Background Primary percutaneous coronary intervention (PPCI) is becoming the primary reperfusion strategy in ST-segment elevation myocardial infarction (STEMI) in the UK. However, activation of the PPCI service is still based on the traditional 12 lead electrocardiogram (ECG) criteria for thrombolysis. Few data exist regarding the diagnostic accuracy of these criteria for patients being considered for PPCI. Our policy is to activate the service in patients who not only have these ECG changes but also in those with a suggestive ECG (eg, ST elevation that does not strictly fulfil these criteria) and a good history for acute myocardial infarction (AMI). We sought to determine the diagnostic accuracy of conventional ECG criteria for a final diagnosis of AMI as well as total coronary artery occlusion for patients that activate our service.

Methods Retrospective analysis on prospectively collected data on consecutive patients between November 2008 and August 2009 who activated our PPCI service. ECGs are sent via telemetry to the PPCI co-ordinator who is able to discuss the history by phone with the attending emergency medical team and then determines whether or not to activate the service (doubtful cases can be discussed with the duty interventional cardiologist).They are then stored separate from case notes for 1 year. All ECGs were reviewed by two experienced cardiology registrars to assess whether they met the criteria for thrombolysis. They were blinded to clinical, procedural details and outcomes.

Results Five hundred and fifty-eight patients activated the PPCI pathway. Of these, 521 (93%) had coronary angiography and 457 (82%) PCI. Twenty-nine patients were excluded (missing ECGs). Mean age of the remaining 529 patients was 64.5±13.0 years; 32% were female. We identified 431 (82%) patients that met the thrombolysis criteria for STEMI (Group 1) and 98(18%) that did not (Group 2). There was a higher incidence of cardiogenic shock in Group 1. Patients in Group 2 were more likely to have had previous MI and anterior territory ECG changes (Abstract 40 Table 1).

The diagnostic accuracy of conventional STEMI criteria (Group 1) is shown in Abstract 40 Table 2. For diagnosing AMI and predicting the presence of a total coronary artery occlusion (CAO), they are sensitive but not very specific. However, 76 of those not meeting the conventional criteria (78%) had an AMI and 32 (33%) had a CAO.

Conclusion Approximately one in seven patients who are having an AMI, roughly half of whom have a CAO, do not strictly meet the criteria for reperfusion therapy based on the original thrombolysis trials. With the planned widespread implementation of PPCI in the UK, our data support the need for further larger studies to determine whether the ECG criteria for activation of the STEMI pathway should be revised.