|
|
||||||||
a Division of Cardiology, Department of Pediatrics, Children's Hospital of Philadelphia, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania
b Division of Cardiothoracic Surgery, Department of Surgery, Children's Hospital of Philadelphia, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania
Accepted for publication February 19, 2008.
* Address correspondence to Dr Tanel, Children's Hospital of Philadelphia, Division of Cardiology, 2nd Floor Main, 34th St and Civic Center Blvd, Philadelphia, PA 19104 (Email: tanel{at}email.chop.edu).
| Pediatric cardiac surgery:
The Annals of Thoracic Surgery CME Program is located online at http://cme.ctsnetjournals.org. To take the CME activity related to this article, you must have either an STS member or an individual non-member subscription to the journal.
|
| Abstract |
|---|
|
|
|---|
Methods: A cross-sectional analysis of Ross procedure survivors (January 1, 1995 to December 31, 2005) followed at our institution was performed, including resting and 24-hour ambulatory electrocardiography (Holter monitoring). Rhythm and conduction disturbances were described, and predictors of arrhythmia were identified.
Results: Of 64 eligible patients, 47 (71%) participated. Median age at surgery was 8.7 years (age range, 34 days to 34 years). Twenty-five patients (53%) had isolated aortic valve disease and 22 (47%) had complex left-sided heart disease. At median follow-up of 8.9 years (range, 2.6–11.1 years), 46 patients (98%) exhibited sinus rhythm. Sinus node dysfunction (SND), defined as a pause of 2 seconds or longer or bradycardia for age, was present in 7 patients (15%). Complete heart block requiring a pacemaker occurred in 2 patients (4%). Ventricular tachycardia (VT) was present in 7 patients (15%), including nonsustained VT in 5 patients on Holter monitoring, and sustained VT in 2 patients requiring defibrillator placement. In multivariate analysis, concurrent arch repair at the time of the Ross operation (p = 0.04), longer cross-clamp time at the time of Ross operation (p = 0.04), and right ventricular outflow tract obstruction on follow-up echocardiogram (p = 0.03) were associated with SND. Longer cross-clamp time (p = 0.03) was also associated with VT, along with older age at surgery (p = 0.06 for trend).
Conclusions: At midterm follow-up after the Ross procedure, rhythm and conduction disturbances occur in one third of patients, including SND in 15%, atrioventricular block in 4%, and VT in 15%. Routine surveillance for late arrhythmias after the Ross procedure is warranted.
Since pioneered by Donald Ross in 1967, the Ross procedure has become an attractive option in the treatment of aortic valve disease in infants, children, and young adults [1]. The Ross procedure can be performed with low early and midterm mortality in children and adolescents with isolated aortic valve disease, and in infants and patients with complex left ventricular outflow tract obstruction [2–7]. Long-term outcomes after the Ross procedure remain uncertain [8–9].
Ventricular arrhythmias are common in patients with aortic valve disease, and aortic valve surgery, including the Ross procedure, may not completely alleviate the potential for ventricular ectopy [10–12]. In addition, the Ross procedure and other surgical interventions may introduce additional factors associated with arrhythmias, including explantation and reimplantation of the coronary arteries, the Konno operation and other procedures performed in the subaortic region, and aortic arch surgery [13–15].
A previous study [16] found that nearly 29% of patients had ventricular arrhythmias and 7% had transient complete heart block in the immediate postoperative period after the Ross procedure. The frequency and type of arrhythmias after the early postoperative period are unknown. The purpose of this study was to assess rhythm and conduction disturbances at midterm follow-up after the Ross procedure.
| Material and Methods |
|---|
|
|
|---|
Data Collection
Medical records were reviewed for age at surgery, sex, diagnosis, previous interventions, indication for the Ross procedure, postoperative complications, repeat interventions, and mortality. Patients were considered to have isolated aortic valve disease if they had aortic stenosis, aortic insufficiency, or both, with no other left-sided heart disease, and to have complex left-sided heart disease if they had multiple levels of left-sided heart obstruction [3]. Information on additional procedures performed concurrent with the Ross procedure and the duration of cardiopulmonary bypass, myocardial ischemia, and circulatory arrest was obtained from the operative record. Data on current antiarrhythmia medications and history of pacemaker or defibrillator implantation was also collected from patient medical records.
The patient's most recent electrocardiogram (ECG) was obtained and evaluated for rate, rhythm, intervals, and hypertrophy. A 24-hour ambulatory ECG (Holter monitor) was obtained for all study subjects. Recordings were evaluated for rate, rhythm, supraventricular and ventricular arrhythmias, longest pause duration, and atrioventricular (AV) block. Two pediatric electrophysiologists (J.R.K. and R.E.T.) interpreted all Holter monitor results. Any differences were reviewed and reconciled.
Supraventricular and ventricular arrhythmias were described as single premature atrial or ventricular complexes, couplets, or nonsustained (3–29 beats) or sustained (
30 beats) runs of supraventricular or ventricular tachycardia. Overall ectopy counts at Holter monitoring were reported as rare (<1%), occasional (1%–5%), or frequent (>5%). Atrioventricular block noted on the ECG or Holter monitor was classified as first-, second- (type I or II), or third-degree block. Sinus node dysfunction was defined by sinus bradycardia on the Holter monitor or by sinus pauses of 2.0 seconds or longer on the ECG or Holter monitor. The lower limit of normal for minimum heart rate on the Holter monitor was defined as 2 SD below reported mean minimum values for healthy children: minimum heart rate 60 beats/min or less for children younger than 6 years, 45 beats/min or less for children 6 to 12 years of age, and 40 beats/min or less for adolescents and young adults [17, 18]. A substantial pause on the ECG or Holter monitor was defined by a duration of 2.0 seconds or longer because the mean maximum sinus pause plus 2 SD is 1.82 seconds in healthy children and 2.02 seconds in healthy young adults [17, 19]. The patient's most recent echocardiogram was reviewed for shortening fraction, left ventricular dimensions, mitral valve regurgitation, and neoaortic and neopulmonary stenosis and insufficiency.
| Statistical Analysis |
|---|
|
|
|---|
0.10 at univariate analysis were evaluated at multivariate analysis using stepwise logistic regression. A p value of <0.05 was considered statistically significant. | Results |
|---|
|
|
|---|
30 d) deaths (total, 3.8%). One of the early deaths was arrhythmia-related and occurred in a 6-year-old boy who underwent the Ross procedure and aortic arch augmentation. Ventricular tachycardia and cardiac arrest occurred in the early postoperative period. Of the 127 survivors, 64 (50%) were followed at our institution. Complete data were available for 47 (71.2%) of the 64 patients; of the remaining 17 patients, 4 denied consent, 1 could not be contacted, and 12 were unable to complete 24-hour Holter monitoring. Patient characteristics are given in Table 1. The study population included 8 patients (17%) who were younger than 1 year at surgery, 12 (47%) who had multiple levels of left-sided heart obstruction, and the majority had either an intervention prior to the Ross operation or additional procedures performed at the time of the Ross operation.
|
Follow-Up and Reintervention
Median duration of follow-up after the Ross procedure was 8.9 years (range, 2.6–11.1 years). Median patient age at follow-up was 16.9 years (range, 2.6–44.1 years). Left ventricular outflow tract reinterventions were performed in 8 patients (17%) at a median of 3.0 years after the Ross procedure (range, 1.0–8.3 years), including mechanical aortic valve replacement in 3 patients, homograft aortic valve replacement in 2, valve-sparing root replacement in 2, and aortic valvuloplasty in 1. Twelve right ventricular outflow tract reinterventions were performed in 10 patients (21%) at a median of 3.5 years after the Ross procedure (range, 4 months to 8.3 years), including conduit replacement in 8 patients, conduit balloon dilation in 2, conduit augmentation in 1, and right pulmonary artery balloon dilation in 1.
Follow-up echocardiographic data (Table 2) were available for all 47 patients at a median of 5 months (range, 0 days to 3.4 years) from the time of Holter monitoring. In 37 patients (79%), the echocardiogram was performed within 1 year of Holter monitoring.
|
|
|
|
|
At Holter monitoring, 6 patients (13%) had first-degree AV block and 4 (8%) had Mobitz type I second-degree AV block (Table 4; Fig 1). All instances of Mobitz type I second-degree heart block occurred during sleep in patients aged 16 to 27 years. No patients had Mobitz type II second-degree AV block. One patient had underlying complete heart block.
During follow-up, 2 patients (4%) underwent pacemaker placement because of complete heart block. One patient underwent a Ross-Konno procedure and subaortic membrane resection at age 8 years. This was complicated by intermittent complete heart block, and a pacemaker was placed 8 days after surgery. Follow-up Holter monitoring demonstrated first-degree AV block with rare pacing. The other patient underwent a Ross-Konno procedure at 1 month of age and mitral valve replacement 1.6 years after the Ross procedure. Mitral valve replacement was complicated by complete heart block, and the patient underwent pacemaker implantation. Follow-up Holter monitoring showed complete heart block and a paced rhythm.
Supraventricular Ectopy
Supraventricular ectopy was not seen on follow-up ECGs. At Holter monitoring, most patients (89%) had only rare premature atrial complexes; none had frequent premature atrial complexes (Table 4). Nine patients (19%) had atrial pairs. Nonsustained supraventricular tachycardia was noted in 3 patients (6%). No patient had sustained supraventricular tachycardia.
Ventricular Ectopy
At follow-up ECG, 2 patients (4%) had isolated premature ventricular complexes. At Holter monitoring, most patients (89%) had rare premature ventricular complexes; no patient had frequent premature ventricular contractions (Table 4). Fourteen patients (30%) had ventricular couplets (median per patient, 2; range, 1–27). Five patients (11%) had nonsustained ventricular tachycardia (duration, 3-24 beats; rate, 126–136 beats/min). No patient had sustained ventricular tachycardia (Fig 1).
During follow-up, 2 patients (4%) underwent defibrillator implantation because of ventricular arrhythmia. One patient who underwent a Ross-Konno procedure, apical aortic conduit division, and mitral valvuloplasty at age 34 years was found to have ventricular tachycardia at Holter monitoring 10 months postoperatively. Due to inducible ventricular fibrillation in the electrophysiology laboratory, this patient subsequently underwent defibrillator implantation. Follow-up Holter monitoring demonstrated rare premature ventricular complexes and 1 couplet, and the patient is receiving mexiletine hydrochloride therapy. Another patient underwent the Ross procedure and aortic annulus enlargement at age 16 years. Seven years postoperatively, ventricular tachycardia and cardiac arrest ensued after a blow to the chest. The patient was successfully resuscitated, and a defibrillator was implanted. Follow-up Holter monitoring demonstrated occasional premature ventricular complexes and 27 couplets, and the patient is receiving mexiletine therapy. Neither of these patients has had significant ventricular arrhythmias detected by the device since defibrillator implantation.
Of the 5 patients with nonsustained ventricular tachycardia, 3 had Holter monitor data available before the Ross procedure. In all 3 patients, Holter monitoring demonstrated isolated premature ventricular complexes with no complex ventricular ectopy. None were taking antiarrhythmia medications. The 2 patients who did not have preoperative Holter monitor data did not have any history of ventricular ectopy. Neither was taking antiarrhythmic medications. In addition, postoperative medical records were reviewed for these 5 patients, and none had significant ventricular ectopy documented during the immediate postoperative period after the Ross operation.
At multivariate analysis, longer aortic cross-clamp time during the Ross procedure (p = 0.03) was significantly associated with ventricular tachycardia (Table 5). There was also a trend for older age at the time of surgery (p = 0.06).
Antiarrhythmia Medications
At follow-up, 4 patients (9%) were taking antiarrhythmic medications (mexiletine, 3; atenolol, 1) because of ventricular arrhythmias.
| Comment |
|---|
|
|
|---|
Aortic valve disease can predispose to ventricular arrhythmias [10–12]. The burden of pressure or volume overload leads to ventricular hypertrophy, which increases myocardial oxygen consumption and reduces coronary flow reserve. This makes the myocardium vulnerable to ischemia and scar formation, which can predispose to ventricular arrhythmias. Aortic valve surgery, including the Ross procedure, may not completely alleviate this potential. In addition, the Ross procedure and other interventions performed concurrent with the surgery may introduce factors that are proarrhythmic. Excision of the pulmonary root, aortic annulus enlargement (Konno or Nicks procedure), and muscle resection in the subaortic area all can result in scar formation, which can serve as the substrate for reentrant arrhythmias [13, 14]. The first septal perforator of the left anterior descending coronary artery can also be damaged during pulmonary root excision, leading to ischemia and ventricular arrhythmias [13, 14]. In addition, explantation and reimplantation of the coronary arteries may result in regions of ischemia and, consequently, ventricular arrhythmias.
In this study, patients who were older at the time of the Ross procedure (those with long-standing supply-demand mismatch) tended to be more likely to have ventricular arrythmias at follow-up (p = 0.06 for trend). In addition, those with a longer aortic cross-clamp time during the Ross procedure (longer duration of myocardial ischemia) were also more likely to have ventricular arrhythmia at follow-up (p = 0.03). None of the patients with significant ventricular ectopy at follow-up had documented ventricular arrhythmias in the immediate postoperative period. It is possible that although the substrate for arrhythmia may be present after surgery, a period of maturation of fibrosis or scar formation may be required, and the arrhythmia does not become manifest until later, as in other patients with repaired congenital heart disease [20, 21]. Alternatively, it may be that longer cross-clamp time is instead a surrogate marker for the overall complexity of the patient's anatomy or the repair. In our analysis, no other markers of complexity (eg, multiple levels of left-sided heart obstruction vs. isolated aortic valve disease, interventions prior to the Ross procedure, or other procedures performed concurrent with the surgery) were found to be significantly associated with ventricular ectopy at follow-up.
Other studies have not systematically investigated arrhythmia with 24-hour monitoring late after the Ross procedure. Chambers and colleagues [22] evaluated the original series of patients who underwent the Ross procedure and found that 1.5% had late ventricular arrhythmias. Daenen and coworkers [23] reported 1 late death due to ventricular arrhythmia in a series of 50 patients with a mean follow-up of approximately 3 years. This patient had significant ventricular dysfunction. Ventricular dysfunction has been associated with postoperative ventricular ectopy in adults who have undergone aortic valve replacement with a mechanical prosthesis [24]. In contrast, we did not find an association between ventricular dysfunction and ventricular arrhythmia in our analysis. Perhaps this is due to the relatively young age at surgery in our study compared with other studies [23, 24].
In our evaluation of AV block, we found that complete heart block necessitating pacemaker placement occurred in only 2 patients (4%) in the early postoperative period: after the Ross-Konno procedure and subaortic membrane resection in 1 patient and after the Ross-Konno procedure and subsequent mitral valve replacement in the other patient. We did not find any significant AV block beyond the early postoperative period. Our data are consistent with those from other studies that have reported complete heart block in zero to 6% of patients after the Ross procedure, including those undergoing a concurrent Konno procedure [6, 7, 23, 25]. Because of the relative infrequency of significant AV block in our study, we were not able to model risk factors associated with this outcome. It has been hypothesized that the Konno procedure or subaortic membrane resection may predispose to conduction abnormalities due to the proximity to the region where the central left bundle divides into anterior and posterior fascicles [13, 14]. While both of the patients who developed complete heart block in our study underwent procedures in the subaortic region, most patients who underwent either a Konno procedure or subaortic membrane resection (23/25 [92%]) did not develop significant AV block. We also found that first-degree AV block and second-degree Mobitz type I AV block occurred in 10 patients (21%). Although not currently pathologic, follow-up is warranted in these patients to assess for progression, particularly in those with concomitant bundle branch block, which was noted in 3 of the 10 patients (30%).
Although chronotropic impairment with exercise has not been reported after the Ross procedure, previous studies have shown that patients with other types of congenital heart defects, such as tetralogy of Fallot, who require great artery surgery, or patients who have undergone cardiac transplantation who require transection of the great arteries frequently have postoperative chronotropic impairment with exercise [26–28]. It has been hypothesized that this is due to sympathetic denervation, which has been shown to occur in patients after surgery involving the great arteries, such as the arterial switch operation, and in adults undergoing ascending aortic aneurysm repair [15, 29]. In our study, we found that 15% of patients had SND at 24-hour Holter monitoring. Although our definition of SND recognizes patients with substantial deviation from normal, none of these patients currently requires intervention. Further follow-up will be necessary to evaluate whether this remains true or whether progression of SND requires pacemaker implantation. At multivariate analysis, we found that SND was associated with arch surgery performed concurrent with the Ross procedure and with a longer aortic cross-clamp time. It is possible that these patients have greater sympathetic denervation after surgery.
Study Limitations
This was a single-center study involving many patients with complex left-sided heart disease, and the results may not be applicable to all patients undergoing the Ross procedure, particularly those with isolated aortic valve disease. We focused only on patients followed at our institution; however, we have previously shown that patients followed elsewhere have similar demographic and clinical characteristics, potentially minimizing selection bias. Finally, arrhythmia assessment with one 24-hour Holter monitoring may not capture all rhythm and conduction abnormalities.
Conclusions
As is recognized in other forms of congenital heart disease, routine surveillance for rhythm and conduction disturbances after the Ross procedure is warranted. Patients who may be particularly at risk for arrhythmia after the Ross procedure include those with long-standing pressure or volume overload, those with complex anatomy, those undergoing additional procedures concurrent with the Ross operation, and those with residual hemodynamic abnormalities.
| References |
|---|
|
|
|---|
Related Article
This article has been cited by other articles:
![]() |
H. Uemura Invited Commentary Ann. Thorac. Surg., June 1, 2008; 85(6): 2078 - 2078. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |