• 
• 

Three Part Question

In a patient with [new onset atrial fibrillation] does [electrical or chemical cardioversion] carry an acceptable [risk of thrombo-embolism]?

Clinical Scenario

A 54 year old man attends the Emergency Department with palpitations of 48 hours duration. He is otherwise fit and well. Examination is unrevealing. An ECG shows atrial fibrillation with a ventricular rate of 156 bpm.
You decide to electrically cardiovert him however you are a little concerned as you are aware of guidelines that differ with regards to the duration of atrial fibrillation that carries an acceptable risk of thrombo-embolism on cardioversion.

Search Strategy

Medline 1966 to October 2004 using Ovid Interface
({Exp atrial fibrillation OR atrial fibrillation.mp} AND {electric countershock OR cardioversion.mp OR chemical cardioversion.mp} AND {exp embolism and thrombosis OR embolism.mp OR thrombosis.mp OR thrombo-embolism.mp}).

Search Outcome

430 papers were found of which 4 were relevant. One abstract and one additional paper were found from bibliographies of the relevant papers.

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Arnold 1992	Patients with atrial fibrillation or flutter of >48 hours duration undergoing successful elective electrical cardioversion Excluded if indication for cardioversion was haemodynamically unstable af, unsuccessful cardioversion and duration of arrhythmia <48 hours	Retrospective review of records	Incidence of thromboembolic events in those taking and those not taking anticoagulants at time of cardioversion.	6 thromboembolic events occurred in 179 non-anticoagulated patients undergoing electrical cardioversion for atrial fibrillation. 5 of these events occurred in patients whose duration of af was between 2 and 7 days duration. 2 of these events occurred in post-operative patients. No events occurred in patients who had been anti-coagulated.	Retrospective review of charts of non-randomised patients. Inclusion / exclusion criteria may limit applicability of results.
Stoddard 1995	143 patients with acute atrial fibrillation (less than 3 days duration as determined by symptoms) referred for TOE	Retrospective review.	Presence of LA thrombus in those with and without recent thrombo-embolic event.	LA thrombus found in 21% of patients with acute af and recent thrombo-embolism. LA thrombus found in 13% of patients with acute af and no recent thrombo-embolism.	Acute af defined as < 3 days duration Only those referred for TOE included in study therefore potential for selection bias Thrombus may have formed between time of referral and time of TOE. Low rate of anticoagulation at time of TOE (24%) Clinical significance of LA thrombus not determined.
Mitchell 1997	1210 patients with atrial fibrillation or flutter who underwent cardioversion using ibutilide,	Retrospective	Incidence of thrombo-embolism in first 7 days in those patients with af/flutter of < 48 hours and > 48 hours duration	1 out of 1064 patients with af/flutter of <48 hours duration had a stroke. 3 out of 145 patients with af/flutter >48 hours duration had a stroke. 2 of these 4 patients received heparin on admission.	Patients predominantly had af therefore data on aflutter more limited. Abstract only of retrospective chart review. Numbers too small to support routine heparin on admission.
Weigner 1997	357 patients with atrial fibrillation less than 48 hours duration, admitted via ED in large University Hospital over a five year period, cardioverting to SR. Patients excluded if duration of af >48 hours or indeterminate. Also excluded if presented with thrombo-embolism. Duration of af determined only on symptomatology.	Prospective study identifying patients. Retrospective study re collection of data.	Incidence of thrombo-embolism in those patients cardioverting to SR by any means.	3 patients out of 357 (0.8%) sustained a thrombo-embolic complication, all of who spontaneously cardioverted.	Small incidence of thrombo-embolism prevents meaningful study of usefulness of anti-coagulation before cardioversion. 50% of patients anticoagulated at time of cardioversion. Only included small numbers of patients with rheumatic valve disease or previous thrombo-embolism. Retrospective collection of data. Short term follow up only
			Difference in incidence of thrombo-embolism between those patients receiving and those not-receiving thrombo-embolic prophylaxis before cardioversion	There was no significant difference between those who were and those who weren't anticoagulated.
Gallagher 2002	1950 patients undergoing DC cardioversion (2639 attempts) for atrial fibrillation or flutter. 443 episodes were of less than 48 hours duration as determined by symptoms	Retrospective review of records	Incidence of thrombo-embolism, in those patients with af <48hours duration with and without prolonged anticoagulation	443 episodes, 258 with af, 185 with aflutter. 1 thrombo-embolic complication in this group. No significant difference if anti-coagulated.	Retrospective chart review.

Clinical Bottom Line

There is little good evidence directly addressing this three-part question and what evidence exists is mostly in relation to electrical cardioversion. There is even less evidence of thrombo-embolic risk for chemical cardioversion but this suggests that the risk is similar to that of electrical cardioversion. Most of these studies suggest there is a low risk of thrombo-embolism with cardioversion of atrial fibrillation of less than 48 hours duration. Stoddard found a high prevalence of left atrial thrombus in patients undergoing TOE but this was in a select population. Consensus suggests that if atrial fibrillation is of less than 48 hours duration then cardioversion of atrial fibrillation is safe and confers a low risk of thrombo-embolism. For those patients with risk factors for thromboembolism, delaying cardioversion for further assessment is appropriate.

References

1. Arnold AZ, Mick MJ, Mazurek RP, Loop FD, Trohman RG. Role of prophylactic anticoagulation for direct current cardioversion in patients with atrial fibrillation or atrial flutter. J Am Coll Cardiol 1992;19:851-5.
2. Stoddard MF, Dawkins PR, Prince CR, Ammash NM. Left atrial appendage thrombus is not uncommon in patients with acute trail fibrillation and a recent embolic event: a transoesophageal echocardiographic study. J Am Coll Cardiol 1995;25:452-459.
3. Mitchell MA, Hughes GS, Ellenbogen KA et al, Cardioversion related stroke rates in atrial fibrillation and atrial flutter [abstract]. Circulation 1997; I-453.
4. Weigner MJ, Caulfield TA, Danias PG, Silverman DI, Manning WJ. Risk for clinical thromboembolism associated with conversion to sinus rhythm in patients with atrial fibrillation lasting less than 48 hours. Ann Intern Med 1997;126:615-620.
5. Fuster V, Ryden LE, Asinger RW, Cannom DS, Crijns HJ, Frye RL, et al. ACC/AHA/ESC guidelines for the management of patients with atrial fibrillation. A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. Eur Heart J 2001;22:1852–923.
6. Gallagher MM, Hennessy BJ, Edvardsson N, Hart CM, Shannon MS, Obel OA, Al-Saady NM, Camm AJ. Embolic complications of direct current cardioversion of atrial arrhythmias: Association with low intensity of anticoagulation at the time of cardioversion. J Am Coll Cardiol 2002;40:926-933.