|
|
||||||||
Ann Thorac Surg 2006;82:2200-2206
© 2006 The Society of Thoracic Surgeons
a Department of Pediatric Cardio-Thoracic Surgery, Royal Liverpool Childrens NHS Trust, Alder Hey Hospital, Liverpool
b Department of Pediatric Cardiology, Royal Liverpool Childrens NHS Trust, Alder Hey Hospital, Liverpool
c Department of Pediatric Cardiology, Royal Manchester Childrens Hospital, Manchester, United Kingdom
Accepted for publication July 10, 2006.
* Address correspondence to Dr Kalavrouziotis, Department of Pediatric Cardio-Thoracic Surgery, Alder Hey Hospital, Eaton Rd, Liverpool L12 2AP, United Kingdom (Email: gkalavrouziotis{at}yahoo.com).
Presented at the Poster Session of the Forty-second Annual Meeting of The Society of Thoracic Surgeons, Chicago, IL, Jan 30Feb 1, 2006.
| Abstract |
|---|
|
|
|---|
METHODS: From July 1993 to December 2005, 29 patients with median age 28 days (range, 11 to 127), and median body weight 3.1 kg (range, 2.6 to 5.9 kg), underwent primary repair of TAC. The anatomical type of TAC was as follows: A1-2, 27; A3, 0; and A4, 2. Right ventricular outflow tract was reconstructed with an aortic (n = 7) or pulmonary homograft (n = 8), or a bovine (n = 11) or porcine valved xenograft (n = 3). Follow-up was complete for all patients.
RESULTS: Hospital mortality was 3.4% (1 death due to respiratory infection). At a mean follow-up of 74 months (range, 2 to 149), 1 patient died suddenly 2 months after surgery (6-year actuarial survival 93%). Of the 27 midterm survivors, 14 (52%) underwent 30 interventional procedures including percutaneous balloon dilation with or without stenting for right ventricular outflow tract or branch pulmonary artery obstruction. Eight of them were reoperated on for right ventricle-to-pulmonary artery conduit replacement (n = 8, 23%), and aortic valve regurgitation (n = 1, 3.4%). The overall freedom from any reintervention at 6 years was 50%. Aortic valve regurgitation was trace in 15 patients, mild in 8, moderate in 4. All midterm survivors but 1 (26 of 27) had good ventricular function.
CONCLUSIONS: Truncus arteriosus communis repair can be performed early with very low perioperative mortality and satisfactory midterm morbidity; the latter is mainly attributed to right ventricular outflow tract reconstruction. Interventional cardiac catheterization delays inevitable conduit replacement.
| Introduction |
|---|
|
|
|---|
The purpose of this study is to report the early and midterm results of early primary repair of TAC by a single surgeon at a single institution in the current era.
| Patients and Methods |
|---|
|
|
|---|
Our study had the approval of our Institutional Ethics Committee, which waived the need for patient consent as our study is a retrospective one and no patient can be identified. The medical notes of patients were reviewed, and the following data were collected: demographic (sex, age, body weight), preoperative (clinical condition, preoperative examinations, medications), intraoperative (TAC type, surgical technique, cardiopulmonary bypass [CPB] technique, duration of CPB/cardiac arrest/circulatory arrest, method of ventricular septal defect closure, conduit for RV-PA continuity reconstruction, additional procedures during TAC repair), and postoperative (early mortality and morbidity, length of stay on mechanical ventilation/intensive care unit [ICU]/hospital, midterm follow-up mortality and morbidity).
Data are expressed as the mean ± SD, range, and median (if significantly different from the mean). "Early" was considered the results that occurred within 30 days of operation or during the same hospitalization. The Kaplan-Meier and the Cox proportional hazards regression methods were used for survival analysis and analysis of freedom from reintervention. Group comparisons were performed using Students t test and
2 test for parametric and nonparametric data, respectively. A p value less than 0.05 indicated a significant difference.
There were 11 male (38%) and 18 female infants. Median age at correction was 28 days (range, 11 to 127; mean, 37.2 ± 29.08), and median body weight was 3.1 kg (range, 2.6 to 5.9 kg; mean, 3.3 ± 0.82 kg). Five patients (17%) were aged 2 weeks or less; 13 patients (44.8%) weighed less than 3 kg.
All patients had unrestrictive pulmonary flow and lung plethora on chest roentgenogram. Twenty-two patients (76%) were in heart failure or moderate to severe respiratory distress, or both; the remaining 7 were in relatively good condition with mild tachypnea, and the diagnosis was made because of cardiac murmur. Hemoglobin oxygen saturations at presentation ranged from 80% to 98% (mean, 94%) in room air. Six patients (21%) with heart failure or severe respiratory distress required mechanical ventilatory support for 1 to 4 days (median, 2) before surgery. Preoperatively, all patients were receiving diuretics (furosemide and spironolactone), 2 were on captopril, and 2 on digitalis. The 2 patients with interrupted aortic arch (IAA) received prostaglandin E1 infusion (0.05 µg · kg1 · min1); 1 of them presented in shock with severely compromised left ventricular function (shortening fraction, 23%), and severe metabolic acidosis, and was mechanically ventilated for 3 days preoperatively. Diagnosis was made with echocardiography in all patients. Cardiac catheterization was performed early in our series to define pulmonary vascular reactivity in 2 patients who presented late with pulmonary hypertension.
According to the congenital heart surgery nomenclature [9], which modified the Van Praagh classification, the anatomical type of TAC was as follows: A12, 27 cases; A3, none; and A4, 2 cases (IAA, type A and type B in each case).
Associated anomalies are summarized in Table 1. The truncal valve had 3 leaflets in 17 cases (59%), 4 leaflets in 9 (31%), and 2 leaflets in 3 (10%). They were thickened and dysplastic in 7 cases (24%). The function of the truncal valve was impaired in 18 patients (62%); regurgitation was present in 11 cases, stenosis in 4, and both lesions in 4.
|
Surgical Procedures
Two techniques of CPB were employed: single venous return (right atrial cannulation) and circulatory arrest with systemic cooling to 15°C to 18°C (n = 22, 76%); or bicaval cannulation and full-flow CPB with systemic cooling to 22°C (n = 7). The PAs were snared at the onset of CPB. Cold blood cardioplegia (4°C) was delivered through the root of the aortic truncal branch every 20 minutes; ice slush was used for topical cooling. The PAs were detached from the truncus. The resultant truncal defect was closed with a bovine pericardial patch (n = 16), or primarily (n = 8), if there were no concerns of potential distortion of the truncal valve or compromise of the coronary arteries; in 5 cases, the truncus was completely transected to facilitate detachment of the PAs, and the aorta was reconstructed by direct end-to-end anastomosis, without any augmentation.
The truncal valve was inspected during surgery in cases with more than mild regurgitation or dysplastic features on preoperative echocardiogram, and in all cases its function was considered not too compromised; therefore, no additional valve procedure was performed.
The ventricular septal defect was closed through a right infundibular ventriculotomy with a bovine pericardial patch in 20 cases (69%), or a synthetic patch (Dacron; C. R. Bard, Haverhill, Pennsylvania) in 9 cases (31%). A running suture technique (n = 22) or interrupted pledgeted suture technique (n = 7) with polypropylene monofilament suture material was utilized.
The right ventricular outflow tract (RVOT) was reconstructed with an aortic (n = 7; diameter range, 11-13 mm) or pulmonary cryopreserved homograft (n = 8, 11-14 mm) in 15 cases; a bovine jugular valve conduit (Contegra, unsupported type; Medtronic, Inc, Minneapolis, MN) in 11 cases (12 mm); and a porcine pulmonic valve conduit (Shelhigh; Shelhigh, Inc, Union, NJ) in 3 (10 to 12 mm). The RVOT-to-homograft anastomosis was augmented with a "hood" of bovine pericardial patch in 11 cases; in 4 cases, the mitral leaflet extension of the aortic homograft was used.
Additional procedures during TAC repair were mainly the reconstruction of IAA in 2 cases. The arterial cannulation was achieved through a 3-mm polytetrafluorethylene tubular graft (Gore-Tex; W. L. Gore & Assoc, Flagstaff, Arizona) anastomosed end to side to the innominate artery, which accommodated the arterial cannula. Therefore, continuous "low flow" cerebral perfusion was maintained during the arch reconstruction. The arterial duct was divided, and all the remnants of ductal tissue in the descending aortic wall were excised. The aortic arch was reconstructed with side-to-end anastomosis to the descending aorta; the posterior wall of the arch was anastomosed directly to the posterior wall of the descending aorta, whereas for the anterior wall anastomosis, an appropriately trimmed patch of bovine jugular venous conduit (Contegra) was incorporated to enlarge the anastomosis. Hence, the posterior wall of the aortic archtodescending aorta anastomosis was constructed of native tissue with growth potential. The xenograft patch was utilized subsequently to enlarge the aortic arch alongside its lesser curvature down to the ascending aorta, where it augmented the site of truncal defect after detaching the PAs. In the case of IAA type B, the left subclavian artery originating from the descending aorta was divided to facilitate the mobilization of the descending aorta and anastomosis to the arch with the least possible tension.
Mean CPB time was 195 ± 65 minutes (range, 128 to 358); the patient with the longest perfusion time had a 2-hour support with CPB before being able to be weaned; mean aortic cross-clamp time was 61 ± 15 minutes (range, 41 to 79); and mean circulatory arrest time was 42 ± 16 minutes (range, 11 to 64).
The sternum was electively left open after TAC repair in all but 2 cases, even when hemodynamic conditions were optimal after weaning off CPB. Delayed sternal closure was performed in the ICU 2 to 7 days (median, 2.5) later. In all patients, antiplatelet therapy (aspirin, 3 to 5 mg/kg daily) was initiated as soon as they tolerated feeding in ICU, and was continued after hospital discharge.
Follow-Up
Patients were followed up at Royal Liverpool Childrens Hospital or Manchester Childrens Hospital, at regular intervals with clinical examination, New York Heart Association class assessment, echocardiogram, and if required, cardiac catheterization. Indications for a cardiac catheterization were clinical symptoms or echocardiographic findings: evidence of RVOT or branch PAs stenosis or regurgitation, or impaired RV function; when needed, cardiac catheterization was combined with an interventional procedure as balloon dilation or stenting. Criteria for interventional cardiac catheterization were stenosis across the RVOT-conduit (pressure gradient > 50 mm Hg, RV pressure
60% of systemic); or stenosis/distortion of the branch PAs resulting in a significant reduction of blood flow to the lung.
Indications for reoperation were severe stenosis or regurgitation, or both, of the aortic/truncal valve; severe regurgitation across the conduit resulting in impaired RV function; and failure of the interventional cardiac catheterization to offer adequate relief of RVOT-PAs stenosis.
Latest follow-up information was obtained from January to February 2006 by patient or attending physician contact and was achieved for all patients.
| Results |
|---|
|
|
|---|
Morbidity
Early reoperations, aside from delayed sternal closure, were performed in 6 patients (21%), for reexploration and control of mediastinal bleeding (n = 5) and cardiac tamponade (n = 1). Of these patients, 3 had prolonged clotting times postoperatively (2 with preoperatively diagnosed von Willebrands disease), and 2 had thrombocytopenia (Table 2). Two patients had cardiac arrest (1 due to cardiac tamponade) requiring open heart massage. Six received inhaled nitric oxide therapy because of having severe pulmonary hypertension. Three patients had neurologic problems (Table 2), but no focal lesions were found in the subsequent brain magnetic resonance imaging scan; the 2 patients with tonicoclonic seizures were discharged on antiepileptic medications. No signs of mediastinitis or sternal wound infection were noted in our series.
|
For the whole series (including the hospital death), the duration of postoperative mechanical ventilation was 7 ± 6.1 days (median, 5.5; range, 1.8 to 34); postoperative ICU stay was 8.3 ± 6.0 days (median, 5.6; range, 3 to 34; 6 patients were preoperatively admitted to ICU for severe heart failure or respiratory distress); and postoperative hospital stay was 17 ± 8.4 days (median, 14; range, 7 to 40).
The 2 patients with IAA (aged 15 and 23 days; body weight 2.4 and 4.7 kg; IAA type B and A, respectively) survived the operation, despite the poor clinical status at presentation of the latter (circulatory collapse, poor LV function). They were mechanically ventilated for 3 and 8 days, respectively, after surgery; they remained in ICU for 4 and 10 days and in the hospital for 9 and 40 days, respectively. They were supported with three inotropes (milrinone, dobutamine, adrenaline) for 4 and 6 days, respectively. The only major complication encountered was reexploration for mediastinal bleeding in the latter patient.
Midterm Results
Follow-up was available for all discharged patients (28, 100%) at a mean time of 74 ± 47 months (range, 2 months to 12.4 years).
Mortality
During the follow-up period, 1 patient (male, 13 days old at operation; body weight, 2.6 kg; TAC type A1; no associated anomalies) died suddenly at home 2 months after discharge from hospital. Autopsy revealed no specific cardiac findings correlated with death. The Kaplan-Meier estimate of survival (including operative mortality) was 93.1% ± 3.6% at 6 years. No risk factor for midterm death could be identified.
Reinterventions
Fourteen patients (51.8% of 27 midterm survivors) underwent a total of 30 interventional cardiac catheterizations: 2 patients had 2, 4 had 3, and 2 had 4. The interventional cardiac catheterizations included percutaneous balloon dilation (n = 30) of the RVOT (n = 10) or PAs (n = 22); and stenting (n = 5) of the RVOT (n = 3) or PAs (n = 2); 2 patients had balloon dilation of both RVOT and PAs. The first interventional cardiac catheterizations was performed at an interval after TAC repair of 24 ± 20 months (range, 1.5 to 69); the second one (2 patients) was performed at 31 ± 23 months (range, 11 to 70); the third one (4 patients) at 51 ± 24 months (range, 22 to 90); and the fourth one (2 patients) at 56 and 74 months.
Reoperations
Of the 14 patients who had undergone interventional cardiac catheterizations, 8 (29.6% of the total survivors) were reoperated on at a mean of 62 months (range, 24 to102) after TAC repair, for RV-PA conduit replacement (n = 8, 29.6%), and aortic valve regurgitation plus subaortic stenosis (n = 1, 3.7%).
The conduits requiring replacement were as follows: aortic homografts, 5 of 7 (79%); pulmonary homografts, 1 of 8 (12.5%); bovine valved conduits (Contegra), 1 of 9 survivors (11%); and porcine valved conduits (Shelhigh), 1 of 3 (33%). The new conduits placed were 2 pulmonary homografts and 6 Contegras. In 5 cases (aortic homograft in 4; Shelhigh in 1), patch enlargement of branch PAs origin was necessary during conduit replacement. All 4 aortic homografts had severe calcification, while the Shelhigh conduit had neointima in-growth within its whole length. No other risk factor for PA origin stenosis could be identified.
Three patients are awaiting surgery (n = 2) or interventional cardiac catheterization (n = 1) in the near future, owing to progressive conduit calcification causing pressure gradient between RV and PAs greater than 50 mm Hg (n = 2), and stenosis at the origin of the left PA; one of the two candidates for surgery will have a second conduit replacement, 8.5 years after the first conduit change.
No mortality or major morbidity was encountered after reoperation. No reintervention has been performed after reoperation during a mean follow-up of 53 months (range, 6 to 120).
The Kaplan-Meier estimate of freedom from all reoperations, conduit reoperations, and truncal valve reoperations was 67.8%, 67.8%, and 96%, respectively, at 6 years. The overall freedom from reintervention at 6 years was 48.2% (Fig 1).
|
|
Pulmonary homografts (n = 8)
Four patients (50%) underwent 9 (mean, 2.2; range, 1 to 3) interventional cardiac catheterizations (30% of the total) for balloon dilation with or without stenting of RVOT or PAs, or both; by now, only 1 patient (12.5%) has had a conduit change at 8 years after TAC repair. Freedom from reintervention or reoperation for pulmonary homografts was 50% at 102 months (range, 71 to 143).
Bovine jugular valve conduits, Contegra (n = 11)
One patient died early. Of the 10 patients discharged from hospital, 1 died 2 months after discharge (death unrelated to conduit), and 2 underwent one interventional cardiac catheterization each for balloon dilation of branch PAs; 1 of them had subsequently replacement of his Contegra conduit at 24 months after TAC repair, owing to severe valve incompetence and stenosis of left PA origin. Freedom from reintervention or reoperation for Contegra conduits was 70% (7 of 10) at 27 months (range, 2 to 60).
Porcine valve conduits, Shelhigh (n = 3)
One patient had 4 interventional cardiac catheterizations and balloon dilation of the PAs bifurcation at 12, 24, 40, and 56 months after TAC repair, and subsequently had conduit replacement and PAs bifurcation enlargement with a Contegra conduit at 79 months after initial repair. Freedom from reintervention or reoperation for Shelhigh conduits was 67% at 34 months (range, 9 to 83).
Aortic/truncal valve
One patient (3.6%) had aortic root replacement with aortic homograft 49 months after TAC repair owing to subaortic stenosis and aortic valve regurgitation; at present, 102 months after the original procedure, he is awaiting surgery for recurrent subaortic obstruction plus moderate-to-severe aortic valve regurgitation, and RV-PA conduit replacement. At the latest echocardiography examination, the aortic/truncal valve had no or trivial regurgitation in 15 patients (55.5%), mild in 8 (29.6%), moderate in 3 (11.1%), and moderate to severe in 1 (3.7%). All patients but 1 (96%) have good RV and LV function. According to the New York Heart Association classification, there are 20 patients (74%) in class I, 6 (22%) in class II, and 1 (4%) in class III.
Other Problems
One patient is on antiepilepsy treatment with valproate sodium (6 years after surgery) and has developmental delay. The patient with postoperative hypotonia and opisthotonus had a normal neurologic examination 2 years after surgery.
| Comment |
|---|
|
|
|---|
A second issue for the ongoing debate about TAC repair is the RV-to-PA continuity reconstruction. Traditionally, the best way has been the use of a homograft conduit. An alternative is the xenograft valved conduit (bovine, porcine), or the RV-to-PA direct connection [12, 13]. In our series, we utilized exclusively valved conduits. We experienced a high incidence of reinterventions (interventional catheterizations and surgical procedures) in the aortic homograft group of patients (freedom from any reintervention, 14% in 10 years), which is in accordance with the experience of others [14, 16]; pulmonary homografts showed a better performance, with 82% freedom from any reintervention at 8.5 years. The allografts Contegra and Shelhigh had a similar performance (70% and 67% freedom from any reintervention, respectively), but we stress that the follow-up time is too short (mean, 27 and 34 months, respectively) for powerful conclusions. All reinterventions, both catheter-based and surgical, had no associated mortality and no major morbidity in our series. Therefore, the midterm survival curve remained stable throughout the follow-up time (93% after 2 months postoperatively).
Associated IAA is recognized as a significant risk factor for mortality after TAC repair, as a recent Congenital Heart Surgeons Society study clearly demonstrated (hospital survival, 44%) [17]. In our limited experience of 2 patients, IAA had not had any impact on mortality or morbidity; both patients are alive and free of any intervention 19 and 12 months after surgery.
In conclusion, our experience encourages the repair of TAC as early as possible in life, since perioperative mortality is very low and midterm results satisfactory. The encountered midterm morbidity is mainly attributed to RV-PA continuity reconstruction. About half of these patients will need repeated interventional cardiac catheterizations procedures, and about one third of them will need reoperation within 6 years after TAC repair; all these procedures can be performed with maximum safety (nill mortality). Pulmonary homografts and bovine valved conduits seem to have better performance in terms of calcification and obstruction, compared with aortic homografts. Interventional catheter-based procedures offer considerable delay of the inevitable RV-PA conduit replacement.
| Acknowledgments |
|---|
|
|
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
O. Raisky, W. B. Ali, F. Bajolle, D. Marini, O. Metton, D. Bonnet, D. Sidi, and P. R. Vouhe Common arterial trunk repair: with conduit or without? Eur. J. Cardiothorac. Surg., October 1, 2009; 36(4): 675 - 682. [Abstract] [Full Text] [PDF] |
||||
![]() |
I. Adachi, S. Y. Ho, M. M. Bartelings, K. P. McCarthy, A. Seale, and H. Uemura Common arterial trunk with atrioventricular septal defect: new observations pertinent to repair. Ann. Thorac. Surg., May 1, 2009; 87(5): 1495 - 1499. [Abstract] [Full Text] [PDF] |
||||
![]() |
F. Lacour-Gayet and S. Goldberg Surgical repair of truncus arteriosus associated with interrupted aortic arch MMCTS, March 28, 2008; 2008(0328): 2451. [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 |