Heart and lungs

Stress and artery health studied

“Stress really does increase the risk of heart attacks and strokes,” reported the Daily Mail. It said research has found that people who become stressed are more likely to suffer from hardened arteries.

This study measured volunteers’ levels of cortisol, a stress hormone, while they carried out tests aimed at raising their stress levels. It found that people who had increased cortisol levels were more likely to have high calcium deposits in the arteries, a marker of coronary heart disease.

Although high calcium deposits may indicate heart disease, this study did not directly investigate if stress increases the risk of a heart attack or stroke. A single measure of stress taken at the same time as a measure of calcium build-up in the arteries cannot show whether a person’s lifetime stress habits have caused the build-up.

Although further research is needed, minimising stress is known to be associated with improved mental and physical wellbeing. 

Where did the story come from?

This research was carried out by Dr Mark Hamer and colleagues from University College London and Wellington Hospital. The study was funded by the British Heart Foundation and the Medical Research Council. The paper was published in the peer-reviewed European Heart Journal.

What kind of research was this?

This preliminary research looked for associations between stress in older humans, measured by levels of cortisol, and coronary artery calcification (CAC), which is measured by computed tomography. The authors say that CAC is an indicator of subclinical coronary atherosclerosis, and is a predictor of future coronary heart disease (CHD) events.

This is a cross-sectional study, therefore it cannot establish causation for CHD but only highlight factors that may be associated. A more reliable method of investigating the question would be a cohort study, in which people who were free from heart disease at the beginning of the study had their stress and anxiety levels measured and were followed over a period of time to see whether they developed heart disease.

What did the research involve?

The study sampled 514 participants from the Whitehall II epidemiological cohort, a previous study that looked at social class and mortality from a wide range of diseases. The participants had no history of CHD and no previous diagnosis or treatment of hypertension (high blood pressure), inflammatory diseases or allergies. They were of white European origin, and aged between 53 and 76 years old (average age 62.9 years). The selection procedure ensured that participants of higher and lower socioeconomic status were included.

The study included information on the participants’ height and weight, whether they smoked or not, and their blood cholesterol and fat levels.

Before any tests were performed, participants were asked to not take any antihistamines or anti-inflammatory medication for seven days. They were also asked not to drink alcohol or do rigorous exercise the day before, nor drink caffeinated beverages or smoke two hours before the test.

The participants’ baseline (starting point) blood pressure was taken, along with a saliva sample. Mental stress was induced using two tests: the Stroop test and the mirror tracing test. The Stroop test asks participants to read out colours that are written down in differently coloured text, while the mirror tracing test involves drawing a shape while only being able to see your hand as a reflection in a mirror. Saliva samples were taken 20, 45 and 75 minutes after the task finished. Cardiovascular measurements were taken continuously during and afterwards.

Levels of the stress hormone cortisol were measured in the saliva samples, while coronary artery calcification was measured using computed tomography.

What were the basic results?

The researchers grouped the participants into two groups: those who had a rise in cortisol in response to the stress tests (responders), and those who did not (non-responders). There were 308 non-responders and 206 responders.

The two groups did not differ in their socioeconomic or smoking status, their height and weight, or blood measurements.

In total, 56% of participants had evidence of coronary artery calcification (CAC). The risk of having CAC increased with age, and men were more likely to have CAC than women.

When the researchers looked at any detectable CAC (greater or equal to one on the Agatston scale) they found no association between cortisol response and CAC. When they looked at participants who had high CAC scores (greater than or equal to 100), there was an association between cortisol response and CAC (odds ratio [OR] 2.20 95% confidence interval [CI] 1.39 to 3.47). These results had been adjusted for other factors that are associated with CAC (age, sex, BMI and a measure of diabetes [the amount of sugar attached to haemoglobin, the oxygen carrying chemical in red blood cells]).

How did the researchers interpret the results?

The researchers concluded that people whose cortisol levels increased when given acute behavioural tasks were associated with high CAC scores.

They suggest that because high CAC scores can predict risk of coronary heart disease, their results may support the theory that psychosocial stress affects the risk of coronary heart disease.


This research found that people whose cortisol levels increased when given acute behavioural tasks were associated with high CAC scores, an indicator of cardiovascular disease. However, the researchers were relatively cautious in the interpretation of their work, and highlighted the following limitations of their study.

  • As the study was cross-sectional, it cannot demonstrate causality, i.e. that one thing causes another. As such, it is not possible to conclude that stress is responsible for increased CAC and, therefore, a higher risk of heart attacks and stroke. It is possible that sub-clinical CHD may make people more easily stressed or affect how they respond to laboratory tasks.
  • The researchers found that only 40% of participants responded to the problem-solving tasks that were used as stressors with increased cortisol levels. It is possible that these tasks may not accurately represent stressors in real life, or induce the same levels of cortisol.
  • Cortisol stress responses were only measured on a single occasion and the participants were only grouped into non-responder and responder groups. Therefore, it was not possible to see whether there was an association between the degree of cortisol elevation and CAC.
  • Although calcium is a component of atherosclerotic plaques and may be considered to be a marker of blood vessel disease, it cannot tell us whether the person currently has, or is at risk of developing, cardiovascular disease (for example, more likely to experience angina or be at risk of a heart attack).

This is well-performed research on a relevant study question. However, further research is needed to assess whether there is an association between stress and coronary heart disease. Nevertheless, minimising stress is known to be associated with improved mental and physical wellbeing.

NHS Attribution