The patient with chest pain is the most common presentation to Emergency Departments. And certainly the low risk chest pain patients seem to take up a large degree of our resources, including time, laboratory testing, and follow up. In the past, we’ve used scoring systems such as TIMI, PURSUIT and GRACE. These scoring systems have either been too complicated, are now outdated, not sensitive enough and certainly were not really designed for use in the Emergency Department. A new score, the HEART score is a means of risk stratification that uses simple scoring, allowing us to sort out our low risk patients and stratifying them at 30 days.
The original study was carried out in the Netherlands, by Six, A.J., et al., Netherlands Heart Journal, Volume 16, Number 6, June 2008, pages 191 – 196. This study was a risk classification study that looked at working out, who was at a lowest risk of having a cardiac event. The study was an observational retrospective study and involved 5 parameters. These parameters are easily remembered by the word HEART. The parameters were:
- History
- ECG
- Age of patient
- Risk factors
- Troponin I
The end points of the study were an acute myocardial infarction, percutaneous coronary intervention, or bypass surgery. 122 patients were involved in the study of which 29 reached the end points above.
A score of less than or equal to 3 had a risk end point of 2.5%. That means that if you scored less than or equal to 3 for the HEART score you had a 2.5% chance of having a cardiac episode at 6 weeks.
What is the HEART Score?
History
If the history is non-specific ie. in terms of onset or duration or localisation the patient scores 0 points.
If there are specific and nonspecific elements in the history the patient scores 1 point.
If there is a high suspicion in the history the patient gets 2 points.
ECG
If the ECG is totally normal the patient gets 0 points.
If the ECG has any of the following:
- Repolarisation abnormalities with no significant ST depression
- Bundle branch block
- Left ventricular hypertrophy indicative changes
- Repolarisation abnormalities due to the digoxin
then these patients all get 1 point.
If the ECG has ST then depression or elevation in the absence of a bundle branch block or left ventricular hypertrophy the patient gets 2 points.
Age
<45 years old is 0 points.
45 – 65 years old is 1 point.
> 65 years old is 2 points.
Risk Factors
If no risk factors the patient gets 0 points.
1 – 2 risk factors the patient gets 1 point.
Greater than 2 risk factors the patient gets 2 points.
Risk factors include diabetes that has been treated, hypertension, being a smoker, increased cholesterol, family history and obesity.
Troponin I
If less than or equal to 0.04 the patient gets 0 points.
If 0.04 – 0.08 the patient gets 1 point.
If greater than 0.08 the patient gets 2 points.
- The troponin is measured at admission to the Emergency Department with no further troponins being done.
A few things to say about the study.
- This is a retrospective study and needs validation (more on this in a minute).
- It is a simple scoring system which is easy to use and easy to remember. This has been a major issue with some of the other systems.
- The system has been designed for use in the Emergency Department which is one of its great benefits and which distinguishes it from some of the other scoring systems.
- There is a small flaw in the scoring however. If for example we have a 42 year old male with non-specific chest pain, a normal ECG, and no risk factors, but a troponin of 1.0, the patients score is still a 2 out of 10. But this is clearly now a high risk patient given that they have a high troponin. So we need to put a caveat on the scoring system and say that a scoring system of less than or equal to 3 will have a risk of 2.5% of having an adverse cardiac event, however if the troponin is raised that patient moves into a higher risk category.
- In this study the proposal that the score of less than or equal to 3 having a 2.5% chance of an adverse cardiac event may seem to be unacceptable to some people. In this study the recommendation was that if the HEART score was less than or equal to 3 that patient didn’t require much more of a work up, however a 2.5% risk in my book is a little bit high.
VALIDATION OF THIS STUDY
Since the initial study the authors have validated this study within the same population. A study published by Backus BE., Six A.J. et al., (The Critical Pathways in Cardiology, Volume 9, Number 3, September 2010, pages 164 – 169), validated the score and found that a HEART score of less than or equal to 3 was associated with a risk of 0.99%. This validation was again a retrospective study.
A Prospect of Validation
Since the retrospective validation Backus et al have prospectively validated the HEART score and published it in the International Journal of Cardiology 2013, October 3: 168 (3) 2153-8. In this study a prospectively validated study of the HEART score, they found that a HEART score of less than or equal to 3 excluded short term cardiac events. The risk of a cardiac event when the HEART score was less than or equal to 3 was 1.7%.
1.7% becomes a more acceptable figure, however we can improve this even further?
Work being done at the moment is showing that if we include a troponin initially on presentation and then a troponin at 3 hours the risk drops to below 1%.
The troponins used in all of these studies are NOT the high sensitivity troponins that are now being used in many Australian Emergency Departments. They are the non-high sensitivity Troponin I’s that have been used in the past and certainly being used in overseas departments in the USA and the UK.
In conclusion I think this is a scoring system that is of great use to us, because it applies to the Emergency Department and because it was developed for the Emergency Department. It has been validated in more than one population group and we are seeing improvements in the diagnostic accuracy. It is a simple scoring system to remember and apply and it can reduce the risk of missing an AMI to very very low percentages. More work needs to be done however risk levels of less than 1% may give us a scoring system we can start using now.
Would you use this?
