Download Resting Heart Rate and the Risk of Heart Failure in Healthy Adults

Resting Heart Rate and the Risk of Heart Failure in Healthy Adults:
The Rotterdam Study
Nanchen et al: Resting Heart Rate and Heart Failure
David Nanchen, MD, MSc;1 Maarten J.G. Leening, MD, MSc;2,3 Isabella Locatelli, PhD;1
Jacques Cornuz, MD, MPH;1 Jan A. Kors, PhD;4 Jan Heeringa, MD, PhD;2
Jaap W. Deckers, MD, PhD;3 Albert Hofman, MD, PhD;2 Oscar H. Franco, MD, PhD;2
Bruno H.Ch. Stricker, MB, PhD;2,4,5,6 Jacqueline C.M. Witteman, PhD;2
Abbas Dehghan, MD, PhD2
Downloaded from http://circheartfailure.ahajournals.org/ by guest on May 6, 2017
1
Department of Ambulatory Care and Community Medicine, University of Lausanne,
Lausanne, Switzerland
2
Department of Epidemiology, Erasmus Medical Center, Rotterdam,
tterdam,, the Ne
Net
Netherlands
3
Department of Cardiology, Erasmus Medical Center, Rotterdam,
rddaam, tthe
he N
Netherlands
ethe
ethe
et
herr
4
Department of Medical Informatics, Erasmus Medical Center,
er, Rotterdam,
Rott
Ro
tter
tt
erda
er
dam,
da
m tthe
m,
Netherlands
5
Department of IInternal
nternal Me
nt
Medi
Medicine,
dici
di
cine
ci
ne,, Er
ne
Eras
Erasmus
asmu
as
muss Med
mu
Me
Medical
edica
call Ce
ca
Cent
Center,
n err, Ro
nt
Rotterdam,
ott
tter
erda
er
dam,
da
m, tthe
h Netherlands
he
6
Inspectorate off H
Health
e lth Ca
ea
Care,
are, T
The
hee H
Hague,
agu
gue,
e,, the
he N
Netherlands
ethe
herl
he
rlaands
rl
dss
Correspondencee tto:
o:
David Nanchen, M
MD
MD,
D M
MSc
Sc
Department of Ambulatory Care and Community Medicine
University of Lausanne
Rue du Bugnon 44
CH-1011 Lausanne, Switzerland
Tel: + 41 21 314 60 53
Fax: +41 21 314 61 08
E-mail: [email protected]
DOI: 10.1161/CIRCHEARTFAILURE.112.000171
Journal Subject Codes: Etiology:[8] Epidemiology, Heart failure:[110] Congestive
1
Abstract
Background—An elevated resting heart rate is associated with re-hospitalization for
heart failure and is a modifiable risk factor in heart failure patients. We aimed to examine
the association between resting heart rate and incident heart failure in a population-based
cohort study of healthy adults without pre-existing overt heart disease.
Methods and Results—We studied 4,768 men and women aged 55 years or older from
the population-based Rotterdam Study. We excluded participants with prevalent heart
Downloaded from http://circheartfailure.ahajournals.org/ by guest on May 6, 2017
failure, coronary heart disease, pacemaker, atrial fibrillation, atrio-ventricular block, and
those using beta-blockers or calcium channel blockers. We used
Cox-models
sed extended C
o
allowing for time-dependent variation of resting heart rate along
Over a
ong ffollow-up.
ollo
ol
l wlo
w-up
up.. O
up
median of 14.6 years
The risk of
yea
yea
ears of follow-up,
foll
fo
llow
ll
ow-u
ow
-up,
-u
p, 656
656 participants
parrti
tici
c pa
pantss developed
dev
evel
eloped
elop
el
opped
e heart
hea
eart
rt failure.
fai
ailu
lure
lu
re. T
re
heart failure wass higher
high
hi
gher
gh
err in
in men
men with
with higher
hig
i he
herr resting
rest
re
sttin
ng heart
hear
he
art rate.
ar
rate
ra
te. For
te
For ea
each
ch iincrement
nccreem of 10
ncre
beats per minute,
multivariable
ratios
e th
e,
thee mu
mult
ltiv
lt
ivar
iv
aria
ar
iabl
ia
blee adjusted
bl
adju
ad
j st
ju
sted
ed
d hazard
haz
azar
ardd ra
ar
rati
tios
ti
os iin
n me
menn we
were
re 11.16
.16
16 ((95%
confidence interval [CI], 1.05-1.28, p=0.005) in the time-fixed heart rate model and 1.13
(95% CI 1.02-1.25, p=0.017) in the time-dependent heart rate model. The association
could not be demonstrated in women (p for interaction = 0.004). Censoring participants
for incident coronary heart disease, or using time-dependent models to account for the
use of beta-blockers or calcium channel blockers during follow-up did not alter the
results.
Conclusions—Baseline or persistent higher resting heart rate is an independent risk
factor for the development of heart failure in healthy older men in the general population.
Key Words: resting heart rate, heart failure, population-based, cardiovascular disease
2
Heart failure is a major public health problem that is associated with substantial mortality
and morbidity.1 In the aging population of Western countries, the incidence of heart
failure hospitalization is increasing despite the improvements in health care for heart
failure patients.2 Therefore, it is important to develop strategies to detect adults at risk of
heart failure in order to better tailor preventive measures and treatment.
Resting heart rate is a very accessible biological parameter with potential
predictive capacity for heart failure and cardiovascular disease.3 Among patients
Downloaded from http://circheartfailure.ahajournals.org/ by guest on May 6, 2017
suffering from heart failure, resting heart rate is a modifiable risk factor to prevent reassociation
ion
o between
hospitalization for heart failure.4 However, studies examiningg the associati
resting heart rate and heart failure have not included adults from
rom th
thee ge
gene
general
nera
ne
rall ppopulation,5
ra
or have included
d aadults
dults w
with
ithh ca
it
card
cardiovascular
rdio
rd
iova
io
vasc
va
scul
sc
u arr ddisease,
ul
isea
easee,6 or
or conduction
cond
co
nduc
nd
u ti
tion
ion disorders
dissor
orde
derss such as
de
n 7 oorr adults
n,
ad
duulltss uusing
sing
si
n aanti-arrhythmics
ng
ntint
i-arrh
irhyt
rh
ythm
ythm
yt
hmic
iccs orr bbeta-blockers.
etaet
a-bl
abloc
bl
ocke
oc
kers
ke
rs.8 Th
rs
This
iss rresearch
e
atrial fibrillation,
question is important
ort
ort
rtan
antt to sstudy
an
tudy
tu
dy
y iin
n a pr
pprospective
ospe
os
p ct
pe
ctiv
ivee ma
iv
mann
manner
nner
nn
er bbecause
ecau
ec
ause
au
se iin
n ca
card
cardiac
rdia
rd
iacc pa
ia
pat
patients,
t
a
subclinical decompensated state might enhance a hemodynamic response that increases
heart rate. This biological interaction between resting heart rate and subclinical heart
failure may limit the interpretation of the role of heart rate in the etiology of heart failure,
because of potential reverse causality. Thus, it is still uncertain whether the association
between resting heart rate and heart failure can be extrapolated to healthy adults from the
general population.
We examined whether higher resting heart rate is independently associated with
the development of heart failure among adults without pre-existing heart disease or heartrate modifying medication use in the general population.
3
Methods
Study sample
This study was performed within the framework of the Rotterdam Study, a prospective
population-based cohort study designed to evaluate the determinants and consequences of
chronic diseases in the elderly. Details regarding the objectives and methods of the
Rotterdam Study have been reported previously.9 Briefly, all inhabitants aged 55 and
over of a well-defined suburb in the city of Rotterdam, the Netherlands, were invited to
Downloaded from http://circheartfailure.ahajournals.org/ by guest on May 6, 2017
participate and 7,983 (78%) were enrolled. The Medical Ethics Committee of the
inf
n
Erasmus Medical Center approved the study and participants gave written informed
m ttheir
heir
he
irr ttreating
reat
re
atin
at
i
consent to participate in the study and to obtain information from
rate
ra
t ly. From
F om 1990
Fr
199
9900 until
unti
un
till 1993,
ti
119993
9 , baseline
baseeli
ba
line
ne data
dat
a a were
w ree collected
we
col
olle
lect
le
ect
cted
ed using
uss
physicians, separately.
memee-in
inte
in
teerrvvieews and
andd established
an
e ta
es
tabllis
ishe
hedd cardiovascular
he
card
ca
rd
dio
iova
vasc
va
scul
sc
ular
ul
a risk
ar
rissk factors
fact
fa
ctor
orss were
w
we
standardized home-interviews
ses
esse
sedd at the
se
the research
res
esea
earc
ea
rchh center.
rc
cent
ce
nter
nt
er.
er
subsequently assessed
Of the 7,129 individuals in the Rotterdam Study who visited the research center at
baseline, heart rate measurement was available for 6,966 participants. To account for
endogenous heart rate variation only, we further excluded 873 participants using betablockers and 173 using calcium channel blockers. Because we aimed to examine the
association between heart rate and the development of heart failure in participants free of
heart disease, we also excluded participants with a pacemaker, second or third degree
atrio-ventricular block on the baseline electrocardiogram (ECG), a history of heart
failure, atrial fibrillation, and those with known coronary heart disease (CHD) defined as
a history of myocardial infarction, coronary artery bypass grafting, or percutaneous
coronary intervention.10 The final sample for analysis comprised 4,768 participants.
4
Measurement of heart rate
In all participants, baseline resting pulse was measured for 30 seconds at the right radial
artery between two consecutive blood pressure measurements at the research center, with
the subject in a sitting position. To obtain heart rate in beats per minute (bpm), the
obtained pulse count was multiplied by two. Repeated examinations of heart rate were
performed every 3-4 years and were available for 3 additional follow-up visits, from 1993
to 1995, from 1997 to 1999, and from 2002 to 2004. For sensitivity analysis, resting heart
Downloaded from http://circheartfailure.ahajournals.org/ by guest on May 6, 2017
rate was further measured from a 10 seconds 12-lead ECG at baseline in 4,127
participants. Repeated measurements of ECG were available for all 3 addit
additional
tio followup visits. All ECGs were recorded with an ACTA Gnosis IV ECG
CG
G rrecorder
ecor
ec
orde
or
derr ((Esaote
de
Biomedica, Florence,
Italy)
All
enc
nce, Ita
aly
y) att a sampling
sam
ampl
plin
pl
ingg frequency
in
frreq
e ue
uenc
nccy of 500
500
0 Hz
H aand
nd sstored
tore
to
ore
redd di
ddigitally.
g
11
ECGs were processed
using
the
validated
modular
ECG
system
c se
cesse
sedd us
usin
ingg th
in
he va
ali
l da
datted m
odul
od
ular
ul
ar E
CG aanalysis
n ly
na
lysi
s s sy
si
syst
stem ((MEANS).
st
stem
ME
ME
MEANS locates th
thee QR
QRS
S co
comp
complexes
mp
ple
lexe
xess an
xe
andd de
dete
determines
term
te
rmin
rm
ines
in
es a sstable
tabl
ta
blee re
bl
refe
reference
fere
fe
renc
re
ncee po
nc
poin
point
intt in each
in
complex. The QRS detector of MEANS operates on multiple simultaneously recorded
leads, which are transformed to a detection function that brings out the QRS complexes
among the other parts of the signal. RR intervals are taken as the intervals between the
reference points in adjacent QRS complexes. The median RR interval was computed,
after exclusion of RR intervals that immediately precede and follow any premature
ventricular complex.
Heart failure assessment
Prevalent and incident heart failure was determined as defined previously.1, 10, 12 For
prevalent cases, a validated score was used based on the heart failure definition of the
5
European Society of Cardiology.13 This score was based on the presence of at least two
signs or symptoms suggestive of heart failure or use of medication for the indication of
heart failure.10 Cases of incident heart failure were obtained by continuously monitoring
participants during follow-up through automated linkage with files from general
practitioners. All available data on heart failure, such as hospital discharge letters and
notes from general practitioners, were copied from the medical records. Heart failure was
adjudicated in accordance with the criteria of the European Society of Cardiology13 based
Downloaded from http://circheartfailure.ahajournals.org/ by guest on May 6, 2017
on the combination of signs and symptoms, and objective evidence of cardiac
y. Two indep
pen
end research
dysfunction, including chest radiographs or echocardiography.
independent
disa
s gr
sa
gree
eeme
ee
m the
physicians adjudicated all potential heart failure cases. In casee off di
disagreement
ard
ard
rdiiologiistt was
was sought
sou
ough
ghtt and
gh
and considered
c nssid
co
der
ered
red ddecisive.
ecis
ec
isiv
is
ive.10 O
iv
Only
n y de
nl
defi
definite
finii and
fi
judgment of a cardiologist
o heart
hear
he
a t failure
ar
fail
fa
illur
uree were
weree included
inc
nclu
ude
dedd inn the
the analyses.
ana
naly
lyse
ly
ses.
se
s Th
Thee date
date of
of incident
inci
in
ci
probable cases of
heart
failure was the fi
first
irs
rstt oc
occu
occurrence
curr
cu
rren
rr
ence
en
ce of ssymptoms
y pt
ym
p om
omss su
sugg
suggestive
gges
gg
esti
es
tive
ti
ve ooff he
hear
heart
artt fa
ar
fail
failure,
ilur
il
uree, oorr th
ur
the
h day of
receipt of a first prescription of a loop diuretic or an angiotensin-converting enzyme
inhibitor for heart failure. For the present analysis, incident heart failure was adjudicated
until 1st January, 2009.
Covariables
Hypertension was defined as a systolic blood pressure •140 mmHg or a diastolic blood
pressure •90 mmHg or use of blood pressure lowering drugs with the indication of
hypertension. Diabetes mellitus was defined as a random or post-load blood glucose
measurement exceeding 11.0 mmol/L, or the use of anti-diabetic drugs. Body mass index
was calculated by dividing measured weight by height squared. Assessment of incident
6
fatal and non-fatal CHD has been previously described.10 For the present analysis,
incident CHD events, including myocardial infarction and myocardial revascularization
procedures have been adjudicated until 1st January, 2009. For all medication use from 1st
January 1991 onwards, eight fully automated pharmacies in the research area provided
data on use, dosage, duration of use, and date of first prescription, using a single
computer network to register all prescriptions, as previously described.9, 10 Moreover,
data on medication use were collected during the baseline home interview.
Downloaded from http://circheartfailure.ahajournals.org/ by guest on May 6, 2017
Statistical analysis
es. This
Thi
hiss was
was done
donn separately
do
Resting heart rate values at baseline were grouped into tertiles.
for men and women,
m n, because
men
me
beccau
ause
s women
wom
omen
en are
are known
kno
n wnn too have
no
have a hhigher
igheer re
ig
rest
resting
s in
ng he
hear
heart
r rate than
ANO
NOVA
VA and
and Ȥ2 tes
tests
ests
es
tss orr th
thee Kr
Krus
Kruskal-Wallis
uska
us
kalka
l-Wa
lWall
Wa
llis
ll
is rank
ran
a k test
tesst were
te
wer
eree used
u
for
men.14 One-way ANOVA
baseline comparisons
iso
sons
ns bbetween
etwe
et
ween
we
en ttertiles.
erti
er
tile
less. W
Wee ccalculated
alcu
al
cula
cu
late
tedd in
te
inci
incidence
cide
denc
ncee ra
nc
rate
rates
tess pe
te
pperr 11,000
,00
00
0 personyears of follow-up in participants categorized according to their resting heart rate and
constructed Kaplan-Meier cumulative incidence curves. The associations between
baseline resting heart rate and heart failure were examined in a time-fixed analysis using
multivariable Cox proportional hazard models. A quadratic mixed-effect model was used
to estimate change in average heart rate per year of follow-up. Among different
polynomial models, we chose the quadratic one which minimized the BIC criterion.
Moreover, we used an extension of the Cox model allowing for introduction of repeated
measurements of heart rate over time in a time-dependent analysis.15 With this version of
the Cox model, the risk of event at any time t depends upon the last available measure of
heart rate before time t.
7
P for trend across categories was obtained by entering heart rate tertiles as a
continuous variable. In the first model, we only adjusted for age. In the second model, we
entered traditional cardiovascular risk factors, including age, smoking status, systolic
blood pressure, antihypertensive treatment, diabetes mellitus, body mass index, total
cholesterol, and high-density lipoprotein cholesterol.16 Among participants, 302 (6.3%)
had missing values for one or more covariables. The multivariable adjusted analysis was
restricted to those participants with complete information on cardiovascular risk factors at
Downloaded from http://circheartfailure.ahajournals.org/ by guest on May 6, 2017
baseline. Participants were followed until the occurrence of heart failure, death, or the
etween heart rrate
a and
at
end of the study period. In order to evaluate the association between
nally
ly
y ccensored
e so
en
sore
redd pparticipants at
re
incident heart failure not mediated through CHD, we additionally
of nnon-fatal
on-faataal CH
HD
D..17 F
Furthermore,
urth
ur
th
herrmo
m ree, we cconducted
ondu
on
d ct
du
cted
e ttime-dependent
imeim
e de
depe
pend
pe
n
analyses
the occurrence of
CHD.
oci
oci
ciat
atio
at
i n between
io
betw
be
tw
wee
eenn heart
hear
he
arrt rate
raate and
and incident
inci
in
cide
ci
dent
de
nt heart
hea
e rt
rt ffailure
ailu
ai
lure
lu
re nnot
o m
ot
ed
to assess the association
mediated
by the
use of heart rate influencing
infl
in
flue
fl
uenc
ue
ncin
nc
ingg medication,
in
medi
me
dica
di
cati
ca
tion
ti
on,, as the
on
the indication
indi
in
dica
di
cati
ca
tion
ti
on for
for the
the pr
pprescription
escr
es
crip
cr
ipti
ip
tion
ti
onn of betablockers or calcium channel blockers might be associated with the risk of heart failure
and these medications affect resting heart rate measurements. This was done by changing
the exposure of the participants whenever they filled-out a prescription of beta-blockers
or calcium channel blockers during follow-up.18 In the sensitivity analyses, we repeated
all analyses using time-fixed and time-dependent resting heart rate measured by ECG,
since ECG-derived heart rate is more accurate.19 We report estimates with 95%
confidence intervals (CIs). All hypothesis tests are two-sided and the significance level
set at 5%. Statistical analyses were performed using STATA statistical software®
(Version 12, STATA Corp, College Station, Tex) and R (Version 2.15.2, http://www.rproject.org).
8
Results
Gender-specific baseline characteristics with respect to resting heart rate categories are
presented in Table 1. Mean age of the participants was 68.5 years and 62% were females.
In men, heart rate was 68 bpm or less for the first tertile, 69 to 78 bpm for the second
tertile, and 79 bpm or more for the third tertile. In women, these rates were 72 bpm or
less, 73 to 80 bpm, and 81 bpm or more, respectively. Participants in the highest heart
rate tertile were more likely to smoke, and to have elevated blood pressure, and diabetes
Downloaded from http://circheartfailure.ahajournals.org/ by guest on May 6, 2017
mellitus. No differences in the use of anti-hypertensive drugs was found across heart rate
categories, but a higher heart rate was more frequently noted in male userss of
o antiasthmatics and female users of oral corticosteroids. Average resti
resting
tiing hheart
eart
ea
rt rrate
at decreased
similarly in both
during
tended
h ge
genderss du
duri
ing ffollow-up,
olllo
low
w-u
up, bbut
ut ten
en
nde
dedd to
o sstabilize
taabi
b lize
ze oover
verr ti
ve
time
me
me
(Supplemental Figu
Figure).
F
ure
re).
)
).
During a m
median
edia
ed
iann (i
ia
((interquartile
nter
nt
erqu
er
q ar
qu
arti
tile
ti
le range)
rangge) 14.6
14.66 (7.6)
(7.6)
6) years
y ar
ye
arss of follow-up,
fol
ollo
lowlo
w-up
wup, 656
up
6
participants developed incident heart failure. Crude incidence rates of heart failure were
higher in men with higher heart rate than in men with lower heart rate, with 13.7 versus
9.9 per 1,000 person-years. In women, crude incidence rates of heart failure were similar
across heart rate categories, except when heart rate was measured with ECG (Figure).
Accordingly, there were remarkable differences in adjusted hazard ratios (HRs) between
men and women. The multiplicative interaction term between heart rate as a continuous
variable and gender was significant both in the age- and gender adjusted model (p=0.011)
and in the multivariable adjusted model (p=0.004). For each increment of 10 beats per
minute, the multivariable adjusted HRs in men were 1.16 (95% confidence interval [CI],
1.05-1.28) in the time-fixed heart rate model and 1.13 (95% CI 1.02-1.25) in the time-
9
dependent heart rate model (Table 2). Further adjustment for use of anti-asthmatic drugs
and corticosteroids use did not change the estimates. In women, resting heart rate was not
associated with a higher heart failure risk (Table 2).
To assess the heart failure risk not mediated by CHD, we additionally censored
328 participants at the occurrence of non-fatal CHD during follow-up. This yielded
similar results (Table 3). We also examined the risk of heart failure taking into account
follow-up time until the first prescription of common heart rate lowering drugs. This
Downloaded from http://circheartfailure.ahajournals.org/ by guest on May 6, 2017
represented 1,700 first prescriptions, including 248 prescriptions prior to the diagnosis of
heart failure. In this analysis, men with a heart rate in the upper
compared
to those
per tertile comp
mpaa
mp
with a heart rate in the lower tertile had a higher risk of heart failu
failure
with
multivariable
lu
ure w
ithh a m
it
adjusted HR of 1.4
CII 11.
1.08
11.47,
.477, 95%
.4
%C
.08 tto
o 2.01
2.01
2.
01. Re
Relative
R
lati
tive
vee risk
risk
issk estimates
esti
es
tima
ti
matees were
ma
we e not
not significant
sii
and
omen ((Table
om
Tabl
Ta
ble 33)
).
unchanged in women
3).
Sensitivity analysis (heart rate measured with ECG)
Using resting heart rate measured by ECG instead of pulse measurement, baseline heart
rate was generally somewhat lower resulting in 62 bpm or less for the first tertile, 63 to
73 bpm for the second tertile, and 74 bpm or more for the third tertile in men, and 69 bpm
or less, 69 to 76 bpm, and 77 bpm or more in women, respectively (Supplemental Table).
However, for each increment of 10 bpm in men, we still found a significant 13% higher
heart failure rate for the time-fixed heart rate model and 18% higher rate for the timedependent heart rate model (Table 4). In women, the association did not reach statistical
significance in the multivariable models. Further excluding 10 men and 7 women with a
resting heart rate lower than 50 bpm, e.g. with bradycardia, yielded similar results.
10
Discussion
In a large population-based cohort study of healthy adults, higher resting heart rate
measured with pulse palpation (and alternatively by ECG) was independently associated
with incident heart failure in men during follow-up, and this association was not mediated
through overt CHD. The association was found significant in men both with single and
repeated heart rate measurement over time. In women, the association could not be
Downloaded from http://circheartfailure.ahajournals.org/ by guest on May 6, 2017
demonstrated. Our study highlights the importance of resting heart rate as an independent
marker of future heart failure in middle-aged and older healthy
hy men from tthe
h general
he
population.
Previous sstudies
reported
that
rate
tu
udies hhave
ave re
av
repo
porrte
po
tedd th
hat
a hheart
e rt rat
ea
atte is a prognostic
p og
pr
ogno
ost
stic
icc and
nd potentially
pot
oe
modifiable marker
k r in patients
ker
pat
atie
ient
ie
ntss suffering
nt
suufffer
erin
ingg from
in
from heart
hea
eart
rt failure
fai
ailu
lure
lu
re or
or CHD.
CHD
CH
D.3 IIn
n a co
convenient
onv
sample of patients
nts w
with
ithh CH
it
CHD,
D, D
Diaz
iazz an
ia
andd co
coll
colleagues
lleeag
ll
gue
uess de
demo
demonstrated
mons
mo
nstr
ns
trat
tr
ated
at
ed tthat
hatt he
ha
hear
heart
artt ra
ar
rate
a was
associated with incident heart failure hospitalization.5 However, both the basic
hemodynamic response to a decompensated state and the excessive neuroendocrine
activation cause tachycardia in patients with subclinical heart failure.20 Thus, the
etiologic nature of the association between heart rate and incident heart failure can only
be examined in subjects without pre-existing overt cardiac disease at baseline. In line
with our results, investigators from the EPIC-Norfolk study recently reported on a
positive association between a single heart rate measurement and heart failure in middleaged adults from the general population.7 In addition to the EPIC-Norfolk study, we
could demonstrate that the association was not influenced by variation of resting heart
rate over time or by the measurement method used to assess heart rate. Furthermore,
11
participants of the EPIC-Norfolk study may still have had subclinical heart disease,
because they relied on use of medication only to define prevalent heart failure cases, as
well as self-reported CHD, and they did not exclude adults with pre-existing atrial
fibrillation or other conduction disorders based on ECG. Finally, the EPIC-Norfolk study
examined only incident heart failure hospitalization, without including diagnoses of heart
failure made by general practitioners or nursing home physicians as in our study.
As confirmed in our data, resting heart rate is higher in women than in men.14
Downloaded from http://circheartfailure.ahajournals.org/ by guest on May 6, 2017
Gender differences have also been proposed in heart failure etiology: in men CHD is
en hypertension
on plays a more
considered the most important determinant, whereas in women
owed
ed
d a stronger
sttro
stro
rong
nger
ng
e association
predominant role.21 In our study sample, resting heart rate showed
in men than in women
wom
wom
omen for
orr the
the development
dev
evel
elop
el
opme
op
mennt of
me
o heart
heart
rtt failure,
fai
ailu
uree, but
b t hy
bu
hype
hypertension
pertten
pe
ensi
sion
si
on was
associated with inci
iincident
cide
ci
dent
de
nt hheart
eart
ea
r ffailure
rt
a lu
ai
ure
r bo
bboth
th iin
n me
menn an
andd wo
wome
women
menn wi
me
with
without
thou
th
outt a si
ou
significant
ign
gnii
gender differencee (p ffor
or iinteraction
nter
nt
erac
er
acti
ac
tion
ti
on 00.77).
.77)
77)). Re
Rest
Resting
stin
st
ingg he
in
hear
heart
artt ra
ar
rate
te iin
n wo
wome
women
menn ma
me
mayy bbe less
precise due to more important endogenous variations and previous experiences with
cardiac volume overload during pregnancy. Indeed, the healthy female heart might have
an advantage to adapt to elevated heart rate and volume overload related to pregnancy,
offering protection against heart failure.22 Our results are in line with those from the
EPIC-Norfolk study examining adults from the general population, in which the
investigators could not find a statistically significant association between heart rate and
heart failure in women not taking heart rate lowering drugs, with a p for trend of 0.055
across heart rate categories.7
We could further demonstrate that higher heart rate was an independent marker of
heart failure that was not preceded by CHD, or by the use of beta-blockers or calcium
12
channel blockers during follow-up. Through mechanisms linked to oscillatory shear
stress in the coronary arteries, an elevated heart rate might promote coronary
atherosclerosis.3 In our study, it seems unlikely that the higher incidence of heart failure
that we observed in adults with a higher heart rate was mediated by CHD. Current
guidelines for the prevention of heart failure classify heart rate as a minor clinical risk
factor, probably because the pathophysiological pathways between heart rate and heart
failure remain speculative.23 Recently, two clinical trials reported the additional benefit of
Downloaded from http://circheartfailure.ahajournals.org/ by guest on May 6, 2017
a specific heart rate lowering drug on outcomes in patients with heart failure and heart
men seemed too derive
d
equal
rate above 75 bpm.24, 25 Interestingly, in the SHIFT trail, women
benefit than men from the heart rate lowering drug to avoid re-hospitalization
e-ho
ospi
pita
taali
liza
zati
za
tion
ti
o for heart
failure.25 Specific
ic hheart
ic
eart ra
rate
tee red
reducing
educ
ed
ucin
uc
ng th
ther
therapy
e ap
er
py sh
should
hould
ld bbee co
cconsidered
ons
nsid
ns
sid
i erred ffor
orr eevaluation
valu
va
l a
in
ithh elevated
elev
el
ev
evat
vated
ted heart
hea
eart
r rate
rat
atee to
to prevent
prev
pr
even
ev
en
nt or ppostpone
ostp
os
tpon
tp
onee th
on
thee development
deve
de
velo
lopm
pm
pmen
men of heart
healthy adults with
failure.
Strengths of our study include the large sample size in a population-based setting
with a long follow-up, the repeated measurement of resting heart rate over time, the
concurrent assessment of resting heart rate by means of both arterial pulse and ECG
recording, as well as the large number of heart failure cases adjudicated using
standardized definitions. The detection of heart failure relied not only on hospitalization,
but also included diagnoses of heart failure made by general practitioners and nursing
home physicians. This systematic approach potentially reduced selection of severe cases
only. Previous studies have evaluated risk factors for incident heart failure based on ECG
measurement.6, 8 The ECG provides important clinical information other than heart rate,
such as presence of left ventricular hypertrophy, bundle branch blocks, or ST-T
13
abnormalities that might together improve cardiovascular risk prediction.26 In our study,
we showed that a single or repeated assessment of heart rate based on arterial pulse was
sufficient to identify men at higher risk of developing heart failure. However, our study
also has some limitations. The association between heart rate and heart failure in our
population of healthy older adults suggests that heart rate is not merely a surrogate
marker of an underlying process leading to heart failure. However, even with a
comprehensive assessment of traditional cardiovascular risk factors, we cannot rule out
Downloaded from http://circheartfailure.ahajournals.org/ by guest on May 6, 2017
residual confounding. For example, we could not account for the circadian variation of
iation of heartt rrate
a is small
resting heart rate in our analyses. However, the circadian variation
ults.27 S
Similarly
im
mil
ilar
arly
ar
ly to previous
between 10 AM and 6 PM, and tends to diminish in older adults.
reports,6, 8 we were
ere also unable
ere
un
nab
a le to
to account
accco
coun
untt for
un
forr the
fo
th
he degree
degr
de
gree
gr
ee of
of physical
physsic
ph
ical
al activity
act
ctiv
ivit
iv
ityy performed
it
a t, and
ant
andd therefore
there
heere
r fo
fore
r the
the
h associations
asssoci
ciat
atio
at
io
ons
n found
fou
ound
nd with
wiitth heart
hear
he
artt rate
ar
rraate
t may
may
y be
by each participant,
confounded by ph
pphysical
phys
ysic
ys
ical
ic
al ffitness.
itne
it
ness
ne
ss. Fi
ss
Fina
Finally,
nall
na
lly,
ll
y we ccould
y,
ould
ou
ld nnot
ot aassess
sses
ss
esss ca
es
card
cardiac
rdia
rd
iacc ch
ia
chan
changes
ange
an
gess or
ge
differentiate between echocardiographic systolic and diastolic dysfunction in our study.
Therefore, further studies should better characterize healthy men and women with
elevated heart rate, for instance using echocardiography or B-type natriuretic peptide
measurements.
In conclusion, in our population-based study of adults free of heart disease, both
single and repeated measurements of resting heart rate based on pulse palpation or
alternatively ECG identifies men at higher risk of developing heart failure, beyond other
cardiovascular risk factors. Whether healthy older individuals with higher resting heart
rate might benefit from preventive therapy that specifically reduces heart rate remains to
be explored.
14
Acknowledgements
The dedication, commitment, and contribution of inhabitants, general practitioners, and
pharmacists of the Ommoord district to the Rotterdam Study are gratefully
acknowledged.
Sources of Funding
This research was supported by the Swiss National Science Foundation [grant number
Downloaded from http://circheartfailure.ahajournals.org/ by guest on May 6, 2017
PBLAP3-132943 to DN], the Netherlands Organization for Scientific Research (NWO)
and the Netherlands Organization for Health Research and Development
(ZonMw)
[Vici
evelopment (Z
Zo
grant number 918-76-619; Veni grant number 916-12-154; ZonMw
80onM
Mw grant
gran
gr
antt number
an
n m
nu
82500-98-10208].
8].
8].
]
The Rotterdam
erdam
erda
dam
da
m Study
Stud
St
udyy is
ud
is funded
funnde
dedd by Erasmus
Era
rasm
sm
muss Medical
Med
edic
ical
ic
aall Center
Cen
nte
terr and
an
nd Erasmus
Erraass
Eras
University Rotterdam;
erd
rdam
am;; The
am
The Netherlands
Neth
Ne
ther
th
erla
er
land
la
ndss Organization
nd
Orga
Or
gani
ga
niza
ni
zati
za
tion
ti
on for
for the
the Health
Hea
ealt
lthh Research
lt
Rese
Re
sear
se
arch
ar
ch
h and
Development (ZonMw); the Research Institute for Diseases in the Elderly (RIDE); the
Ministry of Education, Culture and Science; the Ministry for Health, Welfare and Sports;
the European Commission (DG XII); and the Municipality of Rotterdam.
None of the funders had any role in design and conduct of the study; collection,
management, analysis, and interpretation of the data; and preparation, review, or approval
of the manuscript.
Disclosures
None.
15
References
1.
2.
Downloaded from http://circheartfailure.ahajournals.org/ by guest on May 6, 2017
3.
4.
5.
6.
7.
8.
9.
10.
11.
Bleumink GS, Knetsch AM, Sturkenboom MCJM, Straus SMJM, Hofman A,
Deckers JW, Witteman JCM, Stricker BHCh. Quantifying the heart failure
epidemic: prevalence, incidence rate, lifetime risk and prognosis of heart failure
The Rotterdam Study. Eur Heart J. 2004;25:1614-1619.
Hunt SA, Abraham WT, Chin MH, Feldman AM, Francis GS, Ganiats TG, Jessup
M, Konstam MA, Mancini DM, Michl K, Oates JA, Rahko PS, Silver MA,
Stevenson LW, Yancy CW. 2009 focused update incorporated into the ACC/AHA
2005 Guidelines for the Diagnosis and Management of Heart Failure in Adults: a
report of the American College of Cardiology Foundation/American Heart
Association Task Force on Practice Guidelines: developed in collaboration with
the International Society for Heart and Lung Transplantation. Circulation.
2009;119:e391-479.
Ceconi C, Guardigli G, Rizzo P, Francolini G, Ferrari R. The heart rate story. Eur
Heart J Suppl. 2011;13:C4-C13.
Bohm M, Swedberg K, Komajda M, Borer JS, Ford I,, Dubost-Brama
Dubost Brama
m A,
ma
Lerebours G, Tavazzi L. Heart rate as a risk factor in cchronic
hron
hr
onic
on
ic hheart
eart
ea
rt ffailure
a
(SHIFT): the association between heart rate and outcomes
omess in
n a rrandomised
ando
an
dom
do
placebo-controlled
2010;376:886-894.
controlled
cont
ntro
nt
roll
ro
lle ttrial.
lled
ll
r al. Lancet.
ri
L
t 2010;37
2010
376
376:886-894.
37
886 894
Diaz A, B
Bourassa
o rassaa MG,
ou
MG, Guertin
Guer
Gu
erti
er
tinn MC,
ti
MC, Tardif
MC
T rd
Ta
rdiff JC.
JC. Long-term
Lon
o gg-te
t rm
m prognostic
pro
r gn
gnos
osti
os
t c value of
resting heart
coronary
earrt rate inn ppatients
attients
ts withh ssuspected
usp
speccteed or pproven
ro
ovenn co
oronaary
r aartery
rter
erry disease. Eur
Heart J. 200
22005;26:967-974.
005;
00
5 266:9
5;
:967
67-9
67
-9
974
7 .
Butler J, Kalogeropoulos
Georgiopoulou
V,, Be
Belue
Kal er oulos A,
A G
Geo
eorg
eo
rg
gio
i po
oul
ulou
lou V
Belu
elu
ue R, Rodondi N, Garcia M,
Bauer DC,
Satterfield
S,, S
Smith
AL,
Vaccarino
V,, N
Newman
C, Sa
Satt
tter
tt
erfi
er
fiel
fi
eldd S
el
mith
mi
th A
L, Va
Vacc
ccar
cc
arin
ar
inoo V
in
ewma
ew
mann AB
ma
AB, Harris
Harr
Ha
rrr TB,
Wilson PW,
prediction
in
PW
P
W Kritchevsky
Kriitche
che
h vs
vsky
k SB,
ky
S
SB
B Health
Hea
ealt
lthh ABCS.
lt
ABCS
AB
C In
CS
IIncident
nci
cide
ci
d ntt hheart
de
eart
ea
rtt ffailure
aiilu
l re ppre
re
the elderly: the health ABC heart failure score. Circ Heart Fail. 2008;1:125-133.
Pfister R, Michels G, Sharp SJ, Luben R, Wareham NJ, Khaw K-T. Resting heart
rate and incident heart failure in apparently healthy men and women in the EPICNorfolk study. Eur J Heart Fail. 2012;14:1163-1170.
Rautaharju PM, Prineas RJ, Wood J, Zhang ZM, Crow R, Heiss G.
Electrocardiographic predictors of new-onset heart failure in men and in women
free of coronary heart disease (from the Atherosclerosis in Communities [ARIC]
Study). Am J Cardiol. 2007;100:1437-1441.
Hofman A, van Duijn CM, Franco OH, Ikram MA, Janssen HLA, Klaver CCW,
Kuipers EJ, Nijsten TEC, Stricker BHCh, Tiemeier H, Uitterlinden AG, Vernooij
MW, Witteman JCM. The Rotterdam Study: 2012 objectives and design update.
Eur J Epidemiol. 2011;26:657-686.
Leening MJG, Kavousi M, Heeringa J, van Rooij FJA, Verkroost-van Heemst J,
Deckers JW, Mattace-Raso FUS, Ziere G, Hofman A, Stricker BHCh, Witteman
JCM. Methods of data collection and definitions of cardiac outcomes in the
Rotterdam Study. Eur J Epidemiol. 2012;27:173-185.
Willems JL, Abreu-Lima C, Arnaud P, van Bemmel JH, Brohet C, Degani R,
Denis B, Gehring J, Graham I, van Herpen G, Machado I, Macfarlane PW,
Michaelis J, Moulopoulos SD, Rubel P, Zywietz C . The diagnostic performance
16
12.
13.
14.
Downloaded from http://circheartfailure.ahajournals.org/ by guest on May 6, 2017
15.
16.
17.
18.
19.
20.
21.
22.
23.
of computer programs for the interpretation of electrocardiograms. N Engl J Med.
1991;325:1767-1773.
Mosterd A, Hoes AW, de Bruyne MC, Deckers JW, Linker DT, Hofman A,
Grobbee DE. Prevalence of heart failure and left ventricular dysfunction in the
general population; The Rotterdam Study. Eur Heart J. 1999;20:447-455.
Swedberg K, Cleland J, Dargie H, Drexler H, Follath F, Komajda M, Tavazzi L,
Smiseth OA, Gavazzi A, Haverich A, Hoes A, Jaarsma T, Korewicki J, Levy S,
Linde C, Lopez-Sendon JL, Nieminen MS, Pierard L, Remme WJ. Guidelines for
the diagnosis and treatment of chronic heart failure: executive summary (update
2005): The Task Force for the Diagnosis and Treatment of Chronic Heart Failure
of the European Society of Cardiology. Eur Heart J. 2005;26:1115-1140.
Yamaguchi J, Hozawa A, Ohkubo T, Kikuya M, Ugajin T, Ohmori K, Hashimoto
J, Hoshi H, Satoh H, Tsuji I, Imai Y. Factors affecting home-measured resting
heart rate in the general population: the Ohasama study. Am J Hypertens.
2005;18:1218-1225.
Fisher LD, Lin DY. Time-dependent covariates in the Cox proportional-hazards
regression model. Annu Rev Public Health. 1999;20:145-157.
145 157.
D'Agostino RB, Sr., Vasan RS, Pencina MJ, Wolf PA,
Cobain
Massaro
JM,
A C
A,
obai
ob
ainn M, M
ai
a
Kannel WB. General cardiovascular risk profile for use
se in
n pprimary
r ma
ri
mary
ry ccare:
a the
Framingham
2008;117:743-753.
ham
mH
Heart
eart
ea
rt S
Study.
tudy. Circulation.
Ci l ti
22008
0 8;117:74300
117 743
3-753.
3753
Leening MJ
Elias-Smale
SE,
Kavousi
M,, F
Felix
Deckers
Vliegenthart
M
MJG,
Eli
l as
li
as-S
-S
Sma
male
le S
le
E, K
av
vou
o sii M
e ix
el
x JJF,
F, D
ecke
ec
kers
ke
r JJW,
W, V
l
R, Oudkerk
Hofman
A,, Stey
Steyerberg
EW,
Stricker
BHCh,
Witteman
erk
k M, Hofm
fman
an
nA
yerrbeerg E
W,, S
t ickker BH
tr
HCh, Wi
W
ttem
man
n JCM.
Coronary
Calcification
Risk
Heart
Failure
yC
alci
al
cifi
ci
ficaati
fi
tion
on aand
ndd tthe
he R
iskk of
is
of H
eaart F
ailu
lure
r iinn the
re
the Elderly:
E de
El
d rl
rly:
y The
y:
T
Rotterdam
m Study. JACC: Cardiovascular
Card
Ca
rdio
rd
iova
io
va
asc
scullarr Imaging.
Im
maagi
ging
ng.
g 2012;5:874-880.
201
0 2;5:874-880.
Stricker B
BHCh,
Stijnen
T.. An
Analysis
time-varying
HChh, S
HC
tijn
ti
j en T
jn
Anal
alys
al
ysis
ys
is of iindividual
ndiv
nd
ivid
iv
idua
id
uall dr
ua
drug
ugg uuse
se aass a ti
time
me-v
me
-v
v
determinant
studies.
antt of eexposure
an
xpos
xp
osur
uree in
i pprospective
rosp
ro
spec
ecttiv
i e population-based
popu
po
pula
l ti
la
tion
on-b
-b
bas
ased
ed
d ccohort
oh
hor
ortt st
studi
stud
udii Eur J
ud
Epidemiol. 2010;25:245-251.
Palatini P, Benetos A, Grassi G, Julius S, Kjeldsen SE, Mancia G, Narkiewicz K,
Parati G, Pessina AC, Ruilope LM, Zanchetti A. Identification and management
of the hypertensive patient with elevated heart rate: statement of a European
Society of Hypertension Consensus Meeting. J Hypertens. 2006;24:603-610.
Cohn JN, Levine TB, Olivari MT, Garberg V, Lura D, Francis GS, Simon AB,
Rector T. Plasma norepinephrine as a guide to prognosis in patients with chronic
congestive heart failure. N Engl J Med. 1984;311:819-823.
Lam CS, Carson PE, Anand IS, Rector TS, Kuskowski M, Komajda M, McKelvie
RS, McMurray JJ, Zile MR, Massie BM, Kitzman DW. Sex differences in clinical
characteristics and outcomes in elderly patients with heart failure and preserved
ejection fraction: the Irbesartan in Heart Failure with Preserved Ejection Fraction
(I-PRESERVE) trial. Circ Heart Fail. 2012;5:571-578.
Martínez-Sellés M. What Do Women Have In Their Hearts? Rev Esp Cardiol.
2007;60:1118-1121.
Schocken DD, Benjamin EJ, Fonarow GC, Krumholz HM, Levy D, Mensah GA,
Narula J, Shor ES, Young JB, Hong Y. Prevention of Heart Failure. Circulation.
2008;117:2544-2565.
17
24.
25.
26.
27.
Downloaded from http://circheartfailure.ahajournals.org/ by guest on May 6, 2017
Fox K, Ford I, Steg PG, Tendera M, Ferrari R. Ivabradine for patients with stable
coronary artery disease and left-ventricular systolic dysfunction (BEAUTIFUL): a
randomised, double-blind, placebo-controlled trial. Lancet. 2008;372:807-816.
Swedberg K, Komajda M, Bohm M, Borer JS, Ford I, Dubost-Brama A,
Lerebours G, Tavazzi L. Ivabradine and outcomes in chronic heart failure
(SHIFT): a randomised placebo-controlled study. Lancet. 2010;376:875-885.
Auer R, Bauer DC, Marques-Vidal P, Butler J, Min LJ, Cornuz J, Satterfield S,
Newman AB, Vittinghoff E, Rodondi N, Health ABCS. Association of major and
minor ECG abnormalities with coronary heart disease events. JAMA.
2012;307:1497-1505.
Bonnemeier H, Richardt G, Potratz J, Wiegand UK, Brandes A, Kluge N, Katus
HA. Circadian profile of cardiac autonomic nervous modulation in healthy
subjects: differing effects of aging and gender on heart rate variability. J
Cardiovasc Electrophysiol. 2003;14:791-799.
18
Table 1. Baseline characteristics by gender-specific resting heart rate
Downloaded from http://circheartfailure.ahajournals.org/ by guest on May 6, 2017
Men (n=1,829)
First
Second
Third
tertile
tertile
tertile
(n=571)
(n=1,221)
(n=837)
(n=881)
74 (69-78)
86 (79-128)
68 (40-72)
76 (74-80)
88 (81-128)
67.4 (8.1)
67.6 (8.1)
67.2 (8.3)
0.62
0.662
69.3
6 (9.7)
69
68.9 (9.4)
69.1 (9.4)
0.61
Never
60 (8.7)
47 (8.6)
(8.66)
(8
47
(8.
8 6)
48 (8.6)
<0.0001
<0
<0.001
640 (53.8)
640
417 (51.8)
450 (52.9)
0.38
Former
452 (65.5))
3335
3 ((61.4)
35
61.4
61.44)
287
28
87 (5
((51.3)
1.
1.3)
343
343 (28.8)
238 (29.6)
209 (24.6)
Current
181 (26.1)
164 (30.0)
224 (40.1)
206 (17.3)
149 (18.5)
191 (22.5)
289 (40.9)
257 (46.6)
293 (51.4)
0.001
599 (49.1)
386 (46.2)
493 (56.0)
<0.001
37 (5.3)
51 (9.3)
60 (10.5)
0.001
86 (7.1)
65 (7.9)
105 (12.0)
<0.001
Resting heart rate, median bpm (range)
First
Second
Third
tertile
tertile
tertile
(n=706)
(n=552)
64 (40-68)
Women (n=2,939)
P-value
P-value
Demographics
Age, years
Smoking status, No. (%)
Comorbidities
Hypertension, No. (%)1
Diabetes mellitus, No. (%)2
19
Objective measures
Downloaded from http://circheartfailure.ahajournals.org/ by guest on May 6, 2017
Systolic blood pressure, mmHg
136 (21)
138 (21)
140 (22)
0.009
137 (23)
137 (22)
141 (22)
<0.001
Diastolic blood pressure, mmHg
73 (11)
75 (12)
76 (12)
<0.001
72 (11)
72 (11)
74 (12)
0.052
Body mass index, kg/m2
25.5 (2.9)
25.6 (3.0)
25.2 (3.0)
0.044
26.5 (3.8)
26.3 (4.0)
26.5 (3.9)
0.37
Total cholesterol, mmol/L
6.3 (1.1)
6.3 (1.3)
6.3 (1.2)
0.70
6.8 (1.2)
6.9 (1.2)
6.9 (1.3)
0.47
HDL-cholesterol, mmol/L
1.3 (0.3)
1.2 (0.3)
1.2 (0.3)
0.144
1.5 (0.4)
1.5 (0.4)
1.5 (0.4)
0.37
88(14)
88 (15)
89 (22)
0.16
77 (14)
76 (14)
76 (13)
0.81
Anti-hypertensive drugs, No. (%)
57 (8.1)
52 (9.4)
52 (9.1)
0.67
0.67
221
22 (18.1)
124 (14.8)
151 (17.4)
0.139
Anti-asthmatics, No. (%)
39 (5.5)
30 (5.4)
(5.4)
64 (11.2)
(111.2)
00.002
0.
002
40 (3.3)
29 (3.5)
46 (5.2)
0.184
Corticosteroids, No. (%)
11 (1.6)
10 (1.8)
( .88)
(1
177 (3.0)
(3.
3 0)
0.45
0.4
.45
10
1 (0.8)
8 (1.0)
31 (3.5)
<0.001
Thyroid therapy, No. (%)
1 (0.1)
3 (0.5)
2 (0.4)
(0.
0.4)
4)
0.63
477 (3.8)
28 (3.3)
32 (3.6)
0.76
Creatinine, μmol/L
Medication use
Data are given as mean (standard deviation) unless
unl
nles
esss otherwise
otthe
herw
rwis
rw
isee indicated.
is
iindicated
in
ndi
dica
c te
ca
tedd
1
Defined as a systolic blood pressure •140 mmHg or diastolic blood pressure •90 mmHg or use of blood pressure lowering drugs for
the indication of hypertension.
2
Defined as a random or post-load blood glucose measurement of •11.1 mmol/L or the use of anti-diabetic drugs.
Abbreviations: bpm, beats per minute; HDL, high-density lipoprotein.
20
Table 2. Heart failure risk for each increment of 10 bpm for time-fixed versus time-dependent heart rate measurement during followDownloaded from http://circheartfailure.ahajournals.org/ by guest on May 6, 2017
up
Men (n= 1,829)
Women (n=2,939)
N events
HR
95% CI
P-value
N events
HR
95% CI
P-value
261
1.14
1.16
1.14
(1.04 ; 1.26)
(1.05 ; 1.28)
(1.03 ; 1.27)
0.008
0.005
.005
0.012
.01
0112
395
0.97
0.95
0.94
(0.89 ; 1.05)
(0.87 ; 1.04)
(0.86 ; 1.03)
0.45
0.27
0.21
2611
1.15
1.1
. 5
1.13
1.11
1.
.11
(1.0
(1
(1.04
.044 ; 1.27)
.0
1 27
1.
27))
(1.02
(1.
1 02
0 ; 11.25)
.225)
5
(1.00
(1.
1.00
0 ; 11.22)
00
. 2))
.2
0.005
0.00
0.
.00
005
00.017
0.
017
0.05
0.
05
395
3
1.04
0.98
0.98
(0.95 ; 1.13)
(0.90 ; 1.07)
(0.89 ; 1.07)
0.41
0.72
0.58
Baseline heart rate
Age adjusted
Multivariable adjusted1
Additionally adjusted for treatment2
Time-dependent heart rate
Age adjusted
Multivariable adjusted1
Additionally adjusted for treatment2
1
Adjusted for age, smoking status, systolic blood pressure, diabetes, body mass index, total cholesterol, high-density lipoprotein
cholesterol, and antihypertensive treatment.
2
Additionally adjusted for systemic corticoid and anti-asthmatic treatment.
Abbreviations: N, number; HR, hazard ratio; CI, confidence interval;
21
Table 3. Hazard ratios for developing heart failure, with respect to gender-specific baseline resting heart rate tertile, and after
Downloaded from http://circheartfailure.ahajournals.org/ by guest on May 6, 2017
considering heart rate modifying medication during follow-up, or censoring at the occurrence of myocardial infarction
Men (n=1,829)
Women (n=2,939)
First tertile
Second tertile
Third tertile
First tertile
Second tertile
Third tertile
Heart rate
Heart rate
Heart rate
Heart rate
Heart rate
Heart rate
<69 bpm
69-78 bpm
•79 bpm
<73 bpm
73-80 bpm
•81 bpm
Number of events
84
89
88
17
172
110
113
Incidence rate, per 1,000
9.9
13.6
13.7
111.0
1
10.4
10.8
Age adjusted HR
1.00
1.36
1.
1.36
3
1.48
1.
.48
1.00
1.0
1.
0
1.00
1.00
(95% CI)
(ref)
(1.01
01 ; 1.
1.83
1.83)
83))
83
(1
(1.10
1.1
10 ; 2.
22.00)
00
0)
(re
(r
(ref)
e
(0.78 ; 1.26)
(0.79 ; 1.26)
Multivariable adjusted HR
1.00
1.35
1.51
1.00
0.99
0.98
(95% CI)1
(ref)
(0.99 ; 1.84)
(1.11 ; 2.07)
(ref)
(0.77 ; 1.27)
(0.76 ; 1.26)
1.00
1.34
1.48
1.00
0.99
0.99
(ref)
(0.98 ; 1.83)
(1.09 ; 2.03)
(ref)
(0.77 ; 1.28)
(0.77 ; 1.27)
1.00
0.99
1.00
(ref)
(0.78-1.26)
(0.79-1.27)
1.00
0.98
0.99
(ref)
(0.76-1.26)
(0.77-1.27)
P for
trend
P for
trend
person-years
Additionally adjusted for
2
treatment HR (95% CI)
0.009
0.00
0.
0099
00
0.009
0.013
0.97
0.86
0.92
Using the prescription of beta-blockers and calcium channel blockers as time-dependent variables
Age- adjusted HR
1.00
1.31
1.41
(95% CI)
(ref)
(0.97-1.77)
(1.05-1.91)
Multivariable adjusted HR
1.00
1.32
1.47
(95% CI)1
(ref)
(0.97-1.80)
(1.08-2.01)
Censoring participants at the occurrence of coronary heart disease
22
0.022
0.014
0.98
0.93
Downloaded from http://circheartfailure.ahajournals.org/ by guest on May 6, 2017
Number of events
74
76
76
Age adjusted HR
1.00
1.31
1.46
(95% CI)
(ref)
(0.95-1.81)
(1.06-2.00)
1.00 (ref)
1.33
1.56
(0.95-1.86)
(1.12-2.19)
Multivariable adjusted HR
1
(95% CI)
1
0.020
0.009
103
105
1.00 (ref)
0.99
0.98
(0.77-1.26)
(0.77-1.25)
0.97
0.96
(0.75-1.27)
(0.74-1.25)
1.00 (ref)
Adjusted for age, smoking status, systolic blood pressure, diabetes mellitus, body mass index, total cholesterol, high-density
lipoprotein cholesterol, and antihypertensive treatment.
2
162
Additionally adjusted for systemic corticoid and anti-asthmatic treatment.
Abbreviations: HR, hazard ratio; CI, confidence
nce
cee iinterval;
nter
nt
erva
er
v l; bpm, beats per
p r minute.
pe
minuutee.
23
0.86
0.76
Table 4. Heart failure risk for each increment of 10 bpm for time-fixed versus time-dependent heart rate measurement during followDownloaded from http://circheartfailure.ahajournals.org/ by guest on May 6, 2017
up, as measured by ECG (n=4,127)
Men (n= 1,574)
Women (n=2,553)
N events
HR
95% CI
P-value
N events
HR
95% CI
P-value
219
1.17
1.13
11.12
1.
12
(1.05 ; 1.30)
(1.01
(
; 1.26))
((1.00
1.00
1.
00 ; 11.25)
.25))
.2
0.004
4
0.030
0
00.053
.0
.053
053
33
338
1.07
1.01
1.00
(0.97 ; 1.17)
(0.91 ; 1.11)
(0.90 ; 1.10)
0.165
0.91
0.93
219
1.21
1.21
21
11.18
18
1.16
(1.10
(1.110 ; 1.34)
(1
1 34
1.
34))
(1.06
((11 06 ; 11.31)
331)
1)
(1.05 ; 1.29)
<0
<0.001
0.0
.001
001
00.002
0022
00
0.005
338
33
3
1.09
1.02
1.01
(1.00 ; 1.19)
(0.93 ; 1.12)
(0.92 ;1.11)
0.047
0.72
0.86
Baseline heart rate
Age adjusted
Multivariable adjusted1
Additionally adjusted for treatment2
Time-dependent heart rate
Age adjusted
Multivariable adjusted1
Additionally adjusted for treatment2
1
Adjusted for age, smoking status, systolic blood pressure, diabetes, body mass index, total cholesterol, high-density lipoprotein
cholesterol, and antihypertensive treatment.
2
Additionally adjusted for systemic corticoid and anti-asthmatic treatment.
Abbreviations: N, number; HR, hazard ratio; CI, confidence interval.
24
Figure Legend
Figure
Cumulative incidence of heart failure according to tertile of baseline heart rate, by gender
a) Heart rate measured at the radial artery for 30 seconds (n=4,768)
b) Heart rate measured from a 10 second ECG (n=4,127)
Downloaded from http://circheartfailure.ahajournals.org/ by guest on May 6, 2017
25
Downloaded from http://circheartfailure.ahajournals.org/ by guest on May 6, 2017
Resting Heart Rate and the Risk of Heart Failure in Healthy Adults: The Rotterdam Study
David Nanchen, Maarten J.G. Leening, Isabella Locatelli, Jacques Cornuz, Jan A. Kors, Jan Heeringa,
Jaap W. Deckers, Albert Hofman, Oscar H. Franco, Bruno H.Ch. Stricker, Jacqueline C.M. Witteman
and Abbas Dehghan
Downloaded from http://circheartfailure.ahajournals.org/ by guest on May 6, 2017
Circ Heart Fail. published online April 18, 2013;
Circulation: Heart Failure is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 2013 American Heart Association, Inc. All rights reserved.
Print ISSN: 1941-3289. Online ISSN: 1941-3297
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://circheartfailure.ahajournals.org/content/early/2013/04/18/CIRCHEARTFAILURE.112.000171
Data Supplement (unedited) at:
http://circheartfailure.ahajournals.org/content/suppl/2013/04/18/CIRCHEARTFAILURE.112.000171.DC1
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in
Circulation: Heart Failure 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: Heart Failure is online at:
http://circheartfailure.ahajournals.org//subscriptions/
Supplemental Material
Supplemental Figure. Changes of resting heart rate during follow-up estimated by a quadratic mixed-effect model (n=4,768)
Abbreviation: bpm, beat per minute.
Supplemental Table. Incidence rates and hazard ratios for developing heart failure, with respect to gender-specific resting heart rate
tertile, as measured by baseline ECG (n=4,127)
First tertile
Men (n=1,574)
Third tertile
Heart rate <63 bpm Heart rate 63-73 bpm Heart rate ≥74 bpm
60
66
93
Incidence rate, per 1,000 person-years
8.9
10.3
16.3
Age adjusted HR (95% CI)
1.00 (ref)
1.12 (0.79-1.59)
1.66 (1.20-2.29)
0.002
Multivariable adjusted HR (95% CI)1
1.00 (ref)
1.07 (0.74-1.54)
1.50 (1.07-2.11)
0.015
Additionally adjusted for treatment2
1.00 (ref)
1.07 (0.74-1.54)
1.46 (1.04-2.06)
0.023
Heart rate <69 bpm Heart rate 69-76 bpm Heart rate ≥77 bpm
Number of events
104
110
124
Incidence rate, per 1,000 person-years
8.5
10.4
12.4
Age adjusted HR (95% CI)
1.00 (ref)
1.08 (0.82-1.41)
1.20 (0.93-1.56)
0.166
Multivariable adjusted HR (95% CI)1
1.00 (ref)
0.98 (0.74-1.30)
1.05 (0.80-1.39)
0.69
Additionally adjusted for treatment2
1.00 (ref)
0.98 (0.74-1.30)
1.05 (0.80-1.39)
0.70
Adjusted for age, smoking status, systolic blood pressure, diabetes, body mass index, total cholesterol, high-density lipoprotein
cholesterol and antihypertensive treatment.
2
P for trend
Number of events
Women (n=2,553)
1
Second tertile
Additionally adjusted for systemic corticoid and anti-asthmatics treatment.
Abbreviations: HR, hazard ratio; CI, confidence interval, ECG, electrocardiograph; bpm, beat per minute.