Abstract 5745: Resting Regional Left Ventricular Function by 64-slice Cardiac CT adds Incremental Value to Coronary Plaque and Stenosis in Predicting Acute Coronary Syndrome in Patients with Acute Chest Pain
Background: Cardiac computed tomographic assessment of coronary atherosclerotic plaque and stenosis may improve risk stratification in patients with acute chest pain. Our goal was to assess the diagnostic accuracy of CT derived resting regional LV dysfunction and its incremental value over CAD to predict ACS in patients with acute chest pain but normal initial biomarkers and non-ischemic ECG.
Methods: We included consecutive subjects without a known history of CABG, or coronary stents who presented to the emergency department with acute chest pain but had negative initial troponin and non-ischemic ECG changes. All subjects underwent standard contrast enhanced 64-slice cardiac CT within 3 hours of presentation. Images were reconstructed at 10% intervals throughout the cardiac cycle and analysed using short and long axis cine images of the LV cavity created with 4-dimensional viewing software. Images were assessed qualitatively for regional LV function using the AHA/ACC 17-segment model by 2 experienced readers, blinded to the clinical course and coronary plaque/stenosis findings.
Results: Three-hundred and forty seven consecutive patients (mean age 52.7 ± 12 years, 62% male) were included. Thirty of these patients were ultimately diagnosed with ACS (MI 8, UAP 22) during their index hospitalization. Overall, 44 patients had impaired regional LV function, 21 had ACS. Sensitivity, specificity positive and negative predictive values for ACS were 70%, 92.7%, 47.7% and 97% respectively. In the logistic regression analysis, LV function was not only an independent predictor for ACS but also was associated with over 20 (RR 20.88, 95%-CI 7.81–55.79; p<0.0001) and 16 (RR 16.42, 95-CI: 6.13– 43.95; p<0.0001) fold increase in risk over and above the extent of plaque and stenosis; respectively. Moreover the areas under the ROC curves for the prediction of ACS also improved significantly with the addition of LV dysfunction (AUC from 0.87 to 0.93 for CAD and 0.82 to 0.89 for stenosis P<0.001).
Conclusion: Regional LV dysfunction is significantly associated with and predicts ACS independent of plaque and stenosis in patients with acute chest pain and thus its inclusion may improve the diagnostic accuracy of cardiac CT in this setting.