Cardiology · Medical Literature
Percutaneous left atrial appendage closure in non-valvular atrial fibrillation with concomitant heart failure: a systematic review and meta-analysis of periprocedural and long-term outcomes.
Abstract
Heart failure (HF) frequently coexists with non-valvular atrial fibrillation (NVAF) and may adversely affect outcomes after percutaneous left atrial appendage closure (LAAC). However, data on the safety and efficacy of LAAC in this high-risk subgroup remain inconsistent. A comprehensive search of PubMed, Cochrane Library, ScienceDirect, Epistemonikos, and MedRxiv was conducted from inception to August 2025 for studies comparing outcomes of LAAC in NVAF patients with versus without HF. Random-effects meta-analyses were performed to estimate pooled odds ratios (ORs) for periprocedural outcomes, and hazard ratios (HRs) or ORs for long-term outcomes, with 95% confidence intervals (CIs). Forty studies comprising 351,266 patients were included, with thirty-three contributing to quantitative synthesis. Compared with patients without HF, those with HF had higher odds of periprocedural major adverse events (OR 1.17, 95% CI 1.03-1.32), periprocedural mortality (OR 2.41, 95% CI 1.84-3.16), and major bleeding (OR 1.26, 95% CI 1.08-1.45). No significant differences were observed for periprocedural stroke or pericardial effusion. During long-term follow-up, HF was associated with increased all-cause mortality (HR 1.83, 95% CI 1.50-2.22), major adverse events (HR 1.21, 95% CI 1.00-1.46), and device-related thrombus (OR 1.45, 95% CI 1.17-1.79). Rates of long-term stroke, major bleeding, and peri-device leak were similar between groups. In NVAF patients undergoing LAAC, HF confers higher procedural risk and worse long-term survival, while protection against thromboembolism and major bleeding appears comparable. These findings underscore the need for careful patient selection, optimized procedural care, and individualized post-implant antithrombotic strategies.
Clinical relevance
In patients with non-valvular atrial fibrillation (NVAF) undergoing percutaneous left atrial appendage closure (LAAC), does concomitant heart failure (HF) independently increase the risk of periprocedural and long-term adverse outcomes compared with NVAF patients without HF? __SNAPSHOT__ {"studyDesign":"Systematic review and meta-analysis of observational and comparative studies","setting":"Multinational; databases searched include PubMed, Cochrane Library, ScienceDirect, Epistemonikos, and MedRxiv from inception to August 2025","sampleSize":"n=351,266 total patients across 40 included studies; 33 studies contributed to quantitative synthesis","population":"Adults with non-valvular atrial fibrillation (NVAF) undergoing percutaneous LAAC, stratified by the presence or absence of concomitant heart failure","intervention":"Percutaneous LAAC in NVAF patients with heart failure","comparator":"Percutaneous LAAC in NVAF patients without heart failure","primaryOutcome":"Periprocedural major adverse events, periprocedural mortality, major bleeding, periprocedural stroke, and pericardial effusion; long-term outcomes include all-cause mortality and additional endpoints (full list not available due to abstract truncation)","followUp":"Periprocedural (index hospitalization) and long-term follow-up (duration not reported in available abstract)","fundingSource":"Not reported"}
Key findings
Among 351,266 patients across 40 studies, patients with NVAF and concomitant HF undergoing LAAC demonstrated significantly higher periprocedural risk across multiple endpoints compared with those without HF. Periprocedural major adverse events were more frequent in HF patients (OR 1.17, 95% CI 1.03–1.32). Periprocedural mortality was substantially elevated (OR 2.41, 95% CI 1.84–3.16), representing the most clinically striking periprocedural finding. Major bleeding risk was also increased (OR 1.26, 95% CI 1.08–1.45). Periprocedural stroke and pericardial effusion did not differ significantly between groups (exact ORs and CIs not reported in available abstract). For long-term outcomes, HF was associated with increased all-cause mortality; however, the full set of long-term effect estimates (HRs or ORs, confidence intervals, and additional endpoints) are not available due to abstract truncation and cannot be reported. P-values, NNT, and NNH were not reported in the available abstract. All quantitative estimates are derived directly from the provided abstract; no values have been independently calculated.
Practice implications
This meta-analysis establishes that in NVAF patients undergoing percutaneous LAAC, concomitant HF is associated with meaningfully higher periprocedural mortality (OR 2.41) and major bleeding (OR 1.26), in addition to a modest but statistically significant increase in overall major adverse events (OR 1.17). The markedly elevated periprocedural mortality odds ratio suggests HF patients carry substantial procedural risk that warrants careful pre-procedural risk stratification and shared decision-making. The absence of a significant difference in periprocedural stroke or pericardial effusion is reassuring regarding procedure-specific mechanical complications in this group, but does not negate the overall excess hazard. Clinically, a 2.4-fold higher periprocedural mortality odds is likely clinically important, though absolute risk differences (not reported) are necessary to fully contextualize these findings for individual patients. The long-term outcome data are incompletely reported in the available abstract, limiting conclusions about whether HF patients ultimately derive meaningful stroke reduction or net clinical benefit from LAAC relative to the procedural risk assumed. This analysis does not establish whether LAAC should be withheld in HF patients, but supports the use of heightened caution, optimized pre-procedural HF management, and high-volume center referral. Patients with decompensated HF, very low ejection fraction, or prohibitive surgical risk may be particularly underrepresented. Results are derived predominantly from observational data subject to residual confounding. IMPORTANT NOTE: Only the partial abstract was available for this summary; long-term outcome effect estimates and several secondary endpoints were not retrievable due to abstract truncation, and this represents a significant limitation of this summary. __LIMITATIONS__ Abstract truncation: Long-term outcome data (full HRs/ORs, CIs, and endpoints beyond all-cause mortality) were not available, preventing complete evidence synthesis.,Study design: Predominantly observational studies included in the meta-analysis; residual confounding by HF severity, ejection fraction phenotype (HFrEF vs. HFpEF vs. HFmrEF), NYHA class, and comorbidity burden cannot be excluded.,Heterogeneity: Pooling random-effects ORs and HRs across diverse study designs, populations, and time periods introduces methodological variability; degree of statistical heterogeneity (I²) not reported in available abstract.,Surrogate and composite endpoints: 'Major adverse events' as a composite endpoint may obscure clinically distinct outcomes with differing prognostic implications.,Generalizability: HF subtype, device type (e.g., Watchman vs. Amulet), center volume, and operator experience likely vary substantially across included studies, limiting generalizability to specific clinical contexts.,Funding source not reported, precluding assessment of potential industry influence.,Publication bias: Inclusion of grey literature (MedRxiv) may partially mitigate this, but formal funnel plot assessment results are not available in the abstract.,Absolute risk differences not reported, limiting direct clinical applicability and NNT/NNH calculation.,Long-term follow-up duration not specified in available abstract. __BOTTOM_LINE_CATEGORY__ potentially_practice_informing __BOTTOM_LINE__ In NVAF patients undergoing percutaneous LAAC, concomitant heart failure is associated with a 2.4-fold higher odds of periprocedural mortality and significant increases in major bleeding and overall major adverse events compared to patients without HF. These findings support rigorous pre-procedural risk stratification, optimization of HF status prior to LAAC, and careful shared decision-making in this high-risk subgroup. Practice should not change based on this meta-analysis alone given its observational underpinnings, incomplete long-term data, and unresolved confounding, but it clearly identifies HF as a clinically important procedural risk modifier for LAAC.