Are Computer ECG Reads Better Than Emergency Physicians?
February 4, 2022
Written by Clay Smith
This study suggests that a computer-generated “normal” ECG result obviates the need for immediate review of the tracing by the emergency physician. I’m not so sure about this…
Why does this matter?
If the computer-generated ECG interpretation is “normal,” does the ECG need emergent review by an emergency physician? I know Amal Mattu is getting chest pain already. Let’s see what this study found.
If I had a nickel for every ECG I read on shift…
This was a retrospective review of 989 adult patients at a single center with “normal” computer generated ECG interpretations. Of these, final cardiology interpretations had clinically significant differences in 6.1% (60/989). Just over one third of these were admitted. Clinically significant ECG changes included: ST/T wave changes (75%), possible ischemia (17%), T wave inversions (10%), and prolonged QT (5%). But none of these met the prespecified primary outcome of requiring emergency cardiac catheterization within 4 hours of ED arrival. The authors argue that immediate emergency physician review of an ECG read as “normal” by the computer is unnecessary and can reduce distraction. Anyone who has ever seen Amal Mattu lecture is about to type feisty comments on this blog post…but wait. Dr. Mattu and others, like Dr. Stephen Smith often pose multiple scenarios in which the computer misses STEMI equivalents (or major indicators of badness) such as hyperacute T-waves, de Winter T-waves, Wellen’s warning (not a STEMI equivalent but bad), new LBBB or ECGs meeting modified Sgarbossa criteria, posterior MI, left main coronary occlusion, or simply subtle STEMI ECG changes – especially STEMI in the limb leads. I cover an article in my LLSA 2013 Quick Study Guide on STEMI and STEMI equivalents.
Anyway…I know what this study showed. And I appreciate efforts to reduce distractions. Truly, these subtle STEMI findings and STEMI equivalents are rare. However, it just seems like a bad idea to trust the computer to rule out STEMI for us. Just my opinion, but I think we’d do better to work on the EHR to reduce pop-ups and distractions instead of the ECG.
You will want to take a look at this beauty the computer completely missed on Dr. Smith’s ECG blog. In fact, why not take a look at these 30 examples Dr. Smith has covered over the years!
Emergent cardiac outcomes in patients with normal electrocardiograms in the emergency department. Am J Emerg Med. 2022 Jan;51:384-387. doi: 10.1016/j.ajem.2021.11.023. Epub 2021 Nov 17.