|
|
||||||||
Ann Thorac Surg 2002;73:1778-1785
© 2002 The Society of Thoracic Surgeons
a Department of Cardiac Surgery, Childrens Hospital, and Department of Surgery, Harvard Medical School, Boston, Massachusetts, USA
Accepted for publication February 7, 2002.
* Address reprint requests to Dr Jonas, Department of Cardiac Surgery, Childrens Hospital, 300 Longwood Ave, Boston, MA 02115, USA
e-mail: richard.jonas{at}tch.harvard.edu
| Abstract |
|---|
|
|
|---|
Methods. Retrospective review was made of 54 consecutive patients with APVS between 1960 and 1998. Median age and weight were 4 months and 4.8 kg. RD was present in 23 patients (10 neonates, 16 required ventilation). Fifteen patients had repair with homograft replacement of the PAs and VSD closure (group 1). Twenty-seven patients had transannular patch with VSD closure with PA-plasty (group 2, n = 21) or without PA plasty (group 3, n = 6). Twelve had miscellaneous procedures (group 4); in 6 the VSD was left open.
Results. Operative, 1-, 5-, and 10-year survivals were 83%, 80%, 78%, and 78%, respectively. Risk factors for operative mortality in multivariate analysis were RD (p = 0.04), neonates (p = 0.02), weight less than 3 kg (p = 0.02), open VSD (p = 0.02) and surgery before 1990 (p = 0.04). Since 1990 operative mortality has decreased to 11% (p = 0.04). RD was the only time-related predictor of survival in multivariate analysis (p = 0.004). In patients with RD, survival with homograft was 73% versus 41% with other techniques (p = 0.2). Mean follow-up was 72 ± 50 months. There were no significant differences in freedom from reintervention rates among the surgical groups (p = 0.08).
Conclusions. Aggressive homograft replacement of the central pulmonary arteries has been associated with improved survival in patients with APVS especially in neonates with severe RD.
| Introduction |
|---|
|
|
|---|
The main element responsible for early morbidity in these patients is aneurysmal dilatation of the pulmonary arteries that cause respiratory symptoms by compression of one or both mainstem bronchi. Secondary infection occurs frequently. In very rare cases branching anomalies of the intraparenchymal pulmonary arteries may account for the pathophysiology. Some patients present shortly after birth with severe respiratory problems necessitating intubation and mechanical ventilation. Urgent surgical treatment of this life-threatening condition is indicated as medical therapy is often not successful [1]. Operative mortality in this subgroup of patients is high having been reported to be between 30% and 50% [36]. Patients with less severe symptoms often present later in life and have a low risk of mortality with elective surgery.
In the past various palliative and corrective procedures have been used to decrease the size of the dilated pulmonary arteries including aneurysmorrhaphy or the insertion of a valved conduit. Since 1990 we have used an aggressive approach for neonates with respiratory distress by totally replacing the central pulmonary arteries with a valved homograft conduit. This retrospective study compares this aggressive approach with the historical treatment.
| Material and methods |
|---|
|
|
|---|
Five patients died prior to any surgical treatment. Fifty-four patients underwent surgery; 5 of them had their primary surgery in other institutions and were excluded from survival analysis but are included for risk factors and reintervention analysis. Four patients were lost to follow-up after surgery. Median age and weight at surgery (n = 54) was 4 months (range 1 day to 336 months) and 4.8 kg (range 1.7 kg to 82.4 kg) respectively. Sixteen patients were neonates and 23 were infants (aged 31 days to 1 year). Twenty-five patients were female (46%), 29 were male (54%). Patient characteristics are summarized in Table 1.
|
Surgical technique
The patients were classified according to the surgical strategy applied. In group one (n = 15) patients underwent VSD closure, total resection of the dilated central PAs, and reconstruction of the right ventricular outflow tract (RVOT) and pulmonary arteries with a homograft (Fig 1A, B, C).
Twenty-one patients had VSD closure, reconstruction of the RVOT with a transannular patch with aneurysmorraphy (group 2, n = 21) or without PA-plasty (group 3, n = 6). The remaining 12 patients had miscellaneous procedures (group 4) including 6 patients in whom the VSD was left open.
|
Follow-up and statistical methods
All data procurement was carried out after institutional approval according to guidelines established by the Committee on Clinical Investigation.
Written informed consent was obtained from the patients or their parents. Data were obtained from chart review at Childrens Hospital and written communications with the patients outside cardiologist. Four patients were lost to follow-up after the surgery. Complete two-dimensional echocardiographic data were available on 35 patients at the time of this review. Operative mortality was defined as death less than 30 days after surgery or in the same hospital stay. Reintervention includes procedures in the catheterization laboratory or reoperation.
Univariable and multivariable analyses were performed to compare patient characteristics and outcome data between the different surgical techniques and to identify risk factors associated with patient survival and reintervention. Continuous variables that were normally distributed were compared by the analysis of variance and Students t tests. Variables that were skewed or deviated from a normal-shaped distribution were evaluated by nonparametric Kruskal-Wallis for medians. Fishers exact test was utilized to evaluate differences in proportions or categorical data. Stepwise logistic regression was conducted to determine risk factors of mortality using the likelihood ratio test to assess significance [7]. Candidate variables included age at surgery (neonates versus others), sex, weight, presence of respiratory distress, ventilation at birth, open versus closed VSD, type of surgical technique, postoperative ventilation time, and era of operation. The odds ratio and 70% confidence interval (CI) were calculated to measure association for significant variables based on results of the multivariable analysis. Patient survival and freedom from reintervention rates were estimated by the Kaplan-Meier product-limited method with 70% CI derived by Greenwoods formula [8]. The log-rank test [9] was used to compare survival curves between the different surgical techniques and the respiratory distress subgroups. In addition, the Cox proportional hazards regression model [9] was used to determine significant time-related multivariable predictors of death with the hazard ratio as a measure of risk. Two-tailed p < 0.05 was considered statistically significant. Statistical analysis was performed with the SPSS software package (version 11.0, SPSS Inc, Chicago, IL).
| Results |
|---|
|
|
|---|
|
|
|
|
Postoperative ventilation
The duration of postoperative ventilation in group 1 varied from 1 to 700 days (median 22). Two patients underwent tracheostomy; 1 was ventilated for 2 years and 1 died 4.5 months after the surgery of respiratory infection. In group 2, patients were ventilated from 1 to 14 days (median 3), in group 3 from 1 to 4 days (median 1), and in group 4 from 1 to 16 days (median 2). Comparing the different treatment groups, there was a significantly higher postoperative ventilation time between group 1 and 4 compared with group 3 patients (p < 0.01).
Late results
Follow-up ranged from 1 to 284 months (mean 6.4 years ± 5.4). There were 3 late deaths in the miscellaneous group and 1 death each in groups 1 and 2. In group 1, the patient had severe air trapping in the right middle lobe. He had right pulmonary artery stenosis and was dilated and stented 3 months after the initial surgery. Right middle lobectomy and pulmonary artery pexy were contemplated as these were thought to be causing the difficulty in ventilating the patient. However he improved after tracheobronchitis was treated and was stable enough to be transferred to a hospital nearer to home. Unfortunately he died of a pulmonary infection 2 weeks after the transfer. The patient who died in group 2 had persistent respiratory problems requiring three reoperations and died of respiratory failure at 19 months of age. There were no late deaths in group 3 patients. Two patients in group 4 died of ventricular arrhythmia at 224 months and 284 months postoperatively.
The only significant independent predictor of long-term survival in the time-related survival analysis using the Cox regression model was the presence of respiratory distress (p = 0.004). The monthly risk of death was 10 times higher for patients with respiratory distress compared with patients with no respiratory distress (hazard ratio = 10.8, 70% CI = 4.0 to 29.3).
Reintervention
Six patients (46%) in group 1, 4 patients (22%) in group 2, and 5 patients (55%) in group 4 had reintervention. None of the 6 patients in group 3 required a reintervention. Freedom from reintervention rates at 1 year of follow-up are 70% in the homograft group (70% CI: 60% to 80%) and 83% in the PA+Patch group (70% CI: 76% to 90%; Fig 5)
(p = 0.12). There were no significant differences between the groups regarding the proportion having reintervention (p = 0.08, log-rank test = 6.79). The median time to reintervention was 5 months.
|
| Comment |
|---|
|
|
|---|
Over the past 3 decades multiple treatments have been used in an attempt to salvage infants with respiratory distress. We first adopted the use of homograft replacement in a patient with respiratory distress in November 1989 with good result; since then it has been our preference to replace central pulmonary arteries with valved homograft conduits. There is a trend toward better survival in the homograft group. Seven out of the 10 patients in the homograft group survived (70%) compared with 3 out of 11 patients in other groups (27%). We are aware that improved perioperative care has contributed to better survival but we also believe that homograft replacement has a pivotal role in determining survival as has been shown by other workers [4, 5, 10].
The types of surgical repair have evolved over time with many palliative surgeries attempted to relieve respiratory symptoms. In 1969 Waldhausen and colleagues [11] described the use of Glenn shunt to alleviate bronchial obstruction, followed by Osman and colleagues [12] with aneurysmorrhaphy. Bove and coworkers [13] recommended suspension of pulmonary artery to the retrosternal fascia in addition to aneurysmorrhaaphy and Litwin and colleagues [14] removed the right pulmonary artery from the mediastinum and placed it anterior to the ascending aorta using a tubular prosthetic interposition graft. None of these methods was perfect. In 1983 Opie and coworkers [15] banded the pulmonary artery and Ilbawi and associates [6] ligated the pulmonary artery and constructed a systemic-pulmonary artery shunt to provide pulmonary blood flow. These temporizing measures did not correct the physiology and subsequently made redo operation extremely difficult.
In 1981 Dunnigan and colleagues [16] stressed the importance of complete repair by closure of the VSD and insertion of a prosthetic pulmonary valve in order to reduce pulmonary artery dilation and subsequent bronchial compression. The surgical literature [3, 17] has demonstrated that patients with tetralogy of Fallot and absent pulmonary valve syndrome who have minimal symptoms have done well with transannular patching in the long term despite pulmonary regurgitation; therefore the benefits of valved conduits in all patients remain uncertain. Pinsky and coworkers [18] and Godart and associates [19] recommended that insertion of a pulmonary valve be limited to patients with significant pulmonic branch stenosis or other defects causing elevated pulmonary artery pressure. Ilbawi and coworkers [20], Stellin [21], and Danilowicz [4] strongly believed that pulmonary valve insertion should be part of the intracardiac repair in this group of patients. As early as 1971 Layton and associates [10] used valved homografts successfully in two teenagers and Snir and associates [5] used a valved homograft in 22 patients; the 2 neonates in their series died. This was followed by successful use of a homograft by Danilowicz and coworkers [4] in 1 neonate and 1 3-month-old infant in 1993. It has been our practice since 1990 to use a valved homograft conduit for all neonates with respiratory distress. The rationale for this practice is that pulmonary vascular resistance is likely to be elevated in neonates. Furthermore it is difficult to tailor massively dilated central pulmonary arteries when they are in a decompressed state. Complete or almost complete removal of the central pulmonary arteries and replacement with homograft tissue results in a predictable postoperative size of the central pulmonary arteries that will not cause recurrent bronchial obstruction.
With the use of cardiopulmonary bypass and deep hypothermia, extensive pulmonary artery aneurysmorrhaphy has become possible. Various designs of aneurysmorrhaphy have been proposed [4, 21]. One important feature of reconstruction as illustrated by Stellin and associates [21] is shortening of the main pulmonary artery by plication in order to lift the bifurcation area away from the trachea and draw the orifice of the right pulmonary artery inferiorly. Pulmonary arteriopexy has also been employed to relieve external compression, although we have not used this for primary repair.
Beyond the critical neonatal period, repair with remodelling of the aneurysmal pulmonary arteries to reduce bronchial obstruction, closure of the VSD, and reconstruction of right ventricular outflow tract with a pericardial patch without pulmonary valve implantation has produced good results in our series with no mortality since 1990. Godart and colleagues [19] used this approach in 21 patients (including 9 infants) and was successful in 19 of them.
In older patients the repair of absent pulmonary valve syndrome is as for patients with standard tetralogy of Fallot, without remodeling the dilated pulmonary arteries. This group has shown a good outcome with no reoperations at the time of this review.
Despite replacement of dilated central pulmonary arteries with a valved homograft a small group of neonates persist in having respiratory problems requiring prolonged ventilatory support. We approach this complication by performing bronchoscopy to delineate the cause of the airway obstruction. Two-dimensional echocardiography is done early as is cardiac catheterization to rule out homograft and branch pulmonary artery stenosis. Dilation and stenting of the vessels can then be done as required. We resort to tracheostomy if prolonged ventilatory support is required. We have not used endobronchial stents as a treatment option in this setting. Subramanian and associates [22] reported a successful case of an infant who had complete repair of tetralogy of Fallot with APV with a homograft who was admitted at 5 months old with respiratory distress secondary to severe air trapping. Fischer and associates [23] in their analysis of 17 patients noted that no new respiratory symptoms developed after 3 months of age and that survivors with early symptoms were free of distress after 18 months of age. Our opinion concurs with Fischer that bronchial structures grow and cartilage strengthens so that there is an increasing resistance to extrinsic compression as the child grows older.
There are some limitations in this retrospective review. We adopted the use of homograft conduits exclusively in neonates and infants with respiratory distress only since 1990. Therefore the maximum follow-up is only 7 years and we do not know the long-term outcome of the objective cardiopulmonary function in these patients. This should be done in a future study. Mulla and coworkers [24] in 1995 reported results of cardiopulmonary performance during exercise in patients with tetralogy of Fallot and absent pulmonary syndrome is similar to patients with tetralogy of Fallot repaired with a transannular patch. Interestingly they noted 2 of their patients who received a valved conduit have a higher maximal oxygen consumption.
We understand from the literature [25, 26] that patients with tetralogy of Fallot who were repaired with a transannular patch enjoy a good long-term survival. We are also aware that works from many institutions [27, 28] have revealed poorer exercise function in patients with moderate or severe pulmonary incompetence. The other major concern in the late postoperative follow-up of tetralogy of Fallot repair is sudden death, presumably related to ventricular arrhythmias. The more recent study showed that residual pulmonary incompetence was the best marker for important ventricular arrhythmia during a 24-hour electrocardiogram recording [29]. In short, patients without respiratory distress who had repair similar to that of standard tetralogy of Fallot repair may not be optimal in view of the late effect of pulmonary incompetence.
In conclusion, the treatment of patients with the syndrome of tetralogy of Fallot with absent pulmonary valve should be individualized on the basis of the patients age and clinical symptoms. For neonates who have severe respiratory distress we recommend aggressive total or near total replacement of the dilated central pulmonary artery with a valved pulmonary homograft. We must bear in mind that it may not resolve the problem in a small group of patients where the pathology of airway problem extends beyond the proximal pulmonary artery into the arterioles and related bronchioles. For those who survive the neonatal period elective repair with resection of aneurysmal pulmonary artery, reconstruction of right ventricular outflow tract with pericardial patch, and closure of the VSD may be done. Older patients may be repaired as for tetralogy of Fallot. We advise caution that resulting pulmonary incompetence may be detrimental in the long run, and advocate vigilance in closely observing these patients.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
B. Alsoufi, W. G. Williams, Z. Hua, S. Cai, T. Karamlou, C. C. Chan, J. G. Coles, G. S. Van Arsdell, and C. A. Caldarone Surgical outcomes in the treatment of patients with tetralogy of Fallot and absent pulmonary valve Eur. J. Cardiothorac. Surg., March 1, 2007; 31(3): 354 - 359. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. W. Brown, M. Ruzmetov, P. Vijay, M. D. Rodefeld, and M. W. Turrentine Surgical Treatment of Absent Pulmonary Valve Syndrome Associated With Bronchial Obstruction Ann. Thorac. Surg., December 1, 2006; 82(6): 2221 - 2226. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. M. Chen, J. S. Glickstein, R. Margossian, M. L. Mercando, W. E. Hellenbrand, R. S. Mosca, and J. M. Quaegebeur Superior outcomes for repair in infants and neonates with tetralogy of Fallot with absent pulmonary valve syndrome. J. Thorac. Cardiovasc. Surg., November 1, 2006; 132(5): 1099 - 1104. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. A. Norgaard, N. Alphonso, A. E. Newcomb, C. P. Brizard, and A. D. Cochrane Absent pulmonary valve syndrome. Surgical and clinical outcome with long-term follow-up. Eur. J. Cardiothorac. Surg., May 1, 2006; 29(5): 682 - 687. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. Nolke, A. Azakie, P. V. Anagnostopoulos, N. Alphonso, and T. R. Karl The Lecompte Maneuver for Relief of Airway Compression in Absent Pulmonary Valve Syndrome. Ann. Thorac. Surg., May 1, 2006; 81(5): 1802 - 1807. [Abstract] [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 |