A systematic review and meta‐analysis of the impact of the left atrial appendage closure on left atrial function

Abstract Background Left atrial (LA) appendage closure (LAAC) is effective in patients with atrial fibrillation who are not candidates for long‐term anticoagulation. However, the impact of LAAC on LA function is unknown. The aim of this study is to evaluate the impact of LAAC on atrial function. Methods This meta‐analysis was conducted according to Preferred Reporting Items for Systematic Reviews and Meta‐Analyses guidelines. A search strategy was designed to utilize PubMed/Medline, EMBASE, and Google scholar for studies showing the effect of LAAC on the LA function from inception to November 20, 2021. The standardized mean difference (SMD) was calculated from the means and standard deviations. Results Of 247 studies initially identified, 8 studies comprising 260 patients were included in the final analysis. There was a significant increase in LA emptying fraction following LAAC compared with preoperative function (SMD: 0.53; 95% confidence interval [CI]: 0.04–1.01; p = .03; I 2 = 75%). In contrast, there were no significant differences in LA volume (SMD: −0.07; 95% CI: −0.82–0.69; p = .86; I 2 = 92%) peak atrial longitudinal strain (SMD: 0.50; 95% CI: −0.08–1.08; p = .09; I 2 = 89%), peak atrial contraction strain (SMD: 0.38; 95% CI: −0.22–0.99; p = .21; I 2 = 81%), strain during atrial contraction (SMD: −0.24; 95% CI: −0.61–0.13; p = .20; I 2 = 0%), strain during ventricular systole (SMD: 0.47; 95% CI: −0.32–1.27; p = .24; I 2 = 89%), strain during ventricular diastole (SMD: 0.09; 95% CI: −0.32–0.51; p = .66; I 2 = 65%). Conclusion LAAC is associated with improvement in the left atrial emptying fraction, but did not significantly influence other parameters.


| INTRODUCTION
Standard treatment in nonvalvular atrial fibrillation is oral anticoagulation, either direct oral anticoagulants or warfarin. 1 For patients who are unable tolerate oral anticoagulant agents, or have a contraindication, left atrial appendage closure (LAAC) is considered an effective alternative with less risk of bleeding. The closure of the Left atrial appendage (LAA) has been approved for some devices. In 2015, the Food and Drug Administration authorized the WATCHMAN as the first device for LAA, and the AMPLATZER™ Amulet™ was approved in 2021. 2 These devices prevent systemic embolization by closing off the LAA, which is considered the source of thrombus formation in more than 90% of the cases. 3 The LAA is an actively contracting structure and more distensible than the rest of the atrium and should not be considered to be merely a superfluous appendix and a source of emboli. 4 It is rich in stretch receptors and more sensitive to pressure fluctuation than the left atrial (LA). 5 Also, the LAA has an essential endocrine function as it is a storage for the atrial and brain natriuretic peptides (BNPs). 6 Those characteristics allow the LAA to play a role in the pressure regulation inside the left atrium and fluid balance by enhancing diuresis. Therefore, the LAAC may theoretically affect LA hemodynamics. We, therefore, studied the impact of LAAC on the LA hemodynamics through conducting a meta-analysis of published literature.

| Data source and search strategy
The present meta-analysis was performed in accordance with Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines and the Cochrane handbook. 7 We systematically searched MEDLINE/PubMed, EMBASE, and Google scholar from inception to November 30, 2021. The search included the following key terms: (("left atrial function") AND ("left atrial appendage closure" OR "left atrial appendage exclusion")) AND ("echocardiography" OR "speckle tracking echocardiography"). We further reviewed the references list of the included articles in this review to include other relevant studies.

| Eligibility criteria
We included articles that demonstrate the impact of LAA exclusion on LA functions assessed by speckle tracking echocardiography.
Articles using other methods other than speckle echocardiography for assessment were excluded from our analysis. In-vitro studies, non-English papers, reviews and data that could not be extracted were also excluded.
2.3 | Outcome of interest, data extraction, and quality assessment While the primary outcome in our study is to assess LA reservoir function and LA contractile function, the secondary outcome is to assess LA systolic function by strain rates. LA reservoir function is assessed by LAEF and peak atrial longitudinal strain (PALS), and LA contractile function is assessed by peak atrial contraction strain (PACS).
Initial title and abstract screening were conducted by two reviewers (M.R.M and M.M) and all disagreements were discussed to reach a consensus, otherwise, a third opinion from W.S was obtained.
Potentially eligible articles were imported for full-text review and assessed for inclusion. We extracted data using an Excel sheet.
We assessed the quality of the included studies using the Newcastle-Ottawa Scale for cohort studies, as shown in Table S1. For Newcastle-Ottawa Scale, each asterisk counts as one point. 8 The maximum points are two for comparability and one for all other categories (Table S1). Each star adds to the total score. A score of <5 is considered low quality, 5-6 is medium quality, while 7-9 is high quality. In the included studies, two were low quality, four were medium quality and one was of high quality. We did not perform funnel plots for publication bias since the number of the included studies is <10 in our analysis. 9

| Data synthesis and analysis
We used the random-effects model with 95% confidence intervals (CIs) to calculate the standardized mean difference (SMD) from the means and standard deviations. The means and standard deviations were calculated as described by Wan et al. 10 when unavailable in the selected studies. We used the I 2 to measure heterogeneity over the included studies (<25% considered low heterogeneity, and >50% considered significant heterogeneity. 11 Analyses were performed using R Studio Version 3.6.3.

| Summary of studies
A total of 247 articles were screened from a comprehensive electronic database search. After thorough review, we found 8 retrospective or prospective observational studies (one of them was only published as an abstract) that evaluated the LA function after LAAC. [12][13][14][15][16][17][18][19] The search process is detailed in (Figure 1). The pertinent details of the included studies are illustrated in (Table 1) (Table 2).

| Outcomes
There was significant increase in LA emptying fraction following

| Heterogeneity evaluation and sensitivity analyses
Heterogeneity was noted to be high in most of our studied outcomes.
Baujat plots were utilized to evaluate this heterogeneity. Leave-oneout analyses were performed to detect any influential effects of the included studies, especially those with the highest contribution to heterogeneity.
− Total atrial emptying fraction (TAEF): Yang et al. 13 was the study of most contribution to heterogeneity. Omitting Yang et al. 13 dropped I 2 from 75% to 31% without significant changes to the results. Omitting Murtaza et al., 16 Ijuin et al., 14 or Coisne et al. 15 makes the results nonsignificant ( Figure S3A).
− LA volume: Coisne et al. 15 was the study with most contribution to heterogeneity. However, with omitting it I 2 dropped only from 92% to 85%. Leave-one-out analyses did not show any significant changes in the results ( Figure S3B).  13 was also the study with the highest contribution to heterogeneity. When we excluded it, the I 2 dropped from 89% to 14%. Leave-one-out analyses of the studies evaluating this outcome did not show any significant changes of the results ( Figure S3C).
− PACS: Yang et al. 13 was also the study with the highest contribution to heterogeneity. When we excluded it, the I 2 dropped from 81% to 64%. Leave-one-out analyses of the studies evaluating this outcome did not show any significant changes in the results ( Figure S3D).
− Strain during ventricular systole and diastole: Yang et al. 13 was also the study with the highest contribution to heterogeneity.
When we excluded it, the I 2 dropped from 89% to 63% for the ventricular systole and from 65% to 11% for the strain during ventricular diastole. Leave-one-out analyses of the studies evaluating these outcomes did not show any significant changes in the results ( Figure S3F,G).

| DISCUSSION
Our report investigates differences in LA function following LAAC in 8 studies including 260 patients. We report that following LAAC, LAEF is inversely proportional to LA volume, which decreases after LAA exclusion. 23 Tabata et al. 24    or fibrotic changes that impede LA contractility function. 28 Exhausted frank starling mechanism in the setting of mitral regurgitation or chronic increase in preload is also a potential cause of lack of contractile function improvement after LAA closure which shifts LA pressure-volume curve downward and to the left. 29 Last, the Data included in our analysis exhibits significant heterogeneity, which might be due to the differences in the study population. Variable follow-up durations have been identified as well.
Different techniques for LAA closure have been used, including Watchman, Lariat, and ACP devices. More studies are needed to compare the efficacy of various LAAC devices on LA deformation.

| CONCLUSION
LAA exclusion is associated with improvement of LA reservoir function assessed by TAEF. PALS and LA contractile function did not differ significantly after LAA exclusion. More research is warranted for better understanding of LA hemodynamics and its implications in clinical vignette.

CONFLICTS OF INTEREST
The authors declare no conflicts of interest.

DATA AVAILABILITY STATEMENT
The data that supports the findings of this study are available in the supplementary material of this article.