LAUSR

Background: Following left atrial appendage (LAA) electrical isolation, the decision on whether to continue oral anticoagulation after successful atrial fibrillation ablation is based on the study of its mechanical function on transesophageal echocardiography (TEE). In this cohort, LAA contraction is absent and the incorrect interpretation of emptying flow velocities can lead to unwanted clinical sequelae. Methods: One hundred and sixty consecutive TEE exams performed to evaluate the LAA mechanical function following its electrical isolation were reviewed by an experienced operator blinded to the original diagnosis of LAA dysfunction. The rate of diagnostic discrepancy in the assessment LAA dysfunction and its clinical implications were evaluated. Results: Diagnostic discrepancy with misclassification of the LAA mechanical function occurred 36% (58/160) of TEE exams. In most cases (57/58), such discrepancy was observed in the setting of an incorrect original diagnosis of a normal LAA mechanical function despite absent/reduced or inconsistent LAA contraction. This main source of this wrong diagnosis was the wrong interpretation of passive LAA flows (34/57; 60%), followed by failure to identify dissociated firing (15/57; 26%). In rare cases (8/57; 14%), velocities of surrounding structures were interpreted as LAA flow due to misplacement of the pulsed-wave Doppler sample volume. Following LAA isolation, the proportion of patients who experienced a cerebrovascular event while off oral anticoagulation due to the misclassification of their LAA mechanical function was 70% (7/10 [95% CI, 40%–89%]). Conclusions: Underdiagnosis of LAA mechanical dysfunction is common in TEEs performed following LAA electrical isolation, and it is associated with an increased risk of cerebrovascular events owing to oral anticoagulation discontinuation despite absent/reduced LAA contraction. Careful review of the TEE exam by an operator with specific expertise in LAA imaging and familiar with the functional implications of LAA isolation is necessary before interrupting oral anticoagulation in this cohort.