Pii: s0003-4975(99)01267-9

CURRENT REVIEW
Atrial Fibrillation After Cardiac Operation:
Risks, Mechanisms, and Treatment
Charles W. Hogue, Jr, MD, and Mary L. Hyder, MD
Department of Anesthesiology, Washington University School of Medicine, St. Louis, Missouri
Atrial fibrillation (AF) is a common complication of
who may tolerate these drugs. Preliminary investigations
cardiac operations that leads to increased risk for throm-
showing sotalol and amiodarone to be effective in pre-
boembolism and excessive health care resource utiliza-
venting postoperative AF are encouraging, but early data
tion. Advanced age, previous AF, and valvular heart
have been limited to selective patient populations and
operations are the most consistently identified risk fac-
have not adequately evaluated safety. Newer class III
tors for this arrhythmia. Dispersion of repolarization
antiarrhythmic drugs under development may have a
leading to reentry is believed to be the mechanism of
role in the treatment of postoperative AF, but the risk of
postoperative AF, but many questions regarding the
drug-induced polymorphic ventricular tachycardia must
pathophysiology of AF remain unanswered. Treatment is
be considered. Nonpharmacologic interventions under
aimed at controlling heart rate, preventing thromboem-
consideration for the treatment of AF in the nonsurgical
bolic events, and conversion to sinus rhythm. Multiple
setting, such as automatic atrial cardioversion devices
investigations have examined methods of preventing
and multisite atrial pacing, may eventually have a role
postoperative AF, but the only firm conclusions that can
for selected cardiac surgical patients.
be drawn is to avoid ␤-blocker withdrawal after opera-
(Ann Thorac Surg 2000;69:300 – 6)
tion and to consider ␤-blocker therapy for other patients
2000 by The Society of Thoracic Surgeons
Atrial fibrillation (and/or flutter; AF) is a common patients with postoperative AF especially for those with
complication of cardiac operations and an impor-
low cardiac output [1, 2, 10]. Patients developing postop-
tant source of patient morbidity and increased resource
erative AF are hospitalized 3 to 4 days longer than
utilization [1–9]. The incidence of this arrhythmia is
patients remaining in sinus rhythm leading to increased
dependent on definitions (eg, duration, presence of
hospital cost [1–3]. Recent analysis involving 2,417 pa-
symptoms), patient characteristics, type of operation, and
tients having CABG at 24 U.S. medical centers estimated
method of arrhythmia monitoring [1–9]. In a series of
that postoperative AF increased the cost of operation
3,983 patients undergoing cardiac operations at our insti-
by $1,616 per patient [2]. In a single center study, post-
tution, Creswell and colleagues [1] found the incidence of
operative AF was associated with $10,055 to $11,500
AF detected by continuous electrocardiographic teleme-
of additional hospital charges for CABG [3]. Thus, the
try monitoring to be 32% after coronary artery bypass
health economic implications of postoperative AF are
grafting (CABG), 42% after mitral valve replacement,
substantial.
49% after aortic valve replacement, and 62% after com-
bined CABG and valve procedures. Other researchers
Risk Factors
have reported atrial arrhythmias in 27% to 33% of pa-
tients after CABG [2, 3]. Of more concern is the finding
Patient age has consistently been demonstrated to be the
that the frequency of AF may be increasing. In their
most important risk factor for postoperative AF with
series where surgeons and methods of arrhythmia mon-
incidence rates of more than 50% for patients older than
itoring were constant, Creswell and associates [1] found
80 years undergoing CABG compared with 5% for pa-
that the incidence of AF after CABG increased from 26%
tients less than 50 years [1–3]. This association has been
in 1986 to 36% in 1991. The latter finding was attributable
explained to be attributable to age-related structural
to an increase in the mean age of the surgical patients.
changes in the atrium such as dilation, muscle atrophy,
Postoperative AF is usually well tolerated but tachy-
decreased conduction tissue, and fibrosis [11, 12]. Other
cardia and loss of organized atrial contraction may result
investigators have shown prolonged atrial conduction
in hypotension and congestive heart failure in some
detected by routine and signal-averaged electrocardio-
patients. Even when hemodynamically tolerated, postop-
grams and lowered arrhythmia threshold at the time of
erative AF has consequences. The risk for perioperative
operation to be associated with risk for postoperative AF
stroke has been shown to be nearly threefold higher for
[6, 8, 9]. Recently, we have reported that patients devel-
oping AF after CABG have reduced complexity of heart
rate variability compared with patients remaining in
Address reprint requests to Dr Hogue, Department of Anesthesiology,
sinus rhythm and that this finding along with tachycardia
Washington University School of Medicine, 660 S Euclid Ave, Box 8054,
St. Louis, MO 63110; e-mail: hoguec@notes.wustl.edu.
identified risk for AF with high predictive accuracy [13].
2000 by The Society of Thoracic Surgeons
0003-4975/00/$20.00
Published by Elsevier Science Inc
PII S0003-4975(99)01267-9
Ann Thorac Surg
HOGUE AND HYDER
2000;69:300 – 6
POSTOPERATIVE ATRIAL FIBRILLATION
Whether these methods or other electrophysiologic mea-
only reducing cardiac sympathetic tone will be ineffective
surements can be feasibly applied to routine clinical
for preventing postoperative AF in all patients.
settings for identifying AF risk remains to be established.
The reason for the delay in the onset of AF more than 2 to
Many other risk factors for postoperative AF have been
3 days after operation is not clear. One possibility is that the
identified but the results have been inconsistent between
onset of AF is related to an exaggerated inflammatory
studies [1–3, 14, 15]. The latter risks include history of
response especially involving the pericardium [30, 31]. Me-
rheumatic heart disease, left ventricular hypertrophy, hy-
chanical stretching of the atrium can alter cellular electro-
pertension, preoperative digoxin use, obstructive lung dis-
physiologic properties suggesting that increased intravas-
ease, peripheral vascular disease, and increasing aortic
cular volume due to postoperative mobilization of
cross-clamp duration [1–3, 14, 15]. The atria are inade-
interstitial fluid could contribute to the development of AF
quately cooled during hypothermic cardioplegic arrest [16,
[32]. Tachycardia or brief episodes of AF lead to shortening
17]. Early return of atrial electrical activity during aortic
of the atrial effective refractory period (electrophysiologic
cross-clamping is related to atrial conduction abnormalities
remodeling) promoting the maintenance of AF [33–36].
and AF risk after operation in experimental and clinical
Alterations in calcium-handling proteins have been sug-
investigations [16, 17]. Type of cardioplegia, however, does
gested to be an important mechanism for this electrophysi-
not seem to alter the risk for postoperative AF [18, 19].
ologic remodeling [34–37]. Downregulation of mRNA for
L-type calcium channels and for sarcoplasmic reticular
calcium–ATPase have been demonstrated in atrial tissue
Mechanisms
obtained before cardiac operations in patients with preex-
The electrophysiologic mechanism of postoperative AF is
isting AF and perhaps these mechanisms contribute to
believed to be reentry that results from dispersion of
susceptibility to postoperative AF [38]. The hypothesis that
atrial refractoriness [5, 20, 21]. When adjacent atrial areas
postoperative AF is related to altered gene expression is an
have dissimilar or nonuniform refractoriness, a depolar-
attractive explanation for varying individual susceptibility
izing wavefront becomes fragmented as it encounters
and for the time lag between operation and the onset of the
both refractory and excitable myocardium [5, 20, 21]. This
arrhythmia.
allows the wavefront to return and stimulate previously
refractory but now repolarized myocardium leading to
Treatment
incessant propagation of the wavefront or reentry [5, 20,
Ventricular rate control, anticoagulation, and conversion
21]. Currently, there is not an adequate explanation for
to sinus rhythm are the primary goals of the treatment of
why some patients develop postoperative AF whereas
AF. Rate control can be achieved with ␤-adrenergic
others having the same surgical interventions remain in
receptor or calcium channel blocking drugs. Diltiazem
sinus rhythm. Individuals vulnerable to AF are specu-
and verapamil are effective for heart rate control, but the
lated to have the electrophysiologic substrate (nonuni-
former is usually better tolerated especially for patients
form dispersion of atrial refractoriness) before operation
with impaired left ventricular function, whereas use of
that is then aggravated by surgical perturbations [5].
the latter drug can result in hypotension [39, 40]. Digoxin
It is widely believed that enhanced sympathetic ner-
may be a useful drug for slowing heart rate for situations
vous system activity increases susceptibility to postoper-
where ␤-blockers and calcium channel blocking drugs
ative AF [4, 7, 22–25]. Sympathetic activation, however, is
are contraindicated. Digoxin slows atrioventricular con-
highest the first 24 hours after operation, whereas the
duction to a large degree by enhancement of vagal tone.
onset of AF usually occurs between the second and third
In the setting of high sympathetic nervous system drive
postoperative days [1–3, 26]. Furthermore, the atrial elec-
as occurs perioperatively, its efficacy may be limited [26].
trophysiologic effects of autonomic nervous system stim-
Mural thrombus formation is the most serious compli-
ulation are complex. In contrast to the ventricle where
cation of AF. In an autopsy series involving nonsurgical
sympathetic activation decreases and vagal stimulation
patients, 21% of 642 patients with a history of AF had
increases the threshold for tachycardia and fibrillation,
atrial thrombi compared with 2% of control patients [41].
both sympathetic and parasympathetic activation alter
Atrial thrombi were found in 30% of patients with valvu-
atrial refractoriness, possibly contributing to the arrhyth-
lar heart disease-related AF compared with 14% of pa-
mia substrate [27, 28]. Heightened vagal tone has been
tients with nonvalvular-related AF. In this series, thrombi
demonstrated before AF in nonsurgical patients [29].
were twice as likely to occur in the left compared with the
Recently, we evaluated cardiac sympathovagal balance
right atrium. Left atrial thrombus is reported in 13% to
before the onset of AF in patients recovering from CABG
29% of patients undergoing transesophageal echocardi-
[13]. Either higher or lower measures of heart rate vari-
ography and the potential for thrombus formation occurs
ability were observed before AF, a finding consistent with
early after the onset of AF [42– 45]. Multiple clinical trials
divergent autonomic conditions before arrhythmia onset
have shown that anticoagulation reduces the risk of
[13]. The latter findings support the possibility that in
thromboembolic stroke in nonsurgical patients with
some patients heightened sympathetic tone is present
chronic AF, but the use of anticoagulation for these
before AF but in others, either higher vagal tone or
purposes in cardiac surgical patients has not be evalu-
dysfunctional autonomic heart rate control is present
ated with clinical trials [46 – 48]. The use of anticoagula-
before arrhythmia onset [13]. Thus, measures aimed at
tion therapy in the surgical patient is balanced against
HOGUE AND HYDER
Ann Thorac Surg
POSTOPERATIVE ATRIAL FIBRILLATION
2000;69:300 – 6
Table 1. Summary of Trials Evaluating ␤-Adrenergic Receptor Blocking Drugs, Calcium Channel Blocking Drugs, and Type I
Anti-Arrhythmic Drugs for the Prevention of Postoperative Atrial Fibrillation/Flutter (AF)
Frequency AF
Frequency AF
Study Design
Subjects
Treatment Group
Control Group
Andrews et al [22]
Meta-analysis
Digoxin vs placebo
Verapamil vs placebo
␤-Blockers vs placebo
Kowey et al [23]
Meta-analysis
Digoxin vs placebo
␤-Blockers vs placebo
␤-Blockers & Digoxin vs placebo
Laub et al [63]
Procainamide vs placebo
Gold et al [64]
Procainamide vs placebo
Hannes [59]
Diltiazem vs nitroglycerin
Seitelberger et al [60]
Diltiazem vs nitroglycerin
Merrick et al [62]
Propafenone vs atenolol
a p ϭ NS vs control group.
b p Յ 0.05 vs control group.
DB ϭ double-blind;
R ϭ prospectively randomized.
the individual risk of pericardial hemorrhage. Anticoag-
20% to 34% for placebo-treated patients (p Ͻ 0.01) (Table
ulation has been recommended for patients with AF after
1) [22, 23]. These studies have limitations: the data is 10 to
cardiac procedures when the arrhythmia is persistent or
15 years old, the patients were mostly men, had normal
when there is associated valvular heart disease or im-
or mildly impaired left ventricular function, and few
paired left ventricular function [14, 15].
patients had diabetes [22, 23]. There are also limitations
Electrical cardioversion is indicated whenever AF is as-
inherent with meta-analysis such as publication bias
sociated with hemodynamic deterioration. Some forms of
where small negative studies are less likely to be pub-
atrial flutter (atrial flutter rate, ϷϽ 340/min) can be con-
lished than similar sized positive studies (ie, the studies
verted to sinus rhythm with overdrive atrial pacing. There is
used for this type of retrospective analysis) [57]. Interpre-
no consensus regarding when antiarrhythmic drug therapy
tation of these investigations must also consider the
should be started for postoperative AF. Procainamide, ami-
inconsistent methods of arrhythmia monitoring and the
odarone, and propafenone are effective for these purposes,
inconsistent management of patients whom were receiv-
but only the former two drugs are available in parental
ing ␤-blockers before operation in these trials. That is,
formulations in the United States [49–54]. There are little
some patients receiving placebo postoperatively were
data that have compared the safety and efficacy of each of
subjected to ␤-blocker withdrawal [22, 23]. Failure to
these antiarrhythmic drugs in cardiac surgical patients. In a
restart ␤-blockers after CABG has been shown to be
small prospectively randomized trial, amiodarone and
associated with a greater than twofold higher rate of
propafenone were equally effective in converting AF devel-
postoperative AF [58].
oping after CABG to sinus rhythm [54]. Conversion was
Other pharmacologic approaches evaluated for the
more delayed with amiodarone (19% at 1 hour versus 83%
prevention of postoperative AF are listed in Table 1.
at 24 hours), but control of ventricular rate was observed
Digoxin and verapamil are no more effective than pla-
within 10 minutes. The use of sotalol for treatment of
cebo (Table 1) [22]. A nonblinded, single center study
postoperative AF has also been reported, but a parental
showed that diltiazem when given for 24 hours after
form of this drug is not available in the U.S. and its use was
operation reduced the frequency of AF [59, 60]. In light of
associated with a high frequency of hypotension [55]. Ibuti-
experimental data suggesting that calcium channel
lide is a pure class III antiarrhythmic agent (prolongation of
blockers promote the development of AF, further inves-
repolarization) that received Food and Drug Administra-
tigations of the efficacy of diltiazem seem warranted [61].
tion approval in 1996. The efficacy of ibutilide for converting
Small trials have produced inconsistent results on the
AF has been demonstrated, but these data were derived
efficacy of procainamide in preventing postoperative AF
from clinical trials that did not include cardiac surgical
and propafenone was found to be no more effective than
patients [56]. Although intravenous ibutilide has a rapid
atenolol for AF prophylaxis [62– 64].
onset of action and is hemodynamically well tolerated, a
Stalol and amiodarone possess both membrane-
concern with its use was the high rate of polymorphic
stabilizing properties (class III effects) and ␤-blocking
ventricular tachycardia (8% of patients receiving ibutilide).
properties that make them appealing for AF prophylaxis.
Sotalol has been shown to be effective for this use (Table
Arrhythmia Prophylaxis
2) [9, 25, 65– 67]. Many of these trials, however, had
␤-Blockers have been the most widely studied drugs for
open-labeled, nonblinded study design and the methods
the prevention of postoperative AF [4, 22–25]. Separate
of monitoring the electrocardiogram were inconsistent.
meta-analyses have shown that the frequency of AF after
Furthermore, the management after operation for pa-
cardiac operations in patients receiving ␤-blockers was
tients receiving preoperative ␤-blocker therapy was not
collectively 9% to 10% compared with incidence rates of
always clearly defined [9, 25, 65– 67]. Finally, in some of
Ann Thorac Surg
HOGUE AND HYDER
2000;69:300 – 6
POSTOPERATIVE ATRIAL FIBRILLATION
Table 2. Clinical Investigations of Sotalol for the Prevention of Atrial Fibrillation After Cardiac Surgery
Incidence of Atrial
Drug/Dose
ECG Monitoring
Fibrillation
Jansen et al [65]
Metoprolol: 0.1 mg/kg IV
Telemetry ECG for 48 h
Metoprolol: 15.3%a
then 50 mg TID, PO
then 12 lead ECG TID.
Sotalol: 2.4%a,b
Sotalol: 0.3 mg/kg IV; 80 mg TID, PO
Controls: 36%
Controls: routine care
Suttorp et al [25]
Sotalol: 40 mg TID, PO
Telemetry ECG for 60 h
Sotalol 40 mg: 13.9%
Sotalol: 80 mg TID, PO
then 12 lead ECG with
Sotalol 80 mg: 10.9%
Propranolol: 10 mg QID, PO
symptoms
Propanolol 10 mg: 18.8%
Propranolol: 20 mg QID, PO
Propranolol 20 mg: 13.7%
Nystrom et al [66]
Sotalol: 160 mg BID, PO
Telemetry for 48 h; daily
Sotalol: 10%a
Controls: half-dose of preoperative
12 lead ECG’s
Controls: 29%
␤-blocker
Suttorp et al [67]
Sotalol: 40 mg QID, PO
Telemetry for 60 h; 12
Sotalol: 16%a
Controls: Placebo
lead ECG with symptoms
Controls: 33%
Weber et al [9]
Sotalol: 80 mg BID, PO
Telemetry 24 h; 12 lead
Sotalol: 26%a
Controls: Placebo
ECG with symptoms
Controls: 44%
a p Ͻ 0.05 vs controls.
b p Ͻ 0.05 vs ␤-adrenergic blocking drug.
BID ϭ 2 times a day;
PO ϭ orally;
QID ϭ 4 times a day;
TID ϭ 3 times a day.
the trials the frequency of postoperative AF was no
because most patients either do not develop the arrhyth-
different between patients receiving sotalol versus a
mia or it is transient and of little consequence. ␤-Blockers
␤-blocker, questioning whether any prophylactic effect of
are usually well tolerated but clinically important side
sotalol was as a result of the membrane-stabilizing effects
effects requiring discontinuation of sotalol were reported
or ␤-blocking properties of the drug [25].
in 13% of patients after cardiac operation [9, 25]. Toxicity
Early data on the efficacy of amiodarone for the pre-
(pulmonary fibrosis, worsening of heart failure, hypothy-
vention of postoperative AF have been inconsistent (Ta-
roidism, and hepatic toxicity) occurs in 5% to 10% of
ble 3) [68, 69]. Daoud and colleagues [70], however, found
patients receiving chronic amiodarone therapy, but the
that amiodarone therapy beginning 1 week before car-
lower doses and shorter duration of treatment for peri-
diac operation and continued until hospital discharge
operative arrhythmia control is better tolerated [71–74].
reduced the incidence of AF compared with placebo (25%
In a small series, preoperative low-dose amiodarone
versus 53%, p ϭ 0.003). The management of postoperative
(mean dose, 205 Ϯ 70 mg/day) treatment did not increase
␤-blockers for the 30% of placebo patients receiving these
the risk of pulmonary toxicity but it was associated with
drugs preoperatively is not clear. In the placebo group,
a higher need for inotropic support after cardiac opera-
postoperative AF was more frequent for patients receiv-
tion [73]. Proarrhythmic side effects of antiarrhythmic
ing preoperative ␤-blockers compared with patients not
drugs is a bigger concern. In nonsurgical patients, there
receiving these drugs before operation (61% versus 33%,
is a threefold increased risk for life-threatening ventric-
p ϭ 0.09). In the amiodarone group, the frequency of
ular arrhythmias with class I antiarrhythmics [75]. The
postoperative AF for patients receiving preoperative
proarrhythmic risk associated with sotalol and amioda-
␤-blockers was 27%.
rone is believed to be less but these life-threatening side
effects have been reported in 4.3% to 5.9% of patients
receiving sotalol after myocardial infarction and from
Potential Antiarrhythmic Toxicity
less than 1% to 2% for patients receiving amiodarone [76,
The risk/benefit ratio for administering drugs for AF
77]. Data regarding the safety of sotalol and amiodarone
prophylaxis is different than for arrhythmia treatment
for the prevention of postoperative AF from adequately
Table 3. Clinical Investigations of Amiodarone for the Prevention of Atrial Fibrillation After Cardiac Surgery
Incidence of Atrial
ECG Monitoring
Fibrillation
Hohnloser et al [68] 77 300 mg over 2 h after surgery; 1.2 gm/24
Holter for 48 h then
Amidarone: 5%a
h ϫ 2 days; 900 mg/24 h ϫ 2 days
12 lead ECG with symptoms Control: 21%
Butler et al [69]
12 15 mg/kg after cross-clamp is removed; 200 Holter ϫ 6 days
Amiodarone: 10%b
mg TID ϫ 5 days, PO
Controls: 20%
Daoud et al [70]
12 200 mg TID, PO for 7 days before
Telemetry ϫ 7 days
Amiodaronec: 23% & 25%a
surgery; 200 mg PO daily after surgery
Controls: 42% & 53%
until discharge
a p Ͻ 0.05 vs control.
b p ϭ no significance vs control.
c Frequency of atrial fibrillation during hospitalization and combined for hospitalization and
after discharge, respectively.
PO ϭ orally;
TID ϭ 3 times a day.
HOGUE AND HYDER
Ann Thorac Surg
POSTOPERATIVE ATRIAL FIBRILLATION
2000;69:300 – 6
powered studies are not available, especially for patients
ments for AF. Overdrive atrial pacing can reduce the
at high risk for proarrhythmia (eg, impaired ventricular
frequency of AF in patients with sick sinus syndrome but its
function or myocardial ischemia) [76, 77].
efficacy after CABG has not been confirmed [88]. Pacing the
right and left atrium simultaneously (dual site pacing)
Atrial Fibrillation Prophylaxis and Patient
reduces the recurrence of AF in nonsurgical patients with
Outcomes
intraatrial conduction abnormalities possibly by reducing
dispersion of refractoriness [89]. A randomized study in
Patients most susceptible to postoperative AF often have
patients undergoing CABG found a trend for a lower
other characteristics associated with surgical complica-
frequency of AF in patient having biatrial pacing, especially
tions and longer hospitalization (ie, the elderly, patients
for patients receiving ␤-blockers [90].
with chronic lung disease, peripheral vascular disease,
Implantable automatic atrial defibrillator systems sim-
prolonged aortic cross-clamp time). A possibility exist
ilar to those used to terminate ventricular arrhythmias
that in some situations AF is merely a marker and not
are under investigation for nonsurgical patients with
necessarily the cause for other morbidity and higher
recurring AF but energy levels needed for successful
hospital cost. Few investigators have examined whether
atrial cardioversion (Ϸ1 to 5 J) are associated with dis-
drug prophylaxis for AF improves outcomes. Kowey and
comfort [91]. Innovations in electrode configurations and
colleagues [4] found that acebutolol and digitalis led to a
energy delivery characteristics result in lowered atrial
lower frequency of AF compared with digitalis alone, but
defibrillation thresholds to levels that are tolerable [92,
this treatment did not lead to shorter duration of hospi-
93]. The feasibility of low-energy cardioversion of AF
talization or reduced cost of cardiac operation. The data
with temporary epicardial wire electrodes after cardiac
of Daoud and associates [70] reiterates that patients with
operations has been demonstrated [94]. This latter report
AF have longer hospitalization and higher hospital cost,
and other data from earlier trials in nonsurgical patients
but whether prophylactic amiodarone therapy positively
with AF suggest that, with further advances, temporary
improved patient outcomes was not clearly demonstrated.
automatic atrial defibrillator/pacing systems using tem-
porary epicardial leads could conceivably be developed
Other and Developing Treatments
for the treatment of postoperative AF. This approach
Total and ionized serum magnesium concentrations are
could lead to early termination of postoperative AF,
reduced by cardiopulmonary bypass [78]. Although hy-
lowering the risk of thromboembolism, as well as allow-
pomagnesemia may be related to supraventricular ar-
ing for early pharmacologic therapy.
rhythmias after cardiac operation, it is not clear whether
magnesium replacement reduces this risk [71– 81]. Car-
Conclusions
diopulmonary bypass also results in an euthyroid sick
state [82]. Preliminary trials in patients with reduced left
Postoperative AF is a frequent complication of cardiac
ventricular function undergoing CABG and receiving
operations that increases health care resource utilization
thyroid hormone to improve cardiac performance found
and is associated with other serious adverse events.
a lower incidence of AF in patients receiving triiodothy-
Treatment AF is aimed at ventricular rate control, anti-
ronine compared with controls [83].
coagulation, and restoration of sinus rhythm. At present
Pharmacologic prolongation of atrial repolarization
there is a lack of consensus regarding routine prophy-
(class III effect) is an effective antifibrillatory strategy.
laxis for this arrhythmia other than resuming ␤-blocker
Several new class III antiarrhythmic agents lacking auto-
therapy early after operation and, possibly starting
nomic blocking properties and other toxicity of sotalol
␤-blockers in other patients who may tolerate these
and amiodarone are under development and these com-
drugs [14, 15]. Perhaps, in the future, nonpharmacologic
pounds may have eventual usefulness for patients un-
methods of either AF prevention or early, automatic
dergoing cardiac operations. As a class, these drugs
cardioversion will contribute to improved treatment or
prolong repolarization, refractoriness, increase ventricu-
prevention of this arrhythmia.
lar fibrillation threshold, and slow the rate of ventricular
tachycardia [84]. For the most part, pure class III com-
References
pounds do not have negative inotropic effects but they do
have proarrhythmic potential [84]. These agents have
1. Creswell LL, Schuessler RB, Rosenbloom M, et al. Hazards
actions on different membrane currents such as specific
of postoperative atrial arrhythmias. Ann Thorac Surg 1993;
blockade of the delayed rectifier potassium current (eg,
56:539– 49.
2. Mathew JP, Parks R, Savino JS, et al. Atrial fibrillation
dofetilide, sematilide, d-sotalol), augmentation of the
following coronary artery bypass graft surgery. JAMA 1996;
inactivated sodium channel (eg, ibutilide), or semiselec-
276:300– 6.
tive blockade of the slow component of the delayed
3. Aranki SF, Shaw DP, Adams DH, et al. Predictors of atrial
rectifier potassium current (eg, azimilide) [84 – 86]. Clini-
fibrillation after coronary artery surgery: current trends and
impact on hospital resources. Circulation 1996;94:390–7.
cal trials in nonsurgical patients using ibutilide and
4. Kowey PR, Dalessandro DA, Herbertson R, et al. Effective-
dofetilide have reported conversion rates of more than
ness of digitalis with or without acebutolol in preventing
30% for atrial fibrillation and more than 50% for atrial
atrial arrhythmias after coronary artery surgery. Am J Car-
flutter of relatively recent onset [56, 87].
diol 1997;79:1114–7.
5. Cox JL. A perspective on postoperative atrial fibrillation in
Concerns about drug proarrhythmic side effects have
cardiac operations. Ann Thorac Surg 1993;56:405–9.
sparked interest in developing nonpharmacologic treat-
6. Steinberg JS, Zelenkofske S, Wong S-C, et al. Value of the
Ann Thorac Surg
HOGUE AND HYDER
2000;69:300 – 6
POSTOPERATIVE ATRIAL FIBRILLATION
P-wave signal-averaged ECG for predicting atrial fibrillation
of the complement system during and after cardiopulmo-
after cardiac surgery. Circulation 1993;88:2618–22.
nary bypass surgery: postsurgery activation involves C-re-
7. Kalman JM, Munawar M, Howes LG, et al. Atrial fibrillation
active protein and is associated with postoperative arrhyth-
after coronary artery bypass grafting is associated with
mia. Circulation 1997;96:3542– 8.
sympathetic activation. Ann Thorac Surg 1995;60:1709–15.
32. Mansourati J, Le Grand B. Transient outward currents in
8. Lowe JE, Hendry PJ, Hendrickson SC, Wells R. Intraopera-
young and adult diseased human atria. Am J Physiol 1993;
tive identification of cardiac patients at risk to develop
265:H1466 –70.
postoperative atrial fibrillation. Ann Surg 1991;213:388–91.
33. Wijffels MCEF, Kirchhof CJHF, Dorland R, Allessie MA.
9. Weber UK, Osswald S, Huber M, et al. Selective versus
Atrial fibrillation begets atrial fibrillation: a study in awake
non-selective antiarrhythmic approach for prevention of atrial
chronically instrumented goats. Circulation 1995;92:1954– 68.
fibrillation after coronary surgery: is there a need for pre-
34. Kumagai K, Akimitsu S, Kawahira K, et al. Electrophysiolog-
operative risk stratification? Eur Heart J 1998;19:794–800.
ical properties in chronic lone atrial fibrillation. Circulation
10. Taylor GJ, Malik SA, Colliver JA, et al. Usefulness of atrial
1991;84:1662– 8.
fibrillation as a predictor of stroke after isolated coronary
35. Daoud EG, Bogun F, Goyal R, et al. Effect of atrial fibrillation on
artery bypass grafting. Am J Cardiol 1987;60:905–7.
atrial refractoriness in humans. Circulation 1996;94:1600–6.
11. Kitzman DW, Edwards WD. Age-related changes in the anat-
36. Gaspo R, Bosch RF, Talajic M, Nattel S. Functional mechanisms
omy of the normal human heart. J Gerontol 1990;45:M33–9.
underlying tachycardia-induced sustained atrial fibrillation in
12. Lie JT, Hammond PI. Pathology of the senescent heart:
a chronic dog model. Circulation 1997;96:4027–35.
anatomic observations on 237 autopsy studies of patients of
37. Yue L, Feng J, Gaspo R, Li GR, Wang Z, Nattel S. Ionic
90 to 105 years old. Mayo Clin Proc 1988;63:552– 64.
remodeling underlying action potential changes in a canine
13. Hogue CW Jr, Domitrovich PP, Stein PK, et al. RR interval
model of atrial fibrillation. Circ Res 1997;81:512–25.
dynamics before atrial fibrillation in patients after coronary
38. Lai L, Su M, Lin J, et al. Down-regulation of L-type calcium
artery bypass surgery. Circulation 1998;98:429–34.
channel and sarcoplasmic reticular Ca2ϩ–ATPase mRNA in
14. Page´ P, Pym J. Atrial fibrillation following cardiac surgery.
human atrial fibrillation without significant change in the
Can J Cardiol 1996;12:40A–56.
mRNA of ryandodine receptor, calsequestrin and posphola-
15. Ommen SR, Odell JA, Stanton MS. Atrial fibrillation after
man. J Am Coll Cardiol 1999;33:1231–7.
cardiothoracic surgery. N Engl J Med 1997;336:1429–34.
39. Heywood JT. Calcium channel blockers for heart rate control
16. Smith PK, Buhrman WC, Levett JM, et al. Supraventricular
in atrial fibrillation complicated by congestive heart failure.
conduction abnormalities following cardiac operations: a
Can J Cardiol 1995;11:823– 6.
complication of inadequate atrial preservation. J Thorac
40. Iberti TJ, Benjamin E, Paluch TA, Gentili DR, Gabrielson GV.
Cardiovasc Surg 1983;85:105–15.
Use of constant-infusion verapamil for the treatment of
17. Mullen JC, Khan N, Weisel RD, et al. Atrial activity during
postoperative supraventricular tachycardia. Crit Care Med
cardioplegia and postoperative arrhythmias. J Thorac Car-
1986;14:283– 4.
diovasc Surg 1987;94:558– 65.
41. Aberg H. Atrial fibrillation. 1. A study of atrial thrombosis
18. The Warm Heart Investigators. Randomized trial of normo-
and systemic embolism in a necropsy material. Acta Med
thermic versus hypothermic coronary bypass surgery. Lan-
Scand 1969;185:373–9.
cet 1994;343:559– 63.
42. Mugge A, Daniel WG, Haverich A, Lichtlen PR. Diagnosis of
19. Butler J, Chong JL, Rocker GM, et al. Atrial fibrillation after
noninfective cardiac mass lesions by two-dimensional echo-
coronary artery bypass grafting: a comparison of cardiople-
cardiography: comparison of the transthoracic and trans-
gia versus intermittent aortic cross-clamping. Eur J Cardio-
esophageal approaches. Circulation 1991;83:70– 8.
thorac Surg 1993;7:23–5.
43. Mugge A, Kuhn H, Daniel WG. The role of transesophageal
20. Cox JL, Canavan TE, Schuessler RB, et al. The surgical
echocardiography in the detection of left atrial thrombi.
treatment of atrial fibrillation. II. Intraoperative electro-
Echocardiography 1993;10:405–17.
physiologic mapping and description of the electrophysi-
ologic basis of atrial flutter and atrial fibrillation. J Thorac
44. Brown J, Sadler DB. Left atrial thrombi in non-rheumatic
Cardiovasc Surg 1991;101:406–26.
atrial fibrillation: assessment of prevalence by transesopha-
21. Konings KTS, Kirchhof CJHJ, Smeets JRLM, et al. High-
geal echocardiography. Int J Card Imaging 1993;9:65–72.
density mapping of electrically induced atrial fibrillation in
45. Stoddard MF, Dawkins PR, Prince CR, Ammash NM. Left
humans. Circulation 1994;89:1665– 80.
atrial appendage thrombus is not uncommon in patients
22. Andrews TC, Reimold SC, Berlin JA, et al. Prevention of
with acute atrial fibrillation and a recent embolic event: a
supraventricular arrhythmias after coronary artery bypass
transesophageal echocardiographic study. J Am Coll Cardiol
surgery: a meta-analysis of randomized control trials. Circu-
1995;25:452–9.
lation 1991;84(Suppl 3):236– 44.
46. Petersen P, Godtfredssen J, Boysen G, Andersen ED,
23. Kowey PR, Taylor JE, Rials SJ, Marinchak RA. Meta-analysis
Andersen B. Placebo-controlled, randomised trial of warfa-
of the effectiveness of prophylactic drug therapy in prevent-
rin and aspirin for prevention of thromboembolic complica-
ing supraventricular arrhythmia early after coronary artery
tions in chronic atrial fibrillation: the Copenhagen AFASK
bypass grafting. Am J Cardiol 1992;69:963–5.
study. Lancet 1989;1:175–9.
24. Boudoulas H, Snyder GL, Lewis RP, et al. Safety and ratio-
47. The Boston Area Anticoagulation Trial for Atrial Fibrillation
nale for continuation of propranolol therapy during coro-
Investigators. The effect of low-dose warfarin on the risk of
nary bypass operation. Ann Thorac Surg 1978;26:222–9.
stroke in patients with nonrheumatic atrial fibrillation.
25. Suttorp MJ, Kingma JH, Gin MTJ, et al. Efficacy and safety of
N Engl J Med 1990;323:1505–11.
low- and high-dose sotalol versus propranolol in the pre-
48. Stroke Prevention in Atrial Fibrillation Investigators. Stroke
vention of supraventricular tachyarrhythmias early after
Prevention in Atrial Fibrillation Study: final results. Circula-
coronary artery bypass operations. J Thorac Cardiovasc Surg
tion 1991;84:527–39.
1990;100:921– 6.
49. Hjelms E. Procainamide conversion of acute atrial fibrilla-
26. Reeves JF, Karp RB, Buttner EE, et al. Neuronal and adre-
tion after open-heart surgery compared with digoxin. Scan
nomedullary catecholamine release in response to cardio-
J Thorac Cardiovasc Surg 1992;26:193– 6.
pulmonary bypass in man. Circulation 1982;66:49–55.
50. Gentili C, Giordano F, Alois A, Massa E, Bianconi L. Efficacy
27. Levy MN. Sympathetic-parasympathetic interactions in the
of intravenous propafenone in acute atrial fibrillation com-
heart. Circ Res 1971;29:437– 45.
plicating open-heart surgery. Am Heart J 1992;123:1225– 8.
28. Lewis T, Drury AN, Bulger HA. Observations upon flutter
51. Cochrane AD, Siddins M, Rosenfeldt FL, et al. A comparison
and fibrillation. VII: the effect of vagal stimulation. Heart
of amiodarone and digoxin for treatment of supraventricular
1921;8:141– 69.
arrhythmias after cardiac surgery. Eur J Cardiothoracic Surg
29. Coumel P. Heart rate variability and the onset of tachyar-
1994;8:194– 8.
rhythmias. G Ital Cardiol 1992;22:647–54.
52. Chapman MJ, Moran JL, O’Fathartaigh MS, Peisach AR,
30. Chidambaram M, Akhtar MJ, al-Nozha M, al-Saddique A.
Cunningham DN. Management of atrial tachyarrhythmias
Relationship of atrial fibrillation to significant pericardial
in the critically ill: a comparison of intravenous procain-
effusion in valve-replacement patients. Thorac Cardiovasc
amide and amiodarone. Int Care Med 1993;19:48–52.
Surg 1992;40:70–3.
53. McAlister HF, Luke RA, Whitlock RM, Smith WM. Intrave-
31. Bruins P, te Velthuis H, Yazdanbakhsh AP, et al. Activation
nous amiodarone bolus versus oral quinidine for atrial
HOGUE AND HYDER
Ann Thorac Surg
POSTOPERATIVE ATRIAL FIBRILLATION
2000;69:300 – 6
flutter and fibrillation after cardiac operations. J Thorac
74. Herndon JC, Cook AO, Ramsay MAE, et al. Postoperative
Cardiovasc Surg 1990;99:911– 8.
unilateral pulmonary edema: possible amiodarone pulmo-
54. DiBiasi P, Scrofani R, Paje A, Cappiello E, Mangini A, Santoli
nary toxicity. Anesthesiology 1992;76:308–12.
C. Intravenous amiodarone vs propafenone for atrial fibril-
75. Ruskin JN. The cardiac arrhythmia suppression trial (CAST).
lation and flutter after cardiac operation. Eur J Cardiothorac
N Engl J Med 1989;321:386– 8.
Surg 1995;9:587–91.
76. Soyka LF, Wirtz C, Spangenberg RB. Clinical safety profile of
55. Campbell TJ, Gavaghan TP, Morgan JJ. Intravenous sotalol
sotalol in patients with arrhythmias. Am J Cardiol 1990;65:
for the treatment of atrial fibrillation and flutter after cardio-
74A– 81.
pulmonary bypass: comparison with disopyramide and
77. Hohnloser SH, Klingenheben T, Singh BN. Amiodarone-
digoxin in a randomised trial. Br Heart J 1985;54:86–90.
associated proarrhythmic effects: a review with special ref-
56. Stambler BS, Wood MA, Ellenbogen KA, et al. Efficacy and
erence to torsades de pointes tachycardia. Ann Intern Med
safety of repeated intravenous doses of ibutilide for rapid
1994;121:529–35.
conversion of atrial flutter or fibrillation. Circulation 1996;94:
78. England MR, Gordon G, Salem M, et al. Magnesium admin-
1613–21.
istration and dysrhythmias after cardiac surgery. J Am Med
57. Bailar JC. The promise and problems of meta-analysis.
Assc 1992;268:2395– 402.
N Engl J Med 1997;337:559– 60.
79. Karmy-Jones R, Hamilton A, Dzavik V, et al. Magnesium
58. Ali IM, Sanalla AA, Clark V. Beta-blocker effects on postoper-
sulfate prophylaxis after cardiac operations. Ann Thorac
ative atrial fibrillation. Eur J Cardiothorac Surg 1997;11:1154–7.
Surg 1995;59:502–7.
59. Hannes W, Fasol R, Zajonc H. Diltiazem provides anti-
80. Parikka H, Toivonen L, Pellinen T, et al. The influence of
ischemic and anti-arrhythmic protection in patients under-
intravenous magnesium sulphate on the occurrence of atrial
going coronary bypass grafting. Eur J Cardiothorac Surg
fibrillation after coronary artery bypass operation. Eur
1993;7:239– 45.
Heart J 1993;14:251– 8.
60. Seitelberger R, Hannes W, Gleichauf M, Keilich M, Chris-
81. Wistbacka JM, Koistinen J, Karlqvist KEV, et al. Magnesium
toph M, Fasol R. Effects of diltiazem on perioperative isch-
substitution in elective coronary artery surgery: a double-blind
emia, arrhythmias, and myocardial function in patients
clinical study. J Cardiothorac Vasc Anesth 1995;9:140–6.
undergoing elective coronary bypass grafting. J Thorac Car-
82. Klemperer JD, Klein I, Gomez M, et al. Thyroid hormone
diovasc Surg 1994;107:811–21.
treatment after coronary artery bypass surgery. N Engl
61. Shenasa M, Kus T, Fromer M, LeBlanc RA, Dubuc M,
J Med 1995;333:1522–7.
Nadeau R. Effect of intravenous and oral calcium antagonists
83. Klemperer JD, Klein IL, Ojamaa K, et al. Triiodothyronine
(diltiazem and verapamil) on sustenance of atrial fibrillation.
therapy lowers the incidence of atrial fibrillation after car-
Am J Cardiol 1988;62:403–7.
diac operations. Ann Thorac Surg 1996;61:1323–9.
62. Merrick AF, Odom NJ, Keenan DJM, Grotte GJ. Comparison
84. Singh BN. Antiarrhythmic drugs: a reorientation in light of
of propafenone to atenolol for the prophylaxis of postcar-
recent developments in the control of disorders of rhythm.
diotomy supraventricular tachyarrhythmias: a prospective
Am J Cardiol 1998;81:3D–13.
trial. Eur J Cardiothorac Surg 1995;9:146–9.
85. Colatsky TJ, Follmer CH, Starmer CF. Channel specificity in
63. Laub GW, Janeira L, Muralidharan S, et al. Prophylactic
procainamide for prevention of atrial fibrillation after coro-
antiarrhythmic drug action: mechanism of potassium-
nary artery bypass grafting: a prospective, double-blind,
channel block and its role in suppressing and aggravating
randomized, placebo-controlled pilot study. Crit Care Med
cardiac arrhythmias. Circulation 1990;82:2235– 42.
1993;21:1474– 8.
86. Fermini B, Jurkiewicz NK, Jow B, et al. Use-dependent effect of
64. Gold MR, O’Gara PT, Buckley MJ, DeSanctis RW. Efficacy
the class III antiarrhythmic agent NE-10064 (azimilide) on
and safety of procainamide in preventing arrhythmias after
cardiac repolarization: block of delayed rectifier potassium and
coronary artery bypass surgery. Am J Cardiol 1996;78:975–9.
L-type calcium channel. J Cardiovasc Pharmacol 1995;26:
65. Janssen J, Loomans L, Harink J, et al. Prevention and
259–71.
treatment of supraventricular tachycardia shortly after cor-
87. Falk RH, Pollak A, Singh BN, Friedrich T. Intravenous
onary bypass grafting: a randomized open trial. Angiology
dofetilide, a class III antiarrhythmic agent, for the termina-
1986;37:601–9.
tion of sustained atrial fibrillation or flutter. J Am Coll
66. Nystrom UJ, Edvardsson N, Berggren H, et al. Oral sotalol
Cardiol 1997;29:385–90.
reduces the incidence of atrial fibrillation after coronary artery
88. Chung MK, Augostini RS, Asher CR, et al. A randomized,
bypass surgery. J Thorac Cardiovasc Surg 1993;41:34–7.
controlled study of atrial overdrive pacing for the prevention
67. Suttorp MJ, Kingma JH, Peels HOJ, et al. Effectiveness of
of atrial fibrillation after coronary artery bypass surgery.
sotalol in preventing supraventricular tachyarrhythmias
Circulation 1996;94(Suppl 1):1102.
shortly after coronary artery bypass grafting. Am J Cardiol
89. Saksena S, Prakash A, Hill M, et al. Prevention of recurrent
1991;68:1163–9.
atrial fibrillation with chronic dual site right atrial pacing.
68. Hohnloser SH, Meinertz T, Dammbacher T, et al. Electro-
J Am Coll Cardiol 1996;28:687–94.
cardiographic and antiarrhythmic effects of intravenous
90. Mittleman RS, Hill MRS, Mehra R, et al. Evaluation of the
amiodarone: results of a prospective, placebo-controlled
effectiveness of right atrial and biatrial pacing for the pre-
study. Am Heart J 1991;121:89–95.
vention of atrial fibrillation after coronary artery bypass
69. Butler J, Harriss DR, Sinclair M, et al. Amiodarone prophy-
surgery. Circulation 1996;94(Suppl 1):389.
laxis for tachycardias after coronary artery surgery: a ran-
91. Le´vy S, Ricard P, Gueunoun M, et al. Low-energy cardiover-
domized, double blind, placebo controlled trial. Br Heart J
sion of spontaneous atrial fibrillation: immediate and long-
1993;70:56– 60.
term results. Circulation 1997;96:253–9.
70. Daoud EG, Strickberger SA, Man KC, et al. Preoperative
92. Cooper RAS, Alferness CA, Smith WM, et al. Internal
amiodarone as prophylaxis against atrial fibrillation after
cardioversion of atrial fibrillation in sheep. Circulation 1993;
heart surgery. N Engl J Med 1997;337:1785–91.
87:1673– 86.
71. Faniel R, Schoenfeld P. Efficacy of IV amiodarone in convert-
93. Cooper RAS, Plumb VJ, Epstein AE, et al. Marked reduction
ing rapid atrial fibrillation and flutter to sinus rhythm in
in internal atrial defibrillation thresholds with dual-current
intensive care patients. Eur Heart J 1983;4:180–5.
pathways and sequential shocks in humans. Circulation
72. Mason JW. Amiodarone. N Engl J Med 1987;316:455– 66.
1998;97:2527–35.
73. Dimopoulou I, Marathias K, Daganou M, et al. Low-dose
94. Liebold A, Wahba A, Birnbaum DE. Low-energy cardioversion
amiodarone-related complications after cardiac operations.
with epicardial wire electrodes: new treatment of atrial fibril-
J Thorac Cardiovasc Surg 1997;114:31–7.
lation after open heart surgery. Circulation 1998;98:883–6.

Source: http://thedesignspace.net/unit01a/media/Atrial%20Fibrillation%20after%20cardiac%20operation.pdf