HEART score

Written by: Abiye Ibiebele, MD (PGY-3) Edited by: Kumar Gandhi, MD (PGY-4) Expert commentary by: D. Mark Courtney, MD, MCSI

Chest pain accounts for nearly six million annual visits to the Emergency Departments across the United States and accounts for over ten billion in healthcare dollars, and thus the appropriate management of chest pain is part of the daily reality of Emergency Physicians. From one’s first EM rotation in medical school, one learns to always rule out the dangerous causes of chest pain first, of which acute coronary syndrome (ACS) is often first and foremost. A STEMI is called from the field or quickly on arrival to the emergency department and the patient is quickly evaluated by cardiology either in the emergency department or on the way to the cardiac catheterization lab. However, the diagnostic and management challenge remains for those patients that may have ACS without initial EKG changes or an elevation in troponin. To help us, clinical decision rules have been developed to help us identify which patients who present with chest pain we can safely rule out life-threatening ACS in and help risk stratify patients into different risk categories [1]. What follows is a breakdown of the components of the HEART score, a review of how to appropriately assign points and a brief review of the original study and subsequent research since then.

History

This is the most subjective area of scoring in the HEART score and one of possible contention between different health care providers.

The original study broke down historical elements as specific for ACS and nonspecific for ACS as judged by the clinical experience of practiced providers.1 A 0 score, was given for a completely nonspecific history and a 2 score was given for a primarily specific history. For a mixture of nonspecific and specific elements, a 1 score was given.

The original researchers used clinical gestalt and took into account historical elements such as pattern of pain, onset, duration, relation to exercise, localization, concomitant symptoms and reaction to sublingual nitrates. While this was based on clinical judgment, the historical elements were somewhat based on a prior clinical review which listed specific elements as follows: [3]

Concerning history (read: specific for ACS)

Chest pain radiating to one or both arms
Pressure like pain with associated nausea, vomiting, or diaphoresis
Exertional chest pain
Response of chest pain to nitroglycerin
Chest pain similar to prior MI

Non-concerning history for ACS (read: nonspecific for ACS)

Pleuritic or positional chest pain
Chest pain reproducible with palpation
Stabbing quality of pain
Pain localized to an area on chest smaller than a coin

It is important to note is that for assigning a history score, developers did not take into account risk factors or EKG findings. These are accounted for elsewhere in the HEART score.

EKG

Two points are assigned for ST elevations or depressions, in the absence of a bundle branch block, LVH or use of digoxin [1].

One point is assigned for repolarization abnormalities (new or old) without ST depression. A person can also receive a score of 1 for a bundle branch block or left ventricular hypertrophy [1].

Zero points is assigned for a normal EKG [1].

Age

This component of the HEART score is the most straightforward with scoring as defined in the chart above.

Risk Factors

As in the above chart, having no risk factors results in a score of zero points. Having 1-2 risk factors yields a score of 1. Important thing to note is about having at least 3 risk factors OR a “history of atherosclerotic disease” results in a score of 2 points. [1]

What does “history of atherosclerotic disease” mean? [1]

History of revascularization (PCI or CABG)
History of myocardial infarction
History of ischemic stroke
History of peripheral arterial disease

Thus, a patient with a history of any of the above disease should automatically get a 2 for this section of the HEART score.

What were included as risk factors for study purposes? [1,2]

Hyperlipidemia
Hypertension
Diabetes Mellitus
Cigarette smoking (has to have last smoked within 90 days)
Family history of coronary artery disease (doesn’t matter if family member was over/under 50 years of age)
Obesity (defined as BMI over 30)

Troponin

Also, a straightforward component of the HEART score with scoring as above.

Some notes regarding scoring:

Original study and validation studies did not use high sensitivity troponin when calculating the HEART score [1,4]
Follow up studies have used high sensitivity troponins in what is referred to as a “modified HEART score”. Scoring is similar to conventional troponin testing as in above chart. [5,6]

As those of you who have use the HEART score know, you calculate all the points and if a patient has a score between 0-3, they are considered low risk can be discharged home safely. A score between 4-6 is considered moderate risk and should be admitted for further observation and workup. A score of 7-10 is considered high risk and is recommended to have an early invasive intervention.

Now that we have covered the components of the HEART score, as previously mentioned, below is a brief review of the original study. However, before that, two caveats to using the HEART score:

For original study and following validation studies, patients who presented with only dyspnea or palpitations without associated chest pain were excluded [1,4]
HEART score has been shown to be helpful in distinguishing risk even when looking within special populations (diabetics, elderly, and females) [1,2,4,5]

Original Study

Six AJ, Backus BE, Kelder JC. Chest pain in the emergency room: value of the HEART score. Neth Heart J. 2008;16(6):191-196.

Study design

Retrospective, single center study at a 265 bed community hospital in the Netherlands

Inclusion Criteria

Patients included were any patient admtted to the ER due to chest pain irrespective of age, prehospital assumptoms and previous medical treatments

Patients with STEMI were excluded

Methods

Patient’s charts were reviewed and HEART score was calculated as above. Endpoints as described below were identified and differences between groups were statistically analyzed.

Endpoints

Acute MI, revascularization, death and composite endpoint of all three.

Demographics

Total of 122 patients, mean age 61 years old, 60% male, race not specifically measured but overall hospital population >95% White/ Caucassion.

Results

All patients:

24% of all patients reached one or more of above endpoints

Average HEART score for all patients that did not reach an endpoint: 3.71 +/- 1.83

Average HEART score for all patients that did meet an endpoint: 6.51 +/- 1.84

Significant difference p < 0.0001

Heart Score Groups:

For patients with HEART score 0-3, 2.5% reached an endpoint

For patients with HEART score 4-6, 20.3% reached an endpoint

For patients with HEART score 7-10, 73% reached an endpoint

What to do with low risk patients?

Based on the heart score original study, for low risk patients (HEART score 0-3) the risk of a major adverse cardiac event (MACE) was 2.5%. The original suggestion from the authors of this study is that these patients can be immediately discharged.
Further validation studies have shown the risk of MACE within 45 days to be 1.9% and of all-cause mortality to be 0.05%.
There has also been development of the HEART score pathway, which adds a 3-hour serial troponin to low risk patients. In the initial study, 40 patients were discharged from the ED after two negative troponins and none of those patients had a MACE within 30 days.7
A 2017 systematic review of 9 studies and 11,217 patients revealed that 1.6% of the low risk patients would have a MACE at 6 weeks8. This study derived the sensitivity of the HEART score to be 96.7%. Another systematic review is ongoing to validate these findings and to further delineate the prognostication value of the HEART score9
Ultimately, having a discussion with these patients about what low risk means and coming to a shared decision can be a useful tool but in a patient with good outpatient follow-up, it is reasonable to discharge the patient without further cardiac testing,

Expert Commentary

What clinicians (and patients) care about is post test probability. This is a function of two things, what the pretest probability is and how good your testing tools are. The HEART score, the modified HEART score and the HEART Score Pathway as described above are all ways to try to formalize the process of pretest probability estimation. It is important to realize that the standard ED evaluation is suited to identify STEMI and NSTEMI. However part of the definition of MACE is angina and that is the entire point of stress testing….to discover treatable intermittent cardiac ischemia that may not show up on initial or subsequent ECGs or troponin testing. The HEART score approach has rapidly become the most widely used approach in the US to try to determine who needs a stress test during the index admission vs. who may be able to safely be discharged for follow-up and potential stress testing as an outpatient.

Despite popularity, there are challenges with any HEART score approach. These are as follows. 1) Clinicians like to talk about the HEART score but often don’t document it or do so in a template dot-phrase manner. All decision support pretest probability scores are only as good as the degree of accuracy of the data applied and documented. At times documentation is simply “HEART SCORE low” without the granular details to support this in the record. 2) There is potential for inter-observer disagreement. In work presented in abstract at the 2018 SAEM annual meeting Dr. Steven Ignell reported that residents were less likely to classify patients as low risk than attendings, and that formal application of the HEART score did not result in a larger proportion of low risk classification than by simple gestalt alone (do you think this patient is at 2% risk of 6 week MACE?). However in this work the agreement was better using the HEART score with a weighted Kappa of 0.62 vs 0.29 for gestalt along. So it seems that the HEART score likely identifies a roughly similar cohort of patients but does so with more standardization. The biggest future questions are to what degree in the US a HEART score approach will be optimally integrated with hsTroponin testing, who needs serial hsTroponin testing and who does not, and does timing of pain onset matter? Ongoing and future trials may bring insights into these important process questions.

How to Cite this Post

[Peer-Reviewed, Web Publication] Ibiebele A, Gandhi K. (2019, Dec 9). A Closer Look at the HEART Score. [NUEM Blog. Expert Commentary by Courtney DM]. Retrieved from http://www.nuemblog.com/blog/HEART-score.

References

Six AJ, Backus BE, Kelder JC. Chest pain in the emergency room: value of the HEART score. Neth Heart J. 2008;16(6):191-196.
Backus BE, Six AJ, Kelder JC, et al. Chest pain in the emergency room: a multicenter validation of the HEART Score. Crit Pathw Cardiol. 2010;9(3):164-169.
Swap CJ, Nagurney JT. Value and limitations of chest pain history in the evaluation of patients with suspected acute coronary syndromes. JAMA. 2005;294(20):2623-2629.
Backus BE, Six AJ, Kelder JC, et al. A prospective validation of the HEART score for chest pain patients at the emergency department. Int J Cardiol. 2013;168(3):2153-2158.
Ma CP, Wang X, Wang QS, Liu XL, He XN, Nie SP. A modified HEART risk score in chest pain patients with suspected non-ST-segment elevation acute coronary syndrome. Journal of geriatric cardiology : JGC. 2016;13(1):64-69.
Santi L, Farina G, Gramenzi A, et al. The HEART score with high-sensitive troponin T at presentation: ruling out patients with chest pain in the emergency room. Intern Emerg Med. 2017;12(3):357-364.
Mahler SA, Riley RF, Hiestand BC, et al. The HEART Pathway randomized trial: identifying emergency department patients with acute chest pain for early discharge. Circ Cardiovasc Qual Outcomes. 2015;8(2):195-203.
Van Den Berg P, Body R. The HEART score for early rule out of acute coronary syndromes in the emergency department: a systematic review and meta-analysis. European heart journal Acute cardiovascular care. 2018;7(2):111-119.
Byrne C, Toarta C, Backus B, Holt T. The HEART score in predicting major adverse cardiac events in patients presenting to the emergency department with possible acute coronary syndrome: protocol for a systematic review and meta-analysis. Systematic reviews. 2018;7(1):148.