1
15 October 2017
Contribution of stress to the aetiology and prognosis of cardiovascular
disease
Mika Kivimäki and Andrew Steptoe
Department of Epidemiology and Public Health (M Kivimäki) and Research Department of
Behavioural Science and Health (A Steptoe), University College London, 1–19 Torrington
Place, London WC1E 6BT, UK
Brief biography:
Mika Kivimaki is Professor and Chair of Social Epidemiology at University College London,
UK. He leads several cohort studies and research consortia on the risk factors of
cardiovascular disease and has published more than 900 peer-reviewed papers, including
state-of-the-art reviews and meta-analyses in the Lancet and Nature Reviews with particular
focus on the health effects of chronic stress.
Andrew Steptoe is Head of the Department of Behavioural Science and Health in the Institute
of Epidemiology and Health Care in the Faculty of Population Health Sciences, University
College London, UK. He was British Heart Foundation Professor of Psychology from 2000-
2016, and is well known for research on cardiovascular effects of episodic stress and
depression. He has published more than 550 peer-reviewed articles and is author or editor of
19 books, including Stress and Cardiovascular Disease (London: Springer-Verlag 2012).
Manuscript statistics:
212 words in Abstract, 7239 words in text, 180 references, 2 Boxes, 2 Tables, 5 Figures.
Correspondence to:
Prof Mika Kivimäki, Department of Epidemiology and Public Health, University College
London, 1-19 Torrington Place, WC1E 6BT London, UK
Email: m.kivimaki@ucl.ac.uk
2
Abstract | Cardiovascular disease remains the leading cause of disease burden
globally. This underlies the continuing need to identify new complementary targets
for prevention. Over the last 5-10 years, pooling of multiple datasets into mega
studies has accelerated progress in research on stress as a risk factor for
cardiovascular disease. Severe stressful experiences in childhood, such as physical
abuse and household substance abuse, can damage health and increase the risk of
multiple chronic conditions in adulthood. Compared with stress in childhood and
adulthood classic risk factors, such as smoking, high blood pressure and serum
cholesterol, the harmful effects of stress in adulthood are generally less marked.
However, adulthood stress has an important role as a disease trigger in persons who
already have a high atherosclerotic plaque burden and as a determinant of prognosis
and outcome in those with preexisting cardiovascular or cerebrovascular disease.
Mechanistic studies corroborate in real-life settings earlier laboratory-based
observations on stress-related pathophysiological changes that underlie triggering,
such as lowered arrhythmic threshold, increased sympathetic activation with related
increases in blood pressure as well as proinflammatory and procoagulant responses.
In some clinical guidelines, stress is already acknowledged as a target for prevention
for people with high overall cardiovascular risk or established cardiovascular disease.
However, few scalable evidence-based interventions are currently available.
3
Introduction
Advances in understanding the risk factors of cardiovascular disease have been
extraordinary given that the first cohort studies for cardiovascular disease, such as the
Minnesota Business and Professional Men study
1
, the Framingham Heart study,
2
and the
British Doctors study,
3
were initiated only six or seven decades ago. Since that time
declines in age-specific cardiovascular disease rates have been enormous, more than
those for cancer and neurological disorders.
4
According to the Global Burden of Disease
estimates, age-standardised years of life lived with ishaemic heart disease declined by
47% and that of cerebrovascular disease by 17% between 1990 and 2016. Reductions in
recognized cardiovascular disease risk factors, such as smoking, blood pressure and
serum cholesterol, have helped in preventing the disease.
5,6
In addition, the prognosis of
cardiovascular disease has improved because of improved medical care, including
diagnosis, treatment of acute events, and post-hospital care.
5,6
Despite this progress, coronary heart disease and stroke remain the two leading
causes of disease burden globally.
4
From 2006 to 2016, the number of people dying from
cardiovascular disease actually increased by 15% reaching 17.6 million per year, because
of population ageing and growth.
4
The number of years lost due to ill-health, disability or
early death from cardiovascular disease is 18 times higher, 321.9 million per year.
7
This
underlines the continuing need to identify new complementary targets for prevention and
to evaluate future potential of therapeutic management of patients.
One emerging risk and prognostic factor for cardiovascular disease is stress, the
topic of this review. A large body of prospective research links stress to increased rates of
coronary heart disease, stroke and other presentations of cardiovascular disease, such as
atrial fibrillation. In addition, analyses of 27,461 to 603,838 adults with prospective
follow-up confirm outcome specificity, supporting the associations with cardiovascular
4
disease,
8-11
diabetes
12-14
and mental health
15
but not with cancer of all sites, the lung,
prostate, breast, colon or rectum
16,17
or diseases in the gastrointestinal
18
and respiratory
system.
19,20
In this Review, we assess the current evidence for stress as a risk factor of
cardiovascular disease. Our main focus is on reproducible findings obtained from the
largest stress studies to date and most recent literature-based and individual-participant-
data meta-analyses. Rather than just updating previous field synopses,
21-26
we seek to
determine the clinical importance and population impact of major stressors in relation to
various presentations of cardiovascular disease and provide comparisons to more
established risk factors and treatments for cardiovascular disease, such as smoking,
physical activity, and statin therapy.
To cover the multiple roles of stress in vascular pathology, the Review is
organized according to the disease process, from the long-term development of
atherosclerosis to subclinical disease and the acute triggering of cardiac or
cerebrovascular events in people with advanced underlying disease. In addition to this
aetiological process, stress is thought to impair recovery, accelerate the progression of the
disease, and contribute to cardiovascular death among patients who have survived an
acute coronary syndrome or stroke: in other words, it may act as a prognostic factor.
Evidence on these factors is also included in the Review. We conclude this paper with
implications for cardiovascular disease prevention and treatment.
Measurement of stress
The most commonly studied adult stressors include stress at work and social isolation,
22
although other stressors, such as marital problems, caring for a sick spouse or child at
home and the death of a close person, have also been linked to increased risk.
27-29
Further
5
stressors are natural and man-made disasters (e.g., earthquakes, terrorist attacks, wars,
etc.) that elicit clinical events and cardiac abnormalities.
30,31
Commonly studied episodic
stressors include emotional upset, sporting events and stressful changes at work, such as
layoffs.
32,33
We review evidence on these stressors and also overall perceived stress
whereas other reviews are available for depression,
34
35,36
anxiety,
37
common mental
disorder/psychological distress,
38
vital exhaustion
39
and posttraumatic stress disorder,
40
which, although being correlates of stress, represent consequences rather than sources of
stress. Various personality types or characteristics have been considered as being risk
factors for both psychological distress and cardiovascular disease. These include the type
A behaviour pattern
41
and type D personality
42
as well as anger-proneness and hostility.
43
We do not discuss these personality factors as they have been addressed elsewhere.
41-43
There are at least three distinct ways to measure stress: (a) by self-report
(perception), (b) by external observation, and (c) by biomarkers. The latter include
measuring stress on moment-to-moment basis (e.g., via salivary cortisol) and summary
measures over longer time spans (e.g., 3-month cortisol assay by analyzing hair
samples)(Box). While self-completion questionnaire and interviews are still the most
commonly used methods, new technology is increasingly providing opportunities to make
progress in stress measurement. A wide range of mobile, even wearable, electrochemical
sensors and biosensors has been developed for real-time non-invasive monitoring of
electrolytes and metabolites in sweat or saliva as indicators of a wearer's biological status;
‘mHealth’ applications in mobile telephones and wearable devices for interactive
communication allow more precise monitoring of specific stress exposures and their
timing in relation to physiological changes.
44
For example, PTSD Coach is a smartphone
application designed to help individuals who have post-traumatic stress disorder (PTSD)
symptoms measure and self-manage their symptoms;
45
Surface Enhanced Raman