Review of:Trivedi K, Schuur JD, Cone DC: "Can paramedics read ST-segment elevation myocardial infarction on prehospital 12-lead electrocardiograms?" Prehospital Emergency Care. 13(2):207-214, 2009.
This study used a convenience sample survey of paramedics from a large career system. The authors provided the medics with scenarios of men and women with and without cardiac risk factors, as well as ECGs with and without ST elevation. All ECGs were computer generated. At the time of the study, the system had not implemented a cath lab alert program but did allow medics to "alert" the hospital of a possible ST-segment elevation myocardial infarction (STEMI), and the emergency department (ED) would wait until after patient arrival to activate the cath lab.
The authors surveyed 70% of their medics who had an average of 8.7 years of experience and estimated that it took the medics three minutes to obtain a 12-lead ECG. The medics reported their confidence in interpreting a 12-lead for STEMI at 3.7 on the 5-point Likert scale. Similarly, they reported a 4.0 of placing the ED on "chest pain alert" and 3.7 for "activating the cath lab."
Upon review of their ECG interpretation, the authors determined that the paramedics accurately detected the presence of a STEMI 85% of the time (specificity). Of the STEMI cases, 94% resulted in a simulated cath lab activation. Of the non-STEMI cases, 14.9% resulted in simulated cath lab activation.
The authors concluded their paramedics could accurately interpret prehospital 12-lead ECGs.
Although I agree with the premise of this paper, I disagree with the title and basic conclusion. A more accurate title would be: "Can paramedics detect signs of STEMI in standardized computer generated 12-lead ECGs in a scenario-based setting?"
The differences are significant. Can paramedics interpret 12-lead ECGs? Certainly. So can EMT-Bs, if provided sufficient education. That's why we have textbooks and conferences. The question is whether medics make the call in the field when the patients present with vague symptoms and ECGs that are less dramatic. The ECGs used in this study had more than 4 mm of ST elevation. It would be hard to miss this.
The next obstacle is activating the cath lab. In some systems, the criteria is 1 mm of ST elevation in the inferior leads and 2 mm in the precordial leads. In other systems, the threshold is only 1 mm of ST elevation in the precordial leads. In many systems, the medic must rely on machine interpretation, confirm that interpretation and communicate that to the receiving hospital to activate the cath lab.
In all fairness, the authors recognize the primary limitation of the study is that it was scenario-based. They have moved forward to examine the accuracy of the medics under a clinical perspective.
The issue of over- and under-triage is important. Although I applaud the authors on achieving an 8% over-triage based on this scenario-based study, the reality is that this comes at the cost of a significant under-triage of 9% of very obvious STEMI cases. I would speculate that this will result in an even greater percentage in actual practice unless the authors consider using a combination of computer interpretation and on-line medical control consultation.
Under-triaged patients are missed opportunities to save time and myocardium.