Risk of thromboembolism in new onset or transient atrial fibrillation

R i s k of T h r o m b o e m b o l i s m in N e w O n s e t or T r a n s i e n t Atrial Fibrillation Marcus F, Stoddard Atrial fibrillation (AF), potentially serious cardiac arrhythmia, occurs in 2% to 4% of persons greater than 60 years of age. The risk of systemic thromboembolism from chronic AF has long been recognized. Little is known about the thromboembolic risk of new onset AF. However, the results of prior studies support a significant risk of thromboembolism because of recent onset or paroxysmal AF. The mechanism of thrombus formation, embolization, and resolution in AF is illdefined, particularly that of new onset. The traditional concept holds that atrial thrombus forms only after greater than 2 days of AF and embolizes by dislodgement from increases in shear forces. This prevailing concept further proposes that newly formed atrial thrombus, in the setting of AF, organizes over a span of 14 days. The results of recent transesophageal echocar-

diographic studies have given insight into the temporal sequence of atrial thrombus formation, embolization, and resolution in AF and have expanded the traditional concept of thromboembolism in AF. Namely, left atrial thrombus may form before the onset of AF in the face of sinus rhythm, Conversion to sinus rhythm may increase the thrombogenic millieu of the left atrium. Importantly, atrial thrombus may form in the acute phase of AF. Lastly, thrombus may require more than 14 days to become immobile or to resolve. On the basis of these emerging insights by transesophageal echocardiography, it appears appropriate to consider anticoagulation in patients presenting with new onset or acute AF. Copyright © 1996 by W.B. Saunders Company

TRIAL FIBRILLATION (AF), a potentially serious cardiac arrhythmia, occurs in only 0.4% of the adult population but significantly increases in prevalence to 2% to 4% in persons greater than 60 years of age. 1,2 AF is associated with a significant risk of systemic embolism, heart failure, and death. 3-5 Importantly, thrombi that embolize from a cardiac origin, particularly when caused by AF, are frequently of a relatively large size and may lead to a major life-threatening cerebrovascular accident. 6 Although the incidence of thromboembolism is higher in patients with rheumatic AF, nonrheumatic AF is associated with a significant thromboembolic risk, up to 5% per year. 3,7-9The appropriate treatment of patients with AF to reduce thromboembolic risk is a clinically challenging issue.l° Recent clinical trials support the use of anticoagulant prophylaxis in patients with chronic nonrheumatic AF to lower the risk of stroke. 11-~5 However, long-term anticoagulation therapy is not without risk, particularly in elderly patients in whom this atrial arrhythmia predominates. Although unproven, the postulate that converting AF to sinus rhythm will significantly reduce associated complications, particularly thromboembolic risk, has greatly influenced the practic-

ing cardiologist in the management of patients with AF. Unfortunately, cardioversion of AF is associated with a significant incidence of systemic embolism, ranging from 0.6% to 5.6% of patients undergoing pharmacological or electrical cardioversion. 16-a9 The presumptive mechanism of thromboembolism in patients with AF, occurring spontaneously or around the time of cardioversion, is dislodgement of pre-existing left atrial (LA) thrombus into the systemic circulation. Recent studies using transesophageal echocardiography (TEE) have shown LA thrombus in 12% to 26% of patients with AF 2°-24 and gives support to this proposed mechanism. However, recent studies have shown that cardioversion is associated with worsening LA appendage (LAA) function and development of spontaneous echocardiographic c o n t r a s t (SEC), 23,25,26an independent predictor of thromboembolism.2729In addition, thromboembolism despite exclusion of LA

From the Department of Medicine, Division of Cardiology, University of Louisville, Louisville, KY Address reprint requests to Marcus F. Stoddard, MD, Ambulatory Care Bldg, Third Floor, University of Louisville, 550 S Jackson St, Louisville, KY40203. Copyright © 1996 by W..B. Saunders Company 0033-0620/96/3901-000655. 00/0

Progress in Cardiovascular Diseases. Vol XXXIX, No 1 (July/August), 1996: pp 69-80

69

MARCUS F. STODDARD

70

thrombus by TEE may occur after cardioversion of AF. 3° Lastly, the formation of LAA thrombus immediately after electrical cardioversion has been reported. ~3 These studies support the notion of de novo formation of LA thrombus after cardioversion and subsequent embolization as an alternative or, at least, an additional mode of thromboembolism caused by cardioversion of AF. These data are intriguing, particularly if a similar phenomenon applies to the spontaneous development of AF. The traditional concept is that LA thrombus forms in patients with AF several days after the onset of the arrhythmia.3I Accordingly, patients with new onset AF of less than 3 days in duration are commonly cardioverted without anticoagulation prophylaxis, as recommended by the American College of Chest Physicians.32 The exact incidence of the presence of intracardiac thrombus in and the risk of thromboembolism in new onset or acute AF has not been well defined. This review will discuss the risk of thromboembolism in patients with AF particularly AF of new onset or acute AF, and recent insights from TEE studies. THROMBOEMBOLIC RISK IN AF

Magnitude of Thromboembolic Risk of New Onset AF The risk of stroke and systemic thromboembolism from AF has long been recognized and well documented in patients with chronic and persistent AF. 7 The sequelae of thromboembolic stroke are often devastating. The results of retrospective studies have suggested that greater than half of the victims of a cerebroembolism caused by AF experience severe neurologic deficits or death. 33,34 The results of these prior retrospective studies may be influenced by selection bias. However, recent prospective studies have shown that, in the placebo group of patients with nonrheumatic AF experiencing a stroke, the severity of the stroke is disabling or fatal in 44% to 63% of victims. 11"14Although not all strokes in patients with AF are cardioembolic, the Framingham study 7 supports the notion that the predominant mechanism of stroke in these patients is embolism.

Little is known about the thromboembolic risk of new onset AF. Wolf et aP5 reported on a cohort of patients with recent onset of nonrheumatic AF without other discernible risk factors for emboli from the Framingham study. AF was first diagnosed in 14 of 59 patients of this cohort at initial hospitalization for stroke. It was suggested that a high incidence of stroke occurred during the first 2 years after the onset of nonrheumatic AF among these patients. 35 In a retrospective study of 426 patients with persistent paroxysmal AF, Petersen et aP 6 noted a clustering or distinctly higher incidence of thromboembolic episodes at the onset of paroxysmal AF and during the first year after paroxysmal AF became continuous or chronic. In this study,36 the incidence of systemic embolism was 6.8% the first month after the onset of paroxysmal AF and decreased to 2% per year afterwards. In patients who progressed from paroxysmal to chronic AF, the embolism rate increased to an alarming 13.3% in the first year but decreased to 4% per year thereafter. These data suggest that AF of recent onset has a higher risk of embolism as compared with that for paroxysmal or chronic AF that has been present for more than 1 year. Also, risk increases transiently during the transition from paroxysmal to chronic AF. Furthermore, these data support the notion of a transient increase in the thrombogenic millieu of the left atrium, the presumed intracardiac site of thrombus formation and emboli, during transition from sinus rhythm to AF and from paroxysmal to chronic AF. It is unknown if each episode of AF in patients with AF concurs a transient higher risk of embolism. The cause of paroxysmal AF, as is the case in chronic AF, has a significant impact on thromboembolic risk. The predominant causes of AF in the study by Petersen et aP 6 were heterogeneous, including ischemic heart disease (52%), rheumatic heart disease (13%), thyroid disease (7%), and nonvalvular heart disease (28%). Patients with rheumatic heart disease had a twofold greater risk of thromboembolism as compared with that of those with nonrheumatic AF (ie, 35.7% v 15.9%). 36

THROMBOEMBOLISM

Paroxysmal Versus Chronic A F Thromboembolic Risk Most studies are consistent with the notion that the long-term risk of thromboembolism from paroxysmal AF as compared with that for chronic AF is lower. Takahashi et a137 reported a lower stroke rate in 94 patients with paroxysmal AF as compared with that in 140 patients with chronic AF. Similarly, Petersen et a136 noted a lesser mean incidence of systemic emboli caused by paroxysmal AF as compared with that from chronic AF (2.0% v 5.6% per year, respectively). Shimomura et aP 8 reported a systemic embolism rate per 1,000 patient years of 15 in patients with paroxysmal AF, which is considerably lower than the rate of 28 in those with chronic AF. 36 Although the mean risk of cardioembolism is lower for paroxysmal AF as compared with that for chronic AF, it is still plausible that, during the initial episode and possibly during subsequent recurrent episodes of AF, the embolic risk transiently, but significantly, increases. 36

Recent Embolism and Risk of Recurrence Several studies confirm an increased risk of recurrent thromboembolism within weeks of an initial embolic event. Kelly et a139 noted a recurrence of stroke in 33% of 36 patients with AF over a mean interval of 2.5 months. In the 19 nonanticoagulated patients, the recurrence frequency was 53%. Hart et al4° reported a recurrence of embolism in 20% of 23 nonanticoagulated patients with AF over an 11-day duration after a stroke. A recurrent embolism occurred in 20% of 148 patients with AF within 2 weeks of the initial episode in a study reported by Darling et al. 41 The results of additional studies 42,43 are in agreement with a temporary increased risk of recurrent embolism after an embolic event. However, the rate of the recurrence in these later studies is less impressive as compared with that in the earlier reported studies, ie, a rate of recurrence of 1% to 2% over the subsequent month after an initial embolic event in patients with A F . 42'43

71

MECHANISM OF THROMBOEMBOLISM IN AF: TRADITIONAL CONCEPT

The mechanism of thrombus formation and embolization in AF has not been well defned, particularly for AF of recent or new onset. Presumably, thrombus formation caused by AF in cardiac chambers adheres to Virchow's triad, ie, the presence of endocardial injury, blood flow stasis, and hypercoagulable state. Stagnation of blood flow is an obvious occurrence of impaired atrial mechanical function in AF, partially fulfilling Virchow's triad. The cause of the remaining two components of Virchow's triad in AF is less apparent. Limited data support the notion of a hypercoagulable state in AF. The results of recent studies in patients with AF have shown increased levels of coagulation factors, such as fibrinogen, D-dimer, and fibrinopeptide A, and decreased levels of antithrombin I I I . 4446 Although intriguing, the importance of these coagulation abnormalities in the formation of thrombus in patients with AF remains unclear. It is plausible that activation of coagulation factors and fibrin formation occurs from erythrocyte aggregation and increased blood viscosity because of low shear rates in a mechanically impaired, fibrillating atrium. 47 A paucity of data exists on the postulate of atrial endocardial injury in AF, the last component of Virchow's triad. In theory, stretching of the atria from increased intracavitary pressure might interrupt the endocardial layer, exposing subendocardial tissue to blood and subsequent thrombus formation. 4s Intuitively, disease processes with a greater propensity to raise atrial pressure and stretch the endocardial surface, such as mitral stenosis, would be at highest risk for thrombus formation. However, no direct evidence exists to support this mechanism of atrial endocardial injury. Atrial ischemia could injure the endocardial surface. However, this phenomenon is rare and would be unlikely to explain the frequency of atrial thrombus in AF. The traditional concept further proposes that AF must be persistent in excess of 2 days for atrial thrombus to form solely on the basis of the arrhythmia. Once thrombus forms in the atrium of patients with AF, factors that lead to its

72

MARCUS F. STODDARD

dislodgement and embolization are unknown. The traditional concept is that dynamic shear forces dislodge intracardiac thrombus and is supported by the results of studies investigating morphological determinants of embolization of left ventricular thrombi. 49 The resumption of sinus rhythm and return in atrial mechanical function increases shear forces and predisposes to dislodgement of thrombus previously formed in the body of the atrial cavity. The return in LA mechanical function after conversion to sinus rhythm may be rapid but often times is gradual and may require 3 weeks: ° A longer duration of AF preceding conversion to sinus rhythm is associated with a more protracted recovery time in LA function: ° The traditional concept of the time course of atrial thrombus organization and resolution in AF is based on the hypothesis proposed by Goldman, 51 which states that atrial thrombus organizes by fibroblastic infiltration over a span of 14 days. The traditional concept of atrial thrombus formation, embolization, and organization, as summarized in Fig 1, coupled with observations on the time course of recovery in LA mechanical function after resumption of sinus rhythm help to explain the clustering of embolic events

TRADITIONAL CONCEPT I

AF LA thrombusforn~non and embolism AtrialNb

Onset

LAILAA stas~

Immediate

Thrombus $ Organiz~ thrombus

Aftez> 3 days F_~a,oU After 14 days

Fig 1. Schematic diagram of the traditional concept of LA cavity (LA) and LAA (LAA) thrombus formation, organization, and embolization in AF (fib). immediately after the onset of AF, blood flow stasis ensues and increases the thrombogenic millieu of the left atrium. Thrombus may form after 2 to 3 days if Virchow's triad is met. LA thrombus organizes over 14 days becoming adherent to the endocardial wall and less likely to subsequently embolize. However, before organizing, increases in atrial shear forces from dynamic changes, spontaneous conversion, or electrical cardioversion may dislodge thrombi and lead to embolus.

around the onset of paroxysmal A F . 36 It also partly explains the risk of thrombus formation and embolism in patients with chronic AF, particularly in patients with cardiac disease that increases LA pressure. However, the marked and transient increased risk of embolism in the first year after a transition from paroxysmal to chronic AF does not appear to be easily explained by the traditional concept of thromboembolism in AF. Lastly, the traditional concept implies that LAA function is impaired while in AF relative to the performance when in sinus rhythm. MECHANISM OF THROMBOEMBOLISM IN AF: EXPANDED CONCEPT

Recent TEE studies investigating LA thrombogenesis in AF at least expand, if not completely challenge, the traditional concept of LA thrombus formation, embolization, and organization. Although the mechanism of LA thrombus formation and subsequent embolization is undoubtedly complex and ill-defined, the traditional concept that AF must persist in excess of 2 to 3 days before atrial thrombus forms has become widely accepted) 2In addition, the traditional concept assumes that atrial thrombus in AF forms after the onset of the arrhythmia. Based on these concepts, for example, cardioversion of AF of less than 3 days in duration without anticoagulation prophylaxis is believed to be safe and is commonly performed. 32 However, it is unclear how long it takes for thrombi to form in fibrillating atria, and formation probably depends on multiple factors, such as the type of underlying heart disease, ventricular function, atrial size and function, and intraatrial coagulability. Recent TEE studies give insight into the issue of atrial thrombus formation in acute AF. The results of a recent TEE study by Stoddard et aP 2 challenge the widespread opinion that patients with acute AF of less than 3 days in duration are free of LA thrombi. In this study, 14% (20 of 143 patients) with new onset acute AF of less than 3 days in duration had LA thrombi as detected by TEE (Fig 2). However, the frequency of LA thrombus in chronic AF was significantly higher (Fig 2). In the study by

THROMBOEMBOLISM

73

3O

27%

P25 e 20 r

c 15

47/174 14% . . . . . .

e 10 n 5 t 0

p
Acute

Chronic

Fig 2. The frequency of LA thrombus in patients with acute versus chronic AF was lower (P < .01} but occurred in 14% of patients and would appear clinically relevant,

Stoddard et al, 52 7 of the 20 patients with acute AF and LA thrombus had the onset of A F documented electrocardiographically. Of these 7 patients, 5 had mobile LA and/or LAA thrombi (Figs 3 and 4). These data support an expanded concept that LA thrombus may form within 3 days after the onset of AF. By TEE, Black et a121 showed LAA thrombi in 2 patients with AF of 7 days in duration. It is also possible that patients presenting with new onset or acute AF form LA thrombi

Fig 3. Two-dimensional TEE in a 70-year-old man with new onset AF of less than 3 days in duration. Mobile LA (arrow} and LAA (arrowheads) thrombi were present, This patient subsequently had a peripheral embolus. AO, aorta; LA, left atrium; PA, pulmonary artery. (Reprinted with permission from the American College of Cardiology [Journal of the American

College of Cardiology, 1995, 25:452-459]. 52)

Fig 4. Two-dimensional TEE in a 78-year-old woman with acute AF shows the typical location of a mobile thrombus [arrow) confined to the LAA. AO, aorta; LA, left atrium; PA, pulmonary artery. (Reprinted with permission from the American College of Cardiology [Journal of the American College of

Cardiology, 1995, 25:452-459]. s2)

before the onset of AF. In the study by Stoddard et al,52 50% of patients with acute AF and LA thrombi had immobile thrombus. In 1 of these patients with acute AF and electrocardiographically documented onset of AF, an immobile echodense thrombus completely filled the LAA (Fig 5). Mugge et a153 showed LA thrombus in 23 patients with sinus rhythm. These data support an expanded concept that LA thrombus forms in some patients with AF before the onset of the arrhythmia. The presence of LA thrombus in patients with acute (ie, < 3 days in duration) or recent (ie, duration of few weeks) nonvalvular AF with a recent thromboembolism is more frequent than that in those without a recent embolic event. 5 Stoddard et aP 2 noted a frequency of LA thrombus of 21% (5 of 24) in patients with acute AF and recent systemic emboli, a frequency similar to that in patients with chronic AF with or without a recent embolic event (Fig 6). Using TEE, Manning et a154identified LA thrombus in patients with recent onset of nonvalvular AF in 43% (13 of 30) of patients with a recent thromboembolism as compared with 10% (9 of 87) in those without a recent embolus (Fig 7). These data are consistent with the clinical observation

74

MARCUS F. STODDARD

60 P

50

43%

r 40 c 30

13/30

e

p<0.001

e 20 n 10 t 0

10% 9/87

Emboli

No Emboli

Fig 7. Frequency of LA thrombus in patients with recent onset AF is considerably higher in subjects with a recent thromboembolism as compared with that of those without emboli.

Fig 5. Two-dimensional TEE in an 84-year-old woman with documented onset of acute AF shows an immobile echodense thrombus (arrows) completely filling the LAA. LA, left atrium; AO, aorta. (Reprinted with permission from the American College of Cardiology [Journal of the American College of Cardiology, 1995, 25:452-459[. s2)

of an increased thromboembolic risk at the onset of paroxysmal AF 26 and support the proposed mechanism of atrial thrombus dislodgement as the culprit of thromboembolism in new onset AF. Recent TEE studies support the concept that the LAA, as opposed to the body of the left atrium, is the predominant site for atrial throm-

p40 e 30 r

p<0.004 2 9 %

p=ns

35/122

23 %

c 20

13%

e

nl0 t 0

Acute Chronic Emboli

Acute Chronic No Emboli

Fig 6. Frequencies of LA thrombus in acute as compared with chronic AF w i t h and without thromboembolic event are depicted. Acute AF with a recent embolic event had a similar frequency of LA thrombus as compared with that of chronic AF with recent embolism.

bus formation and the major culprit for thromboembolic events in patients with AF. Twodimensional and Doppler TEE studies have played an integral role in the current understanding of LAA function and thrombogenesis. Pollick et a155 studied 82 patients by TEE and showed an association between LAA thrombus formation and appendage enlargement and impaired contraction. Pozzoli et aP 6 showed an association between impaired LAA function (ie, absent appendage velocities or flow) and that of thrombus formation and SEC. The study by Garcia-Fernandez et aP 7 was one of the first studies to distinguish different Doppler flow patterns in the LAA of patients with AF that was associated with low and high risk for atrial thrombus. Subsequent cross-sectional TEE studies have shown an increased thromboembolic risk in patients with AF and LAA dysfunction as evidenced by LAA decreased emptying or filling velocity,5s,59 reduced emptying fraction,59 or SEC. 6° The method of measuring LAA emptying fraction is shown in Fig 8. Fatkin et al 6° have described three patterns of LAA emptying and filling velocities in patients with AF (Fig 9). The function of the LAA in patients with AF is heterogeneous. LAA function may be preserved, moderate, or severely impaired. The finding of LAA contraction in the face of AF may relate to its smaller size as compared with that of the LA cavity. The heterogeneity of

THROMBOEMBOLISM

75

Fig 8. Example of measuring LAA area using biplane TEE from a horizontal (A) and vertical (B) plane. The appendage is digitized (dashed line) along the endocardium beginning from the limbus of the left upper pulmonic vein. Emptying fraction is calculated as the difference between maximal and minimal area divided by maximal area times 100%. LA, left atrium. (Reprinted with permission from the American College of Cardiology [Journal of the American College of Cardiology, 1994, 23:599-607]. ss}

LAA function in AF is a recently recognized concept because of the emergence of TEE and is intuitively important in its potential to explain episodic thromboembolism. Although investigated in animal models, little data exist on the

hemodynamic determinants of LAA function in humans. 61 One prospective study showed an increased thromboembolic risk in patients with nonvalvular AF with, as compared with that of those without, LA SEC as assessed by TEE. 62

a~

B

Fig 9. Doppler TEE patterns of LAA blood flow velocity obtained by placing a sample volume near the orifice of the LAA are shown. (A) and (B) occur in patients with sinus rhythm and atrial contraction. The " a " peak occurs from atrial contraction after electrocardiographic p-wave. {C), (D}, and (E) occur in patients with AF. In (C), an outflow "'e" wave is followed by high-velocity alternating filling and emptying waves. In (D), fibrillatory empWing and filling velocities are low. In (C), virtual stasis of flow is present. (Reprinted with permission from the American College of Cardiology [Journal of the American College of Cardiology, 1994, 23:961-g69]. e°)

76

However, this study did not assess LAA function as a potential risk for thromboembolism. Limited data exist that report potential differences in LAA function in patients with acute, paroxysmal, and chronic AF. Mugge et aP 9 reported high-velocity flow in the LAA of eight patients with acute (n = 4) or paroxysmal (n = 4) AF. However, in patients with chronic AF, a low-flow LAA velocity profile was noted in 10 of 21 patients with nonrheumatic AF and in 12 of 12 subjects with rheumatic AF. s9 On the basis of SEC, the thrombogenic millieu of the left atrium may differ between patients with AF of acute (ie, < 3 days) and of recent (ie, several weeks) onset. 5344Stoddard et aP 2 noted no difference in the frequency of LA SEC in patients with acute AF with (45%) and without (39%) a recent embolic event. However, Manning et a154 showed a considerably higher frequency of LA SEC in patients with recent onset AF with recent thromboembolism (87%) as compared with the frequency in those with recent onset AF without recent thromboembolism (48%). Counterintuitively, recent TEE studies suggest that the left atrium and LAA become more thrombogenic immediately after electrical cardioversion of AF to sinus rhythm.23,25,z6 The development of new or worsening LA and/or LAA SEC after electrical cardioversion of AF to sinus rhythm was noted in 35% of patients by Grimm et al, 25 in 31% by Fatkin et al, 26 and in 41% by Stoddard et al? 3These subjects also had associated reductions in LAA emptying and filling velocities.23,25 Case reports have also shown that impaired LAA function may occur after pharmacological or spontaneous conversion from AF to sinus rhythm.63 These observations are potentially important in understanding thromboembolism associated with acute, recent onset, or paroxysmal AF. The traditional concept of atrial thrombus formation must be expanded to incorporate the potential for de novo thrombus formation after conversion to sinus rhythm. The time course of LA thrombus organization and resolution in AF as proposed by Goldman51 is unproven. Recent TEE observa-

MARCUS F. STODDARD

tions on the time course of LAA thrombus to become immobile or to resolve in AF are of interest. 23 In a study reported by Stoddard et a123 of 31 patients with AF and LAA thrombus, 21 were followed up with serial TEE while receiving therapeutic anticoagulation. After 3 to 5 weeks of anticoagulation, LAA thrombus resolved in only 1 of 21 patients and remained mobile in 16 patients with initially mobile thrombi (Fig 10). After more than 5 weeks (mean, 9.3--3.6 weeks) of anticoagulation, thrombus resolved or became immobile in 71% (15 of 21 patients).

Predictors of LA Thrombus in Acute A F In the study by Stoddard et al, z3 transthoracic echocardiographic predictors of LA thrombus in patients with acute AF were mitral stenosis (P < .01) and left ventricular ejection fraction <40% (P < .02). No clinical variables were predictive. Importantly, when LA SEC was included in the multivariate regression analysis, it became the only independent predictor (P < .0002) of LA thrombus in patients with acute AF. Similarly, Manning et aP 4 noted a significantly higher frequency of LA contrast in

31 LAA Thrombu~

Io 2 |

NO REPEAT TEE

/% 16

5

Mobile

Immobile /

16 Mobile

6

/ J ,6 %

~

4

Anticoogulotion

I

4 Immobile

Resolved

3 to 5 reeks

%4

Idobile Immobile Resolved

ge=olved

)5 to 1 7 reeks

Fig 10. Schematic diagram depicting the time course for LAA thrombi in patients with AF to resolve or to become immobile as assessed by TEE. In general, more than 5 weeks of anticoagulation therapy was necessary for thrombus to resolve or to become immobile. (Reprinted with permission from the American Heart Journal [Stoddard et ah Transesophageal echocardiographic guidance of cardioversion in patients with atrial fibrillation. 129:1204-1215, 1995]. z3)

THROMBOEMBOLISM

77

patients with LA thrombus and recent onset of AF.

Predictors of Thromboembolism in Acute AF Using multivariate analysis, Stoddard et a123 showed that, in patients with acute AF, mobile LA thrombus (P < .05), congestive heart failure (P < .02), coronary artery disease (P < .05), and left ventricular ejection fraction <40% (P < .05) were independent predictors of an embolic event. SEC was not a predictor of thromboembolism in this patient group. Although these results are at odds with studies showing SEC as a risk factor for thromboembolism in chronic mE, 29'64-66 they are supported by the findings of Manning et al, 54who investigated patients with recent onset AF. In the later study,54the prevalence of LA SEC did not differ between recent onset AF groups presenting with acute thromboembolism versus the group that had LA thrombus without thromboembolism. Although caution must be exercised when extrapolating from these studies, these results suggest that, in acute or recent AF, LA SEC is a factor in thrombus formation but may not be related to subsequent dislodgement of thrombus and embolism. A F LA thrombusformationand e4nbolism S~seayalm

AerialFib f ,1, ~

Electdud

• uml, ~ c , ~ , , , ~ o u

,L

"l"..,....bB

j,

F.,,~i

Omct ham=dim= ~

<3

Fig 11. The expanded concept of LA thrombus formation and embolism in AF (Fib} is schematically shown, in this expanded concept, LAA thrombus may form before the onset of AF while in sinus rhythm if the LAA is dysfunctional. With the onset of AF, the thrombogenic millieu of the left atrium and LAA increased, as reflected by LA/LAA stasis or SEC. In addition, conversion to sinus rhythm electrically, pharmacologically, or spontaneously may increase the thrombogenic millieu of the left atrium factors that, as of yet, are unknown. Importantly, LA and LAA thrombus may form immediately after the onset of AF or conversion to sinus rhythm. Lastly, thrombus may not organize or resolve with 14 days of anticoagulation and can require more than 5 weeks.

Expanded Concept The expanded concept of LA thrombus formation and embolism in acute AF incorporates insights gained from recent TEE studies (Fig 11). Namely, LA and LAA thrombus may form before the onset of AF in the face of sinus rhythm. Conversion to sinus rhythm may increase the thrombogenic millieu of the left atrium and LAA. Thrombus may form in the acute phase of AF in the left atrium or LAA. Lastly, thrombus may require more than 14 days to organize or resolve.

CLINICAL IMPLICATIONS

The management of patients with AF is not the intention of this review, for which the reader is directed to many excellent reviews on management of patients with A F . 67 However, one issue deserves comment based on the clinical implications of recent TEE studies. It appears appropriate to consider anticoagulation in patients presenting with new onset, acute, or recent onset of AF. In many patients, the precise onset of their arrhythmia cannot be defined. Also, clustering of thromboembolism may occur in the first month of AF, particularly if an embolism has already occurred. It is the opinion of this author that patients presenting with AF, be it acute or chronic, should be immediately anticoagulated if contraindications are not present. Decision-making regarding longterm care can be made later. Patients at higher (eg, mitral stenosis) risk for LA thrombus and thromboembolism should be considered for long-term anticoagulation. Low-risk individuals (eg, lone AF) may require no further evaluation. However, TEE may be useful in patients with acute or new onset nonvalvular AF who are not being considered for long-term anticoagulation because of the clinical impression of a low embolic risk. TEE may show an LA thrombogenic millieu predisposed for thrombus formation and subsequent embolization.

ACKNOWLEDGMENT We acknowledge Ms T.O. Jackson for her excellent secretarial assistance in preparing the manuscript.

78

MARCUS F. STODDARD REFERENCES

1. Ostrander LD Jr, Brandt RL, Kjeisberg MO, et al: Electrocardiographic findings among the adult-population of a total natural community, Tecumseh, Michigan. Circulation 31:888-898, 1965

18. Weinberg DM, Mancini JCB: Anticoagulation for cardioversion of atrial fibrillation. Am J Cardio163:745-746, 1989

2. Orndahl G, Thulesius O, Hood B: Incidence of persistent atrial fibrillation and conduction defects in coronary heart disease. Am Heart J 84:120-131, 1972

19. Arnotd AZ, Mick MJ, Mazurek RP, et al: Role of prophylactic anticoagulation for direct current cardioversion in patients with atrial fibrillation or atrial flutter. J Am Coil Cardiol 19:851-855, 1992

3. Wolf PA, Abbott RD, Kannel WB: Atrial fibrillation: A major contributor to stroke in the elderly--the Framingham study. Arch Intern Med 147:1561-t564, 1987 4. Petersen P, Godtfrendsen J: Atrial fibrillation: A review of course and prognosis. Acta Med Scand 216:5-9, 1984 5. Cameron A, Schwartz MJ, Kranmal RA, et al: Prevalence and significance of atrial fibrillation in coronary artery disease (CASS Registry). Am J Cardiol 61:714-717, 1988 6. Yamanouchi H, Tomonaga M, Shimada H, et al: Nonvalvular atrial fibrillation as a cause of fatal massive cerebral infarction in the elderly. Stroke 20:1653-1656, 1989 7. Wolf PA, Dawber TR, Thomas E, et al: Epidemiologic assessment of chronic fibrillation and risk of stroke: The Framingham Study. Neurology 28:973-977, 1978 8. Petersen P: Thromboembolic complications in atrial fibrillation. Stroke 21:4-13, 1990 9. Cairns JA, Connolly SJ: Nonrheumatic atrial fibrillation. Circulation 84:469-481, 1991 10. Pritchett ELC: Management of atrial fibrillation. N Engl J Med 326:1264-1271, 1992 11. Petersen P, Boysen G, Godtfredsen J, et al: Placebocontrolled, randomized trial of warfarin and aspirin for prevention of thromboembolic complications in chronic atrial fibrillation: The Copenhagen AFASAK Study. Lancet 1:175-179, 1989 12. Boston Area Anticoagulation Trial for Atrial Fibrillation: The effect of low-dose warfarin on the risk of stroke in patients with nonrheumatic atrial fibrillation. N Engl J Med 323:1505-1511, 1990 13. Connolly SJ, Laupacis A, Gent M, et al: Canadian Atrial Fibrillation Anticoagulation (CAFA) Study. J Am Coil Cardiol 18:349-355, 1991 14. Stroke Prevention in Atrial Fibrillation Investigators: Stroke Prevention in Atrial Fibrillation Study: Final results. Circulation 84:527-539, 1991 15. Ezekowitz MD, Bridgers SL, James KE, et al: Warfarin in the prevention of stroke associated with nonrheumatic atrial fibrillation. N Engl J Med 327:1406-1412, 1992 16. Bjerkelund CJ, Orning OM: The efficacy of anticoagu. lant therapy in preventing embolism related to D.C. electrical conversion of atrial fibrillation. Am J Cardiol 23:208216, 1969 17. Roy D, Marchand E, Gagne P, et al: Usefulness of anticoagulant therapy in the prevention of embolic complications of atrial fibrillation. Am Heart J 112:1039-1043, 1986

20. Manning WJ, Silverman DI, Gordon SPF, et al: Cardioversion from atrial fibrillation without prolonged anticoagulation with use of transesophageal echocardiography to exclude the presence of atrial thrombi. N Engl J Med 328:750-755, 1993 21. Black IW, Hopkins AP, Lee LCL, et al: Evaluation of transesophageal echocardiography before cardioversion of atrial fibrillation and flutter in nonanticoagulated patients. Am Heart J 126:375-381, 1993 22. Orsinelli DA, Pearson AC: Usefulness oftransesophageal echocardiography to screen for left atrial thrombus before elective cardioversion for atrial fibrillation. Am J Cardio172:1337-1339, 1993 23. Stoddard MF, Dawkins PR, Prince CR, et al: Transesophageal echocardiographic guidance of cardioversion in patients with atrial fibrillation. Am Heart J 129:1204-1215, 1995 24. Manning WJ, Silverman DI, Keighley CS, et al: Transesophageal echocardiographically facilitated early cardioversion from atrial fibrillation using short-term anticoagulation: Final results of a prospective 4.5-year study. J Am Coll Cardiol 25:1354-1361, 1995 25. Grimm RA, Steward WJ, Malone JD, et al: Impact of electrical cardioversion for atrial fibrillation on left atrial appendage function and spontaneous echo contrast: Characterization by simultaneous transesophageal echocardiography. J Am Coil Cardiol 22:1359-1366, 1993 26. Fatkin D, Kuchar DL, Thorburn CW, et al: Transesophageal echocardiography before and during direct current cardioversion of atrial fibrillation: Evidence for "atrial stunning" as a mechanism of thromboembolic complications, J Am Coil Cardiol 23:307-316, 1994 27. Daniel WG, Nellessen U, Schroder E, et al: Left atrial spontaneous echo contrast in mitral valve disease: An indicator for an increased thromboembolic risk. J Am Coil Cardiol 11:1204-1211, 1988 28. Beppu S, Nimura Y, Sakakibara H, et al: Smoke-like echo in the left atrial cavity in mitraI valve disease: Its features and significance. J Am Coil Cardiol 6:744-749, 1985 29. Castello R, Pearson AC, Labovitz AJ, et al: Prevalence and clinical implications of atrial spontaneous contrast in patients undergoing transesophageal echocardiography. Am J Cardiol 65:1149-1153, 1990 30. Black IW, Fatkin D, Sugar KB, et al: Exclusion of atrial thrombus by transesophageat echocardiography does not preclude embolism after cardioversion of atrial fibrillation. Circulation 89:2509-2513, 1994

-I'HROMBOEMBOLISM

31. De Silva RA, Graboys TB, Podrid PJ, et al: Cardioversion and defibrillation. Am Heart J 100:881-895, 1980 32. Laupacis A, Albers G, Dunn M, et al: Antithrombotic therapy in atrial fibrillation. Chest 102:426S-433S, !992 (suppl) 33. Fisher CM: Reducing risks of cerebral embolism. Geriatrics 34:59-66, 1979 34. Sherman DG, Goldman L, Whiting RB, et al: Thromboembolism in patients with atrial fibrillation. Arch Neurol 41:708-710, 1984 35. Wolf PA, Kannel WB, McGee DL, et al: Duration of atrial fibrillation and eminence of stroke: The Framingham Study. Stroke 14:664-667, 1983 36. Petersen P, Godtfredsen J: Emholic complications in paroxysmal atrial fibrillation. Stroke 17:622-627, 1986 37. Takahashi N, Seki A, Imataka K, et al: Clinical features of paroxysmal atrial fibrillations: An observation of 94 patients. Jpn Heart J 22:143-149, 1981 38. Shimomura K, Ohe T, Ueharas S, et al: Significance of atrial fibrillation as a precursor of embolism. Am J Cardiol 63:1405-1407, 1989 39. Kelly RE, Berger JR, Alter M, et al: Cerebral ischemia and atrial fibrillation: Prospective study. Neurology 34:1285-1291, 1984 40. Hart RG, Coull BM, Hart D: Early recurrent embolism associated with nonvalvular atrial fibrillation: A retrospective study. Stroke 14:688-693, 1983 41. Darling RC, Austen WG, Linton RR: Arterial embolism. Surg Gynecol Obstet 124:106-114, 1987 42. Rothrock JF, Dittrich HC, McAllen S, et al: Acute anticoagulation following cardioembolic stroke. Stroke 20: 730-734, 1989 43. McIlraith DM, C6t6 R, Wolfson C: Acute cerebral infarction in non-valvular atrial fibrillation: Early recurrence of stroke and other ischemic events. Stroke 21:174, 1990 (abstr) 44. Uno M, Tsuji H, Sawada S, et ah Fibrinopeptide A (FPA) levels in atrial fibrillation and the effects of heparin administration. Jpn Circ J 52:9-12, 1988 45. Gustafsson C, Blomback M, Britton M, et al: Coagulation factors and the increased risk of stroke in nonvalvular atrial fibrillation. Stroke 21:47-51, 1990 46. Kumagi K, Fukunami M, Ohmori M, et al: Increased intracardiovascular clotting in patients with nonvalvular atrial fibrillation. J Am Coll Cardiol 16:377-380, 1990 47. Stein B, Fuster V, Halperin JL, et al: Antithrombotic therapy in cardiac disease. An emerging approach based on pathogenesis and risk. Circulation 80:1501-1513, 1989 48. Hochman JS, Platia EB, Bulkley BH: Endocardial abnormalities in left ventricular aneurysms: a clinieopathologic study. Ann Intern Med 100:29-35, 1984 49. Fuster V, Halperin JL: Left ventricular thrombi and cerebral embolism an emerging approach. N Engl J Med 320:392-394, 1989 50. Manning WJ, Leeman DE, Gotch PJ, et al: Pulsed Doppler evaluation of atrial mechanical function after electrical cardioversion of atrial fibrillation. J Am Coll Cardiol 13:617-623, 1989

79 51. Goldman MJ: The management of chronic atrial fibrillation: Indications for and method of conversion to sinus rhythm. Prog Cardiovasc Dis 2:465-479, 1960 52. Stoddard MF, Dawkins PR, Prince CR, et al: Left atrial appendage thrombus is not uncommon in patients with acute atrial fibrillation and a recent embolic event: A transesophageal echocardiographic study. J Am Coll Cardio125:452-459, 1995 53. Mggge A, Daniel WG, Hausmann D, et al: Diagnosis of left atrial appendage thrombi by transesophageal echocardiography: Clinical implications and follow-up. Am J Card Imaging 4:173-179, 1990 54. Manning WJ, Silverman DI, Waksmonski CA, et al: Prevalence of residual left atrial thrombi among patients with acute thromboembolism and Newly recognized atrial fibrillation. Arch Intern Med 155:2193-2198, 1995 55. Pollick C, Taylor D: Assessment of left atrial appendage function by transesophageal echocardiography: Implications for the development of thrombus. Circulation 84:223231, 1991 56. Pozzoli M, Febo O, Torbicki A, et al: Left atrial appendage dysfunction a cause of thrombosis? evidence by transesophageal echocardiography-Doppler studies. J Am Soc Echocardiogr 4:435-441, 1991 57. Garcfa-Fernfindez MA, Torrecilla EG, Roman DS, et al: Left atrial appendage Doppler flow patterns-Implications on thrombus formation. Am Heart J 124:955960, 1992 58. Verhorst PMJ, Kamp O, Visser CA, et al: Left atrial appendage flow velocity assessment using transesophageal echocardiography in nonrheumatic atrial fibrillation and systemic embolism. Am J Cardio171:192-196, 1993 59. Mfigge A, Kfihn H, Nikutta P, et al: Assessment of left atrial appendage function by biplane transesophageal echocardiography in patients with nonrheumatic atrial fibrillation: Identification of a subgroup of patients at increased embolic risk. J Am Coll Cardio123:599-607, 1994 60. Fatkin D, Kelly RP, Feneley MP: Relations between left atrial appendage blood flow velocity, spontaneous echocardiographic contrast and thromboembolic risk in vivo. J Am Coll Cardio123:961-969, 1994 61. Holt BD, Shao Y, Gabel M: Influence of acutely altered loading conditions on left atrial appendage flow velocities. J Am Coil Cardiol 24:1117-1123, 1994 62. Leung DYC, Black IW, Cranney GB, et al: Prognostic implications of left atrial spontaneous echo contrast in nonvalvular fibrillation. J Am Coll Cardiol 24:755-762, 1994 63. Grimm RA, Leung DY, Black IW, et al: Left atrial appendage "stunning" after spontaneous conversion of atrial fibrillation demonstrated by transesophageal Doppler echocardiography. Am Heart J 130:174-176, 1995 64. Black IW, Hopkins AP, Lee LCL, et al: Left atrial spontaneous echo contrast a clinical and echocardiographic analysis. J Am Coll Cardiol. 18:398-404, 1991 65. Blackshear JL, Pearce LA, Asinger RW, et al: Mitral regurgitation associated with reduced thromboembolic events in high-risk patients with nonrheumatic atrial fibrillation. Am J Cardiol. 72:840-843, 1993

80 66. Movsowitz C, Movsowitz HD, Jacobs LE, et al: Significant mitral regurgitation is protective against left atrial spontaneous echo contrast and thrombus as assessed by transesophageal echocardiography. J Am Soc Echocardiogr 6:107-114, 1993

MARCUS F. STODDARD

67. Prystowsky EN, Benson DW, Fuster V, et al: Management of patients with atrial fibrillation: A statement for healthcare professionals from the subcommittee on electrocardiography and electrophysiology, American Heart Association. Circulation 93:1262-1277, 1996