In order to best care for our patients, it is important to understand the true risk of thromboembolic events post cardioversion based on whether anticoagulation is prescribed or not.
Paper: Wong BM et al. Thromboembolic Events Following Cardioversion of Acute Atrial Fibrillation and Flutter: A Systematic Review and Meta-Analysis. CJEM 2021. PMID: 33715143 .
Clinical Question: What is the effect of oral anticoagulation use on thromboembolic events at 30 days following cardioversion of acute atrial fibrillation and flutter?
What They Did:
- A systematic review and meta-analysis conducted using the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) and CHARMS (Checklist for Critical Appraisal and Data Extraction for Systematic Reviews of Prediction Modelling Studies) guidelines for the reporting of systematic reviews
- Search of EMBASE, Ovid MED-LINE, and Cochrane Central Register of Controlled Trials from inception to February 6, 2020
- Authors wanted to evaluate the effect of oral anticoagulation on thromboembolic events following cardioversion of low-risk acute atrial fibrillation and flutter patients (defined as those presenting with episodes of <48hrs)
- Primary analysis: Comparison of thromboembolic events based on oral anticoagulation use vs no oral anticoagulation use
- Secondary analysis: Based on baseline thromboembolic risk
- Definitions:
- Oral Anticoagulation = on oral anticoagulation before and after cardioversion OR started on anticoagulation after cardioversion
- No Anticoagulation = Neither anticoagulated before nor after cardioversion
- Low Baseline Risk for Thromboembolism = CHADS2 score = 0 OR CHA2DS2-VASc ≤1
- High Baseline Risk for Thromboembolism = CHADS2 score ≥1 OR CHA2DS2-VASc ≥2
Inclusion / Exclusion:
- Inclusion:
-
- Adults ≥18 years of age
- Acute atrial fibrillation and flutter (i.e. clear symptom onset <48hr, clear onset <7d and receiving appropriate anticoagulation, or clear onset <7d with no left atrial thrombus on TEE)
- Undergoing electrical or pharmacological cardioversion
- Reported thromboembolic events (i.e. stroke, TIA, or systemic thromboembolism) within 30d following cardioversion
- Observation studies, case series or RCTs
- Only English records with full-text articles
- Exclusion
- Commentaries, editorials, letters, case reports systematic reviews, meta-analyses, clinical guidelines, laboratory data, conference abstracts, and studies on patients with valvular heart disease or those without recent onset atrial fibrillation or flutter (i.e. permanent or persistent)
Outcomes:
- Primary: Thromboembolic events (i.e. stroke, transient ischemic attack, or systemic thromboembolism) at 30 days following cardioversion of acute atrial fibrillation and flutter.
- Composite outcome of stroke, transient ischemic attack (TIA), heart failure, acute coronary syndrome, death
Results:
- Initial search 717 citations
- Duplicates removed, 79 articles reviewed
- 59 articles excluded
- Wrong patient population N = 22
- Insufficient data N = 14
- Supplement/abstract N = 14
- Non-applicable outcome measure N = 5
- Duplicate data N = 3
- Incorrect study design N = 1
- 20 studies analyzed (met inclusion criteria)
- 8 prospective cohort
- 8 retrospective cohort
- 4 randomized trials
- 13/20 studies exclusively evaluated patients with acute atrial fibrillation and flutter of <48hrs duration
- Remaining 7 studies included an onset time of <48hrs and >48hrs on anticoagulation
- Overall incidence of thromboembolic events at 30d post-cardioversion was 0.42% (61 events/14,410 cardioversions)
- Primary analysis was conducted on seven studies (2 prospective cohort, 2 randomized trials, and 3 retrospective) reporting thromboembolic events at 30 days and considered high-quality according to the Quality in Prognostic Studies (QUIPS) tool
- Meta-analysis of these seven studies revealed insufficient evidence regarding the effect of oral anticoagulation use on thromboembolic events post-cardioversion
- RR = 0.82, 95% CI 0.27–2.47
- RR<1 suggests decreased risk with oral anticoagulation use
- The rate of thromboembolic events was low in both groups, irrespective of whether oral anticoagulation was used or not.
- Oral anticoagulation group: 3 thromboembolic events in 1380 cardioversions
- No oral anticoagulation: 6 events in 1788 cardioversions.
- Meta-analysis of these seven studies revealed insufficient evidence regarding the effect of oral anticoagulation use on thromboembolic events post-cardioversion
- All 9 thromboembolic events in the primary analysis were in patients who had an onset of < 48 h, received electrical cardioversion, and had a CHA2DS2-VASc score ≥ 2
- NO thromboembolic events at 30d post-cardioversion in 783 cases not receiving oral anticoagulation within the prospective subgroup
- Secondary analysis
- 13 studies (3 prospective cohort, 3 randomized trials, and 7 retrospective) were included in the analysis comparing high versus low baseline risk of thromboembolism according to CHADS2 and CHA2DS2-VASc scores
- High baseline thromboembolic risk was 0.63% (42 events/6706 cardioversions).
- Low baseline risk group was 0.19% (10 events/5285 cardioversions).
- There was an increased risk of thromboembolic events in those at high baseline risk with a CHADS2 score ≥ 1 or CHA2DS2-VASc score ≥ 2 (RR = 2.25, 95% CI 1.25–4.04; I2 = 0%)
- RR >1 indicates increased risk with higher CHADS2 or CHA2DS2-VASc scores
Strengths:
- 1st systematic review and meta-analysis to review the effect of oral anticoagulation use on thromboembolic events within 30 days post-cardioversion of acute atrial fibrillation and flutter
- For studies with incomplete data, 2 attempts were made to obtain further information from the corresponding author. Study excluded if relevant data not received.
- Asks a clinically important question on a topic that has various recommendations from different organizations
- Rigorous study selection using PRISMA and CHARMS guidelines
- Calculated kappa statistic to determine agree for full-text review
- Used Quality in Prognostic Studies (QUIPS) tool to assess risk of bias of individual studies
- Primary analysis was restricted to studies with a low risk of bias
- 2 reviewers assessed for bias
- Performed multiple sensitivity analyses to corroborate the findings of the primary outcome
- Heterogeneity amongst articles in the primary analysis was low (0%)
- Employed a medical librarian to help find relevant articles
- Evaluated a total of 717 citations for study
Limitations:
- Of the 20 studies that were included in the review, 5 were identified by a manual search of reference lists (although these 5 studies were oler and not included in meta-analysis portion), what other studies could have been missed?
- Two researchers reviewed articles; one was consistent and the other variable. This could lead to inconsistencies in study selection, and quality assessment
- Given the lack of high-quality evidence on this topic, risk vs benefit should be weighed. There are no outcomes of bleeding risk with and without anticoagulation presented in this study
- The analysis found few thromboembolic events in both non-anticoagulated and anticoagulated patients, the scarcity of events makes definitive conclusions difficult to make
- 12 studies met inclusion criteria for the primary analysis, but of those 5 were considered low quality and not used in data analysis further limiting results. This reduces the risk of bias in the studies evaluated, but can further limit the data presented on a topic that is already lacking high-quality evidence
- Unable to perform pooled analyses based on the duration of acute atrial fibrillation and flutter onset due to insufficient data
- Only studies in English were included in the studies
Discussion:
- A low event rate in the primary analysis of all comers makes it difficult to determine the value of instituting oral anticoagulation on thromboembolic events post-cardioversion. However, in patients with CHADS2 score ≥1 and CHA2DS2-VASc score ≥2 the risk of thromboembolic events was two-fold greater when compared to those with a CHADS2 score of 0 or CHA2DS2-VASc score of 0 or 1
- Despite a low quality of evidence to provide certainty, there appears to be a low-risk of post-cardioversion thromboembolic events in patients with low-risk acute atrial fibrillation and flutter. However, current evidence does not provide clear guidance on anticoagulation recommendations for low risk patients. This forces physicians to have shared decision making conversations with patients, but in patients who decide to forgo 4 weeks of anticoagulation, the low event rate should provide greater reassurance.
- The 2018 Canadian Cardiovascular Society Guideline (2) recommends that all patients, regardless of risk criteria be anticoagulated, while the 2021 European (4) and 2019 American guidelines (3) only recommend anticoagulant those with CHA2DS2-VASc score ≥2. This meta-analysis and systematic review seeks to clarify the rate of thromboembolic events post-cardioversion and, in doing so, seeks to unify opinion on anticoagulation post cardioversion. The meta-analysis of the seven studies revealed insufficient evidence regarding the effect of oral anticoagulation use on thromboembolic events post-cardioversion.. In order to get definitive data a large randomized controlled trial powered to detect a difference would be required.
Author’s Conclusion: “Primary analysis revealed insufficient evidence regarding the effect of oral anticoagulation use on thromboembolic events post-cardioversion of low-risk acute atrial fibrillation and flutter, though the vent rate is low in contemporary practice. Our finding can better inform patient-centered decision-making when considering 4-week oral anticoagulation use for acute atrial fibrillation and flutter patients.”
Clinical Bottom Line:
This systematic review and meta-analysis tells us that thromboembolic events in patients with acute onset atrial fibrillation and flutter post-cardioversion are exceedingly low at 30 days. There is insufficient evidence regarding the value of oral anticoagulation in low-risk acute afib and aflutter patients post-cardioversion. However, based on this limited evidence it does appear as though the event rate in those not receiving oral anticoagulation post-cardioversion is very low. More importantly, I’m reluctant to deviate from the recommendations from both the American and European guidelines and will continue to anticoagulant those with higher risk (CHADS2 ≥1 and CHA2DS2-VASc ≥2).
References:
- Wong BM et al. Thromboembolic Events Following Cardioversion of Acute Atrial Fibrillation and Flutter: A Systematic Review and Meta-Analysis. CJEM 2021. PMID: 33715143
- Andrande et al. 2018 Focused Update of the Canadian Cardiovascular Society Guidelines for the Management of Atrial Fibrillation. Can J Cardiol 2018. PMID: 30404743
- January CT et al. 2019 AHA/ACC/HRS Focused Update of the 2014 AHA/ACC/HRS Guideline for the Management of Patients With Atrial Fibrillation: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines and the Heart Rhythm Society. J Am Coll Cardiol 2019. PMID: 30703431
- Hindricks G et al. 2020 ESC Guidelines for the diagnosis and management of atrial fibrillation developed in collaboration with the European Association for Cardio-Thoracic Surgery: The Task Force for the diagnosis and management of atrial fibrillation of the European Society of Cardiology (ESC) Developed with the special contribution of the European Heart Rhythm Association (EHRA) of the ESC. Eur Heart J 2020 PMID: 32860505
Post Peer Reviewed By: Anand Swaminathan, MD (Twitter: @EMSwami ) and Salim R. Rezaie, MD (Twitter: @srrezaie)
The post Thromboembolic Events After Cardioversion of Acute Atrial Fibrillation and Flutter: Should we Anticoagulate? appeared first on REBEL EM - Emergency Medicine Blog.