Abstract
Background
Standardized transesophageal echocardiography (TEE) data regarding peri-device leaks and thrombus formation after left atrial appendage occlusion (LAAO) are sparse. Criteria to quantify residual leaks seem to appear highly arbitrarily. We therefore examined patients by long-term TEE follow-up after LAAO in a standardized Echo protocol.
Methods
Our patients were prospectively examined by questionnaires and TEE 6 weeks post-procedurally and ≥12 months after LAAO. TEE was performed according to the Munich consensus document, including multiple TEE views (2D and 3D) and an adequate, lower Nyquist limit (20–30cm/s) to detect low velocity flow. The severity of the leaks were graded as follows: >5mm severe, 3–5mm moderate and 1–2mm mild. Antithrombotic regime was administered according to the PROTECT-AF protocol for the Watchman device in the first 60 patients, and dual antiplatelet therapy for 1–3 months thereafter as well as for the Amplatzer device.
Results
In total, 107 LAAO were performed in our institution in patients with a mean age of 77 years. Mean CHA2DS2-VASc score was 4.3±1.3, mean HAS-BLED score 2.7±0.9. Mean clinical follow-up duration was 52.8 months, mean TEE follow-up 43.1 months. The overall annual stroke rate was 1.49%. None of the patients with a post-procedural stroke showed an intracardial thrombus by TEE after six weeks, during the event and in long-term FU. In long-term TEE follow-up 46% (n=18/39) had a mild to moderate residual leak and 2.6% (n=1/39) a severe leak >5mm.Two of the patients with a permanently existing leak sustained an ischemic stroke: in the first patient with a peri-device leak of 3mm the stroke occurred 80 months post-procedurally, in the second one with a growing leak from 3 to 6,5mm 25 months post-procedurally, both classified most likely as a cardioembolic stroke. One patient without a leak in long-term TEE FU suffered a stroke 5 years after LAAO (p=0.61). In all three patients antithrombotic regime during the event consisted of ASS alone.Device-associated thrombus was observed in 1.9% (n=2/107) six weeks after the procedure and increased to 15.4% (n=6 out of 39 TEE studies) in long-term FU, all appeared immobile. None of those patients suffered a thromboembolic event.
Conclusions
In long-term follow up, residual, mostly mild to moderate leaks and device-associated thrombi after LAAO seem to occur occasionally. However, patients with a residual leak or immobile thrombus do not appear to have a higher risk for thromboembolism so far.