertension.25 It has been suggested that hypertension and dyslipidemia are associated because they are 2 components of the metabolic syndrome, and that a common pathway, possibly insulin resistance or hyperinsulinemia,26 may unite the various abnormalities of the syndromes.27 Some have theorized, however, that hypertension is not an initial component of the metabolic syndrome, a suggestion borne out by factor analysis on the various components.26,27 Hypertension, which often occurs after other components of the metabolic syndrome, may instead be a late-stage manifestation, arising secondary to derangements of other components of the metabolic syndrome, such as dyslipidemia.26,27 Our prospective study finds an independent relationship between increased lipid levels and incident hypertension that predates the development of hypertension by years. This lends support to the theory that hypertension represents an early manifestation of the atherosclerotic process. Alternatively, more research is needed to understand the role of hypertension in the development of dyslipidemia.
Strengths of this study include its long duration, high rates of follow-up, and the high quality of data provided by health professionals. High rates of screening for a common illness, such as hypertension, in this healthy population of physicians would minimize undetected hypertension. Because the specificity of hypertension by self-report has been validated previously,15,28 any misclassification is likely to be underreporting, which would tend to underestimate the true strength of the association. However, we lack information on the number of BP recordings, which may indicate those with greater opportunities to have hypertension identified. In our study, hypertension is predicted on the basis of a single baseline lipid value and does not account for changes in lipids over time that may result from dietary changes or initiation of lipid-lowering medications. At the time of the baseline questionnaire in 1982, however, it was less common for people to receive pharmacological treatment for elevated cholesterol. As rates of treatment increased over the course of follow-up, the pharmacologically lowered cholesterol levels would lead to an underestimation of the true association. Although we adjusted for several covariates, we cannot exclude the possibility of residual confounding based on unmeasured or inadequately measured confounders, such as sodium intake. Also, our study population consists of highly educated men, the majority of whom are white. Though the homogenous nature of the cohort eliminates confounding by race and socioeconomic status, our results may not necessarily be generalizable to minorities and lower socioeconomic populations. The percentage of subjects that developed incident hypertension during follow-up, however, is commensurate with trends seen in the general population. From 1988 to 1994, the prevalence of hypertension in the United States between the ages of 20 and 74 years was &23%, and, in men over the age of 64 years, that proportion increased to 64%.1
Perspectives
In a large prospective cohort of healthy middle-aged and older men, we found a significant association between baseline lipids and the subsequent development of hypertension. This relationship between a common risk factor for atherosclerosis, dyslipidemia, and hypertension suggests that hypertension may be a manifestation of the atherosclerotic process. Currently not enough is known about the causes of hypertension, despite its high prevalence, and more studies are needed to determine whether dyslipidemia actually causes hypertension. By identifying potential risk factors amenable to intervention, we may eventually be able to reduce the burden of hypertension and subsequent cardiovascular disease.
Acknowledgments
This work was supported by research grants CA 34944, CA 40360, and CA 97193 from the National Cancer Institute and grants HL 26490 and HL 34595 from the National Heart Lung, and Blood Institute. The effort of R.O.H. was supported in part by a VA Special Ambulatory Fellowship award. We acknowledge the crucial contributions of the entire staff of the Physicians’ Health Study. We are also indebted to the 22 071 dedicated and committed participants randomized into the Physicians’ Health Study starting in 1982.
References
1.
Chobanian AV, Bakris GL, Black HR, Cushman WC, Green LA, Izzo JL, Jr., Jones DW, Materson BJ, Oparil S, Wright JT, Jr, Roccella EJ. The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure: The JNC 7 Report. JAMA. 2003; 289: 2560–2571.
2.
Whelton PK, He J, Appel LJ, Cutler JA, Havas S, Kotchen TA, Roccella EJ, Stout R, Vallbona C, Winston MC, Karimbakas J. Primary prevention of hypertension: clinical and public health advisory from The National High Blood Pressure Education Program. JAMA. 2002; 288: 1882–1888.
3.
Reaven GM, Lithell H, Landsberg L. Hypertension and associated metabolic abnormalities: the role of insulin resistance and the sympathoadrenal system. N Engl J Med. 1996; 334: 374–381.
4.
Oparil S, Zaman MA, Calhoun DA. Pathogenesis of hypertension. Ann Intern Med. 2003; 139: 761–776.
5.
Anderson KM, Castelli WP, Levy D. Cholesterol and mortality. 30 years of follow-up from the Framingham study. JAMA. 1987; 257: 2176–2180.
6.
Nickenig G. Central role of the AT(1)-receptor in atherosclerosis. J Hum Hypertens. 2002; 16 Suppl 3: S26–S33.
7.
Nickenig G, Harrison DG. The AT(1)-type angiotensin receptor in oxidative stress and atherogenesis: Part II: AT(1) receptor regulation. Circulation. 2002; 105: 530–536.
8.
Nickenig G, Harrison DG. The AT(1)-type angiotensin receptor in oxidative stress and atherogenesis: Part I: oxidative stress and atherogenesis. Circulation. 2002; 105: 393–396.
9.
Haffner SM, Miettinen H, Gaskill SP, Stern MP. Metabolic precursors of hypertension: the San Antonio Heart Study. Arch Intern Med. 1996; 156: 1994–2000.
10.
Hunt SC, Stephenson SH, Hopkins PN, Williams RR. Predictors of an increased risk of future hypertension in Utah. A screening analysis. Hypertension. 1991; 17: 969–976.
11.
Ferrier KE, Muhlmann MH, Baguet JP, Cameron JD, Jennings GL, Dart AM, Kingwell BA. Intensive cholesterol reduction lowers blood pressure and large artery stiffness in isolated systolic hypertension. J Am Coll Cardiol. 2002; 39: 1020–1025.
12.
Borghi C, Dormi A, Veronesi M, Sangiorgi Z, Gaddi A. Association between different lipid-lowering treatment strategies and blood pressure control in the Brisighella Heart Study. Am Heart J. 2004; 148: 285–292.
13.
The Steering Committee of the Physicians’ Health Study Research Group. Final report on the aspirin component of the ongoing Physicians’ Health Study. N Engl J Med. 1989; 321: 129–135.
14.
Rexrode KM, Buring JE, Glynn RJ, Stampfer MJ, Youngman LD, Gaziano JM. Analgesic use and renal function in men. JAMA. 2001; 286: 315–321.
15.
Colditz GA, Martin P, Stampfer MJ, Willett WC, Sampson L, Rosner B, Hennekens CH, Speizer FE. Validation of questionnaire information on risk factors and disease outcomes in a prospective cohort study of women. Am J Epidemiol. 1986; 123: 894–900.
16.
Giles WH, Croft JB, Keenan NL, Lane MJ, Wheller FC. The validity of self-reported hypertension and correlates of hypertension awareness among Blacks and Whites within the stroke belt. Am J Prev Med. 1995; 11: 163–169.
17.
Hennekens CH, Buring JE, Manson JE, Stampfer M, Rosner B, Cook NR, Belanger C, LaMotte F, Gaziano JM, Ridker PM, Willett W, Peto R. Lack of effect of long-term supplementation with beta carotene on the incidence of malignant neoplasms and cardiovascular disease. N Engl J Med. 1996; 334: 1145–1149.
18.
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: 2486–2497.
19.
Castelli WP, Anderson K. A population at risk. Prevalence of high cholesterol levels in hypertensive patients in the Framingham Study. Am J Med. 1986; 80: 23–32.
20.
Selby JV, Newman B, Quiroga J, Christian JC, Austin MA, Fabsitz RR. Concordance for dyslipidemic hypertension in male twins. JAMA. 1991; 265: 2079–2084.
21.
Gaziano JM, Sesso HD, Breslow JL, Hennekens CH, Buring JE. Relation between systemic hypertension and blood lipids on the risk of myocardial infarction. Am J Cardiol. 1999; 84: 768–773.
22.
Kool M, Lustermans F, Kragten H, Struijker Boudier H, Hoeks A, Reneman R, Rila H, Hoogendam I, Van Bortel L. Does lowering of cholesterol levels influence functional properties of large arteries? Eur J Clin Pharmacol. 1995; 48: 217–223.
23.
Glorioso N, Troffa C, Filigheddu F, Dettori F, Soro A, Parpaglia PP, Collatina S, Pahor M. Effect of the 3hydroxy3methylglutaryl (HMG)-coenzymeA (CoA) reductase inhibitors on blood pressure in patients with essential hypertension and primary hypercholesterolemia. Hypertension. 1999; 34: 1281–1286.
24.
Nohria A, Garrett L, Johnson W, Kinlay S, Ganz P, Creager MA. Endothelin-1 and vascular tone in subjects with atherogenic risk factors. Hypertension. 2003; 42: 43–48.
25.
Egan BM. Insulin resistance and the sympathetic nervous system. Curr Hypertens Rep. 2003; 5: 247–254.
26.
Shen BJ, Todaro JF, Niaura R, McCaffery JM, Zhang J, Spiro A, 3rd, Ward KD. Are metabolic risk factors one unified syndrome? Modeling the structure of the metabolic syndrome X. Am J Epidemiol. 2003; 157: 701–711.
27.
Leyva F, Godsland IF, Worthington M, Walton C, Stevenson JC. Factors of the metabolic syndrome: baseline interrelationships in the first follow-up cohort of the HDDRISC Study (HDDRISC-1). Heart Disease and Diabetes Risk Indicators in a Screened Cohort. Arterioscler Thromb Vasc Biol. 1998; 18: 208–214.
28.
Klag MJ, He J, Mead LA, Ford DE, Pearson TA, Levine DM. Validity of physicians’ self-reports of cardiovascular disease risk factors. Ann Epidemiol. 1993; 3: 442–447.
Information & Authors
Information
Published In
Copyright
© 2006.
Versions
You are viewing the most recent version of this article.
History
Received: 20 July 2005
Revision received: 3 August 2005
Accepted: 7 October 2005
Published online: 12 December 2005
Published in print: 1 January 2006
Keywords
Authors
Metrics & Citations
Metrics
Citations
Download Citations
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Select your manager software from the list below and click Download.
- Association between Weight Change and Incidence of Dyslipidemia in Young Adults: A Retrospective Cohort Study of Korean Male Soldiers, Journal of Obesity & Metabolic Syndrome, 33, 1, (36-44), (2024).https://doi.org/10.7570/jomes23056
- Prediabetes, Dyslipidemia, and Hypertension in Acute Coronary Syndrome Patients: A Cross-sectional Study at a Tertiary Hospital in Cirebon District, Indonesia, South East European Journal of Cardiology, 5, (26-29), (2024).https://doi.org/10.3889/seejca.2024.6071
- Recognition of Early Cardiovascular Disease Symptoms in Hypertensive and Dyslipidemic Individuals of Icheon, Korea: Insights into Educational Levels and Health Literacy, Healthcare, 12, 7, (736), (2024).https://doi.org/10.3390/healthcare12070736
- Relationship Between Altered Plasma-Free Amino Acid Levels and Hyperuricemia in Dyslipidemia Without and With Hypertension, Diseases, 12, 11, (267), (2024).https://doi.org/10.3390/diseases12110267
- Genetic polymorphisms affecting telomere length and their association with cardiovascular disease in the Heinz-Nixdorf-Recall study, PLOS ONE, 19, 5, (e0303357), (2024).https://doi.org/10.1371/journal.pone.0303357
- A higher TyG index level is more likely to have enhanced incidence of T2DM and HTN comorbidity in elderly Chinese people: a prospective observational study from the reaction study, Diabetology & Metabolic Syndrome, 16, 1, (2024).https://doi.org/10.1186/s13098-024-01258-3
- Hypertension among persons living with HIV/AIDS and its association with HIV-related health factors, AIDS Research and Therapy, 21, 1, (2024).https://doi.org/10.1186/s12981-023-00576-2
- The association of the platelet/high-density lipoprotein cholesterol ratio with self-reported stroke and cardiovascular mortality: a population-based observational study, Lipids in Health and Disease, 23, 1, (2024).https://doi.org/10.1186/s12944-024-02115-y
- Relationship between serum apolipoprotein B and risk of all-cause and cardiovascular disease mortality in individuals with hypertension: a prospective cohort study, BMC Cardiovascular Disorders, 24, 1, (2024).https://doi.org/10.1186/s12872-024-03949-1
- Association of atherosclerosis indices, serum uric acid to high‐density lipoprotein cholesterol ratio and triglycerides‐glucose index with hypertension: A gender‐disaggregated analysis, The Journal of Clinical Hypertension, 26, 6, (645-655), (2024).https://doi.org/10.1111/jch.14829
- See more
Loading...
View Options
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Personal login Institutional LoginPurchase Options
Purchase this article to access the full text.
eLetters(0)
eLetters should relate to an article recently published in the journal and are not a forum for providing unpublished data. Comments are reviewed for appropriate use of tone and language. Comments are not peer-reviewed. Acceptable comments are posted to the journal website only. Comments are not published in an issue and are not indexed in PubMed. Comments should be no longer than 500 words and will only be posted online. References are limited to 10. Authors of the article cited in the comment will be invited to reply, as appropriate.
Comments and feedback on AHA/ASA Scientific Statements and Guidelines should be directed to the AHA/ASA Manuscript Oversight Committee via its Correspondence page.