Insomnia and risk of cardiovascular disease: A meta-analysis

Abstract

Objective: Increasing evidence suggests an association between insomnia and cardiovascular disease. We performed a systematic review with meta-analysis of all the available prospective studies that investigated the association between insomnia and risk of developing and/or dying from cardiovascular disease.Design: Systematic review and meta-analysis of prospective cohort studies.Methods: We conducted an electronic literature search through MedLine, Embase, Google Scholar, Web of Science, The Cochrane Library and bibliographies of retrieved articles up to December 2011. Studies were included if they were prospective, had assessment of insomnia or sleep complaints at baseline, evaluated subjects free of cardiovascular disease at baseline and measured the association between insomnia and risk of developing and/or dying from cardiovascular disease.Results: After the review process 13 prospective studies were included in the final analysis. These studies included 122,501 subjects followed for a time ranging from three to 20 years. A total of 6332 cardiovascular events occurred during the follow-up. Insomnia was assessed through questionnaire and defined as either difficulty of initiating or maintaining sleep or presence of restless, disturbed nights. The cumulative analysis for all the studies under a random-effects model showed that insomnia determined an increased risk (+45%) of developing or dying from cardiovascular disease during the follow-up (relative risk 1.45, 95% confidence interval 1.29-1.62; p < 0.00001), with no evidence of heterogeneity across the studies (I(2): 19%; p = 0.14).Conclusion: Insomnia is associated with an increased risk of developing and/or dying from cardiovascular disease.

Full text

XML Template (2012) [23.8.2012–6:04pm] [1–8]
{SAGE}CPR/CPR 460020.3d (CPR) [PREPRINTER stage]
E
U
R
O
P
E
A
N
S
O
C
I
E
T
Y
O
F
C
A
R
DIO
L
O
G
Y
®
Original scientific paper
Insomnia and risk of cardiovascular
disease: A meta-analysis
Francesco Sofi
1,2,3
, Francesca Cesari
1
, Alessandro Casini
3
,
Claudio Macchi
1
, Rosanna Abbate
2
and Gian Franco Gensini
1
Abstract
Objective: Increasing evidence suggests an association between insomnia and cardiovascular disease. We performed a
systematic review with meta-analysis of all the available prospective studies that investigated the association between
insomnia and risk of developing and/or dying from cardiovascular disease.
Design: Systematic review and meta-analysis of prospective cohort studies.
Methods: We conducted an electronic literature search through MedLine, Embase, Google Scholar, Web of Science,
The Cochrane Library and bibliographies of retrieved articles up to December 2011. Studies were included if they were
prospective, had assessment of insomnia or sleep complaints at baseline, evaluated subjects free of cardiovascular disease
at baseline and measured the association between insomnia and risk of developing and/or dying from cardiovascular
disease.
Results: After the review process 13 prospective studies were included in the final analysis. These studies included
122,501 subjects followed for a time ranging from three to 20 years. A total of 6332 cardiovascular events occurred
during the follow-up. Insomnia was assessed through questionnaire and defined as either difficulty of initiating or main-
taining sleep or presence of restless, disturbed nights. The cumulative analysis for all the studies under a random-effects
model showed that insomnia determined an increased risk (þ45%) of developing or dying from cardiovascular disease
during the follow-up (relative risk 1.45, 95% confidence interval 1.29–1.62; p<0.00001), with no evidence of hetero-
geneity across the studies (I
2
: 19%; p¼0.14).
Conclusion: Insomnia is associated with an increased risk of developing and/or dying from cardiovascular disease.
Keywords
Insomnia, sleep complaints, cardiovascular disease, sleep, meta-analysis
Received 20 April 2012; accepted 13 August 2011
Introduction
Insomnia, defined as a subjective feeling of either
inability to fall asleep or to maintain sleep or the per-
ception of disturbed sleep, is a highly prevalent condi-
tion in the industrialized countries. It has been
estimated that over 30% of all adults have a current
symptom of sleep disturbance, with a prevalence that
reaches 50% among subjects older than 65 years.
1
During the last years, a considerable number of stu-
dies investigated the possible association between both
qualitative and quantitative sleep disturbances and risk
of cardiovascular disease.
2–5
A recent meta-analysis of
prospective cohort studies by Cappuccio et al. showed a
significant association between both short and long
duration of sleep and cardiovascular outcomes in the
general population.
3
Nevertheless, an increasing inter-
est toward the possible association between the pres-
ence of insomnia and cardiovascular disease has been
reported, with as yet inconsistent results.
5
Some, but
1
Don Carlo Gnocchi Foundation, Centro S. Maria agli Ulivi, Onlus
IRCCS, Florence, Italy
2
Department of Medical and Surgical Critical Care, Thrombosis Centre,
University of Florence, Italy
3
Regional Agency of Nutrition, Azienda Ospedaliero-Universitaria
Careggi, Florence, Italy
Corresponding author:
Francesco Sofi, Department of Medical and Surgical Critical Care,
Thrombosis Centre, University of Florence, Viale Morgagni 85, 50134,
Florence, Italy.
Email: francescosofi@gmail.com
European Journal of Preventive
Cardiology
0(00) 1–8
!The European Society of
Cardiology 2012
Reprints and permissions:
sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/2047487312460020
ejpc.sagepub.com
1
2

XML Template (2012) [23.8.2012–6:04pm] [1–8]
{SAGE}CPR/CPR 460020.3d (CPR) [PREPRINTER stage]
not all, studies found a higher risk of cardiovascular
disease in subjects suffering from sleep complaints com-
pared with those who did not report these disturbances.
The aim of the present study was to systematically
review prospective cohort studies in order to conduct a
meta-analysis of all the available studies that investi-
gated the association between insomnia and cardiovas-
cular incidence and/or mortality in subjects without
cardiovascular disease at baseline.
Methods
Selection of studies
We searched the electronic databases MedLine (source:
PubMed, 1966 to December 2011), Embase (1980 to
December 2011), Web of Science, The Cochrane
Library (source: The Cochrane Central Register of
Controlled Trials, 2011, issue 11), Clinicaltrials.org,
and Google Scholar to identify articles that investigated
the possible association between insomnia and cardio-
vascular disease in adults. Relevant keywords relating
to insomnia and cardiovascular disease in combination
as MeSH terms and text words (‘insomnia’, or ‘sleep
complaints’ or ‘sleep initiation’, or ‘sleep disorders’ or
‘sleep disturbances’ or ‘disorders of initiating and main-
taining sleep’, or ‘poor sleep quality’) were used in com-
bination with words related to cardiovascular disease
(‘coronary heart disease’, or ‘coronary artery dis-
ease’, or ‘myocardial infarction’ or ‘stroke’, or ‘cardio-
vascular disease’, or ‘outcomes’ or ‘mortality’). We
limited the search strategy to prospective studies, with
no language restrictions, supplemented by manually
reviewing the reference list of all retrieved
articles. Two investigators (FS, FC) assessed poten-
tially relevant articles for eligibility. The decision to
include or exclude articles was hierarchical and initially
made on the basis of title, abstract and finally of the
complete article. In the event of conflicting opinions,
resolution of the disagreement was resolved through
discussion.
Eligible articles were included if they met all of the
following criteria: (i) provide as primary or secondary
outcome the association between sleep complaints and
cardiovascular outcomes; (ii) evaluate subjects free of
cardiovascular disease at baseline; (iii) report clear def-
initions of method used to assess sleep complaints; (iv)
report explicit measure of association between insom-
nia, sleep complaints and cardiovascular outcomes.
Accordingly, articles were excluded if they had cross-
sectional, case-control or intervention design, if they
considered outcomes different from those of interest
for the meta-analysis, if they investigated patients
with cardiovascular disease at baseline, if they did not
report the association between sleep complaints and
cardiovascular outcomes or if they failed to present
estimates of association with the disease.
Outcomes of interest for the current meta-analysis
were occurrence of acute myocardial infarction, coron-
ary heart disease, stroke or death from any of the afore-
mentioned cardiovascular causes.
Data extraction
All data were reviewed and separately extracted by two
independent investigators (FS, FC) using a standar-
dized form. The following baseline characteristics for
study populations were collected: first author’s name
and year of publication, cohort, country, baseline
year, number of subjects at baseline, gender, years of
follow-up, age at baseline, outcome of interest, defin-
ition of insomnia, categories of exposure, relative risks
and confidence intervals, as well as adjustment for con-
founding factors at the multivariate model.
Statistical analysis
We used Review Manager (RevMan; version 5.0.23 for
Macintosh; Copenhagen, Denmark; The Cochrane
Collaboration, 2009) to pool results from the individual
studies. We used random effect models to calculate
summary relative risks (RRs) and 95% confidence
intervals (CIs). The natural logarithm of the RRs was
estimated and the RR from each study by the inverse of
its variance was weighted. A two tailed p<0.05 was
considered statistically significant. We used, when
available, the results of the original studies from multi-
variate models with the most complete adjustment for
potential confounders; the confounding variables
included in this analysis are shown in Table 1.
Statistical heterogeneity was evaluated using the I
2
statistic, which assesses the appropriateness of pooling
the individual study results. The I
2
value provides an
estimate of the amount of variance across studies due to
the heterogeneity rather than chance. Where I
2
was
greater than 50%, heterogeneity was considered sub-
stantial. Moreover, to further investigate the heterogen-
eity across the studies we performed sensitivity analyses
by dividing studies into groups according to their main
characteristics. Subgroup analyses were then performed
according to country of cohort (European/non-
European), gender (males/females), mean sample size
of the study populations (<7650; #7650), mean dur-
ation of follow-up (<10 years; #10 years) and adjust-
ment for confounders (only for age/also for other
demographic characteristics and cardiovascular risk
factors). Publication bias was appraised by visual
inspection of funnel plot of effect size against standard
error.
2European Journal of Preventive Cardiology 0(00)

XML Template (2012) [23.8.2012–6:04pm] [1–8]
{SAGE}CPR/CPR 460020.3d (CPR) [PREPRINTER stage]
Table 1. Characteristics of included studies
Source, year (cohort)
Country,
baseline year
Subjects,
nGender
Follow-up,
years
Age,
years Outcome (n)
Definition
of insomnia
Exposure
categories RR (95% CI) Adjustment
Appels et al.,
1987
6
(Rotterdam
Civil Servants Study)
The Netherlands,
1979
3269 M 4.2 39–65 MI (59) Trouble falling
asleep
No Ref. Age
Yes 1.60 (0.92–2.79)
Eaker et al.,
1992
7
(Framingham
Study)
USA, 1965 749 F 20 45–64 MI or CHD
mortality
(39)
Trouble falling
asleep
No Ref. Age, systolic blood pres-
sure, total cholesterol/
HDL, diabetes, cigar-
ettes, BMI
Yes 3.9 (1.7–9.0)
Schwartz et al., 1998
8
(Piedmont Health
Survey)
USA, 1984 2960 M/F 3 65–101 MI (152) Trouble falling
asleep
Never Ref. Age, gender, race, educa-
tion, prescriptions,
self-rated health,
depression
Most of the
time
1.44 (0.92–2.25)
Mallon et al.,
2002
9
(County of
Dalarna registry)
Sweden, 1983 906 M 12 45–65 CHD mortal-
ity (71)
Difficulty initiating
sleep
No Ref. Age, not married, living
alone, smoking habit,
BMI>28, hyperten-
sion, cardiac disease,
respiratory disease,
diabetes, joint pain, GI
disease, depression,
sleep duration, snor-
ing, sleeping pill usage
Yes 3.1 (1.5–6.3)
Mallon et al., 2002
9
(County of Dalarna
registry)
Sweden, 1983 964 F 12 45–65 CHD mortal-
ity (20)
Difficulty initiating
sleep
No Ref. Age
Yes 2.0 (0.8–5.0)
Elwood et al., 2006
10
(Caerphilly cohort
study)
UK, 1984 1874 M 5 55–69 CHD (213) Insomnia None Ref. Age, social class, smoking
habit, alcohol con-
sumption, BMI, neck
circumference
Frequent 1.47 (0.98–2.21)
Elwood et al., 2006
(10) (Caerphilly
cohort study)
UK, 1984 1874 M 5 55–69 Stroke (107) Insomnia None Ref. Age, social class, smoking
habit, alcohol con-
sumption, BMI, neck
circumference
Frequent 1.75 (1.02–3.01)
Phillips et al., 2007
11
(ARIC study)
USA, 1988 11,863 M/F 6.3 44–64 CHD (546) Trouble falling
asleep/sleep
continuity dis-
turbance/non-
restorative
sleep
No Ref. Age, gender, race, educa-
tion, BMI, depression,
lung function, smoking
habit, diabetes
Yes 1.5 (1.1–2.0)
Meisinger et al.,
2007
12
(MONICA/
KORA)
Germany, 1984 3508 M 10.1 45–75 MI (295) Difficulty initiating
sleep
No Ref. Age, survey, BMI, educa-
tion, dyslipidaemia,
alcohol intake, paren-
tal history of MI,
physical activity, smok-
ing habit, hyperten-
sion, diabetes
Yes 1.16 (0.82–1.63)
(continued)
Sofi et al. 3
5

XML Template (2012) [23.8.2012–6:04pm] [1–8]
{SAGE}CPR/CPR 460020.3d (CPR) [PREPRINTER stage]
Table 1. Continued
Source, year (cohort)
Country,
baseline year
Subjects,
nGender
Follow-up,
years
Age,
years Outcome (n)
Definition
of insomnia
Exposure
categories RR (95% CI) Adjustment
Meisinger et al.,
2007
12
(MONICA/
KORA)
Germany, 1984 3388 F 10.1 45–75 MI (85) Difficulty initiating
sleep
No Ref. Age, survey, BMI, educa-
tion, dyslipidaemia,
alcohol intake, paren-
tal history of MI,
physical activity, smok-
ing habit, hyperten-
sion, diabetes
Yes 1.30 (0.81–2.06)
Suzuki et al., 2009
1,3
(Shizuoka Study)
Japan, 1999 11,395 M/F 5.3 65–85 CHD mortal-
ity (310)
Difficulty falling
asleep
No Ref. Age, gender, BMI, smok-
ing habit, alcohol,
physical activity, socio-
economic status,
mental health, hyper-
tension, diabetes
Yes 1.02 (0.73–1.43)
Chien et al., 2010
14
(Chin-Shan commu-
nity study)
Taiwan, 1990 3430 M/F 15.9 >35 CVD (420) Frequency of
insomnia
No Ref. Age, gender, BMI, smok-
ing habit, alcohol,
marital status, educa-
tion, occupation, exer-
cise, family history of
CHD, hypertension,
diabetes, lipids, glu-
cose, uric acid
Nearly every
day
1.78 (1.03–3.08)
Loponen et al.,
2010
1,5
(Helsinki
Heart Study)
Finland, 1981 2753 M 7.3 40–55 CHD (287) Number of
insomnia
symptoms
No
5–6
Ref
1.42 (0.86–2.34)
Age
Chandola et al.,
2010
1,6
(Whitehall II)
UK, 1985 10,308 M/F 15 35–55 CHD mortal-
ity and MI
(1205)
Restless, dis-
turbed nights
No Ref. Age, gender, ethnicity,
employment grade, car
access, housing tenure,
self-rated health
status, cholesterol,
hypertension, BMI,
diabetes, smoking
habit, alcohol, exer-
cise, fruit and vege-
table consumption
More than
usual
1.36 (1.10–1.68)
Hulvej Rod et al.,
2011
1,7
(GAZEL
cohort study)
France, 1989 12,524 M 19 36–52 CHD mortal-
ity (155)
Difficulty falling
asleep
No Ref. Age, socioeconomic
status, marital status,
smoking habit, alcohol,
BMI, night work,
hypertension, diabetes
Yes 1.18 (0.67–2.06)
Laugsand et al.,
2011
1,8
(Nord
Trondelag County)
Norway, 1995 52,610 M/F 11.4 #65 MI (2368) Difficulty initiating
sleep
No Ref. Age, gender, education,
marital status, shift
work, systolic blood
pressure, cholesterol,
diabetes, BMI, physical
activity, smoking habit,
depression, anxiety
Yes 1.53 (1.21–1.92)
MI: myocardial infarction; CHD: coronary heart disease; HDL: high density lipoprotein; BMI: body mass index; GI: gastrointestinal.
4European Journal of Preventive Cardiology 0(00)

XML Template (2012) [23.8.2012–6:04pm] [1–8]
{SAGE}CPR/CPR 460020.3d (CPR) [PREPRINTER stage]
Results
Study identification and selection
Our search strategy yielded 59 articles. Of these, we first
excluded 12 articles because they had a cross-sectional,
case-control or intervention design. The selected art-
icles were then carefully reviewed and a further 21
articles were excluded because they had an outcome
not of interest for this review. Subsequently, 13 articles
were eliminated since they had inadequate statistical
information.
As a result, 13 articles were included and
entered into the final analysis.
6–18
Of these, two articles
analysed the outcomes of interest separately for
males and females
9,12
and one article analysed two dif-
ferent outcomes within the same population,
10
so enter-
ing into the final analysis each as a single paper.
Included papers varied in size from 749 to 52,610
with a follow-up time ranging from three to 20 years.
A total number of 122,501 subjects free from car-
diovascular disease at baseline were analysed.
During the follow-up 6322 cardiovascular events were
reported.
Characteristics of the included papers are sum-
marized in Table 1. Three populations were from
the United States,
7,8,11
nine from Europe (France,
Germany, Norway, Sweden, The Netherlands,
UK),
6,9,10,12,15–18
and two from Asia (Japan,
Taiwan).
13,14
Insomnia was reported as trouble
with falling asleep or difficulty in initiating sleep in
most of the papers,
6–14,17,18
with some of them
reporting also other definitions such as presence of
non-restorative sleep or a number of symptoms
related to insomnia.
11,14–16
As for categories used
to estimate the association between exposure and
outcomes, the vast majority of the papers used the
dichotomous variable presence/absence,
6,7,9,11–13,17,18
with few papers using instead frequency measures
(e.g. most of the time, frequent, more than usual
etc.).
8,10,14–16
Insomnia and cardiovascular outcomes
Meta-analytic pooling under a random-effects model
showed that insomnia was significantly associated
with an increased risk of developing and/or dying
from cardiovascular disease during follow-up. Indeed,
subjects who reported suffering from insomnia had a
45%-increased risk of morbidity and/or mortality from
cardiovascular disease with respect to those who did
not suffer from sleep complaints (RR 1.45, 95% CI
1.29–1.62; p<0.00001) (Figure 1). No significant het-
erogeneity across the studies (I
2
¼19%; p¼0.24) was
evidenced.
Sensitivity analyses
In order to investigate the possible differences across
the studies we performed some sensitivity analyses by
grouping studies according to some characteristics such
as gender of the study populations (males/females),
country of cohorts (European/non-European), size of
the studies (mean size of the study sample: 7650), length
of follow-up (mean duration: 10 years) and adjustment
for possible confounders (only for age/also for other
demographic characteristics and cardiovascular risk
factors). Smaller studies and studies with a longer dur-
ation of follow-up showed a tendency towards a more
elevated estimate of association in terms of significant
increased risk of cardiovascular outcomes, with respect
to larger studies, and with a shorter follow-up time
(Table 2). Studies conducted in non-European coun-
tries reported a significant heterogeneity (I
2
¼61%;
p¼0.008), determined mainly by the study of Eaker
et al. (7) (I
2
¼29%; p¼0.24 after the exclusion).
Publication bias
The funnel plot of effect size versus standard error per-
formed in order to investigate for possible publication
bias reported to be broadly symmetric except for the
study by Eaker et al.,
7
which slightly diverged from
the mean effect size. After exclusion of this study
from the overall analysis, the visual inspection of pub-
lication bias was no longer present and the result did
not change in terms of increased risk of cardiovascular
disease for insomniac subjects (RR 1.41, 95% CI
1.28–1.56; p¼0.00001) (I
2
¼0%; p¼0.54).
Discussion
The present meta-analysis attempted to investigate the
possible association between insomnia and cardiovas-
cular outcomes in subjects free from cardiovascular dis-
ease at baseline. The overall analysis of 13 cohort
prospective studies investigating 122,501 healthy sub-
jects followed for a time ranging from three to 20
years and 6332 incident cases of cardiovascular acci-
dents reported that subjects who reported to suffer
from insomnia have a significantly increased risk of
developing or dying from cardiovascular disease
during follow-up. The cumulative analysis demon-
strated that subjects who reported to have either diffi-
culty in initiating sleep and maintaining sleep or
presence of disturbed nights have a 45%-increased
risk of morbidity and/or mortality from cardiovascular
disease, as compared with subjects with a good sleep
quality.
Sleep disturbances, in terms of poor quantity and/or
quality, have been consistently associated with a poor
Sofi et al. 5

XML Template (2012) [23.8.2012–6:04pm] [1–8]
{SAGE}CPR/CPR 460020.3d (CPR) [PREPRINTER stage]
Study Weight, % Risk ratio (95% CI)
Appels et al.6
Eaker et al.7
Schwartz et al.8
Mallon et al. (M)9
Mallon et al. (F)9
Elwood et al. (CHD)10
Elwood et al. (S)10
Phillips et al.11
Meisinger et al. (M)12
Meisinger et al. (F)12
Suzuki et al.13
Chien et al.14
Loponen et al.15
Chandola et al.16
Hulvej Rod et al.17
Laugsand et al.18
Total (95% CI)
3.8%
1.8%
5.5%
2.3%
1.5%
6.4%
3.9%
10.3%
8.4%
5.1%
8.7%
3.8%
4.5%
3.7%
14.4%
16.0%
1.60 (0.92–2.79)
3.90 (1.69–8.97)
1.44 (0.92–2.25)
3.10 (1.51–6.35)
2.00 (0.80–5.00)
1.47 (0.98–2.21)
1.75 (1.02–3.01)
1.50 (1.11–2.02)
1.16 (0.82–1.64)
1.30 (0.82–2.07)
1.02 (0.73–1.43)
1.78 (1.03–3.08)
1.42 (0.86–2.34)
1.36 (1.10–1.68)
1.18 (0.67–2.07)
1.53 (1.21–1.93)
N
3269
749
2960
906
964
1874
1874
11,863
3508
3388
11,395
3430
2753
12,524
52,610
10,308
n
59
39
152
71
20
213
107
546
295
85
310
420
287
155
2368
1205
123450.50.20.1
Increased riskDecreased risk
122,501 6332 100% 1.45 (1.29–1.62)
Heterogeneity: Chi2
= 18.43, df = 15 (p = 0.24); I2 = 19%
Test for overall effect: Z = 6.36 (p < 0.00001)
Figure 1. Forest plot of studies investigating insomnia in relation to risk of cardiovascular disease.
Table 2. Subgroup analyses
Studies, nRR (95% CI) Heterogeneity; p-value
Country
European countries 11 1.44 (1.28–1.61) I
2
¼0%; p¼0.61
Non-European countries 5 1.53 (1.12–2.10) I
2
¼61%; p¼0.008 (I
2
¼29%; p¼0.24,
after exclusion of the study by Eaker et al.
7
)
Gender
Males 13 1.41 (1.28–1.56) I
2
¼1%; p¼0.43
Females 9 1.44 (1.24–1.68) I
2
¼30%; p¼0.18
Sample size
<7650 subjects 11 1.58 (1.32–1.88) I
2
¼21%; p¼0.25
#7650 subjects 5 1.36 (1.19–1.56) I
2
¼10%; p¼0.35
Duration of follow-up
<10 years 7 1.39 (1.19–1.62) I
2
¼0%; p¼0.60
#10 years 9 1.52 (1.26–1.84) I
2
¼41%; p¼0.09
Adjustment
Only for age 3 1.56 (1.11–2.20) I
2
¼0%; p¼0.81
Multiple adjustment 13 1.44 (1.26–1.65) I
2
¼32%; p¼0.12
6European Journal of Preventive Cardiology 0(00)
5
6

XML Template (2012) [23.8.2012–6:04pm] [1–8]
{SAGE}CPR/CPR 460020.3d (CPR) [PREPRINTER stage]
quality of life and an increased risk of metabolic dis-
turbances such as cardiovascular disease.
2–5
The vast
majority of the literature, nonetheless, including some
data from longitudinal studies and meta-analyses,
reported a significant association between alterations
of duration of sleep and an increased risk of developing
and/or dying from cardiovascular disease. In 2011,
Cappuccio et al. carried out a meta-analysis on the
role of sleep duration on the occurrence of cardiovas-
cular diseases in healthy subjects, reporting that both
short and long duration of sleep are associated with an
increased risk of developing cardiovascular disease.
3
With regard to quality of sleep, fewer studies, with con-
flicting results, investigated its relationship with cardio-
vascular health.
6–18
In 1999, a systematic review on the
relationship between insomnia and cardiovascular
health was carried out by Schwartz et al.; however, it
included a limited number of studies and studies with
both case-control and cohort design.
5
The present
meta-analysis, on the other hand, is the first, to the
best of our knowledge, that has analysed insomnia
and cardiovascular risk only in cohort studies.
The choice of studying insomnia in relation to the
occurrence of cardiovascular outcome has been based
on the hypothesis that sleep complaints may be asso-
ciated with an increased risk of myocardial infarction.
The mechanisms that underlie this possible association
have been reported both by epidemiologic and experi-
mental studies but are not completely understood. On
one hand, epidemiologic studies reported that poor-
quality sleep, as well as short duration or presence of
non-restorative sleep, is associated with metabolic or
endocrine changes through sympathetic activation or
through elevated levels of inflammatory cytokines.
19,20
A population study conducted recently by Friedman
showed a low-grade inflammation state in short
sleeps, with similar mechanisms that can be activated
also during a low-quality sleep.
21
Moreover, another
population-based study reported that polysomno-
graphic correlates of insomnia are significantly more
prevalent among subjects with self-reported health dis-
orders such as depression, low socio-economic status,
low level of physical activity and poor general health,
thus possibly representing only confounding factors of
other health conditions.
22
On the other hand, some
experimental studies provided insights into the mech-
anisms of the metabolic consequences of insomnia.
Indeed, an experimental study published in 2008
demonstrated that reduced quality of sleep with low
levels of slow-wave sleep could determine an increased
risk of type 2 diabetes, through an impaired glucose
tolerance.
23
Moreover, another elegant experimental
study conducted by Spiegel et al.
24
demonstrated that
sleep curtailment in healthy young subjects is able to
determine an increased cortisol secretion and altered
circulating levels of growth hormone, leptin and ghre-
lin, by likely determining an increased risk of being
obese in these subjects.
Few limitations can be identified in this meta-analy-
sis. First of all, the methods used to investigate insom-
nia varied substantially across the included studies.
This might have determined a non-homogenous defin-
ition of insomnia among the studies. Moreover, data
were obtained from questionnaires, so encountering
bias due to misinterpretation of the question proposed
and to the personal perception of non-restorative sleep.
In addition, studies differ in methods used to classify
the presence of insomnia, ranging from studies with a
simple differentiation of present/absent, to others struc-
tured on three or four classes of intensity. Nevertheless,
however, heterogeneity results as well as subgroup ana-
lyses did not show any significant differences among
subjects in terms of risk reduction, thus being independ-
ent of all of these characteristics.
In conclusion, these results highlight the important
role of stress and stress-related behaviours on the
occurrence of cardiovascular disease.
25
The evidence
that low-quality sleepers are at higher risk of myocar-
dial infarction and other cardiovascular disease is extre-
mely important for primary and secondary prevention
since sleep patterns of quantity and quality are, in
actual fact, affected by a variety of cultural, social, psy-
chological and environmental influences. These consid-
erations could be very valuable given that changes in
modern societies include longer working hours, more
shift-work and more stressful appointments, likely
determining a relevant modification of sleep patterns
in the general population.
Funding
This research received no specific grant from any
funding agency in the public, commercial, or not-for-profit
sectors.
Conflict of interest
None declared.
Author contributions
Conception and design: FS, GF; analysis and interpretation
of the data: FS, FC, CM, AC; drafting of the paper: FS, FC,
RA; critical revision of the paper for important intellectual
content: AC, CM, RA, GFG; final approval of the paper: FS,
FC, AC, RA, GFG; statistical expertise: FS.
References
1. Ohayon MM. Epidemiology of insomnia: What we know
and what we still need to learn. Sleep Med Rev 2002; 6:
97–111.
Sofi et al. 7
3

XML Template (2012) [23.8.2012–6:04pm] [1–8]
{SAGE}CPR/CPR 460020.3d (CPR) [PREPRINTER stage]
2. Hamer M, Batty GD and Kivimaki M. Sleep loss due to
worry and future risk of cardiovascular disease and all-
cause mortality: The Scottish Health Survey. Eur J
Cardiovasc Prev Rehabil. Epub ahead of print 3
October 2011.
3. Quan SF. Sleep disturbances and their relationship to
cardiovascular disease. Am J Lifestyle Med 2009; 3:
55–59.
4. Cappuccio F, Cooper D, D’Elia L, et al. Sleep duration
predicts cardiovascular outcomes: A systematic review
and meta-analysis of prospective studies. Eur Heart J
2011; 32: 1484–1492.
5. Cappuccio FP, D’Elia L, Strazzullo P, et al. Quantity and
quality of sleep and incidence of type 2 diabetes. Diabetes
Care 2010; 33: 414–420.
6. Schwartz S, McDowell Anderson W, et al. Insomnia and
heart disease: A review of epidemiologic studies.
J Psychosom Res 1999; 47: 313–333.
7. Appels A, de Vos Y, van Diest R, et al. Are sleep com-
plaints predictive of future myocardial infarction?
Act Nerv Super (Praha) 1987; 29: 147–151.
8. Eaker ED, Pinsky J and Castelli WP. Myocardial infarc-
tion and coronary death among women: Psychosocial
predictors from a 20-year follow-up of women in the
Framingham Study. Am J Epidemiol 1992; 135: 854–864.
9. Schwartz SW, Cornoni-Huntley J, Cole SR, et al. Are
sleep complaints an independent risk factor for myocar-
dial infarction? Ann Epidemiol 1998; 8: 384–392.
10. Mallon L, Broman JE and Hetta J. Sleep complaints pre-
dicts coronary artery disease mortality in males: A
12-year follow-up study of a middle-aged Swedish popu-
lation. J Intern Med 2002; 251: 207–216.
11. Elwood P, Hack M, Pickering J, et al. Sleep disturbances,
stroke, and heart disease events: Evidence from the
Caerphilly cohort. J Epidemiol Community Health 2006;
60: 69–73.
12. Phillips B and Mannino DM. Do insomnia complaints
cause hypertension or cardiovascular disease? J Clin
Sleep Med 2007; 3: 489–494.
13. Meisinger C, Heier M, Lowel H, et al. Sleep duration and
sleep complaints and risk of myocardial infarction in
middle-aged men and women from the general popula-
tion: The MONICA/KORA Augsburg cohort study.
Sleep 2007; 30: 1121–1127.
14. Suzuki E, Yorifuji T, Ueshima K, et al. Sleep duration,
sleep quality and cardiovascular disease mortality among
the elderly: A population-based cohort study. Prev Med
2009; 49: 135–141.
15. Chien KL, Chen PC, Hsu HC, et al. Habitual sleep dur-
ation and insomnia and the risk of cardiovascular events
and all-cause death: Report from a community-based
cohort. Sleep 2010; 33: 177–184.
16. Loponen M, Hublin C, Kalimo R, et al. Joint effect of
self-reported sleep problems and three components of the
metabolic syndrome on risk of coronary heart disease.
J Psychosom Res 2010; 68: 149–158.
17. Chandola T, Ferrie JE, Perski A, et al. The effect of short
sleep duration on coronary heart disease risk is greatest
among those with sleep disturbance: A prospective study
from the Whitehall II cohort. Sleep 2010; 33: 739–744.
18. Hulvej Rod N, Vahtera J, Westerlund H, et al. Sleep
disturbances and cause-specific mortality: Results from
the GAZEL cohort study. Am J Epidemiol 2011; 173:
300–309.
19. Laugsand LE, Vatten LJ, Platou C, et al. Insomnia and
risk of acute myocardial infarction. A population study.
Circulation 2011; 124: 2073–2081.
20. Copinshi G. Metabolic and endocrine effects of sleep
deprivation. Essent Psychopharmacol 2005; 6: 341–347.
21. Taheri S, Lin L, Austin D, et al. Short sleep duration is
associated with reduced leptin, elevated ghrelin, and
increased body mass index. PLoS Med 2004; 1: e62.
22. Friedman EM. Sleep quality, social well-being, gender,
and inflammation: An integrative analysis in a national
sample. Ann N Y Acad Sci 2011; 1231: 23–34.
23. Budhiraja R, Roth T, Hudgel DW, et al. Prevalence and
polysomnographic correlates of insomnia comorbid with
medical disorders. Sleep 2011; 34: 859–867.
24. Tasali E, Leproult R, Ehrmann DA, et al. Slow-wave
sleep and the risk of type 2 diabetes in humans. PNAS
2008; 105: 1044–1049.
25. Spiegel K, Tasali E, Penev P, et al. Sleep curtailment in
healthy young men is associated with decreased leptin
levels, elevated ghrelin levels, and increased hunger and
appetite. Ann Intern Med 2004; 141: 846–850.
26. Kornerup H, Osler M, Boysen G, et al. Major life events
increase the risk of stroke but not of myocardial infarc-
tion: Results from the Copenhagen City Heart Study. Eur
J Cardiovasc Prev Rehabil 2010; 17: 113–118.
8European Journal of Preventive Cardiology 0(00)
4