Abstract DP203: Cardiac CT Angiography in Transient Ischemic Attack: Diagnostic Yield

Introduction: Transient ischemic attack (TIA) carries a risk of early recurrent ischemic stroke. Echocardiography remains the standard for detecting cardioembolic sources of stroke, but has low yield and limited availability in rural hospitals. Cardiac CT angiography (CCTA) may have increased sensitivity for cardioembolic sources of ischemic stroke, but the added value of CCTA in the diagnostic evaluation of TIA patients is less clear. We evaluated the diagnostic yield of CCTA in the routine evaluation of TIA patients within our hospital system. Methods: We assessed the diagnostic yield of CCTA compared to echocardiography in hospitalized patients presenting with TIA. In addition to routine echocardiography, CCTA was incorporated into the standard inpatient diagnostic evaluation for acute ischemic stroke and TIA in February 2024. Cardiac CT angiogram was either integrated with head/neck CTA using a single contrast bolus or as a stand-alone study. The diagnostic yield was defined as the proportion of imaging studies leading to a change in clinical management. Proportions were compared using McNemar’s mid P-test. Patient demographics, TIA characteristics, and CCTA acquisition type were reported descriptively. Results: We analyzed 35 consecutive TIA patients (median age 76 [IQR: 68-82]; 63% male; median ABCD2 score 5 [IQR: 4-5]; median NIHSS score 0 [IQR: 0-0.5]) who underwent both CCTA and echocardiography. Echocardiography changed clinical management in 1/35 patients (2.9%), whereas CCTA changed management in 4/35 patients (11.4%). The absolute difference in paired proportions was 8.6% (95% CI, -3.6% to 20.8%, p=0.21 by McNemar’s mid-p test). The CCTA findings which changed management were not seen on echocardiography and the echocardiography finding was not found on CCTA. Conclusion: In this real-world pilot study, CCTA changed management more frequently than echocardiography (11.4% vs 2.9%) in patients with TIA, although the difference was not statistically significant. These findings align with previous studies of ischemic stroke, however, larger studies are necessary to further assess the diagnostic yield and impact on clinical outcomes.