5 Audit of stress cardiac magnetic resonance referrals in northern ireland

Introduction Current guidelines regarding investigation and treatment of stable ischaemic heart disease are conflicting. The recently updated NICE guidelines advise CTCA for investigation of chest pain and reserve non-invasive ischaemia tests for patients with known coronary artery disease (figure 1). The ESC guidelines advise non-invasive ischaemia tests for patients with a pre-test probability of 15 – 85% using Diamond Forrester score. In practice, choice of investigation often depends on local availability and expertise and options for non-invasive ischaemia testing include DSE, MPI and CMR. This audit aimed to assess safety outcomes and sensitivity/specificity of stress CMR in Northern Ireland. Methodology A retrospective audit of patients undergoing CMR from 1st Jan 2016–1st January 2017 was undertaken. Baseline demographics, risk factors and CMR outcome (positive or negative for ischaemia) were recorded. For patients with a positive CMR, findings at angiography were assessed. CMR positive patients were defined as new inducible ischemia on stress CMR imaging. Patients with matched perfusion defects due to prior infarction were not included in analysis. For patients with a negative study, the electronic care record was reviewed to assess incidence of adverse outcomes including acute coronary syndrome (ACS) and mortality. Statistical analysis was performed using Chi square test for parametric categorical variables and Mann Whitney U test for parametric continuous variables, p Results In total, 105 patients were referred for stress CMR, 13 patients were excluded due to patient factors (cancellation, claustrophobia, body habitus) and a further 16 were excluded due to adenosine contraindications resulting in 74 patients for analysis. Of these, 22 patients were referred without previous angiogram or known coronary artery disease (11 due to angina, 8 due to impaired left ventricular systolic function demonstrated on Echocardiogram, 3 due to arrhythmias and elevated troponins). From that population, 19 patients had negative CMR and 3 patients had positive CMR with coronary artery disease confirmed on post CMR angiogram. In total, 54 (72.9%) patients were male. Mean age was 64.8 yrs (SD 9.6 yrs), 45 (60.8%) were hypertensive, 23 (31.0%) were diabetic, 36 (48.6%) had hypercholesterolaemia. Follow-up was undertaken from 14–26 months. Twenty nine (39.2%) had a positive CMR. Table 1 compares demographics of stress CMR positive and negative patients. Of the 45 stress CMR negative patients, no patient had myocardial infarction or mortality during follow-up. One patient had ongoing chest pain with subsequent angiography showing LAD disease requiring PCI. Of the 29 CMR positive patients, 20 (68.9%) had invasive angiography, of which 19 (95.0%) had PCI and 1 (5.0%) had bypass surgery. The remaining 9 patients were felt inappropriate for angiography due to comorbidities. There were no mortalities in the stress CMR positive group. Sensitivity was 0.96 for coronary artery disease according to post CMR angiogram. Conclusion Stress CMR correctly identified patients with obstructive coronary disease which required further therapy with a high degree of sensitivity. Patients with negative studies appear to have a low risk of cardiac events in the short to medium term.