Thursday June 2 2011
This small study had several limitations
“Patients who feel scared of dying during heart attack symptoms may be more likely to suffer another,” the Daily Mirror has reported.
The news is based on a small study in 208 people who were admitted to hospital with chest pain. The patients were asked three questions designed to assess their level of fear, whether they thought they might die and feelings of stress. Researchers compared their answers to the results of blood tests, taken when the patients were admitted to hospital, that measured levels of a chemical associated with inflammation, as well as heart rate or stress hormones three weeks later. Inflammation is known to both damage the heart and occur in response to heart damage.
The study found that people who were more distressed when admitted to hospital had higher levels of inflammation markers as well as lower levels of stress hormones three weeks later. However, the study had several limitations. Principally, it did not assess the risk of a second heart attack, but only looked at markers of inflammation at the start of the study. Also, about 50% of participants chose not to take part in the follow-up tests three weeks after hospital admission. These were mainly people who were unmarried and from poorer backgrounds. This means that the data from this study need to be interpreted cautiously.
Given the limited scope of this early research, a link between inflammatory markers in the blood and emotional distress needs further investigation.
Where did the story come from?
The study was carried out by researchers from University College London, the University of Stirling, the University of Bern and St George’s Hospital in London. It was supported with grants from the British Heart Foundation, the Medical Research Council and the Swiss National Foundation.
The research paper was published in the peer-reviewed European Heart Journal.
The Daily Mirror uncritically reported the researchers’ main findings. The BBC included quotes that highlighted some of the study’s limitations.
What kind of research was this?
This study featured a cross-sectional analysis that looked for a link between emotional responses when people were admitted to hospital for acute coronary syndrome (ACS) and the level of inflammatory response at the same time. Short-term changes in heart rate variability and stress hormone levels were also measured three weeks after hospital admission.
ACS is defined as a blockage or narrowing of the coronary arteries and includes heart attacks. As inflammatory responses are known to both damage the heart and occur in response to heart damage, the researchers wanted to see if fear of dying was linked to inflammatory changes. If it was, this could explain why, for example, depression after ACS is associated with recurrent cardiac events and impaired quality of life.
The study had two main aims:
- to evaluate whether acute distress and fear of dying were associated with levels of an inflammatory marker (TNF alpha) at the time of hospital admission for ACS
- to discover whether TNF alpha and fear of dying during ACS were related to reduced heart rate variability and levels of cortisol (a stress hormone) three weeks later
TNF alpha (Tumour necrosis factor) is a maker of inflammation that, along with other inflammatory markers, rises during heart attacks. Levels of inflammatory markers are known to predict both short- and long-term risk of recurrent cardiac events and heart problems. Acute psychological stress also stimulates an increase in the concentration of TNF alpha within 1-2 hours of stress, according to the researchers.
The study’s design was appropriate for investigating the researchers’ very specific questions. However, from reading the coverage in the media, it would be possible to think that hard outcomes such as deaths from heart attack or second heart attacks were measured, when they were not. Also, because fear of dying and inflammatory response were assessed at the same time, it is not possible to say whether the fear of dying might cause changes in inflammatory markers, or vice versa. Other factors that were not measured may also have influenced the results.
What did the research involve?
To look at the theoretical link between distress and inflammatory markers, the researchers recruited 208 patients admitted to a South London hospital with clinically verified ACS between June 2007 and October 2008.
Patients were included if they had chest pain accompanied by typical ECG changes, had markers of heart muscle damage (troponin T or troponin I or CK) raised beyond normal values and were 18 years or over without other illnesses. In addition, they needed to be able to complete interviews and questionnaires in English.
Although 666 potentially eligible patients were admitted to hospital during the recruitment period, many were excluded from taking part for various reasons. These included patients being discharged or transferred too quickly, being too clinically fragile to take part, the blood test (TNF alpha) being unavailable, not speaking English, being confused or declining to take part. This only left 208 participants for the study. Complete heart rate data at three weeks was only available for 106 people (50%) and data on cortisol levels for 110 (53%).
The researchers gave all recruits a three-item questionnaire, which asked them to score on a scale of one to five (from “not true” to “extremely true”) the following statements:
- I was frightened when the symptoms came on.
- I thought that I might be dying when the symptoms came on.
- I found my cardiac event stressful.
They divided the patients into three groups – those with no distress and fear, moderate distress and fear, and intense distress and fear – and took blood tests for TNF alpha levels.
After an average of three weeks (21.9 days, variance +/-8.4 days), researchers visited the participants at home and measured cortisol output by collecting saliva samples over a day, and also measured heart rate variability (differences in heart rate during five-minute recording spells). Both of these measurements are thought to indicate levels of stress. The researchers then looked at whether there were relationships between fear of dying, TNF alpha levels at the start of the study, and either heart rate variability or cortisol levels after ACS.
The researchers adjusted their results for other factors that might have influenced this link, such as age, gender, marital status, ethnicity, social deprivation, statin and aspirin use before admission to hospital, pain during the ACS, a measure of heart attack severity (the GRACE score), and number of days spent in hospital.
What were the basic results?
Most participants were men (84%). Out of 208 participants, intense distress and fear of dying was reported by 45 (21.6%), moderate distress by 116 (55.8%), and low distress and fear of dying by 47 (22.6%). Fear of dying was more common in patients who were younger, of lower socioeconomic status and unmarried.
Fear of dying was associated with blood levels of the inflammatory marker TNF alpha at time of admission after the researchers adjusted their results for sociodemographic factors, clinical risk and pain intensity. This means that the odds of having a high level of TNF alpha were greater in patients with a high fear of dying on the questionnaire than those with a low fear of dying (adjusted odds ratio 4.67, 95% confidence interval 1.66 to 12.65).
Higher levels of TNF alpha at admission were associated with reduced heart rate variability three weeks later, after the researchers adjusted for clinical and sociodemographic factors and medication, while greater fear of dying was associated with reduced cortisol output. All results were statistically significant, which means that they were unlikely to have occurred due to chance.
How did the researchers interpret the results?
The researchers say that “intense distress and fear of dying and heightened inflammation” may be related early responses to severe muscle injury and have implications for future heart attack risk.
They go on to suggest that understanding the relationship between psychological and biological factors in heart attacks would open the possibility of new avenues for patient management.
This early research looked at the potential links between psychological and biological symptoms in heart disease. This area needs further study. The study has limitations, some of which the authors mentioned:
- Of those who completed the study and were followed for three weeks, 77% participated in the interview, but only about 50-55% had their heart rate variability and cortisol tested. Participation was lower in unmarried patients and those from poorer backgrounds. The researchers say that these groups of people are more likely to withdraw from medical research and surveys, but claim that those who did not participate did not differ from the remaining participants in their fear of dying and inflammatory responses. This suggests that the loss of these participants is less likely to have influenced the results.
- The researchers only assessed inflammation and fear of dying at the start of the study and not at the three-week follow-up. Therefore, it is not possible to say for sure that the three-week measurements taken were associated with persistent inflammation or fear of dying.
- Some of the measurements, for example the heart rate variability, were not done under standard conditions. Measuring these factors in the patients’ homes may have led to inaccuracies.
- The analysis involved scores of fear and distress, but these emotions can be hard to quantify as individuals may experience or interpret them in different ways.
Most importantly, this study did not look at clinical outcomes such as heart attack or death. Therefore, it is not possible to say from this study whether a fear of dying influences these outcomes. Also, as fear of dying and inflammation were assessed at the same time, it is not possible to say for certain whether fear of dying caused the increase in the inflammatory marker.
Overall, this study provides further avenues for research, but the picture is not yet complete enough to conclude that fear of dying is itself linked to inflammatory markers in the blood in a way that predicts the long-term risk of heart attacks.