Search tips
Search criteria 


Logo of expclincardiolExperimental and Clinical Cardiology HomepageSubscription PageSubmissions Pagewww.pulsus.comExperimental and Clinical Cardiology
Exp Clin Cardiol. 2001 Winter; 6(4): 200–205.
PMCID: PMC2859000
Clinical Cardiology

Factors related to the early and late success of direct current cardioversion of chronic nonrheumatic atrial fibrillation: An echocardiographic study

Marcin Szulc, MD PhD,1 Hubert Gurba, MD PhD,1 Agnieszka Kuch-Wocial, MD PhD,1 Piotr Pruszczyk, MD PhD,1 Zbigniew Miskiewicz, MD PhD,1 Inga Chlewicka, MD PhD,1 Jolanta Stanislawska, MD PhD,2 Aleksander Górecki, MD,2 Grazyna Swiecicka, MD,2 Grzegorz Opolski, MD PhD FESC,2 and Adam Torbicki, MD PhD FESC3



To assess factors related to the success of restoration and one-year maintenance of sinus rhythm in chronic (more than 48 h) nonrheumatic atrial fibrillation (AF).


One hundred and fifty consecutive patients aged 62±9 years with AF lasting 123±254 days were evaluated clinically with transthoracic and transesophageal echocardiography before elective direct current cardioversion. Heart chamber dimensions and left ventricular ejection fraction were measured. The presence of left atrial thrombi and spontaneous echocardiographic contrast as well as flow velocities in the left atrial appendage were assessed. The first cardioversion was followed by standardized two-step antiarrhythmic treatment including a second cardioversion, if necessary. Twenty patients (13%) spontaneously reverted to sinus rhythm (S) during anticoagulation preceding cardioversion, 81 (54%) were successfully cardioverted (Y), and in 49 (33%) cardioversion failed initially (N). No differences were noted between the two latter groups. However, S patients had smaller left atria measured in the short and long axes (42±4 mm, P=0.05, and 53±7 mm, P=0.005, respectively) than both the Y (45±4 and 61±8 mm) and the N patients (46±4 and 61±8 mm). One-year follow-up was obtained in 95 patients: 64 (67%) were in sinus rhythm while 31 (33%) had AF. Again, no initial differences predicting the maintenance of sinus rhythm were found.


Spontaneous reversion of AF seems more likely with smaller left atria. Echocardiography, including trans-esophageal echocardiography, is unlikely to identify patients in whom attempts to restore and maintain sinus rhythm will fail or succeed.

Keywords: Atrial fibrillation, Cardioversion, Echocardiography

Nonrheumatic atrial fibrillation (AF) is a substantial epidemiological problem affecting about 3% of people aged 60 years or more. There are two generally accepted management strategies for patients with chronic nonrheumatic AF: long term anticoagulation with heart rate control and cardioversion with subsequent efforts to maintain sinus rhythm.

It has been reported that restoration of sinus rhythm results in a decreased left atrial diameter and an increased left ventricular ejection fraction (1). Also, spontaneous echocardiographic contrast initially present in the left atrium was reported to disappear one year after successful cardioversion (2), suggesting a decreased risk of peripheral embolism.

However, benefits resulting from cardioversion should be expected only if sinus rhythm is maintained long enough. Although the initial success of cardioversion may reach 94% (3), after six months sinus rhythm remains in only 40% to 60% of patients, depending on the underlying disease and type of anti-arrhythmic treatment applied (4). Despite intensive efforts to maintain sinus rhythm, this proportion decreased to about 30% over a four-year follow-up (5).

Thus, cardioversion would be best suited to patients with a high probability of long term maintenance of sinus rhythm after cardioversion. However, the identification of such patients remains obscure.

The objective of the present study was to assess the role of transthoracic (TTE) and transesophageal echocardiography (TEE) in predicting initial and late success of electrical cardioversion in patients with chronic nonrheumatic AF.



The study group consisted of 150 consecutive patients, 69 female and 81 male, with nonrheumatic AF lasting more than 48 h or of unknown duration who were scheduled for elective direct current (DC) cardioversion. Their average age was 62.1±9.1 years and ranged from 33 to 79 years. AF had lasted from two days to four years (mean 122.9±254.2 days) (Table 1).

General characteristics of the study population (n=150)

Study protocol:

All patients initially scheduled for elective cardioversion were referred to the echocardiographic laboratory. The result of echocardiographic examination described below did not influence the decision of whether to cardiovert a patient, with the exception of mitral stenosis, which excluded patients from the trial.

In addition to echocardiography, clinical data were collected and a standard 12-lead electrocardiogram was recorded. After the initial examination and a period of prophylactic anticoagulation, patients were referred for electrical cardioversion of AF. After successful cardioversion, preventive treatment with one of three randomly selected antiarrhythmic drugs (disopiramide, propafenone or sotalol) was instituted. In patients after an unsuccessful first cardioversion, another attempt was made to restore the sinus rhythm. In all of these patients, the second DC cardioversion was preceded by oral administration of 6 g amiodarone for several days. The amiodarone treatment was continued in patients with a successful second cardioversion. If AF relapsed during the one-year follow-up after the first successful cardioversion, then patients underwent the same treatment as when the first cardioversion failed; that is, another cardioversion preceded by amiodarone treatment was attempted. In the case of the second recurrence of AF independent of obvious and removable causes (such as electrolyte disturbances), AF was considered permanent. Moreover, apart from the initial assessment, all investigated patients were examined 45 min and one year after the cardioversion. A standard 12-lead electrocardiogram was recorded at every visit. After one year, data concerning rhythm disturbances during the follow-up period were collected.

The presence of sinus rhythm evaluated 45 min after the first cardioversion or spontaneous return of sinus rhythm during the period of anticoagulation was considered to be an early success. Late success was defined as maintenance of sinus rhythm for one year, regardless of whether one or two cardioversions had been attempted and of the type of anti-arrhythmic drugs used.

The protocol of the study was accepted by the local institutional ethics committee. Informed consent was obtained from all patients participating in the trail. The study was performed in accordance with the Declaration of Helsinki.

Clinical evaluation:

The following clinical parameters were assessed: age, sex, duration of the current episode of AF, and history of coronary artery disease, heart failure, diabetes, hypertension and embolic episodes.

The evaluation was based on the history and available documentation. Any doubts about AF duration, diagnosis of coronary artery disease, heart failure, hypertension and diabetes were classified as missing data.

Echocardiographic examination:

Echocardiographic examination was performed with the Sonos 500 and Sonos 1000 Hewlett Packard (USA) systems. TTE was done with a 2.5 MHz probe, and during TEE a 5 MHz single-or biplane probe was used.

During TTE the heart was visualized in the parasternal long and short axes and in the apical views.

TEE was performed after local anaesthesia of the posterior pharynx with a 10% lidocaine solution spray. In a few cases sedation with 5 mg diazepam given intravenously was needed. The transesophageal probe was inserted to a level 25 to 35 cm from the dental arcs. The heart was always visualized in the horizontal plane, and in some patients also in vertical plane. All examinations were videotaped.

During TTE examinations, left ventricular end-diastolic diameter, left ventricular end-systolic diameter (only for left ventricular ejection fraction assessment) and left atrial transversal diameter were measured in the M mode according to American Society of Echocardiography recommendations. The left atrial longitudinal diameter was measured in the four-chamber apical view in the late systolic period as the distance between the mitral leaflets and the most distant point of the left atrial wall between the pulmonary vein orifices. All measurements were averaged from five consecutive cardiac cycles. Also, abnormal left ventricle regional contractility, and the presence and range of mitral insufficiency were assessed. Segmental left ventricular wall movement abnormalities were classified as present or absent. Mitral insufficiency was assessed by colour Doppler mapping, and its range was graded from I to IV. Left ventricular ejection fraction was estimated using the Teichholz equation. In patients with abnormal left ventricle regional contractility, the modified Simpson rule was used. During TEE, the presence of thrombi and of spontaneous echocardiographic contrast was evaluated in the left atrial cavity or its appendage. Thrombus was diagnosed by TEE if a mass distinct from its walls was noticed within the left atrium. Spontaneous echocardiographic contrast was defined as moving smoke-like echoes within the left atrium. Diagnosis of both thrombus and spontaneous echocardiographic contrast required consensus of two online observers. In case of doubt a third independent observer examined the video recording.

Flow in the left atrial appendage was recorded by pulsed wave Doppler (6,7). The Doppler gate was located in the left atrial appendage orifice. An effort was made to make a long axis of the left atrial appendage possibly parallel to the ultrasound beam. The blood outflow and inflow velocities were based on the average of 10 measurements of either flow.

Anticoagulant treatment before cardioversion:

Patients chronically treated with anticoagulant drugs for any reason were excluded from the study.

On the basis of TEE results, patients were treated with anticoagulant according to different regimens. Patients with thrombi or spontaneous echocardiographic contrast were treated conservatively: four weeks of anticoagulation treatment was the rule. Patients not at risk for embolism (without spontaneous echocardiographic contrast or thrombi in TEE) were randomly assigned to conservative or (every second patient) to short anticoagulation treatment. The latter consisted of a two-day treatment with a low molecular weight heparin (nadroparine) before cardioversion.

The rule in the treatment of patients with thrombus in the left atrium was to postpone the cardioversion until the thrombus was resolved as assessed by TEE every four weeks.


The cardioversion was performed as follows: after short term anesthesia, a maximum of three subsequent synchronized DC discharges of 200, 300 and 300 J were delivered. If any AF remained or if AF recurred during the first 45 min after initial restoration of the sinus rhythm, then the procedure was assessed as a failure. Only the presence of sinus rhythm in the 45 min after DC cardioversion was assessed as a procedural success.

Statistical methods:

The averages of variables with normal distribution were compared using Student’s t test. If the variable had a non-normal distribution, then Mann-Whitney’s modification of Wilcoxon’s test was used. For categorical parameters the χ2 test was used. For comparison of variables between more than two groups, analysis of variance with subsequent Duncan’s multiple range test was performed. The significance level was set at P≥0.05. The calculations were performed using SPSS for Windows statistical software (SPSS Inc, USA).


Early success of cardioversion:

According to the result of the first cardioversion, all investigated patients were divided into three groups. Group Y comprised 81 patients with successful cardioversion. Group S consisted of 20 patients who spontaneously reverted to sinus rhythm while waiting for cardioversion. Group N consisted of 49 patients in whom cardioversion failed. Thus, the sinus rhythm was initially restored in 101 patients (groups Y and S).

These three groups did not differ in age or sex. The duration of AF tended to be shorter in group S, although the difference was not statistically significant. Hypertension, diabetes, heart failure and coronary artery disease were present at similar frequencies in all groups.

Left atrial diameters assessed in both axes were significantly smaller in group S than in group Y and group N, as well as in groups Y+N combined (Table 2). Interestingly, the difference in atrial diameter between the combined groups Y+S and group N did not reach statistical significance.

Initial characteristics of the patient groups: successful first cardioversion (group Y), spontaneous reversion while awaiting cardioversion (group S), failure of first cardioversion (group N) and groups Y+S combined

Left ventricular diameter, ejection fraction, left ventricular wall movement disturbances and the incidence of mitral insufficiency, spontaneous echocardiographic contrast and thrombi in the left atrium were similar in the groups. A tendency to a higher velocity of blood inflow and outflow in the left atrial appendage in group S than in the other groups was observed. However, the difference did not reach statistical significance (Table 3).

Initial characteristics of patients maintaining sinus rhythm (Year-Y) or with atrial fibrillation (Year-N) one year after attempted cardioversion. All comparisons were not significant

One-year follow-up:

One-year follow-up in accordance with the adopted procedure was possible in 95 patients. The sex distribution and age average in the whole group followed up at one year were similar to those in the whole initial study group. Sixty-four people (67.4%) maintained sinus rhythm after one year, in contrast to 31 (32.6%) in whom sinus rhythm was never restored or who experienced a return of AF after successful cardioversion (Year-N). Age, sex, hypertension, diabetes, heart failure and coronary artery disease were not different in the two groups. The AF duration was longer in the Year-N group, although the difference was not statistically significant. Neither of the parameters determined by TTE differed between the groups.

Among the parameters obtained with TEE, the tendency to lower velocities in the left atrial appendage in the Year-N group deserves more attention. However, as before, observations were not statistically significant. Spontaneous contrast and thrombi occurred at similar frequencies in both groups. The results are grouped in Table 3.


Before deciding whether to perform cardioversion of chronic AF, the physician should answer two fundamental questions. The first concerns the probability of restoring sinus rhythm. The second refers to the expected maintenance of sinus rhythm after a successful cardioversion.

The early success of cardioversion in the present study, which was defined as the return of sinus rhythm after the first cardioversion or during the preparation for it, amounted to 67%. This is slightly lower than rates reported in previous papers claiming success in 80% to 94% of patients. This difference in the earlier studies most likely resulted from applying antiarrhythmic treatment before the first cardioversion and by considering even short runs of sinus rhythm just after the electrical discharge to be an early success. In the present study, the procedure was considered to be successful only if sinus rhythm was present during the 45 min after cardioversion.

The second DC cardioversion attempt after pretreatment with amiodarone in patients who initially failed to cardiovert succeeded in another 20% of subjects. Thus, sinus rhythm never recovered in only 13% of patients.

The number of patients maintaining sinus rhythm after one year is high (67%). This is most likely connected to the intensive controlled antiarrhythmic treatment following cardioversion and with the protocol enabling the second cardioversion when the first one failed or AF relapsed in the course of the follow-up.

The present study did not find any relation between basic clinical and echocardiographic parameters and the success of the first cardioversion or the tendency of AF to recur after restoration of sinus rhythm. As for the early success of cardioversion, the results obtained are in accordance with previous reports. Furthermore, the high percentage of cardioversion success, which in certain studies reached 90%, from a practical point of view makes it pointless to identify the few patients who did not respond to cardioversion.

An ability to foresee the maintenance of sinus rhythm after successful cardioversion would be much more significant. The results obtained in the present study contradict some of the previous data and common beliefs. Of the many echocardiographic indexes assessed, left atrial diameter is mentioned most frequently as an indicator of early and late success of cardioversion. This is because of the assumption that the dilation of the left atrium reflects the extent of damage related to the extent of the underlying disease and enhances the risk of electrical dishomogeneity and re-entry. Studies done in recent years suggesting that left atrial dimensions return to the initial size following the restoration and maintenance of sinus rhythm (1) have raised doubts about the commonly accepted cause-effect relation between left atrial size and AF.

On the other hand, in their classic paper Henry et al (8) found not only that the left atrium is larger in patients with AF than in those in sinus rhythm but also that a transversal left atrial diameter exceeding 45 mm predicts the recurrence of AF within six months of follow-up. However, 30% of the study group had rheumatic mitral valve disease, which might have influenced the outcome. Even more important, in their early 70s the patients assessed by Henry et al (8) were not treatable with either IC group (especially propafenone) or group III drugs (sotalol, amiodarone).

Our results may suggest that modern antiarrhythmic therapy reduces the importance of the underlying heart disease, manifested by increased left atrial dimensions as assessed by echocardiography, as a factor in late cardioversion success.

Many studies (4,911) were unable to confirm the relation between the left atrial diameter and both early and late success of cardioversion. In most of these studies at least some of the cardioverted patients were receiving group IC and group III drugs. While Brodsky et al (12), in a group of 44 patients, proved the relation between the atrial width and the late success of cardioversion, they found the separating diameter to be as high as 60 mm. This may, again, be explained by the 30% prevalence of rheumatic heart disease among their patients. An even greater (65 mm) left atrial diameter as a predictor of the recurrence of AF was proposed by Volgman et al (13). In the present study, none of 150 patients with nonrheumatic AF, even of long duration, had a left atrial diameter exceeding 60 mm. Therefore, the observation of Brodsky et al (12) can have practical significance for patients with nonrheumatic AF only in exceptional cases.

The absence of a prognostic role for the remaining echocardiographic indexes such as left atrial diameter and left ventricular ejection fraction is in accordance with the results of the studies cited above.

Moreover, the present study endeavoured to assess the prognostic significance of the parameters obtained during transesophageal examination. This examination was used to assess the risk of the peripheral embolism before the planned cardioversion. Apart from the occurrence of thrombus and spontaneous echocardiographic contrast, the inflow and outflow blood velocities to and from the left atrial appendage were determined. No relation was found between the occurrence of thrombi or spontaneous echocardiographic contrast and the early or late success of cardioversion, although spontaneous echocardiographic contrast occurred somewhat more frequently in the group in whom cardioversion failed. Similarly, although the flow velocities in the left atrial appendage tended to be lower in patients with both early and late failure of cardioversion, this difference did not reach statistical significance. This disagrees with a report from Verhorst et al (14), who assessed the prognostic role of TEE. While this study also failed to identify echocardiographic factors related to the early success of cardioversion, the group maintaining sinus rhythm after one year had smaller initial dimensions and smaller left atrial area assessed by TEE, as well as higher velocities in the left atrial appendage and less frequent spontaneous contrast in the left atrium. Interestingly, these highly significant differences were noted despite the relatively small size of the population studied (40 patients). Further studies aimed at verifying the usefulness of TEE in estimating cardioversion success seem necessary.

Among the clinical parameters influencing early and late cardioversion success, the duration of AF is mentioned most often. As regards the early success of cardioversion, no difference in fibrillation duration was established. Patients maintaining sinus rhythm after one year had an AF duration almost two times shorter than that in patients in whom cardioversion failed. However, the difference did not reach statistical significance (P=0.25), which can be explained mostly by a large standard deviation.

Interestingly, in group S, with smaller left atria than those in either group Y or group N, the duration of AF was also shorter and blood velocities in the appendage were higher. Although the differences in the last two parameters did not reach statistical significance and have no practical significance, it shows that the echocardiographic indexes assessed in the study are basically related to the pathophysiology of AF.

The population size of 150 patients might have been too small to detect weak relations between echocardiographic indexes and AF. However, this objection concerns even more other studies that have been published so far. It appears that questions related to the prediction of the early and late success of cardioversion can ultimately be answered only by a large multicentre trial.

The dropout rate of about 30% of the original study participants during the one-year follow-up also raises doubts. However, the basic characteristics of the groups of patients examined for early and late success of cardioversion were almost identical, ruling out any nonrandom selection bias.

The present study showed that echocardiography and clinical assessment have little value in predicting either early or late success of cardioversion of chronic nonrheumatic AF. It is still unclear whether this is caused by modern antiarrhythmic medications that increase the chance of maintaining sinus rhythm in people with more advanced heart disease or by the lack of a relation between the echocardiographically assessed parameters and the risk of arrhythmia.


Spontaneous reversion of AF seems more likely with smaller left atria. Echocardiography including TEE is unlikely to identify patients in whom attempts to restore and maintain sinus rhythm will fail or succeed.


1. Sanfilippo AJ, Abascal VM, Sheehan M, et al. Atrial enlargement as a consequence of atrial fibrillation: a prospective echocardiographic study. Circulation. 1990;82:792–7. [PubMed]
2. Gurba H, Pruszczyk P, Szulc M, et al. The effect of successful cardioversion of chronic non-valvular atrial fibrillation on left spontaneous echocardiographic contrast. Eur Heart J. 1995;16(Suppl):350. (Abstract)
3. Lown B. Electrical reversion of cardiac arrhythmias. Br Heart J. 1967;29:469–89. [PMC free article] [PubMed]
4. Van Gelder IC, Crijns HJ, van Gilst WH, Verwer R, Lie K. Prediction of uneventful cardioversion and maintenance of sinus rhythm from direct-current electrical cardioversion of chronic atrial fibrillation and flutter. Am J Cardiol. 1991;68:41–6. [PubMed]
5. Van Gelder IC, Crijns HJGM, Hillege HL, Gosselink ATM, Lie KI. Limited long term efficacy of electrical cardioversion in patients with atrial fibrillation. Shock therapy revisited. Eur Heart J. 1995;16(Suppl):194. [PubMed]
6. Garcia-Fernandez MA, Torrecilla EG, San Roman D, et al. Left atrial appendage Doppler flow patterns: implications on thrombus formation. Am Heart J. 1992;124:955–61. [PubMed]
7. 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. 1991;4:435–41. [PubMed]
8. Henry WL, Morganroth J, Pearlman AS, et al. Relation between echocardiographically determined left atrial size and atrial fibrillation. Circulation. 1976;53:273–9. [PubMed]
9. Dethy M, Chassat C, Roy D, Mercier LA. Doppler echocardiographic predictors of recurrence of atrial fibrillation after cardioversion. Am J Cardiol. 1988;62:723–6. [PubMed]
10. Dittrich HC, Erickson JS, Schneiderman T, Blacky AR, Savides T, Nicod PH. Echocardiographic and clinical predictors for outcome of elective cardioversion of atrial fibrillation. Am J Cardiol. 1989;63:193–7. [PubMed]
11. Perez Y, Duval AM, Carville C, et al. Facteurs echographiques predictifs du maintien du rythme sinusale un an apres cardioversion pour trouble du rythme auriculaire non valvulaire. Arch Mal Coeur Vaiss. 1997;90:911–8. [PubMed]
12. Brodsky MA, Allen BJ, Capparelli EV, Luckett CR, Morton R, Henry WL. Factors determining maintenance of sinus rhythm after cardioversion of chronic atrial fibrillation with left atrial dilatation. Am J Cardiol. 1989;63:1065–8. [PubMed]
13. Volgman AS, Soble JS, Neumann A, et al. Effect of left atrial size on recurrence of atrial fibrillation after electrical cardioversion: atrial dimension versus volume. Am J Card Imaging. 1996;10:261–5. [PubMed]
14. Verhorst PMJ, Kamp O, Welling RC, Van Eenige MJ, Visser CA. Transesophageal echocardiographic predictors for maintenance of sinus rhythm after electrical cardioversion of atrial fibrillation. Am J Cardiol. 1997;79:1355–9. [PubMed]

Articles from Experimental & Clinical Cardiology are provided here courtesy of Pulsus Group