Article, Cardiology

Emergency medicine considerations in atrial fibrillation

a b s t r a c t

Background: Atrial fibrillation is an abnormal heart rhythm which may lead to stroke, heart failure, and death. Emergency physicians play a role in diagnosing AF, managing symptoms, and lessening complications from this dysrhythmia.

Objective: This review evaluates recent literature and addresses ED considerations in the management of AF. Discussion: Emergency physicians should first assess patient clinical stability and evaluate and treat Reversible causes. Immediate cardioversion is indicated in the hemodynamically unstable patient. The American Heart As- sociation/American College of Cardiology, the European Society of Cardiology, and the Canadian Cardiovascular Society provide recommendations for management of AF. If hemodynamically stable, rate or rhythm control are options for management of AF. Physicians may opt for Rate control with medications, with Beta blockers and calcium channel blockers the predominant medications utilized in the ED. Patients with intact left ventricular function should be rate controlled to b110 beats per minute. Rhythm control is an option for patients who possess longer Life expectancy and those with AF onset b48 h before presentation, anticoagulated for 3-4 weeks, or with transesophageal echocardiography demonstrating no intracardiac thrombus. Direct oral anticoagulants are a safe and reliable option for anticoagulation. Clinical judgment regarding disposition is recommended, but literature supports discharging stable patients who do not have certain comorbidities.

Conclusion: Proper diagnosis and treatment of AF is essential to reduce complications. Treatment and overall management of AF include rate or rhythm control, cardioversion, anticoagulation, and admission versus dis- charge. This review discusses ED considerations regarding AF management.

Introduction

Atrial fibrillation is one of the most common dysrhythmias, af- fecting up to 1-2% of the population and 9% in those over age 80 [1-5]. Not only is AF the most common dysrhythmia overall, but it is also the most common dysrhythmia diagnosed in the ED. A study in 2013 demonstrates over a 29% increase in ED AF visits, with the dysrhythmia accounting for up to 0.5% of all ED visits [6]. When associated with other concomitant pathologies such as CHF, AF patients who present to the ED tend to be older and have higher mortality [6]. As the population continues to age, the dysrhythmia will likely increase in prevalence.

The dysrhythmia is strongly associated with stroke and heart failure [1-6]. Men, Caucasians, and the elderly demonstrate greater risk for AF, although women more commonly present with stroke from AF [7-15].

* Corresponding author at: 3841 Roger Brooke Dr., San Antonio, TX 78234, United States.

E-mail address: [email protected] (B. Long).

The risk of stroke approaches 5% annually with no anticoagulation, which decreases to b1% with appropriate management [10,16]. Mortal- ity in patients with AF is close to double that of patients with normal sinus rhythm [8-15]. Hypertension, diabetes mellitus, obesity, ethanol use, coronary artery disease (CAD), valvular heart disease, thyroid dis- ease, autonomic or electrolyte disturbances, and prior cardiac surgery contribute to AF [7-13,15-22], and up to 70% of patients with AF have associated heart disease [16-22]. A recent retrospective review of 564 ED patients with recent onset AF found hypertension to be the most common comorbidity, followed by ischemic heart disease [23]. Other contributing factors include channelopathies, stimulant use, pulmonary disease, enhanced Vagal tone, extreme exercise, smoking, and chronic kidney disease (CKD) [7,8,11,21-27].

In a normal heart, impulses originate from the sinus node, followed by regular atrial and ventricular activation and contraction [8,28]. AF re- sults from depolarization of multiple microreentry circuits, which reach the AV node at 300-600 atrial impulses per minute. The AV node refrac- tory period is responsible for the irregularly irregular ventricular re- sponse [8-10]. On electrocardiogram (ECG), P waves will be absent

https://doi.org/10.1016/j.ajem.2018.01.066 0735-6757/

and the R-R intervals irregular. These irregular atrial beats cause ineffec- tive atrial contraction, leading to thrombus formation predominantly in the left atrial appendage [8-10,16,22,23]. The irregular beats also can lead to rapid ventricular activity, which if not well controlled, decrease Myocardial blood flow, decrease cardiac output, and cause long term damage to the myocardium [8-11,28]. The QRS complex is narrow in those without bundle branch block (BBB), though QRS width N120 ms is found in those with ventricular BBB. Some patients with complete heart block and AF may present with regular rhythm and no discernable p waves. Patients with WPW syndrome and AF may demonstrate an ECG resembling ventricular tachycardia, though AF with preexcitation demonstrates an irregularly irregular rhythm [8-13,16].

Atrial fibrillation is comprised of several categories [8-13,16]. Parox- ysmal AF consists of episodes that terminate spontaneously or with in- tervention within 7 days of onset, while persistent AF is present for longer than 7 days [8-13,15,16]. Recurrent AF is defined by more than two episodes. Longstanding persistent AF is continuous AF for greater than one year. Permanent AF is defined as the presence of continuous AF, with joint decision between patient and clinician to stop further at- tempts to maintain sinus rhythm. If permanent AF is eventually treated with rhythm control, it is redefined as longstanding persistent AF [8-13,15,16]. Prolonged AF makes restoration of normal sinus rhythm difficult [8-11].

Methods

This is a narrative review of AF emergency evaluation and manage- ment. The objective is to evaluate recent literature and address current considerations in the management of AF in the ED. The literature search was limited to inclusion of recent studies from the prior 20 years. Rather than discussing AF in its entirety, the authors have investigated specific components of the condition relevant to emergency physicians includ- ing ED evaluation, rate and rhythm control, anticoagulation, and patient disposition.

Discussion

When evaluating and managing the patient with AF with rapid ven- tricular response (RVR), the physician should consider if the patient is unstable and whether this is due to primary AF versus another cause. Hypotension and tachycardia may not be due to AF solely, but rather sepsis, myocardial infarction, gastrointestinal hemorrhage, alcohol withdrawal, pulmonary embolism, and other causes [29,30]. This is termed complex AF [29,30]. Inflammation and oxidative stress, seen in sepsis, may play a role in the development of AF, as they may directly change the electrical activity of the cardiac myocyte [7,8,25]. Up to 25% of hospitalized patients with sepsis develop AF [7]. If the etiology of tachycardia and hypotension is due to another primary etiology, the patient will likely not improve with interventions targeting specifically AF alone. Attempts to control the heart rate or rhythm in these patients are usually less successful and may be harmful. Rather, the etiology must be properly evaluated and treated [29,30]. Scheuermeyer et al. evaluated 416 patients with AF or atrial flutter in a retrospective de- scriptive cohort study [29]. Of these patients evaluated, 105 underwent rate control and 30 rhythm control, and 55 adverse events occurred in this group of 135 patients (40.7%), while the 281 patients not managed with rhythm or rate control demonstrated 20 adverse events (7.1%) [29]. Patients with complex AF demonstrated a 5.7-fold increase in ad- verse events and 11.7-fold increase in Major adverse events with rate or rhythm control, and patients with sepsis or heart failure with AF demonstrated the highest number of adverse events, though others in- cluded acute coronary syndrome (ACS), acute renal failure, obstructive lung disease, gastrointestinal bleeding, and stroke [29,30]. These pa- tients may need the relative tachycardia to compensate. If no underlying cause of AF is suspected on evaluation, management should focus on symptom improvement and reduction in potential complications

[7-12,30]. If another condition is suspected on focused history and ex- amination and the patient has AF with RVR, the Underlying etiology should be managed (Fig. 1), as treatments focusing on AF alone may re- sult in Patient harm.

Stable patients

Most patients with AF with RVR who present to the ED possess a well-perfusing blood pressure [8,11,13,30]. Focused history and physi- cal examination are warranted, with the physician inquiring on onset, frequency, duration of symptoms, associated symptoms, prior episodes, medications (anticoagulants, antiarrhythmics, rate control agents), past medical history (heart disease), and instigating factors. Evaluation in- cluding electrolyte panel, complete blood cell count, chest x-ray, and ECG is advised. Thyroid panel may assist if the patient demonstrates other symptoms associated with thyroid abnormality. Additional test- ing including Brain natriuretic peptide or troponin is not recom- mended on a routine basis, but rather, depends on the clinical situation. Pregnancy testing in reproductive aged-women is recommended, and evaluation for pulmonary embolism should be based on the clinical sit- uation [30].

Evaluation for ACS can be challenging, as a significant number of pa- tients with AF have coexisting coronary artery disease (CAD) [30-34]. Chest pain is present in 20% of patients presenting with AF with RVR, which is not usually due to a primary ischemic event [31,32]. Patients without significant ST-segment changes are at low risk for acute myo- cardial infarction [31,32]. One prospective cohort study suggested chest pain and ST segment depression b2 mm (mm) are common find- ings in AF, but they have limited ability to diagnose or predict ACS [32]. ST segment elevation or depression >=2 mm was found to be a reliable predictor of concomitant ischemia [32]. Similarly, a 2007 retrospective study of ED patients with chest pain and AF demonstrated the presence of AF did not change the risk of ACS in patients with chest pain [33].

Clinical judgment is required when assessing risk of ACS with AF. Pa- tient assessment and ECG before and after rate or rhythm control should be strongly considered in the evaluation of ACS [30-34]. Patients with significant ST changes after treatment require consideration for ACS.

Echocardiography in the ED

transesophageal echocardiography allows assessment for in- tracardiac thrombus before cardioversion and cardiac function. TEE should be conducted if symptom duration is N48 h or unknown before cardioversion to evaluate for thrombus [8-13]. However, if required for situations such as hemodynamic instability, cardioversion can be performed emergently without TEE. In younger, healthy patients with known time of onset b48 h, Transthoracic echocardiography is low yield and likely not beneficial in the ED, though older patients with greater likelihood of cardiac abnormality may benefit from echo- cardiography. The American and European guidelines recommend TTE for patients with AF as part of the initial evaluation to assess for struc- tural heart disease, cardiac function (right and left heart), and atrial size, which can occur as outpatient [8-13].

Rate versus rhythm control

Symptom improvement may occur via rate or rhythm control, al- though controversy surrounding rate versus rhythm control still exists. Rhythm control for AF of <=48 h duration is viable, with stroke risk ap- proaching b1% if the patient is cardioverted within 48 h [35]. Patients who present after 48 h require rate control before rhythm control is considered due to increased risk of stroke [8-13,16]. Patients with no in- tracardiac thrombus on TEE or those on anticoagulation for 3-4 weeks may also undergo rhythm control. Patients can spontaneously convert to sinus rhythm on their own. Several studies conducted in the last sev- eral years sought to determine if rate control or rhythm control is the

Fig. 1. AF management pathway.

better option for overall survival and quality of life [36-45]. In 2002, the Atrial Fibrillation Follow-up Investigation of Rhythm Management (AF- FIRM) trial, a randomized multicenter comparison study, was conduct- ed, though the study population included patients from Clinical sites other than the ED. [39,40] Two treatment strategies were compared, in- cluding pharmacologic or non-pharmacologic rhythm methods and rate control medications including diltiazem, verapamil, and beta blockers [39,40]. There were no significant differences between the rate and rhythm control groups for overall mortality. More of the rhythm control patients were hospitalized and sustained more Adverse drug effects than the rate control patients, but there were no differences in stroke when controlled for anticoagulation [39,40]. A similar 2002 study showed rate control to be non-inferior to rhythm control in terms of death, heart failure, thromboembolic complications, requirement of a pacemaker, and severe adverse drug effects [41]. A 2008 study of pa- tients with AF and heart failure suggested rhythm control does not con- tribute to improved survival as compared to rate control with anticoagulation [42].

A meta-analysis released in 2013 found no difference between rate versus rhythm control in mortality, stroke, embolism, worsening heart failure, myocardial infarction, and bleeding [44]. Patients younger than 65 years in subgroup analysis demonstrate rhythm control to be superi- or in prevention of all-cause mortality (relative risk 3.03; 95% confi- dence interval 1.59-5.75) [44]. The Okcun 2004 study was a single center RCT evaluating AF N48 h, finding rhythm control to be associated with fewer deaths (15% versus 43%), with no difference in embolic events [45]. The J-RHYTHM study in 2009 was a multicenter RCT con- ducted in patients over 18 years with AF b48 h, finding rhythm versus rate control was associated with lower events (defined by total mortal- ity, symptomatic cerebral infarction, systemic embolism, hospitalization for heart failure, major bleeding, or physical/psychological disability re- quiring treatment strategy alteration), 15% versus 22%, respectively [46]. The American Heart Association/American College of Cardiology (AHA/ACC), the Canadian Cardiovascular Society (CCS), and the Europe- an Society of Cardiology (ESC) suggest that ultimately patients should be cardioverted in the long run, which can improve quality of life and decrease symptoms [8-13,16].

In patients b65 years of age with known onset of b48 h, rhythm con- trol can be safe and useful. Spontaneous cardioversion may occur, espe- cially in younger patients. Rate control is advised in patients with valvular disease or chronic AF, though patients with CHF may undergo rhythm control [8-13,16].

Rhythm control options

Guidelines suggest rhythm control may be beneficial for younger pa- tients with greater life expectancy [8-16]. Older patients likely warrant rate control before rhythm control is considered, though initial rhythm control is a valid option in specific circumstances including known onset of AF b48 h, TEE with no intracardiac thrombus, or on anticoagulation at therapeutic levels for 3-4 weeks [39-41]. Rate control is the first option for several situations before rhythm control is considered (Table 1). Rhythm control includes cardioversion, whether by pharmacologic or electrical means. ED studies demonstrate electrical cardioversion to be 90% effective and pharmacologic cardioversion to be 60% effective [8-11,36-38,47]. Some patients with New onset AF may covert to sinus rhythm within a few hours, negating the need for medical or electrical cardioversion.

Airakinsen et al. evaluated adult patients age N 18 with acute onset AF (b48 h) treated with cardioversion in the ED. [48] The primary study outcome was thromboembolic event within 30 days after cardio- version. A total of 7660 cardioversions were performed, but the analysis for embolic complications included 2481 patients with no peri-proce- dural or post-procedural anticoagulation. At 30 days, the authors noted 38 Thromboembolic events: 31 strokes, 4 TIAs, 2 pulmonary em- boli, and 1 combined stroke and a systemic embolism. Median time to a thromboembolic event was 2 days. Further analysis revealed heart fail- ure, diabetes, and age N60 years were associated with thromboembolic events [48]. The authors conclude that while overall embolic events are low (1.5% of patients), risk is elevated in patients with older age, Female gender, heart failure, and diabetes. While some patients may clearly present with AF in less than a 48 h period, they require risk stratification [48].

Canadian physicians have repeatedly demonstrated cardioversion to be efficacious and safe. The “Ottawa Aggressive Protocol” consists of acute rhythm control and Discharge home for Hemodynamically stable patients with recent onset (b48 h) rapid AF or atrial flutter [37]. Stiell et al. evaluated this protocol in 660 patient visits, with a mean patient age 64.5 years. Using intravenous (IV) procainamide or electric cardio- version, 96.8% of patients were discharged home, and 93.3% remained in normal sinus rhythm. If procainamide is used, 1 g (g) IV over 60 min is provided. In patients cardioverted with procainamide, 44% dem- onstrate return to normal sinus rhythm with 500 mg. If systolic blood pressure decreases below 100 mm Hg, the infusion is discontinued. One hour after cardioversion, the patient can be considered for dis- charge. No anticoagulation was provided upon discharge for most pa- tients [37]. There were few adverse effects, including hypotension in 6.7%, bradycardia in 0.3%, and a 7 day relapse rate of 8.6%, with no strokes and no deaths. However, no long-term follow-up for patients occurred, and this is not a common strategy in the U.S. [37,38].

Ibutilide, vernakalant, and flecainide were evaluated in a prospective observational study in patients with AF onset b48 h and average age

66.8 years [49]. All patients were anticoagulated with Low Molecular Weight Heparin, and 72.5% of patients converted with one medication [49]. Ibutilide is a Vaughan-Williams Class III antidysrhythmic, often given in doses of 1 mg IV over 10 min, which may be repeated in anoth- er dose of 1 mg IV. It may result in QTc prolongation and ventricular tachycardia in 3% of patients, and ejection fraction (EF) must be N30%. Before administering ibutilide, serum potassium and magnesium should be assessed and repleted if needed. Prolonged QTc is a contrain- dication to ibutilide. Magnesium can improve the ability to cardiovert to sinus rhythm by 60% [50], and it can also improve the efficacy of ibutilide in cardioversion [51].

The AHA provides a Class I Recommendation with Level of evidence A for flecainide, dofetilide, propafenone, and ibutilide for pharmacologic cardioversion [7], while amiodarone receives a Class IIa Recommenda- tion, Level of Evidence A [8]. These guidelines fail to mention procain- amide, which is also absent in the Canadian guidelines [8,12,13,16]. The European guidelines recommend flecainide, propafenone, ibutilide, vernakalant, or amiodarone [9-11]. With the efficacy demonstrated in the Ottawa studies and availability of procainamide in the ED, this re- view recommends procainamide for pharmacologic cardioversion.

Rate control

Rate control is recommended for stable patients with AF duration N48 h [8-13,30]. For rate control, beta blockers and nondihydropyridine Calcium-channel blockers are the most commonly used agents in the ED. Others include digoxin and amiodarone. Digoxin has been used in the ED for AF, though it does not adequately control heart rate in the ED setting unless the patient is sedentary [30,52]. Digoxin may be used for patients in whom beta blocker or calcium channel blocker ther- apy is not effective, as well as in patients with decompensated heart fail- ure [8-13,52,53]. Amiodarone suppresses AV nodal conduction through its sympatholytic and calcium antagonistic properties, and it may be used for rate control in patients with reduced EF. Amiodarone requires up to 6-7 h to achieve rate control [8-13]. Agents including beta blockers, calcium-channel blockers, digoxin, and amiodarone should

Table 1

Rhythm control/cardioversion contraindications [8-13].

Contraindications to ED cardioversion (electrical or medication)

Unknown duration of AF
AF duration >=48 h
Patient is high risk for stroke: mechanical heart valve, rheumatic heart disease, or recent stroke or transient ischemic attack
Patient is high risk for ventricular dysrhythmia: electrolyte abnormality such as severe hypomagnesemia or hypokalemia, digoxin toxicity

not be used in patients with pre-excitation and AF, in which the medi- cations have the potential to decrease the refractoriness of bypass tracts and accelerate the ventricular rate [54,55]. This review will discuss beta blockers and nondihydropyridine calcium-channel blockers, which demonstrate greater rate control efficacy in the ED. Beta blockers, pre- dominantly metoprolol, or nondihydropyridine calcium channel blockers, primarily diltiazem, can be used (Table 2).

Heart rate target varies in the literature, with early guidelines recommending strict heart rate b80 beats per minute (bpm). The most recent European and U.S. guidelines recommend rate control b110 bpm, with Canadian guidelines recommending b100 bpm [8-13,16]. These guidelines are based on the Rate Control Efficacy in Per- manent Atrial Fibrillation: a Comparison between Lenient versus Strict Rate Control II (RACE II) trial [56]. This trial found lenient rate control to be noninferior in preventing Cardiovascular death, CHF hospitaliza- tion, stroke, embolism, bleeding, or life-threatening dysrhythmia over 3 years. The lenient control group met criteria for Heart rate control in 98% of cases with 75 Hospital visits, compared to the strict control group meeting heart rate target in 78% of cases with 684 hospital visits (close to nine times as many visits) [56].

Several studies have compared medication class efficacy (calcium channel blocker versus beta blocker) in rate control. A randomized, open-label study in 1989 compared esmolol and verapamil in 45 pa- tients, finding a decline in heart rate from 139 bpm to 100 bpm with esmolol and 142 bpm to 97 bpm with verapamil [57]. Close to 50% of pa- tients with esmolol converted to sinus rhythm, while 12% of the verap- amil group converted [57]. A 2005 study compared 20 patients receiving metoprolol (0.15 mg/kg, maximum 10 mg) to 20 patients re- ceiving diltiazem (0.35 mg/kg, maximum 25 mg) [58]. Success was de- fined by decrease in heart rate b100, decrease in ventricular rate by 20%, or conversion to sinus rhythm. The diltiazem group demonstrated greater success at 2 min, (50% versus 15%) [58]. One of the best designed studies evaluating this question was published in 2015, which evaluat- ed 52 patients in a prospective, randomized, double blind trial, with pri- mary outcome defined by heart rate b100 bpm [59]. Dosing included diltiazem 0.25 mg/kg (maximum 30 mg) and metoprolol 0.15 mg/kg (maximum 10 mg), with a second dose provided after 15 min if heart rate control was not obtained. The primary outcome was reached in 95.8% of patients in the diltiazem group and 46.4% of patients in the metoprolol group within 30 min. However, this was a convenience sam- ple, and the maximum dose of metoprolol was 10 mg (5 mg three times can be provided) [59]. A 2012 prospective study found rate control was successful for 71% of patients receiving calcium channel blockers and 79% for those receiving beta blockers, though authors did not report sta- tistical analysis [60].

Metoprolol and diltiazem are both effective, but the highest quality study to date suggests diltiazem performs better in controlling heart rate. A systematic review released in 2015 evaluating diltiazem versus metoprolol demonstrated that diltiazem possesses approximately an 80% greater likelihood for controlling heart rate [61]. If the patient is not on a rate controlling agent, diltiazem may offer better ability to de- crease ventricular rate. Several patient factors must be considered [62-66]. Treatment with beta-blockers may improve survival in patients with heart failure with reduced EF but not heart failure with preserved EF [62-66]. However, the use of beta blockers acutely in patients with decompensated heart failure and AF may result in cardiogenic shock [62-66]. Calcium channel blockers such as diltiazem are associated with decreased long-term outcomes in heart failure patients with re- duced EF, though use for short-term rate control may be efficacious. In ischemic heart disease, beta blockers are associated with reduction in ventricular dysrhythmia and sudden cardiac death, though these bene- fits attenuate over time in the post-MI setting [62-66]. Beta blockers are recommended in thyrotoxicosis. In hypertension, calcium channel blockers are one of the first-line medications [67,68].

Magnesium has been evaluated for rate control in several studies [69-74]. Chiladakis et al. evaluated diltiazem 25 mg over 15 min

Table 2

Rate control agents^a.

Medication Form Standard dose Note

Beta blocker

Metoprolol IV 5 mg slow push, repeat up to 15 mg HR effect typically seen 5 min after dose PO 12.5-100 mg 2x/day 37.5 mg, 50 mg are options

Carvedilol PO 3.125 mg 2x/day, titrated to max dose 25 mg 2x/day Used in heart failure

Bisoprolol PO 2.5-5 mg once/day Often used in reactive airways Esmolol IV 500 mcg/kg bolus, then 50-300 mcg/kg/min

Calcium-channel blocker

Diltiazem	IV	0.25 mg/kg slow push, may give a second 0.35 mg/kg dose 15 min from first dose	Maximal HR control 2-7 min after dose, infusion 5-15 mg/h after 2nd dose
	PO	120-240 mg 1-2x/day (extended)
		30-90 mg 4x/day (immediate)
Verapamil	IV	0.075-0.15 mg/kg; may give additional 10 mg	Greater chance of hypotension
	PO	40-80 mg up to 3x/day

Digitalis glycoside

Digoxin IV 0.25 mg, up to maximum 1.5 mg over 1 day Little effect on HR, requires hours for effect PO 0.125-0.25 mg

Others

Amiodarone IV 150-300 mg May be used in critically ill patients or those with reduced EF for rate

PO 100-200 mg every day control, though several hours are needed for effect

Magnesium IV 2 g over 15 min Adverse effects including flushing and hypotension

^a Abbreviations: g - grams; HR - heart rate; IV - intravenous; kg - kilogram; PO - per os; mcg - microgram; mg - milligram; EF - ejection fraction.

followed by diltiazem infusion compared to magnesium 2.5 g IV over 15 min, then 7.5 g over 6 h [69]. This study found similar efficacy in re- ducing rate at 1 h [69]. Davey et al. compared digoxin and magnesium

2.5 g IV over 20 min and 2.5 g over 2 h [70]. Magnesium was more likely to achieve heart rate b100 bpm (65% versus 24%, RR 1.89; 95% CI 1.38- 2.59) [70]. Joshi et al. evaluated verapamil 5 mg versus magnesium 2 g and found verapamil more likely to achieve heart rate b100 bpm (55.6% versus 19.5%) [73]. In a prospective, randomized trial, Gullestad found verapamil 5 mg IV was more likely than magnesium 1.2 g over 5 min to achieve heart rate b100 bpm (48% versus 28%) [72]. Two meta-analyses concluded magnesium to be safe and effective compared to placebo and digoxin, though the majority of the rate control data comes from placebo-controlled trials [50,73]. The meta-analysis by Onlan found magnesium approximately doubled the chance of rate con- trol [50]. Another ED trial compared normal saline to magnesium 2.5 g over 15 min, with no change in heart rate. Adverse effects include flush- ing and mild hypotension, and magnesium does not appear effective in patients with chronic AF [50,73].

Considerations in anticoagulation

Thromboembolic risk is an important consideration due to increased stroke risk. Stroke rates may approach 2.75% in males and 2.55% in fe- males with anticoagulation, though this increases in patients over age 65 [8-13,16]. Annual stroke risk may reach 5% in older patients with no anticoagulation and close to 10% if the patient has experienced prior stroke [8-13,16]. Unfortunately, many patients who meet criteria for Oral anticoagulation (OAC) do not receive appropriate therapy [30,75]. Several scores are available for assessment of stroke risk, includ- ing CHADS₂ (Congestive Heart Failure, Hypertension, Age N75 years, Di- abetes Mellitus and Prior Stroke or Transient ischemic attack ) and CHA₂DS₂-VASc (CHADS₂ plus vascular disease, age 65-74 years and fe- male gender) [1-13,16,75-77]. In patients with CHADS₂ N 2 who war- rant anticoagulation, 38% receive only aspirin, and 40% of those with CHA₂DS₂-VASc >= 2 receive only aspirin (Table 3) [75]. The ED is a vital component to initiating proper therapy including anticoagulation, as patients discharged from the ED with anticoagulation are more likely to be receiving it later at 1 year when prescribed in the ED. [30,75].

The AHA/ACC, CCS, and ESC guidelines recommend using a risk score

such as CHADS₂ or CHA₂DS₂-VASc to determine if a patient is eligible for oral anticoagulation (Table 3) [8-13,16,76,77]. These societies

recommend patients with AF should be risk stratified using a prediction model [8-13,16].

The AHA/ACC proposes the use of CHA₂DS₂-VASc as the risk score of choice in determining anticoagulation needs [8]. The AHA/ACC recom- mends that any patient with prior stroke, TIA, or CHA₂DS₂-VASc score

>=2 should be anticoagulated with warfarin, dabigatran, rivaroxaban, or apixaban [8]. If a patient has non-valvular AF and CHADs₂-Vasc score 0, no Oral anticoagulant therapy is recommended. Warfarin is the rec- ommended anticoagulant in patients with AF and mechanical heart valve [8]. If patients have a CHA₂DS₂-VASc score 1, the AHA/ACC leaves the choice to the patient and the clinician and recommends anticoagulation, aspirin, or no anticoagulation [8].

The ESC endorses the use of the CHA₂DS₂-VASc score [9-11]. The ESC denotes low risk as a CHA₂DS₂-VASc score 0 for males and 1 for females. These patients do not require anticoagulation per the ESC. Unlike the AHA/ACC, the ESC recommends anticoagulation for any patient with a CHA₂DS₂-VASc score of >=1 for men and >=2 for women [9-11]. If anticoagulation is prescribed, the ESC recommends a vitamin K antago- nist, a direct thrombin inhibitor, or an oral factor Xa inhibitor [9-11]. The ESC does not recommend one specific anticoagulant, and clinicians should consider Patient variables, cost, and drug compliance tolerability in determining the medication used [9-11].

The Canadian Guidelines recommend using the CHADS₂ score to es- timate initial stroke risk but still consider age >=65 as an initial risk com- ponent that is not included in the CHADS₂ score. Anyone age >=65 with AF should receive OAC therapy [12,13,16]. If a patient is b65 and has any CHADS₂ risk factors, OAC is recommended. If a patient is b65 and has no CHADS₂ risk factor, then the CHA₂DS₂-VASc score is used as a complement to determine OAC need. If patients are b age 65 with no CHADS₂ risk factors but have vascular disease, aspirin is recommended. The Canadian guidelines do not consider female gender or vascular dis- ease alone as sufficient reasons for OAC [12,13,16].

The most common anticoagulation regimen previously included warfarin, a vitamin K antagonist, for anticoagulation. However, warfarin requires repeat laboratory assessments and has significant interactions with food and other medications. Direct oral anticoagulants , also known as novel oral anticoagulants, include direct thrombin inhib- itors (dabigatran) and factor Xa inhibitors (apixaban, edoxaban, rivaroxaban) [78-81]. Apixaban, rivaroxaban, and dabigatran are ap- proved by the FDA for use in nonvalvular AF (Table 4) [78-81].

DOAC therapy offers several advantages over warfarin, such as no re- quirement for routine anticoagulation monitoring (unless the patient

Table 3

CHA₂DS₂-VASc score [76,77].

Criteria Scoring

Age b65 0

65-74 1

>=75 2

Sex Female 1

Male 0

CHF history Yes 1

No 0

Hypertension Yes 1

No 0

Stroke/TIA/thromboembolism history Yes 1

No 0

Vascular disease history Yes 1

No 0

Diabetes history Yes 1

No 0

AHA/ACC score interpretation 0 points - Low risk, no anticoagulation

1 point - Low-moderate risk, consider antiplatelet or Anticoagulation medication N2 points - Moderate-high risk, offer anticoagulation

has renal disease, in which the patient needs repeat renal function mon- itoring) and less interaction with diet and medications [78-81]. Assess- ment of the patient’s renal function is recommended before these agents are used [78-81]. When compared to warfarin in patients with AF, DOACs are associated with decreased stroke risk overall and lower mortality [82-87]. DOACs are also associated with decreased major bleeding and intracranial hemorrhage (RR 0.49, 95% CI 0.38-0.64) as compared to warfarin, though GI bleeding may be increased with DOACs (RR 1.25, 95% CI 1.01-1.55) [80]. The Randomized Evaluation of Long-Term Anticoagulation (RE-LY) trial [82], Rivaroxaban Once daily oral direct factor Xa inhibition Compared with vitamin K antago- nism for prevention of stroke and Embolism Trial in AF (ROCKET-AF) trial [83], and Apixaban for Reduction in Stroke and Other Thromboem- bolic Events in Atrial Fibrillation (ARISTOTLE) trial [84] demonstrate the efficacy and safety of rivaroxaban, apixaban, and dabigatran when com- pared to warfarin (Table 5). However, warfarin is recommended in pa- tients with mechanical heart valves and mitral stenosis [78-87]. Patients with Renal impairment typically require dose adjustment [78-80,86,87]. The use of anticoagulation should be considered when treating pa- tients with AF in the ED, whether the patient is stable or unstable or if rate or rhythm control is utilized. Patients who are high risk for stroke, based on the recommendations above, should be considered candidates

for anticoagulation, even in emergent situations [8-13,16].

All three guidelines recommend use of the HAS-BLED score (Table 6) to identify those at increased risk of bleeding who may have factors that can be modified to decrease this risk [8-13,16]. This score was devel- oped from the Euro Heart Survey from 3978 patients to assess one-

year bleeding risk in AF. Patients with <=1 point experience a 3.4% chance of bleeding in validation, with 5.8% chance with score N3 points [88-90].

Unstable patients

hemodynamic stability of the patient with AF and RVR is not a di- chotomous state, but a continuum. Per the AHA, hemodynamic instabil- ity is defined by systolic blood pressure b 90 mm Hg, altered mental status, cardiac ischemia, or severely decompensated heart failure (for example pulmonary edema) due to the underlying rhythm [8]. The clearly unstable patient requires immediate resuscitation [8-13,16]. Anticoagulation is recommended if dysrhythmia onset is unknown or N48 h. If due to AF primarily, electrical cardioversion is advised, no mat- ter the duration of atrial fibrillation [ 8-13,16,30]. Most patients hemo- dynamically unstable due to AF demonstrate heart rate N 140-150 bpm. Physicians should evaluate for other conditions resulting in AF with RVR, which may be compensatory for an underlying condition (Fig. 1) [29,30]. If electrical cardioversion is chosen, 100-200 J biphasic should be provided, though authors recommend 200 J as initial starting dose [30]. Anterior-posterior pad placement may demonstrate greater efficacy than anterior-lateral placement [91]. For sedation, authors uti- lize etomidate 0.1 mg/kg IV with fentanyl IV. Ketamine may be used as well. Anticoagulation should be provided if warranted in the hemo- dynamically unstable patient, but this should not delay cardioversion. Anticoagulation is warranted if the onset of AF is unknown or perma- nent AF [ 8-13,16,30].

Chronic AF may not respond to electrical cardioversion, requiring other therapies. Hypotension warrants careful attention. Definitive therapy of hypotension requires correction of the underlying condition [8-13,16,30]. Small IV boluses (250-500 ml) of fluid can be used. Physicians should be wary of causing pulmonary edema. To improve perfusion and blood pressure, vasopressors may be needed. Norepi- nephrine, starting at 5 mcg/min IV and titrating to improved clinical sta- tus or improved blood pressure is warranted. Phenylephrine can be used, but only in patients without significant heart failure, as it is contra- indicated in patients with significant systolic dysfunction. If electrical cardioversion is ineffective and AF is the predominant cause of hemody- namic instability, heart rate control is vital. Diltiazem, metoprolol, and amiodarone are options. Diltiazem can be given as 0.25 mg/kg IV (or 25 mg) over 10-15 min. Otherwise, small doses of 2.5 mg per minute can be given. Once heart rate improves, a diltiazem drip of 5-15 mg/h or 30-60 mg by mouth is needed. Extended release diltiazem may be used, with maximum dosing 360 mg per day. Calcium may have benefi- cial properties in pretreatment before a calcium channel blocker [92-96]. While calcium may reduce hypotension with verapamil [92,95], literature suggests calcium may not demonstrate the same ef- fects with diltiazem [96].

Other medications include amiodarone, which can be given as 150 mg IV over 10 min, then 1 mg/min IV infusion for the first 6 h. In this setting, amiodarone is used for rate control, not cardioversion, which can take up to 6 h [8-13,16]. This medication is given a grade IIA

Table 4

DOAC therapy [78-81].

Medication Rivaroxaban (Xarelto(R)) Apixaban (Eliquis(R)) Edoxaban (Savaysa(R)) Dabigatran (Pradaxa(R))

Mechanism Dose

Factor Xa inhibitor

20 mg once/day with evening meal?

5 mg twice/day??

60 mg once/day?

Factor IIa inhibitor 150 mg twice/day?

Renal elimination

66%

25%

35%

80%

Notes - Caution warranted in those on medications affecting Cytochrome P450 3A4 or p-glycoprotein.

- - DOAC therapy should be avoided in patients with severe renal or liver disease, those who cannot comply with consistent dosing, recent bleeding, and platelets b70,000/mm³.
  - Cost may be prohibitive for long-term therapy.
  - patients with cancer should be provided low molecular weight heparin over DOAC.

* Rivaroxaban, edoxaban, and dabigatran dosing depends on patient creatinine clearance (CrCl). Patients with CrCl 15-50 ml/min should receive rivaroxaban 15 mg/day or edoxaban 30 mg/day. Patients with CrCl 15-30 ml/min should 75 mg twice/day.

?? Apixaban dose 2.5 mg if age >= 80 years, body weight <= 60 kg, or serum creatinine >=1.5 mg/dL.

Table 5

Prominent studies evaluating DOAC therapy in AF [78-87].

Study	Medication compared to Warfarin	Age - DOAC vs. Warfarin (mean in years)	Patients	Stroke or systolic embolic event RR (95% CI)	Major bleeding RR (95% CI)
RE-LY	Dabigatran	71.5 vs. 71.6	18,113	0.66 (0.53-0.82)	0.94 (0.82-1.07)
ROCKET-AF	Rivaroxaban	73 vs. 73	14,264	0.88 (0.75-1.03)	1.03 (0.90-1.18)
ARISTOTLE	Apixaban	70 vs. 70	18,201	0.80 (0.67-0.95)	0.71 (0.61-0.90)
ENGAGE AF-TIMI	Edoxaban	72 vs. 72	21,105	0.88 (0.75-1.02)	0.80 (0.71-0.90)

recommendation, level B evidence, recommendation by the AHA/ACC for rate control in critically ill patients. Magnesium may be given as 2- g IV over 10-15 min, which may increase the chance of spontaneous reversion to normal sinus rhythm. Digoxin may be utilized in unstable patients at doses of 0.25 mg IV [8-13,16].

Disposition

Significant variation in patient disposition is present when manag- ing patients with AF [97-100]. A 2015 study by Barrett et al. showed 69% of U.S. patients with a primary diagnosis of AF resulted in hospital- ization, while only 37% were hospitalized in Canada [98]. Physicians in Canada are more likely to cardiovert and discharge stable patients home [30,31,36-38]. Studies have shown that discharge from the ED is safe for most patients [30,31,37]. The introduction of anticoagulation with DOACs allows safe, early discharge in patients requiring anticoagulation as well. It has been suggested the patients should be ad- mitted if they have another ED diagnosis such as pneumonia, CAD, heart failure, or failure to achieve rate or rhythm control [30,31,100]. Other- wise, patients are likely safe to be discharged with close outpatient fol- low-up [100]. If anticoagulation is needed, DOACs are reliable and safe [78-87].

In 2011, Barrett et al. developed a clinical decision model that risk stratifies patients with symptomatic AF [97]. The authors’ aim was to estimate a patient’s risk of experiencing an adverse event 30 days after ED visit. Primary adverse outcomes included an ED return visit within 30 days, unscheduled hospitalization, cardiovascular complica- tion, or death. Researchers found older age, a smoking history, inade- quate ED rate control, shortness of breath, and beta blocker treatment were associated with an increased risk of 30-day adverse events [97]. Of the total 832 patients studied, 216 (25.9%) experienced at least 1 of the 30-day adverse events. The authors combined these risk factors into a clinical prediction model called the Risk Estimator decision aid for Atrial Fibrillation (RED-AF). RED-AF assigns points according to age, sex, preexisting disease such as heart failure and hypertension, physical examination findings, and amount of rate control [97].

Table 6

Risk of hemorrhage [88-90].

HAS-BLED score for major bleeding risk (each factor scores 1 point)

Hypertension (uncontrolled, N160 mm Hg systolic)
Abnormal renal or liver function (Renal disease (dialysis, transplant history, Cr N 2.26 mg/dL or N200 umol/L); Liver disease defined by cirrhosis or bilirubin N2x normal or AST/ALT/AP N3x normal)
Stroke history
Bleeding event or predisposition to bleeding
Labile INR (unstable/high INRs, time in therapeutic range b 60%)
Elderly (age >= 65 years)
Drugs or alcohol (drugs defined by anticoagulants, alcohol use defined by >=8 drinks/week)

Score b 1 warrants consideration of anticoagulation, as patient has low bleeding risk Score 2 warrants consideration of anticoagulation, but patient has moderate bleeding risk

Score N 3 is high risk for bleeding, and factors associated with higher bleeding risk should be addressed

In a subsequent prospective cohort study, Barrett et al. validated the RED-AF score as an aid to clinical decision making [98]. The primary out- come was >=1 AF-related adverse outcome such as ED revisits, re-hospi- talization, Cardiovascular complications, and death within 30 days. Of the included patients, 24% had >=1 adverse event within a 30-day period, and a RED-AF score of 87 was determined to be the optimum score with a sensitivity of 96% and specificity of 19% [98]. The RED-AF score also had a positive predictive value of 27% and a negative predictive value of 19%. Overall the authors conclude the RED-AF score is “moderately better” than chance for determining adverse event, and clinicians should not completely rely on the score. Future studies are necessary to determine whether the score may aid in clinician assessment of risk and Disposition decisions [98]. Clinicians should continue to utilize cur- rent society recommendations, consider discharge in otherwise stable patients without comorbid conditions, and continue to use their own judgment when determining the proper disposition of stable patients with rapid AF. If secondary causes of AF are ruled out; follow-up can be arranged; and chest pain, ST changes, CHF, and uncontrolled rate are not present, the patient may be appropriate for discharge [8-13,16,30,99,100].

Conclusions

AF is a common dysrhythmia that may lead to stroke, heart failure, and death. Recent literature has evaluated several components of ED care, including evaluation, rate versus rhythm management, cardiover- sion, anticoagulation, and disposition. The emergency physician should first assess hemodynamic status and evaluate for secondary causes of AF with RVR. Cardioversion is warranted in the patient with hemodynamic instability due to AF. Several studies suggest that cardioversion in the ED may be useful in stable patients under age 65 years with known AF onset b48 h and limited comorbidities, as well as those with negative TEE for intracardiac thrombus or on anticoagulation for 3-4 weeks. Pa- tients with new onset AF may spontaneously revert to sinus rhythm without medical or electrical cardioversion. Other patients warrant rate control, with medication choice based on patient and situational factors. Anticoagulation is an important consideration, with use of scor- ing systems to calculate risk of bleeding and thromboembolism. Clinical judgment regarding disposition is recommended, but some literature supports discharging stable patients in specific circumstances.

Conflicts of interest

None.

Acknowledgements

This manuscript did not utilize any grants, and it has not been pre- sented in abstract form. This clinical review has not been published, it is not under consideration for publication elsewhere, its publication is approved by all authors and tacitly or explicitly by the responsible au- thorities where the work was carried out, and that, if accepted, it will not be published elsewhere in the same form, in English or in any other language, including electronically without the written consent of the copyright-holder. This review does not reflect the views or opinions of

the U.S. government, Department of Defense or its Components, U.S. Ar- my, U.S. Air Force, or SAUSHEC EM Residency Program.

References

AJEM

Emergency medicine considerations in atrial fibrillation

a b s t r a c t

Introduction

Methods

Discussion

Stable patients

Echocardiography in the ED

Rate versus rhythm control

Rhythm control options

Rate control

Considerations in anticoagulation

Unstable patients

Disposition

Conclusions

Camm AJ, Lip GY, De Caterina R, et al. 2012 Focused update of the ESC guidelines for the management of atrial fibrillation. Eur Heart J 2012;33(21):2719-47.

Kuipers S, Klouwenberg PMK, Cremer OL. Incidence, risk factors and outcomes of new onset atrial fibrillation in patients with sepsis: a systematic review. Crit Care 2014;18:688.

Atzema CL, Barrett TW. Managing atrial fibrillation. Ann Emerg Med 2015;65: 532-9.

AFFIRM investigators. Clinical factors that influence response to treatment strate- gies in atrial fibrillation: the atrial fibrillation follow-up investigation of rhythm management (AFFIRM) study. Am Heart J 2005;149(4):645-9.

Okcun B, Yigit Z, Arat A. Comparison of rate and rhythm control in patients with atrial fibrillation and nonischemic heart failure. Japan Heart J 2004;45:591-601.

Michael JA, Stiell IG, Agarwal S, et al. Cardioversion of paroxysmal atrial fibrillation in the emergency department. Ann Emerg Med 1999;33:379-87.

David D, Segni ED, Klein HO, et al. Inefficacy of digitalis in the control of heart rate in patients with chronic atrial fibrillation: beneficial effect of an added beta adren- ergic blocking agent. Am J Cardiol 1979;44:1378-82.

Demircan C, Cikriklar HI, Engindeniz Z, et al. Comparison of the effectiveness of intra- venous diltiazem and metoprolol in the management of rapid ventricular rate in atrial fibrillation. Emerg Med J 2005;22(6):411-4 Erratum in: Emerg Med J 2005;22(10):758.

Hsu JC, Maddox TM, Kennedy K, et al. Aspirin instead of oral anticoagulant pre- scription in atrial fibrillation patients at risk for stroke. J Am Coll Cardiol 2016 Jun 28;67(25):2913-23. https://doi.org/10.1016/j.jacc.2016.03.581.

Guigliano RP, Ruff CT, Braunwald E, et al. Edoxaban versus warfarin in patients with atrial fibrillation. N Engl J Med 2013;369(22):2093-104.

Villines TC, Peacock WF. Safety of direct oral anticoagulants: insights from

Atzema CL, Austin PC, Chong AS, et al. Factors associated with 90 day death after emergency department discharge for atrial fibrillation. Ann Emerg Med 2013; 61(5):539-48.

Leave a Reply Cancel reply

Emergency medicine considerations in atrial fibrillation

a b s t r a c t

Introduction

Methods

Discussion

Stable patients

Echocardiography in the ED

Rate versus rhythm control

Rhythm control options

Rate control

Considerations in anticoagulation

Unstable patients

Disposition

Conclusions

David D, Segni ED, Klein HO, et al. Inefficacy of digitalis in the control of heart rate in patients with chronic atrial fibrillation: beneficial effect of an added beta adren- ergic blocking agent. Am J Cardiol 1979;44:1378-82.

Demircan C, Cikriklar HI, Engindeniz Z, et al. Comparison of the effectiveness of intra- venous diltiazem and metoprolol in the management of rapid ventricular rate in atrial fibrillation. Emerg Med J 2005;22(6):411-4 Erratum in: Emerg Med J 2005;22(10):758.

Hsu JC, Maddox TM, Kennedy K, et al. Aspirin instead of oral anticoagulant pre- scription in atrial fibrillation patients at risk for stroke. J Am Coll Cardiol 2016 Jun 28;67(25):2913-23. https://doi.org/10.1016/j.jacc.2016.03.581.

Leave a Reply Cancel reply

See Also