We read with interest the recent paper by Sedaghat et al. (1) in which they report a high incidence of thrombus between the left upper pulmonary vein and the Amulet device (St. Jude Medical, St. Paul, Minnesota) on transesophageal echocardiogram performed 11.0
8.2 weeks following left atrial appendage (LAA) occlusion. These thrombi were observed in 4 of 24 cases (16.7%) in spite of a strategy of dual antiplatelet therapy for 3 months post-implantation. The high incidence of thrombus in this Amulet series contrasts with our own experience (2) in a very similar population of patients (mean age: 75
9 years, median CHA2DS2VASc [Congestive Heart Failure, Hypertension, Age $75 Years, Diabetes Mellitus, Previous Stroke or Transient Ischemic Attack, Vascular Disease, Age 65 to 74 Years, Sex Category]: 4) undergoing LAA occlusion with the Amulet device. Our patients received 6 weeks of double antiplatelet therapy and no thrombus was visualized in any of 32 patients on follow-up transesophageal echocardiogram at 6 to 8 weeks post-implantation. Our results are in line with the other published series of Amulet device implants, in which the prevalence of thrombi on follow-up imaging ranged from 0% to 4% (3). This discrepancy between the findings of Sedaghat et al. (1) and other published series could simply be explained by statistical fluctuation, given the low number of patients. However, a genuine reason for the higher prevalence of thrombus in their patients might be found in their observation that thrombus was associated with incomplete coverage of the ridge between the LAA and the left upper pulmonary vein. In our experience (2), and that of others (3), intraprocedural echocardiography (either transesophageal or intracardiac) is invaluable in helping achieve complete LAA occlusion. The mere use of fluoroscopy, even in a biplane lab, cannot provide the level of detail of LAA anatomy contributed by echocardiography (4,5). It is notable that Sedaghat et al. (1) do not mention the use of echocardiography to guide device implantation in their methods. However, given that their procedures were all performed under conscious sedation and with an average procedure duration of