The HEART score is an excellent predictor of major adverse cardiac events in adult ED patients with chest pain, particularly mortality and MI, and should be the primary clinical decision instrument used for the risk stratification of this patient population, according to the authors of a recent review of 30 studies.1

A major concern for EPs evaluating chest pain is acute coronary syndrome (ACS), which carries high morbidity and mortality. “However, even when we rule out ACS, we are sometimes concerned by the patient or the story,” says Shannon Fernando, MD, MSc, the study’s lead author. Sometimes, it is unclear if the ED patient can be safely discharged.

“In the ED, the hardest thing we do is send people home,” says Fernando, a resident in the department of emergency medicine and a fellow in the division of critical care at the University of Ottawa. To alleviate fears of a short-term bad outcome, the easiest thing to do is to admit the patient for further testing or arrange outpatient testing on an urgent basis.

“However, there is increasing evidence that this testing, aside from being costly and resource-intensive, can also result in adverse events for the patients,” Fernando notes.

EDs commonly use the HEART score to risk-stratify patients “but without strong evidence of how it functions,” Fernando says. The study’s findings were reassuring in this regard. Researchers found that the sensitivity of a HEART score above 3 for prediction of future major adverse cardiac events was 95.9%. This compared favorably with the sensitivity of a Thrombolysis in Myocardial Infarction (TIMI) score above the low-risk threshold, which was only 87.8%. “This was not surprising to us,” Fernando reports.

The findings support existing evidence that the HEART score is superior to the TIMI score in predicting future major adverse cardiac events for ED patients with chest pain. “Therefore, if you are going to use a clinical decision instrument for risk-stratification of ED patients with chest pain, you should preferentially be using the HEART score,” Fernando concludes.

Generally, the “acceptable” miss rate for future major adverse cardiac events is recognized as 1% to 2%. If one applies the HEART score indiscriminately among patients with chest pain, roughly 4.1% of patients will be scored as “low-risk” but still go on to experience a major adverse cardiac event, Fernando acknowledges.

“The HEART score has never been compared extensively with clinical gestalt and therefore may actually be inferior to clinician judgment,” he notes. Some evidence suggests that individual clinicians vary in how they score patients. It is also unknown how the HEART score performs in the context of high-sensitivity troponin assays. “Pathways that incorporate the HEART score with a specific troponin assay will likely be more useful for clinicians than simply relying upon the HEART score itself,” Fernando offers. Ultimately, EDs are seeking a better approach than admitting all chest pain patients for observation and testing. The HEART score attempts to reduce the number of patients requiring such testing. “But there are still important shortcomings that clinicians should be aware of before utilizing this tool,” Fernando warns.

A low HEART score helps the ED defense team justify the EP’s decision to discharge a chest pain patient. “It’s not foolproof, and it should not supplant clinical gestalt. But I think juries can understand formulas like this,” says Jesse K. Broocker, JD, a partner at Atlanta-based Weathington McGrew. It is difficult to argue an EP is negligent if the pathway was applied thoughtfully.

There is not much room for debate about most of the HEART score components. The age, troponins, and risk factors are difficult to argue over since they are fairly objective findings. “Experts can haggle over an ECG read,” Broocker notes. “Getting a cardiology overread is always the safe play when in doubt on nonspecific findings.” The presenting complaint and patient history, factors in the HEART score, are much more open to interpretation. “This is where plaintiffs can make hay because it is so subjective,” Broocker explains.

An ED chart noting the absence of “classic” signs of sweating and shortness of breath as well as describing pain that does not radiate, is reproducible on palpation, and is not related to exertion paints a picture of a history that was not suspicious for ACS. “But plaintiffs will look for those one to two things that can be associated with cardiac ischemia,” Broocker says. This opens the door to argue that the EP’s job is to consider the worst possibilities on the differential. Laying out a full history with great detail can refute this. “Take care not to haphazardly click through the EMR,” Broocker adds. “Plaintiff lawyers use clerical mistakes as evidence that the doc was not paying attention.”


  1. Fernando SM, Tran A, Cheng W, et al. Prognostic accuracy of the HEART score for prediction of major adverse cardiac events in patients presenting with chest pain: A systematic review and meta-analysis. Acad Emerg Med 2018; Oct 29. doi: 10.1111/acem.13649. [Epub ahead of print].