PATHOPHYSIOLOGY AND NATURAL HISTORY

ATRIAL FIBRILLATION

Early atrial fibrillation during evolving myocardial

infarction: a consequence of impaired left atrial
perfusion
HANOCH HOD, M.D., ALLAN S. LEW, M.D., F.R.A.C.P., MATYAS KELTAI, M.D., C.SC.,
BOJAN CERCEK, M.D., IVOR L. GEFT, M.D., M.R.C.P., PREDIMAN K. SHAH, M.D.,
AND WILLIAM GANZ, M.D.

ABSTRACT Seven of 214 patients (3%) with acute myocardial infarction (120 inferior and 94
anterior) developed atrial fibrillation within 3 hr of the onset of chest pain. All seven patients had an
inferior infarction and in all seven the left circumflex artery was occluded proximal to the origin of its
left atrial circumflex branch. In five patients this occlusion was acute and was the cause of inferior
infarction and in the remaining two patients the occlusion was old and the inferior infarction was due to
an acute occlusion of the right coronary artery that also supplied extensive collaterals to the previously
occluded left circumflex artery. All seven patients also had impaired perfusion to the atrioventricular
nodal artery, as evidenced by total occlusion proximal to its origin or by stenosis proximal to its origin
associated with second- or third-degree atrioventricular block. In contrast, early atrial fibrillation did
not occur in any of the 18 patients with inferior myocardial infarction due to acute occlusion of the distal
left circumflex artery or in any of the five patients with inferior infarction due to acute occlusion of the
proximal left circumflex artery if perfusion to the atrioventricular nodal artery was not impaired. Early
atrial fibrillation did not occur in any of the 90 patients with inferior infarction due to acute occlusion of
the right coronary artery, including 12 patients with occlusion proximal to the sinus nodal artery, but
without coexistent occlusion of the left circumflex artery. Our data indicate that early atrial fibrillation
in acute myocardial infarction occurs when there is coexistent occlusion of the left circumflex artery
proximal to the origin of its left atrial circumflex branch and impaired perfusion of the atrioventricular
nodal artery. Since these two arteries both contribute to left atrial perfusion, our data suggest that acute
left atrial ischemia is the pathophysiologic mechanism of early atrial fibrillation.
Circulation 75, No. 1, 146-150, 1987.

ATRIAL FIBRILLATION complicates acute myocardial infarction in about 20% of patients. It usually
occurs later than 24 hr after the onset of infarction as a
consequence of either pericarditis or heart failure.1 In
contrast, atrial fibrillation during the early hours of
myocardial infarction is rare` and its pathogenesis is
poorly understood. The purpose of this study was to
investigate the pathogenesis of atrial fibrillation in the
early hours of an acute myocardial infarction in a population of patients who were admitted within 3 hr of the
From the Division of Cardiology, Department of Medicine, CedarsSinai Medical Center and Department of Medicine, UCLA School of
Medicine, Los Angeles.
Supported in part by NIH SCOR grant No. 17651.
Address for correspondence: William Ganz, M.D., Room No. 5314,
Division of Cardiology, Cedars-Sinai Medical Center, 8700 Beverly
Blvd., Los Angeles, CA 90048.
Received March 26, 1986; revision accepted Aug. 14, 1986.
Dr. Hod is the recipient of a H. J. Mirish Fellowship.
Dr. Keltai was an NIH-sponsored exchange scientist.

onset of acute myocardial infarction and in whom there

was subsequent angiographic documentation of coronary anatomy.

Methods
Patient population. The patient population was drawn from
244 consecutive patients with acute myocardial infarction who
were enrolled in our ongoing studies of intravenous or intracoronary streptokinase in acute myocardial infarction. Patients
were included if they satisfied the following criteria: (1) chest
pain of less than 3 hr duration at the time of evaluation by our
team, (2) ST segment elevation in two or more leads, (3) persistence of both the chest pain and the electrocardiographic
changes after sublingual nitroglycerin, (4) absence of contraindications to thrombolytic and anticoagulant therapy, and (5)
informed consent from the patient after approval by the patient's

physician.

Twenty-two patients who received intravenous streptokinase

but did not undergo coronary angiography were excluded from
this study. None of these 22 patients developed atrial fibrillation. An additional eight patients who developed atrial fibrillation were also excluded for the following reasons: four had
documented chronic atrial fibrillation, two developed atrial fi-

146

CIRCULATION
Downloaded from http://circ.ahajournals.org/ by guest on December 28, 2015

PATHOPHYSIOLOGY AND NATURAL HISTORY-ATRIAL FIBRILLATION

brillation immediately after direct current countershock for ventricular fibrillation,9 10 and two developed transient atrial fibrillation during right atrial pacing."' 12
The study population consisted of the remaining 214 patients,
167 men and 47 women, who were 59 + 11 years old (range 31
to 80). Seventy-five patients received intracoronary and 139
received intravenous streptokinase.
Definition of early atrial fibrillation. Atrial fibrillation was
defined as early if it commenced within 3 hr of the onset of chest
pain. Of note, no patient in this study developed atrial fibrillation between 3 and 24 hr after the onset of chest pain.
Coronary anatomy. Coronary angiography was performed
within 2 hr of admission in the 75 patients who received intracoronary streptokinase and 2 to 7 days after admission in the 139
patients who received intravenous streptokinase.
Artery of infarction. The artery of infarction and the site of
occlusion were obvious in 170 patients: the 75 patients who
received intracoronary streptokinase and the 95 patients who
received intravenous streptokinase and either had an anterior
infarction with severe stenosis or occlusion of the left anterior
descending artery (64 patients) or an inferior infarction with
stenosis or occlusion of the right or the left circumflex coronary
artery, but not both (31 patients). In the remaining 44 patients
with inferior infarction who received intravenous streptokinase
and had stenoses in both the right coronary and the left circumflex artery, the artery of infarction and the site of occlusion were
identified by (1) angiographic criteria of an ulcerated atheromatous plaque, i.e., indistinct luminal margins and subintimal
ulceration of the coronary artery after thrombolysis'3' 14 and (2)
the regional pattern of left ventricular dysfunction on contrast
and radionuclide ventriculograms and by the regional distribution of decreased perfusion assessed by prestreptokinase 20`T1
scintigraphy.
Assessment of atrial blood supply. Coronary angiograms
were examined to identify the following potential sources of
atrial blood supply:
(1) The sinus nodal artery that arises from the proximal right
coronary artery in about 60% of patients and from the proximal
left circumflex artery in 40% of patients and supplies most of the
right atrium, including the sinoatrial node, and also contributes
to the blood supply of the left atrium. 15, 16
(2) The left atrial circumflex artery that arises from the proximal left circumflex artery and supplies most of the left atrium. 15, 17
(3) The atrioventricular nodal artery that originates from the
right coronary artery in 90% of patients and from the left circumflex artery in 10% of patients and supplies the atrioventricu-

lar node and also the left atrium directly and/or by anastomoses
with other left atrial branches.'5-18

Results
Early atrial fibrillation. Only seven of the 214 patients
(3%) developed atrial fibrillation within 3 hr of the
onset of chest pain. Two patients were already in atrial
fibrillation when first seen by the paramedical officers
and five patients were in sinus rhythm with second- or
third-degree atrioventricular block on admission and
subsequently developed early atrial fibrillation.
Early atrial fibrillation and the location of infarction. All

seven patients who developed early atrial fibrillation

had acute inferior myocardial infarction. None of the
94 patients with acute anterior infarction developed
early atrial fibrillation. Therefore, this analysis is
limited to the 120 patients with acute inferior myocardial infarction. The clinical data on the seven patients
who developed early atrial fibrillation are summarized
in table 1.
Early atrial fibrillation and the coronary anatomy
Early atrialfibrillation and acute occlusion ofthe left circumflex coronary artery (table 2). In 28 of the 120 patients the

inferior infarction was due to acute occlusion of the left

circumflex artery. In 10 of these 28 patients the occlusion was proximal to the origin of the left atrial circumflex branch and five of these 10 patients developed
early atrial fibrillation. All five of these patients also
had impaired perfusion of the atrioventricular nodal
artery, as evidenced by either total occlusion proximal
to its origin or severe stenosis associated with secondor third-degree atrioventricular block. None of the five
patients with acute occlusion of the left circumflex
artery proximal to the left atrial circumflex branch but
without impaired perfusion of the atrioventricular nodal artery and none of the 18 patients with acute occlusion distal to the origin of the left atrial circumflex

TABLE 1
Clinical and angiographic data on patients with early atrial fibrillation

Compromised perfusion to

Patient
No.
1
2
3
4
5
6
7

Age (yr)/sex Infarct artery

51lM

LCX
LCX
LCX
LCX
LCX

70/M
71/M

RCAA
RCAA

42/M

66/M
62/M
65/M

LACx

AVN artery

SNA

Yes
Yes
Yes
Yes
Yes
Yes
Yes

Yes
Yes
Yes
Yes
Yes
Yes

No
Yes
No
Yes
No
No

Yes

No

Initial ECG rhythm Precipitating factors

AF
AF
SR, AVB
SR, AVB
SR, AVB
SR, AVB
SR, AVB

None
None
NTG and BP fall
None
NE, dopamine

Dopamine
NE

AF = atrial fibrillation; AVB = atrioventricular block; AVN = atrioventricular nodal; BP =blood pressure; LACx = left
atrial circumflex branch; NE = norepinephrine; NTG = nitroglycerin; RCA = right coronary artery; SNA = sinus nodal artery;
SR = sinus rhythm.
AOld complete occlusion of the left circumflex artery proximal to the origin of the left atrial circumflex branch.

Vol. 75, No. 1, January 1987

Downloaded from http://circ.ahajournals.org/ by guest on December 28, 2015

147

HOD et al.
TABLE 2
Relationship of early atrial fibrillation to compromised perfusion of
the left atrial circumflex (LACx) and atrioventricular nodal (AVN)
arteries in patients with acute occlusion of the left circumflex coronary artery

No.

Compromised perfusion to

of patients who

opeder wra
developed early atrial

LACx

AVN artery

fibrillation

Yes
Yes
No
No

Yes
No
Yes
No

515
0/5
0/5
0/13

branch developed early atrial fibrillation, even though

five of these 18 patients had impaired perfusion of the
atrioventricular nodal artery.
Early atrialfibrillation and acute occlusion of the right coro(table 3). Only two of the 92 patients in whom
the inferior infarction was due to acute occlusion of the
right coronary artery developed early atrial fibrillation
and these were the only two patients with acute right
coronary artery occlusion in whom there was a coexistent old occlusion of the left circumflex artery proximal to the origin of its left atrial circumflex branch. Of
interest, postreperfusion angiography in these two panary artery

tients demonstrated filling of the occluded left circumflex artery from the right coronary artery. In both these
patients, the atrioventricular nodal artery arose from
the right coronary artery such that there was impaired
perfusion to both the left atrial circumflex branch and
the atrioventricular nodal artery.
None of the 90 patients with inferior infarction due
to acute occlusion of the right coronary artery but without compromised perfusion to the left atrial circumflex
branch developed early atrial fibrillation, even though
84 of them had impaired perfusion to the atrioventricular nodal artery (three with second-degree and 11 with
third-degree atrioventricular block). Of note, 12 of
these 84 patients had acute occlusion proximal to the
origins of both the sinus nodal artery and the atrioventricular node artery. In the remaining six patients neither the left atrial circumflex branch nor the atrioventricular nodal artery were compromised.
Potential precipitating factors of early atrial fibrillation.

Three patients who were initially in sinus rhythm with

atrioventricular block developed atrial fibrillation during the course of intravenous administration of dopamine and/or norepinephrine given for hypotension (table 1). Early atrial fibrillation did not occur during
intravenous administration of catecholamines in any of
the 13 patients with anterior infarction or in any of the
21 patients with inferior infarction but without coexis-

tent impairment of left atrial circumflex and the atrio-

ventricular nodal artery perfusion. Of note, 14 of these

latter 21 patients had acute right coronary occlusion
proximal to the origin of the atrioventricular nodal
artery and six had occlusion proximal to the origins of
both the sinus nodal artery and the atrioventricular
nodal artery. In one patient atrial fibrillation commenced during a nitroglycerin-induced transient hypotensive episode. In the remaining three patients (two of
whom were already in atrial fibrillation when seen by
the paramedical officers) a precipitating factor could
not be identified.
Six of the seven patients who developed early atrial
fibrillation had no evidence of valvular heart disease
on clinical, echocardiographic, or contrast ventriculographic assessment. One patient with a left atrial diameter of 4.5 cm as determined by echocardiography at
the time of atrial fibrillation had mild mitral insufficiency at delayed contrast ventriculography. Of note,
the severity of mitral insufficiency and the degree of
left atrial enlargement in this patient was not greater
than in those patients with mitral insufficiency complicating anterior infarction, none of whom developed
early atrial fibrillation. None of the seven patients had
a history or signs of hyperthyroidism and all seven
patients had serum potassium on admission to the hospital of greater than 3.8 meq/liter.
Reperfusion and atrial fibrillation. In all seven patients
the early atrial fibrillation was transient. Normal sinus
rhythm returned within 60 min of the onset of resolution of chest pain and ST segment elevations in five
patients and after 150 min in one patient. Since one of
these patients received 0.25 mg digoxin, one received
5 mg verapamil, and one received both 0.25 mg digoxin and 5 mg verapamil before reversion to sinus
rhythm, the relative contributions of reperfusion and
antiarrhythmic therapy to cessation of atrial fibrillation
in these three patients cannot be determined. In the one
patient in whom reperfusion was not achieved, atrial
TABLE 3
Relationship of early atrial fibrillation to compromised perfusion of
the left atrial circumflex (LACx) and the atrioventricular nodal
(AVN) arteries in patients with acute occlusion of the right coronary
artery

Compromised perfusion to
LACx

AVN artery

Yes
Yes
No
No

Yes
No
Yes
No

148

No. of patients who

developed early atrial
fibrillation

2/2
0/84
0/6

CIRCULATION
Downloaded from http://circ.ahajournals.org/ by guest on December 28, 2015

PATHOPHYSIOLOGY AND NATURAL HISTORY-ATRIAL FIBRILLATION

fibrillation persisted for 16 hr after the onset of chest

pain.
Discussion
Our study confirms previous reports`8 that atrial
fibrillation is a rare event during the early hours of an
evolving acute myocardial infarction. The rarity of
early atrial fibrillation is consistent with our finding
that its development requires both occlusion of the
proximal left circumflex artery and total or near-total
occlusion proximal to the origin of the atrioventricular
nodal artery. Acute occlusion of the left circumflex
artery occurs in only about 15% of patients with myocardial infarction and occlusion proximal to the left
atrial circumflex branch obviously occurs even less
frequently. Since the atrioventricular nodal artery
originates from the right coronary artery in 90% of
patients and from the circumflex coronary artery in
only 10% of patients, early onset atrial fibrillation will
most often require coexistent total or near-total occlusion of both the right and the proximal circumflex
coronary arteries. This was the case in six of the seven
patients in our study; the seventh patient had proximal
occlusion of a dominant left circumflex artery that
gave origin to the atrioventricular nodal artery.
Our conclusions appear to contradict those of
James,19 who reported that atrial fibrillation in acute
myocardial infarction was caused by coexistent occlusion of the sinus nodal artery and the atrioventricular
nodal artery. However, review of the anatomic findings in the nine patients who developed atrial fibrillation in James' study reveals that eight of them had
occlusion of the left circumflex artery either directly or
as a consequence of left main coronary artery occlusion. The ninth patient had hemochromatosis, which
may be an alternative explanation for the development
of atrial fibrillation.20
Our observation that early atrial fibrillation was accompanied by impaired perfusion to the atrioventricular nodal artery is consistent with James' finding19 21 of
an association between atrioventricular conduction delay and occurrence of atrial arrhythmias in acute myocardial infarction. Whether this reflects an electrophysiologic role for atrioventricular conduction delay
in the genesis of early atrial fibrillation or whether
atrioventricular block is simply a marker of compromised left atrial perfusion caused by obstruction of the
atrioventricular nodal artery is not known. However,
the finding that atrioventricular block alone or in combination with compromised perfusion to the sinus nodal artery did not result in early atrial fibrillation, even
during infusion of catecholamines, unless there was

also compromised perfusion of the left atrial circumflex branch suggests that atrioventricular block per se
does not have a primary role in the development of
early atrial fibrillation.
Of note, since the obstruction that impaired perfusion to the atrioventricular nodal artery was also always proximal to the origin of the posterior descending
artery, and since according to Kugel'8 this artery may
contribute to the left atrial blood supply, it is not possible to rule out that compromised perfusion of the posterior descending artery also contributed to the development of early atrial fibrillation.
Our anatomic findings suggest that acute left atrial
ischemia is the primary cause of early atrial fibrillation. This conclusion is supported by our observations
that (1) early atrial fibrillation reverted to sinus rhythm
in close temporal relation to the resolution of myocardial ischemia, and (2) administration of catecholamines, known to accentuate myocardial ischemia,22

preceded early atrial fibrillation in three patients with

compromised left atrial perfusion. However, a direct
effect of catecholamine administration on atrial conduction2' in the genesis of early atrial fibrillation cannot be excluded.
Potential limitations of study. The lack of atrial perfusion studies precludes a precise definition of the extent
and distribution of impaired atrial perfusion and the
lack of electrophysiologic studies precludes the determination of the electrophysiologic mechanism and the
site of origin of atrial fibrillation. Therefore, we cannot
rule out the possibility that some patients who developed early atrial fibrillation due to impaired left atrial
perfusion may also have had some impairment of right
atrial perfusion, but our data suggest that coexistent
right atrial involvement would not have been a prerequisite for the development of early atrial fibrillation.
Clinical implication. The spontaneous onset of early
atrial fibrillation in patients with acute myocardial infarction is rare and usually short-lived in patients in
whom reperfusion of the artery of infarction is successful. It is probably due to left atrial ischemia secondary
to occlusion of the circumflex artery proximal to the
origin of its left atrial circumflex branch and coexistent
impairment of atrioventricular nodal artery perfusion.
We are grateful to Gloria Solomon and Ahuva Hod for research assistance and to Dwana Williams for assistance with the
preparation of the manuscript.

References
1. Liberthson RR, Salisbury KW, Hutter AM, DeSanctis RW: Atrial
tachyarrhythmias in acute myocardial infarction. Am J Med 60:
956, 1976

Vol. 75, No. 1, January 1987

Downloaded from http://circ.ahajournals.org/ by guest on December 28, 2015

149

HOD et al.
2. Cristal N, Peterburg I, Szwarcberg (Shaher) J: Atrial fibrillation
developing in acute phase of myocardial infarction. Prognostic
implications. Chest 70: 8, 1976
3. Hunt D, Sloman G, Pennington C: Efects of atrial fibrillation on
prognosis of acute myocardial infarction. Br Heart J 40: 303, 1978
4. Lofmark R, Orinius E: Supraventricular tachyarrhythmias in acute
myocardial infarction. Acta Med Scand 203: 517, 1978
5. Sugiura T, Iwasaka T, Ogawa A, Shiroyama Y, Tsuji H, Onoyama
H, Inada M: Atrial fibrillation in acute myocardial infarction. Am J
Cardiol 56: 27, 1985
6. Klass M, Haywood LJ: Atrial fibrillation associated with acute
myocardial infarction: a study of 34 cases. Am Heart J 79: 752,
1970
7. Adgey AAJ, Geddes JS, Webb SW, Allen JD, James RGG, Zaidi
SA, Pantridge JF: Acute phase of myocardial infarction. Lancet 2:
7723, 1971
8. Simoons ML, van der Brand M, de Zwaan C, Verheugt FWA,
Remme WJ, Serruys PW, Bar F, Res J, Krauss XH, Vermeer F,
Lubsen I: Improved survival after early thrombolysis in acute myocardial infarction. A randomized trial by the Interuniversity Cardiology Institute in The Netherlands. Lancet 2: 578, 1985
9. Liberthson RR, Nagel EL, Hirschman JC, Nussenfeld SR: Prehospital ventricular fibrillation. N Engl J Med 291: 317, 1974
10. Sano T, Scher AM: Multiple recording during electrically induced
atrial fibrillation. Circ Res 14: 117, 1964
11. Haft JI, Lau SH, Stein E, Kosowsky BD, Damato AN: Atrial
fibrillation produced by atrial stimulation. Circulation 37: 70, 1968

12. Furman S, Cooper JA: Atrial fibrillation during A-V sequential

pacing. Pace 5: 133, 1982
13. Ambrose JA, Winters SL, Stern A, Eng A, Teichholz LE, Gorlin
R, Fuster V: Angiographic morphology and the pathogenesis of
unstable angina pectoris. J Am Coll Cardiol 5: 609, 1985
14. Ambrose JA, Monsen C, Winters S, Arora R, Rentrop P, Rudin D,
Maratea S, Haft J, Gorlin R, Fuster V: Coronary morphology
following streptokinase in myocardial infarction and unstable angina. J Am Coll Cardiol 7: 107A, 1986 (abst)
15. James TN, Burch GE: The atrial coronary arteries in man. Circulation 17: 90, 1958
16. James TN: Anatomy of the coronary arteries in health and disease.
Circulation 32: 1020, 1965
17. Homby RI: Clinical-anatomical correlates in coronary artery disease. Mount Kisco, 1979, Futura Publishing Company, Inc., p 11
18. Kugel MA: Anatomical studies on coronary arteries and their
branches. Am Heart J 3: 260, 1927
19. James TN: Myocardial infarction and atrial arrhythmias. Circulation 14: 761, 1961
20. Buja LM, Roberts WC: Iron in the heart. Etiology and clinical
significance. Am J Med 51: 209, 1971
21. James TN, Hershey EA: Experimental studies on the pathogenesis
of atrial arrhythmias in myocardial infarction. Am Heart J 63: 196,
1962
22. Weiner N: Norepinephrine, epinephrine, and sympathomimetic
amines. In Gilman AG, Goodman LS, Rall TW, Murad F, editors:
The pharmacological basis of therapeutics. New York, 1985, Macmillan Publishing Company, p 145

150

CIRCULATION
Downloaded from http://circ.ahajournals.org/ by guest on December 28, 2015

Early atrial fibrillation during evolving myocardial infarction: a consequence of impaired

left atrial perfusion.
H Hod, A S Lew, M Keltai, B Cercek, I L Geft, P K Shah and W Ganz
Circulation. 1987;75:146-150
doi: 10.1161/01.CIR.75.1.146
Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright 1987 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7322. Online ISSN: 1524-4539

The online version of this article, along with updated information and services, is located on
the World Wide Web at:
http://circ.ahajournals.org/content/75/1/146

Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally
published in Circulation can be obtained via RightsLink, a service of the Copyright Clearance Center, not the
Editorial Office. Once the online version of the published article for which permission is being requested is
located, click Request Permissions in the middle column of the Web page under Services. Further
information about this process is available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Circulation is online at:
http://circ.ahajournals.org//subscriptions/

Downloaded from http://circ.ahajournals.org/ by guest on December 28, 2015