Originally published at R.E.B.E.L. EM on November 1, 2014. Reposted with permission.

Follow Dr. Salim R. Rezaie on twitter @srrezaie

Every year there are 6 million visits to the Emergency Department (ED) for chest pain, and approximately 2 million hospital admissions each year.  This is approximately about 10% of ED visits and 25% of hospital admissions with 85% of these admissions receiving a diagnosis of a non-ischemic etiology to their chest pain (CP).  This over triage has enormous economic implications for the US health care system estimated at $8 billion in annual costs.

Why do we do this?

Well, it could be that the single greatest contributor to financial losses in malpractice claims against emergency physicians comes from failure to accurately diagnose acute myocardial infarction (AMI).

So the question is:

Are there specific aspects of the history that can increase or decrease the likelihood that a patient has acute coronary syndrome (ACS) and/or AMI?

There are 5 studies that were recommended by Dr. Amal Mattu, on his EMCast Podcast (July 2012) that evaluated the components of history that were more likely to correlate with ACS and/or AMI. Each will be reviewed below.

 (Edwards et al. 2011)*

The main objective was to see if there was any correlation between severity of CP and the risk of AMI at presentation, or composite end points (death, revascularization, or acute myocardial infarction) at 30 days. Severe chest pain was defined as 9 – 10 on a pain scale of 0 to 10.

• Risk of AMI with Pain Score of 1 – 8 (82% of patients) = 3.0%
• Risk of AMI with Pain Score of 9 – 10 (18% of patients) = 3.9%
• Not statistically significant different
• Bottom Line:  Severity of pain is not related to likelihood of AMI at presentation, or composite end points (death, revascularization, or AMI) at 30 days.

 (Body et al. 2009)*

The objective was to assessing the value of individual symptoms for predicting a diagnosis of AMI or the occurrence of adverse events (death, AMI, revascularization via PCI or CABG) within 6 months.

• Strongest positive predictor of AMI
  • Diaphoresis with CP
• Other positive predictors of AMI and adverse events
  • Nausea and vomiting with CP
  • CP with radiation to both shoulders > right shoulder > left shoulder
  • Central chest pain
• Strongest negative predictor of AMI
  • Pain located in the left anterior chest
• Other negative predictors of AMI and adverse events
  • CP described as pain being the same as previous AMI
  • Presence of CP at rest
• Bottom Line:  Many “atypical” symptoms are more likely to render the diagnosis of ACS than traditional “typical” symptoms

(Swap and Nagurney 2005)*

The authors wanted to identify the elements of a CP history that might be most helpful to the clinician in identifying ACS. They performed a literature search from 1970 to 2005.

• Bottom Line: No characteristics of chest pain alone, or in combination, identify a group of patients that can be safely discharge home without further diagnostic testing. Also beware the chest pain which radiates to the RIGHT shoulder (LR = 4.7).

(Goodacre et al. 2002)*

In this prospective, observation cohort study of 893 patients, the authors assessed the performance of clinical features used in the diagnosis of CP, specifically in patients who were clinically stable and had a non-diagnostic EKG.

•  Predictive of ACS/AMI:
  • Exertional pain
  • Pain radiating to both arms > right arm
•  NOT predictive of ACS/AMI:
  • Presence of chest wall tenderness
  • Nausea or vomiting
  • Diaphoresis
• Bottom Line:  Clinical features have a limited role in triage decision-making for ACS/AMI.

(Panju et al. 1998)*

•  Bottom Line:  History alone can help, but can NOT rule out ACS/AMI!

SUMMARY

Bibliography