DASH Diet Lowers Blood Pressure and Lipid-Induced Oxidative Stress in Obesity
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
DASH Diet Lowers Blood Pressure and Lipid-Induced
Oxidative Stress in Obesity
Heno F. Lopes, Kelley L. Martin, Khaled Nashar, Jason D. Morrow,
Theodore L. Goodfriend, Brent M. Egan
Abstract—Evidence suggests that obesity may raise blood pressure (BP) through oxidative stress–sensitive mechanisms
and that the Dietary Approaches to Stop Hypertension combination diet (DASH-CD) may decrease BP by enhancing
antioxidant capacity. To address this question, 12 obese patients with high-normal–to–stage 1 hypertension (hyperten-
sives) and 12 lean normotensives were studied on their usual diets and after following the DASH-CD and a
low-antioxidant diet in random sequence for 4 weeks each. Acute oxidative stress was induced by a 4-hour infusion of
intralipid and heparin. Ferric-reducing activity of plasma (FRAP) and plasma F2-isoprostanes were measured as
biomarkers of antioxidant capacity and oxidative stress, respectively. BP was lower in obese hypertensives on the
DASH-CD than on the usual and low-antioxidant diets (⫺8.1⫾1.5/⫺7.4⫾1.6 mm Hg, P⬍0.05). BP did not change
significantly in lean normotensives after 4 weeks on the DASH-CD but tended to rise on the low-antioxidant diet. FRAP
on usual diets was higher in lean subjects than in obese subjects. FRAP increased in obese but not lean volunteers on
the DASH-CD compared with usual diet, and the group difference disappeared. F2-isoprostanes increased from baseline
during intralipid and heparin in both groups on the low-antioxidant diet but not in obese hypertensives on the
DASH-CD. Among free-living obese hypertensives, the DASH-CD raises antioxidant capacity, lowers BP, and reduces
oxidative stress induced by acute hyperlipidemia. The findings are consistent with evidence that elevated BP in obese
subjects may reflect an imbalance between antioxidant capacity and oxidative stress that is improved by the DASH-CD.
(Hypertension. 2003;41:422-430.)
Key Words: obesity 䡲 insulin resistance 䡲 fatty acids 䡲 F2-isoprostanes 䡲 oxidative stress 䡲 antioxidants
P atients with high-normal blood pressure (BP) and hyper-
tension are more obese and insulin resistant than are
normotensives. Reduced antioxidant capacity and oxidative
without low-fat dairy products, reduced BP significantly
more in hypertensives than in normotensives.16 Both diets are
rich in a variety of antioxidants, which may contribute to the
stress represent a potential mechanism linking obesity with BP-lowering effects. Collectively, published observations
insulin resistance, elevated BP, and cardiovascular dis- raise the possibility that the DASH intervention lowers BP by
ease.1–10 For example, acutely raising lipids with a short-term increasing antioxidant capacity, particularly in obese insulin-
infusion of intralipid and heparin, which mimics and/or resistant hypertensives. The principal focus of the present
exacerbates the dyslipidemia seen with insulin resistance, study was to determine the impact of the DASH prescription
increases BP and raises F2-isoprostanes, a biomarker of on BP and indices of antioxidant capacity and oxidative stress
oxidative stress.11,12 Moreover, in experimental models, in- at baseline and in response to an acute lipid stressor in
creasing oxidative stress with a long-term low-dose infusion free-living obese subjects with high-normal and stage 1
hypertension (hypertensives) and lean normotensive
of angiotensin and reducing antioxidant capacity by depletion
volunteers.
of glutathione produce severe hypertension.2,13 In these mod-
els, BP declines when antioxidant defenses are augmented by
Methods
providing either superoxide dismutase13,14 or vitamins C and
E.13 Vitamin C also lowers BP in humans, which implicates Human Volunteers
All subjects read and signed a written informed consent document
a role for antioxidant capacity in human hypertension.15
approved by the Institutional Review Board and the General Clinical
In the Dietary Approaches to Stop Hypertension (DASH) Research Center (GCRC) Advisory Committee. Twenty-four volun-
Study, a diet rich in fruits and vegetables, either with or teers were enrolled, including 12 obese (body mass index ⱖ27
Received September 3, 2002; first decision September 17, 2002; revision accepted December 10, 2002.
From the Departments of Pharmacology (H.F.L., K.N., B.M.E.) and Medicine (B.M.E.) and the General Clinical Research Center (K.L.M., B.M.E.),
Medical University of South Carolina, Charleston; Unidade de Hipertensão-Heart Institute (InCor), HCFM-USP (H.F.L.), São Paulo, Brazil; Departments
of Pharmacology and Medicine, Vanderbilt University School of Medicine (J.D.M.), Nashville, Tenn; and Departments of Pharmacology and Medicine,
University of Wisconsin and William S. Middleton Veterans Hospital (T.L.G.), Madison.
Correspondence to Brent M. Egan, MD, Departments of Medicine and Pharmacology, Medical University of South Carolina, 96 Jonathan Lucas St,
CSB 826H, Charleston, SC 29425. E-mail eganbm@musc.edu
© 2003 American Heart Association, Inc.
Hypertension is available at http://www.hypertensionaha.org DOI: 10.1161/01.HYP.0000053450.19998.11
422
Downloaded from http://hyper.ahajournals.org/ by guest on February 3, 2015Lopes et al Dietary Antioxidants and Blood Pressure 423
kg/m2) dyslipidemic (triglycerides ⬎150 mg/dL and/or HDL-choles- Plasma F2-isoprostanes, a biomarker of oxidative stress, were
terol ⬍45 for women or ⬍40 mg/dL for men) subjects with measured on all 3 dietary phases by using gas chromotography/
high-normal to stage 1 hypertension (BP 130 to 159/85 to negative ion chemical ionization (GC/NICI) mass spectrometry as
99 mm Hg) and 12 lean (body mass index ⱕ25 kg/m2) normoten- previously described.24,25
sives (BP⬍130/85 mm Hg) with normal lipids matched for age, race,
and gender. After 3 consecutive weekly screening visits to ensure Study Design and Protocol
eligibility, including BP consistently within the defined ranges, After baseline studies on their usual diets, subjects entered an
subjects were referred to a nutritionist for instructions on the study open-label randomized crossover study from the DASH-CD, which
diets. is high in antioxidants, to a low-antioxidant diet, which is compara-
ble to the usual diet in most subjects, for 4 weeks each. On each of
Diets the 3 diets, subjects were admitted to the outpatient GCRC at 8:00
A registered dietitian (K.L.M.) instructed each subject on how to AM after an overnight fast. Intravenous access was established in
plan and prepare these diets for home consumption. Each volunteer both arms, with 1 side for obtaining blood samples and the other for
received written meal plans, recipe lists, and portion control infor- infusion of intralipid and heparin. Baseline hemodynamic data were
mation to match dietary goals. Subjects and investigators could not obtained at 5-minute intervals for 30 minutes. At the end of 30
be blinded owing to obvious differences in the 2 diets. The diets were minutes, blood samples were drawn for serum lipids and plasma
designed to maintain consistent intake of sodium, caffeine, and insulin, total antioxidant capacity (FRAP), F2-isoprostanes, NEFAs,
alcohol, which could influence BP. Each subject’s isocaloric energy and nitrite-nitrate. An infusion was then started of 20% intralipid
intake was estimated by using the Harris Benedict equation,17 and (Baxter Healthcare Corp) at 0.8 mL 䡠 m⫺2 䡠 min⫺1 and heparin (200-U
calories were adjusted according to clinic weights to minimize bolus, followed by infusion at 1000 U/h). Heparin was given to
weight changes. Subjects were monitored by the dietitian at least activate lipoprotein lipase and accelerate the hydrolysis of fatty acids
every 2 weeks to review their diet. from triglycerides.11 Hemodynamic measurements were made, and
Subjects received a disposable camera to photograph all foods and blood samples were obtained for triglycerides, NEFAs, F2-
beverages, and they also kept a food diary for 3 days, which included isoprostanes, and nitrite-nitrate at 2 hours and 4 hours of the
2 working days and 1 nonworking day, before each study. These intralipid and heparin infusion.
tools were used in combination, because the accuracy of self-reports
is limited.18 Analysis of the diaries and photographs was conducted Statistical Analysis
with Nutritionist Five software (First DataBank). Group comparisons for descriptive dichotomous variables, eg, gen-
Physiological data, BP, and heart rate were measured in triplicate der, were made by using the 2 test. For descriptive numeric
with subjects in the sitting position for the 3 qualifying visits and variables such as age, body mass index, nutrient intake, casual BP,
weekly thereafter for the remainder of the study. BP was measured and heart rate, the Student unpaired t test was used for between-
by a trained observer using a mercury sphygmomanometer and group comparisons. Changes in hemodynamic variables, plasma
appropriately sized cuff. Systolic BP was defined by the first total antioxidant capacity (FRAP), plasma F2-isoprostanes, plasma
Korotkoff sound, diastolic BP by the disappearance of the last (fifth) nitrite-nitrate, and measures of insulin sensitivity were assessed by
Korotkoff sound. Heart rate was measured by palpating the radial
pulse for 60 seconds between the second and third BP measurement,
with subjects seated. TABLE 1. Selected Demographic and Biochemical Data in
Obese and Lean Volunteers
Hemodynamic Measurements Obese Lean
The H.D.I./PulseWave CR-2000 was used to noninvasively assess Hypertensives Normotensives
BP, heart rate, systemic vascular resistance, stroke volume, cardiac Variables (n⫽12) (n⫽12) P
output, large artery elasticity index, small artery elasticity index, and
total vascular impedance.11 In brief, a tonometer was placed securely Age, y 35⫾2 39⫾2 NS
over the left radial artery for pulse waveform analysis, and a properly Gender, f/m 6/6 6/6 NS
sized cuff was placed on the right upper extremity for BP
Race, black/white 7/5 6/6 NS
measurements.
Systolic BP, mm Hg 130⫾1 111⫾2 ⬍0.0001
Metabolic Data and Assay Methods Diastolic BP, mm Hg 92⫾1 72⫾2 ⬍0.0001
The homeostatic model assessment of insulin resistance (HOMAir) Body mass index, kg/m2 34.1⫾1.9 22.7⫾0.5 ⬍0.0001
index was obtained as an index of insulin action.19 Plasma insulin
was measured by radioimmunoassay (RIA).20 Cholesterol, mg/dL
Total 229⫾6 173⫾7 ⬍0.0001
Nonesterified Fatty Acids and Lipids HDL 47⫾3 61⫾5 ⬍0.05
Blood for nonesterified fatty acids (NEFAs) was drawn into pre- LDL 152⫾5 95⫾6 ⬍0.0001
chilled Eppendorf tubes containing paraoxon.11 Plasma was stored at
⫺70°C before analyzing the total plasma NEFAs by using the 63Ni VLDL 29⫾4 16⫾1 ⬍0.01
method.5 Triglycerides were measured by the fluorometric method. Triglycerides, mg/dL 171⫾30 81⫾6 0.01
Total cholesterol was measured by the colorimetric method, and Glucose, mg/dL 90⫾3 83⫾3 0.08
HDL-cholesterol was prepared from whole plasma by precipitation
with phosphotungstate-MgCl2.21 LDL and VLDL cholesterol were Sodium, mmol/L 139.3⫾0.7 142.0⫾0.7 0.01
calculated.22 Potassium, mmol/L 4.34⫾0.15 4.32⫾0.08 NS
Urea nitrogen, mg/dL 12⫾1 12⫾1 NS
Ferric Reducing/Antioxidant Power of Plasma Assay
Creatinine, mg/dL 0.90⫾0.05 0.92⫾0.03 NS
Total antioxidant capacity was measured by the ferric reducing/
antioxidant power of plasma (FRAP) assay23 under fasting condi- Calcium, mg/dL 9.4⫾0.1 9.3⫾0.1 NS
tions, with volunteers on their usual diet, and after 4 weeks each on Phosphorus, mg/dL 3.4⫾0.1 3.5⫾0.1 NS
the DASH-CD and low-antioxidant diets. The FRAP assay was
Uric acid, mg/dL 5.5⫾0.3 4.5⫾0.3 ⬍0.05
performed in triplicate on each sample, and the mean of the 3 values
was used in the analysis. Data presented as mean⫾SEM.
Downloaded from http://hyper.ahajournals.org/ by guest on February 3, 2015424 Hypertension March 2003
TABLE 2. Nutritional Analysis and Comparison of the DASH-CD and
Low-Antioxidant Diets
DASH Combination Diet Low-Antioxidant Diet
Goal Actual Goal Actual
Nutrient (Mean) (Mean⫾SEM) (Mean) (Mean⫾SEM) *P
Kcals, kcal/d 2000 1850⫾86 2000 2067⫾98 0.10
Protein, % total kcal 16 16⫾1 16 16⫾1 0.93
Fat, % total kcal 27 29⫾1 37 37⫾1 ⬍0.0001
Unsaturated fat, % total kcal 21 21⫾0 21 24⫾0 0.016
Saturated fat, % total kcal 6 8⫾1 16 13⫾1 ⬍0.0001
Cholesterol, mg/d 150 200⫾19 300 394⫾50 ⬍0.001
Carbohydrate, % total kcal 55 56⫾1 48 47⫾1 ⬍0.0001
Dietary fiber, g/d 31 22⫾1 9 12⫾1 ⬍0.0001
Alcohol, g/d ⱕ30 5⫾1 ⱕ30 3⫾1 0.23
Na⫹, mg/d 3000 2792⫾167 3000 3437⫾210 0.02
K⫹, mg/d 4700 2878⫾177 1700 1898⫾115 ⬍0.0001
2⫹
Ca , mg/d 1240 763⫾58 450 619⫾39 0.04
Mg⫹⫹, mg/d 500 294⫾21 165 159⫾9 ⬍0.0001
Vitamin C, mg/d 266 179⫾16 133 69⫾11 ⬍0.0001
Vitamin A, IU/d 14409 9762⫾1644 6192 3750⫾374 0.0003
Fruit, exchanges/d 4–5 3.8⫾0.4 1–2 1.0⫾0.2 ⬍0.0001
Vegetables, exchanges/d 4–5 2.3⫾0.3 2 0.9⫾0.1 ⬍0.0001
Low fat dairy, exchanges/d 2 0.6⫾0.1 0 0.2⫾0.1 ⬍0.0009
*P value comparing actual data on the DASH-CD and low-antioxidant diets.
using the general linear model for repeated measures. The Pearson In comparing nutrient intake between obese and lean
correlation was used to assess the relationship of antioxidant capacity subjects on their usual diets, obese subjects had significantly
to BP. All statistical analyses were performed with the SPSS/PC 10.0
lower caffeine intakes (30⫾6 mg/d) than did lean volunteers
statistical software package (SPSS). A probability value ⬍0.05 was
accepted as statistically significant. (79⫾12 mg/d, P⬍0.01). None of the other variables listed in
Table 2 showed significant differences between the 2 groups
on the usual diets. On the DASH-CD, obese subjects had
Results
significantly lower values than did lean volunteers for dietary
Demographic and Biochemical Data K⫹ (2415⫾224 versus 3327⫾231 mg/d), vitamin C (253⫾32
Demographic and biochemical data for the 12 lean normo- versus 337⫾25 mg/d), and fruits (3.0⫾0.5 versus 4.6⫾0.6
tensives and 12 obese hypertensives are shown in Table 1. exchanges/d). No significant differences between the 2
Both groups had similar values for age, gender, and race. groups were observed for the other variables listed in Table 2.
Obese hypertensives had higher values for total and LDL On the low-antioxidant diet, a lower dietary fiber intake in
cholesterol, triglycerides, and uric acid and lower values for lean volunteers than in obese volunteers (13.3⫾1.4 versus
HDL cholesterol than did lean normotensives. 9.7⫾0.9 g/d, P⬍0.05) was the only significant difference
between the 2 groups. The lipid profiles within each group
Nutritional Content of the Diets did not differ between the 3 dietary periods (Table 3).
The nutrient goals and average daily intakes for the usual diet,
DASH-CD, and low-antioxidant diets estimated from food Twenty-four–Hour Urinary Electrolytes
diaries and photographs are provided in Table 2. Caffeine The mean of the 4 weekly values for 24-hour urinary Na⫹ for
intake was similar on the DASH-CD and low-antioxidant subjects on the usual diet, DASH-CD, and the low-
diets (43⫾7 versus 55⫾7 mg/d, P⫽NS). Although expected antioxidant diets were not statistically different at 156⫾11,
differences between the DASH-CD and the low-antioxidant 138⫾14, and 158⫾16 mmol/d, respectively. Moreover, val-
diet were found, actual intakes for the DASH-CD deviated by ues for 24-hour urinary Na⫹ did not differ between lean and
32% to 49% from goal for percentage of calories as saturated obese subjects on the different diets (usual, 156⫾18 versus
fat, cholesterol, dietary fiber, K⫹, Ca2⫹, Mg2⫹, vitamins C and 157⫾13 mEq/d; DASH-CD, 135⫾21 versus 141⫾20 mEq/d;
A, and vegetables. The greatest discrepancy between ex- low antioxidant, 158⫾27 versus 158⫾19 mmol/d; respective-
pected and actual intake was present for low-fat dairy ly). The mean of 4 weekly values for 24-hour urinary K⫹
products. Actual intake for the low-antioxidant diet deviated were higher (P⬍0.05) in subjects on the DASH-CD
from goal by 31% to 48% for cholesterol, dietary fiber, Ca2⫹, (54⫾5 mmol/d) compared with the low-antioxidant diet
and vitamins C and A. (39⫾3 mmol/d) weekly. Twenty-four– hour urinary Ca2⫹ and
Downloaded from http://hyper.ahajournals.org/ by guest on February 3, 2015Lopes et al Dietary Antioxidants and Blood Pressure 425
TABLE 3. Data on Weight and Markers of Insulin Resistance and Antioxidant Capacity Shown
Separately for Obese Hypertensives and Lean Normotensives on the 3 Diets
Obese Hypertensives (n⫽12) Lean Normotensives (n⫽12)
Variables Usual Diet DASH-CD Low Antioxidant Usual Diet DASH-CD Low Antioxidant
Weight, kg 101⫾6 100⫾6 101⫾6 69⫾3 69⫾3 69⫾3
Cholesterol, mg/dL
Total 209⫾8 206⫾8 205⫾6 158⫾7† 154⫾6† 162⫾6†
HDL 44⫾3 42⫾3 44⫾4 54⫾4 50⫾3 56⫾5
LDL 131⫾8 125⫾8 123⫾5 88⫾5† 89⫾4† 88⫾5†
Triglyceride, mg/dL 168⫾29 191⫾42 194⫾32 83⫾10† 79⫾10† 94⫾14†
Glucose, mg/dL 91⫾26 93⫾27 92⫾27 85⫾25 85⫾24† 81⫾24†
Insulin, U/mL 12.5 ⫾ 1.1 15.1⫾2.3 17.3⫾4.1 6.5⫾0.9† 5.6⫾1.3† 6.6⫾1.4†
HOMAir 51⫾15 63⫾18 75⫾22 25⫾7† 22⫾6† 24⫾7†
FRAP, mol/L 271⫾15 334⫾27* 300⫾20 333⫾29† 350⫾35 345⫾39
F2-Isoprostanes, pg/mL 42⫾5 40⫾5 40⫾5 39⫾5 35⫾4 41⫾5
Values are mean⫾SEM. HOMAir indicates homeostatic model assessment of insulin resistance; FRAP, ferric
reducing activity of plasma.
* P⬍0.05 vs baseline; †P⬍0.05 lean vs obese.
Mg2⫹ values did not differ between the usual, DASH-CD, and Acute Hyperlipidemia
low-antioxidant diets. The intralipid and heparin infusion increased plasma triglyc-
erides and NEFAs after 2 and 4 hours on all 3 diets in both
Insulin Sensitivity lean normotensives and obese hypertensives (Figures 2a and
Baseline values derived from the homeostatic model assess- 2b). The changes in plasma triglycerides during intralipid and
ment of insulin resistance (HOMAir) were greater in obese heparin infusion were similar within and between both groups
volunteers than in lean volunteers (Table 3). on all 3 diets. NEFAs tended to increase more in lean
Antioxidant capacity, as measured by FRAP, was greater volunteers than in obese volunteers during the infusion of
in lean subjects than in obese subjects on normal diets. FRAP intralipid and heparin.
increased in obese volunteers but not in lean volunteers after After 4 hours of the intralipid and heparin infusion, systolic
4-weeks on the DASH-CD (Table 3), and the group differ- BP increased significantly in obese hypertensives and in lean
ence disappeared. normotensives on all 3 diets. Despite the increase with acute
hyperlipidemia, systolic and diastolic BP in obese hyperten-
BP Responses to the Dietary Interventions sives remained lower on the DASH-CD than on the usual and
Systolic, diastolic, and mean BP declined significantly in low-antioxidant diets (Figures 3a and 3b).
obese hypertensives after weeks 3 and 4 on the DASH-CD Small artery elasticity index decreased in lean normoten-
compared with values on the baseline and low-antioxidant sives and in obese hypertensives during the intralipid and
diets (Figure 1a through 1c). After 4 weeks on the DASH-CD, heparin infusion when they were on usual and low-
baseline BP in obese hypertensives was 8.1⫾1.5/ antioxidant diets. However, small artery elasticity did not
7.4⫾1.6 mm Hg lower than on the low-antioxidant diet. decline during the infusion of intralipid and heparin in obese
Plasma antioxidant activity also correlated negatively with subjects on the DASH-CD (Figure 3c), as it did in lean
systolic BP in obese hypertensives after 4 weeks on the normotensives.
DASH-CD (r⫽⫺0.60, P⬍0.05). In obese hypertensives, the
differences in systolic and diastolic BP on the usual and low- F2-Isoprostanes
antioxidant diets were not significant (0.6⫾1.3/⫺0.6⫾1.3 mm Hg). Plasma F2-isoprostanes increased in both groups during the
In lean normotensives, systolic BP tended to rise during 4-hour infusion of intralipid and heparin when subjects were
weeks 2 and 3 of the low-antioxidant diet, but the difference consuming their usual and low-antioxidant diets (Figure 2c).
compared with that of the DASH-CD was not significant. However, plasma F2-isoprostanes did not increase signifi-
Diastolic and mean BPs were higher (P⬍0.05) after 3 weeks cantly in obese hypertensives during intralipid and heparin
on the low-antioxidant diet than on the DASH-CD (Figure 1b after 4 weeks on the DASH-CD.
and 1c), but the difference between the 2 diets was not
significant at 4 weeks. Discussion
Mean values for heart rate in obese subjects were 5 bpm In this study, obese subjects with BP values in the high-
greater after 4 weeks on the low-antioxidant diet than on the normal–to–stage 1 hypertension range (hypertensives) and
DASH-CD, but the difference was not statistically significant evidence for the metabolic syndrome had a significant decline
(P⫽0.15). Heart rate did not differ between the various in BP (⫺8.1⫾1.5/7.4⫾1.6 mm Hg) when they followed a
dietary phases in lean normotensives. DASH-CD prescription over 4 weeks (Figure 1). In contrast,
Downloaded from http://hyper.ahajournals.org/ by guest on February 3, 2015426 Hypertension March 2003
Figure 1. The changes from baseline in
systolic (a), diastolic (b), and mean (c)
BPs are depicted over 4 weeks on DASH
and low-antioxidant (LAO) diets for
obese hypertensives (left) and lean nor-
motensives (right).
BP did not change significantly in lean normotensives after 4 By comparison, in the original DASH study, subjects re-
weeks on the DASH-CD. Although systolic BP was ceived prepackaged meals.1 Second, many subjects were
⬇5 mm Hg higher during the second and third weeks of the reluctant to increase low-fat dairy products to the target level.
low-antioxidant diet than of the DASH-CD in lean normo- Although attempts were made to encourage consumption of
tensives, the difference was not significant. Ca2⫹-fortified foods, actual Ca2⫹ intake was ⬇60% of goal.
The decrease of systolic BP in obese hypertensives is Despite limited adherence, subjects consumed a statistically
similar in magnitude to that seen in hypertensive patients with higher intake of total fruits, vegetables, and low-fat diary
the high fruits-and-vegetables component of DASH diet and products on DASH-CD than on the low-antioxidant diet.
less than that seen among hypertensive patients in the original The DASH-CD may lower BP by raising antioxidant
DASH-CD.16 The more limited effect of our DASH-CD capacity and improving the balance between antioxidant
prescription on BP compared with the sentinel study may defenses and oxidative stress. Antioxidant capacity, measured
reflect the incomplete dietary adherence of our volunteers by the FRAP assay, was higher in lean normotensives than in
(Table 2). The failure of BP to decline significantly in lean obese hypertensives on their usual diets (Table 3). After 4
volunteers on the DASH-CD did not reflect poorer dietary weeks on the DASH-CD, plasma antioxidant capacity in-
adherence than that of obese subjects, as noted in the Results. creased significantly in obese hypertensives but not in lean
In fact, if anything, the lean normotensives had better dietary normotensives, and the group difference was eliminated.
adherence, as evidenced by greater consumption of fruit and Although F2-isoprostanes, a marker of oxidative stress, did
estimated daily K⫹ intake compared that of with the obese not change with diet, oxidative stress may raise BP2,13,14 and
hypertensives on the DASH-CD. Our findings are consistent has emerged as a possible mechanism by which obesity and
with the DASH study results, in which hypertensives had a insulin resistance raise BP.26 Moreover, oxidative stress is a
significantly greater reduction of BP than did common signal transduction mechanism by which various
normotensives.16 risk factors may induce vascular remodeling.9,27
The explanation for the discrepancies in goal versus actual The DASH-CD is high in antioxidants contained in fruits,
dietary intake was probably a function of study design. vegetables, and whole grains, and this diet raised antioxidant
Subjects were instructed to plan and prepare their own meals. capacity (Table 3) compared with that of usual diets in the
Downloaded from http://hyper.ahajournals.org/ by guest on February 3, 2015Lopes et al Dietary Antioxidants and Blood Pressure 427
Figure 2. Changes from baseline are
depicted in triglycerides (a), NEFAs (b),
and plasma F2-isoprostanes (c) at 2 and
4 hours of acute hyperlipidemia in obese
hypertensives (left) and lean normoten-
sives (right) on the usual diet (䢇),
DASH-CD (䡩), and low-antioxidant diet
(䡲).
obese volunteers. In the present study, we do not have DASH-CD than on the usual diet. Vitamin C decreases
evidence that the DASH-CD reduced oxidative stress under NADPH oxidase, a major superoxide-generating enzyme, in
basal conditions using F2-isoprostanes as a biomarker. There spontaneously hypertensive rats.31 These data suggest that
are multiple biomarkers of oxidative stress in vivo, which increasing antioxidant capacity can reduce the genesis of
have various advantages and limitations.28 Among the more reactive oxygen.
notable limitations is the lack of tissue and signal transduc- Our results indicate that borderline and stage 1 hyperten-
tion pathway specificity, particularly in target tissues critical sive, obese, dyslipidemic subjects have a lower total antiox-
in BP regulation. It is not clear in this study whether oxidative idant capacity than that of lean normotensive volunteers with
stress did not change on the DASH-CD or if plasma F2- normal lipids, when both groups are consuming their usual
isoprostanes did not reflect a true change. Based on the data, diets. These data raise the possibility of an imbalance be-
the benefits of the DASH-CD on BP16 may be mediated in tween the production or reactive oxygen species and the
part by improving the balance between antioxidant capacity capacity of antioxidant defenses in obese subjects, which
and oxidative stress in obese subjects. could contribute to elevated BP.2,13 The DASH-CD may
The difference in total antioxidant capacity between obese improve this imbalance in obese hypertensive patients.
hypertensives and lean normotensives on their usual diets was Obesity is associated with hypertension32 and insulin
noted previously. Tse et al29 documented that hypertensive resistance.33 Weight loss lowers BP and reduces insulin
patients had lower levels of ascorbic acid and albumin- resistance in obese hypertensive subjects.34,35 Insulin resis-
corrected thiol levels, than those of normotensive volunteers. tance is also a risk factor for cardiovascular disease,36 and
In another study,30 plasma ascorbic acid was inversely related markers of insulin resistance are associated with hyperten-
to systolic and diastolic BP of hypertensive patients indepen- sion.1 Thus, effects of the DASH-CD on weight and insulin
dently of other traditional factors, and changes of ascorbic resistance could account for some of the BP reduction
acid with nutritional depletion and repletion preceded observed in obese hypertensives. However, weight did not
changes of BP. In the present study, vitamin C consumption change significantly throughout the study in either group.
was ⬇3 times greater in both lean and obese subjects on the Although evidence for insulin resistance was more apparent
Downloaded from http://hyper.ahajournals.org/ by guest on February 3, 2015428 Hypertension March 2003
Figure 3. Values for systolic BP (a), dia-
stolic BP (b), and small artery elasticity
index (SAEI) are depicted at baseline and
after 2 and 4 hours of acute hyperlipid-
emia in obese hypertensives (left) and
lean normotensives (right) on the usual
diet (䢇), DASH-CD (䡩), and low-
antioxidant diet (䡲).
in obese volunteers than in lean volunteers (Table 3), these hypertensive subjects following usual, DASH-CD, and low-
indices were not significantly different within either group antioxidant diets also had an increase in systolic BP during an
among the 3 diets. From the available data, the BP decline in intralipid and heparin infusion. Similar to our previous study,
obese hypertensives on the DASH-CD was probably not intralipid and heparin increased plasma F2-isoprostanes in
explained by changes of weight or insulin sensitivity. lean normotensives and in obese, dyslipidemic, hypertensive
Components of the DASH-CD in addition to antioxidants subjects on their normal diets, as well as during the low-
may contribute to the BP decline. Based on food records, the antioxidant diet. In contrast, F2-isoprostanes did not increase
DASH-CD contained more K⫹, Ca2⫹, and Mg2⫹ and less Na⫹ significantly during acute hyperlipidemia when obese sub-
than did the usual and low-antioxidant diets, and 24-hour jects were consuming the DASH-CD. One implication of
urine K⫹ values were also greater on the DASH-CD than on these findings is that the DASH-CD enhances antioxidant
the low-antioxidant diet. Although 24-hour urine Na⫹ tended capacity, which, in turn, attenuates the oxidative stress
to be lower on the DASH-CD than the other dietary periods, response induced by acute hyperlipidemia.
the differences were not significant. Although the differences Although oxidative stress, as measured by F2-isoprostanes,
in cation consumption may have contributed to the BP did not increase during the infusion of intralipid and heparin
decline in obese hypertensives on the DASH-CD, previous among obese subjects consuming a DASH-CD, their BP rose
studies suggest that the magnitude of changes achieved is
similarly during acute hyperlipidemia on all 3 diets (Figure
modest and insufficient to explain the BP reduction ob-
2). These observations suggest that BP and oxidative stress,
served.37,38 To further complicate efforts to separate the effect
as measured by F2-isoprostanes, are not linked in the setting
of dietary minerals and antioxidant on BP, K⫹ and Mg2⫹ have
of acute hyperlipidemia. However, small-artery distensibility,
antioxidant effects.39,40
a measure of endothelial function,41 did not fall during acute
We reported that an infusion of intralipid and heparin
hyperlipidemia in obese hypertensives on DASH-CD, as it
raises BP and increases oxidative stress as measured by
did on the other 2 diets (Figure 3). Thus, the antioxidant
F2-isoprostanes in lean normotensives and obese hyperten-
sives.11,12 In the present study, lean normotensive and obese actions of the DASH-CD may have important vasculoprotec-
Downloaded from http://hyper.ahajournals.org/ by guest on February 3, 2015Lopes et al Dietary Antioxidants and Blood Pressure 429
tive effects, which could contribute to BP benefits over the 6. Baynes JW. Role of oxidative stress in development of complications
longer-term. in diabetes. Diabetes. 1991;40:405– 412.
7. Laight DW, Desai KM, Gopaul NK, Anggard EE, Carrier MJ.
In conclusion, the DASH-CD emerges as a clinically useful F2-isoprostane evidence of oxidant stress in the insulin resistant, obese
tool in the outpatient management of free-living obese sub- Zucker rat: effects of vitamin. E Eur J Pharmacol. 1999;377:89 –92.
jects with high-normal and stage 1 hypertension, as well as 8. Ceriello A. Oxidative stress and glycemic regulation. Metabolism.
2000;49:27–29.
other evidence for the cardiovascular risk factor cluster. Even 9. Egan BM, Greene EL, Goodfriend TL. Insulin resistance and cardio-
though dietary adherence is imperfect, acute lipid-induced vascular disease. Am J Hypertens. 2001;14:116S–125S.
oxidative stress and impairment of small-artery distensibility 10. El Midaoui A, de Champlain J. Prevention of hypertension, insulin
resistance, and oxidative stress by ␣-lipoic acid. Hypertension. 2002;
are blunted in obese hypertensives following a DASH-CD.
39:303–307.
Moreover, after 4 weeks on the DASH-CD, antioxidant 11. Stojiljkovic MP, Zhang D, Lopes HF, Lee CG, Goodfriend TL, Egan
capacity rises and BP falls in obese hypertensives but not in BM. Hemodynamic effects of lipids in humans. Am J Physiol. 2001;
lean normotensives. These findings suggest that the 280:R1674 –R1679.
12. Stojiljkovic MP LH, Zhang D, Morrow JD, Goodfriend TL, Egan BM.
DASH-CD lowers BP in obese hypertensive patients by Raising plasma fatty acids increases F2-isoprostanes in humans: impli-
improving the balance between antioxidant capacity and cations for the cardiovascular risk factor cluster. J Hypertens. 2002;
oxidative stress. 20:1–7.
13. Vaziri ND, Wang XQ, Oveisi F, Rad B. Induction of oxidative stress
by glutathione depletion causes severe hypertension in normal rats.
Perspective Hypertension. 2000;36:142–146.
A diverse and expanding literature implicates excessive 14. Laursen JB RS, Galis Z, Tarpey M, Freeman BA, Harrison DG. Role
oxidative stress and/or inadequate antioxidant defenses in the of superoxide in angiotensin II–induced but not catecholamine-
induced hypertension. Circulation. 1997;95:557–559.
pathogenesis of cardiovascular risk and disease. Given the 15. Duffy SJ GN, Holbrook M, Huang A, Frei B, Keaney JF Jr, Vita JA.
epidemic status of the metabolic syndrome, which is driven Treatment of hypertension with ascorbic acid. Lancet. 1999;354:
principally by unhealthy lifestyle patterns, this topic is 2048 –2049.
16. Appel LJ, Moore TJ, Obarzanek E, Vollmer WM, Svetkey LP, Sacks
assuming progressively greater importance. Unfortunately,
FM, Bray GA, Vogt TM, Cutler JA, Windhauser MM, Lin PH,
achieving healthy lifestyle patterns can be challenging. Karanja N. A clinical trial of the effects of dietary patterns on blood
Therefore, it would seem useful to define the minimum pressure: DASH Collaborative Research Group. N Engl J Med. 1997;
intervention required to elicit clinically relevant risk reduc- 336:1117–1124.
17. Harris JA, Benedict FG. A Biometric Study of Basal Metabolism in
tion. The present study indicates that the DASH-CD prescrip- Man. Washington DC: Carnegie Institute of Washington (Carnegie
tion in obese free-living individuals with evidence for the Institute of Washington publication No. 279), 1919.
metabolic syndrome, despite suboptimal adherence, results in 18. Trabulsi J, Schoeller DA. Evaluation of dietary assessment
instruments against doubly labeled water, a biomarker of habitual
clinically important improvements of BP. Further studies to energy intake. Am J Physiol. 281:E891–E899.
separate the effects of various foods from their mineral and 19. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF,
antioxidant content may help establish whether dietary Turner RC. Homeostasis model assessment: insulin resistance and
changes are required or if supplements are adequate to -cell function from fasting plasma glucose and insulin concentrations
in man. Diabetologia. 1985;28:412– 419.
achieve substantial risk reduction. 20. Besch W, Woltanski KP, Keilacker H, Diaz-Alonso JM, Schulz B,
Amendt P, Kohnert KD, Ziegler M. Measurement of insulin in human
Acknowledgments sera using a new RIA kit, 1: insulin determination in the absence of
insulin antibodies– conventional assay and micro modification. Exp
This research was supported by grants from the National Institutes of
Clin Endocrinol. 1987;90:264 –270.
Health (HL58794, HL04290, HL55782, GM42056, DK48831, 21. De Bruin TW, Brouwer CB, Gimpel JA, Erkelens DW. Postprandial
DK26657, and CA77839) and the General Clinical Research Center decrease in HDL cholesterol and HDL apo A-I in normal subjects in
(RR-01070). H.F.L. was supported by Fundação de Amparo à relation to triglyceride metabolism. Am J Physiol. 1991;260:
Pesquisa do Estado de São Paulo (FAPESP) and Fundação E. J. E492–E498.
Zerbini (InCor), São Paulo, Brazil. J.D.M. is the recipient of a 22. Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concen-
Burroughs Wellcome Fund Clinical Scientist Award in Translational tration of low-density lipoprotein cholesterol in plasma, without use of
Research. We thank Kim L. Seymour-Edwards for excellent admin- the preparative ultracentrifuge. Clin Chem. 1972;18:499 –502.
istrative assistance and Jackie P. Nguyen and the GCRC staff and 23. Benzie IF, Strain JJ. The ferric reducing ability of plasma (FRAP) as
nurses for their expert technical assistance. a measure of “antioxidant power”: the FRAP assay. Anal Biochem.
1996;239:70 –76.
24. Morrow JD, Hill KE, Burk RF, Nammour TM, Badr KF, Roberts LJ
References 2nd. A series of prostaglandin F2–like compounds are produced in
1. Landsberg L, Krieger DR. Obesity, metabolism, and the sympathetic vivo in humans by a non-cyclooxygenase, free radical-catalyzed
nervous system. Am J Hypertens. 1989;2:125S–132S. mechanism. Proc Natl Acad Sci U S A. 1990;87:9383–9387.
2. Rajagopalan S, Kurz S, Munzel T, Tarpey M, Freeman BA, Griendling 25. Awad JA, Morrow JD, Takahashi K, Roberts LJ 2nd. Identification of
KK, Harrison DG. Angiotensin II–mediated hypertension in the rat non-cyclooxygenase-derived prostanoid (F2-isoprostane) metabolites
increases vascular superoxide production via membrane in human urine and plasma. J Biol Chem. 1993;268:4161– 4169.
NADH/NADPH oxidase activation: contribution to alterations of 26. Haas JA, Krier JD, Bolterman RJ, Juncos LA, Romero JC. Low-dose
vasomotor tone. J Clin Invest. 1996;97:1916 –1923. angiotensin II increases free isoprostane levels in plasma. Hyper-
3. Lee KU. Oxidative stress markers in Korean subjects with insulin tension. 1999;34:983–986.
resistance syndrome. Diabetes Res Clin Pract. 2001;54(suppl 27. Kunsch C, Medford RM. Oxidative stress as a regulator of gene
2):S29 –S33. expression in the vasculature. Circ Res. 1999;85:753–766.
4. Facchini FS, Hua NW, Reaven GM, Stoohs RA. Hyperinsulinemia: 28. Therond P, Bonnefont-Rousselot D, Davit-Spraul A, Conti M,
the missing link among oxidative stress and age-related diseases? Free Legrand A. Biomarkers of oxidative stress: an analytical approach.
Radic Biol Med. 2000;29:1302–1306. Curr Opin Clin Nutr Metab Care. 2000;3:373–384.
5. Giugliano D, Ceriello A, Paolisso G. Oxidative stress and diabetic 29. Tse WY, Maxwell SR, Thomason H, Blann A, Thorpe GH, Waite M,
vascular complications. Diabetes Care. 1996;19:257–267. Holder R. Antioxidant status in controlled and uncontrolled hyper-
Downloaded from http://hyper.ahajournals.org/ by guest on February 3, 2015430 Hypertension March 2003
tension and its relationship to endothelial damage. J Hum Hypertens. restriction reduced blood pressure in obesity hypertensives by
1994;8:843– 849. improvement of autonomic nerve activity and insulin sensitivity.
30. Block G, Mangels AR, Norkus EP, Patterson BH, Levander OA, J Cardiovasc Pharmacol. 2001;38(suppl 1):S69 –S74.
Taylor PR. Ascorbic acid status and subsequent diastolic and systolic 36. Despres JP, Lamarche B, Mauriege P, Cantin B, Dagenais GR,
blood pressure. Hypertension. 2001;37:261–267. Moorjani S, Lupien PJ. Hyperinsulinemia as an independent risk
31. Chen X, Touyz RM, Park JB, Schiffrin EL. Antioxidant effects of factor for ischemic heart disease. N Engl J Med. 1996;334:952–957.
vitamins C and E are associated with altered activation of vascular 37. Sacks FM, Brown LE, Appel L, Borhani NO, Evans D, Whelton P.
NADPH oxidase and superoxide dismutase in stroke-prone SHR. Combinations of potassium, calcium, and magnesium supplements in
Hypertension. 2001;38:606 – 611. hypertension. Hypertension. 1995;26:950 –956.
32. Garrison RJ, Kannel WB, Stokes J 3rd, Castelli WP. Incidence and 38. Gu D, He J, Wu X, Duan X, Whelton PK. Effect of potassium
precursors of hypertension in young adults: the Framingham Off- supplementation on blood pressure in Chinese: a randomized,
spring Study. Prev Med. 1987;16:235–251. placebo-controlled trial. J Hypertens. 2001;19:1325–1331.
33. Ferrannini E, Natali A, Bell P, Cavallo-Perin P, Lalic N, Mingrone G. 39. Ishimitsu T, Tobian L, Sugimoto K, Everson T. High potassium diets
Insulin resistance and hypersecretion in obesity: European Group for reduce vascular and plasma lipid peroxides in stroke-prone sponta-
the Study of Insulin Resistance (EGIR). J Clin Invest. 1997;100: neously hypertensive rats. Clin Exp Hypertens. 1996;18:659 – 673.
1166 –1173. 40. Maulik D, Zanelli S, Numagami Y, Ohnishi ST, Mishra OP,
34. Reisin E, Abel R, Modan M, Silverberg DS, Eliahou HE, Modan B. Delivoria-Papadopoulos M. Oxygen free radical generation during
Effect of weight loss without salt restriction on the reduction of blood in-utero hypoxia in the fetal guinea pig brain: the effects of maturity
pressure in overweight hypertensive patients. N Engl J Med. 1978; and of magnesium sulfate administration. Brain Res. 1999;817:
298:1– 6. 117–122.
35. Nakano Y, Oshima T, Sasaki S, Higashi Y, Ozono R, Takenaka S, 41. Cohn JN. New approaches to screening for vascular and cardiac risk.
Miura F, Hirao H, Matsuura H, Chayama K, Kambe M. Calorie Am J Hypertens. 2001;14:218S–220S.
Downloaded from http://hyper.ahajournals.org/ by guest on February 3, 2015DASH Diet Lowers Blood Pressure and Lipid-Induced Oxidative Stress in Obesity
Heno F. Lopes, Kelley L. Martin, Khaled Nashar, Jason D. Morrow, Theodore L. Goodfriend
and Brent M. Egan
Hypertension. 2003;41:422-430; originally published online February 3, 2003;
doi: 10.1161/01.HYP.0000053450.19998.11
Hypertension is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 2003 American Heart Association, Inc. All rights reserved.
Print ISSN: 0194-911X. Online ISSN: 1524-4563
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://hyper.ahajournals.org/content/41/3/422
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Hypertension can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial
Office. Once the online version of the published article for which permission is being requested is located,
click Request Permissions in the middle column of the Web page under Services. Further information about
this process is available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Hypertension is online at:
http://hyper.ahajournals.org//subscriptions/
Downloaded from http://hyper.ahajournals.org/ by guest on February 3, 2015You can also read
