Sie sind auf Seite 1von 22

Epidemiology/Population

Healthy Lifestyle Factors and Risk of Cardiovascular Events


and Mortality in Treatment-Resistant Hypertension
The Reasons for Geographic and Racial Differences in Stroke Study
Keith M. Diaz, John N. Booth III, David A. Calhoun, Marguerite R. Irvin, George Howard,
Monika M. Safford, Paul Muntner, Daichi Shimbo
See Editorial Commentary, pp 459460
AbstractFew data exist on whether healthy lifestyle factors are associated with better prognosis among individuals with
apparent treatment-resistant hypertension, a high-risk phenotype of hypertension. The purpose of this study was to assess the
association of healthy lifestyle factors with cardiovascular events, all-cause mortality, and cardiovascular mortality among
individuals with apparent treatment-resistant hypertension. We studied participants (n=2043) from the population-based
Reasons for Geographic and Racial Differences in Stroke (REGARDS) study with apparent treatment-resistant hypertension
(blood pressure 140/90 mmHg despite the use of 3 antihypertensive medication classes or the use of 4 classes of
antihypertensive medication regardless of blood pressure control). Six healthy lifestyle factors adapted from guidelines for the
management of hypertension (normal waist circumference, physical activity 4 times/week, nonsmoking, moderate alcohol
consumption, high Dietary Approaches to Stop Hypertension diet score, and low sodium-to-potassium intake ratio) were
examined. A greater number of healthy lifestyle factors were associated with lower risk for cardiovascular events (n=360)
during a mean follow-up of 4.5 years. Multivariable-adjusted hazard ratios [HR (95% confidence interval)] for cardiovascular
events comparing individuals with 2, 3, and 4 to 6 versus 0 to 1 healthy lifestyle factors were 0.91 (0.681.21), 0.80 (0.57
1.14), and 0.63 (0.410.95), respectively (P-trend=0.020). Physical activity and nonsmoking were individual healthy lifestyle
factors significantly associated with lower risk for cardiovascular events. Similar associations were observed between healthy
lifestyle factors and risk for all-cause and cardiovascular mortality. In conclusion, healthy lifestyle factors, particularly
physical activity and nonsmoking, are associated with a lower risk for cardiovascular events and mortality among individuals
with apparent treatment-resistant hypertension.(Hypertension. 2014;64:465-471.) Online Data Supplement

Key Words: diet

hypertension

espite the use of antihypertensive medications, inadequately controlled blood pressure (BP) remains a challenge
in the management of hypertension for many patients. In 2008,
the American Heart Association published a scientific statement
on a subclass of hypertensive patients considered to have treatment-resistant hypertension. In this statement, treatment-resistant hypertension was defined as uncontrolled BP despite the use
of antihypertensive medications from 3 or more classes, or the
use of 4 or more classes to achieve BP control.1 The prevalence
of apparent treatment-resistant hypertension (aTRH) in the 2005
to 2008 National Health and Nutrition Examination Survey was
estimated to be 11.8% among the US adults with hypertension,
an increase from 5.5% in 19981994 and 8.5% in 1999 to 20042;
indicating that a growing proportion of the US adults with hypertension are resistant to antihypertensive medication regimens.

lifestyle

smoking

Data from the Cardiovascular Research Network has shown


that patients with aTRH had an almost 50% higher risk for
cardiovascular events compared with patients whose BP had
been controlled on 3 medications.3 Other longitudinal studies
have yielded similar results for cardiovascular outcomes and
all-cause mortality.4,5 Taken together, available data suggest
that individuals with aTRH are at high risk for adverse cardiovascular events and mortality, highlighting a need for efforts
toward improving outcomes in this population.
Guidelines endorse lifestyle changes, including weight
loss, regular exercise, smoking cessation, moderation of alcohol consumption, and a high-fiber, low-fat, and low-salt diet,
for individuals with hypertension.6,7 However, few data exist
on whether these lifestyle factors, either individually or combined as part of an overall healthy lifestyle, are associated

Received March 18, 2014; first decision March 29, 2014; revision accepted April 29, 2014.
From the Department of Medicine, Columbia University Medical Center, New York, NY (K.M.D., D.S.); and Departments of Epidemiology (J.N.B.,
M.R.I., P.M.), Biostatistics (G.H.), and Medicine (D.A.C., M.M.S.), University of Alabama at Birmingham.
The online-only Data Supplement is available with this article at http://hyper.ahajournals.org/lookup/suppl/doi:10.1161/HYPERTENSIONAHA.
114.03565/-/DC1.
Correspondence to Keith M. Diaz, Department of Medicine, Columbia University Medical Center, 622 West 168th St, PH9-319, New York, NY 10032.
E-mail kd2442@columbia.edu
2014 American Heart Association, Inc.
Hypertension is available at http://hyper.ahajournals.org

DOI: 10.1161/HYPERTENSIONAHA.114.03565

Downloaded from http://hyper.ahajournals.org/


by guest on November 22, 2015
465

466HypertensionSeptember 2014
with better prognosis among individuals with aTRH. If lifestyle factors are associated with better outcomes in aTRH, it
will highlight the need to invest more resources in developing
and testing behavioral interventions to mitigate their elevated
risk. The purpose of this study, therefore, was to investigate
the associations of healthy lifestyle factors with cardiovascular events, all-cause mortality, and cardiovascular mortality
among individuals with aTRH in the Reasons for Geographic
and Racial Differences in Stroke (REGARDS) study.

Methods
Study Population
The REGARDS study has been described previously.8 Briefly,
REGARDS is a population-based cohort study of 30239 white and
black adults 45 years of age from across the contiguous United
States who were enrolled between 2003 and 2007. The current
analysis was restricted to individuals with aTRH (defined below).
After excluding participants without BP measurements or data on
antihypertensive medications at baseline, without aTRH, and who
were missing outcome data during follow-up, 2043 individuals with
aTRH were available for analyses (Figure S1 in the online-only Data
Supplement). Characteristics of participants included and excluded
from analyses are presented in Table S1. The REGARDS study protocol was approved by Institutional Review Boards at participating
centers. All participants provided informed consent.

Data Collected at Baseline


Data were collected via a telephone interview, self-administered questionnaires, and an in-home examination. The in-home examination was conducted by trained health professionals during a single visit and included
anthropometrics, BP measurements, ECG, collection of blood and urine
samples, and review of medication pill bottles. A detailed summary of
baseline measures are provided in the online-only Data Supplement.

BP Measurement, Medication Use, and aTRH


Definition
BP was measured during the in-home examination using a standardized protocol.8 BP was measured 2 times by a trained examiner using
an aneroid sphygmomanometer after 5 minutes of seated rest. Based
on the average of 2 measurements, uncontrolled hypertension was
defined as systolic BP 140 mmHg or diastolic BP 90 mmHg.
During the in-home examination, medications taken in the past 2
weeks were recorded and subsequently coded into drug classes.6
Single-pill combinations were classified into their respective classes
as individual components. aTRH was defined as: (1) taking 3 or more
antihypertensive medication classes with uncontrolled BP; or (2) taking 4 or more classes regardless of BP control.

Healthy Lifestyle Factors


Six lifestyle factors were evaluated: waist circumference, physical
activity, cigarette smoking, alcohol consumption, Dietary Approaches
to Stop Hypertension (DASH) diet score, and sodium-to-potassium
(Na/K) intake. Normal waist circumference, physical activity 4
days/week, nonsmoking status, moderate alcohol consumption, high
DASH diet score, and low Na/K intake were considered to be healthy
lifestyle factors. The 6 healthy lifestyle factors were adapted from
lifestyle modifications recommended for the treatment of hypertension.6,7 Moderate alcohol consumption was considered to be a
healthy lifestyle factor as prior studies have shown reduced cardiovascular risk associated with moderate versus heavy or no alcohol
consumption.9
Consistent with World Health Organization recommendations,
waist circumference was measured using a tape measure midway between the lowest rib and iliac crest with the participant standing.10
Waist circumference was dichotomized as normal (102 cm in men;
88 cm in women) or abdominal obesity (>102 cm in men; >88 cm
in women).11 Physical activity was assessed during the telephone

interview using the question: How many times per week do you engage in intense physical activity, enough to work-up a sweat? with
response options of none, 1 to 3 times/week, and 4 or more times/
week. Smoking status was determined by responses to 2 questions
during the telephone interview: Have you smoked at least 100 cigarettes in your lifetime? and Do you smoke cigarettes now, even
occasionally? Current smoking was defined as a positive response to
both questions. Self-reported alcohol consumption, assessed during
the telephone interview, was categorized as none (no weekly alcohol
consumption), moderate (114 and 17 alcoholic beverages/week for
men and women, respectively), or heavy (>14 and >7 alcoholic beverages/week for men and women, respectively). Participants completed
a self-administered Block Food Frequency Questionnaire following
the in-home study visit to estimate average dietary intake for the previous year. Nutrient analysis was conducted by NutritionQuest. A
DASH dietary score was created using methods described by Fung
et al.12 Using the distribution of DASH dietary scores from participants with aTRH, a high DASH diet score was defined as being in
the highest quartile (27). Similar methods were used to dichotomize
Na/K intake, with low Na/K intake defined as being in the lowest
quartile (0.71). Information on the validation and reproducibility
of the physical activity questionnaire and Block Food Frequency
Questionnaire are provided in the online-only Data Supplement.

Outcomes
The primary outcome was combined fatal and nonfatal cardiovascular
events. Fatal cardiovascular events were defined as death within 28
days of a definite or probable myocardial infarction, or sudden death;
or death within 28 days of a confirmed stroke. Nonfatal cardiovascular events were defined as nonfatal definite or probable myocardial
infarction or stroke. All-cause mortality and cardiovascular mortality
were secondary and tertiary outcomes, respectively. Vital status and
cardiovascular events were ascertained during biannual telephone follow-up interviews of the participant, proxy, or next-of-kin. Report of
a potential event triggered medical record retrieval followed by expert
adjudication. A detailed description of the identification and adjudication of outcomes is provided in the online-only Data Supplement.

Statistical Analysis
We accounted for missing data in the lifestyle factors (waist circumference: n=11; physical activity: n=33; smoking: n=4; alcohol use: n=43;
DASH diet score and Na/K intake: n=699) using multiple imputation.
Almost all of the data missing for the Food Frequency Questionnairederived variables (DASH diet score and Na/K intake) were because of
participants not returning the questionnaire. Missing data were imputed
with 10 data sets using chained equations.13 For primary analyses, participants were grouped according to the number of healthy lifestyle factors: 0 to 1, 2, 3, or 4 to 6. Participant characteristics were calculated
by the number of healthy lifestyle factors. Cox proportional-hazards
regression models were then used to calculate the hazard ratio (HR) for
cardiovascular events associated with the number of healthy lifestyle
factors (2, 3, or 46) in comparison to participants with 0 to 1 healthy
lifestyle factors. Crude HRs were initially calculated. Subsequently,
HRs were calculated with adjustment for age, race, sex, education,
and geographic region of residence (Model 1) and further adjustment
for total cholesterol, high-density lipoprotein (HDL) cholesterol, estimated glomerular filtration rate <60 mL/min/1.73 m2, albuminuria,
atrial fibrillation, left ventricular hypertrophy on ECG, diabetes mellitus, statin use, history of coronary heart disease, and history of stroke
(Model 2). P-trend tests were conducted by including the number of
lifestyle factors for each participant as an ordinal variable in regression models. Analyses were then repeated in a fully adjusted model
testing interactions for race (black vs white) and history of stroke or
coronary heart disease (yes vs no). Also, HRs were calculated in subgroups defined by race and history of stroke or coronary heart disease.
Three sensitivity analyses were conducted. First, analyses were
repeated using participants with complete data (ie, without imputation; n=1300). Second, as the use of a diuretic has been suggested
to be required for diagnosis of aTRH,1 we repeated analyses restricting the study population to participants on a diuretic (n=1809).

Downloaded from http://hyper.ahajournals.org/ by guest on November 22, 2015

Diaz et al Lifestyle Factors in Resistant Hypertension 467


Finally, as aTRH may be partially explained by reduced medication
adherence,14 we restricted the study population to participants who
reported a high level of medication adherence (n=1364), defined as
a Morisky Medication Adherence Scale score of 0 (see online-only
Data Supplement).
To evaluate the association of each individual healthy lifestyle factor with cardiovascular events, Cox proportional-hazards regression
models were repeated, testing each lifestyle factor separately. P-trend
tests were conducted for lifestyle factors with more than 2 levels
that we hypothesized would have a linear association with outcomes
(physical activity, DASH diet score, Na/K intake).
The association between the number of healthy lifestyle factors
and all-cause mortality, a secondary outcome, was examined. This
association was also examined in a sensitivity analysis without imputation. Next, the association of each individual healthy lifestyle factor with all-cause mortality was tested. Finally, associations between
the number of healthy lifestyle factors and, separately, each healthy
lifestyle factor and cardiovascular mortality was examined. Data
analyses were conducted using STATA/IC version 12.1 (StataCorp,
College Station, TX).

Results
Participant Characteristics
Among the 2043 participants with aTRH, 662 (32.4%) had
0 to 1 healthy lifestyle factors, 671 (32.8%) had 2 healthy
lifestyle factors, 430 (21.0%) had 3 healthy lifestyle factors,
and 280 (13.7%) had 4 to 6 healthy lifestyle factors. Overall,
658 participants (32.2%) did not have abdominal obesity, 453
(22.2%) engaged in physical activity 4 times/week, 1797
(88.0%) were nonsmokers, and 539 (26.4%) were moderate

alcohol drinkers. The mean DASH diet score and Na/K intake
ratio were 23.70.1 and 0.920.01, respectively. Participants
with more healthy lifestyle factors were, on average, older,
more likely to be male, white, and graduate high school, less
likely to have diabetes mellitus, estimated glomerular filtration rate <60 mL/min/1.73 m2, and albuminuria, were taking
fewer classes of antihypertensive medications, and had lower
total cholesterol and diastolic BP (Table1).

Cardiovascular Events
There were 360 (17.6%) cardiovascular events during a mean
follow-up of 4.5 years (maximum: 7.8 years). In unadjusted
and adjusted models, a greater number of healthy lifestyle factors were associated with a lower risk for cardiovascular events
(Table2). The association between the number of healthy
lifestyle factors and risk of cardiovascular events did not vary
by race (p-interaction=0.934) or history of stroke or coronary
heart disease (p-interaction=0.628; Tables S2 and S3).
In a sensitivity analysis restricted to participants with complete data, a greater number of healthy lifestyle factors were
associated with a lower risk of cardiovascular events (Table
S4). In a second sensitivity analysis excluding participants not
taking a diuretic, a greater number of healthy lifestyle factors were also associated with a lower risk of cardiovascular
events; the fully adjusted HR (95% confidence interval) comparing participants with 2, 3, and 4 to 6 versus 0 to 1 healthy
lifestyle factors were 0.92 (0.671.26), 0.80 (0.551.16), and

Table 1. Characteristics of Reasons for Geographic and Racial Differences in Stroke (REGARDS) Study
Participants (n=2043) With Apparent Treatment-Resistant Hypertension Stratified by the Number of Healthy
Lifestyle Factors
Number of Healthy Lifestyle Factors
Variable
Age, y
Male, %

01 (n=662)

2 (n=671)

3 (n=430)

46 (n=280)

P-Trend

66.10.4

67.20.4

68.70.4

70.30.5

<0.001

38.3

50.4

56.1

59.5

<0.001

Black, %

69.7

63.7

52.9

42.5

<0.001

Education < high school, %

24.5

19.7

17.3

8.7

<0.001

Non-belt and nonbuckle, %

35.1

36.8

33.7

30.9

1 (ref)

Residence in stroke belt, %

22.8

18.6

22.0

18.6

0.197

Residence in stroke buckle, %

42.1

44.6

44.3

50.5

0.086

Diabetes mellitus, %

53.6

47.5

39.8

32.1

<0.001

Total cholesterol, mg/dL

182.51.8

181.41.7

178.12.0

176.22.7

HDL cholesterol, mg/dL

Region of residence

0.031

48.00.7

47.80.6

48.10.8

50.21.0

0.071

eGFR < 60 mL/min/1.73 m2 (%)

29.1

29.4

26.2

22.3

<0.001

Albuminuria, %

36.3

35.4

29.6

28.9

<0.001

Atrial fibrillation, %

17.0

15.0

14.0

15.1

0.369

Left ventricular hypertrophy, %

21.7

17.0

16.3

17.7

0.128

Systolic BP, mmHg

142.20.8

141.70.8

141.11.1

143.31.1

0.606

Diastolic BP, mmHg

80.00.5

79.60.5

79.20.6

75.00.1

0.048

No. of antihypertensive medications

3.70.0

3.60.0

3.70.0

3.60.0

0.028

Statin use, %

51.9

48.9

52.5

59.5

0.061

History of CHD, %

35.2

35.3

37.0

38.7

0.508

History of stroke, %

15.5

14.6

11.5

9.9

0.557

Data are presented as meanSE or percentage. BP indicates blood pressure; CHD, coronary heart disease; eGFR, estimated glomerular
filtration rate; and HDL, high-density lipoprotein.

Downloaded from http://hyper.ahajournals.org/ by guest on November 22, 2015

468HypertensionSeptember 2014
Table 2. Hazard Ratios for Cardiovascular Events Associated With Cumulative Number of Healthy
Lifestyle Factors in Reasons for Geographic and Racial Differences in Stroke (REGARDS) Study
Participants With Apparent Treatment-Resistant Hypertension
Number of Healthy
Lifestyle Factors

No. of Events

Person Years
at Risk

01

122

2869.3

Hazard Ratio (95% Confidence Interval) for Cardiovascular Events


Unadjusted
1 (ref)

Model 1
1 (ref)

Model 2
1 (ref)

128

3093.5

0.98 (0.741.29)

0.87 (0.661.15)

0.91 (0.681.21)

71

2001.2

0.84 (0.591.18)

0.69 (0.490.98)

0.80 (0.571.14)

46

39

1316.8

0.70 (0.471.03)

0.54 (0.360.80)

0.63 (0.410.95)

P-trend=0.042

P-trend=0.001

P-trend=0.020

Healthy lifestyle factors: waist circumference<102 cm in men and <88 cm in women, physical activity4 times/week, nonsmoking,
moderate alcohol consumption, highest quartile (fourth) for Dietary Approaches to Stop Hypertension diet score, lowest (first) quartile for
Na/K intake ratio. Model 1: adjusted for age, race, sex, education, and geographic region of residence. Model 2: adjusted for covariates in
Model 1 plus total cholesterol, high-density lipoprotein cholesterol, estimated glomerular filtration rate < 60 mL/min/1.73 m2, albuminuria,
atrial fibrillation, left ventricular hypertrophy, diabetes mellitus, statin use, history of coronary heart disease, and history of stroke.

0.59 (0.370.92), respectively (P-trend=0.017). In a third


sensitivity analysis, associations were similar after restricting
the sample to participants who had a high level of medication
adherence; the fully adjusted HR (95% confidence interval)
for cardiovascular events comparing participants with 2, 3,
and 4 to 6 versus 0 to 1 healthy lifestyle factors were 0.88
(0.621.26), 0.71 (0.471.09), and 0.54 (0.330.90), respectively (P-trend=0.010).
When the association of each healthy lifestyle factor with
cardiovascular events was examined, higher levels of physical
activity and nonsmoking status were significantly associated
with a reduced risk for cardiovascular events in unadjusted
and adjusted models (Table3). No other healthy lifestyle factor was associated with risk of cardiovascular events.

All-Cause Mortality
There were 452 (22.1%) deaths during a mean follow-up of
5.4 years (maximum: 9.0 years). A greater number of healthy
lifestyle factors were associated with a lower risk for all-cause
mortality (Table4). This association remained significant in
a sensitivity analysis restricted to participants with complete
data (Table S5). When each healthy lifestyle factor was examined, physical activity, nonsmoking status, and a high DASH
diet score were significantly associated with a lower risk for
all-cause mortality in adjusted models (Table S6).

Cardiovascular Mortality
There were 174 (8.5%) cardiovascular deaths during follow-up.
A greater number of healthy lifestyle factors were significantly
associated with a lower risk for cardiovascular mortality (Table
S7). Higher levels of physical activity and nonsmoking status
were each associated with a significantly lower risk for cardiovascular mortality in unadjusted and adjusted models (Table S8).

Discussion
In this population-based study of individuals with aTRH, having more healthy lifestyle factors was associated with a lower
risk for cardiovascular events and all-cause and cardiovascular mortality. Furthermore, physical activity and nonsmoking
were individual healthy lifestyle factors associated with a 33%
and 46% lower risk for cardiovascular events, 42% and 47%
lower risk for all-cause mortality, and 50% and 53% lower risk

for cardiovascular mortality, respectively. Finally, a higher


DASH diet score was associated with a 28% lower risk for
all-cause mortality. To our knowledge, this is the first study to
examine the association of healthy lifestyle factors with outcomes among individuals with aTRH.
As the prevalence of aTRH is expected to increase,1 effective treatments to improve outcomes among individuals with
aTRH are needed. Treatment methods being explored for the
management of aTRH include invasive, irreversible procedures
or implantable devices such as renal denervation and carotid
baroreceptor stimulation. However, it is important to determine
the efficacy of less invasive approaches to spare individuals the
inconvenience and possible complications that come from these
procedures. Hypertension guidelines generally recommend
lifestyle modification, including weight reduction, increasing
physical activity, moderation of alcohol consumption, adoption
of the DASH diet, dietary salt reduction, and smoking cessation, as adjunctive therapy to antihypertensive medication.6,7
These recommendations, in part, stem from studies which have
demonstrated an association between lifestyle factors and morbidity/mortality among hypertensive individuals.15,16 However,
it was previously unknown whether lifestyle factors are associated with better prognosis specifically in individuals with
aTRH. In our study of individuals with aTRH, having a greater
number of healthy lifestyle factors were associated with a lower
risk for cardiovascular events and mortality. These findings suggest that lifestyle interventions may be beneficial for reducing
morbidity and mortality risk in individuals with aTRH. As there
was a graded inverse association between the number of healthy
lifestyle factors and risk for cardiovascular events and mortality, there may be incremental benefits to increasing the number of healthy lifestyle factors among individuals with aTRH.
Moreover, only a small percentage of individuals had 4 or more
healthy lifestyle factors (13.7%); a finding that corresponds to
the low percentage of individuals with multiple healthy lifestyle factors reported for general population-based samples
and populations with hypertension.1719 The low prevalence of
healthy lifestyle factors in this study highlights a great potential
to reduce the increased morbidity and mortality risk associated
with aTRH through a multifaceted lifestyle intervention.
In our study, higher levels of physical activity were associated with a lower risk of cardiovascular events, all-cause

Downloaded from http://hyper.ahajournals.org/ by guest on November 22, 2015

Diaz et al Lifestyle Factors in Resistant Hypertension 469


Table 3. Hazard Ratios for Cardiovascular Events Associated With Individual Lifestyle Factors in Reasons for Geographic
and Racial Differences in Stroke (REGARDS) Study Participants With Apparent Treatment-Resistant Hypertension
Hazard Ratio (95% Confidence Interval) for
Cardiovascular Events
Lifestyle Factor

No. of Events

Person Years at Risk

Unadjusted

Model 1

Model 2

Yes

224

6294.9

No

136

2985.9

None

186

3812.6

13 times/wk

113

3280.9

0.72 (0.570.91)

0.70 (0.560.89)

0.82 (0.651.04)


4 times/wk

61

2187.4

0.58 (0.430.77)

0.55 (0.400.74)

0.67 (0.500.91)

P-trend <0.001

P-trend <0.001

P-trend=0.006

Abdominal obesity
1 (ref)
1.28 (1.041.59)

1 (ref)
1.05 (0.831.32)

1 (ref)
1.12 (0.891.42)

Physical activity
1 (ref)

1 (ref)

1 (ref)

Current smoking
Yes

66

1016.2

No

294

8264.6

285.6

1 (ref)
0.55 (0.420.71)

1 (ref)
0.47 (0.350.61)

1 (ref)
0.54 (0.410.72)

Alcohol consumption*
Heavy
Moderate

1 (ref)

1 (ref)

1 (ref)

90

2624.6

0.88 (0.531.45)

0.90 (0.551.48)

0.94 (0.571.57)

263

6364.6

1.06 (0.661.69)

1.15 (0.711.87)

1.10 (0.671.80)

Quartile 1 (20)

82

2386.7

Quartile 2 (>2023)

96

2420.1

1.16 (0.821.64)

1.06 (0.761.48)

1.07 (0.771.48)

Quartile 3 (>2326)

89

2126.7

1.20 (0.861.69)

1.09 (0.781.53)

1.08 (0.771.52)

Quartile 4 (>26)

93

2347.3

1.14 (0.821.60)

1.03 (0.731.43)

1.06 (0.751.50)

P-trend=0.423

P-trend=0.864

P-trend=0.746

None
DASH diet score

1 (ref)

1 (ref)

1 (ref)

Na/K intake
Quartile 4 (0.71)

98

2461.6

Quartile 3 (>0.710.88)

103

532.6

1.08 (0.801.47)

1 (ref)

1.01 (0.741.37)

1 (ref)

1.01 (0.731.38)

1 (ref)

Quartile 2 (>0.881.08)

83

2310.3

0.90 (0.621.29)

0.83 (0.571.20)

0.86 (0.581.27)

Quartile 1 (>1.08)

76

2125.9

0.89 (0.641.24)

0.82 (0.581.16)

0.82 (0.571.18)

P-trend=0.344

P-trend=0.171

P-trend=0.221

Model 1: adjusted for age, race, sex, education, and geographic region of residence. Model 2: adjusted for covariates in Model 1 plus total
cholesterol, high-density lipoprotein cholesterol, estimated glomerular filtration rate <60 mL/min/1.73 m2, albuminuria, atrial fibrillation, left
ventricular hypertrophy, diabetes mellitus, statin use, history of coronary heart disease, and history of stroke. DASH indicates Dietary Approaches
to Stop Hypertension.
*P-trend values for alcohol consumption are not presented as we hypothesized that the association between alcohol consumption and outcomes
would not be linear given that moderate alcohol consumption (the middle category) was considered to be a healthy lifestyle factor in our analyses.

mortality, and cardiovascular mortality. This finding may


provide support for implementation of regular exercise or
increased physical activity as a treatment modality for aTRH.
Notably, there was a linear association between higher levels
of physical activity and lower risk of clinical outcomes, suggestive that even modest amounts of physical activity may be
beneficial for individuals with aTRH. Mechanisms underlying
this association, however, are unknown. One possible explanation is the effect of physical activity on BP. Recently, in a
study of 50 participants with aTRH, Dimeo et al showed that 8
to 12 weeks of aerobic exercise reduced daytime systolic and
diastolic ambulatory BP by 5.9 and 3.3 mmHg, respectively.20
Alternatively, the association between physical activity and
outcomes in aTRH could also be attributed to the non-BP
effects of increased physical activity (eg, insulin sensitivity,
lipid metabolism, endothelial function, immune function).21

Our study also found that nonsmoking was associated with


a lower risk of cardiovascular events, all-cause mortality, and
cardiovascular mortality in individuals with aTRH. Previous
findings from observational studies have suggested that smoking cessation may have a larger effect on reducing the risk of
morbidity and mortality than any other intervention or treatment.22 Our findings remain consistent with previous findings
and reinforce the role of cigarette smoking as a major health
hazard that is pertinent even to individuals with aTRH.
Adoption of the DASH diet, a diet rich in fruits, vegetables,
low-fat dairy products, and low in saturated and total fat, has
been recommended as an important lifestyle modification to
lower morbidity and mortality risk among individuals with
hypertension.6,7 In this study, a high DASH diet score was
associated with a lower risk for all-cause mortality; suggestive
that adoption of a DASH-like diet may reduce mortality risk

Downloaded from http://hyper.ahajournals.org/ by guest on November 22, 2015

470HypertensionSeptember 2014

Table 4. Hazard Ratios for All-Cause Mortality Associated With Cumulative Number of Healthy Lifestyle Factors in
Reasons for Geographic and Racial Differences in Stroke (REGARDS) Study Participants With Apparent TreatmentResistant Hypertension
Number of Healthy
Lifestyle Factors

Hazard Ratio (95% Confidence Interval) for All-Cause Mortality


No. of Deaths

Person Years at Risk

Unadjusted

Model 1

Model 2

01

161

3364.8

1 (ref)

1 (ref)

1 (ref)

146

3646.1

0.82 (0.641.04)

0.71 (0.550.91)

0.71 (0.550.93)

96

2361.3

0.82 (0.621.10)

0.63 (0.470.86)

0.70 (0.520.94)

46

49

1593.5

0.62 (0.440.88)

0.46 (0.330.66)

0.54 (0.370.77)

P-trend=0.009

P-trend<0.001

P-trend=0.001

Healthy lifestyle factors: waist circumference <102 cm in men and <88 cm in women, physical activity 4 times/week, nonsmoking, moderate
alcohol consumption, highest quartile (fourth) for Dietary Approaches to Stop Hypertension diet score, lowest (first) quartile for Na/K intake ratio.
Model 1: adjusted for age, race, sex, education, and geographic region of residence.
Model 2: adjusted for covariates in Model 1 plus total cholesterol, high-density lipoprotein cholesterol, estimated glomerular filtration rate <60 mL/
min/1.73 m2, albuminuria, atrial fibrillation, left ventricular hypertrophy, diabetes, statin use, history of coronary heart disease, and history of stroke.

for individuals with aTRH. However, caution is warranted in


interpreting this finding as the association was not significant
when risks for cardiovascular events or cardiovascular mortality were assessed. Results from previous observational studies
are conflicting. Some have reported a reduced risk of fatal and
nonfatal cardiovascular events with consumption of a DASHlike diet,12 whereas other studies have reported a reduced risk
for all-cause mortality, but not cardiovascular mortality.15
Reasons for the lack of association between consumption
of a DASH-like diet and risk for cardiovascular-related outcomes are unclear, but have been attributed to several factors
related to macronutrient composition of the DASH diet and
blood lipid changes that occur with adoption of the DASH diet
including: a reduction in HDL cholesterol, higher consumption of carbohydrates, and less consumption of mono- and
polyunsaturated fats.15
It has been previously reported in a study of 12 participants
with aTRH that a low-salt diet compared with high-salt diet
reduced systolic and diastolic BP by 22.7 and 9.1 mmHg,
respectively.23 These striking BP reductions have led some
to suggest that emphasis be placed on sodium reduction to
control BP and reduce cardiovascular risk in individuals with
aTRH.24,25 In this study, however, the ratio of sodium-to-potassium intake was not associated with a significant reduction in
clinical outcomes. Rigorous studies are needed to determine
whether dietary sodium reduction and the concomitant BP
reductions it induces can translate to mitigating morbidity/
mortality risk in aTRH.
Several limitations must be noted when interpreting our
findings. First, medication dosing is not available in the
REGARDS study. Thus, we were unable to confirm optimal
dosing of antihypertensive medications. Second, BP levels
were defined by readings during a single visit. Third, physical activity, cigarette smoking, dietary measures, and alcohol
use were assessed via self-report. Fourth, Food Frequency
Questionnaire data to derive the DASH diet score and ratio of
sodium-to-potassium intake were not available on 699 (34.2%)
of 2043 participants. Nonetheless, dietary data were available
for 1344 participants, which is still a sizable cohort. Also,
results were similar when analyses were conducted among
participants with complete data. Fifth, it is possible that some

individuals with a history of aTRH undertook several lifestyle


modifications, lowered their BP, and were no longer classified as having aTRH at the REGARDS baseline visit. As these
individuals may have a low risk for outcomes, the protective
association of healthy lifestyle factors with risk of outcomes
in our study may have been underestimated. Finally, the study
was observational; therefore, the causal nature of the associations cannot be established.
Despite these limitations, there are several strengths to our
study. First, the REGARDS study is one of the largest population-based studies conducted in the United States and includes
a biracial sample of participants recruited from across the
United States. Therefore, results of this study may be highly
generalizable to the US adult population. Second, BP was
measured by trained technicians using a standardized protocol.8 Third, medication usage was recorded through direct
inspection of medication pill bottles. Finally, all outcomes
were adjudicated by a centralized events committee.

Perspectives
In a geographically diverse, biracial population-based sample
of the US adults with aTRH, a greater number of healthy lifestyle factors were associated with a lower risk of cardiovascular events, all-cause mortality, and cardiovascular mortality.
Among the individual healthy lifestyle factors investigated,
physical activity and nonsmoking status were each associated with a decreased risk for cardiovascular events, all-cause
mortality, and cardiovascular mortality. A higher DASH diet
score was also associated with a decreased risk for all-cause
mortality. These data support the concept that the prognosis
among individuals with aTRH may be improved by interventions targeting healthy lifestyle factors. Future randomized
controlled trials are warranted to examine whether lifestyle
interventions lower morbidity and mortality risk among
adults with aTRH.

Acknowledgments
We thank the staff, participants, and other investigators of the
REGARDS study for their valuable contributions. A full list of participating REGARDS investigators and institutions can be found at
http://www.regardsstudy.org.

Downloaded from http://hyper.ahajournals.org/ by guest on November 22, 2015

Diaz et al Lifestyle Factors in Resistant Hypertension 471

Sources of Funding
This research is supported by a cooperative agreement U01NS041588 from the National Institute of Neurological Disorders
and Stroke (NINDS), National Institutes of Health (NIH), and
Department of Health and Human Service. Additional funding
was provided by R01-HL80477 from the National Heart, Lung
and Blood Institute (NHLBI) and by General Mills for coding of
the Food Frequency Questionnaire. Dr Diaz was supported by an
NHLBI Diversity Supplement (P01-HL047540-19S1). The content
is solely the responsibility of the authors and does not necessarily
represent the official views of the NINDS or NIH. Representatives
of the funding agency have been involved in review of the article
but not directly involved in the collection, management, analysis, or
interpretation of the data.

Disclosures
None.

References
1. Calhoun DA, Jones D, Textor S, Goff DC, Murphy TP, Toto RD, White A,
Cushman WC, White W, Sica D, Ferdinand K, Giles TD, Falkner B, Carey
RM; American Heart Association Professional Education Committee.
Resistant hypertension: diagnosis, evaluation, and treatment: a scientific
statement from the American Heart Association Professional Education
Committee of the Council for High Blood Pressure Research. Circulation.
2008;117:e510e526.
2. Egan BM, Zhao Y, Axon RN, Brzezinski WA, Ferdinand KC. Uncontrolled
and apparent treatment resistant hypertension in the United States, 1988 to
2008. Circulation. 2011;124:10461058.
3. Daugherty SL, Powers JD, Magid DJ, Tavel HM, Masoudi FA, Margolis
KL, OConnor PJ, Selby JV, Ho PM. Incidence and prognosis of resistant
hypertension in hypertensive patients. Circulation. 2012;125:16351642.
4. Irvin MR, Booth JN, Shimbo D, Lackland DT, Oparil S, Howard G,
Safford MM, Muntner P, Calhoun DA. Apparent treatment-resistant
hypertension and risk for stroke, coronary heart disease and all-cause
mortality [published online ahead of print March 17, 2014]. J Am Soc
Hypertens. doi:10.1016/j.jash.2014.03.003. http://www.ashjournal.com/
article/S1933-1711(14)00100-4. Accessed March 18, 2014.
5. Bangalore S, Fayyad R, Laskey R, Demicco DA, Deedwania P, Kostis JB,
Messerli FH; Treating to New Targets Steering Committee and Investigators.
Prevalence, predictors, and outcomes in treatment-resistant hypertension in
patients with coronary disease. Am J Med. 2014;127:7181.e1.
6. Chobanian AV, Bakris GL, Black HR, Cushman WC, Green LA, Izzo JL
Jr, Jones DW, Materson BJ, Oparil S, Wright JT Jr, Roccella EJ; National
Heart, Lung, and Blood Institute Joint National Committee on Prevention,
Detection, Evaluation, and Treatment of High Blood Pressure; National
High Blood Pressure Education Program Coordinating Committee. The
Seventh Report of the Joint National Committee on Prevention, Detection,
Evaluation, and Treatment of High Blood Pressure: the JNC 7 report. J Am
Med Assoc. 2003;289:25602572.
7. Mancia G, Fagard R, Narkiewicz K, et al; Task Force Members. 2013
ESH/ESC Guidelines for the management of arterial hypertension: the

Task Force for the management of arterial hypertension of the European


Society of Hypertension (ESH) and of the European Society of Cardiology
(ESC). J Hypertens. 2013;31:12811357.
8. Howard VJ, Cushman M, Pulley L, Gomez CR, Go RC, Prineas RJ,
Graham A, Moy CS, Howard G. The reasons for geographic and racial
differences in stroke study: objectives and design. Neuroepidemiology.
2005;25:135143.
9. Ronksley PE, Brien SE, Turner BJ, Mukamal KJ, Ghali WA. Association
of alcohol consumption with selected cardiovascular disease outcomes: a
systematic review and meta-analysis. BMJ. 2011;342:d671.
10. World Health Organization (WHO). Waist Circumference and WaistHip Ratio: Report of a WHO Expert Consultation. Geneva, Switzerland:
World Health Organization; 2011.
11. Expert Panel on Detection, Evaluation, and Treatment of High Blood
Cholesterol in Adults. 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). J Am Med Assoc. 2001;285:24862497.
12. Fung TT, Chiuve SE, McCullough ML, Rexrode KM, Logroscino G, Hu
FB. Adherence to a DASH-style diet and risk of coronary heart disease
and stroke in women. Arch Intern Med. 2008;168:713720.
13. White IR, Royston P, Wood AM. Multiple imputation using chained equations: Issues and guidance for practice. Stat Med. 2011;30:377399.
14. Fagard RH. Resistant hypertension. Heart. 2012;98:254261.
15. Parikh A, Lipsitz SR, Natarajan S. Association between a DASH-like diet
and mortality in adults with hypertension: findings from a populationbased follow-up study. Am J Hypertens. 2009;22:409416.
16. Rossi A, Dikareva A, Bacon SL, Daskalopoulou SS. The impact of physical activity on mortality in patients with high blood pressure: a systematic
review. J Hypertens. 2012;30:12771288.
17. Chiuve SE, McCullough ML, Sacks FM, Rimm EB. Healthy lifestyle
factors in the primary prevention of coronary heart disease among men:
benefits among users and nonusers of lipid-lowering and antihypertensive
medications. Circulation. 2006;114:160167.
18. Forman JP, Stampfer MJ, Curhan GC. Diet and lifestyle risk factors associated
with incident hypertension in women. J Am Med Assoc. 2009;302:401411.
19. King DE, Mainous AG III, Carnemolla M, Everett CJ. Adherence to healthy
lifestyle habits in US adults, 1988-2006. Am J Med. 2009;122:528534.
20. Dimeo F, Pagonas N, Seibert F, Arndt R, Zidek W, Westhoff TH. Aerobic
exercise reduces blood pressure in resistant hypertension. Hypertension.
2012;60:653658.
21. Warburton DE, Nicol CW, Bredin SS. Health benefits of physical activity:
the evidence. CMAJ. 2006;174:801809.
22. Critchley JA, Capewell S. Mortality risk reduction associated with smoking cessation in patients with coronary heart disease: a systematic review.
J Am Med Assoc. 2003;290:8697.
23. Pimenta E, Gaddam KK, Oparil S, Aban I, Husain S, DellItalia LJ,
Calhoun DA. Effects of dietary sodium reduction on blood pressure in
subjects with resistant hypertension: results from a randomized trial.
Hypertension. 2009;54:475481.
24. Agarwal R. Resistant hypertension and the neglected antihypertensive:
sodium restriction. Nephrol Dial Transplant. 2012;27:40414045.
25. Appel LJ. Another major role for dietary sodium reduction: improving blood pressure control in patients with resistant hypertension.
Hypertension. 2009;54:444446.

Novelty and Significance


What Is New?

This is the first study to examine whether healthy lifestyle factors are
associated with better prognosis among individuals with apparent treatment-resistant hypertension.

Summary
Lifestyle modification interventions may be beneficial for reducing
morbidity and mortality risk in individuals with apparent treatmentresistant hypertension.

What Is Relevant?

Among individuals with apparent treatment-resistant hypertension, hav-

ing more healthy lifestyle factors was associated with a lower risk for
cardiovascular events and mortality. Physical activity and nonsmoking
were individual healthy lifestyle factors associated with a lower risk for
these outcomes.

Downloaded from http://hyper.ahajournals.org/ by guest on November 22, 2015

R3

ONLINE SUPPLEMENT
HEALTHY LIFESTYLE FACTORS AND RISK OF CARDIOVASCULAR EVENTS
AND MORTALITY IN TREATMENT-RESISTANT HYPERTENSION: THE REGARDS
STUDY
Keith M. Diaz, John N. Booth III, David A. Calhoun, Marguerite R. Irvin, George Howard,
Monika M. Safford, Paul Muntner, Daichi Shimbo
From the Department of Medicine, Columbia University Medical Center, New York, NY
(K.M.D., D.S.); Departments of Epidemiology (J.N.B., M.R.I., P.M.), Biostatistics (G.H.), and
Medicine (D.A.C., M.M.S.), University of Alabama at Birmingham, Birmingham, AL.

Correspondence:
Keith M. Diaz
Columbia University Medical Center
622 West 168th Street, PH9-319
New York, NY 10032
212-304-5231
Fax: 212-305-3172
Email: kd2442@columbia.edu

R3

Supplemental Methods
Data Collected at Baseline: REasons for Geographic And Racial Differences in Stroke
(REGARDS) study data were collected via a computer-assisted telephone interview, a selfadministered questionnaire, and in-home examination. Trained health professionals conducted
in-home examinations which included anthropometric and blood pressure (BP) measurements,
electrocardiogram (ECG), collection of blood and spot urine samples, and a review of
medication pill bottles for medications taken in the prior 2 weeks. History of coronary heart
disease (CHD) was defined as a self-reported or ECG evidence of myocardial infarction (MI) or
a self-reported coronary revascularization procedure. History of stroke was defined on the basis
of self-report. Atrial fibrillation was defined as self-reported history or evidence on ECG using
standardized procedures1. Left ventricular hypertrophy was defined by ECG using SokolowLyon criteria2. Diabetes was defined as a serum glucose 126 mgdL for participants who had
fasted 8 hours prior to their blood draw, a serum glucose 200 mgdL for those who had not
fasted, or a self-report of a prior diagnosis of diabetes with use of insulin or oral hypoglycemic
medications. Chronic kidney disease was defined by the presence of a reduced estimated
glomerular filtrate rate (eGFR, <60 ml/min/1.73 m2) calculated using the Chronic Kidney
Disease Epidemiology Collaboration (CKD-EPI) equation3 and/or albuminuria (urinary albumin
to urinary creatinine ratio 30 mg/g). Medication adherence was assessed by summing each
point assigned for responding yes (totaling 0-4) to each of the 4-items on the Morisky
Medication Adherence Scale (MMAS), with a higher score indicating worse adherence4,5.
Validity and Reproducibility of Physical Activity and Food Frequency Questionnaires: The
physical activity question (How many times per week do you engage in intense physical activity,
enough to work up a sweat?) used in REGARDS is a widely used measure of physical activity
that has been validated against maximum aerobic capacity, has moderate reproducibility (0.600.70), can include either aerobic or resistance training, and has been used in similar studies, such
as the National Health And Nutrition Examination Survey (NHANES)6-8. For dietary intake
assessment, versions of the Block Food Frequency Questionnaire have been extensively
validated against 24-hour recalls and 4-day food records among diverse populations of adults,
have been demonstrated to have moderate to high reproducibility (0.57-0.90), and have been
used in several other population-based studies including NHANES9-11.
Outcome Ascertainment: For the current study, cardiovascular events, all-cause mortality, and
cardiovascular mortality were evaluated as primary, secondary and tertiary outcomes,
respectively. Vital status and cardiovascular events were ascertained during bi-annual telephone
follow-up interviews of the participant, proxy, or next-of-kin. Trained experts adjudicated all
cardiovascular events and deaths (described below) using information retrieved from
interviewing the participant, proxy, or next-of-kin; and by reviewing autopsy reports, death
certificates, medical records and administrative databases (e.g. Social Security Death Index,
National Death Index) for the circumstances surrounding each event. Agreement on independent
review for all endpoints had >0.80.
Cardiovascular events: Cardiovascular events included combined fatal and non-fatal
cardiovascular events. Fatal cardiovascular events were defined as death within 28 days of a
definite or probable MI, or sudden death; or death within 28 days of a confirmed ischemic,
hemorrhagic or clinical (defined below) stroke event. Non-fatal cardiovascular events were

R3

defined as a non-fatal definite or probable MI or a non-fatal confirmed stroke. Cardiovascular


events occurring through December 31, 2010 were available for the current analysis.
Cardiovascular events were adjudicated using published guidelines12,13. MI was
determined using a combination of clinical signs and symptoms consistent with ischemia,
imaging/ECG findings, and a characteristic rising and falling pattern of cardiac biomarkers
(usually troponin, but in the absence of troponin, creatine phosphokinase-MB fraction) over 6 or
more hours to a peak of at least twice the upper limit of normal. The interpretation of ECG was
guided by the Minnesota code to classify ECGs as evolving diagnostic, positive, non-specific or
not consistent with ischemia14,15.
For stroke events, a panel of neurologists used the World Health Organization (WHO)
definition16 to adjudicate stroke events based on self-reported hospitalizations or physician
evaluations of stroke, and newly reported stroke symptoms detected on the Questionnaire for
Verifying Stroke-Free Status that resulted in a medical evaluation17. Events not meeting the
WHO definition were classified as clinical strokes if symptoms lasted <24 hours and
neuroimaging was consistent with acute infarct or hemorrhage. We included both WHO and
clinic strokes.
All-Cause Mortality: All-cause mortality was defined as any REGARDS participant who died
after enrollment regardless of the cause of death. Dates of death were confirmed through review
of death certificates, medical records, and administrative databases (e.g. Social Security Death
Index, National Death Index). Deaths occurring through March 29, 2012 were included in the
current analysis.
Cardiovascular Morality: For cardiovascular mortality, the cause of death was adjudicated as
caused by a definite, probable or possible MI; sudden death; heart failure; stroke; or other
cardiovascular cause including ruptured aortic aneurysm, pulmonary embolism or other heartrelated condition. Deaths from cardiovascular-related causes occurring through December 31,
2010 were available for the current analysis.

R3

SUPPLEMENTAL REFERENCES
1.

2.
3.
4.
5.
6.
7.
8.

9.
10.
11.

12.

Meschia JF, Merrill P, Soliman EZ, Howard VJ, Barrett KM, Zakai NA, Kleindorfer D,
Safford M, Howard G. Racial disparities in awareness and treatment of atrial fibrillation:
The reasons for geographic and racial differences in stroke (regards) study. Stroke.
2010;41:581-587.
Sokolow M, Lyon TP. The ventricular complex in left ventricular hypertrophy as
obtained by unipolar precordial and limb leads. Am Heart J. 1949;37:161-186.
Levey AS, Stevens LA, Schmid CH, Zhang YL, Castro AF, III, Feldman HI, Kusek JW,
Eggers P, Van LF, Greene T, Coresh J. A new equation to estimate glomerular filtration
rate. Ann Intern Med. 2009;150:604-612.
Morisky DE, Green LW, Levine DM. Concurrent and predictive validity of a selfreported measure of medication adherence. Med Care. 1986;24:67-74.
Shalansky SJ, Levy AR, Ignaszewski AP. Self-reported morisky score for identifying
nonadherence with cardiovascular medications. Ann of Pharmacother. 2004;38:13631368.
Kohl HW, Blair SN, Paffenbarger RS, Macera CA, Kronenfeld JJ. A Mail survey of
physical activity habits as related to measured physical fitness. Am J Epidemiol.
1988;127:1228-1239.
Philippaerts RM, Lefevre J. Reliability and validity of three physical activity
questionnaires in flemish males. Am. J. Epidemiol. 1998;147: 982-990.
Zhao G, Li C, Ford ES, Fulton JE, Carlson SA, Okoro CA, Wen XJ, Balluz LS
Leisure-time aerobic physical activity, muscle-strengthening activity and mortality risks
among US adults: the NHANES linked mortality study. Br J Sports Med. 2014;48:244249.
Block G, Hartman AM, Naughton D. A reduced dietary questionnaire: development and
validation. Epidemiology. 1990;1:58-64.
Boucher et.al. Validity and reliability of the Block 98 food-frequency questionnaire in a
sample of Canadian women. Public Health Nutr. 2006;9:84-93.
National Center for Health Statistics. Plan and Operation of the Third National Health
and Nutrition Examination Survey, 198894. Hyattsville, MD, National Center for Health
Statistics, 1994. (Vital and Health Statistics Ser. 1, no. 32). U.S. Department of Health
and Human Services Publication No. (PHS) 94-1308; 1994.
Luepker RV, Apple FS, Christenson RH, Crow RS, Fortmann SP, Goff D, Goldberg RJ,
Hand MM, Jaffe AS, Julian DG, Levy D, Manolio T, Mendis S, Mensah G, Pajak A,
Prineas RJ, Reddy KS, Roger VL, Rosamond WD, Shahar E, Sharrett AR, Sorlie P,
Tunstall-Pedoe H, Epidemiology AHACo, Prevention, Committee AHAS, World Heart
Federation Council on E, Prevention, European Society of Cardiology Working Group on
E, Prevention, Centers for Disease C, Prevention, National Heart L, Blood I. Case
definitions for acute coronary heart disease in epidemiology and clinical research studies:
A statement from the aha council on epidemiology and prevention; aha statistics
committee; world heart federation council on epidemiology and prevention; the european
society of cardiology working group on epidemiology and prevention; centers for disease
control and prevention; and the national heart, lung, and blood institute. Circulation.
2003;108:2543-2549.

R3

13.

14.
15.
16.
17.

Thygesen K, Alpert JS, White HD, Joint ESCAAHAWHFTFftRoMI, Jaffe AS, Apple
FS, Galvani M, Katus HA, Newby LK, Ravkilde J, Chaitman B, Clemmensen PM,
Dellborg M, Hod H, Porela P, Underwood R, Bax JJ, Beller GA, Bonow R, Van der Wall
EE, Bassand JP, Wijns W, Ferguson TB, Steg PG, Uretsky BF, Williams DO, Armstrong
PW, Antman EM, Fox KA, Hamm CW, Ohman EM, Simoons ML, Poole-Wilson PA,
Gurfinkel EP, Lopez-Sendon JL, Pais P, Mendis S, Zhu JR, Wallentin LC, FernandezAviles F, Fox KM, Parkhomenko AN, Priori SG, Tendera M, Voipio-Pulkki LM,
Vahanian A, Camm AJ, De Caterina R, Dean V, Dickstein K, Filippatos G, FunckBrentano C, Hellemans I, Kristensen SD, McGregor K, Sechtem U, Silber S, Tendera M,
Widimsky P, Zamorano JL, Morais J, Brener S, Harrington R, Morrow D, Lim M,
Martinez-Rios MA, Steinhubl S, Levine GN, Gibler WB, Goff D, Tubaro M, Dudek D,
Al-Attar N. Universal definition of myocardial infarction. Circulation. 2007;116:26342653.
Prineas RJ, Crow RS, Blackburn HW. The minnesota code manual of
electrocardiographic findings : Standards and procedures for measurement and
classification. Boston, Mass.: J. Wright; 1982.
Prineas RJ, Crow RS, Blackburn HW. The minnesota code manual of
electrocardiographic findings : Standards and procedures for measurement and
classification. London: Springer; 2010.
Stroke--1989. Recommendations on stroke prevention, diagnosis, and therapy. Report of
the who task force on stroke and other cerebrovascular disorders. Stroke. 1989;20:14071431.
Meschia JF, Brott TG, Chukwudelunzu FE, Hardy J, Brown RD, Meissner I, Hall LJ,
Atkinson EJ, O'Brien PC. Verifying the stroke-free phenotype by structured telephone
interview. Stroke. 2000;31:1076-1080.

R3

Table S1: Characteristics of REGARDS participants included and not included


in the present analysis.
Included
Excludeda P-difference
(n=2,043) (n=28,196)
Variable
Age (years)
67.6 0.2
64.7 0.1
<0.001
Male (%)
49.2
44.6
<0.001
Black (%)
60.5
40.1
<0.001
Education < High School (%)
19.2
12.1
<0.001
Region of residence
Non-belt & non-buckle (%)
34.8
34.6
1 (ref)
Residence in stroke belt (%)
20.7
20.9
0.824
Residence in stroke buckle (%)
44.5
44.5
0.929
Diabetes (%)
45.7
20.3
<0.001
Total Cholesterol (mg/dL)
180.3 0.9 192.9 0.2
<0.001
HDL-cholesterol (mg/dL)
48.3 0.3
52.1 0.1
<0.001
27.6
10.4
<0.001
eGFR < 60 ml/min1.73m2 (%)
Albuminuria (%)
33.6
14.2
<0.001
Atrial Fibrillation (%)
15.4
8.4
<0.001
Left Ventricular Hypertrophy (%)
18.5
9.3
<0.001
Systolic BP (mmHg)
142.0 0.4 126.6 0.1
<0.001
Diastolic BP (mmHg)
78.0 0.1
76.3 0.1
<0.001
No. of antihypertensive
3.6 0.0
1.0 0.0
<0.001
medications
Statin use (%)
52.1
29.9
<0.001
History of CHD (%)
36.0
16.7
<0.001
History of stroke (%)
13.6
5.9
<0.001
Data are presented as mean standard error or percentage.
BP, blood pressure; CHD, coronary heart disease; eGFR, estimated glomerular
filtration rate; HDL, high-density lipoprotein.
a
Participants were excluded if they were missing blood pressure measurements
or data on antihypertensive medications at baseline, did not have aTRH, or
were missing outcome data during follow-up.

R3

Table S2: Hazard ratios for cardiovascular events associated with cumulative
number of healthy lifestyle factors in REGARDS study participants with
apparent treatment resistant hypertension stratified by race.
Number of
Person
P-Interaction
Healthy
No. of
Hazard Ratioa
(Blacks vs.
years at
(95% CI)
Lifestyle
events
Risk
Whites)
Factors
Blacks
0-1
82
1,976.2
1 (ref)
2
80
1,975.3
0.94 (0.671.34)
3
36
1,060.1
0.83 (0.521.34)
4-6
13
549.9
0.59 (0.521.34)
P-trend=0.118
0.934
Whites
0-1
40
893.1
1 (ref)
2
48
1,119.3
0.85 (0.531.36)
3
35
941.1
0.72 (0.441.18)
4-6
26
766.9
0.60 (0.351.04)
P-trend=0.050
Healthy lifestyle factors: waist circumference <102 cm in men and <88 cm in
women, physical activity 4 times/week, non-smoking, moderate alcohol
consumption, highest quartile (4th) for DASH diet score, lowest (1st) quartile for
Na/K intake ratio.
a
Adjusted for the following covariates: age, sex, education, and geographic
region of residence, total cholesterol, HDL-cholesterol, eGFR < 60 ml/min/1.73
m2, albuminuria, atrial fibrillation, left ventricular hypertrophy, diabetes, statin
use, history of coronary heart disease, and history of stroke.

R3

Table S3: Hazard ratios for cardiovascular events associated with cumulative
number of healthy lifestyle factors in REGARDS study participants with
apparent treatment resistant hypertensionstratified by history of stroke or
coronary heart disease (CHD).
Number of
P-Interaction
Person
a
Healthy
No. of
Hazard Ratio
(No stroke or
years at
Lifestyle
events
(95% CI)
CHD history vs.
Risk
Factors
stroke or CHD)
No history of stroke or CHD
0-1
52
1,699.0
1 (ref)
2
54
1,740.5
0.96 (0.641.45)
3
30
1,213.8
0.78 (0.451.34)
4-6
15
756.6
0.52 (0.271.00)
P-trend=0.040
0.628
History of stroke or CHD
0-1
70
1,170.3
1 (ref)
2
74
1,353.0
0.88 (0.571.36)
3
41
787.4
0.84 (0.541.31)
4-6
24
560.2
0.71 (0.421.20)
P-trend=0.189
Healthy lifestyle factors: waist circumference <102 cm in men and <88 cm in
women, physical activity 4 times/week, non-smoking, moderate alcohol
consumption, highest quartile (4th) for DASH diet score, lowest (1st) quartile for
Na/K intake ratio.
a
Adjusted for the following covariates: age, race, sex, education, and geographic
region of residence, total cholesterol, HDL-cholesterol, eGFR < 60 ml/min/1.73
m2, albuminuria, atrial fibrillation, left ventricular hypertrophy, diabetes, and
statin use.

R3

Table S4: Hazard ratios for cardiovascular events associated with cumulative number of healthy
lifestyle factors in REGARDS study participants with apparent treatment resistant hypertension
restricted to individuals with complete data for the healthy lifestyle factors (n=1,300).
Number of
Person
Hazard Ratio (95% CI) for Cardiovascular Events
Healthy Lifestyle No. of
years
Factors
events at Risk
Unadjusted
Model 1
Model 2
0-1
78
2,011.4
1 (ref)
1 (ref)
1 (ref)
2
94
2,211.7 1.10 (0.811.48) 0.97 (0.711.31) 0.91 (0.651.27)
3
36
1,225.0 0.76 (0.511.13) 0.58 (0.380.88) 0.61 (0.390.96)
4-6
19
629.4
0.78 (0.471.29) 0.56 (0.330.94) 0.56 (0.320.99)
P-trend=0.095
P-trend=0.001
P-trend=0.007
Healthy lifestyle factors: waist circumference < 102 cm in men and < 88 cm in women, physical
activity 4 times/week, non-smoking, moderate alcohol consumption, highest quartile (4th) for
DASH diet score, lowest (1st) quartile for Na/K intake ratio.
Model 1: Adjusted for age, race, sex, education, and geographic region of residence.
Model 2: Adjusted for covariates in model 1 plus total cholesterol, HDL-cholesterol, eGFR < 60
ml/min/1.73 m2, albuminuria, atrial fibrillation, left ventricular hypertrophy, diabetes, statin use,
history of coronary heart disease, and history of stroke.

R3

Table S5: Hazard ratios for all-cause mortality associated with cumulative number of healthy
lifestyle factors in REGARDS study participants with apparent treatment resistant hypertension
restricted to individuals with complete data for the healthy lifestyle factors (n=1,300).
Number of
Person
Hazard Ratio (95% CI) for All-Cause Mortality
Healthy Lifestyle No. of
years
Factors
events at Risk
Unadjusted
Model 1
Model 2
0-1
100
2,369.1
1 (ref)
1 (ref)
1 (ref)
2
95
2,621.2 0.84 (0.631.11) 0.72 (0.540.96) 0.71 (0.510.98)
3
49
1,458.9 0.76 (0.541.08) 0.51 (0.360.74) 0.52 (0.350.79)
4-6
19
750.8
0.58 (0.350.94) 0.38 (0.230.61) 0.42 (0.250.71)
P-trend=0.016
P-trend <0.001
P-trend <0.001
Healthy lifestyle factors: waist circumference < 102 cm in men and < 88 cm in women, physical
activity 4 times/week, non-smoking, moderate alcohol consumption, highest quartile (4th) for
DASH diet score, lowest (1st) quartile for Na/K intake ratio.
Model 1: Adjusted for age, race, sex, education, and geographic region of residence.
Model 2: Adjusted for covariates in model 1 plus total cholesterol, HDL-cholesterol, eGFR < 60
ml/min/1.73 m2, albuminuria, atrial fibrillation, left ventricular hypertrophy, diabetes, statin use,
history of coronary heart disease, and history of stroke.

10

R3

Table S6: Hazard ratios for all-cause mortality associated with individual lifestyle factors in
REGARDS study participants with apparent treatment resistant hypertension.
Person
Hazard Ratio (95% CI) for All-Cause Mortality
No. of years at
Lifestyle Factor
Events
Risk
Unadjusted
Model 1
Model 2
Abdominal Obesity
Yes
271
7,402.8
1 (ref)
1 (ref)
1 (ref)
No
181
3,562.8 1.38 (1.141.66) 1.05 (0.861.29) 1.10 (0.891.35)
Physical Activity
None
258
4,533.6
1 (ref)
1 (ref)
1 (ref)
1-3 times/week
117
3,854.3 0.51 (0.410.64) 0.50 (0.400.62) 0.58 (0.460.73)
4 times/week
77
2,577.8 0.51 (0.390.65) 0.48 (0.370.61) 0.58 (0.450.75)
P-trend <0.001
P-trend <0.001
P-trend <0.001
Current Smoking
Yes
79
1,212.6
1 (ref)
1 (ref)
1 (ref)
No
373
9,953.1 0.57 (0.450.73) 0.45 (0.350.59) 0.53 (0.400.69)
Alcohol Consumptiona
Heavy
10
340.6
1 (ref)
1 (ref)
1 (ref)
Moderate
114
3,054.6 0.66 (0.440.99) 0.67 (0.450.99) 0.80 (0.441.49)
None
328
7,570.5 0.78 (0.541.14) 0.79 (0.551.14) 0.98 (0.791.22)
DASH diet score
Quartile 1
118
2,767.3
1 (ref)
1 (ref)
1 (ref)
( 20)
Quartile 2
132
2,848.1 1.10 (0.841.45) 0.96 (0.731.25) 0.97 (0.741.28)
(>20-23)
Quartile 3
101
2,489.2 0.95 (0.691.32) 0.82 (0.601.12) 0.78 (0.571.07)
(>23-26)
Quartile 4
101
2,861.1 0.82 (0.601.13) 0.69 (0.510.95) 0.72 (0.530.98)
(>26)
P-trend=0.158
P-trend=0.017
P-trend=0.019
Na/K intake
Quartile 4
128
2,911.2
1 (ref)
1 (ref)
1 (ref)
(0.71)
Quartile 3
123
2,890.4 0.98 (0.741.31) 0.91 (0.691.19) 0.92 (0.701.20)
(>0.71-0.88)
Quartile 2
100
2,665.0 0.84 (0.611.15) 0.76 (0.541.05) 0.78 (0.551.11)
(>0.88-1.08)
Quartile 1
101
2,559.2 0.90 (0.661.21) 0.81 (0.601.08) 0.80 (0.591.08)
(>1.08)
P-trend=0.302
P-trend=0.080
P-trend=0.094
Model 1: Adjusted for age, race, sex, education, and geographic region of residence.
Model 2: Adjusted for covariates in model 1 plus total cholesterol, HDL-cholesterol, eGFR < 60
ml/min/1.73 m2, albuminuria, atrial fibrillation, left ventricular hypertrophy, diabetes, statin use, history
of coronary heart disease, and history of stroke.
a
P-trend values for alcohol consumption are not presented as we hypothesized that the association
between alcohol consumption and outcomes would not be linear given that moderate alcohol
consumption (the middle category) was considered to be a healthy lifestyle factor in our analyses.

11

R3

Table S7: Hazard ratios for cardiovascular mortality associated with cumulative number of
healthy lifestyle factors in REGARDS study participants with apparent treatment resistant
hypertension.
Hazard Ratio (95% CI) for Cardiovascular
Number of
Person
Mortality
Healthy Lifestyle No. of
years
Factors
events at Risk
Unadjusted
Model 1
Model 2
0-1
66
3,064.0
1 (ref)
1 (ref)
1 (ref)
2
61
3,294.9 0.83 (0.581.19) 0.72 (0.501.04) 0.75 (0.511.09)
3
30
2,138.9 0.62 (0.381.01) 0.49 (0.290.81) 0.55 (0.330.93)
4-6
17
1,408.9 0.54 (0.300.96) 0.40 (0.230.70) 0.48 (0.270.85)
P-trend=0.009
P-trend <0.001
P-trend=0.004
Healthy lifestyle factors: waist circumference < 102 cm in men and < 88 cm in women, physical
activity 4 times/week, non-smoking, moderate alcohol consumption, highest quartile (4th) for
DASH diet score, lowest (1st) quartile for Na/K intake ratio.
Model 1: Adjusted for age, race, sex, education, and geographic region of residence.
Model 2: Adjusted for covariates in model 1 plus total cholesterol, HDL-cholesterol, eGFR < 60
ml/min/1.73 m2, albuminuria, atrial fibrillation, left ventricular hypertrophy, diabetes, statin use,
history of coronary heart disease, and history of stroke.

12

R3

Table S8: Hazard ratios for cardiovascular mortality associated with individual lifestyle factors in
REGARDS study participants with apparent treatment resistant hypertension.
Person Hazard Ratio (95% CI) for Cardiovascular Mortality
No. of years at
Lifestyle Factor
Events
Risk
Unadjusted
Model 1
Model 2
Abdominal Obesity
Yes
107
6,680.9
1 (ref)
1 (ref)
1 (ref)
No
67
3,225.7 1.30 (0.961.76) 1.01 (0.741.41) 1.08 (0.771.51)
Physical Activity
None
97
4,111.4
1 (ref)
1 (ref)
1 (ref)
1-3 times/week
53
3,476.1 0.60 (0.430.84) 0.59 (0.420.82) 0.72 (0.511.00)
4 times/week
24
2,319.2 0.42 (0.270.66) 0.40 (0.250.62) 0.50 (0.320.80)
P-trend <0.001
P-trend <0.001
P-trend=0.002
Current Smoking
Yes
35
1,104.3
1 (ref)
1 (ref)
1 (ref)
No
139
8,802.4 0.49 (0.340.71) 0.39 (0.270.58) 0.47 (0.310.70)
Alcohol Consumptiona
Heavy
4
309.8
1 (ref)
1 (ref)
1 (ref)
Moderate
43
2,746.2 0.67 (0.351.31) 0.69 (0.361.35) 0.76 (0.391.51)
None
127
6,850.6 0.81 (0.441.51) 0.83 (0.441.58) 0.82 (0.421.58)
DASH diet score
Quartile 1
40
2,541.6
1 (ref)
1 (ref)
1 (ref)
( 20)
Quartile 2
47
2,561.2 1.05 (0.631.75) 0.93 (0.551.59) 0.95 (0.561.62)
(>20-23)
Quartile 3
48
2,267.3 1.27 (0.732.19) 1.11 (0.631.95) 1.09 (0.631.88)
(>23-26)
Quartile 4
39
2,536.5 0.85 (0.511.41) 0.73 (0.441.23) 0.79 (0.461.34)
(>26)
P-trend=0.741
P-trend=0.384
P-trend=0.520
Na/K intake
Quartile 4
45
2,677.7
1 (ref)
1 (ref)
1 (ref)
(0.71)
Quartile 3
48
2,535.9 1.21 (0.672.17) 1.12 (0.622.01) 1.13 (0.641.99)
(>0.71-0.88)
Quartile 2
41
2,400.9 1.00 (0.621.61) 0.90 (0.541.47) 0.96 (0.591.58)
(>0.88-1.08)
Quartile 1
42
2,292.1 1.03 (0.621.70) 0.92 (0.541.56) 0.93 (0.541.57)
(>1.08)
P-trend=0.870
P-trend=0.514
P-trend=0.603

Model 1: Adjusted for age, race, sex, education, and geographic region of residence.
Model 2: Adjusted for covariates in model 1 plus total cholesterol, HDL-cholesterol, eGFR < 60
ml/min/1.73 m2, albuminuria, atrial fibrillation, left ventricular hypertrophy, diabetes, statin use, history
of coronary heart disease, and history of stroke.
a
P-trend values for alcohol consumption are not presented as we hypothesized that the association
between alcohol consumption and outcomes would not be linear given that moderate alcohol
consumption (the middle category) was considered to be a healthy lifestyle factor in our analyses.

13

R3

N=30,239

N=14,853

Not currently taking


antihypertensive medication or
missing medication data
n=15,386
Missing blood pressure data
n=87

N=14,766
Individuals who do not have aTRH
n=12,683

N=2,083
Missing follow-up information,
n=40

N=2,043
Figure S1. Inclusion criteria for examining the association of healthy lifestyle factors with
cardiovascular events, all-cause mortality, and cardiovascular mortality among individuals with
apparent treatment resistant hypertension (aTRH) in the REGARDS study.

14

Healthy Lifestyle Factors and Risk of Cardiovascular Events and Mortality in


Treatment-Resistant Hypertension: The Reasons for Geographic and Racial Differences in
Stroke Study
Keith M. Diaz, John N. Booth III, David A. Calhoun, Marguerite R. Irvin, George Howard,
Monika M. Safford, Paul Muntner and Daichi Shimbo
Hypertension. 2014;64:465-471; originally published online June 9, 2014;
doi: 10.1161/HYPERTENSIONAHA.114.03565
Hypertension is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright 2014 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/64/3/465

Data Supplement (unedited) at:


http://hyper.ahajournals.org/content/suppl/2014/06/09/HYPERTENSIONAHA.114.03565.DC1.html

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 November 22, 2015