Heart and lungs

Blood test may predict heart risk

A simple blood test “can spot heart failure years in advance,” The Daily Telegraph has reported. The test can identify problems “before any outward symptoms appear, allowing doctors to advise those at risk to change their lifestyle”, the newspaper added.

The story is based on a new study that found that blood levels of a protein called cardiac troponin T (cTnT) could help predict the risk of heart failure and death from cardiovascular disease in older adults.

Testing cTnT levels is already used to help identify patients who are suspected of having a heart attack or other damage to the heart but in this study, researchers have used a new, highly sensitive test that was able to detect cTnT in two-thirds of the study participants.

The finding that blood levels of cTnT are associated with the risk of heart failure and death is of interest, and the test may have some role alongside other predictive measurements. However, as the researchers point out, cTnT measurements did not greatly improve their ability to predict these problems when combined with identifying established risk factors in patients. Furthermore, the tested blood samples were collected several years ago meaning that more widespread use of medication, such as statins for cardiovascular problems, may further reduce the predictive value of such a test.

Where did the story come from?

The study was carried out by researchers from the University of Maryland, Baltimore, and the University of Texas, Dallas in the US. It was mainly funded by the US National Heart, Lung and Blood Institute, with additional contributions from the US National Institute of Neurological Disorders and Stroke. 

The study was published in the peer-reviewed Journal of the American Medical Association.

The research was reported at length in The Daily Telegraph, which described the methods used in the study and included comments from independent experts that highlight their reservations about the usefulness of such a test. However, the Telegraph’s headline and introduction perhaps exaggerated the findings of the study, without mentioning its limitations.

What kind of research was this?

The researchers point out that older adults comprise the majority of new heart failure patients but that assessing risk factors to predict those at highest risk has only limited accuracy in this population.

Various biomarkers in the blood, including types of proteins called troponins, have been tested as an additional way to predict risk but studies so far have had inconsistent results. In addition, cardiac troponin T (cTnT) circulates at such low levels that standard tests have been unable to detect it in the blood of many people. However, recently researchers have developed a more sensitive cTnT test (or assay) that has the ability to pick up extremely low levels of cTnT. The test has detected blood levels of cTnT in almost all patients with established heart failure or other types of heart disease.

This was a longitudinal cohort study to find out whether the new assay could detect levels of cTnT in older people (65 years or more) who had not previously been diagnosed with heart failure. It also investigated if these initial measurements of cTnT levels, or changes in these levels, were associated with the risk of future heart failure or death from heart disease or stroke.

Cohort studies, which follow people over a number of years, are a useful way of finding out whether a particular factor (in this case, cTnT levels before diagnosis) is associated with particular outcomes (in this case, heart failure or death from heart disease or stroke).

What did the research involve?

The researchers included 5,613 participants recruited from an ongoing study of cardiovascular health in their study. They were all 65 years or older and free of heart failure at the start. However, 1,392 of this group had no blood sample available, leaving 4,221 participants who could have their cTnT measured using the new test at the start of the study (1989-90 for the main cohort, or 1992-3 for a supplemental cohort of black participants).

After two to three years, 2,918 of the participants had their cTnT levels measured again (the rest had for various reasons been excluded). Researchers followed up participants for an average of 11.8 years to establish any diagnoses of heart failure and cardiovascular death.

The researchers diagnosed heart failure by visiting the participants and evaluating Medicare insurance claim data. Both heart failure and the cause of any deaths were determined by an expert panel, which used physician diagnoses, medical records and other relevant data.

To carry out their analysis, researchers divided participants into five groups depending on the levels of cTnT found, with the first group being those with undetectable levels. They used standard statistical methods to analyse the relationship between cTnT levels and the risk of heart failure or death, adjusting their findings to take account of factors that might affect the results (called confounders). These included age, race, sex, traditional risk factors such as smoking, and other biological markers.

The researchers also compared the risk of heart failure for people whose cTnT levels changed more than 50% at the second measurement, with those whose levels had changed by 50% or less.

What were the basic results?

They found cTnT levels were detectable in 2,794 participants (66.2%). During follow up, 1,279 people were diagnosed with new onset heart failure and there were 1,103 cardiovascular deaths, with a higher risk of both associated with higher cTnT levels.

Below are the main findings:

  • participants with the highest cTnT levels were over twice as likely to develop heart failure as those with undetectable cTnT levels (adjusted hazard ratio [aHR], 2.48; 95% confidence interval [CI] 2.04 to 3.00)
  • those with the highest cTnT levels were also nearly three times as likely to die from heart disease or stroke as those with undetectable cTnT levels (aHR, 2.91, 95% CI 2.37 to 3.58)
  • among individuals with initially detectable cTnT, a subsequent increase of more than 50% was associated with a greater risk for heart failure (aHR, 1.61, 95% CI 1.32 to 1.97) and cardiovascular death (aHR 1.65, 95% CI 1.35 to 2.03)
  • a cTnT decrease of more than 50% was associated with lower risks of both heart failure (aHR 0.73, 95% CI, 0.54 to 0.97) and death related to heart or stroke (aHR 0.71, 95% CI 0.52 to 0.97) compared with participants with a change of 50% or less.
  • supplementing known clinical risk factors with data on baseline cTnT levels was associated with only modest improvement in predicting those at risk

How did the researchers interpret the results?

The researchers say that in their study of a population of older adults, both the baseline cTnT levels and changes in cTnT levels (as measured with a highly sensitive test) were associated with a higher risk of heart failure and cardiovascular death. They also point out that the changes in low levels of cTnT, which were associated with a change in risk, were common, indicating that serial measurements of this protein may improve risk assessment in older adults.


This is a useful study but it has several limitations, as the researchers have noted:

  • blood samples were available only in three-quarters of the original cohort, which could have introduced a bias in the result
  • this study was initiated over two decades ago, so the way we currently use medications, such as statins, may blunt the predictive value of the test
  • other ‘confounders’ both measured and unmeasured, may have influenced the results

In conclusion, this interesting study shows a significant association between elevated troponin T, as measured by a new assay, and the risk of heart failure and death from heart disease or stroke in older adults with no prior diagnosis of heart failure. However, the results made only a modest difference to predictions based on clinical risk factors and its usefulness for the future is therefore uncertain.

Individuals who reduce their modifiable risk factors: smoking, high blood pressure, lipid levels or blood sugar, will not need to wait for further research. It is known that these are linked with heart disease and deaths from heart disease and stroke.

NHS Attribution