Author + information
- Received April 6, 2003
- Revision received August 20, 2003
- Accepted September 8, 2003
- Published online February 4, 2004.
- ↵*Reprint requests and correspondence:
Dr. Jack C. Salerno, University of Washington School of Medicine, Children's Hospital and Regional Medical Center, 4800 Sand Point Way NE, PO Box 5371/4G-1, Seattle, Washington 98105, USA.
- Naomi J Kertesz, MD, FACC,
- Richard A Friedman, MD, FACC and
- Arnold L Fenrich Jr, MD, FACC
Objectives We assessed the clinical presentation, natural history, and treatment response of atrial ectopic tachycardia (AET) in children <3 years of age (group 1) compared with those ≥3 years of age (group 2).
Background Atrial ectopic tachycardia is a common cause of chronic supraventricular tachycardia in children and can be resistant to pharmacologic therapy. Radiofrequency ablation (RFA) can eliminate AET arising from a single focus.
Methods A retrospective review identified all children at Texas Children's Hospital diagnosed with AET from March 1991 to November 2000. Data obtained included clinical presentation, echocardiographic evaluation, response to antiarrhythmic therapy, spontaneous resolution, and outcomes of radiofrequency and surgical ablation.
Results Sixty-eight children were identified (22 children <3 years and 46 children ≥3 years of age). Control of AET with antiarrhythmic therapy was achieved in 91% of the younger children but only 37% of the older children (p < 0.001). There was a higher rate of spontaneous resolution in the younger group (78%) compared with the older group (16%) (p < 0.001). Radiofrequency ablation was performed in 35 of the older children, with ultimate success in 74%. Surgical intervention was required for six children.
Conclusions Younger children respond to antiarrhythmic therapy and have a high incidence of AET resolution, thus warranting a trial of antiarrhythmic therapy. In children ≥3 years, AET is unlikely to resolve spontaneously, and antiarrhythmic medications are frequently ineffective. Thus, RFA should be considered early in the course of treatment for these children; however, surgical intervention may be necessary.
Atrial ectopic tachycardia (AET) is a common cause of chronic supraventricular tachycardia (SVT) in children (1)and is thought to be due to rapid discharges from an automatic atrial focus distinct from the sinus node. Historically, AET has been resistant to pharmacologic therapy (2–4). More recently, limited treatment success has been reported with class Ic and class III antiarrhythmics (5–7). The natural history of AET includes possible progression to congestive heart failure (2,4), and it represents one of the few reversible etiologies of cardiomyopathy (4,8,9). Radiofrequency ablation (RFA) of SVT in children has been successful with a low risk of significant complications (10,11), and there is a high short-term success rate for AET arising from a single focus (11,12). To date, studies reporting the clinical course of AET have been limited by small numbers. The purpose of our study was to determine if the clinical course of AET, with regard to infants and older children, was similar to that of accessory-pathway-mediated SVT with a bimodal distribution characterized by the possibility of early disappearance with late recurrence (13).
A retrospective review of the pediatric cardiology database at Texas Children's Hospital identified all children who were diagnosed with AET during the RFA era from March 1991 to November 2000. Atrial ectopic tachycardia is characterized electrophysiologically such that: 1) the P wave of the first beat is identical to that of subsequent beats; 2) the cycle length of the first few beats progressively shorten; 3) atrioventricular block can occur in the presence of continuing SVT; and 4) tachycardia typically cannot be induced or terminated with standard pacing protocols (14,15). Inclusion in the study was based upon either electrocardiogram (ECG) or 24-h Holter findings that met established criteria for AET detailed above. Patients were excluded if they had ECG appearance of multifocal atrial tachycardia at presentation or more than one mechanism for tachycardia. Data obtained included patient demographics, clinical presentation, echocardiographic evaluation, response to antiarrhythmic therapy, occurrence of spontaneous resolution, and outcomes of RFA and surgical ablation. Based on evidence that younger children may have a different response to medication as well as a higher rate of resolution, three years of age was chosen to divide the patient population into two groups. Additionally, our policy has been to use three years as the cutoff for RFA. This takes into account the pediatric radiofrequency registry data and the higher risks of complications in children <15 kg (16). Because 15 kg is the 50th percentile for a three-year-old and a typical three-year-old would be of sufficient size to reduce the risks of RFA, this age cutoff has been used for many years. In our study, group 1 includes all children who presented with AET at <3 years of age, whereas group 2 includes all children who presented with AET at ≥3 years of age.
Although determination of objective measures of ventricular function were complicated by the underlying arrhythmia, a cardiomyopathy was defined as a shortening fraction <28% or an ejection fraction <40% (Simpson's biplane method). Control of AET during therapy was defined as either normal sinus rhythm, or rate control of AET with normal heart rate variability on a 24-h Holter. Resolution was defined as normal sinus rhythm on a 24-h Holter after antiarrhythmic therapy was discontinued for at least one month with no recurrence of symptoms. When discussing coexistent heart disease, the presence of a patent ductus arteriosus was not included as an anatomic abnormality if evident only on the first postnatal echocardiogram performed at less than one month of age.
Each patient was studied in the fasting state. All antiarrhythmic medications were discontinued a minimum of five half lives before the study. In the majority of cases, conventional mapping was performed using standard electrode and ablation catheters with bipolar electrograms displayed on a computerized system (Cardiolab, Prucka Engineering, Sugar Land, Texas). At the end of the study period, three-dimensional electroanatomic mapping using the CARTO system (Biosense-Webster, Diamond Bar, California) became available at our institution.
Baseline comparisons between the age groups were performed using a two-tailed Fisher exact test for categorical variables and a two sample ttest for numerical values. After baseline comparisons were performed, the variables that were found to differ between groups were then included in a multiple logistic regression analysis in order to control for their effects. Age was analyzed as both a grouping variable and in terms of years.
There were 22 patients diagnosed with AET at <3 years of age (median, 7 days; range, 1 day to 2 years). The mean follow up time was 25 months (range, 2 to 180 months). There were two patients in the younger group who were lost to follow-up. Five patients who were diagnosed with SVT in utero met electrophysiologic criteria for AET postnatally and were included with their diagnosis made at one day of age.
Common reasons for referral in the younger group included an irregular heart rate (10 patients), tachycardia (6 patients), feeding difficulties (2 patients), and for other reasons (4 patients). At presentation, echocardiograms were performed in 20 of 22 (91%). There was echocardiographic evidence of decreased ventricular function in three patients. Most of these patients with decreased function had shortening fractions ranging from 20% to 27%; however, a severe cardiomyopathy (ejection fraction by Simpson's biplane method of 18%) resulting in referral for transplant evaluation was seen in a single patient. Additionally, there were five anatomic abnormalities detected by echocardiography. The abnormalities included a perimembranous ventricular septal defect in three patients, and a secundum atrial septal defect in two patients. None of these patients with septal defects have required surgical intervention, with four of the five defects having closed spontaneously. The fifth patient is currently being treated with diuretic therapy for mild pulmonary overcirculation.
Antiarrhythmic medications were used as the initial therapy in all 22 of the younger patients, with a median of three drugs (range, 1 to 5) per patient. Drug doses were typically increased to achieve clinical control or until measured serum levels were therapeutic. Control of AET while on antiarrhythmic therapy was achieved in 20 of 22 (91%). Drugs used in the younger group and their effects are detailed in Table 1. Most patients in the younger group (17 of 22) received digoxin and/or a beta-blocker as the first line of medical therapy. Additional first line medications in the younger patients included flecainide (3 patients), amiodarone (1 patient), and a combination of digoxin and sotalol (1 patient). Responders to first line therapy included eight of the 17 who were treated with digoxin and/or a beta-blocker, as well as two of the three who were on flecainide. The patients started on sotalol or amiodarone as first line therapy did not achieve control until additional medications were added.
Drug toxicity occurred in two patients in the younger group. One child who was receiving flecainide developed a new right bundle branch block with an accelerated ventricular rhythm, and a second child had a toxic flecainide level (flecainide level 1,700 ng/dl on 120 mg/m2/day) without evidence of clinical toxicity.
Excluding four patients in the younger group who were still on drug therapy, 14 of the possible 18 (78%) patients had complete resolution. The mean duration of therapy in those patients whose AET did spontaneously resolve was 12 ± 7 months.
Only two of the 22 patients in the younger group who were diagnosed at less than three years of age underwent RFA; both were successful in the short term. Their ages at the time of RFA were six months and nine years. The six-month-old underwent RFA due to a progressive cardiomyopathy despite medical therapy with amiodarone. The ablation was successful with an early recurrence of AET. The AET focus was modified by the RFA as it was more easily controlled with medication after ablation, followed by resolution once antiarrhythmics were discontinued. The older of these two patients had been on multiple antiarrhythmic combinations and ultimately underwent a successful RFA secondary to family preference.
None of the patients in the younger group required surgical intervention.
All three of the patients who presented with myocardial dysfunction in the younger group had improved function after intervention. The two patients who presented with mild-to-moderate myocardial dysfunction both had improved function while on antiarrhythmic therapy. The single patient in the younger group with a severe cardiomyopathy was discussed above. He was tried on multiple medical regimens without success and ultimately underwent RFA at six months of age. Due to recurrence of AET after RFA, he was placed on, and responded to, amiodarone with normalization of function.
There were 46 patients diagnosed with AET at ≥3 years of age (median, 10 years; range, 3 to 18 years). The mean follow-up time was 20 months (range, 1 week to 120 months). There were three patients in the older group who were lost to follow-up.
Common reasons for referral included irregular heart rate (16 patients), tachycardia (13 patients), palpitations (11 patients), and for other reasons (6 patients). At presentation, echocardiograms were performed in 40 of 46 (87%) in the older group. There was echocardiographic evidence of decreased function in eight patients with a severe cardiomyopathy resulting in referral for transplant evaluation in two patients (ejection fractions by Simpson's biplane method of 16% and 20%). The single anatomic abnormality in the older group consisted of a quadricuspid aortic valve that was neither stenotic nor regurgitant.
Antiarrhythmic medications were used as the initial therapy in 35 of 46 (76%) of the older patients, with a median of two drugs (range, 1 to 5) per patient. Similar to the younger group, drug doses were typically increased to achieve clinical control or until measured serum levels were therapeutic. Control of AET while on antiarrhythmic therapy was achieved in only 13 of 35 (37%) of the older children. Drugs used in the older group and their effects are detailed in Table 2. As with the younger group, most patients (30 of 35) in the older group were initially started on digoxin and/or beta-blocker therapy. Additional first line medications included procainamide (1 patient), ethmozine (1 patient), flecainide (1 patient), and a combination of digoxin and quinidine (2 patients). There were 13 patients in the older group who achieved control, nine of whom received digoxin and/or beta-blocker therapy. Additional medications used to achieve control in the older group included ethmozine (1 patient), flecainide with a beta blocker (1 patient), and digoxin with flecainide (2 patients). There were no toxic side effects in the older group.
Excluding four patients who were still on therapy, only five of the possible 31 (16%) patients in the older group had complete resolution. The mean duration of therapy in those patients whose AET did spontaneously resolve was similar to the younger group at 12 ± 7 months. The children in the older group who did not have resolution of their AET and ultimately underwent RFA received an average of 33 ± 2 months of antiarrhythmic therapy before RFA.
In total, 39 patients from the older group underwent an electrophysiology study; however, in six of these patients, their AET was either suppressed secondary to sedation or quiescent preventing mapping of the focus during the study. Ultimately, RFA was performed in 35 patients in the older group, including two patients who had no AET during the initial attempt in the electrophysiology laboratory. Indications for RFA in the older group included tachycardia despite antiarrhythmic therapy (17 patients), patient choice (16 patients), and cardiomyopathy (6 patients). As depicted in Figure 1, RFA was successful in the short term in 30 of 35 (86%). There were 10 patients who had recurrence of AET after RFA. Successful repeat ablations were achieved in four of those who had recurrence and in two of the initial failures, giving ultimate success in 26 of 35 (74%) in the older group. Figure 2details the location of successful ablations, the recurrences, and the failures. The ectopic focus was left-sided in 22 of 35 patients, of which 12 were in the left atrial (LA) appendage, nine near the entry of the pulmonary veins, and one on the free wall of left atrium (LA). The remainder (13 of 35) of the foci were right-sided with six in the right atrial (RA) appendage, four on the free wall of the right atrium (RA), and three at the margin of the superior vena cava and RA appendage. As demonstrated in Figure 3, most patients (70%) who had a recurrence after ablation did so in the first three months after ablation. Six of the recurrences were located in an atrial appendage, three in the RA appendage, and three in the LA appendage. The remainder of the recurrences were located near the right upper pulmonary veins (3 patients), and one on the free wall of the LA. Of the five failures, there were two in the RA appendage, one in the LA appendage, and two at the junction of the RA and RA appendage. Although six recurrences and three failures were located in an atrial appendage, there was not statistical significance between location of the focus and outcome of ablation. In the 10 patients with recurrence after RFA (Fig. 3), additional therapy included successful repeat RFA in four patients, successful surgical appendectomy in three patients, and successful medical therapy in three patients (1 was lost to follow-up after successful medical therapy).
Overall, there were 44 RFAs performed with three complications: 1) a thrombus on the anterior leaflet of the mitral valve without sequelae after a precautionary six-week course of coumadin; 2) the RFA catheter adhered to the atrial appendage requiring manual removal of the ablation catheter, with termination of the procedure; 3) an intramural hematoma of the pulmonary vein with echocardiographic resolution and no evidence of pulmonary vein stenosis after therapy with aspirin.
Due to concerns of performing RFA in young children, there were only three patients in our study who were three years of age or younger at the time of their RFA. The youngest patient was described above, having undergone RFA at six months of age and a weight of 7 kg. The two other children, both from the older group, underwent RFA at three years of age, weighing 14 and 17 kg, respectively. One of these patients had a successful RFA; the other required surgical appendectomy after recurrence. There were no complications in these three children.
Whereas none of the patients in the younger group required surgical intervention, six patients in the older group required surgery after first undergoing an attempt at RFA; one patient for failure of the first RFA, two for failure after the second RFA, and three for recurrence after ablation. One of the patients required a second surgical intervention after an RA appendectomy, due to recurrence at the margin of the RA appendage. Therefore, a total of seven surgical interventions were performed with three cryoablations and four atrial appendectomies. Ultimately, there was success in all six patients who underwent a surgical procedure at latest follow-up. There were no complications after surgery.
All eight of the patients in the older group who presented with myocardial dysfunction had improved function after intervention. There were six patients with mild-to-moderate myocardial dysfunction; of these, three had improved function after surgical therapy, two after successful RFA, and one with antiarrhythmic therapy. One of the two patients in the older group who presented with a severe cardiomyopathy had RFA of a focus in the high RA with resolution of the AET and subsequent normalization of function. The second patient had RFA of an LA focus with early recurrence due to the emergence of a second LA focus requiring a second catheterization and RFA one week later. The second RFA resulted in a reduction, but not termination, of the rate of AET focus. Sinus rhythm was achieved with the initiation of amiodarone after ablation. Ventricular function had begun to normalize after the second RFA and subsequently normalized completely after treatment with amiodarone. While attempting to wean the amiodarone several years later, there was recurrence of AET, and, thus, the patient had a third RFA that was successful in eliminating the focus completely.
Control of AET while on antiarrhythmic therapy (Table 3) was achieved in 20 of 22 (91%) of the younger children in our study and only 13 of 35 (37%) of the older children (p < 0.001). Additionally, there was a higher frequency of spontaneous resolution of AET in the younger group with 14 of the possible 18 (78%) having complete resolution, compared with only five out of the possible 31 (16%) in the older group (p < 0.001). After controlling for the baseline effects that differed between groups (Table 4), age in years was still predictive of both control on antiarrhythmic therapy (p < 0.02) and resolution (p < 0 .04). Statistical control included variables that may only indicate inherent age effects and not differences in disease state, in which case statistical control may represent overadjustment. In spite of this, when these factors were adjusted, the effect of age persisted.
The results of our study are consistent with the findings of Mehta et al. (5), von Bernuth et al. (6), and Bauersfeld et al. (17)who also found a high incidence of resolution of AET after withdrawal of antiarrhythmic therapy, particularly if diagnosed early in life. Spontaneous resolution of AET after successful pharmacologic control has been reported to be 72% in infants <6 months of age and 30% in older children as compiled from eight contemporary studies by Saul et al. (18). Unlike the findings of Koike et al. (4), who reported resolution of AET in several older patients, we found spontaneous resolution in our older patient population to be distinctly rare. Thus, our study confirms that response to medical therapy and spontaneous resolution is age-dependent.
A standard approach to drug therapy for AET remains problematic; however, our results show that control of AET while on antiarrhythmic therapy was achieved in 20 of 22 (91%) of the younger children and only 13 of 35 (37%) of the older children (p < 0.001). Therefore, drug therapy should be considered as the primary therapeutic modality, particularly in younger children. Although there is a large variability in the choice of medications, digoxin and beta blockers remain our first-line medications followed by flecainide. Although several authors have reported success with amiodarone (6,7), we tend to reserve amiodarone as a second-line medication due to possible side effects and its long half life, which makes it difficult to discontinue and/or change to a new antiarrhythmic medication if amiodarone is ineffective. In our experience, amiodarone has had some beneficial effect as an additional agent and has primarily been used for the younger children due to their higher chance of resolution and a defined end point of therapy.
As this was a retrospective study, all children who met criteria for the diagnosis of AET were included, which resulted in enrollment in the study of several children who had nonsustained and nonincessant episodes of AET. Our study includes seven patients in the younger group and three patients in the older group who had nonsustained and nonincessant episodes of AET. All seven of the patients in the younger group responded to medication with six continuing on to resolution of AET after withdrawal of antiarrhythmic therapy. In the older group, all three patients with nonsustained AET responded to medication, with two currently on antiarrhythmic therapy and one having resolution of AET. If these patients were excluded from the study, the number of responders to medication (p < 0.05) and the likelihood of resolution (p < 0.05) would remain statistically significant between the two groups.
Due to the difference in resolution and response to antiarrhythmic therapy, one might speculate that there may be a different underlying substrate for AET in younger children versus older children. The younger children may have a focus of abnormal cells with more rapid phase 4 depolarization that either involutes or degenerates with time resulting in resolution. The older children, on the other hand, might have a focus that has become irritated due to inflammation or scarring such as that from myocarditis.
We did not find any evidence of late recurrence after spontaneous resolution in our younger patient group. This is in opposition to reentrant SVT that presents with a bimodal distribution (13)where younger patients are likely to have resolution with recurrence at a later age.
It is interesting that the occurrence of AET appears to be less common in adults than in children (3,15). It may not be as rare in adults as previously believed, as there is evidence that focal atrial fibrillation in adults results from a focus in the pulmonary veins (19)or other atrial structures and, thus, may be a similar etiology to AET in children. The atrial myocardium in children may be more stable and, therefore, less vulnerable to AET degenerating into atrial fibrillation.
In our study, the ectopic focus was more likely to be left-sided; however, both atrial appendages were a common source of the ectopic focus. Although six recurrences and three failures were located in an atrial appendage, there was no statistical significance between location of the focus and outcome of ablation, possibly due to the small sample size. A higher recurrence/failure rate in the atrial appendage may be related to its trabeculated nature, which makes catheter manipulation and adequate tissue contact difficult. Although there have been reports of AET due to multiple foci, we encountered multiple foci in only a minority of our patient population. As demonstrated by several patients in our study, AET can be easily suppressed in the catheterization laboratory by sedation, and therefore, all attempts should be made to minimize sedation but ensure patient comfort.
Due to our reluctance to perform RFA in younger children, there was a bias to make every effort to treat the younger children with antiarrhythmic agents. For the older children, there appears to be a more favorable risk-to-benefit ratio with RFA as compared with antiarrhythmic agents, which have their own inherent risks. Although a study by Blaufox et al. (20)did support the use of RFA by experienced physicians in selected infants, it is generally believed that a favorable risk-to-benefit ratio may not exist for the younger children. Additionally, concern regarding the late enlargement of radiofrequency lesions placed in the immature atrial myocardium should be kept in mind with younger children (21). We recommend that, before considering RFA, all efforts should be made to treat younger children medically at least through infancy. However, if they develop a cardiomyopathy, we would recommend an electrophysiology study with consideration of RFA or, if necessary, surgical intervention, especially if the focus is in an atrial appendage, as surgery can be performed without the need for cardiopulmonary bypass and with fairly low risk (22).
In our study, AET was not associated with any significant congenital heart defects; however, there was evidence by echocardiography of decreased myocardial function in 11 patients. All of these patients had improved myocardial function after intervention; therefore, echocardiographic assessment at the time of diagnosis is important.
In children ≥3 years of age, AET is unlikely to resolve spontaneously, and antiarrhythmics are frequently ineffective. As RFA has been used successfully and safely (10,11)in this age group, it should be considered early in the course of treatment, especially if there is an associated tachycardia-induced cardiomyopathy. For children younger than three years of age, we recommend every effort be made to establish control of AET with antiarrhythmic agents, thus allowing for spontaneous resolution of AET or, alternatively, delaying RFA until they are older. Although AET is rarely associated with congenital heart disease, echocardiography is important to assess for left ventricular dysfunction, which may result from chronic tachycardia. In our experience, AET is most often left-sided, and the atrial appendages are a common focus. Due to the large number of trabeculations, RFA in an atrial appendage may fail or recur, and surgical appendectomy/cryoablation may be required.
- atrial ectopic tachycardia
- electrocardiogram/electrocardiographic/ electrocardiography
- left atrial/atrium
- right atrial/atrium
- radiofrequency ablation
- supraventricular tachycardia
- Received April 6, 2003.
- Revision received August 20, 2003.
- Accepted September 8, 2003.
- American College of Cardiology Foundation
- Gillette P.C.
- Von Bernuth G.,
- Engelhardt W.,
- Kramer H.H.,
- et al.
- Kugler J.D.,
- Baisch S.D.,
- Cheatham J.P.,
- et al.
- Van Hare G.R.,
- Witherell C.L.,
- Lesh M.D.,
- et al.
- Walsh E.P.,
- Saul J.P.,
- Hulse E.,
- et al.
- Gillette P.C.,
- Garson A.T.
- Kugler J.D.,
- Danford D.A.,
- Houston K.,
- et al.
- Haissaguerre M.,
- Jais P.,
- Shah D.C.,
- et al.
- Blaufox A.D.,
- Felix G.L.,
- Saul J.P.
- Saul J.P.,
- Hulse J.E.,
- Papagiannis J.,
- et al.
- Ott D.A.,
- Gillette P.C.,
- Garson A.T.,
- et al.