Prevalence of Thrombus, Spontaneous Echo Contrast, and Atrial Stunning in Patients Undergoing Cardioversion of Atrial Flutter
A Prospective Study Using Transesophageal Echocardiography
Background Several studies have shown that patients undergoing cardioversion of atrial fibrillation have a high prevalence of of atrial thrombus and spontaneous echo contrast and frequently develop atrial stunning after restoration of sinus rhythm. These findings are strongly associated with increased risk of embolism in these patients. However, little is known about the prevalence of these markers of thromboembolism in patients undergoing cardioversion of atrial flutter.
Methods and Results We performed transesophageal echocardiography in 47 consecutive, nonanticoagulated patients scheduled for elective cardioversion of atrial flutter. In patients who underwent successful cardioversion, mitral inflow velocity was recorded by transthoracic pulsed Doppler examination immediately after restoration of sinus rhythm. All patients were men (mean age, 65±10 years). Mean duration of flutter was 4±9 weeks. Atrial thrombus and/or spontaneous echo contrast were found in 16 patients (34%). Left atrial thrombus was seen in 5 patients (11%), either with (n=4) or without spontaneous contrast. Of 40 patients with successful cardioversion, atrial mechanical activity was absent in 28% immediately after restoration of sinus rhythm.
Conclusions Our findings suggest that contrary to traditional teaching, atrial thrombus and spontaneous contrast are not uncommon in patients with atrial flutter and cardioversion may be associated with increased risk of thromboembolism.
Several studies have demonstrated a high prevalence of atrial thrombus and spontaneous echo contrast (SEC) in patients undergoing cardioversion of atrial fibrillation.1 2 3 4 In addition, it has been shown that atrial mechanical activity may not return immediately after restoration of sinus rhythm.5 These findings are strongly associated with increased risk of embolic events, and anticoagulation is recommended before cardioversion of atrial fibrillation.6 7 8 Unlike atrial fibrillation, patients with atrial flutter have traditionally been considered at low risk for thromboembolism due to the presence of organized mechanical atrial activity.9 10 Recently, a few studies have shown that atrial thrombus and SEC also occur in patients with atrial flutter.3 11 12 However, the true prevalence of atrial thrombus and SEC in atrial flutter cannot be estimated from these studies because of small numbers, nonconsecutive patient enrollment, referral bias, and the confounding influence of prior anticoagulation. Therefore, we performed a prospective study of consecutive, nonanticoagulated patients with atrial flutter to determine the prevalence of atrial thrombus and SEC and to identify clinical and echocardiographic risk factors associated with these findings. We also evaluated mechanical atrial activity immediately after cardioversion.
All patients >18 years of age with atrial flutter of ≥2 days' duration scheduled to undergo elective cardioversion were screened for enrollment in the study. Exclusion criteria were anticoagulation with heparin or warfarin in the preceding 72 hours, history of atrial fibrillation, mitral stenosis, or any contraindication to transesophageal echocardiography (TEE).
The study protocol was approved by our institutional review board, and all patients referred for elective cardioversion of atrial flutter were screened for participation in the study. Each patient underwent a complete history and physical examination. The diagnosis of atrial flutter was made from a 12-lead ECG with the use of standard criteria.13 Every effort was made to exclude patients with prior atrial fibrillation. For each enrolled patient, hospital records were searched for any admissions in the past 2 years, which were then reviewed for evidence of prior atrial fibrillation. In addition, with the use of a computerized retrieval system (MUSE Network, Marquette Electronics), ECGs performed in the hospital in the preceding 2 years were searched to identify whether any of the patients had an ECG interpreted as atrial fibrillation. All screened patients underwent either 24-hour Holter or telemetry monitoring before cardioversion, and those with intermittent atrial fibrillation were excluded from the study.
After giving written informed consent, all study patients underwent transthoracic and transesophageal echocardiographic examination. The TEE results were communicated to the patients' physicians, who then made decisions regarding anticoagulation and cardioversion. Patients were asked to return at 1 and 4 weeks after the TEE for a brief history and examination to assess for occurrence of clinically evident embolic events. In cases in which cardioversion was deferred, an additional follow-up also was performed 4 weeks after cardioversion.
A two-dimensional transthoracic echocardiogram was performed with either a Hewlett-Packard sonos 2500 with 2.5-MHz phased array transducer or a Vingmed CFM 750 with 2.5-MHz mechanical annular array transducer. Standard parasternal and apical images were obtained and recorded on half-inch VHS tape. With the use of electronic calipers, left atrial dimension was measured in the parasternal view as a line drawn from the posterior aortic root to the posterior left atrial wall.14 Left ventricular systolic function was assessed visually in all views and graded as normal or abnormal.
Immediately after cardioversion, atrial mechanical function was assessed by pulsed Doppler examination of the mitral inflow, using the apical four-chamber view. The Doppler sample volume was placed between the tips of the mitral leaflets and recorded at a speed of 100 mm/s. Atrial mechanical activity was considered present if an atrial filling velocity (A wave) was detected in late diastole (after the ECG P wave).5 Heart rate was <90 bpm in all patients at the time of the Doppler examination.
After topical pharyngeal anesthesia and light sedation, TEE was performed with the use of standard techniques.15 The studies were done on one of the following ultrasound systems with a 5-MHz monoplane, biplane, or multiplane probe: Hewlett-Packard sonos 2500, Acuson 128 XP/10, or Vingmed CFM 750. Both atria and their appendages were carefully examined for thrombus or SEC. A thrombus was defined as an echo-dense intracavitary mass distinct from the underlying endocardium and not caused by pectinate muscles.16 Spontaneous echo contrast was defined as dynamic smokelike echoes within the atrial cavity with a characteristic swirling motion that could not be eliminated by changes in gain settings.17 Left atrial appendage mechanical function was assessed in the following manner. Left atrial appendage velocities were recorded by pulsed Doppler interrogation at the orifice of the appendage.18 The peak appendage emptying velocity within a cardiac cycle was identified and measured using computerized software (Vingmed). During a cardiac cycle, the maximal and minimal left atrial appendage areas were visually identified and planimetered to derive an appendage emptying fraction (maximal area–minimal area/maximal area).18 For analysis, appendage velocities and areas from 6 consecutive cardiac cycles were averaged. All echocardiographic studies were analyzed by one investigator (I.A.), who was unaware of the clinical characteristics of the patients.
All values are reported as mean±SD, and all statistical tests are two-tailed. Univariate and multivariate analyses were done to identify clinical and echocardiographic risk factors associated with atrial thrombus and SEC. The risk factors analyzed included age, duration of flutter, hypertension, congestive heart failure, prior cerebrovascular accident, left atrial size, and left ventricular function. For univariate analysis, logistic regression and Fisher's exact test were used to assess continuous and dichotomous variables, respectively, with statistical significance set at a value of P=.05. Odds ratios and 95% confidence intervals were determined for each variable. For multivariate analysis, all risk factors were entered into a logistic model through stepwise variable selection. A value of P=.1 was chosen for entry and staying in the stepwise regression. Since our sample size was small, choosing a conservative probability value allowed more variables into the model, erring on the side of overparameterization, which adds variability to the prediction but ensures identification of as many risk factors as possible.
Patients were enrolled in the study between February 1995 and March 1996. During this period, 65 patients were scheduled for elective cardioversion of atrial flutter. Of these, 13 were taking warfarin, 3 showed intermittent atrial fibrillation on Holter monitoring, 1 was found to have a previous ECG that showed atrial fibrillation, and 1 had mitral stenosis. The remaining 47 patients, all men (mean age, 65±10 years), formed the study group. The mean duration of flutter was 4±9 weeks (range, 2 days to 15 months). The duration of atrial flutter was estimated from the date of onset of symptoms (n=18) or the date of the first ECG documentation of flutter (n=29). Clinical and echocardiographic characteristics of the study patients are listed in Table 1.⇓ A history of pulmonary or peripheral (other than cerebral) embolism was not obtained in any patient. All 5 patients with a remote history of cerebrovascular accident were in sinus rhythm at the time of the cerebrovascular event. None of the study patients had previously taken coumadin. Twenty-three patients (49%) were taking aspirin and 30 (64%) were on a ventricular rate–controlling medication at the time of the study. None of the patients were on any class I or III antiarrhythmic drugs at the time of the TEE and during the follow-up period.
Prevalence of Atrial Thrombus/SEC
All TEE examinations were completed without any complications. Images were obtained using a multiplane probe in 31 patients; a monoplane (n=9) or biplane (n=7) probe was used in the rest. Both atria and atrial appendages were well visualized in all patients. The Figure⇓ shows the prevalence of atrial thrombus and SEC in the study population. Atrial thrombus was located either in the left atrial appendage (n=3) or body of the left atrium (n=2). In 3 patients, SEC was present in both right and left atria. Right atrial thrombus was not seen in any patient. There was no difference in the prevalence of aspirin use between patients with and without thrombus or SEC on TEE (44% versus 52%, P=.6). The clinical and echocardiographic characteristics of patients found to have atrial thrombus are shown in Table 2.⇓
Atrial Appendage Function in Patients With and Without Atrial Thrombus/SEC
Left atrial appendage Doppler evaluation showed multiple, regular emptying and filling velocities typical of atrial flutter in all patients.12 For all study patients, the peak left atrial appendage velocity was 45±17 cm/s and appendage emptying fraction was 36±16%. Appendage velocities were lower in patients with, compared with those without, thrombus or SEC on TEE (35±15 cm/s versus 50±16 cm/s, P=.01). Similarly, left atrial appendage emptying fraction was lower in patients with thrombus or SEC (24±13% versus 42±13%, P=.002).
Patients with negative TEE. All 31 patients with no thrombus or SEC on TEE underwent successful direct current cardioversion without anticoagulation, and no clinically recognized embolic events occurred during 4 weeks of follow-up.
Patients with SEC. Of the 11 patients with SEC without thrombus, 4 underwent electrical cardioversion without anticoagulation with no embolic events on follow-up. At the discretion of their attending physicians, cardioversion was deferred and anticoagulation started in the other 7 patients. Of these, 1 patient suffered a cerebrovascular accident 4 days after TEE and subsequently was not cardioverted, 1 failed cardioversion, 2 were successfully cardioverted after 3 weeks of anticoagulation with no embolic events, and 3 patients did not return for follow-up.
Patients with thrombus. Cardioversion was deferred and anticoagulation initiated in all 5 patients with atrial thrombus on TEE. Two (40%) of these patients had prior history of a cerebrovascular accident. On follow-up, 1 converted spontaneously 7 days later, 3 were cardioverted successfully after 3 weeks of anticoagulation with no embolic events, and 1 patient was lost to follow-up. Repeat TEE was not performed in any patient.
Prevalence of Atrial Stunning After Cardioversion
Pulsed Doppler examination was performed in 40 patients immediately after successful electrical cardioversion. Atrial mechanical activity was absent in 11 patients (28%) despite the restoration of sinus rhythm. No thrombus or SEC was present on precardioversion TEE in 9 of the 11 patients (82%) found to have atrial stunning after cardioversion.
Predictors of Atrial Thrombus/SEC
The results of univariate and multivariate analyses for identifying predictors of atrial thrombus/SEC in atrial flutter are given in Table 3.⇓ On univariate analysis there was a trend toward higher prevalence of atrial thrombus and SEC in patients with congestive heart failure (58% versus 26%, P=.05) and those with a history of cerebrovascular accident (80% versus 29%, P=.07). On multivariate analysis, congestive heart failure emerged as a significant predictor of atrial thrombus/SEC in our study population (P=.03).
Traditional teaching has been that atrial flutter is an organized rhythm with low risk of thromboembolism.6 9 10 In contrast, our study showed that a third of the patients who presented for cardioversion of atrial flutter demonstrated atrial thrombus and/or SEC. Furthermore, immediately after successful cardioversion, 28% of the patients had absent mechanical atrial activity. In studies on atrial fibrillation, the risk of embolism has been strongly related to the presence of atrial thrombi before cardioversion.2 10 Embolism is believed to occur as a result of dislodgment of a thrombus after restoration of mechanical atrial activity. Similarly, SEC represents a prothrombotic state and is a strong marker of thromboembolism.17 19 In addition, it has been shown that atrial stunning after cardioversion may cause development or worsening of SEC, leading to new thrombus formation.7 18 The high prevalence of these markers of thromboembolism in our study indicates that cardioversion of atrial flutter may be associated with significant risk of embolic events.
Comparison With Previous Studies
Few prior studies have examined the prevalence of thrombus and SEC in patients with atrial flutter and have reported conflicting results. Using TEE, Bikkina et al11 recently studied 24 patients with atrial flutter and reported a 21% prevalence of atrial thrombus. This high prevalence probably is due to patient selection bias, since the study was limited to hospitalized patients and 46% of the subjects underwent TEE to evaluate a cardiac source of embolism. After excluding patients with suspected cardiac embolism, the prevalence of atrial thrombi in the study by Bikkina et al was 11%, similar to our findings. Santiago et al12 studied 17 patients with atrial flutter and found SEC in 6% and thrombi in none. Their study does not reflect true prevalence of these findings in atrial flutter since only patients referred for echocardiography due to a clinical indication were included. Furthermore, 36% of the patients were on anticoagulation at the time of the study. Black et al3 studied 7 nonanticoagulated patients referred for elective cardioversion of atrial flutter and reported atrial thrombus in 14% and SEC in 43%. Our study significantly adds to the few prior reports on this subject. In contrast to prior studies, we enrolled a larger series of consecutive, nonanticoagulated patients, including both inpatients and outpatients, and all subjects in our study underwent TEE as part of the protocol and not because of any clinical indications. Besides assessing the prevalence of thrombus and SEC, we also studied left atrial appendage function before cardioversion.
We also determined the prevalence of atrial stunning immediately after cardioversion of atrial flutter. To our knowledge, only one prior study has evaluated mechanical atrial activity after cardioversion of atrial flutter. Jordaens et al20 studied 22 patients after cardioversion of atrial flutter and reported absent atrial activity in 4 (18%). We found an even higher prevalence of atrial standstill in the present study, thus confirming that similar to atrial fibrillation, patients with atrial flutter are also at substantial risk of developing atrial stunning after cardioversion.
Comparison With Studies on Atrial Fibrillation
Compared with atrial flutter, the prevalence of atrial thrombus and SEC is believed to be much higher in patients with atrial fibrillation.10 Orsinelli et al2 compared TEE findings in 39 patients, 31 with atrial fibrillation and 8 with atrial flutter. Atrial thrombus was present in 26% of patients with atrial fibrillation compared with 13% with atrial flutter. Similarly, Black et al21 reported atrial thrombus in 18 of 150 patients (12%) with atrial fibrillation and 1 of 20 (5%) with atrial flutter. It is difficult to draw any definite conclusions from these studies for several reasons, including retrospective design, patient selection bias (only those referred for TEE, mostly to exclude a cardiac source of embolism, were studied), and enrollment of only a small number of patients with atrial flutter. Few studies have performed TEE in consecutive, nonanticoagulated patients referred for cardioversion of atrial fibrillation or flutter. Manning et al1 recently reported TEE findings in 230 consecutive patients referred for elective cardioversion of atrial fibrillation. They found atrial thrombus in 15% and SEC in 49% of the patients, a prevalence rate only slightly higher compared with our findings in atrial flutter. The majority of thrombi in studies on atrial fibrillation have been reported to occur in the left atrial appendage.1 2 3 In contrast, 40% of the thrombi in our patients with atrial flutter were located in the body of the left atrium. The reason for this finding is unclear. It is possible that regular appendage contractions in flutter reduce the likelihood of thrombus formation in this location.
Embolism After Cardioversion of Atrial Flutter
There are few data on the incidence of embolic events after cardioversion of atrial flutter. In a retrospective study of 122 patients who underwent cardioversion of atrial flutter, Arnold et al9 found no embolic events. However, 26% of their patients were anticoagulated at the time of cardioversion. Pagadala et al22 studied 85 patients referred for cardioversion of chronic atrial flutter. Embolism occurred in 1 of 39 nonanticoagulated patients compared with none of 46 on anticoagulation. The incidence of embolism cannot be estimated from our study because all patients with thrombus and 64% of those with SEC were placed on anticoagulation, and cardioversion was deferred for 3 weeks. Despite this strategy, 1 patient with SEC in whom cardioversion was deferred did suffer a cerebrovascular accident. No embolic events after cardioversion occurred in patients without evidence of thrombus or SEC on precardioversion TEE. Clearly, further work is needed to define the relation of atrial thrombus and SEC in atrial flutter to clinical risk of embolism.
Immediately after cardioversion, atrial mechanical activity was absent in 9 of 31 patients (29%) without thrombus or SEC on TEE. Studies on atrial fibrillation indicate that atrial stunning after cardioversion may predispose the atrium to thrombus formation, and such patients are at increased risk of embolic events.7 18 Our protocol allowed patients' physicians to make decisions regarding anticoagulation before and after cardioversion, which may have reduced the risk of embolism in these patients. Although the present study did not evaluate the incidence of embolization, our data suggest that patients with atrial flutter may be at increased risk for embolic events.
Strengths of Our Study
This is the largest study to date on the prevalence of atrial thrombus and SEC in atrial flutter. To avoid selection bias, we prospectively studied consecutive, nonanticoagulated patients scheduled for elective cardioversion of atrial flutter. The high sensitivity and specificity of TEE for detection of atrial thrombus is well established.23 24 The majority of patients (81%) were studied with biplane or multiplane probes, which are superior to the monoplane probe for examination of the atria and atrial appendages.25 26 Every effort was made to ensure enrollment of patients with pure atrial flutter as detailed in the “Methods” section. It is still possible that intermittent atrial fibrillation contributed to the formation of thrombus and SEC in our study patients. In fact, several studies suggest that patients with atrial flutter have frequent episodes of atrial fibrillation.27 However, recommendations regarding anticoagulation and risk estimates of thromboembolism continue to be different for atrial flutter and fibrillation. On the contrary, our data demonstrate that like atrial fibrillation, patients presenting with atrial flutter also have a high prevalence of atrial thrombus and SEC, suggesting that the two arrhythmias may have similar potential for thromboembolism.
Limitations of Our Study
Since we only studied male patients referred for cardioversion, our results may not reflect the prevalence of thrombus/SEC in all patients with atrial flutter. Our estimation of the duration of atrial flutter may not be accurate particularly in patients without symptoms. However, the interpretation of TEE results was done without knowledge of the estimated duration of flutter. Follow-up for embolic events could not be obtained in all cases because 4 patients did not come for return visits and could not be contacted despite multiple attempts. However, the purpose of the study was to determine the prevalence of atrial thrombus, SEC, and postcardioversion atrial stunning and not the incidence of embolic events in patients presenting for cardioversion of atrial flutter. We also attempted to identify clinical and echocardiographic predictors of thrombus/SEC in atrial flutter. The variables assessed were those found predictive of these findings in patients with atrial fibrillation.28 29 Congestive heart failure was the only significant predictor of thromus/SEC in our study. Our results in this respect should be viewed with caution because of the small sample size. A larger study may identify additional clinical and echocardiographic predictors of these findings. It is interesting that the prevalence of thrombus/SEC tended to be higher in those with prior cerebrovascular accident even though these patients were in sinus rhythm at the time of the cerebrovascular event. It is possible that these patients had paroxysmal atrial flutter that was not identified at the time of the cerebrovascular accident.
Our study demonstrated that 34% of male patients who presented for elective cardioversion of atrial flutter had atrial thrombus and/or SEC. In addition, 28% of patients showed evidence of absent mechanical atrial activity immediately after restoration of sinus rhythm. These findings suggest that cardioversion of atrial flutter may be associated with a significant risk of thromboembolism. A large study is required to define the incidence of embolism in atrial flutter and reassess the need for anticoagulation in these patients.
Presented in part at the 68th Scientific Sessions of the American Heart Association, Anaheim, Calif, November 15, 1995.
- Received August 5, 1996.
- Revision received September 25, 1996.
- Accepted October 5, 1996.
- Copyright © 1997 by American Heart Association
Laupacis A, Albers G, Dunn M, Feinberg W. Antithrombotic therapy in atrial fibrillation: third ACCP conference on antithrombotic therapy. Chest.. 1992;102:426S-433S.
Grimm RA, Stewart WJ, Maloney JD, Cohen GI, Pearce GL, Salcedo EE, Klein AL. Impact of electrical cardioversion for atrial fibrillation on left atrial appendage function and spontaneous echo contrast: characterization by simultaneous transesophageal echocardiography. J Am Coll Cardiol.. 1993;22:1359-1366.
Black IW, Hopkins AP, Lincoln CL, Lee CL, Walsh WF. Thromboembolic risk of atrial flutter. J Am Coll Cardiol.. 1992;19:314A. Abstract.
Pagadala P, Gummadi SS, Olshansky B. Thromboembolic risk of chronic atrial flutter: is the risk underestimated? Circulation. 1994;90(suppl I):I-398. Abstract.
Manning WJ, Waksmonski CA, Haering JM, Weintraub RM, Rooney PS, Maslow AS, Douglas PS. Sensitivity and specificity of transesophageal echo for left atrial thrombi: a prospective, consecutive surgical study. Circulation. 1994;90(suppl I):I-224. Abstract.
SPAF Investigators. Predictors of thromboembolism in atrial fibrillation: clinical features of patients at risk. Ann Intern Med.. 1992;116:1-5.
SPAF Investigators. Predictors of thromboembolism in atrial fibrillation: echocardiographic features of patients at risk. Ann Intern Med.. 1992;116:6-12.