16-27: Percutaneous Left Atrial Appendage Closure with WATCHMAN device: results from the TRAPS Registry
Steven Williams 1 3 4 5 7 8
Henry Chubb 1 3 4 5 7 8
James Harrison 1 3 4 5 7 8
Nick Linton 1 3 4 5 7 8
John Whitaker 1 3 4 5 7 8
Radoslaw Kiedrowicz 1 2 3 5 7 8
Jaswinder Gill 1 3 4 5 7 8
Michael Cooklin 1 3 4 5 7 8
Christopher Aldo Rinaldi 1 3 4 5 7 8
Steven Niederer 1 3 4 5 7 8
Reza Razavi 1 3 4 5 7 8
Matthew Wright 1 3 4 5 7 8
Mark O'Neill 1 3 4 5 7 8
0 P.O. Riunito-Ospedale Civile , Cirie` (TO) , Italy
1 Patrizio Mazzone
2 Szczecin , Poland
3 Frankfurt am Main , Germany
4 London , United Kingdom
5 Stefano Bordignon , Laura Perrotta, Daniela Dugo, Fabrizio Bologna, Alexander Fu ̈ rnkranz, Julian Chun, and Boris Schmidt
6 Mirano Hospital , Mirano (VE) , Italy
7 Zabrze , Poland
8 Katarzyna Mitrega , Witold Streb, Magdalena Szymala, Tomasz Podolecki, and Zbigniew Kalarus
Introduction: Left atrial fibrosis is thought to be associated with outcome after pulmonary
vein isolation (PVI) in atrial fibrillation (AF) patients. Aim of this study is to determine (1) the
incidence of bipolar left atrial (LA) electroanatomical scar (EA-S) in patients with AF
undergoing PVI and (2) its correlation with LA conduction delay.
Methods: Consecutive AF patients referred for PVI using an electroanatomical 3D mapping
system were enrolled. The map was acquired during SR. Patients in which SR could not be
restored after 2 DC cardioversion were excluded. After double transseptal puncture “time to
LAA” was determined (TTLAA: P wave begin to first intracardiac LAA activation measured
with a circumferential mapping catheter (CMC)). EA-S was defined as an area with a bipolar
amplitude ,0.5 mV. Two groups were defined: group A, without EA-S, and group B, with
presence of left atrial EA-S.
Results: One hundred patients (61 male, 69 + 9 years, 41 paroxysmal AF (PAF), 59
persistent AF ( persAF)) were enrolled, 49/100 had no EA-S, group A, 51/100 presented an EA-S,
group B. In group B patients were significantly older (74 + 6 vs. 64 + 9 years; p , 0.001),
were mostly female (29/51 (57%) vs. 10/49 (20%), p , 0.001) and had a history of arterial
hypertension (44/51 (86%) vs 30/49 (61%), p , 0.001). EA-S were localized in the left atrial
anterior wall in 48/51 patients (94%), while EA-S in the lateral wall were rare (6/51 pts,
11,8%). Septal, posterior, roof and inferior EA-S were identified in the 65%, 39%, 39% and
22% of patients of group B, respectively. TTLAA was significantly longer in patients with
LAEA-S (122 + 43 vs. 84 + 22 ms, group B and group A, respectively p , 0.001). P wave
duration was longer in group B (133 + 28 vs. 119 + 18 ms, p ¼ 0.009). PVI was achieved
in all patients without major complications. Follow up is ongoing to clarify the clinical role
after a PVI procedure of EA-S and interatrial conduction delay.
Conclusion: Age, female sex and hypertension are risk factors for EA-S. Sinus rhythm
conduction time to LAA is significantly longer in patients presenting with left atrial EA-S.
Conflict of interest: none
Purpose of the study: The WATCHMAN device for Left Atrial Appendage occlusion
(LAAO) has been found effective and non-inferior to the oral anticoagulation (OAC) in patients
with atrial fibrillation, and is now adopted in clinical practice.
Method used: The TRAPS registry is an observational, multicenter registry involving 4 Italian
centers. Patients who underwent LAAO with WATCHMAN device were enrolled, and clinical,
demographic and procedural data were collected.
Summary of results: In this analysis we included 148 patients. Mean age was 73 + 8 years,
57% of patients were male, 22% had heart failure, 10% had a history of transient ischemic
attack, 23% a history of ischemic stroke and 72% a history of bleeding. The baseline CHADS2
score was 2.4 + 1.3, the CHADsVASc score was 3.9 + 2.8 and the HAS-BLED score was
3.3 + 1.1. 57% of the patients were on OAC at the time of implantation. The device was
successfully positioned in all patients with no or minimal (, 5mm) leakage, assessed by
peri-procedural transesophageal echo. Following intra-procedural complications were reported: 2
pericardial effusions treated with pericardiocentesis, 1 device-associated thrombus formation
treated with aspiration, and 1 vascular access dissection. Moreover, device embolization was
reported in 1 patient early after the implantation and the device was successfully snared in the
iliac bifurcation. The overall rate of adverse events within 7 days was therefore (5/148 ¼
3.3%). In the present experience, device size was chosen 20% greater than the LAA diameter
in 68% of patients. Comparing the first half with the second half of the study cohort, the
procedure duration decreased from 89min to 64min ( p , 0.0001).
Conclusion: The success rate in LAAO with WATCHMAN was high, and the rate of adverse
events was low. We reported an experience-related improvement in procedural duration.
Conflict of interest: none
PERCUTANEOUS LEFT ATRIAL APPENDAGE CLOSURE WITH
WATCHMAN DEVICE: RESULTS FROM THE TRAPS REGISTRY
RESIDUAL LEAKS AFTER LEFT ATRIAL APPENDAGE CLOSURE IN
CORRELATION TO THE SHAPE OF OSTIUM AND LENDING ZONE
ELECTRICAL REMODELING DOES NOT REPRODUCIBLY MIRROR
STRUCTURAL CHANGE ON ATRIAL MRI
Purpose of the Study: Left atrial (LA) late gadolinium enhancement (LGE) CMR has been proposed to guide
ablation therapy and predict procedure outcome. AF electrical and structural remodeling are known to occur to
varying extents in patients with AF. The extent to which pre-procedural LGE CMR reflects LA
electrophysiological remodeling is unknown.
Methods Used: Intracardiac contact mapping during programmed stimulation was used to characterize conduction
delay, activation dispersion, electrogram morphology and AF vulnerability in 8 patients undergoing first-time
pulmonary vein isolation. Pre-procedural LGE CMR imaging was performed and quantified by LGE signal intensity
(SI; image intensity ratio, IIR), % LGE area and LA sphericity indices.
Summary of Results: Although electrogram voltage was significantly reduced at areas of high SI (IIR . 0.97
0.9mV vs 1.3mV P ¼ 0.049), IIR showed no relationship with electrogram duration (17.7ms vs 18.3ms, P ¼
0.673), conduction delay (322ms vs 316ms, P ¼ 0.3939) or conduction block (262ms vs. 262ms, P ¼ 0.8667).
There was no relationship between % LGE area or sphericity and any of the LA remodeling parameters (all P .
0.05). LA extrastimulus conduction curves showed significant variation between patients with similar LGE CMR
Conclusions: Although these results confirm that electrogram voltage may be significantly reduced at areas of
high LGE signal intensity, other markers of LA remodeling showed significant variation between patients with
similar LGE appearances. These findings may have important implications for patient-tailored ablation strategies
guided by the integration of electrical and structural information.
Conflict of interest: none
DM t.2 [n]
AF paroxysmal [n]
Background: Left atrial appendage occlusion is a method of stroke prevention in atrial fibrillation patients. The
protective effect of LAA elimination may be deteriorated by periocclcuder leakage. Inadequate occlude sizing is a
known risk factor for a leak occurrence, however additional factors may be important.
Aim: The current study was aimed to assess the relation between morphology of left atrial appendage ostium and
the shape of landing zone chosen for occluder implantation, and the residual leak incidence.
Methods: 40 consecutive patients (79 + 8 years; 57% male) underwent LAA occlusion with Amplatzer Cardic
Plug or Amplatzer Amulet. During LAA occlusion the patients were anesthetized and the transesophageal
echocardiography (TOE) was performed. The shape of ostium and lending zone was assessed based on two
perpendicular measurements (r1, r2) of each using 3-D TOE. The patients were divided to group with elliptic shape of LAA
ostium and lending zone (r1 - r2 . 4 mm) or the regular one (r1 - r2 4 mm). Statistical analysis was
performed using Yates x2 test.
Results: Mean values of LAA ostium diameters measured in two perpendicular axis using 3-D TOE were 29.6
+ 5.3 and 24.8 + 4.5 mm. Mean values of LAA lending zone dimensions also measured in two perpendicular
axis were 21.4 + 3.8 and 17.1 + 4.1 mm. In patients with elliptic shape of LAA ostium (n ¼ 18; 45%) 27%
(n ¼ 5) had residual leak in TOE, whereas in patients with regular shape of LAA ostium none residual leak was
observed. The elliptic shape of LAA lending zone was found in 19 patients (47.5%) and 26% of those patients had
diagnosed residual leak. Patients with regular shape of lending zone had no residual leak. The comparison of
different shapes of LAA ostium or lending zone and residual leak incidence showed the significant differences for
the elliptic and regular shape (P ¼ 0.03, P ¼ 0.04 respectively).
Conclusions: In patients with regular shape of LAA ostium or lending zone statistically less residual leaks were
observed.Table 1Group characteristic
HA – atrial hypertension; DM t.2 – diabetes mellitus type 2; CHD – coronary heart disease; AF – atrial
Conflict of interest: none
70 + 8