Prospective, randomized comparison of 3-dimensional computed tomography guidance versus TEE data for left atrial appendage occlusion (PRO3DLAAO)

Document Type


Publication Date


Publication Title

Catheterization and cardiovascular interventions : official journal of the Society for Cardiac Angiography & Interventions


BACKGROUND: Preliminary data comparing 3-dimensional computed tomography (3D-CT) to transesophageal echocardiography (TEE) for left atrial appendage occlusion (LAAO) indicates that 3D-CT provides more accurate measurements and improves case planning. Therefore, we conducted a pilot study comparing 3D-CT to TEE in occluder selection accuracy and procedural efficiency.

METHODS: From May 2016 to February 2017, 24 patients were prospectively randomized to undergo LAAO using either TEE or 3D-CT. The primary endpoint was device accuracy while the secondary endpoints included # devices per case, # guide catheters used per case, # fluoroscopy angles used, procedure time, fluoroscopy time, radiation dose, and major adverse events (stroke, MI, device embolization, perforation, death).

RESULTS: Procedure success was 100% and 92% for the 3D-CT and 2D-TEE cohorts respectively. Accuracy for 1st device selection 92% and 27% (P = .01) for 3D-CT and 2D-TEE respectively but with intra-procedural upsizing in the 2D-TEE cohort, the 2D-TEE cohort accuracy increased to 64% while the 3D-CT groups 92% was accurate (P = .33). Case planning using 3D-CT was significantly more efficient with respect to device utilization (CT 1.33 ± 0.7 vs. 2D-TEE 2.5 ± 1.2 P = .01), guide catheters (CT 1 vs. 2D-TEE 1.7 ± 0.8 P = .01) and procedure time (3D-CT 55 ± 17 min vs. 2D-TEE 73 ± 24 min P < .05). One major adverse event, a stroke occurred in the 2D-TEE group.

CONCLUSION: In this single-center pilot study, CT guided LAAO case planning was associated with improved device selection accuracy and procedural efficiency. This study data supports the notion that comprehensive 3D assessment significantly simplifies LAAO, minimizing the time and number of steps needed.

PubMed ID






First Page


Last Page