ATRIAL FIBRILLATION

See also "Cerebrovascular Disease"

I. General

  1. 350-500 atrial impulses/minute; irregular and slower ventricular response rate (usually 140-180 without tx)
  2. Associated with increased risk of embolic CVA and heart failure
  3. Can be associated with Ashmann's phenomenon: aberrant conduction of an impulse after a normal impulse following a pause; usually shows a RBBB pattern; arises from increased refractory period of bundle branches, R>L, after the pause

III. Risk factors

  1. Advanced age
  2. Rheumatic MV disease
  3. CAD
  4. Cardiomyopathy
  5. Thyrotoxicosis
  6. EtOH lowers threshold; cocaine can cause
  7. Pulmonary embolus
  8. Hypoxia
  9. Sepsis
  10. Pericarditis
  11. Atrial myxoma

IV. Clinical presentation: palpitations, no cannon waves, heart sounds vary beat-to-beat

  1. Workup: Cardiac enzymes, TFTs, drug tox-screen, BUN/Cr, LFTs, K, glucose, ABG, EKG for rhythm and ischemic changes, echo for valves, pericardial fluid, LA size (should be <4.0cm), & LVEF

V. Treatment

  1. Initial treatment-If symptomatic, start with rate control (though cardiovert if hemodynamically unstable, etc.-See below)
  1. Dig IV(0.25-0.75 or 0.5 PO x1 then 0.26 PO Q6-8h, max 1mg total dose) to control rate to 70-90/min
  1. Check serum [K], don't give if hypokalemic
  1. Beta-blockers may provide added rate control (Propanolol 0.025-0.1mg/kg IV, then 10-60mg PO Q6h; esmolol 50-200ug/kg/min). Note that some see beta-blockers as first choice, since they may help to convert to SR as well as controlling rate
  2. Also can use calcium-channel blockers for rate control: (verapamil 0.075-0.15mg/kg over 2-3min; can use PO verapamil 80-120Q8h; IV dilt 0.25mg/kg over 20min, then IV gtt 5-15mg/h IV for up to 24h, monitoring HR and BP closely)
  1. DON'T give Ca-blockers IF there is any suspicion of Wolff-Parkinson-White syndrome.
  1. If rate control is not indicated or does not result in spontaneous reversion to sinus rhythm, must choose between approaches of "rhythm control" (i.e. attempt to convert to and maintain sinus rhythm) vs. "rate control" (i.e. leave in Afib with treatment to control heart rate and with anticoagulation; see below)
    1. Rhythm-control approach is not in generall associated with reduced mortality, and requires meds that may be toxic, so is generally reserved for patients for whom the symptoms are troublesome.
  1. Rhythm control
  1. DC electrocardioversion
  1. Generally considered to be urgently indicated in pts with recent-onset AFib with rapid ventricular response and evidence of acute myocardial ischemia, symptomatic hypotension, or heart failure.
  2. Better success with <6mos AF, small LA, MS rather than MR if mitral valve disease exists
  3. Worse success if CHF with NYHA Class III or worse or age >56yo
  4. Can repeat cardioversion if AFib recurs after cardioversion; if so, some recommend adding an antiarrhythmic drug prophylactically
  5. Usual approach is before attempting conversion, anticoagulate if AFib may have been present >48h (to INR 2.0-3.0 x 3wks and continue 4wks post-conversion)
  6. Use of transesophageal echocardriography (TEE) rule out intra-atrial thrombus and clear for immediate cardioversion
    1. 1222 pts w/AFib > 48h all underwent DC cardioversion, randomized to prior tx with warfarin x 3wks vs. (Heparin IV x 24h OR warfarin x 5d then TEE and cardioversion if no thrombi seen; if thrombi seen, warfarin x 3wks then DC cardioversion); all pts got warfarin x 4wks after cardioversion. Rates of embolic events at 8wks were 0.8% in TEE group and 0.5% in conventional group (nonsig.) and total mortality slightly higher in TEE group (2.4% vs. 1.0%, nonsig.); but hemorrhagic events sig. less common in TEE group (2.4% vs. 1.0%). (NEJM 344:1411, 2001--JW)
    2. In an uncontrolled series of 533 pts with AFib > 48h pre-tx'd with heparin to therapeutic PTT; 463 pts had no thrombus on TEE and were cardioverted; only 1 had a clinically significant thromboembolic event; among 278 pts followed x 1y, recurrence rate of AF was 41% for those who had had AF x < 3wks at time of cardioversion vs. 58% for those with AF lasting > 3wks at time of cardioversion (sig.) (AM. J. Med. 110:694, 2001--JW)
  7. Use of TEE rule out intra-atrial thrombus in patients who have been anticoagulated to therapeutic INR x 3wks: In one nonrandomized study (using two consecutive cohorts), a strategy of doing TEE before cardioversion and postponing cardioversion if left atrial thrombus was detected was NOT ass'd with any difference in post-cardioversion embolic events (JACC 39:1436, 2002--abst)
  1. Antiarrhythmic medication for rhythm control-Both IA's and Sotalol may actually increase mortality with long-term use for maintenance of sinus rhythm in AFib (Am. J. Cardiol. 83:1629, 1999--AFP)
  1. Class IA antiarrhythmics
    1. Quinidine (e.g. 200mg PO Q2h x 5 on day 1, 300mg Q2h x 5 on day 2, with 1st dose 7am & last dose 3pm; or 200mg Q4h x 24h then 300mg Q4h x 24h); if doesn't convert after 48h quinidine, use electrocardioversion. Quinidine causes much diarrhea and can cause torsades des pointes
    2. Procainamide--alternative first-line to Quinidine
    3. Disopyramide--3rd-line; has a good side-effect profile
  1. Class IC antiarrhytmics
    1. Propafenone (2nd-line)
    2. Flecainide (3rd-line)--use only if no structural heart disease
  1. Class II antiarrhythmics (beta-blockers)
  1. 394 pts successfully undergoing DC Cardioversion for AFib randomized to Metoprolol XR 100-200mg/d vs. placebo. 6mo relapse rate was sig. less in the metoprolol group (49% VS. 60%) (JACC 36:139, 2000--JW)
  1. Class III antiarrhythmics
    1. Sotalol (combined class II/III)
    2. Amiodarone (Cordarone)-Appears more effective at maintaining sinus rhythm than either Sotalol or Propafenone
      1. 70 pts with symptomatic AFib (chronic or paroxysmal) successfully converted in the ER randomized to amiodarone (loading with 800-1600mg/d x 7-14d then taper to maintenance of 200/d) vs. sotalol (80mg BID titrated to max of 360 BID; dose lowered if pt noted side fx or QTc was found to be > 500ms). At 1y f/u, 70% of amiodarone group c/w 40% of sotalol group were still in sinus rhythm (summary doesn't say if sig.); results not affected by age, sex, LA size, or underlying cardiac disease. MI < 1wk prior, decreased LVEF, unstable renal or hepatic disease, & hyperthyroidism were exclusion criteria (Am. J. Cardiol. 81:995, 1998--AFP)
      2. 403 pts with at least 1 symptomatic episode of AFib in previous 6mos randomized to amiodarone, sotalol, or propafenone. Over mean 15mo f/u, amiodarone group had sig. fewer recurrences of AFib than combined incidence in sotalol & propafenone groups (35% vs. 63%; no diff. between sotalol & propafenone). Amiodarone had sig. fewer CVA's but no sig. diff. in mortality (NEJM 342:913, 2000--JW)
      3. Possible adverse effects include cataract formation and pulmonary fibrosis
    3. Dofetilide-Associated with torsades de pointes (risk about 0.8%, higher in pts with LV dysfunction). Recc'd to start in-hospital with 72h of cardiac monitoring. Start at 500ug BID, less in pts with renal failure or pts who develop QTc > 500msec during tx.
    4. Azimilide-Associated with torsades des pointes
  1. Antiarrhythmics with mixed properties
    1. Dronedarone (Multaq) 400mg BID with meals-Less effective, but also less toxic, than amiodarone
      1. Has properties of antiarrhythmics class I-IV
      2. Chemically similar to amiodarone but has less tissue absorption and may be less toxic with long-term use (i.e. no association with thyroid, pulmonary, or liver disease)
      3. May worsen heart failure
      4. May cause bradycardia, prolongation of QT interval, nausea, diarrhea, rash, and elevated serum Cr levels.
      5. In a meta-analysis of nine trials (4 dronedarone vs. placebo, 4 amiodarone vs. placebo, 1 head-to-head), incidence of recurrent AFib was sig. lower with amiodarone (OR 0.49) but amiodarone recipients had sig. higher incidence of adverse effects requiring drug discontinuation (OR 1.8) and nonsig. higher mortality (OR 1.6) (J. Am. Coll. Cardiol. 54:1089, 2009-JW)

      6. In a study in 504 pts with atrial fibrillation requiring cardioversion randomized to dronedarone vs. amiodarone, over median 7mo f/u, dronedarone group had higher incidence of AFib recurrence (64% vs. 42%) but lower incidence of drug discontinuation due to adverse effects (5% vs. 11%) ("DIONYSOS" studi; J. Cardiovasc. Electrophysiol. 21:597, 2010-JW)  

      7. In a study in 4,628 pts with atrial fibrillation and one of (HTN, DM, prior CVA or TIA, left atrial enlargement, or LVEF of 40% or less) randomized to dronedarone 400mg BID vs. placebo, over mean 21mo f/u, incidence of (first hospitalization for a cardiac event or death from any cause) was sig. lower in dronedarone group (32% vs. 39%) though no sig. diff. in mortality was noted.  Dronedarone recipients had sig. higher incidence of bradycardia, QT prolongation, QI complaints, and elevated serum Cr, but not in evident pulmonary, liver, or thyroid doxicity ("ATHENA" trial; NEJM 360:668, 2009-JW)

  2. A comparison of antiarrhythmics for cardioversion of AF--115 pts with AF onset < 24h previously, randomized to sotalol vs. amiodarone vs. digoxin; reversion to sinus rhythm at 48h was 88%, 77%, and 58%, respectively; difference between first two and dig was sig. (Ann. Emerg. Med. 36:1, 2000--JW)
  3. In a meta-analysis of 44 randomized studies of anti-arrhythmic drugs for maintaining sinus rhythm, AFib recurrence was sig. lower with type IA, IC, and III (except dronedarone) antiarrhythmics c/w controls; amiodarone was sig. better than IA drugs.  Mortality was sig. higher among IA recipients than controls.  (Arch. Int. Med. 166:719, 2006--JW)
  1. Studies of outcomes with rate-control approach vs. rhythm-control approach
    1. 252 pts 18-75yo with symptomatic AFib 7-360d in duration randomized to rate control (diltiazem) vs. rhythm control (pharmacologic and/or DC cardioversion); all pts received anticoagulation.  At 1y, there were no sig. diffs. in % of pts still symptomatic or quality of life scores between the two groups.  Pts in rhythm-control group had sig. higher incidence of drug side effects (64% vs. 47%) (Pharmacological Intervention in Atrial Fibrillation ("PIAF") Trial; Lancet 356:1789, 2000--AFP)
    2. 3060 pts with onset of AF < 6mos previously and (age > 65yo or other risk factors for CVA) randomized to rhythm control (anti-arrhythmic drugs and/or DC cardioversion) vs. rate control (with beta-blockers, Ca-blockers, and/or digoxin); all pts also received warfarin.  Over mean 3.5y f/u, no sig. diff. in all-cause mortality; rhythm-control group had sig. higher rates of hospitalization and drug discontinuation due to adverse effects (NEJM 347:1834, 2002--JW)
    3. 522 pts with recurrent or persistent AF after attempted cardioversion randomized to rhythm control vs. rate control (similar interventions to study above); over mean 2.3y f/u, rhythm-control group had nonsig. higher incidence of (CV death, heart failure, thromboembolism, bleeding, need for pacemaker, or severe drug side f/x) than rate-control group (22.6% vs. 17.2%) (NEJM 347:1883, 2002--JW)
    4. In a secondary analysis of a study in 665 pts with persistent atrial fibrillation randomized to amiodarone, sotalol, or placeo, over 1y f/u, those pts who were in sinus rhythm had sig. better quality of life scores and sig. lower symptom scores than those still in AF.  In the already-published report on this study (NEJM 352:1861, 2005), data was presented showing that amiodarone & sotalol were equally efficacious at converting to sinus rhythm but that amiodarone was assocaited with sig. lower incidence of recurrence of AFib  ("Sotalol Amiodarone Atrial Fibrillation Efficacy Trial ("SAFE-T"), presentation by D.r Steven Singh at the Heart Rhythm Society, reported in Family Practice News, July 2005)
    5. In a study in 192 women and 330 men with persistent atrial fibrillation randomized to rate control (with digoxin, beta-blockers, and calcium-blockers) vs. rhythm control (using sotalol, class IC anti-arrhythmic drugs, and amiodarone), over mean 2.3y f/u, among women, incidence of (cardiovascular death or other cardiovascular event) was sig. higher in rhythm control group (33% vs. 11%); among men there was no sig. diff. ("Rate Control versus Electrical Cardioversion" ("RACE") Study; J. Am. Coll. Cardiol. 46:1298, 2005--JW)
  2. RF catheter ablation
    1. As of 2005, generally considered 2nd-line if no response to cardioversion or drugs
    2. Usually focuses on arrhythmogenic triggers in the pulmonary veins, or electrical pathways between those areas and the left atrium
    3. If the AV node is the target of ablation, requires implantation of a permanent pacemaker
    4. Shown to be associated with improvement in LVEF and symptoms in pts with AF and Heart Failure in nonrandomized studies (NEJM 351:2373, 2004--JW)
  3. Pre-anticoagulate for at least 1mo; switch to unfractionated or low-molecular-weight heparin for the procedure itself and continue anticoagulation for at least 2mos after ablation.
  4. Approximately 40% of patients episodic atrial fibrillation for several weeks after ablation, so it is recommended to continue anticoagulation for at least 2mos post-procedure, longer if other cardiovascular risk factors are present (e.g. indefinite anticoagulation if pt has a "CHADS score: of 2 or more (Congestive heart failure, Hypertension, Age >75 years, Diabetes, previous Stroke or TIA)
  5. Risk factors for failure of ablation (total failure or initial success but recurrence of AF): Persistent AFib, longer duration of AFib, older age, underlying heart disease
  6. 1-2% risk of life-threatening complications, including CVA, cardiac rupture and cardiac tamponade.
    7.
  7. May hold promise as a first-line treatment
    1. In a study in 70 pts 18-75yo with symptomatic atrial fibrillation x > 3mos and no prior anti-arrhythmic Rx randomized to RF ablation vs. anti-arrhythmic drugs; over 1y f/u, RF ablation group had sig. lower incidence of symptomatic AF (13% vs. 63%) and hospitalization (9% vs. 54%); at 6mos, there was sig. greater improvement in quality-of-life scores in RF ablation group (JAMA 293:2634, 2005--abst)
    2. In a study in 198 pts with paroxysmal atrial fibrillation who had failed treatment with antiarrhythmic drugs randomized to further antiarrhythmic drug therapy (flecainide, amiodarone, or sotalol) vs. circumferential pulmonary-vein ablation.  Over 1y, incidence of AF recurrence was sig. lower in ablation group (22% vs. 86%) (J. Am. Coll. Cardiol. 48:2340, 2006--JW)
    3. In a study in 167 pts with AFib not responsive to at least one antiarrhythmic drug randomized to antiarrhythmic drug therapy vs. catheter ablation, after 9mos, incidence of treatment failure was sig. lower with catheter ablation group (33% vs. 84%) and mean uality-of-life scores were higher in catheter ablation recipients (JAMA 303:333, 2010-abst)
  1. Implantable automatic cardioverter (Circ. 98:1651, 1998--JW; Circ. 102:1407, 2000--JW)
  2. Angiotensin Receptor Blockers to reduce risk of adverse outcomes in patients with atrial fibrillation
    1. In a study in 9,016 pts with atrial fibrillation with at least two episodes in prior 6mos randomized to irbesrtan vs. placebo, over median 4.5y f/u, there was no sig. diff. in incidence of (MI, CVA, or vascular death) or recurrent atrial fibrillation (for sthose pts who were in sinus rhythm at study enrollment ("ACTIVE I" trial; NEJM 364:928, 2011-JW).

VI. Risk of CVA in pts with Atrial Fibrillation

  1. RR = 5.0 in pts with AFib, risk in intermittent AFib is similar to that in sustained AFib (JACC 35:183, 2000--AFP)
  1. Risk factors for CVA in pts with AFib (per ACC/AHA guidelines; JACC 48:854, 2006--JW)
    1. "High Risk" factors
      1. Prior cerebral ischemic events (CVA, TIA, or embolism)
      2. Mitral Stenosis
      3. Prosthetic heart valve
    2. "Moderate Risk" factors
      1. Age > 75yo
      2. Hypertension
      3. Diabetes mellitus
      4. Heart Failure
      5. LVEF < 35%
        1. Mod-severe LV systolic dysfunction as measured by transthoracic echo ass'd with OR 2.5 (in a study of 1066 pts with AF, mean age 67, 78% male; LA diameter and septal thickness were not predictive) (Arch. Int. Med. 158:1316, 1998--JW)
  1. Antithrombotics to reduce risk of CVA in pts with atrial fibrillation and no rheumatic heart disease
  1. Comparisons of  Warfarin  vs. Antiplatelets
    1. In a meta-analysis of 6 studies (total n = 4052) of warfarin vs. aspirin in pts with nonrheumatic AF, warfarin was ass'd with sig. lower incidence of CVA (HR 0.55) and cardiovascular events (HR 0.71), but sig. higher incidence of major bleeding (HR 1.71).  Pts with paroxysmal AF had similar risk reductions for ischemic CVA (HR 0.32); there were no sig. diffs in overall mortality.  Greatest benefit in CVA reduction if >75yo, male, prior cerebrovascular event, or diabetic (JAMA 288:2441, 2002--abst)
    2. In a study in 6,706 pts with atrial fibrillation and at least one stroke risk factor randomized to warfarin (adjusted to target INR 2.0-3.0) vs. (clopidogrel 75mg/d + aspirin 75-100mg/d), over median 1.3y f/u, the warfarin group had sig. lower incidence of CVA (1.4%/yr vs. 2.4%/yr) and of (CVA, non-CNS embolism, MI, or vascular death; 3.9% vs. 5.6%); no sig. diff. in incidence of major bleeding ("ACTIVE" Trial; Lancet 367: 1903, 2006--JW)
  2. Comparisons of Warfarin + Antiplatelets vs. Warfarin alone
    1. Low-dose unadjusted warfarin (adjusted at start to give INR 1.2-1.5) plus ASA 325mg QD associated with sig. more ischemic CVA than adjusted-dose warfarin (INR 2.0-3.0); no sig. diff in risk of major hemorrhage (Lancet 348:633, 1996-JW)
  3. Comparisons of different antiplatelets
    1. In a study in 7,554 pts pts with atrial fibrillation + at least one CVA risk factor who were considered unsuitable for warfarin, all of whom received aspirin 75-100mg/d, randomized clopidogrel 75mg/d vs. placebo, over median 3.6y f/u, incidence of (CVA, MI, non-CNS systemic embolism, or vascular death) was sig. lower in the clopidogrel group (6.8% vs. 7.6%)  ("ACTIVE A" Trial; NEJM 360:2066, 2009-JW)
  4. Studies comparing various degrees of anticoagulation with Warfarin
    1. In a case-control study, INR < 2.0 ass'd with increased risk of CVA c/w INR of 2.0; INR > 2.0 not ass'd with add'l benefit (NEJM 335:540, 1996-JW)
  5. Comparisons of Warfarin vs. Ximelagatran
    1. In a trial of 254 pts with non-valvular atrial fibrillation and at least one other CVA risk factor (HTN, age > 75yo, prior CVA or TIA, LV dysfunction, or ((DM or CAD) + age > 65yo) randomized to ximelagatran 20-60mg/d  vs. adjusted-dose warfarin (target INR 2.0-3.0) x 12wks, there were no systemic embolic events in either group and no sig. diff. in incidence of adverse events; 4% of ximelagatran pts (and none of the warfarin pts) had transient transaminase elevations ("SPORTIF-II" trial, JACC 41:1445, 2003--JW)
    2. In a trial of 3410 pts with nonvalvular atrial fibrillation and one other CVA risk factor (see summary of SPORTIF-II above) randomized to Ximelagatran 36mg BID vs. adjusted-dose warfarin (target INR 2.0-3.0); over mean 17mo f/u, incidence of (stroke or systemic embolism) was nonsig. lower in ximelagatran group (1.6% vs. 2.3%); no diff. in incidence of death or major bleeding; minor bleeding sig. less common in ximelagatran recipients.  6% of Ximelagatran recipients had ALT elevations but in most of those pts normalized w/continued use ("SPORTIF-III trial"--Lancet 362:1691, 2003--JW)
    3. In a study of 3,922 pts with nonvalvular atrial fibrillation randomized to ximelagatran 36mg BID vs. warfarin (to INR 2.0-3.0), over mean 20mo f/u, incidence of (any CVA or systemic embolic event) and major bleeding were not sig. diff in the two groups, though total bleeding was sig. lower with ximelagatran (37% vs. 47%/yr) (JAMA 293:690, 2005--abst)
  1. Low risk patients may not need anticoagulation
    1. In a prospective cohort uncontrolled study of 892 "low risk" pts (nl LVEF and no h/o CHF; SBP < 160 mm Hg; no prior ischemic cerebral event or embolic episode; age < 75yo if female; no prosthetic heart valves, mitral stenosis, or other conditions requiring anticoagulation) with AFib tx'd with ASA 325mg/d and no warfarin w/avg f/u 2y; risk of CVA or cerebral embolic events was 2.2%/yr; rate of disabling ischemic CVA was 0.8%/yr; rate of major bleeding events was 0.5%/yr (JAMA 279:1273, 1998--JW)
  1. If anticoagulation contraindicated, use ASA 325d
    1. Anticoagulation vs. ASA--In a meta-analysis of 5 randomized trials involving total 3298 pts, anticoagulation was ass'd with nonsignificantly lower risks for CVA death (OR 0.74), vascular death (OR 0.86), and combined fatal & nonfatal events (OR 0.79) (BMJ 322:321, 2001--JW)
  1. ACC/AHA recommendations 2006 (JACC 48:854, 2006--JW)
    1. If pt has no "risk factor" (see above), ASA
    2. If has 1 "moderate risk factor", ASA or warfarin
    3. If has 2 or more moderate risk factors" or 1 or more "high risk factor" then warfarin

(Sources include Core Content Review of Family Medicine, 2012)