1. Introduction
Atrial fibrillation (AF) is the most common sustained cardiac arrhythmia in the adult population globally, with an estimated prevalence between 2% and 4%[
1]. Both incidence and prevalence of AF are expected to increase further, due to population aging and increasing burden of other comorbidities, such as hypertension, heart failure, obesity, diabetes mellitus, coronary artery disease, and chronic kidney disease[2-7]. AF is associated with increased morbidity[
8], mortality[
9] and higher healthcare costs, rendering it a major socioeconomic burden[
10]. AF has progressively become a more common reason for emergency department (ED) visits, representing 0.5% of presenting reasons[
11].
Registry data have indicated that about 60% of AF patients who present to the ED are admitted, leading to an increase in AF-related hospital admissions due to the aforementioned rise in ED visits, emphasizing the need for more-efficient management of AF in the acute phase[11-14]. Several trials of AF treatment strategies, including cardioversion, have demonstrated that AF in ED can be treated safely and effectively, avoiding admission[11-26].
In the present study, the epidemiology and characteristics of AF patients presenting to the ED will be reviewed, as well as the impact of diverse management strategies on AF-related hospital admissions. Lastly, the design and initial data of the HEROMEDICUS protocol, which constitutes an electrophysiology-based aggressive rhythm control strategy in patients with AF in the ED setting will be presented.
4. AF Cost of Hospital Stay
The annual cost of AF treatment was estimated to be
$6.65 billion in 2005[
10] and later estimates were even higher[
22]. In 2007, inpatient charges for AF hospitalization accounted for 1.12% of
$530 billion and in 2014 this further increased to 1.22% of
$828 billion[
11]. There is a huge difference between charges spent for hospitalized AF patients and charges spent for those discharged from the ED. In 2007, annual charges for AF hospitalization were estimated at
$7.39 billion, increasing by 37% to
$10.1 billion in 2014. Per patient hospitalization charges increased from
$17.317 in 2007 to
$22.113 in 2014[
11].
A smaller European study examined the Impact on hospital resources of two different approaches to recent onset AF using a random sample of 300 patients with AF, collected from an ED electronic database from an urban community teaching hospital. Median charges for patients cardioverted and discharged from the ED were
$5.460 (IQR
$4.677-
$6.190) whereas median charges for admitted patients with no attempt at cardioversion at the ED were at least 4 times higher (
$23.202, IQR
$19.663-
$46.877)[
16].
5. Strategies for AF Management in the ED
There is no specific algorithm in ESC guidelines on management of hemodynamically stable patients presenting to the ED with primary AF, as well as guidance on when to admit or discharge them straight from the ED. There are multiple protocols that have examined the safety and efficacy of diverse strategies implemented in the ED for patients presenting with AF as the primary diagnosis. The basic goal of all different pathways was the reduction of hospital admissions.
Burton et al published in 2004 data on the feasibility of electrical cardioversion in the management of AF in the ED[
18]. In this multicenter cohort study the objective was to identify the outcomes and complications associated with electrical cardioversion in AF patients[
18]. Study population included 388 patients with recent-onset AF (<48hours) and electrical cardioversion was successful in 332 patients (86%). In total, the admission rate was as low as 14% since 333 patients were discharged home from the ED (301 after successful cardioversion and 32 with electrical cardioversion failure). Interestingly, 39 patients returned to the ED because of AF recurrence within 10 days[
18].
A multicenter observational cross-sectional study (HERMES-AF) of patients with recent-onset AF which was carried out in 124 Spanish EDs from May 23
rd to June 5
th, 2011, studied the impact of a rhythm control strategy on admission rates and symptoms[
24]. Inpatient admission rate was 14% in the rhythm control strategy. The study population was 421 patients of which 352 were allocated to the rhythm control strategy with rate control being chosen in 69 patients. In total, 60 (14.2%) patients were admitted. Control of symptoms was achieved in 95.2% of patients in the rhythm and in 88.4% of patients in the rate control group[
24].
In a prospective, 2-stage study at 2 tertiary care hospitals in the Northeastern United States, utilization of a multidisciplinary AF pathway was associated with a decrease in hospital admission rates[
25]. During the first stage of the study (June 27
th to October 4
th, 2016) AF patients who presented to the ED were treated according to the routine care. During the second stage of the study (October 5
th to March 20
th, 2017) patients who presented to the ED with primary AF were screened and treated according to the AF pathway[
25]. Implementation of the AF treatment pathway was associated with decreased rates of hospital admission (15%) compared to the group who received routine care (55%)[
25]. This was the first study that included an expert electrophysiologist besides emergency medicine doctors in decision making and AF management in ED.
Canadian emergency physicians are known for publishing widely on the “AF in the ED” field and for managing these patients quickly and efficiently[
20,
23,
26]. Academic centers in Canada have implemented a series of aggressive protocols of AF management in the ED setting[
20,
23,
26]. An initial retrospective cohort study by Stiell et al reported the results of pharmacological cardioversion with intravenous (i.v.) procainamide in 341 patients with recent-onset AF or atrial flutter. Sinus rhythm was restored in 52% of AF patients and 28% of atrial flutter patients[
20]. Admission rates were only 5.6% and adverse event rate was reported to be 10%, though without necessitating hospital admission. Return rate to the ED among those who were successfully cardioverted was 2.9% within 7 days[
20]. Subsequently, electrical cardioversion was included in this procainamide-based protocol for those not successfully cardioverted with i.v. procainamide[
26]. This modified protocol – called the “Ottawa aggressive protocol” – was evaluated in a retrospective cohort study. The sinus rate restoration was achieved in 92% of cases and minor adverse events rate was 7.6%[
26]. Certain steps included in the Ottawa aggressive protocol need to be mentioned. Specifically, no heparin or warfarin administration was necessary if onset was clearly less than 48 hours. Importantly, procedural sedation and analgesia (i.v. propofol and fentanyl) was given by emergency physicians without the use of transesophageal echocardiography unless onset was unclear[
20,
26].
Towards the same direction, ED patients at 6 urban Canadian centers with uncomplicated AF of less than 48 hours duration and CHADS
2 score of 0 or 1 were randomized in a 1:1 ratio to either chemical cardioversion with procainamide infusion, followed by electrical cardioversion if unsuccessful or to electrical cardioversion followed by procainamide infusion if unsuccessful[
23]. The primary endpoint was the proportion of patients being discharged within 4 hours of arrival. Results of the study showed that both strategies appeared to be effective and well tolerated but the electrical-first strategy was associated with less time spent in the ED[
23].
All the above led to the development of Acute AF/AFL Best Practices Checklist by the Canadian Association of Emergency Physicians (CAEP)[
27]. Consequently, the RAFF-3 trial[
28] sought to investigate the impact of implementing this guidance[
28]. It was conducted as a stepped-wedge cluster randomized trial at 11 large community and academic hospital EDs in 5 Canadian provinces and enrolled consecutive AF/AFL patients. The study intervention was introduction of the CAEP Checklist with the use of a knowledge translation-implementation approach that included behavior change techniques and organization/system−level strategies. The RAFF-3 trial led to optimized care of AF/AFL patients with decreased ED lengths of stay by 20.9%, increased ED rhythm control by drug or electricity, and no increase in adverse events[
28].
Most published AF treatment protocols have been validated in high volume-tertiary care centers. De Meester et al demonstrated that utilization of an AF treatment protocol could reduce admission rates even in a community-level hospital ED, based primarily on rate rather than rhythm control[
21]. Specifically, in this retrospective cohort study, there was a pre implementation period from March 2013-February 2014 with 586 patients included and a post implementation period from March 2015-February 2016 with 522 patients included. Primary outcome was hospital admissions and indeed these were greatly reduced in the post-implementation group (67.4% vs 80.4% in the pre-implementation group)[
21].
Study |
AF duration |
Intervention |
Study type |
Hospital type |
Number of patients |
SR rate (%) |
Admission rate (%) |
Adverse events rate (%) |
ED return rate (%) |
Burton[18] |
<48 h |
DCCV |
Retrospective cohort |
Tertiary |
388 |
86 |
14 |
8 |
10 (10d) |
Stiell[20] |
Acute onset |
PROC iv |
Retrospective cohort |
Tertiary |
341 |
52(AF), 28(AFL) |
5.6 |
10 |
2.9 (7d) |
Stiell[26] |
Recent onset |
PROC iv +/- DCCV |
Retrospective cohort |
Tertiary |
600 (PROC), 243(DCCV) |
58 (PROC) 92 (DCCV) |
3.2 |
7.6 |
8.6(7d) |
Scheuermeyer[23] |
<48 h |
PROC iv +/- DCCV or DCCV +/- PROC |
Multicenter randomized study |
Tertiary care |
41 PROC 43 DCCV |
100 PROC 98 DCCV |
0 |
25 |
12.2 (3d) 2.2 (30d) |
Martin[24] |
Recent onset |
CC or DCCV |
Multicenter observational cross sectional |
Tertiary care, community |
421 |
70 |
14 |
n/a |
n/a |
Ptaszek[25] |
New or recurrent |
CC or DCCV |
Prospective 2-stage at 2 hospital study |
Tertiary care, community |
104 routine care 104 AF pathway |
61 routine care, 76 AF pathway |
55 routine care 15 AF pathway |
n/a |
11(in 4 months) |
De Meester[21] |
New or recurrent |
DCCV if rate control failed |
Retrospective cohort |
Community |
1108 |
n/a |
67 |
n/a |
1 (3d) 3.6 (30d) |
CC: chemical cardioversion, DCCV: direct current cardioversion, PROC: procainamide |
8. HEROMEDICUS Protocol
The HEROMEDICUS protocol was developed by the National and Kapodistrian University of Athens to standardize the management of AF in ED. Our purpose is not only to reduce hospital admissions in a safe and effective environment but also to ensure the highest percentage of sinus rhythm rate through an aggressive rhythm control strategy. Most importantly the protocol is designed and executed by expert electrophysiologists in collaboration with cardiologists in the absence of emergency medicine doctors. Study protocol was approved by Ethics Committee and all included patients provide an informed consent form.
All patients who visit ED with a primary diagnosis of AF or AFL are included in the study, even if time from arrhythmia initiation cannot be specified. Patients with ischemic heart disease are randomized in the setting of FLECA-ED study and excluded from the HEROMEDICUS protocol[
38,
39]. Patients with indications of acute coronary syndrome or acute heart failure are also excluded. Enrollment started in September 2023 and will end in September 2025. Retrospective data will be collected from ED archive and hospital data in the periods September 2021- September 2022 and September 2022- September 2023 in order to assess the impact of Covid pandemia on AF visits and admissions from ED.
A special application has been created by P.T for the purpose of the HEROMEDICUS protocol and installed in mobile phones of cardiologists in charge in ED. Initial baseline clinical parameters are recorded and inclusion and exclusion criteria are implemented by means of bedside echocardiogram. Furthermore, emphasis in given in prior history of AF episodes and ED visits and hospital admissions, along with use of antiarrhythmics and anticoagulant status (warfarin and time in therapeutic INR range, type, and dose of direct oral anticoagulant)
-
Decision for rate control and subsequent (>21 days) electrical cardioversion in case of poor anticoagulant status and long (>48hrs) AF detection
- ⇒
Use of verapamil in case of AFL or atrial tachycardia
- ⇒
Use of β-blockers in case of AF
-
Chemical cardioversion if immediate pill in the pocket administration of propafenone or flecainide did not exceed 300mg and 200mg, respectively.
- ⇒
Use of iv flecainide (150mg IV within 10min)
-
Electrical cardioversion
- ⇒
Nil by mouth for the preceding 6hrs
- ⇒
Sedation and analgesia provided by cardiologist in ED (iv use of midazolam and pethidine)
- ⇒
Use of biphasic synchronized cardioversion (registration of delevered energy)
Modification of outpatient antiarrhythmic use based on discharge ECG.
Outpatient follow up in AF clinic.
Programmed electrical cardioversion in case of poor anticoagulant status and long (>48hrs) AF detection (use of transesophageal echocardiogram only in cases of suspected severe valvulopathy).
Preference of iv flecainide was based on the largest published so far network meta-analysis that indirectly compared and ranked antiarrhythmic agents focusing exclusively on adults with paroxysmal AF[
40]. Forty-one randomized controlled trials (6013 patients) were included with i.v. vernakalant and i.v. flecainide found to have the highest conversion rate within 4hr, possibly allowing discharge from the ED and reducing hospital admissions[
40]. Vernakalant was not selected due to its prohibitive cost in the setting of ED.
9. Conclusion
Management of AF in the setting of ED varies between countries and health care systems. It is noteworthy that rate control rather than rhythm control is historically a preferred strategy in the United States while in other western countries such as Canada, Australia, and the United Kingdom rhythm control is attempted in patients with AF who visit the ED[
26,
30,
41]. The most plausible reason to justify a conservative rather than aggressive strategy in the management of AF is the absence of specific guidelines from diverse societies. Towards this direction, in the RACE 7 ACWAS, a multicenter, randomized, open-label, noninferiority trial, a wait-and-see approach was noninferior to early cardioversion in achieving a return to sinus rhythm at 4 weeks in patients presenting to the ED with recent-onset, symptomatic AF[
42]. It should also be taken into account that in certain hospitals there is a lack of ED resources to perform electrical cardioversion, as well as a lack of payer scrutiny of short hospitalizations or readmissions due to AF, in contrast to other conditions such as angina, syncope, and decompensated heart failure[
42]. On the other hand, the very encouraging results yielded by aggressive AF management protocols, resulting in significantly fewer hospitalizations without evidence of worse outcomes, along with data suggesting patient preference for sinus rhythm restoration, constitute the fundamental reasons for pursuing an early cardioversion strategy[43-48].