Incidence and Risk Factors of Relapse After Elective Electrical Cardioversion for Atrial Fibrillation: a Protocol for Systematic Review and Meta-analysis


 BackgroundAtrial Fibrillation (AF) is the most frequent cardiac arrhythmia and it is associated with life-threatening complications such as hemodynamic instability and thromboembolic events. Electrical cardioversion remains its mainstay of treatment and can be performed either urgently or electively. Relapse after electrical cardioversion has been reported by several primary studies with divergent results, but no detailed summary exists for a critical appraisal of its global incidence and risk factors.Therefore, we propose the first systematic review and meta-analysis to synthesize the existing data on the global incidence and risk factors of relapse after electrical cardioversion for AF.MethodsWe will include case-control and cohort studies reporting on the incidence and risk factors of relapse after electrical cardioversion for atrial fibrillation. Electronic databases including PubMed, Embase, WHO Global Health Library and Web of Science will be searched for relevant records published between January 01, 2000, and December 15, 2020. Pairs of independents reviewers will perform study selection and data extraction, as well as an assessment of the methodological quality of included studies. Appropriate meta-analysis will then be used to pool studies judged to be clinically homogenous. Egger’s test and funnel plots will be used to detect publication bias. Findings will be reported and compared by the human development level of countries. ResultsThis systematic review and meta-analysis to synthesize the existing data on the global incidence and risk factors of relapse after electrical cardioversion for AFConclusionThis review is expected to provide relevant data to help in evaluating the burden and risk factors of relapse after electrical cardioversion for atrial fibrillation. The overall findings of this research will be published in a peer-reviewed journal. Systematic Review RegistrationThe protocol has been registered in Prospero with this registration number: CRD42020209301


Abstract
Background Atrial Fibrillation (AF) is the most frequent cardiac arrhythmia and it is associated with life-threatening complications such as hemodynamic instability and thromboembolic events. Electrical cardioversion remains its mainstay of treatment and can be performed either urgently or electively. Relapse after electrical cardioversion has been reported by several primary studies with divergent results, but no detailed summary exists for a critical appraisal of its global incidence and risk factors.Therefore, we propose the rst systematic review and meta-analysis to synthesize the existing data on the global incidence and risk factors of relapse after electrical cardioversion for AF.

Methods
We will include case-control and cohort studies reporting on the incidence and risk factors of relapse after electrical cardioversion for atrial brillation. Electronic databases including PubMed, Embase, WHO Global Health Library and Web of Science will be searched for relevant records published between January 01, 2000, and December 15, 2020. Pairs of independents reviewers will perform study selection and data extraction, as well as an assessment of the methodological quality of included studies.
Appropriate meta-analysis will then be used to pool studies judged to be clinically homogenous. Egger's test and funnel plots will be used to detect publication bias. Findings will be reported and compared by the human development level of countries.

Results
This systematic review and meta-analysis to synthesize the existing data on the global incidence and risk factors of relapse after electrical cardioversion for AF

Conclusion
This review is expected to provide relevant data to help in evaluating the burden and risk factors of relapse after electrical cardioversion for atrial brillation. The overall ndings of this research will be published in a peer-reviewed journal.

Systematic Review Registration
The protocol has been registered in Prospero with this registration number: CRD42020209301 Background Atrial Fibrillation (AF) is a supraventricular tachyarrhythmia characterized by uncoordinated atrial activation with consequent deterioration of atrial mechanical function(1). The two major consequences are hemodynamic instability and thromboembolic events. AF is the most common sustained cardiac rhythm disturbance which increases with aging (1); observed in 3% of adults aged 20 years and above (2) (3). More representative regional gures report prevalence rates varying between 1% and 4% in Australia, Europe and the USA and an estimated prevalence varying between 0.49%and1.9% in Asia. AF prevalence is highest in the Caucasian race. In Western Europe, Australia and North America 70% of people with AF are aged > 65 years, whereas the mean age of AF patients in other geographical locations is often lower (4). AF can be classi ed as a rst-detected episode, whether or not it is symptomatic or self-limiting.
This implies that there can be uncertainty about the duration of the episode and previous undetected episodes. When a patient has had two or more episodes, AF is considered recurrent. Once terminated, recurrent AF is categorized as paroxysmal, and when sustained, persistent. In the latter case, termination by pharmacological therapy or electrical cardioversion does not change its terminology. Persistent AF (> 7 days) can be either the rst presentation or a culmination of recurrent episodes of paroxysmal AF.
Persistent AF includes cases of long-standing AF (greater than one year), in which cardioversion is not indicated or has not been attempted, usually leading to long-lasting AF (1). Management of AF includes anticoagulation according to the CHA 2 DS 2 -VASc score, heart rate regulation and cardiac rhythm control strategies (1)(5). One of the main means to control the cardiac rhythm during AF is through electrical cardioversion or direct-current cardioversion. This entails an electrical shock synchronized with the intrinsic activity of the heart(1). The rate of relapse after elective cardioversion is quite high and reported between 40 to 50%(6) and is in uenced by the duration of AF, advanced age, female gender, tobacco abuse, chronic alcoholism, obesity,hypertension, obstructive sleep apnoea, diabetes mellitus, chronic obstructive pulmonary disease, renal impairment, hyperthyroidism, structural heart disease, left atrial size, heart failure and C-reactive protein (CRP) level (7)(8)(9).
Although the prevalence/incidence and risk factors of relapse after electrical cardioversion for AF has been widely reported by several isolated primary studies illustrating controversial ndings, no detailed and comprehensive overview using rigorous methodological procedure has been carried out to date to scale up the evidence to enable decision making. Therefore, we propose the rst systematic review and meta-analysis to critically synthesize the existing data on the global incidence and risk factors of relapse after electric cardioversion for atrial brillation.

Review question
What are the global incidence and the risk factors of relapse after electrical cardioversion for atrial brillation?

Objective
The systematic review and meta-analysis aim to determine the incidence and risk factors of relapse after electric cardioversion for atrial brillation Methods This review is reported in accordance with the Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols 2015 Guidelines(10) applicable to observational studies. This is illustrated in more detail in online additional le 1.

Types of studies
We will include case-control and cohort studies. Commentaries, editorials, letters and reviews will not be considered. Studies with inaccessible full-texts either online or after contacting its corresponding author will be excluded.

Types of patients
We will consider studies including adults and children.

Type of study settings
We will consider hospital-based studies in both rural and urban areas.

Types of outcomes
The incidence of relapse after elective cardioversion for AF will be de ned as the number of individuals who developed a new episode of AF after a planned electric shock between January 01, 2000, and December 15, 2020. Risk factors will be derived from primary reported data. Data on pharmacological cardioversion and urgent cardioversion procedure will be excluded.

Search strategy for identifying relevant studies
We will search the following bibliographic databases: PubMed, Embase, WHO Global Health Library and Web of Science for relevant records published in English or French between January 01, 2000, and September 15, 2020. The designed search strategy for PubMed using both text words and medical subject heading (meSH) terms related to electric cardioversion, direct current cardioversion, electric shock, atrial brillation, relapse, recurrence, failure is provided in Table 1. This search strategy will be adapted to t other databases.

Searching for other sources
We will scan the references of all relevant articles for additional data sources missed during our search strategy, and their full-texts will be sorted. Citations of important reviews will also be scanned. Lastly, the search strategy will extend to include grey literature from conference proceedings, book chapters, theses, government and non-governmental organizations' reports.
Selection of studies for inclusion in the review Two reviewers (MNT and JNT) will independently evaluate the records obtained from the searching process, with the aid of an evaluation form to ensure a reliable application of the selection criteria. These reviewers will screen the titles and abstracts of records obtained. Next, the full-texts of potentially eligible articles will be retrieved by at least one author. The two reviewers will independently review the full text of each potentially eligible article, compare their ndings and resolve any discordance by the arbitration of a third author (CN). For duplicates articles, only the study reporting the largest sample size will be considered.

Data extraction and management
A data extraction form will be used by two pairs of independent reviewers (MNT, JNT, CN and CT), to collect information on the last name of the rst author, year of publication, region and country of recruitment, human development index ranking of country economic level, study area (rural vs. urban), period of participants' inclusion, study setting(cardiology department, emergency unit, intensive care unit), age groups (children, adult, elders), number of sites, study design, mean or median age, sampling method(probabilistic sampling versus non probabilistic sampling), sample size, number of cases of AF, types of AF (paroxysmal or persistent), timing of data collection (prospective or retrospective), male proportion, speci c characteristics of the study population (cardiomyopathy, coronary heart disease, congenital heart disease, valvular heart disease or any other speci c condition), incidence rate of relapse, risk factors of relapse. For multicenter studies conducted in different countries, the incidence and risk factors will be reported separately for the individual countries.Countries of recruitment of participants will be categorized according to WHO regional classi cation: Africa, The Americas, South-East Asia, Europe, Eastern Mediterranean, and Western Paci c.

Data synthesis and analysis
After data collection, a meta-analysis will be conducted where there will be clinical homogeneity based on the pro le of the population. Unadjusted SE, the incidence for the study-speci c estimates will be resumed based on the crude information of the numerators, and denominators provided by each study. Sub-group analyses will be performed by separate pooling of studies conducted on speci c groups of patients like cardiomyopathy, coronary heart disease,congenital heart disease andvalvular heart disease. To maintain the effect of studies with extremely small or large estimates on the overall estimate to a minimum, the variance of the study-speci c incidence/risk factors will be stabilized with the Freeman-Tukey double arcsine transformation19, before pooling the data using a random-effects meta-analysis model. Heterogeneity will be assessed using the χ2 test on Cochrane's Q statistic and quanti ed by calculating I 2 (11).Values of 25%, 50% and 75% for I 2 will, respectively, represent low, medium and high heterogeneity. We will assess the presence of publication bias using funnel plots inspection and Egger's test (12). Where substantial heterogeneity will be detected, a meta-regression and sub-group analyses will be performed to investigate the possible sources of heterogeneity using the aforementioned variables and the study methodological quality. In case of substantial clinical heterogeneity, a narrative summary of our ndings will be done. The inter-rater agreement for study inclusion between investigators will be assessed using Cohen's k coe cient (13).Data analyses will be done using the 'meta' package of the statistical software R (V.3.5.1, The R Foundation for Statistical Computing, Vienna, Austria).

Presentation and reporting of results
The study selection process will be summarized in a ow diagram. Quantitative data will be presented in evidence tables of individual studies as well as in summary tables and forest plots where appropriate.
The quality scores and risk of bias for each eligible study will be reported accordingly. This may be tabulated and accompanied by narrative summaries.

Conclusion
This review is expected to provide relevant data to help in evaluating the burden and risk factors of relapse after electrical cardioversion for atrial brillation. The overall ndings of this research will be published in a peer-reviewed journal. The authors have no speci c grant to declare for this research from any funding agency in the public, commercial or not-for-pro t sector.
-Authors' contributions : MNT, JB and JNT had the idea, designed and conceived the protocol. MNT wrote the rst draft of the manuscript. MNT, AM and SK are the guarantors of the review. All authors critically revised the methodology and intellectual content and approved the nal version of this manuscript.
-Acknowledgements : Not applicable Table   Table 1 Search strategy on Pubmed