Does the choice of drug in pharmacologic cardioversion correlate with the guidelines? Systematic review

Background. Atrial fibrillation (AF) is the most common sustained arrhythmia, the most common cause of supraventricular tachycardia in the global population and the most common arrhythmia requiring treat-ment in an emergency department. Objectives. To systematically review recent literature and quantify the correlation between the choice of pharmacological cardioversion (PCV) drug and the national or international guidelines. Materials and methods. A systematic review was performed in accordance with the PRISMA statement methodology. The PubMed search engine was used to search for articles regardless of type or language and published in the last 6 years (May 2014–May 2020). In addition, we searched for AF guidelines and recommendations published online by cardiology and emergency medicine societies. Results. The search strategy returned a total of 2615 abstracts. A total of 2598 full texts were screened; 2540 full texts were excluded with reasons and 58 articles from 32 countries were included in the analysis. In 17 of the 58 articles (29%), we noted discrepancies with the AF guidelines, specifically regarding the PCV drug used, the patients’ comorbidities and the contraindications associated with the PCV drug. The most common clinical situation for the use of a contraindicated drug was when ibutilide was administered to patients with heart failure. The analysis did not reveal any statistically significant correlations, although the correlation between the sample size and guideline adherence was close to statistical significance (p < 0.06). Conclusions. Our systematic analysis revealed substantial non-adherence to AF treatment guidelines.


Background
Atrial fibrillation (AF) is the most common sustained arrhythmia and the most common cause of supraventricular tachycardia in the world. 1,2Furthermore, acute AF is a common complaint among emergency department (ED) patients and is the most common arrhythmia requiring treatment in the ED. 3 It commonly occurs because AF is often caused by common diseases (see Table 1).However, ongoing academic discussions seek to answer whether a patient with AF who does not have any cardio-pulmonary disease should be diagnosed with "lone AF". 4 According to the latest AF guidelines published by the European Society of Cardiology (ESC), the term/diagnosis of "lone AF" should not be used because AF always has an underlying cause. 5ere are 2 widely accepted and separate goals of AF treatment: rate control and rhythm control.In the case of paroxysmal AF, a clinician has a choice of 2 methods to restore sinus rhythm (SR): pharmacological (chemical) cardioversion (PCV) or electric cardioversion (ECV).According to a large international emergency physician survey, PCV is the first line of treatment for recent-onset AF. 6 The efficacy of PCV in restoring sinus rhythm varies among published studies and is subject to ongoing debate.
When deciding to perform PCV, clinicians have several antiarrhythmic drugs to choose from, which are listed in national and international guidelines (Table 2).Little is known about adherence to AF guidelines when it comes to PCV, particularly in the ED.

Objectives
The aim of this study was to systematically review the most recent literature in an attempt to answer the following clinical question: Do recently published articles about PCV reveal any correlation between the choice of PCV drug and national or international guidelines?

Materials and methods
A systematic review was performed in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement methodology. 7he PubMed search engine was used to find articles regardless of type or language and published in the last 6 years (May 2014-May 2020).The unusual six-year timespan was purposefully chosen because the American Heart Association (AHA)/American College of Cardiology (ACC)/Heart Rhythm Society (HRS) and National Institute for Health and Care Excellence (NICE) guidelines were published in December and August of 2014, respectively. 6,7he following search terms were applied: atrial fibrillation AND pharmacological cardioversion AND antazoline OR amiodarone OR dronedarone OR flecainide OR ibutilide OR procainamide OR propafenone OR vernakalant.
Data were extracted from the articles and entered into Excel spreadsheets (Microsoft Office 2007; Microsoft Corp., Redmond, USA) and subsequently exported to STA-TISTICA v. 12.0 (StatSoft Inc., Tulsa, USA) for analysis.The following statistical tests were performed: Mann-Whitney U test (for continuous variables) and Fisher's twotailed test (for categorical variables).Values of p <0.05 were considered statistically significant.
In addition, we searched for AF guidelines and recommendations published online by cardiology and emergency medicine societies.Our search returned guidelines from Australia (National Heart Foundation of Australia (NHFA)/ Cardiac Society of Australia and New Zealand (CSANZ)), 8 Canada (Canadian Association of Emergency Physicians (CAEP), Canadian Cardiovascular Society (CCS)), 9,10 Europe (European Resuscitation Council (ERC), ESC) 5,11 UK (NICE), 12 and USA (American College of Emergency Physicians (ACEP), AHA/ACC/HRS) (Table 2). 13,14We used these recommendations as a reference point to answer the research question described earlier.

Results
Our search returned 58 articles from 32 countries; most articles were published in 2017-2018 (Fig. 2,3).  Unfounately, not all relevant data was provided by the authors, thus making it impossible to perform a full meta-analysis.Detailed results of the systematic review are summarized in Table 3 (Fig. 2,3).
Despite the incomplete data, the analyzed articles revealed a surprising trend of non-adherence to AF treatment guidelines.49,55,60,63 According to the data presented in the articles, it appeared that a total of 239 patients underwent PCV using a drug that was contraindicated given their specific comorbidities.In the described cases, the most common culprit PCV drug was ibutilide, followed by vernakalant, amiodarone, propafenone, and flecainide.The most commonly described clinical situation for the use of contraindicated drug was ibutilide when administered to a patient with HF, which is contraindicated according to the ACEP, CAEP and ESC guidelines (Table 2). 10,12,13,31,39,55,63n 9 of the 17 articles, using a contraindicated drug during PVC was performed in the ED (Table 4). 21,22,26,27,32,36,39,55,63ue to incomplete data, it was impossible to assess whether an additional 338 patients were administered a PCV drug that was contraindicated or not. 21,36(Table 4).Analysis using the Mann-Whitney U test and Fisher's test did not reveal any statistically significant correlations between adherence to AF guidelines and demographic variables such as sample size, patient age, and male sex (Table 5).However, it is noteworthy that the correlation between the sample size and guideline adherence was close to statistical significance (p < 0.059).It appears that the larger the sample size, the less adherence was observed.The analysis using Fisher's two-tailed tests did not reveal any statistically significant correlations between adherence to AF guidelines and the type of study/article, region/country or department where the PCV was performed (Table 5). [76][77][78]

Discussion
Although we found articles describing PCV performed on all of the inhabited continents of the world, we are aware that they do not necessarily reflect daily clinical practice.The articles we analyzed did not contain enough data to answer the question why the AF guidelines were not followed.We do not want to speculate about the particular authors' intent or the circumstances during the described PCV.However, given our institutional experience with PCV, we can think of several possible reasons, most of which are rather mundane or perhaps even temporary, e.g., the availability of antiarrhythmic drugs, institutional/personal experience with particular drug(s), and interest in comparing the efficacy of a new drug (e.g., vernakalant) compared to a "tried and tested" drug.
The very same issue of non-adherence with AF guidelines was addressed in the literature, although the answers   were not definitive.Authors suggested reasons such as lack of quality evidence (see Table 2 for information about the level of evidence in the analyzed AF guidelines), impossibility to establish AF onset, concerns about thromboembolic events, concerns about negative inotropic or proarrhythmic effect of PCV drugs, time constrains (excluding secondary causes of AF is time-consuming and adds more complexity to decision-making), and the fact that a significant number of ED patients with AF spontaneously revert to SR. 35,45,49,[78][79][80][81] Finally, patient preference, or perhaps the physician's attitude, towards a given therapeutic option may influence the decisions about adopting a wait and observe approach or rhythm control or rate control, as well as electrical or pharmacological cardioversion. 46n a survey of 561 physicians, Heidbuchel et al. found 8 major barriers to AF guidelines implementation that were knowledge-related (e.g., diagnosing AF based on duration instead of etiology, uncertainty during decision-making, use and interpretation of risk assessment scores, difficulties in choosing stroke prevention treatment), skill-related (e.g., difficulties in EKG interpretation/detection of AF, difficulties in discussing with patients their treatment strategy) and systemic (e.g., poor cooperation between specialists and general practitioners, local regulations regarding the use of novel anticoagulants). 82

Limitations
Our systematic review had several limitations, most notably, the high heterogeneity and incompleteness of the obtained data which did not allow us to perform a meta-analysis.Specifically, we were unable to extract enough data about the patients (e.g., patient age is provided only as an average value, comorbidities listed as totals without mention if any patients had more than 1 comorbidity).Therefore, it was not possible to assess if AF guidelines were followed during PCV of those patients.Furthermore, although reports of single cases are universally defined as weak evidence, we had little data to choose from and decided to include them in the analysis.Had there been more data from large trials available, we would have chosen them over case reports, thus making our statistical analysis and conclusions more robust.Finally, we are aware that there might be national AF guidelines which we were unable to find.

Conclusions
Our review of the published clinical literature about PCV reveals significant non-adherence to AF treatment guidelines.Specifically, the drugs used for PCV in patients with AF and comorbidities such as heart failure and thyroid disease are inconsistent with the guidelines.

Fig. 1 .
Fig. 1.Flowchart of the literature search strategy

Fig. 2 .
Fig. 2. Number of analyzed articles describing pharmacological cardioversion in the particular country's patient population (illustration by Zu).

Fig. 3 .
Fig. 3. Number of analyzed articles published in a given year

Table 1 .
Etiology of atrial fibrillation (AF) (according to Benjamin et al. and Kirchhof et al.)

Table 2 .
Pharmacological cardioversion recommendations published in national and international guidelines Drug

Table 3 .
Detailed results of the systematic review

Table 3 .
Detailed results of the systematic review -cont.

Table 3 .
Detailed results of the systematic review -cont.

Table 3 .
Detailed results of the systematic review -cont.

Table 3 .
Detailed results of the systematic review -cont.

Table 3 .
Detailed results of the systematic review -cont.

Table 3 .
Detailed results of the systematic review -cont.

Table 3 .
Detailed results of the systematic review -cont.

Table 4 .
Articles describing PCV performed using a contraidicated drug HF/LV dysfunction received flecainide; 3 patients with HF/LV dysfunction received ibutilide, 4 patients with HF/ LV dysfunction received vernakalant (elsewhere in the article it is stated that patients with HF and "severely reduced left ventricular ejection fraction" were excluded) CAD -coronary artery disease; CER -cardiac emergency room; CMP -cardiomyopathy; ED -emergency department; EF -ejection fraction; HF -heart failure; HVD -heart valve disease; ICU -intensive care unit; LV -left ventricular; NYHA -New York Heart Association; PED -pediatrics department; THY -thyroid disease.