Background: Lets face it. All of us have been interrupted by the onslaught of triage ECGs for interpretation. This constant flow of pink paper with black scribble causes frequent task switching, interrupts train of thought, and ultimately can lead to medical errors, which affects the patients in front of us. On the other hand, it is important to avoid delays in care and, in accordance with the American Heart Association guidelines, ECGs in triage should be obtained and interpreted by an attending emergency physician within 10 minutes of arrival to the emergency department for any patients with concerns of acute coronary syndrome. Is there a way to maybe minimize the number of interruptions?
Triage ECGs: Reducing Interruptions in a Busy ED
What They Did:
- Single center, prospective cohort study of all ECGs performed at triage in the ED
- Comparison of computer interpretation of “normal ECG” vs attending cardiologist’s final interpretation
- Each ECG interpreted as “normal” by the ECG machine were compared to cardiologist’s final interpretation
- If cardiologist interpretation differed from “normal,” the ECG was then interpreted by two board certified EM physicians that were blinded to patient presentation, patient care, and the goals of the study. EPs were asked to evaluate the ECG for clinical significance
- Clinical Significance: Any change from “normal ECG” computer read that would alter triage care
- All patients ≥18 years of age with a triage ECG
- <18 years of age
- 885 triage ECGs collected over 16 weeks
- 222 (26%) interpreted as “normal ECG” by computer
- 5 (0.57%) interpreted as STEMI
- NPV: 99% (95% CI 97 – 99%)
- 13/222 (5.9%) had different interpretation by attending cardiologist
- 1/13 (7.7%) found to have clinical significance (EP determined patient should be placed in bed immediately)
- Patient was ultimately discharged home for a next day stress test which was interpreted as normal
- 1/222 (0.45%) “normal ECGs” changed triage decision making
- Asks a real-life physician oriented question about interruptions in the ED
- Board certified cardiologists and EM physicians interpreting ECGs were blinded to patient presentations, patient care, and goals of the study
- ECGs not only read as abnormal, but clinical relevance of abnormality assessed
- Single Center Study
- This was a small sample size. This study did not meet the 379 “normal” ECGs necessary to obtain a NPV of 100% with a lower limit 95% CI of >99%
- The incidence of STEMI in this study was low (0.57%)
- Original interpreting physicians were not blinded to computer interpretation of ECGs
- The ECG software used may not be the same ECG software used at other institutions
- One advantage that physicians have over ECG interpretation is that we are able to get a history and do a physical exam.
- This study is not saying that physicians should not review all ECGs, but instead a delayed strategy in “normal” ECGs may be feasible
Author Conclusion: “Our data suggest that triage ECGs identified by the computer as normal are unlikely to have clinical significance that would change triage care. Eliminating physician review of triage ECGs with a computer interpretation of normal may be a safe way to improve patient care by decreasing physician interruptions.”
Clinical Take Home Point: Although this is a small, single center study, who’s results need to be replicated, this seems like a very feasible intervention to try and reduce one of the most common interruptions encountered by EM physicians at the work place.
- Hughes KE et al. Safety of Computer Interpretation of Normal Triage Electrocardiograms. Acad Emerg Med 2017; 24(1): 120 – 24. PMID: 27519772
For More Thoughts on This Topic Checkout:
- Ryan Radecki at EMLit of Note: Can We Trust Our Computer ECG Overlords?
- Stephen W. Smith at Dr. Smith’s ECG Blog: How Unreliable are Computer Algorithms in the Diagnosis of STEMI?
- Vince Diguilio at EMS 12-Lead: Triaging Triage ECGs
Post Peer Reviewed By: Anand Swaminathan (Twitter: @EMSwami)
Latest posts by Salim Rezaie (see all)
- Episode 38 – Do All Submassive PE’s Require Treatment with Thrombolysis? - May 15, 2017
- More on the Easy IJ - May 4, 2017
- Episode 37 – Definitions and Identification of Sepsis: Sepsis 2.0 vs Sepsis 3.0 - May 1, 2017