Medical Malpractice Insights: So you think you’re smarter than the EKG computer…

Author: Chuck Pilcher, MD FACEP (Editor, Med Mal Insights) // Editors: Alex Koyfman, MD (@EMHighAK) and Brit Long, MD (@long_brit)

Welcome to this month’s case from Medical Malpractice Insights – Learning from Lawsuits, a monthly email newsletter for ED physicians. The goal of MMI-LFL is to improve patient safety, educate physicians, and reduce the cost and stress of medical malpractice lawsuits. To opt in to the free subscriber list, click here.

Chuck Pilcher, MD, FACEP

Editor, Med Mal Insights


You vs. Watson: Who wins?

So you think you’re smarter than the EKG computer. Young patients can have an MI too.

 

Facts: A successful 34-year-old male with no prior heart history experiences chest pain for a few days. He presents to the ED with crushing chest pain radiating down his L arm and cold sweats. He is evaluated immediately. His EKG and blood work are interpreted as normal by the ED physician, but the EKG computer reports ischemic changes. The patient is admitted to an internist and discharged 2 days later with a diagnosis of anxiety and stress, having had neither a cardiology consult, stress test, nor angiogram. He is, however, told to follow up with a cardiologist of his choice. He does as told, in fact seeing the cardiologist 3 times over the next 3 months, each time complaining of ongoing chest pain. He is given a prescription for NTG; it helps. The cardiologist never accesses the hospital record or looks at the original EKG, relying only on his own evaluation. One morning the patient’s pain suddenly worsens. His wife calls 911, and he returns to the original hospital where he is diagnosed with a massive MI. He survives, but with significant myocardiopathy which leads to renal failure, making him ineligible for a recommended heart transplant. He consults an attorney, and a lawsuit is filed against the original ED physician and others.

Plaintiff: The ED physician misread my EKG on my very first visit. Everyone agrees it showed significant ischemia, including not only my three experts but your own defense expert. If you would have recognized the problem, I would have had a coronary angiogram during my first admission. I would then have had the triple bypass surgery or stents that I needed. Just because I’m 34 doesn’t mean I can’t have a heart attack. Now I can barely walk, much less work, and I need a heart transplant that I can’t get because my kidneys have failed.

Defense: The ED doc continued to insist that the EKG really was normal. “I always disregard the computer reading because I have more experience than the EKG computer.” The internist said “You saw a cardiologist like I told you to. He did nothing. It’s not my fault.” The cardiologist said “Your EKG was normal when I saw you. You didn’t need an angiogram because you had no signs of ischemia.”

Result: $6 million pre-trial settlement.

Takeaways:

  • While EKG computer readings have limited accuracy, one ignores them at one’s peril, especially when they report an abnormality. They can be wrong, but so can we. We can’t assume we’re smarter than that computer. If we do, we must clearly document our reason for disagreement. (1)
  • Also keep in mind that the EKG computer interpretation may miss subtle EKG findings. If you are concerned with “nonsignificant” EKG changes, obtain repeat EKG and maintain your suspicion of ACS. See this post from Dr.  Smith’s ECG blog on subtle EKG changes.
  • In chest pain patients with equivocal findings, such as HEART score of 4 or greater, further testing for risk stratification is recommended.
  • When doing follow-up on a recently hospitalized patient, obtain and read the medical record.
  • And finally, there is no lower age limit below which a patient cannot have an MI (or PE or aortic dissection). (4)

 

References:

  1. Romeo Fairley has a nice post on emDocs February 3, 2016 http://www.emdocs.net/ekg-analysis-where-do-we-go-wrong-and-how-can-we-improve/ in which he says “Computer automated analysis has been shown to be inaccurate 6% to 42% of the time.”
  2. http://www.emdocs.net/clinical-pathway-use/
  3. http://www.emdocs.net/r-e-b-e-l-em-management-and-disposition-of-low-risk-chest-pain/
  4. Bezold LI. Myocardial Infarction in Children. Medscape eMedicine, May 22, 2014. https://emedicine.medscape.com/article/897453-overview.

 

Failure is the key to success; each mistake teaches us something.

Morihei Ueshiba

Leave a Reply

Your email address will not be published. Required fields are marked *