Journal of the American College of Cardiology Vol. 39, No. 9, by the American College of Cardiology Foundation ISSN /02/$22.

Transcription

1 Journal of the American College of Cardiology Vol. 39, No. 9, by the American College of Cardiology Foundation ISSN /02/$22.00 Published by Elsevier Science Inc. PII S (02) CLINICAL STUDIES Embolic Events in Patients With Atrial Fibrillation and Effective Anticoagulation: Value of Transesophageal Echocardiography to Guide Direct-Current Cardioversion Final Results of the Ludwigshafen Observational Cardioversion Study Karlheinz Seidl, MD, Monika Rameken, MD, Axel Drögemüller, MD, Margit Vater, MD, Andreas Brandt, MD, Harald Schwacke, MD, Caroline Bergmeier, MD, Ralf Zahn, MD, Jochen Senges, MD, FACC Ludwigshafen, Germany Atrial Fibrillation OBJECTIVES BACKGROUND METHODS RESULTS CONCLUSIONS The primary objective was to evaluate the usefulness of transesophageal echocardiography (TEE)-guided cardioversion to prevent thromboembolic complications in patients with atrial fibrillation (AF) and effective anticoagulation (International Normalized Ratio of 2 or 3) at least three weeks before cardioversion. Transesophageal echocardiography has been proposed as a method of screening patients for left atrial thrombi before direct-current cardioversion of AF. The usefulness of TEE as a screening tool has always been evaluated in patients without long-term anticoagulation before cardioversion. This prospective, single-center, observational study, performed on an intention-to-cardiovert basis, comprised 1,076 consecutive, unselected patients with AF. The initial two years were designed to be the control phase, during which the conventional approach was used. After that, cardioversion guided by TEE was performed in consecutive patients. The prevalence of left atrial thrombi was 7.7% in patients with persistent AF and effective anticoagulation. During the first four weeks after electrical cardioversion, six thromboembolic complications were observed in patients in whom the TEE-guided approach was employed (6 [0.8%] of 719 patients), compared with three thromboembolic complications in patients in whom the conventional approach was used (3 [0.8%] of 357 patients). None of the patients in whom electrical cardioversion was not performed experienced an embolic event. There were no differences in the rate of embolic events between the two treatment groups. In patients with AF and effective anticoagulation, TEE-guided electrical cardioversion does not reduce the embolic risk. However, TEE revealed left atrial thrombi in 7.7% of patients with AF and effective anticoagulation, before direct-current cardioversion. (J Am Coll Cardiol 2002;39: ) 2002 by the American College of Cardiology Foundation From the Department of Cardiology, Herzzentrum Ludwigshafen, Ludwigshafen, Germany. Manuscript received October 10, 2001; revised manuscript received January 24, 2002, accepted February 6, Transesophageal echocardiography (TEE) has been demonstrated to be more sensitive than transthoracic echocardiography for the detection of left atrial thrombi and spontaneous echo contrast, a putative marker of thromboembolic risk (1 5). Clinical data are emerging in support of the usefulness and safety of the TEE-guided approach to cardioversion (1 11). However, all previous studies have evaluated the usefulness in patients with no anticoagulation or only short-term anticoagulation before direct-current cardioversion (6 11). The primary objective of this study was to evaluate the usefulness of TEE-guided cardioversion to prevent thromboembolic complications in patients with atrial fibrillation (AF) and effective anticoagulation (International Normalized Ratio [INR] of 2 to 3) at least three weeks before direct-current cardioversion in clinical practice. METHODS Patient selection. Patients who were candidates for electrical cardioversion of AF were eligible for enrollment in the study. Patients were excluded if 1) anticoagulation was ineffective (INR 2); or 2) they had hemodynamically unstable AF that had to be cardioverted. All patients gave written, informed consent. Study design. This prospective, single-center, observational study was designed on an intention-to-cardiovert basis. During the first two years (control phase), patients were treated according to the recommendations of the American College of Chest Physicians (12): three weeks of warfarin therapy before cardioversion, followed by four weeks of warfarin therapy, and a follow-up examination at the end of four weeks. Effective anticoagulation was

2 JACC Vol. 39, No. 9, 2002 May 1, 2002: Seidl et al. Cardioversion of Atrial Fibrillation 1437 Abbreviations and Acronyms ACUTE Assessment of Cardioversion Using Transesophageal Echocardiography study AF atrial fibrillation INR International Normalized Ratio LV left ventricular TEE transesophageal echocardiography TTE transthoracic echocardiography achieved if the patient had an INR between 2 and 3 at least three weeks before electrical cardioversion. After the initial two years, TEE-guided cardioversion was performed in consecutive patients. If thrombi were detected, cardioversion was postponed. In patients with moderate to severe spontaneous echo contrast, the decision to cardiovert was left to the treating physician. The anticoagulation regimen was the same as the regimen in the conventional (control) group. Patients with ineffective anticoagulation. Ineffective anticoagulation was established if warfarin was given but the INR was 2, if no anticoagulation was performed because of AF for 48 h or if there were contraindications against anticoagulation. Patients with ineffective anticoagulation were excluded; however, follow-up was performed. The reasons for performing direct-current cardioversion in patients with ineffective anticoagulation were AF for 48 h, hemodynamically unstable AF or contraindications against anticoagulation. After cardioversion, a combination of warfarin therapy and intravenous heparin therapy was given to maintain adequate anticoagulation after cardioversion in patients without contraindications against anticoagulation. Echocardiography before cardioversion. Conventional transthoracic echocardiography (TTE) was performed in both study groups before cardioversion (13 16). During the TEE study, special attention was paid to assess the presence or absence of left atrial thrombi and spontaneous echo contrast (13 16). A thrombus was considered to be present if a mass detected in the appendage or body of the atrium appeared to be distinct from the underlying endocardium, was not caused by pectinate muscles and was detected in more than one imaging plane (17,18). Spontaneous echo contrast was defined as dynamic smoke-like echoes within the atrial cavity, with the characteristic swirling motion that could not be eliminated by changes in gain settings. The degree of spontaneous echo contrast was characterized independently as absent, mild, moderate or severe (17,19). The severity of spontaneous echo contrast was graded according to the following criteria: mild was defined as minimal echogenicity located in the left atrial appendage or sparsely distributed in the main cavity of the left atrium, which was possible to detect only transiently during the cardiac cycle, but imperceptible at operating gain settings for two-dimensional echocardiographic analysis. Moderate was defined as a dense, swirling pattern in the left atrial appendage, generally associated with somewhat lesser intensity in the main cavity, which may fluctuate in intensity, but detectable constantly throughout the cardiac cycle. Severe was defined as an intense echo density and a very slow swirling pattern in the left atrial appendage, usually with a similar density in the main cavity (18 21). The delay between these echocardiographic studies and the scheduled cardioversion was 3 h. Clinical outcomes. Determination of a thromboembolic event was based on the clinical findings and supported by radiologic studies. The clinical safety outcomes were clinically apparent ischemic stroke, transient ischemic attack, systemic embolization, death related to cardioversion and episodes of bleeding. Other outcome variables were the prevalence of thrombi, the number of patients without thrombi who had cardioversion and the immediate rhythm after cardioversion. These outcomes were assessed for four weeks after cardioversion. In patients in whom no cardioversion was performed because of spontaneous conversion to sinus rhythm or contraindications against cardioversion, these variables were assessed for four weeks after the date of the planned cardioversion. Statistical analysis. Clinical echocardiographic and outcome data are reported as the mean value SD. For dichotomous outcome variables, percentages are given. The Mann-Whitney U test was used for the comparison of continuous variables between two groups. For the comparison of dichotomous outcomes, the chi-square test or Fisher exact test for small numbers was performed, and crude odds ratios with 95% confidence intervals were computed. All statistical tests were two-tailed, with statistical significance set at p RESULTS Patient group. Between January 1994 and July 2000, 1,269 consecutive patients were registered. Anticoagulation was ineffective in 193 patients. Therefore, 1,076 patients with AF and effective anticoagulation were entered into the final analysis. The TTE group (conventional therapy group) consisted of 357 patients, and the TEE group consisted of 719 patients. Patients in the TEE group were older and more often had a history of hypertension, a history of stroke or transient ischemic attack, structural heart disease, an ejection fraction 40% and a larger left atrial diameter. In addition, patients in the TEE group more often had episodes lasting longer than two days, compared with patients in the conventional treatment group. Regarding anticoagulant therapy, no differences between the two groups were detected (Table 1). Conventional therapy group. CARDIOVERSION SUCCESS. Electrical cardioversion was performed in 355 (99%) of 357 patients. Cardioversion was successful in 278 (78%) of 355 patients (Table 1). In two patients, no electrical cardioversion was performed

3 1438 Seidl et al. JACC Vol. 39, No. 9, 2002 Cardioversion of Atrial Fibrillation May 1, 2002: Table 1. Patient Characteristics: Comparison Between Patients Studied by Transesophageal Versus Transthoracic Echocardiography Total Group (N 1,076) TTE Group (n 357) TEE Group (n 719) p Value Age in years (n) (1,076) (357) (719) Male 74% (797/1,076) 78% (278/357) 72% (519/719) 0.05 Hypertension 54% (583/1,070) 50% (176/351) 57% (407/719) 0.05 Diabetes 12% (124/1,069) 9% (33/351) 13% (91/718) NS History of TIA/stroke 10% (110/1,076) 8% (27/357) 12% (83/719) 0.05 Duration of AF 2 days 85% (662/777) 76% (170/225) 89% (492/552) Duration of AF in median days (n) 28 (777) 21 (225) 28 (552) Organic heart disease 57% (608/1,069) 51% (180/350) 60% (428/719) 0.01 CAD 40% (242/608) 33% (60/180) 43% (182/428) DCM 15% (92/608) 12% (22/180) 16% (70/428) NS HHD 42% (254/608) 43% (77/180) 41% (177/428) NS VHD 27% (165/608) 27% (49/180) 27% (116/428) NS Others 5% (29/608) 8% (14/180) 4% (15/428) 0.02 EF 40% 15% (152/1,044) 7% (24/326) 18% (128/178) LA diameter in mm (n) 46 6 (1,007) 45 6 (306) 46 6 (701) Spontaneous echo contrast 52% (374/719) ND 52% (374/719) Mild 63% (237/374) 63% (237/374) Moderate 27% (100/374) 27% (100/374) Severe 10% (37/374) 10% (37/374) LA thrombus 7.7% (55/719) ND 7.7% (55/719) INR (n) (1,076) (357) (711) NS CV performed 87% (941/1,076) 99% (355/357) 82% (586/719) CV not performed 13% (135/1,076) 1% (2/357) 18.5% (133/719) Spontaneous CV 2% (26/1,076) 0.3% (1/357) 3% (25/719) 0.01 Successful CV 83% (783/941) 78% (278/355) 86% (505/586) 0.01 Data are presented as the mean value SD or percentage (n/n) of patients). AF atrial fibrillation; CAD coronary artery disease; CV electrical cardioversion; DCM dilated cardiomyopathy; EF ejection fraction; HHD hypertensive heart disease; INR International Normalized Ratio; ND not done; NS not significant; LA left atrial; TIA transient ischemic attack; TEE transesophageal echocardiography; TTE transthoracic echocardiography; VHD valvular heart disease. (1 patient had spontaneous restoration of sinus rhythm, and in 1 patient, the treating physician decided against cardioversion) (Fig. 1A). OUTCOME. In the conventional therapy group, a total of three thromboembolic events occurred in the first four weeks after cardioversion (0.8%). In the initial 48 h after electrical cardioversion, no thromboembolic complications occurred in these patients. All three patients experienced a thromboembolic stroke. No bleeding complications were observed in the conventional therapy group (Table 2). Transesophageal echocardiography group. TRANS- ESOPHAGEAL ECHOCARDIOGRAPHY BEFORE CARDIOVER- SION. A thrombus was found in the left atrium in 55 (7.7%) of 719 patients. In 374 (52%) of the 719 patients, TEE revealed spontaneous echo contrast. The grade of spontaneous echo contrast was mild in 63%, moderate in 27% and severe in 10% of patients (Table 1). CARDIOVERSION SUCCESS. In 586 patients (82%), electrical cardioversion was performed. It was successful in 505 (86%) of the 586 patients (Table 1). Cardioversion was not performed in 133 patients (18.5%) in the TEE-guided group: 25 (3%) had spontaneous restoration of sinus rhythm; 104 patients did not have cardioversion because TEE revealed left atrial thrombi or moderate to severe spontaneous echo contrast; 3 patients had no cardioversion because the treating physician decided against it; and 1 patient declined cardioversion (Fig. 1A). OUTCOME. In the TEE-guided group, a total of six thromboembolic events (Table 2) occurred in the first four weeks after cardioversion (0.8%). Two patients had a transient ischemic attack in the first 48 h after cardioversion. After 48 h and during the first four weeks after cardioversion, another four thromboembolic events occurred in the TEEguided group. Two patients had an ischemic stroke, one patient had a transient ischemic attack and one patient had a peripheral embolism. None of the patients with thromboembolic complications after cardioversion had evidence of left atrial thrombi before cardioversion. In all patients with a thrombus, the INR was increased to 3 to 3.5. After four weeks, resolution of thrombus was observed in 55% of patients. During four weeks of follow-up, two bleeding complications occurred. One patient had a hemorrhagic stroke three weeks after the planned electrical cardioversion. This patient had not been cardioverted because of a thrombus in the orifice of the left atrial appendage (INR 3.1). One patient had lung bleeding during the first four weeks of follow-up (INR 4.7) (Table 2). None of the 133 patients without electrical cardioversion experienced a thromboembolic complication.

4 JACC Vol. 39, No. 9, 2002 May 1, 2002: Seidl et al. Cardioversion of Atrial Fibrillation 1439 Figure 1. Distribution of patients by (A) effective anticoagulation and (B) ineffective anticoagulation. AC anticoagulation; CV cardioversion; pt(s) patient(s); TEE transesophageal echocardiography.

5 1440 Seidl et al. JACC Vol. 39, No. 9, 2002 Cardioversion of Atrial Fibrillation May 1, 2002: Table 2. The Precardioversion Clinical and Echocardiographic Characteristics of Patients With an Embolic Complication or Bleeding Embolic Complications Bleeding Pt. 1 Pt. 2 Pt. 3 Pt. 4 Pt. 5 Pt. 6 Pt. 7 Pt. 8 Pt. 9 Pt. 10 Pt. 11 Gender M F M M F M M M F M M Age (years) Organic heart disease None None None HHD CAD CAD None None None CAD Valvular LV function Normal Normal Normal Normal Normal Normal Normal Normal Normal Discrete Normal Type of echocardiography TEE TEE TEE TEE TEE TEE TTE TTE TTE TEE TEE LA diameter (mm) SEC None Mild None None Moderate None Mild None LA thrombus None None None None None None 1 None Effective AC Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes CV performed Yes Yes Yes Yes Yes Yes Yes Yes Yes No Yes Success Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Acute complications TIA TIA None None None None None None None None None Lung bleeding Hemorrhagic infarction Complications at 4 weeks None None TIA Peripheral embolic event Recurrence of AF at 4 weeks Yes NA No No Yes NA NA No Yes No Rhythm at time of event SR NA SR SR AF NA NA SR AF AF SR AC anticoagulation; F female; LV left ventricular; M male; NA not applicable; Pt. patient; SEC spontaneous echo contrast; SR sinus rhythm; other abbreviations as in Table 1. Comparison of patients with and without a thromboembolic event. Although the number of patients with embolic complications was too small to permit meaningful statistical comparisons between the groups, it is notable that patients with embolic complications did not have a longer duration of AF (32 vs. 28 days, p NS), nor a larger left atrium (46 4 vs mm, p NS), nor a greater prevalence of spontaneous echo contrast (33% vs. 52%, p NS), mitral valve disease (p NS) or left ventricular (LV) dysfunction (0% vs. 15%, p NS), compared with patients without thromboembolic complications. Patients with ineffective anticoagulation. In 193 patients, anticoagulant therapy was ineffective at the time of the intended cardioversion. Patients with ineffective anticoagulation were younger (59 12 vs years, p 0.001), less often had a history of hypertension (46% vs. 54%, p 0.032), more often had AF that lasted 48 h (61% vs. 15%, p 0.001), less often had organic heart disease (49% vs. 57%, p 0.036), less often had impaired LV function (ejection fraction 40%; 9% vs. 15%, p 0.035) and more often had a smaller left atrium (44 6 vs mm, p 0.001), compared with patients with effective anticoagulation. Transesophageal echocardiography was performed in 126 patients. A thrombus was found in 10 patients (7.9%). No differences were observed when comparing patients who had a thrombus detected with effective anticoagulation to patients who had a thrombus detected with ineffective anticoagulation. Outcomes. In 150 patients (78%), electrical cardioversion was performed. Cardioversion was successful in 128 (85%) of 150 patients (Fig. 1B). In 2 (1.3%) of 150 patients, a transient ischemic attack occurred in the first four weeks after cardioversion. In both patients, TEE was performed before cardioversion. None of the patients had evidence of a left atrial thrombus before cardioversion. However, TEE revealed mild to moderate spontaneous echo contrast. None of the 43 patients without electrical cardioversion experienced a thromboembolic complication. DISCUSSION Main findings. This prospective, single-center, observational study evaluated the usefulness of TEE-guided cardioversion to prevent thromboembolic complications in patients with AF and effective anticoagulation (INR 2 to 3) at least three weeks before direct-current cardioversion. During follow-up, there were no significant differences in the rate of embolic events between the two treatment groups (TTE vs. TEE). Overall, nine thromboembolic complications occurred. Six thromboembolic complications were observed in the patients in whom the TEE-guided approach to cardioversion was employed (6 [0.8%] of 719 patients), compared with three thromboembolic complications in the patients in whom the conventional approach was used (3 [0.8%] of 357 patients). Severe hemorrhagic events occurred in two patients in the

6 JACC Vol. 39, No. 9, 2002 May 1, 2002: Seidl et al. Cardioversion of Atrial Fibrillation 1441 TEE group. No severe bleeding complications were observed in the conventional treatment group. Thromboembolism after cardioversion. In contrast to our study, the objective of the recently published Assessment of Cardioversion Using Transesophageal Echocardiography (ACUTE) study (11) was to compare the conventional anticoagulation strategy with the strategy of using TEE to guide short-term anticoagulant therapy in patients with AF and planned electrical cardioversion. In the ACUTE study (11), there were no significant differences in the rate of embolic events between the two treatment groups (0.8% in the TEE-guided group vs. 0.5% in the conventional group). The rate of embolic events was comparable in both the ACUTE study (11) and present study. Patients in the TEE group were older and more often had organic heart disease, impaired LV function, episodes of AF lasting 48 h and a larger left atrium. However, all of these differences are minor and not clinically relevant. Transesophageal echocardiography in patients with effective anticoagulation. In the ACUTE study (11), a thrombus was detected more often than in our study (13.8% vs. 7.7%). This observation is explained by the different anticoagulation strategy. Several investigators (22 24) have reported resolution of thrombus in patients with AF after more than two weeks of adjusted-dose warfarin therapy. In the Stroke Prevention in Atrial Fibrillation (SPAF-III) trial, thrombi were noted in 10% of all patients (22,23). Thrombi were seen with a similar frequency in both treatment groups (adjusted-dose warfarin [INR 2 to 3] and combination therapy with low-dose warfarin therapy [INR 1.2 to 1.5] and aspirin), when TEE was performed within two weeks of randomization, but the incidence in the adjusted-dose warfarin group was much lower when TEE was done later (22). Similarly, Collins et al. (24) reported thrombus resolution in 86% of patients with AF after one month of warfarin therapy (INR 2 to 3). In our study, thrombus resolution was found in 55% of patients after one month of adjusted-dose warfarin therapy (INR 3 to 3.5). A similar prevalence of thrombi in the TEE-guided group should be expected in the conventional treatment group. These patients were not excluded from cardioversion, because they were not identified. Nevertheless, the embolic rate was similar. One might argue that in our study group, patients in the conventional group differed from those in the TEE-guided group. In contrast, when comparing the patient group in the conventional arm of the ACUTE study with our patients in the TEE-guided group, no major differences were observed. This supports the hypothesis that the prevalence of thrombi is 7.7% in patients after at least three weeks of effective anticoagulation. Thromboembolism after cardioversion of AF has been attributed to the dislodgment of preformed atrial mural thrombi during the resumption of atrial contraction (8,25,26). However, our observation suggests that thromboembolism after cardioversion may more often arise as a consequence of the effects of cardioversion than from the dislodgment of a pre-existent thrombi. Our understanding is that the recommendation of three weeks of anticoagulation before cardioversion is empirical, and there is no such study to prove that three weeks of anticoagulation is necessary. Investigators in this field may wish to study whether a shorter period of anticoagulation and cardioversion without TEE still provides a similar rate of embolic events. It is possible that the reason we do not have a significant difference in the embolic rate between cardioversion with and without TEE is that the duration of anticoagulation before cardioversion may not make much difference clinically, although this will be difficult to prove. Severe bleeding complications. In the ACUTE study (11), the rate of hemorrhagic events was significantly lower in the TEE-group than in the conventional group (2.9% vs. 5.5%). A comparison of severe hemorrhagic events showed no significant differences. In the present study, hemorrhagic events occurred in two patients in the TEE group. No bleeding complications were observed in the conventional treatment group. In both patients with bleeding complications, the INR was 3. Study limitations. This study was not randomized; it is a clinical, observational study from a single center. The study included a control period with conventional cardioversion and a TEE-guided approach. Patients who had the TEEguided approach to direct-current cardioversion were sicker in terms of more often having structural heart disease, LV dysfunction and a larger left atrium. However, these differences are minor and not clinically relevant. Furthermore, the low prevalence of thromboembolic events limits the conclusions that can be drawn from our data. Conclusions. The results of this study, in combination with the results of the ACUTE study, may have two clinical implications: 1) TEE before direct-current cardioversion is not needed in patients with effective anticoagulation at least three weeks before cardioversion; and 2) if TEE is performed early, more thrombi will be detected and more patients will be excluded from cardioversion. However, in patients with a normal TEE study, early cardioversion can be performed safely. The impact of thrombi detected during TEE on long-term prognosis must be evaluated. Reprint requests and correspondence: Dr. Karlheinz Seidl, Herzzentrum Ludwigshafen, Bremserstrasse 79, Ludwigshafen, Germany. Seidlk@klilu.de. REFERENCES 1. Pearson AC, Labovitz AJ, Tatineni S, Gomez CR. Superiority of transesophageal echocardiography in detecting cardiac source of embolism in patients with cerebral ischemia of uncertain etiology. J Am Coll Cardiol 1991;17: Acar J, Cormier B, Grimberg D, et al. Diagnosis of left atrial thrombi in mitral stenosis: usefulness of ultrasound techniques compared with other methods. Eur Heart J 1991;12 Suppl B: Lin SL, Hsu TL, Liou JY, et al. Usefulness of transesophageal echocardiography for the detection of left atrial thrombi in patients with rheumatic heart disease. Echocardiography 1992;9:161 8.

7 1442 Seidl et al. JACC Vol. 39, No. 9, 2002 Cardioversion of Atrial Fibrillation May 1, 2002: 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 Coll Cardiol 1988;11: Castello R, Pearson AC, Labovitz AJ. Prevalence and clinical implications of atrial spontaneous echo contrast in patients undergoing transesophageal echocardiography. Am J Cardiol 1990;65: Bjerkelund CJ, Orning OM. The efficacy of anticoagulant therapy in preventing embolism related to DC electrical conversion of atrial fibrillation. Am J Cardiol 1969;23: Weinberg DM, Mancini GBJ. Anticoagulation for cardioversion of atrial fibrillation. Am J Cardiol 1989;63: Arnold AZ, Mick MJ, Mazurek RP, Loop FD, Trohman RG. Role of prophylactic anticoagulation for direct current cardioversion in patients with atrial fibrillation or atrial flutter. J Am Coll Cardiol 1992;19: Stein B, Haperin JL, Fuster V. Should patients with atrial fibrillation be anticoagulated prior to and chronically following cardioversion? In: Cheitlin MD, editor. Dilemmas in Clinical Cardiology. Philadelphia, PA: F.A. Davis, 1990: Moreyra E, Finkelhor RS, Cebul RD. Limitations of transesophageal echocardiography in the risk assessment of patients before nonanticoagulated cardioversion from atrial fibrillation and flutter: an analysis of pooled trials. Am Heart J 1995;129: Klein AL, Grimm RA, Murray RD, et al., for the Assessment of Cardioversion Using Transesophageal Echocardiography Investigators. Use of transesophageal echocardiography to guide cardioversion in patients with atrial fibrillation. N Engl J Med 2001;344: Laupacis A, Albers G, Dalen J, Dunn M, Feinberg W, Jacobson A. Antithrombotic therapy in atrial fibrillation. Chest 1995;108:352S 9S. 13. Sahn DJ, DeMaria A, Kisslo J, Weymann A. Recommendations regarding quantitation in M-mode echocardiography: results of a survey of echocardiographic measurements. Circulation 1978;58: Seward JB, Khandheria BK, Oh JK, et al. Transesophageal echocardiography: technique, anatomic correlations, implementation, and clinical applications. Mayo Clin Proc 1988;63: Seward JB, Khandheria BK, Edward WD, Oh JK, Freeman WK, Tajik AJ. Biplanar transesophageal echocardiography: anatomic correlations, image orientation, and clinical applications. Mayo Clin Proc 1990;65: Seward JB, Khandheria BK, Freeman WK, et al. Multiplane transesophageal echocardiography: image orientation, examination technique, anatomic correlations, and clinical applications. Mayo Clin Proc 1993;68: Irani WN, Grayburn PA, Afridi I. Prevalence of thrombus, spontaneous echo contrast, and atrial stunning in patients undergoing cardioversion of atrial flutter: a prospective study using transesophageal echocardiography. Circulation 1997;95: Fatkin D, Kelly RP, Feneley MP. Relations between left atrial appendage blood flow velocity, spontaneous echocardiographic contrast and thrombo-embolic risk in vivo. J Am Coll Cardiol 1994;23: Fatkin D, Kuchar DL, Thorburn CW, Feneley MP. Transesophageal echocardiography before and during direct current cardioversion of atrial fibrillation: evidence for atrial stunning as a mechanism of thromboembolic complications. J Am Coll Cardiol 1994;23: Deppu S, Numura Y, Sakakibara H, Nagata S, Park YD, Izumi S. Smoke-like echo in the left atrial cavity in mitral valve disease: its features and significance. J Am Coll Cardiol 1985;6: Daniel WG, Nellsessen U, Schroder E, et al. Left atrial spontaneous echo contrast in mitral valve disease: an indicator for an increased thromboembolic risk. J Am Coll Cardiol 1988;11: The Stroke Prevention in Atrial Fibrillation Investigators Committee on Echocardiography. Transesophageal echocardiographic correlates of thromboembolism in high-risk patients with nonvalvular atrial fibrillation. Ann Intern Med 1998;128: Adjusted-dose warfarin versus low-intensity, fixed-dose warfarin plus aspirin for high-risk patients with atrial fibrillation: Stroke Prevention in Atrial Fibrillation III randomised clinical trial. Lancet 1996;348: Collins LJ, Silverman DI, Douglas PS, Manning WJ. Cardioversion of nonrheumatic atrial fibrillation: reduced thromboembolic complications with 4 weeks of precardioversion anticoagulation are related to atrial thrombus resolution. Circulation 1995;92: Mancini GBJ, Goldberger AL. Cardioversion of atrial fibrillation: consideration of embolization, anticoagulation, prophylactic pacemaker, and long-term success. Am Heart J 1982;104: Goldman MJ. The management of chronic atrial fibrillation: indications for and method of conversion to sinus rhythm. Prog Cardiovasc Dis 1960;2: