Intra-cardiac thrombus detection by electrocardiogram-gated cardiac computed tomography in hyperacute ischemic stroke

Traditional workup with transthoracic echocardiogram (TTE) has limited yield in the diagnosis of suspected cardioembolic stroke. The current study aimed to determine the diagnostic yield of hyperacute electrocardiogram (ECG)-gated cardiac computed tomography (CT) in comparison with TTE in the detection of intra-cardiac thrombus amongst patients with ischemic stroke within an ethnically and socioeconomically diverse population.

A prospective observational analysis was undertaken of patients who had undergone cardiac CT with prospective ECG-gating as part of a hyperacute stroke imaging protocol at a single comprehensive stroke centre in Sydney, Australia between November 2022 and December 2024. Data related to demographics, clinical assessment, treatment and follow-up functional status were collected in addition to high-risk sources of cardio-aortic stroke.

A total of 117 patients underwent imaging with ECG-gated cardiac CT as part of their hyperacute stroke assessment. Of 69 (59%) patients with a final diagnosis of ischemic stroke, intra-cardiac thrombus was identified on cardiac CT in 18 patients (26%). Intra-cardiac thrombus was associated with a higher prevalence of premorbid hypertension (94% vs. 67%; p = 0.03), premorbid heart failure (33% vs. 2%; p < 0.001), premorbid atrial fibrillation or atrial flutter (56% vs. 24%; p = 0.01) and intracranial arterial occlusion (83% vs. 49%; p = 0.01). Amongst patients who underwent both cardiac CT and TTE (n = 49; 71%), intra-cardiac thrombus was more likely to be identified on cardiac CT than TTE (29% vs. 2%; p < 0.001). In an unadjusted analysis, more patients had poor functional outcome (modified Rankin scale 4-6) at 3 months if intra-cardiac thrombus was present on cardiac CT (67% vs. 28%; p = 0.03).

ECG-gated cardiac CT identified intra-cardiac thrombus in 26% of patients who had a final diagnosis of ischemic stroke. The diagnostic yield of intra-cardiac thrombus detection was substantially higher with cardiac CT compared to TTE. Cardiac CT should be considered as part of hyperacute stroke imaging protocols. The unadjusted analysis on functional outcomes is hypothesis generating, limited by imprecision from a small sample size; further investigation requires a larger cohort with multivariable adjustment. Further studies in the hyperacute setting are required to explore benefits in different populations and with alternative imaging acquisition protocols.