Abstract
Background and aims: Cardiac CT integrated into acute stroke imaging protocol detects cardiac thrombi in approximately 6-8% of patients with acute ischemic stroke (AIS). We assessed the yield of cardiac CT in patients with transient ischaemic attacks (TIA).
Methods: We compared the data from the AIS of HEARTS collaboration with contemporaneous TIA patients presenting to four hospitals in the Netherlands, Australia, and New Zealand. We compared the rate of cardiac thrombi detected between TIA and AIS patients using standard descriptive statistics and determined the association using multivariable logistic regression analysis.
Results: We included data of 313 TIA and 3919 AIS patients. Compared with AIS patients, TIA patients had similar age (median 73 vs 74, p=0.227), similar sex distribution (female 40.9% vs 42%, p=0.72), lower baseline NIHSS (median 1 vs 6 p<0.001), and significantly less atrial fibrillation (AF) (14.7% vs 23.3%, p<0.001). A cardiac thrombus was detected in only 1.3% (n=4) of TIA patients, compared with 6.2% (n=243) in AIS (p<0.001). This association remained statistically significant after adjusting for age, sex, AF, heart failure, ischaemic heart disease (OR 0.24, 95% CI 0.09-0.64, p=0.005).
Among TIA patients, those with cardiac thrombi more frequently had AF at baseline (50% vs 14.2%) and MACE at 90-day follow up (25% vs 2.6%). There was no difference in recurrent stroke (0% vs 6.6%) and functional outcome (mRS 0-2: 75% vs 89%).
Conclusions: The yield of cardiac CT in the acute stroke imaging protocol to detect cardiac thrombi is low in TIA patients compared AIS patients.