Author + information
- Received June 10, 1998
- Revision received August 24, 1998
- Accepted October 2, 1998
- Published online February 1, 1999.
- Paul B Sparks, MBBS∗,†,1,
- Shenthar Jayaprakash, MD∗,
- Harry G Mond, MD, FACC∗,
- Jitendra K Vohra, MD, FACC∗,
- Leeanne E Grigg, MBBS∗ and
- Jonathan M Kalman, MBBS, PhD, FACC∗,†,* ()
- ↵*Reprint requests and correspondence: Dr. Jonathan M. Kalman, The Royal Melbourne Hospital Department of Cardiology, Parkville, Victoria, 3050 Australia
This study examined the effect of brief duration atrial fibrillation on left atrial and left atrial appendage mechanical function in humans with structural heart disease.
Left atrial dysfunction and the development of spontaneous echo contrast (SEC) may follow the cardioversion of atrial fibrillation (AF) to sinus rhythm. This phenomenon has been termed “stunning” and is implicated in the development of atrial thrombus and embolic stroke. The effects of brief duration AF on left atrial mechanical function in humans are unknown.
Twenty-four patients (23 men, aged 59.1 ± 12.7 years) with significant structural heart disease (ejection fraction 31.2 ± 9.0%, left atrial diameter 4.9 ± 0.4 cm) undergoing implantation of a ventricular cardiodefibrillator underwent transesophageal echocardiography to evaluate left atrial appendage emptying velocities (LAAeV) and SEC before, during and after a 15-min period of AF induced by rapid right atrial pacing. Atrial fibrillation was then permitted to terminate spontaneously within 5 min or was reverted with an endocardial direct current shock. Velocities and SEC were assessed in sinus rhythm pre-AF, during AF and immediately, 5 and 10 min after reversion to sinus rhythm.
Atrial fibrillation terminated spontaneously in 10 patients after 16.1 ± 1.0 min. Endocardial direct current (DC) cardioversion of 10.4 ± 6.4 J was required in 14 patients after AF lasting 20 min. Mean LAAeV pre-AF (50.0 ± 17.5 cm/s) was not significantly different to LAAeV immediately (52.8 ± 16.7 cm/s), 5 min (54.3 ± 16.4 cm/s) or 10 min (53.7 ± 15.7 cm/s) after reversion to sinus rhythm. Atrial stunning defined as a reduction in LAAeV of >20% was not observed in any patient. Fourteen of 24 patients (58%) developed SEC during AF, which resolved within 30 s of AF termination. There were no significant differences between LAAeV in those patients reverting with DC shock (pre-AF 50.6 ± 16.2 cm/s vs. immediately post-AF 54.7 ± 16.6 cm/s) or in those patients with spontaneous reversion (pre-AF 48.9 ± 20.2 cm/s vs. immediately post-AF 49.8 ± 17.3 cm/s).
Significant left atrial stunning was not observed after brief duration AF in humans with structural heart disease. Transient left atrial SEC develops in a significant proportion of these patients during AF but resolves rapidly on reversion to sinus rhythm. These findings suggest that the risk of thromboembolism may be low after brief duration AF that terminates either spontaneously or with an endocardial DC shock even in patients with significant structural heart disease. These findings have important implications for recipients of implantable devices that are capable of atrial defibrillation in response to AF.
The cardioversion of atrial fibrillation (AF) to sinus rhythm is associated with mechanical dysfunction of the left atrium (LA) and the development of spontaneous echo contrast (SEC) (1–3). This phenomenon has been termed “stunning” and is implicated in the development of embolic stroke after cardioversion of AF. The mechanism of atrial stunning is incompletely understood, but some evidence suggests that it represents a form of tachycardia-mediated atrial mechanical remodeling (4,5).
Recent evidence clearly demonstrates that electrical remodeling of the human atrium may occur with AF duration as short as 7 to 8 min and be sufficient to promote subsequent episodes of AF (6–8). The underlying mechanisms responsible for both electrical and mechanical remodeling of the atrium may be similar. Whether atrial mechanical remodeling or “stunning” also develops after such brief durations of AF is unknown.
We prospectively determined the effects of brief duration AF on LA and left atrial appendage (LAA) mechanical function in patients with structural heart disease using transesophageal echocardiography (TEE).
Twenty-four patients in sinus rhythm undergoing nonthoracotomy implantation of a ventricular implantable cardioverter defibrillator (ICD) for accepted indications were recruited for this research investigation. All patients were either survivors of resuscitated sudden cardiac death or had rapid ventricular tachycardia or ventricular fibrillation induced at electrophysiologic study. Demographic details are described in Table 1. All patients underwent transthoracic echocardiography to determine LA size and gated single-photon emission computed tomographic scanning to assess left ventricular function prior to ICD implantation. All patients had significant structural heart disease with left ventricular dysfunction; the mean left ventricular ejection fraction was 31.2 ± 9.0%. Left atrial enlargement was present in all patients (4.9 ± 0.4 cm, range 4.2 cm to 5.5 cm), and 9/24 patients (33%) had a clinical history of paroxysmal AF. No patient had a history of stroke or transient ischemic attacks related to chronic or paroxysmal AF. All patients were receiving chronic antiarrhythmic therapy, which was not ceased prior to ICD implantation. No patient received therapy with specific calcium antagonists. All patients gave written informed consent to the study, which was approved by the Board of Medical Research of The Royal Melbourne Hospital.
Implantable cardioverter defibrillator implantation procedure
A standardized general anesthetic was administered to all patients. This comprised induction with an intravenous propofol (Diprivan) infusion, followed by anesthesia maintained with the volatile agent isoflurane and the muscle relaxant atracurium. All devices were implanted in the left prepectoral position with the ventricular endocardial leads placed at the right ventricular apex.
Induction of atrial fibrillation
After ICD implantation, AF was induced by rapid right atrial pacing at an output of 5 mA at a drive cycle length of 150–200 ms employing one of two methods: i) pacing from a 6-F quadripolar Josephson temporary electrode introduced via the right femoral vein, placed in the right atrial appendage and connected to a Medtronic 5325 programmable stimulator (n = 19); or ii) pacing from the permanent atrial lead of the ICD if a dual chamber device with atrial pacing capability was selected for implantation (n = 5). After induction of AF, the surface electrocardiogram was monitored continuously to ensure that AF persisted. If AF reverted spontaneously prior to a total of 15 min duration, AF was immediately reinduced by rapid right atrial pacing.
After 15 min of AF, the arrhythmia was permitted to terminate spontaneously within 5 min. If AF did not terminate spontaneously during this 5-min interval, an endocardial direct current (DC) shock was delivered from the ICD to revert AF to sinus rhythm after a total of 20 min. Patients in whom AF could not be induced and maintained for more than a few seconds after 20 rapid atrial pacing attempts were excluded from further study.
Implantable cardioverter defibrillator shock vector, waveform and direct current shock delivery
Direct current cardioversion of AF to sinus rhythm was achieved by employing increasing endocardial shock energies. An initial attempt at 5 J was made, followed by a 10-J then 20-J shock if the prior attempt was unsuccessful. All endocardial shocks were delivered between the right ventricular cathode and the ICD generator anode and synchronized to the R wave of the intrinsic QRS complex. Shock energy, synchronization and delivery were controlled via telemetry of the ICD system. No patients required transthoracic DC shock delivery or additional antiarrhythmic agents for cardioversion of AF.
Left atrial and LAA function were assessed at six defined time points:
1. In sinus rhythm 1 min before induction of AF.
2. After induction of AF at 1 min.
3. After induction of AF at 10 min.
4. Immediately after reversion of AF to sinus rhythm.
5. In sinus rhythm 5 min after reversion.
6. In sinus rhythm 10 min after reversion.
All recordings were performed during apnea to minimize potential velocity variations caused by mechanical ventilation. Transesophageal echocardiography was performed using a 5-MHz phased array multiplane probe attached to a Hewlett-Packard Sonos 2500 ultrasound system. Images were recorded on half-inch super VHS tape and analyzed off-line using Hewlett-Packard software. All velocity measurements were performed off-line and averaged over 20 consecutive cardiac cycles.
Left atrial appendage emptying velocities were assessed using pulse wave Doppler placing the sample volume 1 cm into the mouth of the LAA. A multiplane probe was used to scan the appendage from 0° to 180° in the transverse plane to establish an angle at which maximum flow velocities could be obtained. This angle was retained for subsequent analysis. Left atrial function was assessed with TEE pulse wave Doppler interrogation of the mitral inflow A waves with the sample volume placed at the level of the mitral valve leaflet tips. Mitral inflow A waves were not assessed during AF and were evaluated at time points 1, 4, 5 and 6 only.
Left atrial appendage stunning was defined as a reduction in LAA velocities in sinus rhythm of ≥20% after AF compared with velocities in sinus rhythm before AF (1).
Spontaneous echo contrast was evaluated at the six time points as described above. This was defined as swirling patterns of echogenicity in the LA and LAA distinct from white noise artifact. Gain settings were reduced in a sequential manner to distinguish echocardiographic contrast from noise artifact and maintained for the duration of the study. Spontaneous echo contrast was graded independently by two observers based on the description of Daniel et al. (9). The development of SEC preceding, during and after AF was determined by consensus. Analysis was performed off-line, and each echocardiographer was blinded to the other’s interpretations. Left atrial appendage thrombus was defined as a mass adherent to the wall of the LAA displaying either independent motion or different echogenic density (9).
All measurements are presented as mean ± standard deviation. A repeated measures analysis of variance was used to compare continuous variables. Scheffé F test was used for multiple comparisons. In comparing patient subgroups, differences between continuous variables were assessed with the unpaired Student ttest. Differences between categoric variables were evaluated with 2 × 2 contingency tables with the chi-square test. Statistical significance was established at p < 0.05. The study was prospectively designed with 90% power to detect a mean fall in LAA velocity of 14 ± 13 cm/s after spontaneous or electrical cardioversion from AF (1). At least 10 patients in each group were required to detect this difference (10).
Atrial fibrillation induction and termination
Five patients who consented to the study were excluded from analysis, as AF could not be induced despite 20 attempts at rapid right atrial pacing. A dual chamber ICD was implanted in 5/24 patients (21%) and was used to induce AF via the permanent atrial electrode positioned in the right atrial appendage. In the remaining 19 patients (79%), AF was induced via a temporary pacing electrode positioned in the right atrial appendage. The mean number of AF inductions for the patient population was 3.3 ± 3.5 (range 1 to 15). The mean ventricular rate during AF was 87 ± 10/min at 1 min and 91 ± 11/min at 10 min.
Spontaneous termination of AF occurred in 10 patients (42%). The mean duration of AF prior to termination in this group was 16.1 ± 1.0 min. Endocardial shocks were delivered in the remaining 14 patients in whom spontaneous reversion did not occur by 20 min. The mean duration of AF in this group was 20.9 ± 1.0 min, which was significantly longer than in the group of those in whom AF terminated spontaneously (p < 0.001). The mean endocardial DC shock energy delivered for reversion was 10.4 ± 6.4 J.
Left atrial appendage function
Mean LAA velocity prior to AF induction was 50.0 ± 17.4 cm/s. During AF, the mean LAA velocity decreased significantly to 37.9 ± 10.7 cm/s at 1 min (p < 0.01) and to 38.1 ± 10.9 cm/s at 10 min (p < 0.01) compared with sinus rhythm prior to induction of AF. There were no significant differences between mean LAA velocities immediately (52.8 ± 16.7 cm/s), 5 min (54.3 ± 16.4 cm/s) and 10 min after reversion from AF (53.7 ± 15.7 cm/s) compared with the mean LAA velocity prior to AF induction (Fig. 1).
Mode of reversion
There were no significant differences between LAA velocities before and immediately, 5 min and 10 min after AF in those patients reverting specifically with endocardial DC shocks. Similarly, there were no significant differences between LAA velocities before and immediately, 5 min and 10 min after cardioversion from AF in those patients reverting spontaneously (Table 2). There were no significant differences between the electrical reversion versus spontaneous reversion groups in terms of ejection fraction (32.8 ± 9.3% vs. 28.3 ± 8.3%), LA dimensions (4.9 ± 0.4 cm vs. 5.0 ± 0.4 cm), age (59.8 ± 15.5 years vs. 58.1 ± 7.1 years) or antiarrhythmic therapy (43% sotalol vs. 40% sotalol).
There were no significant differences between LAA velocities before and immediately, 5 min and 10 min after reversion from AF in those patients taking chronic amiodarone therapy. Similarly, there were no significant differences between LAA velocities before and immediately, 5 min and 10 min after reversion from AF in those patients taking chronic sotalol therapy (Table 3, Fig. 2). ⇓⇓There were no significant differences between the amiodarone versus sotalol groups in terms of ejection fraction (31.1 ± 8.9% vs. 31.3 ± 9.8%), LA size (5.0 ± 0.3 cm vs. 4.7 ± 0.5 cm), age (60.8 ± 6.7 years vs. 56.9 ± 18.0 years) and endocardial shock reversion (57% vs. 70%).
Left atrial appendage stunning
Immediately after reversion of AF to sinus rhythm, no patients demonstrated evidence of LAA stunning (defined as a decrease in LAA velocity of ≥20% ). However, three of the 24 patients demonstrated LAA velocity decreases in the range of 15% to 20%. The mean decrease in LAA velocity in these three patients was 16.3 ± 1.8%, which represented a mean absolute reduction of 10.6 ± 1.2 cm/s. After 5 min of sustained sinus rhythm, these velocities returned to within 5% of the pre-AF value in all three patients. Left atrial SEC did not develop after reversion to sinus rhythm in these patients. Interestingly, these three patients reverted to sinus rhythm spontaneously without requiring a DC shock.
Left atrial function
For the total study population, the mean late mitral A wave velocity prior to AF induction was 38.3 ± 11.8 cm/s. This was not significantly different from that immediately (39.4 ± 10.0 cm/s), 5 min (40.0 ± 10.3 cm/s), or 10 min (39.7 ± 10.4 cm/s) after AF termination.
Mode of reversion
In those patients reverting from AF with endocardial DC shocks, there were no significant differences between late mitral A wave velocities before and immediately, 5 min and 10 min after reversion. Similarly, there were no significant differences in late mitral inflow A wave velocities before and immediately, 5 min and 10 min after AF in those patients reverting spontaneously (Table 2).
There were no significant differences between late mitral A wave velocities before and immediately, 5 min and 10 min after cardioversion from AF in those patients taking chronic amiodarone therapy. Similarly, there were no significant differences in late mitral inflow A wave velocities before and immediately, 5 min and 10 min after cardioversion from AF in those patients taking chronic sotalol therapy (Table 3).
Left atrial spontaneous echo contrast
One patient demonstrated light LA SEC before induction of AF. During AF, 14 patients (58%) demonstrated new or increased LA SEC. Light SEC was present in 11 patients, and dense SEC was present in three patients. In all patients, change in SEC grade developed within 30 s and did not change in terms of classification (i.e., light vs. dense) during AF. No patient developed LA or LAA thrombus. After reversion to sinus rhythm, LA SEC resolved in all but the single patient with baseline SEC within 30 s.
In those patients who developed new or increased LA SEC during AF, the mean LAA velocities during AF after 1 and 10 min of AF were 34.7 ± 10.5 cm/s and 34.4 ± 10.7 cm/s respectively. These values were significantly less than the mean LAA velocities after 1 min (43.7 ± 9.0 cm/s) and 10 min (45.2 ± 7.6 cm/s) of AF in those patients who did not develop LA SEC (p < 0.05). There were no significant differences in the SEC group versus the group with no SEC in terms of LA size (4.9 ± 0.4 cm vs. 5.0 ± 0.4 cm), ejection fraction (32.3 ± 9.7% vs. 28.5 ± 6.3%) or antiarrhythmic therapy (36% sotalol vs. 50% sotalol).
The mean heart rate in sinus rhythm for the total population before AF induction (62 ± 8/min) was not significantly different from that immediately after reversion to sinus rhythm (61 ± 8/min), 5 min after AF (62 ± 7/min) or 10 min after AF (63 ± 9/min).
All patients were reviewed 3 months after the research protocol and ICD implantation. Four of the 24 patients (17%) developed spontaneous AF during this 3-month follow-up period, all of whom had a history of paroxysmal AF.
The present study produced the following findings: 1) a 15 to 20 min period of AF that reverts spontaneously or with an endocardial shock is not associated with LA stunning in humans with structural heart disease; 2) LA SEC develops in many of these patients during acute AF but resolves rapidly on reversion to sinus rhythm; and 3) chronic therapy with sotalol or amiodarone is not associated with LA stunning after cardioversion from acute AF.
Background: mechanical and electrical remodeling of the atria
Left atrial stunning is considered an important mechanism responsible for embolic stroke occurring after the cardioversion of AF, and anticoagulation is recommended if AF has been present for more than 48 h to reduce the risk of stroke (2,3,11–13). Whether LA stunning occurs after cardioversion of AF that has been present for <48 h is unknown.
Although the effects of short duration AF on atrial mechanical function are unknown, recent work has shown that even brief episodes of AF may be sufficient to induce electrical remodeling of the atria (6–8). After 8 min of AF in patients without structural heart disease, Daoud et al. demonstrated that atrial effective refractory periods decrease, and AF becomes more easily inducible and sustainable (7).
Whether the mechanisms responsible for electrical remodeling in AF also contribute to mechanical remodeling is unknown. Nevertheless, the observation that “AF begets AF” as a result of electrical remodeling has created an imperative to intervene early after AF onset to prevent these effects (14,15). Indeed, the rationale for an implantable atrial defibrillator is at least in part premised on the notion that the cycle of electrical remodeling and propensity to AF may be reversed by early restoration of sinus rhythm, and preliminary evidence suggests this may be true (16). It is imperative to know whether short duration AF is associated not only with electrical but also mechanical remodeling of the atrium. Reversion to sinus rhythm early after AF onset may be associated with atrial stunning and hence a propensity to embolic stroke.
Risk factors for mechanical remodeling and atrial stunning
Structural heart disease
We hypothesized that patients undergoing ICD implantation with both LA enlargement and left ventricular dysfunction would be at high risk for atrial stunning after brief duration AF (11,17,18). Despite this patient selection no evidence of atrial stunning after AF of up to 20 min duration was demonstrated. Our results conflict with those of a preliminary study examining the effects of brief duration AF in patients without structural heart disease in which LAA velocities decreased from 70 cm/s to 62 cm/s after 15 min of AF (19). Although this result was statistically significant, it may be questioned whether the reduction is of clinical significance, as LAA velocities associated with SEC and thrombus are usually less than 20 cm/s (1–3).
In the present study, only 8/24 patients (33%) had a history of paroxysmal AF, and mean LAA velocities at baseline and during AF were greater than those usually associated with LA thrombus and embolic stroke (2). In the presence of more advanced atrial dysfunction at baseline, brief duration AF may have been associated with atrial stunning. This is an important issue that should be the focus of future research.
Although there is evidence that calcium antagonists may attenuate electrical remodeling (6,8,15), the effect of antiarrhythmic agents on atrial mechanical function has not yet been established. Transient atrial mechanical dysfunction after 10 min of AF has been demonstrated in a porcine model (20). This was attenuated by verapamil and exacerbated by BAY K8644, suggesting transsarcolemmal calcium flux to be an important determinant of atrial stunning. Sotalol may aggravate atrial stunning after cardioversion of AF in humans (21).
In the present study, all patients were taking sotalol or amiodarone, and no evidence of atrial stunning after brief duration AF was found in either group. It has been shown that neither of these drugs prevents atrial electrical remodeling (8). Although it cannot be excluded that the common actions of these agents (class III or beta-blockade) prevented atrial stunning, we believe this to be unlikely in the face of available evidence (8,21).
Mode of reversion
Controversy exists as to whether electrical energy is an important determinant of atrial stunning after cardioversion (2,18,22). Recently it has been demonstrated that endocardial and external shocks delivered in sinus rhythm do not affect LA function (23). Furthermore, significant LA stunning occurs when chronic atrial flutter is reverted with radiofrequency ablation (in the absence of DC shocks), thus suggesting that stunning results from the atrial arrhythmia and not the mode of reversion (24). In the present study, LA stunning was not observed in patients who reverted either spontaneously or with endocardial shocks.
Development of SEC during AF
In this study, LA SEC developed in 58% of patients within 30 s of the development of AF. The mean LAA velocity during sinus rhythm was 47.8 cm/s, but fell to 34.4 cm/s with AF at 10 min (Table 4). The rapid appearance of SEC is consistent with previous reports demonstrating SEC in 39% to 87% of patients with paroxysmal AF and AF lasting <3 days (2,25,26). In patients with chronic AF, SEC is associated with LA thrombus and is an independent predictor of stroke and mortality (2,27). The implications of SEC developing early after a paroxysm of AF are unknown. Nevertheless, it is a concern that 10 min of AF was sufficient to allow SEC to form in a proportion of patients. Indeed, SEC was seen as a result of AF onset and not as a result of atrial mechanical remodeling, which is a consequence of chronic AF (4,13). Whether these acute changes in LA function are sufficient for thrombus to form without the requirement for atrial mechanical remodeling is unknown.
Resolution of SEC after cardioversion
The severity of LA stunning after cardioversion relates to the chronicity of preceding AF, and a tachycardia-mediated atrial cardiomyopathy is considered a potential cause (1,4,5,13,15). In the current study, cardioversion from brief AF was not associated with a reduction in LAA velocities but a prompt return to pre-AF velocities and rapid resolution of SEC. Presumably, AF in these circumstances is not of sufficient duration to induce the atrial mechanical remodeling observed with cardioversion of chronic AF (4,13).
The efficacy of low energy endocardial defibrillation for AF has been demonstrated by several groups, and a multicenter study examining the feasibility of an implantable atrial defibrillator is in progress (28). Data demonstrate that the majority of patients with paroxysmal forms of AF can be successfully cardioverted with energy levels of <5 J (28). Levy and Camm have emphasized the importance of specific and safe therapy if the implantable atrial defibrillator is to become an acceptable mode of therapy for paroxysmal AF (29). Whether endocardial DC cardioversion of brief duration AF by such a device is complicated by LA stunning and hence stroke has not been previously determined. The present study demonstrates that patients with an implantable atrial defibrillator as a stand-alone device or as a component of a ventricular ICD are not at risk of developing atrial mechanical dysfunction as a result of endocardial cardioversion after AF of brief duration. Thromboembolic stroke risk related to atrial stunning in these circumstances is presumably low, and anticoagulation after endocardial shock reversion of brief duration AF may not be necessary.
Programmed stimulation during electrophysiologic studies may be complicated by AF. Endocardial defibrillation through predeployed intracardiac catheters allows rapid reversion to sinus rhythm and facilitates continuation of the procedure (28). Findings from the present study suggest that cardioversion in these circumstances where AF has lasted less than 20 min would not be associated with atrial stunning.
In the present study, immediate reinduction of AF with rapid atrial pacing was performed within one R-R interval if spontaneous reversion to sinus rhythm occurred. Such paroxysms of AF may not necessarily lead to the same electrical and mechanical remodeling as produced by continuous AF of the same duration. However, a very similar stimulation protocol to that used in other studies that have demonstrated electrical remodeling of the atria after either rapid atrial pacing or AF induction was used (6–8), and as atrial electrical remodeling persists for 6 to 8 min after reversion to sinus rhythm, a very brief pause in AF or rapid atrial pacing would be unlikely to prevent atrial electrical remodeling. Wijffels et al. also demonstrated that significant atrial electrical remodeling developed after periods of intermittent AF and atrial pacing that were separated by brief periods of spontaneous termination to sinus rhythm (14).
This study has not defined the duration of AF after which atrial mechanical dysfunction develops and atrial stunning becomes a significant problem after cardioversion. The present study suggests that AF lasting more than 20 min but less than 48 h may be sufficient to induce atrial mechanical remodeling and be associated with stunning after reversion (2,13,25). This underscores the potential importance of an atrial defibrillator that will be capable of effecting cardioversion within a period of less than 20 min and potentially obviate the need for chronic anticoagulation to prevent embolic stroke in these circumstances. As only one 15- to 20-min episode of AF was induced, the cumulative effect of sequential episodes of AF and multiple cardioversions cannot be determined. It is conceivable that many such episodes of AF followed by cardioversion may lead to atrial stunning.
Significant LA stunning does not occur after brief duration AF in humans with significant structural heart disease. Transient LA SEC develops in a significant proportion of these patients during AF but resolves rapidly on reversion to sinus rhythm. These findings suggest that the risk of thromboembolism may be low after brief duration AF that terminates either spontaneously or with an endocardial DC shock even in patients with significant structural heart disease taking concomitant antiarrhythmic therapy. Our findings are consistent with the current American Heart Association guidelines suggesting that anticoagulation would not be required at the time of cardioversion of brief duration AF in patients with structural heart disease (11). These findings have potentially important implications for recipients of implanted devices who may receive cardioverting endocardial shocks after AF of brief duration.
↵1 Dr. Sparks is the recipient of a Postgraduate Medical Research Scholarship from The National Heart Foundation of Australia, Australian Capital Territory, Australia.
- atrial fibrillation
- direct current
- implantable cardioverter defibrillator
- left atrium/left atrial
- left atrial appendage
- spontaneous echo contrast
- transesophageal echocardiography
- Received June 10, 1998.
- Revision received August 24, 1998.
- Accepted October 2, 1998.
- American College of Cardiology
- Grimm R.A,
- Stewart W.J,
- Arheart K.L,
- Thomas J.D,
- Klein A.L
- Fatkin D,
- Kuchar D.L,
- Thorburn C.W,
- Feneley M.P
- Grimm R.A,
- Stewart W.J,
- Maloney J.D,
- et al.
- Zipes D.P
- Ausma J,
- Wijffels M,
- Thone F,
- Wouters L,
- Allessie M,
- Borgers M
- Daoud E.G,
- Knight B.P,
- Weiss R,
- et al.
- Daoud E.G,
- Bogun F,
- Goyal R,
- et al.
- Yu W.C,
- Chen S.A,
- Lee S.H,
- et al.
- Bland M
- Prystowsky E.N,
- Benson W,
- Fuster V,
- et al.
- Laupacis A,
- Albers G,
- Dunn M,
- Feinberg W
- Manning W.J,
- Silverman D.I,
- Katz S.E,
- et al.
- Wijffels M.C.E.F,
- Kirchhof C.J.H.J,
- Dorland R,
- Allessie M.A
- Goette A,
- Honeycutt C,
- Langberg J.L
- ↵Daoud EG, Knight BP, Weiss R, et al. Atrial fibrillation-induced acute atrial mechanical dysfunction in humans (abstr). Circulation 1997;96:I–453.
- Leistad E,
- Aksnes G,
- Verburg E,
- Christensen G
- Pollak A,
- Falk R.H
- Harjai K.J,
- Mobarek S.K,
- Cheirif J,
- Boulos L,
- Murgo J.P,
- Abi-Samra F
- Sparks P.B,
- Kalman J.M,
- Vohra J,
- et al.
- Sparks P.B,
- Kalman J.M,
- Jayaprakash S,
- et al.
- Stoddard M.F,
- Dawkins P.R,
- Prince C.R,
- Ammash N.A
- Levy S,
- Ricard P,
- Lau C.-P,
- et al.
- Levy S,
- Camm A.J