| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Division of Cardiac Surgery, University of Verona, Verona
b Division of Cardiology, University of Verona, Verona
c Division of Cardiology Bolzano City Hospital, Bolzano, Italy
Accepted for publication March 21, 2008.
* Address correspondence to Dr Luciani, Division of Cardiac Surgery, University of Verona, O. C. M. Piazzale Stefani 1, Verona, 37126, Italy (Email: giovanni.luciani{at}univr.it).
Presented at the Poster Session of the Forty-fourth Annual Meeting of The Society of Thoracic Surgeons, Fort Lauderdale, FL, Jan 28–30, 2008.
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsCommentReferences |
|---|
Methods: Repairs occurred in 104 consecutive patients (51 men), aged 29 ± 23 years (range, 1 to 70 years). Seven had isolated SVD and 97 had associated lesions that required concomitant operations. Five patients had preoperative arrhythmias; 24 (23%) were in New York Heart Association (NYHA) class III to V. Single-patch repair was done in 91 patients, caval translocation (Warden) in 7, and double-patch in 6.
Results: Ten late deaths during 38 years of follow-up (mean, 15 ± 20 years). Survival was 97% ± 2% and 79% ± 7% at 10 and 30 years. Thirty-one (29%) long-term survivors experienced 47 complications, including chronic/recurrent supraventricular tachycardia in 28, heart failure in 5, permanent pacing in 8, cerebrovascular accident in 3, and unrelated cardiac reoperation in 3. At 30 years, freedom from adverse cardiac events was 47% ± 9%, from supraventricular tachycardia, 50% ± 9%; from permanent pacing, 83 ± 6%; and from cerebrovascular accident, 96% ± 2%. Follow-up age was 42 ± 23 years (range, 5 to 82 years); 74 patients (79%) were in NYHA class I, and 15 and 5 were in class II and III to IV, respectively. Baseline cardiac rhythm was sinus in 75 patients (84%), atrial fibrillation in 11 (12%), and paced in 8. Nine patients had moderate/severe pulmonary hypertension, and 8 had left ventricular dysfunction. Only older age at operation was associated with lower survival (p = 0.003), freedom from cardiac events (p = 0.001), supraventricular tachycardia (p = 0.009), and permanent pacing (p = 0.002). Repair before age 20 was associated with lower NYHA class at follow-up (p = 0.01).
Conclusions: SVD repair at an older age is associated with increased risk of late mortality, adverse cardiac events, and worse functional outcome. Repair during childhood is strongly advised.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
Because of its distinct anatomic features, among which proximity with pulmonary and systemic (generally superior vena cava) veins and with the sinus node [6], repair of SVD has been associated with greater operative and late morbidity compared with secundum ASD [6–8]. In particular, obstruction to superior caval or, less commonly, pulmonary venous drainage and sinus node dysfunction continue to be reported to date [9–12].
These observations, however, have been questioned by other historic and contemporary studies [3, 13]. Some reports have further proposed that the natural history of unrepaired SVD may also be worse than that of ASD due to the higher prevalence and earlier onset of pulmonary hypertension [14], a factor known to adversely affect outcome of atrial level septal defects regardless of repair [15]. Given the controversial information on early and late outcome after repair of SVD and the paucity of specific clinical studies with very long-term follow-up [16], we undertook a retrospective review of 38 years of experience with the surgical treatment of SVD at our institution.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
Patients
Between 1970 and 2008, 104 consecutive patients with SVD underwent repair at the University of Verona. No patients were denied repair during the study period. Patients with posterior or inferior SVD, as well as those with isolated partial anomalous pulmonary venous connection (PAPVC) to the right atrium, were excluded from the study. The demographic variables of the population are reported in Table 1.
|
|
Operative Techniques
Repair of SVD was performed through a median sternotomy. Direct bicaval cannulation was used for venous return and ascending aortic cannulation was used for systemic perfusion, with the aid of moderately (28° to 30°C rectal temperature) hypothermic cardiopulmonary bypass. Myocardial protection during intracardiac repair was achieved using antegrade infusion of cold (0° to 4°C) crystalloid cardioplegia. After 1995, this was changed to antegrade cold blood cardioplegia.
Three techniques were adopted during the study period according to the surgeon's preference (Table 2). The general institutional policy, however, favored pericardial patch/baffle repair (single patch) for isolated SVD or SVD and PAPVC to the right atrium or lower superior vena cava (SVC); pericardial patch with atriocaval plasty (double patch) for SVD and PAPVC to the upper SVC in adults; and caval translocation (Warden operation) [17] for SVD and PAPVC to the SVC in children, as reported in Table 2. Associated secundum ASD was repaired separately either by direct suture or patch apposition.
|
Follow-Up Methods
Follow-up gathered by direct physical examination or telephone interview between December 2007 and January 2008 was 95% complete. Clinical status, including New York Heart Association (NYHA) functional class, medications, and latest electrocardiographic (ECG) and echocardiographic examinations were collected and reviewed. A major cardiovascular event (MACE) was defined as any of the following complications: cardiac death, congestive heart failure, myocardial infarction, infective endocarditis, pericarditis, stroke or transient ischemic attack, sustained or paroxysmal atrial fibrillation or flutter, heart block or sinus node dysfunction (sinus bradycardia < 50 beats/min, ectopic atrial rhythm, junctional/nodal rhythm, wandering pacemaker with < 60 beats/min) requiring permanent pacemaker implantation, or any cardiac reoperation. Atrial fibrillation or flutter, whether sustained or paroxysmal, was defined as occurring after hospital discharge or 30 days after operation.
Statistical Analysis
Categoric variables were reported as absolute numbers and percentage. Continuous variables were expressed as means ± standard deviation. Comparison of continuous variables was performed using the two-tailed Student t test for unpaired data and comparison of discrete data using the Pearson 
2 test or Fisher exact test, as appropriate. Time-related events were described using the Kaplan-Meier estimate. Survival of an age- and sex-matched population in Italy was based on Istituto Nazionale di Statistica (Institute of Statistics) life tables [18] and compared with observed patient survival using the log-rank test.
Study end points were survival, freedom from cardiac death, MACE, cerebrovascular accident, supraventricular tachycardia (SVT; including atrial fibrillation and flutter), permanent pacing, and reoperation. Multivariate analysis was performed using Cox proportional hazard method to identify risk factors for time related occurrence of death and morbid events. Variables entered in the analysis included age, sex, preoperative NYHA, associated ASD, associated PAPVC, associated coronary artery disease, preoperative atrial arrhythmia (SVT, sinus node), preoperative pacing, year of operation, and duration of follow-up. Significance was inferred at a value of p < 0.05.
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
Actuarial overall survival was 97% ± 2% at 10 years and 79% ± 7% at 30 years, which was significantly lower than expected for an age- and sex-matched control population (100% and 93% ± 3%, p = 0.041; Fig 2). Freedom from cardiac death was 99% ± 1% at 10 years and 89% ± 5% at 30 years. Univariate analysis showed older age at repair, but neither follow-up nor preoperative clinical status, to be associated with cardiac death (p = 0.001) and lower overall survival (p = 0.0001). The latter result was confirmed by multivariate analysis (p = 0.003; Table 3).
|
|
|
|
|
The only variable significantly associated with MACE, SVT, and permanent pacing was older age at repair (Table 3). Although operative technique was not associated with outcome end points, possibly due to difference in follow-up and sample size, no MACE was recorded in patients undergoing the Warden operation.
Functional Status
The functional status of patients is reported in Table 4. Age at follow-up of 94 long-term survivors averaged 43 ± 23 years (range, 5 to 82 years). Clinical conditions proved satisfactory: 89 patients (95%) were either in NYHA class I or II, which attests to a significant overall improvement compared with their preoperative status (77% vs 95%, p = 0.003). Nonetheless, moderate to severe cardiac failure (NYHA class III to IV) was recognized in 5 patients (6%), who required long-term medication and multiple hospital admissions. At univariate analysis, repair of SVD before age 20 was associated with significantly lower NYHA class at follow-up (p = 0.01).
|
Results of echocardiographic examinations were available in 82 late survivors (87%). These showed negligible prevalence of residual intracardiac shunts, no evidence of SVC or pulmonary venous obstruction, and a low prevalence of moderate to severe pulmonary hypertension and right and left ventricular dysfunction [19], each at about 10%.
|
Comment |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
Although one prior study documents late outcome after repair of SVD [16], most long-term clinical series report the repair of any atrial level septal defect [1–5]. This policy is based on the observation that long-term results are favorable and, by some, perceived to be substantially comparable with results documented after repair of secundum ASD [2–5]. Such an assumption, however, has been repeatedly questioned by historic as well as recent studies that have showed a higher prevalence of early and late complications [7–12] inherent with the very anatomic nature of the defect [6]. Accordingly to date, systemic and pulmonary venous obstructions as well as atrial arrhythmias continue to be reported with variable prevalence [9–12]. This circumstance is further underscored by the numerous surgical techniques proposed over time to repair SVD, suggesting not only variability in anatomy but also suboptimal clinical outcome [7, 8, 10, 11, 17, 20–23].
These considerations not withstanding, most clinical series portray results using one or more of the following surgical techniques: patch/baffle repair (single patch), patch repair with atriocaval plasty (double patch), or caval translocation (Warden operation) [7–13, 16, 17]. The same is true of two large clinical series on late outcome after repair of SVD: the one from the Mayo clinic [16] and the present one.
Survival
Thirty-year survival after operation was satisfactory (79%) and comparable with what was described in the two studies from the Mayo Clinic experience, 74% by Murphy and associates [2] dealing with secundum ASD and SVD and the more recent one by Attenhofer Jost and associates [16] exclusively on SVD. It is of interest that in the present series and in the one by Murphy and colleagues [5], mean age at repair was similar (29 vs 26 years) but younger than the age of 34 years in the series by Attenhofer Jost and associates [16]. This may explain why survival in the first two studies was worse than predicted for age- and sex-matched population, but in the latter study it was not. Among other elements that may account for this observation are longer average and maximum follow-up in the present analysis compared with the one by Attenhofer Jost and colleagues [16]: 15 vs 12 years and 38 vs 32 years, respectively.
These results are not directly comparable with series describing exclusively the outcome in children, where early and late mortality are essentially unknown, but so is long-term follow-up [8, 9, 13]. Other studies that include pediatric and adult patients are also limited by follow-up periods that rarely exceed the 10-year landmark [10–12, 20–23]. Late deaths were evenly divided between malignancy as expected based on age at follow-up (average, 43; but as old as 83 years) and cardiac failure. These data are difficult to compare with those from the Mayo Clinic, where 9 of 16 patients died of an unknown cause, and only two deaths were possibly related to SVD repair [16]. Deaths from unknown events, even late after heart operations, are generally classified as cardiac until proven otherwise. This postulation would make the prevalence of cardiac death in the series by Attenhofe Jost and colleagues [16] absolutely comparable with this study (11 of 16 vs 5 of 10). In addition, a prior report by the same institution on ASD repair in general confirms cardiovascular events as a leading cause of late death (18 of 27 [67%]) [2].
Similar to the studies cited [2, 16], older age at repair, but not year of operation nor duration of clinical follow-up, emerged as risk factor for both long-term and cardiac death. This observation suggests a modest role of surgical experience, in agreement with the very low operative risk that has been associated with repair [2, 7, 8]. Explanations for older age as a predictor of late overall and cardiac death are speculative and may include longstanding right heart volume and, sometimes, pressure overload, pulmonary vascular disease [14], decrease in left heart compliance owing to reverse ventricular remodeling and systemic hypertension, and increasing prevalence of arrhythmias caused by right heart dilation, as observed in the present and prior studies [2, 3, 5]. It is noteworthy that left heart or biventricular failure, which led to listing for heart transplantation in 3 of 5 patients who eventually died during follow-up, was more common than right heart failure caused by pulmonary vascular disease. This observation suggests that the operation may reverse or slow the progression of pulmonary hypertension yet not necessarily prevent left ventricular dysfunction, particularly in patients undergoing repair late in life.
Cardiovascular Complications
Thirty-year freedom from MACE in the present series was 47% and somewhat disappointing despite general clinical improvement, as documented subsequently. The estimate is not readily comparable with prior studies where the cumulative result is not reported [16]. Most of the morbidity was from arrhythmias (SVT, sinus node dysfunction, need for permanent pacing) persisting, recurring, or presenting anew at variable time intervals after repair. In addition, arrhythmias influenced the occurrence of cerebrovascular events and, possibly, congestive heart failure, which were the other two relevant causes of MACE in the present study. Prevalence of atrial fibrillation or flutter and of the need for pacemaker implant was slightly higher than that described by Attenhofer Jost and colleagues [16], 27% vs 19% and 9% vs 6%, possibly due to shorter follow-up observation and the incompleteness of follow-up ECG, in the latter series.
Similar to the latter work and to the study by Gatzoulis and associates [5], older age at repair was predictive of late atrial arrhythmias as well as the need for permanent pacing. Because these were the leading cause of morbidity, prevalence of MACEs were also associated with older age at operation. The mechanism by which age may influence prevalence of arrhythmias late after operation is inherent with chronic right heart volume overload and ensuing atrial dilatation [5]. These findings have led some authors to advocate the addition of a Maze procedure at the time of repair in older patients [5, 16], but the results of this strategy remain elusive. Additional factors, however, need to be considered in patients with SVD, including sinus node dysfunction with bradycardia-induced atrial arrhythmias and reentry tachycardia due to scarring [16], which are intimately related to operative technique.
Clearly, the present study failed to show any association between arrhythmias—or any other cardiovascular event—and type of repair; thus, we cannot formulate recommendations in terms of optimal technique. Reasons for this limitation include nonrandom use of three operative techniques; more recent adoption of the Warden procedure, resulting in different follow-up; and selective application of the latter in younger children (Table 2), whereas complications were mostly found in older patients. Nonetheless, it is worth mentioning that no evidence of atrial arrhythmia, including fibrillation, flutter, or sinus node dysfunction, was identified in the Warden group, but 1 patient (17%) in the double-patch group required permanent pacing for sick sinus syndrome.
Several recent studies have addressed the issue of ideal technique of SVD repair, but none have arrived at conclusive evidence [10–13]. Whereas scattered reports of superior caval, and more rarely, pulmonary venous obstruction after the Warden repair exist, the consensus is moving away from operations entailing incisions across the superior atriocaval junction, as are required for most double-patch techniques [9, 10, 12, 13]. Injury to the sinus node tissue and its arterial supply, which in 50% to 60% of cases derives from an artery coursing laterally across the atriocaval junction [24], seems to be more prevalent with the latter techniques. Consequently, several institutions [10, 12, 13], including our own, have adopted a simplified approach to SVD repair that recommends single patch/baffle for PAPVC to the right atrium or lower SVC and the Warden operation for PAPVC to the upper SVC.
Functional Status
Clinical conditions at follow-up were generally satisfactory, considering patient age, and improved when compared with preoperative findings. This is in line with several studies [2–5], including the one reporting on SVD only [16]. Attenhofer Jost and associates [16] found as many as 77% in NYHA class I and improved late after repair, which closely matches the 79% prevalence of NYHA I in the present study. More important, subjective clinical improvement persists despite the high prevalence of atrial arrhythmias, in some studies even comparable with preoperative estimates [4, 5]. Symptomatic benefit was more evident in patients repaired before age 20, as observed in some historic reports [2]. This result is at variance with the study by Attenhofer Jost and colleagues [16] showing older age at operation as predictive of clinical improvement at follow-up. Interpretation of the latter finding is counterintuitive, however, because older age proved to be associated with late mortality and arrhythmias, as observed in this study and elsewhere before [2, 5].
Echocardiographic examination was performed in 87% of late survivors and corroborated clinical findings, showing satisfactory (normal to mildly depressed) ventricular function in nearly 90% of patients and negligible sequelae of repair. A few patients, 5% to 10%, show little or no benefit late after operation, evolving towards refractory heart failure, similar to what described by others [2, 4, 16]. This observation parallels the prevalence of significant right ventricular dilatation (12%) and left ventricular dysfunction (10%) at follow-up investigations, a finding reported by some studies in otherwise asymptomatic children late after repair [3].
Study Limitations
Similarly to all previous studies, the current study was retrospective and nonrandomized, thereby lacking a control group receiving medical therapy that would have allowed definition of natural (unoperated) history of SVD. However, because repair can essentially not be denied nowadays, it is unlikely that any such study will ever be available in the future. In addition, because age proved to be such a formidable risk factor for late death and morbid events, the effects of aging per se, independent from the congenital heart disease, cannot be assessed. The same limitation also afflicts the other long-term study on SVD repair [16].
Use of three different operative techniques in a nonrandom fashion also resulted in the inability to conclude in favor of one or more techniques. This circumstance reflects both evolution of surgical treatment of SVD during the last 40 years and variability in anatomy of the lesion (site of anomalous pulmonary vein connection), as observed by others [10–13].
Conclusions
The present study allows to extend to a longer follow-up period and to current clinical practice prior clinical findings [16], namely that repair of SVD is associated with low risk and satisfactory late survival, albeit worse than for a matched control population. Although clinical improvement is generally to be expected, cardiac morbidity increases with time after operation and is primarily due to arrhythmias. Older age at repair represents a distinct risk factor for death and late complications, suggesting the indication to surgical repair should be in childhood, and in any case, not deferred after diagnosis.
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
Related Article
This article has been cited by other articles:
|
|
 |
|
  H. Uemura Invited Commentary Ann. Thorac. Surg., July 1, 2008; 86(1): 160 - 160. [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 |
