Automated STEMI Diagnosis Gives Rate of Cath-Lab Activation

TORONTO — Using an automated ECG reading to diagnose patients with ST-segment-elevation MI (STEMI) did not increase the rate of inappropriate mobilization of the cardiac catheterization lab team in a new study [1].

Speaking with heartwire after presenting here at the Canadian Cardiovascular Congress 2012, Dr Brian Potter (Centre Hospitalier de l'Université de Montréal, QC) said, "With this system, we've put the diagnosis and activation decision right at the point of first medical contact."

The study showed that "diagnosing STEMI up front is highly effective in achieving the door-to-balloon target [and] we do not see an excess of false positives or inappropriate activation . . . when we take the physician diagnosis out of the equation," he added.

In patients with STEMI, the door-to-balloon time frequently exceeds the recommended delay of <90 minutes, Potter said. Although several strategies have been proposed to speed up the process, inappropriate mobilization of the cardiac cath team remains "the elephant in the room."

According to Potter, this occurrence has multiple negative implications. It increases healthcare costs. Patients can incur harm from a needless invasive procedure. It also entails intangible costs such as "STEMI fatigue," where healthcare providers mistrust referrals from emergency responders.

Researchers aimed to determine the rate of false positives and inappropriate cardiac cath-lab activation in a new system in which ECGs were performed by first responders using an automated ECG device (made by Zoll Medical) that would diagnose the presence or absence of STEMI.

Data were collected from all patients who complained of chest pain or dyspnea and had an automated ECG test done by a first responder between January 2010 and January 2012. Patients lived in an area served by a community hospital in Laval, QC.

Inappropriate activation was defined as cath-lab mobilization that resulted from a nondiagnostic ECG (two independent cardiologists judged that there was no ST elevation). Human error was defined as improper referral to the cath lab by an ambulance technician, and machine error was characterized as an incorrect, automated diagnosis of STEMI from an ECG tracing.

Investigators identified 155 patients who met the study criteria and had complete data. A total of 19 of these patients (12%) had ECGs that led to inappropriate activation of the cath lab--which is comparable with the 9% to 12% rate reported in the literature.

A total of 120 patients (83%) had a final diagnosis of STEMI, and eight patients (5%) were diagnosed as having false-positive ECGs.

Of the 19 patients inappropriately referred to the cardiac cath lab, human error was implicated in nine cases; in six of these cases, the quality of the ECG tracing was poor. Machine error led to improper referral in 10 cases; five of these patients had a heart rate >140 bpm.

These findings suggest that making simple changes to triage criteria--such as excluding patients with tachycardia and educating first responders to reject poor ECG tracings--can have a big impact on improving quality of care and using resources wisely, Potter said.

"New, Simple System" Shows Promise

Comoderator of the session, Dr Michel Le May (University of Ottawa Heart Institute, ON), commented that this work has "pushed the envelope" and "opens up a new way of dealing with cardiac cath-lab activation."

Speaking with heartwire , comoderator Dr Jean G Diodati (University of Montreal, QC), said: "It's very exciting to see that a relatively simple system like this could work to achieve lower rates of inappropriate activation."

At a time when the healthcare system is being stretched, "I think it is a very important goal to have inappropriate activation as low as possible," he added.

This research suggests that good way to decrease false positives and inappropriate activation would be to send patients with tachycardia to an emergency room (for an ECG read by a physician), rather than taking them directly to a cath lab, he noted.

The investigators declared no conflicts of interest.