What question did this study set out to answer?

To evaluate the spatial distribution of peri-device leaks (PDLs) using gated cardiac CT following WATCHMAN implantation.

February 2, 2026

Real-world spatial distribution of peri-device leaks on cardiac CT after WATCHMAN implantation

Key Result

At first follow-up, 6.1% of patients had peri-device leaks, primarily located posterior-inferiorly in the LAA ostium, especially for leaks ≥3 mm.

Key Points

To evaluate the spatial distribution of peri-device leaks (PDLs) using gated cardiac CT following WATCHMAN implantation.
Enrolled patients underwent WATCHMAN FLX or FLX Pro implantation in a multicenter registry.
Gated cardiac CT was performed during the first follow-up (45-90 days) after implantation.
PDLs were categorized by size and location on a clock-face framework.
Of 676 patients, 41 (6.1%) had defined anatomical PDLs.
42 leaks were identified; among larger leaks (≥3 mm), the most common sites were at 7:00 (30%).
For smaller leaks (<3 mm), frequent sites were 1:00 and 11:00, each at 36.4%.

Structured PICO

What is the spatial distribution of peri-device leaks on gated cardiac CT at first follow-up in patients post-WATCHMAN implantation?

Population

676 patients undergoing WATCHMAN FLX or FLX Pro implantation enrolled in a prospective multicenter registry with post-implant gated cardiac CT imaging.

Intervention

Gated cardiac CT at first follow-up (45-90 days) using a 12-point clock-face framework for anatomical mapping of peri-device leaks.

Outcome

Spatial distribution of peri-device leaks (PDLs) categorized by minimum width (<3 mm or ≥3 mm) using a 12-point clock-face centered on the left atrial appendage ostium.surrogate

Cardiac CT with clock-face mapping reveals that larger peri-device leaks post-WATCHMAN implantation predominantly occur in the posterior-inferior region, providing a reproducible framework for spatial assessment.

Main Result

Absolute Event Rate: 0% vs 0%

Abstract

Abstract Background The WATCHMAN device requires post-implant imaging to assess peri-device leak (PDL) and device-related thrombus, typically with transesophageal echocardiography (TEE). Cardiac CT has emerged as a non-invasive alternative, offering high-resolution views of left atrial appendage (LAA) anatomy. However, real-world data on CT-based leak localization are limited, and standardized methods to describe leak distribution are lacking. Purpose To evaluate the spatial distribution of PDLs using gated cardiac CT at first follow-up post-WATCHMAN implantation and to apply a reproducible clock-face framework for anatomical mapping. Methods Patients undergoing WATCHMAN FLX or FLX Pro implantation were enrolled in a prospective multicenter registry and followed per local standard of care. Gated cardiac CT was performed first follow up (45-90 days) and interpreted by a central imaging core laboratory. PDLs were defined as contrast opacification within the LAA and categorized by minimum width: 3 mm or ≥3 mm. Leak locations were mapped using a 12-point clock-face centered on the LAA ostium, with 12:00 representing anterior and 6:00 posterior, following standard axial CT orientation. Results Among 676 patients with post-implant CT imaging, 41 (6.1%) had core lab–adjudicated PDLs with defined anatomical location and size. A total of 42 leaks were recorded, including one patient with two distinct sites. Among 30 patients with leaks ≥3 mm, 7:00 was the most common site (9/30; 30%), followed by 6:00, 8:00, and 11:00 (each 4/30; 13.3%). Among 11 patients with leaks 3 mm, the most frequent sites were 1:00 (4/11; 36.4%) and 11:00 (4/11; 36.4%), followed by 12:00 and 2:00. Leaks were less frequently observed at lateral positions (3:00 and 9:00). (Figure) Conclusion At first follow-up, CT-identified peri-device leaks were most frequently located in the posterior-inferior region of the LAA ostium, particularly for larger leaks. Smaller leaks tended to occur more anteriorly. Clock-face mapping offers a reproducible and intuitive framework for spatial assessment of PDLs and may improve procedural planning, cross-modality comparisons, and device refinement.

Bookmark