Haemodynamic changes after interventional closure of the left atrial appendage may facilitate peri-device leaks

This commentary refers to ‘Residual leaks following percutaneous left atrial appendage occlusion and outcomes: a meta-analysis’, by A. Samaras et al., https://doi.org/10.1093/eurheartj/ehad828 and the discussion piece ‘Left atrial appendage occlusion leaks matter: the cryptic interplay of post-procedural haemodynamic changes and device surveillance’, by A. Samaras and A. Tzikas, https://doi.org/10.1093/eurheartj/ehae459. With great interest, we read the meta-analysis by Samaras et al. who showed that after interventional closure of the left atrial appendage (LAA), peri-device leaks (PDL) assessed by trans-oesophageal echocardiography (TEE) regardless of their size are associated with adverse events, primarily thromboembolism.1 The authors have to be congratulated for their analysis refuting the arbitrary classification into ‘significant’ and ‘non-significant’ PDL, which was a source of confusion for many years. In Table 1, baseline data such as diabetes mellitus, hypertension, and coronary artery disease are presented; however, we miss information about pre-existing heart failure (HF). Left atrial appendage occlusion interacts unfavourably with left atrial compliance and reservoir function.2 In addition, the LAA is an important structure for the release of natriuretic peptides, and LAA occlusion may impede the physiological regulation in HF. After LAA occlusion in 282 patients, a 19% incidence of acute HF has been reported during a follow-up period of 40 months.3 A further study in 98 patients found that 16% of the patients experienced an unexpected HF admission after LAA occlusion. Especially patients with a history of HF were at risk of acute HF events after LAA occlusion.4 The development of HF after LAA occlusion may play an important role in the pathogenesis of PDL. An acute increase of the invasively measured left atrial pressure (LAP) in 48% of 302 patients after interventional LAA occlusion has been reported. Patients with an increase in LAP ≥ 28% were more likely to develop a PDL at 45 days.5 The authors hypothesize that an increase in LAP after LAA occlusion results in acute left atrial dilation, which may promote the development of a PDL by having a negative impact on the device seal in the peri-procedural period. In addition, subsequent left atrial remodelling after LAA occlusion may result in late re-opening of an initially closed LAA.5 Therefore, it would be of interest to know if mechanisms differ for early PDL and late re-openings. In 12 of the 48 studies included in the meta-analysis by Samaras et al., TEE examinations performed 12 months after LAA occlusion have been reported. An additional analysis of risk factors and outcomes of patients with growth of a PDL or late re-opening would be of great value. How to explain the increased all-cause mortality and major bleeding rate in patients with a PDL, which the authors describe in their meta-analysis?1 Is the worse outcome a consequence of more comorbidities, the development of HF after the procedure, and/or the longer duration of anticoagulant therapy in patients with PDL compared with patients without? Regarding the clinical implications, optimization of pre-procedural planning and technical refinements of the occluder implantation are suggested.1 But procedural measures alone may not completely eliminate PDL. Since atrial remodelling after LAA occlusion may contribute to persistence and growth of a PDL and promote re-opening of an initially closed LAA, more long-term data are urgently needed. To increase our knowledge on pathogenesis, evolution over time, and risk factors for PDL, systematic studies after LAA closure with routine TEE examinations after 12 months and at yearly intervals thereafter are necessary since even small PDL are associated with adverse events. All authors declare no disclosure of interest for this contribution.