Catheter-based left atrial appendage closure guided by intracardiac echocardiography in a patient with recurrent esophageal variceal bleedings
A 76-year-old patient with permanent atrial fibrillation and recurrent oesophageal variceal bleeding can no longer receive oral anticoagulation, leaving stroke prevention as a major challenge. How would you manage this complex balance between thromboembolic and bleeding risk?
A 76-year-old male patient presented with a problem of recurrent gastrointestingal bleedings from esophageal varices while under direct oral anticoagulant (DOAC) therapy. Oral anticoagulant therapy was indicated for stroke prevention in this patient with known permanent atrial fibrillation and a CHADS-VASc risk score of 4. Other co-morbidities included diabetes mellitus type 2, congestive heart failure with a reduced left ventricular ejection fraction (LVEF 45%) and liver cirrhosis Child-Pugh class A (Slide 2). Because of two recent episodes of melena, DOAC therapy was paused and the patient does not take any anti-thrombotic therapy at this moment. As an alternative stroke prevention therapy, the patient was now referred for catheter-based left atrial appendage (LAA) closure.
We planned this patient for an ECG-gated, contrast-enhanced and thin-sliced cardiac CT in order to plan for a catheter-based LAA closure. Pre-procedural planning with cardiac CT has several advantages over TEE imaging (Slide 5), as it provides:
A complete 3D LAA assessment, resulting in a deep understanding of the overall LAA morphology, complexity, orientation, number of LAA lobes, the overall tapering in LAA dimension and a detailed visualisation of the LAA trabeculations (Slides 6-8).
Accurate measurements of the LAA dimensions in a normovolemic patient, as cardiac CT can be performed without fasting the patient. LAA dimensions can only be measured reliably by making use of 3D multi-plane reconstructed (MPR) imaging, whether by cardiac CT or 3D-TEE imaging. Cardiac CT also provides the option to make and repeat measurements offline by different specialists and/or ask for computational modelling of different LAA occluder sizes and/or positions in case of a complex LAA morphology (Slides 9-13).
The optimal fluoroscopic implantation view - i.e., the fluoroscopic view which results in minimal LAA foreshortening, minimal overlap between the left atrium and LAA and an aligned LAA ostium (and landing zone) (Slides 14-15).
Information on the optimal transseptal puncture site to obtain the best possible coaxiality between delivery sheath and LAA central axis. A general recommendation is to puncture as inferior as possible. A mid-posterior puncture is recommended in case of a classical LAA orientation, whereas a mid-anterior puncture is recommended in case of a reverse chicken wing (Slides 16-18).
The cardiac CT of our patient demonstrated a single-lobe, angulated ('chicken wing') LAA, measuring 25.7 x 30.7 mm at the ostium (perimeter-derived mean diameter 28.2 mm) and 25.0 x 27.2 mm (mean diameter 26.0 mm) a few millimetres distal into the LAA, indicating only a mild tapering in LAA dimensions. Aiming for a 10-20% oversizing, we opted for a WATCHMAN Flx Pro 31 mm closure device. The optimal fluoroscopic implantation view was estimated to be a 54o right anterior oblique (RAO) and 7o cranial view. An inferior and mid-posterior transseptal puncture should result in the best coaxiality, as assessed by CT image analysis.
The procedural plan was to perform a catheter-based LAA closure in local anesthesia guided by fluoroscopy and 3D-ICE guidance (Slides 19-20). The following materials were used:
Imaging
Pre-procedural: cardiac CT-scan: ECG-gated, thin-sliced (< 0.75 mm), and contrast-enhanced CT-scan with an appropriate end-systolic phase imaging
Intra-procedural: fluoroscopy (Siemens Artis icono)
Intra-procedural: VeriSight Pro 3D-ICE probe (Philips)
Transseptal puncture and LAA closure device (Slides 21-24)
Procedural safety is key when performing a catheter-based LAA closure. In order to do so, a step-wise procedural execution should be strived for (Slide 25):
Step 1. Transseptal puncture
ICE provides excellent images for guidance of the transseptal puncture. The ICE catheter should be lined up at the level of the fossa ovalis, often best with a discrete retroflex of the catheter to obtain some distance between the catheter and the fossa.
Imaging too superiorly on the septum is often seen in early experience. A very posterior transseptal puncture is typically not needed, not always safe and can make crossing the ICE catheter through the interatrial septum more challenging (Slides 26-28).
In our patient, an inferior and mid (not posterior, not anterior) transseptal puncture was performed by use of the VersaCross Connect LAAC Access Solution system. Next, the TruSteer Access System could immediately be pushed over to the left atrium and the VersaCross wire was positioned in the left upper pulmonary vein (Slide 29). Following two-three passages of the TruSteer over the interatrial septum, the sheath was pulled back to the level of the diaphragm, after which the VeriSight Pro 3D-ICE catheter could easily be crossed over to the left atrium and positioned in the left upper pulmonary vein or mid-left atrium (en face view of the LAA).
Step 2. Ruling out thrombus
Using the X-plane function of the 3D-ICE catheter, thrombus in the LAA could be ruled out (Slides 30-31).
Step 3. LAA imaging & sizing
With the 3D-ICE catheter positioned in the left upper pulmonary vein and at the level (depth-wise) of the LAA ostium, the LAA ostium could be measured making use of the 3D-MPR function of 3D-ICE probe. Hereby, it is important to realise that the measured 'LAA ostium' for a LAA closure procedure with the WATCHMAN Flx Pro device does not necessarily correspond to the anatomical ostium of the LAA. Typically, the measurement of the 'LAA ostium' should be made where the implanter aims to position the crown of the WATCHMAN occluder; this is typically a few millimeters distal from the ostium into the LAA. In our case, the LAA ostium was measured 25.4 x 26.9 mm by 3D-ICE imaging intra-procedurally (Slides 32-33). This corresponded well with the pre-procedural LAA sizing by cardiac CT, and confirmed our device choice of a WATCHMAN Flx Pro 31 mm device (15-20% over-sizing) (Slide 34).
Step 4. LAAC device deployment
Deployment of the WATCHMAN Flx Pro occluder is a controlled, 3-step process (Slides 35-36):
First, the device is exposed from the delivery sheath until a ball shape is visible. When the ball shape is obtained, the anchors are not yet functional, making it the right moment to adjust the co-axiality between the delivery sheath and LAA central axis, if needed. In nearly all cases, this implicates a counter-clockwise (anterior) torque of the delivery sheath - in 90-95% of cases, the flush port of the delivery sheath will be pointing upwards (12 o'clock). The delivery sheath co-axiality can be checked on echocardiography or fluoroscopy by injection of contrast into the LAA.
Next, the device is further exposed and slowly deployed into the anatomy. Depending on the obtained depth into the LAA with the delivery sheath, this often implicates a partial 'pushing' of the device in combination with a partial 'retraction' of the delivery sheath.
Finally, the delivery cable is pushed softly into the deployed WATCHMAN Flx Pro device after final device deployment in order to obtain the small central indentation at the proximal end of the device. This assures good engagement of the WATCHMAN Flx Pro anchors into the surrounding LAA.
A new aspect of the WATCHMAN Flx Pro device is the three proximal markers at the level of the 'shoulders' or 'crown' of the device. This makes it easier to assess device co-axiality (in the CT-calculated optimal implantation view) and possible protrusion of the inferior 'shoulder' of the WATCHMAN device into the left atrium.
Step 5. LAAC assessment before device release
Before final device release, the PASS criteria have to be checked (Slides 37-39):
Position: plane of maximum diameter at or just distal to the LAA ostium. Basically, we assure that the WATCHMAN LAA occluder does not leave LAA trabeculations uncovered, is not protruding more than 30% into the left atrium with one of its 'shoulders', and is implanted with a reasonably good co-axiality within the LAA anatomy.
Anchor: gently pull back, then release the deployment knob to exclude movement of the device within the anatomy and confirm movement of the occluder and LAA together during the tug-test.
Compression: device compression can typically easily be assessed at fluoroscopy by looking for the shape of the implanted device (marshmallow: 10%, bell: 20%, hot dog: 30%). In case of doubt, one can also measure/estimate the device compression by 3D-MPR imaging. A compression range of 10-30% should be aimed for.
Seal: ensure all lobes and trabeculations of the LAA are distal to the closure device and sealed. Peri-device leaks ≥ 3 mm are not recommended.
Summary
In contemporary practice, the goal is to have a 100% safe and effective LAA closure procedure. In order to obtain this goal, pre-procedural planning by means of cardiac CT has shown its value by offering precise LAA imaging and sizing as well as by providing information on the optimal transseptal puncture site and fluoroscopic implantation view. Ideally, intra-procedural imaging relies on 3D-imaging, which can be provided by 3D-ICE or standard 3D-TEE. A safe and accurate transseptal puncture is one of the key conditions for a successful LAA closure procedure and outcome, whereas a steerable delivery sheath can be valuable to optimise co-axial device implantation in challenging LAA anatomies.
06 Jul 2026
Catheter-based left atrial appendage closure guided by intracardiac echocardiography in a patient with recurrent esophageal variceal bleedings
A 76-year-old patient with permanent atrial fibrillation and recurrent oesophageal variceal bleeding can no longer receive oral anticoagulation, leaving stroke prevention as a major challenge. How would you manage this complex balance between thromboembolic and bleeding risk?
Frequency of the problem:
Expert level:
Author
Interventional cardiologist / Cardiologist
København, Denmark
Latest contributions
Catheter-based left atrial appendage closure guided by intracardiac echocardiography in a patient with recurrent esophageal variceal bleedings TAVIBy O. De Backer
A 76-year-old male patient presented with a problem of recurrent gastrointestingal bleedings from esophageal varices while under direct oral anticoagulant (DOAC) therapy. Oral anticoagulant therapy was indicated for stroke prevention in this patient with known permanent atrial fibrillation and a CHADS-VASc risk score of 4. Other co-morbidities included diabetes mellitus type 2, congestive heart failure with a reduced left ventricular ejection fraction (LVEF 45%) and liver cirrhosis Child-Pugh class A (Slide 2). Because of two recent episodes of melena, DOAC therapy was paused and the patient does not take any anti-thrombotic therapy at this moment. As an alternative stroke prevention therapy, the patient was now referred for catheter-based left atrial appendage (LAA) closure.
Powered by Quiz MakerBy O. De Backer
We planned this patient for an ECG-gated, contrast-enhanced and thin-sliced cardiac CT in order to plan for a catheter-based LAA closure. Pre-procedural planning with cardiac CT has several advantages over TEE imaging (Slide 5), as it provides:
The cardiac CT of our patient demonstrated a single-lobe, angulated ('chicken wing') LAA, measuring 25.7 x 30.7 mm at the ostium (perimeter-derived mean diameter 28.2 mm) and 25.0 x 27.2 mm (mean diameter 26.0 mm) a few millimetres distal into the LAA, indicating only a mild tapering in LAA dimensions. Aiming for a 10-20% oversizing, we opted for a WATCHMAN Flx Pro 31 mm closure device. The optimal fluoroscopic implantation view was estimated to be a 54o right anterior oblique (RAO) and 7o cranial view. An inferior and mid-posterior transseptal puncture should result in the best coaxiality, as assessed by CT image analysis.
The procedural plan was to perform a catheter-based LAA closure in local anesthesia guided by fluoroscopy and 3D-ICE guidance (Slides 19-20). The following materials were used:
Imaging
Transseptal puncture and LAA closure device (Slides 21-24)
Procedural safety is key when performing a catheter-based LAA closure. In order to do so, a step-wise procedural execution should be strived for (Slide 25):
Step 1. Transseptal puncture
ICE provides excellent images for guidance of the transseptal puncture. The ICE catheter should be lined up at the level of the fossa ovalis, often best with a discrete retroflex of the catheter to obtain some distance between the catheter and the fossa.
Imaging too superiorly on the septum is often seen in early experience. A very posterior transseptal puncture is typically not needed, not always safe and can make crossing the ICE catheter through the interatrial septum more challenging (Slides 26-28).
In our patient, an inferior and mid (not posterior, not anterior) transseptal puncture was performed by use of the VersaCross Connect LAAC Access Solution system. Next, the TruSteer Access System could immediately be pushed over to the left atrium and the VersaCross wire was positioned in the left upper pulmonary vein (Slide 29). Following two-three passages of the TruSteer over the interatrial septum, the sheath was pulled back to the level of the diaphragm, after which the VeriSight Pro 3D-ICE catheter could easily be crossed over to the left atrium and positioned in the left upper pulmonary vein or mid-left atrium (en face view of the LAA).
Step 2. Ruling out thrombus
Using the X-plane function of the 3D-ICE catheter, thrombus in the LAA could be ruled out (Slides 30-31).
Step 3. LAA imaging & sizing
With the 3D-ICE catheter positioned in the left upper pulmonary vein and at the level (depth-wise) of the LAA ostium, the LAA ostium could be measured making use of the 3D-MPR function of 3D-ICE probe. Hereby, it is important to realise that the measured 'LAA ostium' for a LAA closure procedure with the WATCHMAN Flx Pro device does not necessarily correspond to the anatomical ostium of the LAA. Typically, the measurement of the 'LAA ostium' should be made where the implanter aims to position the crown of the WATCHMAN occluder; this is typically a few millimeters distal from the ostium into the LAA. In our case, the LAA ostium was measured 25.4 x 26.9 mm by 3D-ICE imaging intra-procedurally (Slides 32-33). This corresponded well with the pre-procedural LAA sizing by cardiac CT, and confirmed our device choice of a WATCHMAN Flx Pro 31 mm device (15-20% over-sizing) (Slide 34).
Step 4. LAAC device deployment
Deployment of the WATCHMAN Flx Pro occluder is a controlled, 3-step process (Slides 35-36):
Step 5. LAAC assessment before device release
Before final device release, the PASS criteria have to be checked (Slides 37-39):
Summary
In contemporary practice, the goal is to have a 100% safe and effective LAA closure procedure. In order to obtain this goal, pre-procedural planning by means of cardiac CT has shown its value by offering precise LAA imaging and sizing as well as by providing information on the optimal transseptal puncture site and fluoroscopic implantation view. Ideally, intra-procedural imaging relies on 3D-imaging, which can be provided by 3D-ICE or standard 3D-TEE. A safe and accurate transseptal puncture is one of the key conditions for a successful LAA closure procedure and outcome, whereas a steerable delivery sheath can be valuable to optimise co-axial device implantation in challenging LAA anatomies.
The Essentials - LAAC
Disclaimer
This case report does not reflect the opinion of PCR or PCRonline, nor does it engage their responsibility.