Session S31.4

Effectiveness of ECG Interpretation Programs in the Ambulance Setting

E Clark*, M Sejersten, P Clemmensen, PW Macfarlane

University of Glasgow
Glasgow, UK

Aim: This study was aimed at assessing the effectiveness of automated ECG reporting in the setting of patients with acute chest pain being transported to one of several hospitals by ambulance in the Zealand region of Denmark.
Methods: Prehospital 12 lead ECGs were recorded using a LIFEPAK 12 (Physio-Control) and then transmitted digitally to the attending cardiologist. If a diagnosis of ST Elevation Myocardial Infarction (STEMI) were confirmed, the patient was taken directly to an interventional centre while other patients were taken to a local hospital. 200 randomly selected ECGs out of a total of 1000, along with associated patient data collected by the University of Copenhagen, were made available in digital form to the University of Glasgow for a pilot study. 22 ECGs were excluded for various reasons, e.g. the patient had a pacemaker. The data for the other 178 patients (114 male, 64 female, age 64.7±13.6 years) were then studied.
Results: The ECG report from the LIFEPAK 12 was compared with the hospital discharge diagnosis. The sensitivity for a report of STEMI was 73% and specificity was 92%. The ECGs were re-analysed using the Glasgow ECG analysis program which uses age and sex based criteria for STEMI. Corresponding results were 78% and 94% (p>0.05). 63/178 patients were sent directly to an interventional centre by the attending cardiologist. Of these 63, the LIFEPAK 12 reported 49 as having an acute MI of whom 45 were given a discharge diagnosis of STEMI. Of the remaining 14/63, 9 were later diagnosed as having a STEMI. Corresponding figures for the Glasgow program were 51/63, of whom 48 were given a discharge diagnosis of STEMI, while 6/12 remaining patients were subsequently diagnosed as having a STEMI. There were 9 patients who were not sent to an interventional centre by the cardiologist despite the ECG report showing a diagnosis of acute MI. Four were subsequently diagnosed as having STEMI and transferred to an interventional centre for further treatment.
Conclusion: In this investigation, the ECG report of STEMI was highly specific and relatively sensitive. Further work on extending diagnostic criteria may improve accuracy and enhance efficiency of this referral method.

(Abstract Control Number: 16)