17 Correlation of incidental left atrial appendage thrombus on non-cardiac gated CT with transoesophageal echocardiography. (30th September 2020)

• Record Type: Journal Article
• Authors: Waters, M; O' Connor, C; Moore, D; Loo, B
• Is Part Of: Heart. Volume 106(2020)Supplement 4
• Page Start: A12
• Page End: A12
• Publication Date: 2020-09-30
• Subjects: Heart -- Diseases -- Treatment -- Periodicals; Cardiology -- Periodicals; 616.12
• DOI: 10.1136/heartjnl-2020-ICS.17
• Languages: English
• ISSNs: 1355-6037

Abstract

Abstract : Background: Transoesophageal echocardiography is recognized as the gold standard for detection of a left atrial appendage (LAA) thrombus. LAA thrombus is visualized as filling defect on computed tomography (CT), however it can be difficult to differentiate between slow flow and thrombus. This poses challenges in both the reporting and interpretation of left atrial filling defects on non-cardiac gated CTs, and whether to classify this as thrombus or an issue with imaging technique when reporting. Due to the risk of stroke with a LAA thrombus, the reporting of a LAA thrombus on CT makes it difficult for physicians to ignore and usually leads to anti-coagulation, thereby exposing the patient to the risk of bleeding. In this study we assessed the correlation between left atrial appendage thrombus or filling defects on a non-cardiac gated CT, with confirmation on TOE. Methods: We retrospectively analyzed TOE cases performed at our center between March 2016 and March 2020. Patients were included if they had been referred for LAA assessment following identification of a suspected LAA thrombus on CT thorax. Results: 757 TOEs were conducted at our center during the study period. This study includes 19 patients who were referred for TOE following incidental left atrial appendage thrombus on CT. Baseline characteristics include; mean age 69.5 years, male 68.4%, hypertensive 79%, history of atrial fibrillation 63.2%, on long-term anticoagulation 63.2%, on anti-coagulation following CT prior to TOE 89.5%, previous stroke 21%, median CHA2DS2-VASC 3. The mean time from CT to TOE was 16 days. 15.8% (n = 3/19) of the patients referred had evidence of a LAA thrombus on TOE. 66% (n = 2) of these patients had a history of prior stroke, the median CHA2DS2-VASC score was 5 and overall 50% of the patients in our cohort who had a history of stroke had a LAA thrombus at TOE. The mean peak LAA filling velocity was only mildly reduced at 37 cm/s, and 63% (n = 12) of LAA were windsock morphology. In the follow-up period from TOE to data collection, there was no reported subsequent strokes in the 19 patients analyzed, mean follow-up time 14 months. Discussion: In patients who have an incidental left atrial thrombus or filling defect reported on non-gated CT thorax, only a minority have a thrombus at TOE and it was no associated with subsequent stroke risk. A history of stroke or CHA2DS2-VASC of >5 conferred a higher probability of thrombus being present and we recommend that physicians take this into consideration when deciding on anti-coagulation. The LAA filling velocities were only mildly reduced and the radiographic findings are likely related to CT gating rather than an physiological process in the left atrium. Incidental LAA thrombus on CT correlates poorly with presence of thrombus on TOE, and if there is concern a TOE should be performed to assess for a thrombus.