Atlas Contents

Anatomy for Left Atrial Appendage Closure

1 Anatomical identity of the left atrial appendage

The left atrial appendage is not a passive sac attached to the left atrium. It is the trabeculated remnant of the primitive atrium and forms one of the four morphological components of the chamber, together with the venous component, body and mitral vestibule. In attitudinal orientation it projects leftward from the anterolateral atrial wall, superior to the left atrioventricular groove. Its tip may extend anteriorly over the pulmonary trunk, superiorly behind the arterial pedicle or posteriorly along the left ventricular surface. Only a small portion usually contributes to the left cardiac border in a frontal projection [1,2].

The smooth wall of the atrial body ends at a variably defined mouth, beyond which the endocardium becomes trabeculated by pectinate muscles. The appendage has reservoir, contractile and neuroendocrine functions in addition to its well recognised role in thrombus formation. Atrial fibrillation, atrial dilatation and loss of effective contraction favour stasis, spontaneous echo contrast and thrombus. Closure therefore removes a major source of embolism but also excludes a dynamic part of the atrial chamber. The anatomical objective is complete exclusion of the cavity without compromising adjacent myocardium, vessels or valves [1,3,4].

2 Ostium neck landing zone and lobes

Three regions should be distinguished during planning. The ostium is the anatomical mouth where the smooth atrial body becomes the trabeculated appendage. The neck extends from this mouth to the entrance of the first major lobe. Within the neck, the device specific landing zone is the circumferential region selected for anchoring or compression. The ostium and landing zone are therefore related but not interchangeable planes. A plug device is generally seated within the neck, whereas a lobe and disc system anchors more distally and uses the disc to cover the mouth from the atrial side [1,4–7].

The ostium is most often oval and its long axis is oblique to the mitral annulus. In the anatomical and imaging series reported by Cabrera and colleagues, an oval configuration was found in 68.9 percent, while foot like, triangular, water drop and round configurations were less frequent. Mean dimensions from fixed specimens or particular imaging cohorts should not be transferred directly to an individual patient. The orifice changes during the cardiac cycle and may enlarge and become more circular with atrial remodelling. Measurements must identify the maximal and minimal diameters, perimeter, area, depth and orientation in the phase and loading conditions relevant to implantation [1].

The appendage may have one or several lobes with early bifurcation, sharp bends or a short usable neck. The descriptive categories chicken wing, windsock, cactus and cauliflower are useful for communication but are not substitutes for multiplanar reconstruction. A chicken wing may be straightforward when the proximal segment is long, yet difficult when an early bend shortens the landing zone. A cauliflower configuration may combine limited depth, several small lobes and dense trabeculation. Device choice should therefore be based on the actual ostial and landing zone geometry, not on the morphological nickname alone.

3 Wall structure and muscular architecture

The appendage wall is thin and markedly nonuniform. Pectinate muscles commonly exceed 1 mm in thickness, whereas the intervening wall can be translucent or paper thin. The posterior and distal walls may contain less than 1 mm of myocardium. In some hearts, extra appendicular pectinate muscles extend inferiorly from the mouth towards the mitral vestibule. The troughs between these trabeculations can be extremely thin and may create an irregular sealing surface. They also explain why deep advancement of a wire, pigtail, sheath or device can perforate the appendage even when the manoeuvre appears gentle [1,3,4].

Transmurally, the wall consists of overlapping and interdigitating aggregates of cardiomyocytes sharing different preferential orientations; these are not discrete insulated layers. Epicardially, the leftward extension of the broad interatrial myocardial connection divides around the appendage neck. Subendocardially, cardiomyocytes from the septoatrial region curve around the mouth and continue into the pectinate muscles. These arrangements contribute to the local contractile pattern and help explain why the ostium and neck are dynamic rather than fixed rings. Chronic atrial fibrillation may add dilatation, fibroelastosis, fibrosis and adipose infiltration, altering both mechanical behaviour and imaging appearance [1–4].

4 Neighbouring structures

The left superior pulmonary vein and appendage mouth frequently lie at a similar level and are separated by the left lateral ridge. This ridge is an infolding of the atrial wall rather than an independent muscular cord. It is often narrow, with thicker myocardium superiorly and a less substantial inferior portion. An indistinct or narrow ridge can make the superior and posterior boundaries of an oval ostium difficult to define on two dimensional imaging. The epicardial aspect contains the Marshall structures and autonomic nerves; in some individuals, a left sided sinus node artery also courses through this region [1–3].

Inferior to the neck, the left circumflex coronary artery and great cardiac vein run within adipose tissue in the left atrioventricular groove. Their distance from the landing zone varies. Excessive oversizing, deep anchoring or an unfavourable lobe and disc relationship can compress the circumflex artery, while an external ligation or clip can threaten the same vessel. The mitral vestibule and hinge of the mural leaflet are nearby; a device that projects too far into the atrium must not interfere with leaflet motion or pulmonary venous flow.

The left phrenic nerve descends on the fibrous pericardium and may pass anterior to the appendage body or directly over its neck. This relationship is especially important for epicardial ligation or clipping. Prior cardiac surgery may add adhesions and place bypass grafts near the appendage tip. A persistent left superior caval vein, although uncommon, can occupy the groove between the appendage and left pulmonary veins before joining the coronary sinus. These structures must be recognised before an epicardial approach [1,8].

5 Imaging based anatomical planning

Preprocedural imaging must answer four questions: is thrombus present; where are the true ostium and landing zone; what device dimensions and depth are available; and what structures are at risk. Transoesophageal echocardiography provides dynamic multiplane assessment, Doppler information and exclusion of thrombus. Three dimensional transoesophageal imaging improves appreciation of the noncircular mouth. Cardiac computed tomography supplies isotropic multiplanar anatomy, perimeter and area measurements, optimal fluoroscopic angles, the transseptal to appendage trajectory and relationships with the circumflex artery, pulmonary vein and mitral valve. Intracardiac echocardiography can provide intraprocedural guidance when used by an experienced team [1,5–7,9].

Slow flow can imitate thrombus on first pass computed tomography. A delayed acquisition helps distinguish persistent thrombus from incomplete contrast mixing. Pectinate muscles should likewise not be mistaken for thrombus. Measurements must be made in a plane perpendicular to the intended axis rather than on a convenient axial slice. The operator should record the ostial plane, device specific landing zone, maximal and minimal diameters, perimeter, area, usable depth before the first major bend, lobe distribution and expected working projection. Agreement between imaging and the intended device instructions is more important than any single diameter [5–7,9].

6 Transseptal access and procedural trajectory

Percutaneous closure usually begins with femoral venous access and passage through the true atrial septum, which is limited to the floor of the oval fossa and its immediate anteroinferior buttress. The surrounding superior, posterior and inferior rims are largely infoldings containing extracardiac adipose tissue. Puncture outside the fossa can enter the transverse pericardial sinus, aorta or atrial wall. The septum should therefore be crossed under transoesophageal or intracardiac echocardiographic guidance rather than from fluoroscopy alone.

An inferior and posterior position within the oval fossa commonly provides a direct path towards the leftward and anteriorly situated appendage. It is not universally optimal. An appendage with a posteriorly directed proximal axis or reverse chicken wing configuration may require a more central or anterior puncture to avoid an acute angle. Conversely, using a patent foramen ovale merely because it is present may create an excessively superior and anterior trajectory. Preprocedural computed tomography and intraprocedural imaging should be used to select the point that permits coaxial engagement, stable sheath support and controlled deployment [5,6,9].

After transseptal crossing, the sheath is advanced under imaging and pressure control. Angiography defines the proximal lobes and confirms the working projection. A soft pigtail rather than a stiff wire should explore the cavity whenever possible. Repeated deep injections or forceful advancement against a distal trabeculated wall should be avoided. The sheath and device should follow the native axis of the neck; coaxial alignment distributes radial force more evenly and reduces eccentric protrusion, incomplete apposition and instability.

7 Device selection deployment and release

Current endocardial devices can be considered broadly as plug systems and lobe and disc systems. Plug devices seal within the neck and depend on adequate depth, circumferential contact, compression and anchoring. Lobe and disc devices anchor within a device specific landing zone while the disc covers the ostium. The second design may accommodate some shallow or complex anatomies, but an excessively deep lobe can approach the circumflex artery and an oversized or malaligned disc can leave gaps or impinge on neighbouring structures. Selection must respect the instructions for the actual device generation; historical dimensions for earlier devices should not be applied automatically to current systems [5–7,10].

Deployment is an anatomical test, not simply delivery of a prescribed size. Before release, imaging must confirm device position relative to the ostium and landing zone, stable anchoring, appropriate compression or conformation, circumferential apposition and satisfactory seal. The mitral valve, left pulmonary veins and circumflex artery should remain unobstructed, and the pericardial space should be checked for effusion. If the device is eccentric or unstable, the cause may be a noncircular neck, an unrecognised proximal lobe, inadequate depth, trabecular interference or a noncoaxial puncture. Repositioning without correcting the anatomical cause may reproduce the same failure.

Epicardial clips and ligation systems exclude the appendage from its outer surface and avoid an endocardial implant, but they require a suitable external neck. A medially rotated tip, adhesions, epicardial fat, the left phrenic nerve, circumflex artery or bypass grafts may prevent safe capture. Surgical or epicardial closure must be verified because a residual stump or incomplete ligation remains a patent cavity rather than anatomical exclusion [1,8].

8 Anatomical mechanisms of complications and follow up

Perforation and tamponade usually reflect the combination of a thin wall and focal mechanical force from a wire, sheath, injection or anchoring element. Embolisation is favoured by undersizing, shallow seating, poor alignment or an anchoring zone that is shorter or more irregular than anticipated. Circumflex compromise can result from deep or oversized implantation. Peridevice leak may follow incomplete circumferential contact at an oval ostium, uncovered secondary lobes, a disc positioned within rather than across the mouth, or channels created by prominent trabeculations. The mechanism and location of a leak are more informative than diameter alone [10,11].

Device related thrombus forms on the atrial surface of the implant and is influenced by device position, exposed metal, incomplete healing, residual flow and patient factors. Immediate colour flow through a porous device should not automatically be labelled a peridevice gap; anatomical separation between device and tissue must be distinguished from transient transdevice permeability. Follow up transoesophageal echocardiography or cardiac computed tomography assesses position, conformation, leak, residual appendage patency, device related thrombus and surrounding structures. Findings should always be interpreted against the original anatomy and the final deployment images [6,10–12].

Key Anatomical Messages

Anatomical focus

Practical implication

Ostium

The anatomical mouth of the appendage and the plane that must ultimately be sealed.

Neck

The segment between the ostium and entrance of the first major lobe.

Landing zone

A device specific anchoring plane within the neck and not a synonym for the ostium.

Wall

Pectinate muscles overlie exceptionally thin intervening myocardium with variable perforation risk.

Transseptal puncture

Usually inferior and posterior within the oval fossa but tailored to the individual appendage axis.

Circumflex artery

Courses inferior to the neck and may be compressed by deep or oversized implantation.

Seal

Depends on circumferential tissue contact and coverage of relevant lobes rather than diameter alone.

Follow up

TEE or CT should relate leak and thrombus to the original anatomy and final device position.

References

  1. Cabrera JA, Saremi F, Sánchez-Quintana D. Left atrial appendage: anatomy and imaging landmarks pertinent to percutaneous transcatheter occlusion. Heart. 2014;100:1636–1650. https://doi.org/10.1136/heartjnl-2013-304464
  2. Ho SY, Cabrera JA, Sánchez-Quintana D. Left atrial anatomy revisited. Circ Arrhythm Electrophysiol. 2012;5:220–228. https://doi.org/10.1161/CIRCEP.111.962720
  3. Cabrera JA, Ho SY, Climent V, Sánchez-Quintana D. The architecture of the left lateral atrial wall: a particular anatomic region with implications for ablation of atrial fibrillation. Eur Heart J. 2008;29:356–362. https://doi.org/10.1093/eurheartj/ehm606
  4. Cabrera JÁ, González-Casal D, Sánchez-Quintana D. Left atrial appendage closure: reaching more patients every day. JACC Case Rep. 2025;30:103291. https://doi.org/10.1016/j.jaccas.2025.103291
  5. Glikson M, Wolff R, Hindricks G, et al. EHRA EAPCI expert consensus statement on catheter based left atrial appendage occlusion an update. Europace. 2020;22:184. https://doi.org/10.1093/europace/euz258
  6. Saw J, Holmes DR, Cavalcante JL, et al. SCAI HRS expert consensus statement on transcatheter left atrial appendage closure. Heart Rhythm. 2023;20:e1–e16. https://doi.org/10.1016/j.hrthm.2023.01.007
  7. Fukutomi M, Fuchs A, Bieliauskas G, et al. Computed tomography based selection of transseptal puncture site for percutaneous left atrial appendage closure. EuroIntervention. 2022;17:e1435–e1444. https://doi.org/10.4244/EIJ-D-21-00555
  8. Sánchez-Quintana D, Cabrera JA, Climent V, et al. Anatomic evaluation of the left phrenic nerve relevant to epicardial and endocardial catheter ablation: implications for phrenic nerve injury. Heart Rhythm. 2009;6:764–768. https://doi.org/10.1016/j.hrthm.2009.02.029
  9. Rosseel L, Wunderlich NC, Berti S, et al. Transseptal puncture for left atrial appendage closure: a consensus document from the Second European Left Atrial Appendage Club. J Soc Cardiovasc Angiogr Interv. 2026;5:104190. https://doi.org/10.1016/j.jscai.2025.104190
  10. Cepas-Guillén P, Holmes DR, Cavalcante J, et al. Imaging assessment after percutaneous left atrial appendage closure: from immediate to long term follow up. Eur Heart J Cardiovasc Imaging. 2025;26:993–1006. https://doi.org/10.1093/ehjci/jeaf078
  11. Alkhouli M, De Backer O, Ellis CR, et al. Peridevice leak after left atrial appendage occlusion: incidence mechanisms clinical impact and management. JACC Cardiovasc Interv. 2023;16:627–642. https://doi.org/10.1016/j.jcin.2022.12.006
  12. Simard T, Hibbert B, Alkhouli MA, Abraham NS, Holmes DR. Device related thrombus following left atrial appendage occlusion. EuroIntervention. 2022;18:224–232. https://doi.org/10.4244/EIJ-D-21-01010

Media

Generic selectors
Exact matches only
Search in title
Search in content
Post Type Selectors
Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore