Pediatric Cardiology

Sleep Quality and Elevated Blood Pressure in Adolescents

Sogol Javaheri, MA; Amy Storfer-Isser, MS; Carol L. Rosen, MD; Susan Redline, MD, MPH

BackgroundWe assessed whether insufficient sleep is associated with prehypertension in healthy adolescents.
Methods and ResultsWe undertook a cross-sectional analysis of 238 adolescents, all without sleep apnea or severe
comorbidities. Participants underwent multiple-day wrist actigraphy at home to provide objective estimates of sleep patterns.
In a clinical research facility, overnight polysomnography, anthropometry, and 9 blood pressure measurements over 2 days
were made. Exposures were actigraphy-defined low weekday sleep efficiency, an objective measure of sleep quality (low
sleep efficiency 85%), and short sleep duration (6.5 hours). The main outcome was prehypertension (90th percentile for
age, sex, and height), with systolic and diastolic blood pressures as continuous measures as secondary outcomes.
Prehypertension, low sleep efficiency, and short sleep duration occurred in 14%, 26%, and 11% of the sample, respectively.
In unadjusted analyses, the odds of prehypertension increased 4.5-fold (95% CI, 2.1 to 9.7) in adolescents with low sleep
efficiency and 2.8-fold (95% CI, 1.1 to 7.3) in those with short sleep. In analyses adjusted for sex, body mass index percentile,
and socioeconomic status, the odds of prehypertension increased 3.5-fold (95% CI, 1.5. 8.0) for low sleep efficiency and
2.5-fold (95% CI, 0.9 to 6.9) for short sleep. Adjusted analyses showed that adolescents with low sleep efficiency had on
Downloaded from http://circ.ahajournals.org/ by guest on April 15, 2017

average a 4.01.2-mm Hg higher systolic blood pressure than other children (P0.01).
ConclusionsPoor sleep quality is associated with prehypertension in healthy adolescents. Associations are not explained
by socioeconomic status, obesity, sleep apnea, or known comorbidities, suggesting that inadequate sleep quality is
associated with elevated blood pressure. (Circulation. 2008;118:1034-1040.)
Key Words: blood pressure epidemiology hypertension pediatrics sleep

H ypertension is an increasingly prevalent health problem

in adults and adolescents alike. Between 1988 and 1999,
prehypertension (ie, a blood pressure [BP] 90th percentile
for height, age, and sex) and hypertension were estimated to
increase in children by 2.3% and 1%, respectively.1 Child-
hood hypertension is associated with hypertension in adult-
hood, a risk factor for cardiovascular disease incidence and
death.25 It also is associated with end-organ damage, notably
left ventricular hypertrophy, in both children and adults.6,7
Clinical Perspective p 1040
Several studies have implicated insufficient sleep as a risk
factor for hypertension in adults.8 12 Although the cause is
unclear, experimental studies indicate that shorter sleep
results in metabolic and endocrine dysfunction, which may
contribute to cardiovascular disease.1317 Studies in both adult
and pediatric populations also have reported associations of
shorter sleep duration with obesity and impaired glucose
tolerance.14,18,19 These findings have a potentially large public
impact given the frequency of sleep curtailment.20
Few studies have addressed the relationship between sleep
and hypertension in children. A higher level of diastolic but not
systolic BP was reported in children with obstructive sleep apnea
compared with primary snorers.21 The Tucsons Childrens
Assessment of Sleep Apnea Study found that elevations in
systolic and diastolic BPs were independently associated with
sleep efficiency, respiratory disturbance index (a measure of
sleep apnea), and obesity in 230 children 6 to 11 years of age.22
To the best of our knowledge, no studies have examined the
association between insufficient sleep and BP in adolescents free
of sleep apnea. In this report, we examine the relationship
between prehypertension and systolic and diastolic BP levels
and objective measures of sleep quality and duration in a
community-based cohort of adolescents. First, we hypothesize
that adolescents with poor sleep quality or short sleep duration
will be at increased odds of prehypertension. Second, we posit
that adolescents with short sleep duration or poor sleep quality
will have higher systolic and diastolic BP readings on average
compared with adolescents with better-quality sleep. We ex-
cluded adolescents with clinically significant levels of sleep
apnea to minimize the influence of this exposure on BP and
sleep duration measurement.
Methods
Study Population
The sample was derived from the Cleveland Childrens Sleep and
Health Study, a longitudinal cohort study. Data for this analysis are from

Received January 14, 2008; accepted July 7, 2008.

From the Case School of Medicine (S.J., A.S.-I., C.L.R., S.R.) and Center for Clinical Investigation (A.S.-I., C.L.R., S.R.), Case Western Reserve University
School of Medicine; Department of Pediatrics, Rainbow Babies and Childrens Hospital (C.L.R.); and Case Center for Transdisciplinary Research on Energetics
and Cancer, Case Comprehensive Cancer Center and Department of Medicine, University Hospitals of Cleveland (A.S.-I., S.R.), Cleveland, Ohio.
The online-only Data Supplement can be found with this article at http://circ.ahajournals.org/cgi/content/full/CIRCULATIONAHA.108.766410/DC1.
Correspondence to Susan Redline, Case Western Reserve University, Cleveland, OH 44106-6003. E-mail sxr15@case.edu
2008 American Heart Association, Inc.
Circulation is available at http://circ.ahajournals.org DOI: 10.1161/CIRCULATIONAHA.108.766410

1034
 Javaheri et al
238 adolescents free of severe illnesses who participated in an exami-
nation performed between 2002 and 2006 aimed at participants 13 to 16
years of age. Details of the study population have been reported
elsewhere23,24 and are reviewed in the online Data Supplement.
Study Protocol
Adolescents underwent 5- to 7-day wrist actigraphy and completed a
daily sleep log at home during the week before a clinical research
center examination and when free of acute illness. After this period
of in-home monitoring, participants were studied in a dedicated
clinical research center where overnight polysomnography and
physiological and anthropometric assessments were performed using
a standardized protocol.23,25 Examinations at the research center
began at approximately 5 PM and ended the following day at 11 AM.
Informed consent was obtained from the childs legal guardian, and
written assent was obtained from the child. The study was approved
by the governing institutional review board.
Measurements
Actigraphy
Sleep-wake estimation was made with wrist actigraphy (Octagonal
Downloaded from http://circ.ahajournals.org/ by guest on April 15, 2017
Sleep Watch 2.01, Ambulatory Monitoring Inc, Ardsley, NY) analyzed
with the Action-W software and the time-above-threshold algorithm.24
Using weekday data (minimum, 3 days), we calculated mean sleep
duration and mean sleep efficiency, an objective measure of sleep
continuity and quality defined as the percentage of time in bed estimated
to be asleep (ie, total time estimated to be asleep divided by the total
time in bed for the major sleep period times 100). Adolescents with a
sleep efficiency 85% were considered to have low sleep efficiency.
Given the lack of data on cutoffs for defining short sleep duration in this
age, we used the lowest decile of mean sleep duration on weekdays to
define short sleep duration, which approximated 6.5 hours.
Blood Pressure
Three BP readings were obtained at each of 3 times (9 PM [supine]
the night of the polysomnography and 8 AM [supine] and 9:30 AM
[sitting] the following morning) following published guidelines.23
After a 10-minute rest period, BP was obtained by trained nurses
using a calibrated sphygmomanometer. The mean systolic and
diastolic BP values used in primary analyses were based on the
average of all 9 measurements. Prehypertension was identified if the
systolic and/or diastolic BP was 90th percentile for age, sex, and
height.4 Hypertension was defined as systolic or diastolic BP 95th
percentile. One adolescent using antihypertensive medication was
classified as having hypertension.
Other Measurements
A rigid stadiometer was used to measure height, and a calibrated
digital scale was used to measure weight. Body mass index (BMI)
was calculated by dividing the weight in kilograms by height in
meters squared and converted into age- and sex-adjusted percentiles
(http://www.cdc.gov/growthcharts/). Overweight was defined as
BMI 95th percentile. Adolescents who were reported to snore
loudly at least 1 to 2 times per week during the past month were
categorized as snorers. The apnea hypopnea index was defined as
all obstructive apneas and hypopneas with a 3% desaturation per
sleep hour from the polysomnogram. Socioeconomic status (SES)
measures included parent report of educational level and family
income. Additionally, the census tract of the childs residence when
initially enrolled in the study was linked to the corresponding 2000
US Census Bureau database, and median income of the census tract
was ascertained (http://wagda.lib.washington.edu/data/type/census/).
A composite SES z score was created by averaging the sample z
scores for these 3 measures. Tanner staging was performed by a
physician to determine pubertal status.26,27 Preterm status was
ascertained from birth records and defined as a gestational age 37
weeks. Attention deficit hyperactivity disorder (ADHD) was defined
as parent-reported doctors diagnosis of ADHD and either currently
present condition or medication/stimulant use for ADHD during the
past year. Using a standardized questionnaire, adolescents reported
Sleep and Blood Pressure in Adolescents 1035
the frequency with which they consumed caffeine after 6 PM during
the past month; those reporting frequently or always consuming
caffeine were coded as consuming caffeine in the evening.
Statistical Analysis
Between-group differences for the binary outcome, prehypertension,
were assessed with the Pearson 2 test for categorical variables, the
2-sample t test for normally distributed variables, and the Wilcoxon
rank-sum test for nonnormally distributed measures. To assess con-
founding, associations between the primary exposures, low sleep effi-
ciency (85%) and short sleep duration (6.5 hours), and sociodemo-
graphic characteristics also were examined. Spearman and Pearson
correlations assessed the strength of the linear relationship between
sleep characteristics obtained from polysomnography and actigraphy.
Logistic regression analyses were used to examine whether adolescents
with short sleep duration or low sleep efficiency were at increased odds
of prehypertension. Given the relatively small number of adolescents
with prehypertension, covariate adjustment was limited to the SES z
score and the 2 variables most strongly associated with prehypertension:
sex and BMI percentile. Multiple linear regression, adjusted for age, sex,
race, preterm status, BMI percentile, and SES z score, was used to
examine the linear associations between sleep duration or sleep effi-
ciency with continuously measured systolic and diastolic BP levels.
Additional analyses included low sleep efficiency from the polysom-
nography as the exposure. Residual confounding by snoring or the
apnea hypopnea index also was assessed by including these measures
as covariates in the adjusted analyses.
The authors had full access to the data and take full responsibility
for the integrity of the data. All authors have read and agree to the
manuscript as written.
Results
Characteristics of the analytic sample are shown in Table 1.
The average participant was 13.70.7 years of age. As
designed, the sample had an 50% representation of boys,
blacks, and children born prematurely. One fifth of the
sample was overweight. Approximately one-fourth reported
their household income as less than $20 000 per year.
Sixty-one adolescents (26%) had low sleep efficiency. Aver-
age weekday sleep duration was 7.71 hours, and 11% of the
sample slept 6.5 hours.
Sample characteristics stratified by prehypertension also are
shown in Table 1. Overall, 33 children (14%) met the criteria for
prehypertension, including 19 who had prehypertension and 14
who were hypertensive. Compared with normotensive adoles-
cents, those with prehypertension tended to more often be male,
tended to have a higher BMI, and were more frequently from
neighborhoods with a low median income (P0.05 to 0.10).
Both low sleep efficiency (P0.0001) and short sleep duration
(P0.06) were 2-fold more prevalent in those with prehyper-
tension compared with normotensive adolescents.
The distribution of various BP measures is further detailed
in Table 2. Using the mean of 9 BP readings, we classified
11% of the sample as having elevated systolic BP and 5%
as having elevated diastolic BP. All measures of systolic BP
were significantly higher among the adolescents with low
sleep efficiency compared with those with higher sleep
efficiency. Adolescents with low sleep efficiency also had a
higher prevalence of elevated diastolic BP and had higher 8
AM diastolic BP values. Adolescents with short sleep duration
did not differ from those with longer sleep duration in regard
to systolic BP but had a higher average diastolic BP and
higher prevalence of elevated diastolic BP (24.0% versus
2.4%; P0.001).
 1036 Circulation September 2, 2008

Table 1. Sample Characteristics

All Normotensive Prehypertension
(n238) (n205) (n33) P
Characteristics
Age, y 13.70.7 13.70.7 13.70.9 0.9782
Male, n (%) 123 (51.7) 101 (49.3) 22 (66.7) 0.0634
White race, n (%) 107 (45.0) 94 (45.9) 13 (39.4) 0.4887
Preterm status, n (%) 136 (57.1) 115 (56.1) 21 (63.6) 0.4167
Loud snoring, n (%) 41 (17.8) 36 (18.2) 5 (15.6) 0.7259
ADHD, n (%) 17 (7.2) 13 (6.4) 4 (12.1) 0.2683
Caffeine consumption after 6 PM, n (%) 61 (25.9) 53 (26.1) 8 (24.2) 0.8204
BMI percentile 72.2 (47.193.8) 70.9 (45.292.8) 87.9 (63.196.8) 0.0536
Overweight (BMI percentile 95th), n (%) 50 (21.0) 40 (19.5) 10 (30.3) 0.1579
Tanner stage 4, n (%) 172 (73.5) 148 (73.6) 24 (72.7) 0.9131
SES measures
Household income, n (%)
$20 000
Downloaded from http://circ.ahajournals.org/ by guest on April 15, 2017

62 (27.9) 53 (27.8) 9 (29.0)

$20 000$49 999 64 (28.8) 51 (26.7) 13 (41.9) 0.2689
$50 000 96 (43.2) 87 (45.5) 9 (29.0)
Neighborhood median income census tract, $1000 38.6 (23.652.4) 41.0 (24.155.2) 30.9 (23.243.1) 0.0617
Caregiver education, n (%) 0.5803
High school 19 (8.2) 15 (7.6) 4 (12.1)
High school or GED 47 (20.4) 41 (20.7) 6 (18.2)
High school 165 (71.4) 142 (71.7) 23 (69.7)
SES z score 0.000.80 0.030.81 0.170.74 0.1763
Actigraphy sleep characteristics, n (%)
Low sleep efficiency (85%) 61 (25.6) 43 (21.0) 18 (54.6) 0.0001
Short sleep (6.5 h) 25 (10.5) 18 (8.8) 7 (21.2) 0.0585
Sleep duration, n (%)33
6.5 h 25 (10.5) 18 (8.8) 7 (21.2)
6.517.49 h 75 (31.5) 65 (31.7) 10 (30.3) 0.1644
7.508.49 h 87 (36.6) 78 (38.0) 9 (27.3)
8.5 h 51 (21.4) 44 (21.5) 7 (21.2)
Sleep duration, h 7.711.03 7.750.99 7.461.23 0.1367
Polysomnography sleep characteristics
Arousal index 7.4 (6.09.4) 7.4 (6.19.4) 7.5 (6.010.4) 0.8343
Time in stage 34, % 33.811.6 34.111.4 32.313.0 0.4005
Time in REM, % 17.94.8 17.94.8 17.54.9 0.6563
Sleep efficiency, % 90.4 (85.594.0) 91.1 (86.594.2) 85.9 (80.389.6) 0.0002
Sleep efficiency (85%), n (%) 55 (23.1) 40 (19.5) 15 (45.5) 0.0010
GED indicates general educational development. Values are mean (%) or median (interquartile range) as appropriate.

To assess confounding, associations among the sleep
exposures and sociodemographic characteristics were exam-
ined (see the online Data Supplement). Adolescents with low
sleep efficiency had a higher BMI, were more often male, and
were from households with lower incomes and lower levels
of caregiver education. These characteristics were not signif-
icantly associated with short sleep duration. Approximately
two thirds (68.0%) of adolescents with short sleep duration
also had low sleep efficiency, whereas 27.9% of adolescents
with low sleep efficiency also had short sleep duration. The
correlation between mean weekday sleep efficiency and sleep
efficiency from the night of the polysomnography was low
(r0.13, P0.04), as was the correlation between mean
weekday sleep duration and sleep duration from the night of
the polysomnography (r0.06, P0.37). Approximately
one third (32.8%) of adolescents with low sleep efficiency as
assessed on actigraphy also had low sleep efficiency from the
polysomnography.
Results of the logistic regression models of the association
between each sleep measure and the odds of prehypertension
are shown in Tables 3 and 4. After adjustment for sex, BMI
percentile, and SES z score, those with low sleep efficiency
had 3.5 times the odds of prehypertension compared with
those without low sleep efficiency (95% CI, 1.54 to 7.96).
 Javaheri et al Sleep and Blood Pressure in Adolescents 1037

Table 2. BP in Subgroups Defined by Sleep Quality

Sleep Efficiency 85% Sleep Efficiency 85% Mean Sleep Duration Mean Sleep Duration
(n177) (n61) P 6.5 h (n213) 6.5 h (n25) P
Systolic BP
Systolic BP percentile 61.2 (42.777.8) 71.2 (53.191.1) 0.0012 63.2 (46.780.8) 66.4 (42.277.4) 0.9109
Elevated systolic BP (90th 11 (6.2) 16 (26.2) 0.0001 23 (10.8) 4 (16.0) 0.5005
percentile), n (%)
Mean systolic BP, mm Hg* 112.67.5 118.49.9 0.0001 113.87.9 116.512.6 0.3185
9 PM 114.68.8 121.911.7 0.0001 116.09.2 120.815.4 0.1420
8 AM 111.48.7 116.69.4 0.0001 112.68.9 114.010.9 0.4726
9:30 AM 111.98.2 116.811.5 0.0028 113.08.7 114.613.8 0.5635
Diastolic BP
Diastolic BP percentile 54.8 (41.670.6) 58.1 (42.674.3) 0.1695 55.2 (41.670.6) 58.0 (46.070.8) 0.3127
Elevated diastolic BP 5 (2.8) 6 (9.8) 0.0350 5 (2.4) 6 (24.0) 0.0003
(90th percentile), n (%)
Mean diastolic BP, mm Hg* 65.76.2 67.47.4 0.0816 65.86.3 68.68.2 0.0463
9 PM 64.19.1 65.211.0 0.4719 64.19.3 67.212.0 0.1224
Downloaded from http://circ.ahajournals.org/ by guest on April 15, 2017

8 AM 66.37.1 68.47.1 0.0416 66.67.2 69.16.6 0.0908

9:30 AM 66.77.1 68.59.0 0.1661 66.97.2 69.510.5 0.2308
Prehypertension, n (%) 15 (8.5) 18 (29.5) 0.0001 26 (12.2) 7 (28.0) 0.0585
Hypertension, n (%) 5 (2.8) 9 (14.8) 0.0019 9 (4.2) 5 (20.0) 0.0089
Values are meanSD or median (interquartile values) as appropriate.
*Average of all 9 values.

Short sleep duration was associated with a 2.8-fold increased
odds of prehypertension in unadjusted analyses (95% CI, 1.07
to 7.34), but this association was modestly attenuated after
adjustment for sex, BMI percentile, and SES z score (odds
ratio [OR], 2.54; 95% CI, 0.93, 6.90).
The unadjusted and adjusted associations between contin-
uously measured systolic and diastolic BP levels with sleep
efficiency are shown in Table 5. After adjustment for age,
sex, race, term status, BMI percentile, and SES z score, the
model predicts that each 5% increase in sleep efficiency was
associated with a 1.50.40-mm Hg decrease in systolic BP
(P0.001). Weaker associations were observed between
sleep efficiency and diastolic BP; ie, each 5% increase in
sleep efficiency was associated with a 0.650.35-mm Hg
decrease in diastolic BP (P0.05). When low sleep effi-
ciency was modeled as a dichotomous exposure, the adjusted
model estimates that adolescents with low sleep efficiency
had a mean systolic BP that was on average 3.99
1.24 mm Hg higher compared with those with higher sleep
efficiency (P0.002). Including sleep duration as a continu-
ously measured covariate did not alter the primary associa-
tions of sleep efficiency and BP (data not shown).
Similar to the models of prehypertension, sleep duration
was more weakly associated with continuously measured
systolic and diastolic BPs compared with sleep efficiency
(Table 6).
Additional Analyses
The primary analyses also were repeated with low sleep
efficiency ascertained via polysomnography as the exposure.
Consistent with the results of the primary analysis, after
adjustment for sex, BMI percentile, and SES z score, those
with polysomnography sleep efficiency 85% had nearly 3
times the odds of prehypertension as those with better sleep
(OR, 2.83; 95% CI, 1.28, 6.24). Also consistent with the
results of the actigraphy-defined sleep exposures, in adjusted
analyses, each 1% increase in sleep efficiency was associated
with a 0.200.06-mm Hg decrease in systolic BP (P0.001).
Similarly, those with low polysomnography sleep efficiency
had systolic BP that was 3.261.25 mm HG higher on
average compared with those with better sleep (P0.01).
Although analyses were restricted to children without
clinically significant sleep apnea, additional analyses as-
sessed potential residual confounding by snoring or the
apnea hypopnea index (ie, in an apnea hypopnea index
range of 0 to 4.9). The results show that loud snoring was not
significantly associated with prehypertension, systolic BP, or
diastolic BP. In contrast, although the apnea hypopnea index

Table 3. Association Between Low Sleep Efficiency and Odds of Prehypertension

Unadjusted OR (95% CI) P Adjusted OR (95% CI) P
Low sleep efficiency (85%) 4.52 (2.119.70) 0.0001 3.50 (1.547.96) 0.0028
Male 1.78 (0.784.04) 0.1702
BMI percentile (per 10-unit increase) 1.08 (0.931.26) 0.3098
SES z score 0.85 (0.511.42) 0.5339
 1038 Circulation September 2, 2008
Table 4. Association Between Short Sleep and Odds of Prehypertension
Unadjusted OR (95% CI)
Short sleep (6.5 h) 2.79 (1.077.34)
Male
BMI percentile (per 10-unit increase)
SES z score
does not confound the association between the outcomes and
the sleep exposures, it was associated with increased odds of
prehypertension after adjustment for low sleep efficiency,
sex, and BMI percentile; ie, for each 1-unit increase in the
apnea hypopnea index, the odds of prehypertension in-
creased by 47% (OR, 1.47; 95% CI, 1.00 to 2.17). Similarly,
the apnea hypopnea index was significantly associated with
systolic BP in adjusted models; after adjustment for subject
characteristics and low sleep efficiency, for each 1-unit
increase in the apnea hypopnea index, mean systolic BP
Downloaded from http://circ.ahajournals.org/ by guest on April 15, 2017
increases by 1.65 mm Hg on average (P0.009).
Discussion
To the best of our knowledge, this is the first reported associa-
tion between low sleep efficiency and short sleep duration
objectively measured in the childs usual sleep environment with
elevated BP (prehypertension or hypertension) in adolescents
without clinically significant levels of sleep apnea. Specifically,
adolescents with poor sleep quality, as measured by a sleep
efficiency of 85%, were at 3.5-fold increased odds of being
prehypertensive or hypertensive. Similar findings were observed
when single-night polysomnography was used to quantify sleep
efficiency. The association between low sleep efficiency and
prehypertension persisted even after adjustment for sex, SES,
and adiposity. The results did not change appreciably after
adjustment for snoring or the apnea hypopnea index. Short
sleep duration also was associated with a 2.5-fold increase in the
odds of prehypertension or hypertension. However, it was not
clear whether this association was attributable to the low sleep
efficiency found in a majority of the adolescents with short sleep
duration.
In adults, poor sleep quality identified by questionnaires has
been reported in association with an increased prevalence of
hypertension12 and an increased rate of nondipper hyperten-
sion.28 However, poor sleep quality in adults often occurs in the
presence of primary sleep disorders such as sleep apnea or
insomnia or secondary to numerous comorbidities. Therefore,
adult studies reporting associations with disturbed or reduced
sleep and hypertension have been cautiously interpreted because
of concerns about residual confounding.10 One large prospective
study reported associations of short sleep duration in women but
Table 5. Association Between Actigraphy Sleep Efficiency
(per 1% Increase) and Continuously Measured BP
Unadjusted Adjusted*
SE P SE P SE
Systolic BP 0.420.07 0.0001 0.300.08 0.0002
Diastolic BP 0.130.06 0.03 0.130.07 0.05
*Adjusted for age, sex, race, term status, BMI percentile, and SES z score.
P Adjusted OR (95% CI) P
0.0366 2.54 (0.936.90) 0.0679
2.20 (0.994.88) 0.0523
1.13 (0.981.31) 0.1042
0.75 (0.461.23) 0.2526
not men8; another study showed no association of hypertension
and sleep duration in the elderly, a group with a high prevalence
of morbidities.29 Because adolescents with major comorbidities,
including those with clinically significant levels of sleep apnea,
were excluded from our analyses (to minimize confounding and
to reduce measurement error), it is unlikely that major confound-
ing resulting from medical illnesses, medications, or sleep-
related hypoxemia explains the strong association between low
sleep efficiency and elevated BP. Given that the association
between BP and low sleep efficiency persisted even after
adjustment for average sleep duration, our findings also suggest
that recurrent arousals or awakenings from sleep (which reduce
sleep efficiency) are associated with elevated BP. Our findings
are consistent with a report from a sample of preadolescent
children studied with single-night polysomnography that dem-
onstrated an association between low sleep efficiency and
elevated BP after adjustment for the apnea hypopnea index.22
The 3.5-fold increased odds of prehypertension or hyper-
tension in children with low sleep efficiency, if causal,
suggests associations with a potential large public health
impact. Although the overall prevalence of low sleep effi-
ciency in general pediatric samples is unknown, our preva-
lence of 26% is likely an underestimate given the exclusion of
children with sleep disorders and significant comorbidities.
Our finding of an increased prevalence of low sleep effi-
ciency among vulnerable population subgroups such as
poorer children and those of minority ethnicity may be of
special concern because these groups are known to be at risk
for hypertension and other adverse health outcomes.
Low sleep efficiency was associated with an average
adjusted increase in systolic BP of 4 mm Hg. Although
limited data are available in children to interpret the clinical
significance of this absolute elevation, large cohort studies
suggest a log-linear increase in morbidity in association with
incremental changes in systolic BP.30
Short sleep duration was associated with a 2.5-fold increased
odds of prehypertension, an association attributable partly to low
sleep efficiency. Short sleep duration has been increasing in all
ages31 and is associated with an increased risk for obesity.13,16 18
Thus, efforts to optimize sleep in childhood may improve the BP
profile of children through obesity-dependent and -independent
Table 6. Association Between Actigraphy Sleep Duration
(per 1-Hour Increase) and Continuously Measured BP
Unadjusted Adjusted*
P SE P
Systolic BP 1.740.53 0.0012 0.980.52 0.06
Diastolic BP 0.600.41 0.15 0.410.44 0.34
*Adjusted for age, sex, race, term status, BMI percentile, and SES z score.
 Javaheri et al
pathways. Further work is needed to dissect the relative influ-
ences of sleep curtailment from sleep disruption on health
outcomes, which will be important in determining whether
future interventions would be best directed at improving sleep
time, sleep consolidation, or both.
The cause of low sleep efficiency in healthy adolescents is
unclear. Sensitivity analyses did not indicate an association
between low sleep efficiency and common childhood disor-
ders such as asthma or ADHD or caffeine or tobacco use, nor
were these variables confounders in the association between
sleep efficiency and BP (data not shown). It is possible that
unknown psychological disorders may have confounded our
results, but this seems unlikely given the strong associations
and community sampling design.
Although children with significant sleep apnea were excluded
from our analyses, the apnea hypopnea index (in a range of 0 to
4.9) was significantly associated with prehypertension and sys-
tolic BP after adjustment for sleep efficiency. This suggests that
even mild sleep-disordered breathing may contribute to abnor-
Downloaded from http://circ.ahajournals.org/ by guest on April 15, 2017
mal BP levels, a result consistent with reports of more severely
affected children from sleep clinic samples.21
Strengths of this report are the inclusion of a community-
based sample of children, minimizing referral biases, and the
use of objective measures of sleep duration and multiple
measures of BP, minimizing measurement error and reporting
biases. By characterizing numerous risk factors and comor-
bidities, we were able to restrict the analytical sample to
children without disorders likely to confound associations
with sleep quality. Although former preterm children were
overrepresented by design, there was no evidence of any
differences in the exposures, responses, or associations be-
tween preterm and full-term children, suggesting that our
results should be generalizable to other pediatric samples.
There are no established cutoffs to define thresholds of
sleep duration or sleep efficiency that increase morbidity in
adolescents. In adults, sleep durations of 6 hours have been
associated with a variety of adverse health outcomes,15,19,32,33
and sleep efficiencies of 85% are considered low. Our
choice for defining short sleep duration as 6.5 hours was to
approximate the cutoff associated with hypertension risk in
adults,10 which, in our sample, represented the lowest decile.
However, examination of a larger sample may permit a more
comprehensive assessment of dose response and threshold
levels for each sleep exposure.
A limitation of this cross-sectional study is that BP status
was determined from measurements made on 2 consecutive
days. Because BP may vary from day to day, repeated
measurements over time are needed to identify children with
persistent elevations in BP. Another limitation is that the
reported associations do not provide proof of causality. We
also cannot exclude the possibility that elevated BP operates
as a risk factor for poor sleep. It is important, however, to
interpret our findings in light of the biological plausibility of
the observed associations and experimental data that show
acute effects of sleep disruption on BP. Mechanisms linking
poor sleep efficiency or sleep deprivation with hypertension
may be through disruptions in cortisol secretion,19,34,35 stim-
ulation of the renin-angiotensin system and sympathetic
nervous system as measured by increased secretion of cate-
Sleep and Blood Pressure in Adolescents 1039
cholamines36 and abnormalities in sympathovagal balance,37
and abnormal secretion of vasoactive hormones, including
endothelin, vasopressin, and aldosterone.38 Experimental
sleep disruption has been associated with elevated BP in sleep
in normal subjects.39 Although some experimental models
suggest that sustained elevations in BP require sleep frag-
mentation to occur in a background of intermittent hypox-
emia40 (as occurs with sleep apnea), sleep fragmentation may
be associated with elevated BP even in adults with a low
apnea hypopnea index41 or with simple snoring.11 Prospec-
tive and interventional studies are needed to provide further
evidence of causality and to address whether improving sleep
quality and duration reduces BP and risk of hypertension.
Conclusions
Extensive analyses using objective measures of sleep quality
and duration and multiple measures of BP provide evidence for
a strong association of low sleep efficiency with increased risk of
prehypertension and hypertension in a healthy sample of ado-
lescents. Our data suggest that low sleep efficiency may be
associated more consistently with prehypertension than short
sleep duration. Future research is needed to address whether
prevention of hypertension in children should include not only
weight management and exercise but also optimization of sleep.
Our data underscore the need to monitor the quantity and quality
of sleep as part of health supervision in children.
Sources of Funding
This work was supported by National Institutes of Health (NIH)
grants HL07567, HL60957, RO1 NR02707, M01 RR00080, and
1U54CA116867.
Disclosures
Dr Rosen has received a subcontract from Advanced Brain Moni-
toring Inc to provide clinical research services funded through NIH
Small Business Innovative Research and has received honoraria for
society-sponsored educational talks. Dr Redline has received a
subcontract from Cleveland Medical Devices Inc to provide clinical
research services as part of NIH Small Business Innovative Research
funding. The other authors report no conflicts.
References
1. Din-Dzietham R, Liu Y, Bielo MV, Shamsa F. High blood pressure
trends in children and adolescents in national surveys, 1963 to 2002.
Circulation. 2007;116:1488 1496.
2. MacMahon S, Peto R, Cutler J, Collins R, Sorlie P, Neaton J, Abbott R,
Godwin J, Dyer A, Stamler J. Blood pressure, stroke, and coronary heart
disease, part 1: prolonged differences in blood pressure: prospective
observational studies corrected for the regression dilution bias. Lancet.
1990;335:765774.
3. MacMahon SW, Cutler JA, Neaton JD, Furberg CD, Cohen JD, Kuller
LH, Stamler J. Relationship of blood pressure to coronary and stroke
morbidity and mortality in clinical trials and epidemiological studies.
J Hypertens Suppl. 1986;4:S14 S17.
4. National High Blood Pressure Education Program Working Group on
High Blood Pressure in Children and Adolescents. The fourth report on
the diagnosis, evaluation, and treatment of high blood pressure in children
and adolescents. Pediatrics. 2004;114(suppl 4th report):555576.
5. Chobanian AV, Bakris GL, Black HR, Cushman WC, Green LA, Izzo JL
Jr, Jones DW, Materson BJ, Oparil S, Wright JT Jr, Roccella EJ. The
seventh report of the Joint National Committee on Prevention, Detection,
Evaluation, and Treatment of High Blood Pressure: the JNC 7 report.
JAMA. 2003;289:2560 2572.
6. Shen QB, Xu DL, Lin S, Lai WY. Sleep-disordered breathing and left
ventricular remodeling in patients with chronic heart failure [in Chinese].
Nan Fang Yi Ke Da Xue Xue Bao. 2006;26:486 489.
 1040 Circulation September 2, 2008

7. McNiece KL, Gupta-Malhotra M, Samuels J, Bell C, Garcia K, Poffenbarger
T, Sorof JM, Portman RJ. Left ventricular hypertrophy in hypertensive
adolescents: analysis of risk by 2004 National High Blood Pressure Edu-
cation Program Working Group staging criteria. Hypertension. 2007;50:
392395.
8. Cappuccio FP, Stranges S, Kandala NB, Miller MA, Taggart FM, Kumari
M, Ferrie JE, Shipley MJ, Brunner EJ, Marmot MG. Gender-specific
associations of short sleep duration with prevalent and incident hyper-
tension: the Whitehall II Study. Hypertension. 2007;50:693700.
9. Gangwisch JE, Heymsfield SB, Boden-Albala B, Buijs RM, Kreier F,
Pickering TG, Rundle AG, Zammit GK, Malaspina D. Short sleep duration
as a risk factor for hypertension: analyses of the first National Health and
Nutrition Examination Survey. Hypertension. 2006;47:833839.
10. Gottlieb DJ, Redline S, Nieto FJ, Baldwin CM, Newman AB, Resnick
HE, Punjabi NM. Association of usual sleep duration with hypertension:
the Sleep Heart Health Study. Sleep. 2006;29:1009 1014.
11. Lofaso F, Coste A, Gilain L, Harf A, Guilleminault C, Goldenberg F.
Sleep fragmentation as a risk factor for hypertension in middle-aged
nonapneic snorers. Chest. 1996;109:896 900.
12. Fiorentini A, Valente R, Perciaccante A, Tubani L. Sleeps quality dis-
orders in patients with hypertension and type 2 diabetes mellitus. Int
J Cardiol. 2007;114:E50 E52.
13. Cizza G, Skarulis M, Mignot E. A link between short sleep and obesity:
building the evidence for causation. Sleep. 2005;28:12171220.
Downloaded from http://circ.ahajournals.org/ by guest on April 15, 2017
24. Johnson NL, Kirchner HL, Rosen CL, Storfer-Isser A, Cartar LN, Ancoli-
Israel S, Emancipator JL, Kibler AM, Redline S. Sleep estimation using
wrist actigraphy in adolescents with and without sleep disordered
breathing: a comparison of three data modes. Sleep. 2007;30:899 905.
25. Redline S, Storfer-Isser A, Rosen CL, Johnson NL, Kirchner HL, Eman-
cipator J, Kibler AM. Association between metabolic syndrome and
sleep-disordered breathing in adolescents. Am J Respir Crit Care Med.
2007;176:401 408.
26. Marshall WA, Tanner JM. Variations in pattern of pubertal changes in
girls. Arch Dis Child. 1969;44:291303.
27. Marshall WA, Tanner JM. Variations in the pattern of pubertal changes in
boys. Arch Dis Child. 1970;45:1323.
28. Yilmaz MB, Yalta K, Turgut OO, Yilmaz A, Yucel O, Bektasoglu G,
Tandogan I. Sleep quality among relatively younger patients with initial
diagnosis of hypertension: dippers versus non-dippers. Blood Press.
2007;16:101105.
29. van den Berg JF, Tulen JH, Neven AK, Hofman A, Miedema HM,
Witteman JC, Tiemeier H. Sleep duration and hypertension are not
associated in the elderly. Hypertension. 2007;50:585589.
30. Lawes CM, Rodgers A, Bennett DA, Parag V, Suh I, Ueshima H,
MacMahon S. Blood pressure and cardiovascular disease in the Asia
Pacific region. J Hypertens. 2003;21:707716.
31. New Poll Finds Children Have Poor Sleep Habits, Parents Paying a High
Price. Washington, DC: National Sleep Foundation; 2004.

32. McEwen BS. Sleep deprivation as a neurobiologic and physiologic

14. Flint J, Kothare SV, Zihlif M, Suarez E, Adams R, Legido A, De Luca F.
Association between inadequate sleep and insulin resistance in obese
children. J Pediatrics. 2007;150:364 369.
15. Spiegel K, Knutson K, Leproult R, Tasali E, Van Cauter E. Sleep loss: a
novel risk factor for insulin resistance and type 2 diabetes. J Appl Physiol.
2005;99:2008 2019.
16. Taheri S. The link between short sleep duration and obesity: we should
recommend more sleep to prevent obesity. Arch Dis Child. 2006;91:
881 884.
17. Taheri S, Lin L, Austin D, Young T, Mignot E. Short sleep duration is
associated with reduced leptin, elevated ghrelin, and increased body mass
index. PLoS Med. 2004;1:e62.
18. Chaput JP, Despres JP, Bouchard C, Tremblay A. Short sleep duration is
associated with reduced leptin levels and increased adiposity: results from
the Quebec Family Study. Obesity. 2007;15:253261.
19. Spiegel K, Leproult R, Van Cauter E. Impact of sleep debt on metabolic
and endocrine function. Lancet. 1999;354:14351439.
20. Bonnet MH, Arand DL. We are chronically sleep deprived. Sleep. 1995;
18:908 911.
21. Marcus CL, Greene MG, Carroll JL. Blood pressure in children with
obstructive sleep apnea. Am J Respir Crit Care Med. 1998;157:
1098 1103.
22. Enright PL, Goodwin JL, Sherrill DL, Quan JR, Quan SF. Blood pressure
elevation associated with sleep-related breathing disorder in a community
sample of white and Hispanic children: the Tucson Childrens Assessment of
Sleep Apnea study. Arch Pediatr Adolesc Med. 2003;157:901904.
23. American Heart Association Recommendations for Human Blood
Pressure Determination and Sphygmomanometers: Report of a Subcom-
mittee of the Postgraduate Education Committee. Dallas, Tex: American
Heart Association; 1980.
stressor: allostasis and allostatic load. Metabolism. 2006;55:S20 S23.
33. Vgontzas AN, Zoumakis E, Bixler EO, Lin HM, Follett H, Kales A,
Chrousos GP. Adverse effects of modest sleep restriction on sleepiness,
performance, and inflammatory cytokines. J Clin Endocrinol Metab.
2004;89:2119 2126.
34. Spiegel K, Leproult R, LHermite-Baleriaux M, Copinschi G, Penev PD,
Van Cauter E. Leptin levels are dependent on sleep duration: relationships
with sympathovagal balance, carbohydrate regulation, cortisol, and thy-
rotropin. J Clin Endocrinol Metab. 2004;89:57625771.
35. Spath-Schwalbe E, Gofferje M, Kern W, Born J, Fehm HL. Sleep
disruption alters nocturnal ACTH and cortisol secretory patterns. Biol
Psychiatry. 1991;29:575584.
36. Lusardi P, Zoppi A, Preti P, Pesce RM, Piazza E, Fogari R. Effects of
insufficient sleep on blood pressure in hypertensive patients: a 24-h study.
Am J Hypertens. 1999;12:63 68.
37. Tochikubo O, Ikeda A, Miyajima E, Ishii M. Effects of insufficient
sleep on blood pressure monitored by a new multibiomedical recorder.
Hypertension. 1996;27:1318 1324.
38. Kato M, Phillips BG, Sigurdsson G, Narkiewicz K, Pesek CA, Somers VK.
Effects of sleep deprivation on neural circulatory control. Hypertension.
2000;35:11731175.
39. Davies RJ, Belt PJ, Roberts SJ, Ali NJ, Stradling JR. Arterial blood
pressure responses to graded transient arousal from sleep in normal
humans. J Appl Physiol. 1993;74:11231130.
40. Brooks D, Horner RL, Kozar LF, Render-Teixeira CL, Phillipson EA.
Obstructive sleep apnea as a cause of systemic hypertension: evidence
from a canine model. J Clin Invest. 1997;99:106 109.
41. ODriscoll DM, Morrell MJ. The interaction between respiratory and
autonomic function during sleep-related changes in pharyngeal airway
patency. Auton Neurosci. 2005;120:18 25.

CLINICAL PERSPECTIVE
Childhood hypertension is a risk factor for adult hypertension and for target-organ damage. Early recognition and intervention
of childhood hypertension are believed to be important in reducing the risk of cardiovascular morbidity in adulthood. Traditional
approaches for intervention focus on the role of overweight as a contributing cause of hypertension and include weight reduction,
increased physical activity, and nutritional changes. The present report identifies a significant association between increased
blood pressure and poor sleep quality (ie, increased wake time during the sleep period), found in 26% of a community sample
of adolescents. Independent of obesity, sex, and socioeconomic status and unrelated to sleep apnea, adolescents with poor sleep
had a 3.5-fold increased risk of prehypertension or hypertension. This finding suggests that approaches for optimizing sleep
quality and duration in children may complement other behavioral approaches for preventing or treating pediatric hypertension.
Monitoring sleep quality and duration in children as part of their health supervision may help to identify children who are at risk
for both sleep problems and hypertension and who would benefit from behavioral interventions aimed at improving sleep.
 Sleep Quality and Elevated Blood Pressure in Adolescents
Sogol Javaheri, Amy Storfer-Isser, Carol L. Rosen and Susan Redline

Circulation. 2008;118:1034-1040; originally published online August 18, 2008;

doi: 10.1161/CIRCULATIONAHA.108.766410
Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Downloaded from http://circ.ahajournals.org/ by guest on April 15, 2017

Copyright 2008 American Heart Association, Inc. All rights reserved.

Print ISSN: 0009-7322. Online ISSN: 1524-4539

The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://circ.ahajournals.org/content/118/10/1034

Data Supplement (unedited) at:

http://circ.ahajournals.org/content/suppl/2008/08/25/CIRCULATIONAHA.108.766410.DC1

Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Circulation 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 Circulation is online at:

http://circ.ahajournals.org//subscriptions/