Left atrial appendage patency and device-related thrombus after percutaneous left atrial appendage occlusion: a computed tomography study.

Aims: Transoesophageal echocardiography studies have reported frequent peri-device leaks and device-related thrombi (DRT) after percutaneous left atrial appendage (LAA) occlusion. We assessed the prevalence, characteristics and correlates of leaks and DRT on cardiac computed tomography (CT) after LAA occlusion.

Methods and results: Consecutive patients underwent cardiac CT before LAA occlusion to assess left atrial (LA) volume, LAA shape, and landing zone diameter. Follow-up CT was performed after >3 months to assess device implantation criteria, device leaks and DRT. CT findings were related to patient and device characteristics, as well as to outcome during follow-up. One-hundred and seventeen patients (age 74 ± 9, 37% women, CHA2DS2VASc 4.4 ± 1.3, and HASBLED 3.5 ± 1.0) were implanted with Amplatzer cardiac plug (ACP)/Amulet (71%) or Watchman (29%). LAA patency was detected in 44% on arterial phase CT images and 69% on venous phase images. The most common leak location was postero-inferior. LAA patency related to LA dilatation, left ventricular ejection fraction impairment, non-chicken wing LAA shape, large landing zone diameter, incomplete device lobe thrombosis, and disc/lobe misalignment in patients with ACP/Amulet. DRT were detected in 19 (16%), most being laminated and of antero-superior location. DRT did not relate to clinical or imaging characteristics nor to implantation criteria, but to total thrombosis of device lobe. Over a mean 13 months follow-up, stroke/transient ischaemic attack occurred in eight patients, unrelated to DRT or LAA patency.

Conclusion: LAA patency on CT is common after LAA occlusion, particularly on venous phase images. Leaks relate to LA/LAA size at baseline, and device malposition and incomplete thrombosis at follow-up. DRT is also quite common but poorly predicted by patient and device-related factors.