Severe Airflow Limitation After the Unifocalization Procedure
Clinical and Morphological Correlates
Background—While unifocalization techniques have improved the treatment options in patients with pulmonary atresia, ventricular septal defect (PA-VSD), and major aortopulmonary collaterals (MAPCAs), severe airflow limitation contributes to significant early postoperative morbidity and mortality. Although this has been attributed to bronchospasm, characteristically it is refractory to bronchodilators, suggesting that other mechanisms may play a role.
Methods and Results—The clinical course and preoperative angiograms of patients who underwent unifocalization were reviewed. Patients who developed airflow limitation early after surgery underwent fiberoptic bronchoscopy. In addition, the anatomy of the MAPCAs was examined in 14 heart-lung blocks from patients with PA-VSD. Twenty-two procedures were performed in 16 children. Three developed marked airflow limitation early after surgery, necessitating prolonged high-pressure ventilation. Bronchoscopy demonstrated tracheobronchial epithelial necrosis in 2 and signs of tracheobronchial ischemia in the third. Two were successfully extubated after 15 and 16 days, but the third died after 57 days of ventilatory support. Review of the preoperative angiograms demonstrated an extensive peribronchial arterial supply arising from a MAPCA in 1 of the patients who developed severe airway necrosis after unifocalization. This was also obvious in a second patient, but the MAPCA was not included in the unifocalization. In 7 autopsy specimens, MAPCAs contributed to a peribronchial or peritracheal vascular network. Dissection of the distribution of these branches in 2 specimens revealed extensive intrapulmonary peribronchial anastomoses.
Conclusions—Airflow limitation early after unifocalization is related to airway ischemia resulting from interruption of the tracheobronchial blood supply during mobilization of MAPCAs.
In patients with pulmonary atresia and ventricular septal defect, the pulmonary blood supply is often derived from multiple sources. Major aortopulmonary collateral arteries (MAPCAs), which most often arise from the aorta and its direct branches, are probably derived from the bronchial circulation1 2 3 4 5 and potentially make an important contribution to pulmonary blood flow. The development of surgical techniques in which these multiple sources of pulmonary blood supply can be perfused from a single source (unifocalization)6 7 has considerably enhanced the management of this difficult patient group. However, although these unifocalization procedures can be performed with low perioperative mortality, the early postoperative period may be complicated in some patients by the development of severe airflow limitation.8 9 Although this airflow limitation has been attributed to bronchospasm, characteristically it is refractory to bronchodilator treatment with adrenergic agonists and halothane.10 If it is severe, it may result in death.
The pathophysiological basis for airflow limitation early after unifocalization is unknown. The refractory nature of this phenomenon makes it likely that processes other than classic “bronchospasm” are responsible for its development. In this article, we provide a clinical-morphological assessment of patients with pulmonary atresia and ventricular septal defect who underwent a unifocalization procedure, with particular reference to the pathogenesis of airflow limitation.
We reviewed the clinical course of 16 patients with pulmonary atresia and ventricular septal defect who underwent unifocalization procedures between January 1996 and May 1999 at our hospital. The demographic details of the patients are given in Table 1⇓. In 15 patients, the unifocalization was performed through a lateral thoracotomy; in 1, it was performed as part of a total correction through a median sternotomy. Six patients underwent successive procedures on both lung sides, so that 22 unifocalization procedures in all were studied. In 3 patients, the early postoperative course was characterized by the development of severe ventilatory difficulties, which were refractory to bronchodilator treatment.
In these 3 patients who developed airflow limitation, we performed fiberoptic bronchoscopy to ascertain the mechanism for the airflow limitation. As will be discussed, these examinations demonstrated appearances consistent with severe airway ischemia.
Review of Angiograms
To delineate the vascular supply of the proximal tracheobronchial tree in our group, we reviewed the preoperative angiograms of all 16 patients. The number, origin, course, and distal blood supply of the MAPCAs were noted, with particular attention given to any supply from them to the tracheobronchial tree.
We dissected a series of stored heart-lung blocks of 14 patients with pulmonary atresia and ventricular septal defect. We examined only those hearts in which atresia of the pulmonary trunk was found in association with concordant AV connections and in which the great arteries were “normally related,” with the aorta posterior and to the right of the remnant of the pulmonary trunk. We paid particular attention to the pulmonary vascular supply and to the course of the MAPCAs on the basis of the normal bronchopulmonary segments and their relationship to the tracheobronchial tree. In 2 cases, it was possible to perform a detailed dissection of small peribronchial anastomoses that arose from distal branches of the MAPCAs.
Sixteen patients, 11 male and 5 female, underwent 22 surgical procedures for pulmonary atresia, ventricular septal defect, and MAPCAs, which consisted of a staged unifocalization procedure in 15 patients via a lateral thoracotomy and a complete correction through cardiopulmonary bypass in 1 patient (patient 9) via a median sternotomy (Table 1⇑). There was associated DiGeorge syndrome in 3 patients, Alagille syndrome in 1, and the VACTERL association in 1. The median age at unifocalization was 2.8 years (range, 0.1 to 13.2 years). The number of collaterals that were unifocalized ranged from 1 to 3 (median, 2.5).
In 19 procedures, the MAPCAs were connected to each other and then to an expanded polytetrafluoroethylene (PTFE) shunt (range, 3.5 to 6 mm) that was anastomosed to the subclavian artery7 (Figure 1⇓). In 2 cases, the mobilized azygous vein was used to augment the anastomosis between collaterals,11 and in 1 case, the MAPCAs were unifocalized to the already shunted central pulmonary artery. In 1 patient, the unifocalization was performed as part of a complete correction. Significant postoperative complications included chylous pleural effusion (n=1), pneumonia (n=1), hemothorax requiring surgical exploration (n=1), and phrenic nerve palsy (n=3), and 3 patients experienced severe airflow limitation with expiratory wheeze and signs of air trapping resembling bronchospasm, necessitating ventilation with high inspiratory pressures (peak inspiratory pressure >30 cm H2O), prolonged expiratory time to prevent auto–positive end-expiratory pressure (as measured from online ventilatory flow-pressure studies), and hypoxia requiring oxygen supplementation. All were unresponsive to bronchodilator therapy, inhaled nitric oxide, and steroids. Compared with the patients without airflow limitation in whom the median duration of ventilatory support was 2.1 days, 2 patients with airflow limitation required 15 and 16 days of ventilation, and 1 patient died after 57 days of ventilatory support. Consent for postmortem examination of this patient was refused. The other 2 survivors of severe airflow limitation were in good health on follow-up without symptoms of residual airflow obstruction.
In the 3 patients with airflow limitation, bronchoscopy demonstrated extensive necrosis of the trachea and main bronchi with sloughing of the mucosa, resulting in severe airway obstruction in 2 patients (Figure 2⇓). In the third patient with less severe disease, there was severe mucosal pallor, suggesting ischemia and minor mucosal sloughing.
Review of Preoperative Angiograms
In all, the courses of 48 MAPCAs were studied (Table 1⇑). These originated from the descending aorta (81%) or the right or left subclavian artery (19%). In 2 patients (patients 2 and 16), an MAPCA was seen to supply an extensive peribronchial network (Figure 3⇓, top and bottom). One of these patients developed severe airway necrosis in the postoperative period after this MAPCA was unifocalized; in the second patient, the MAPCA was not included in the unifocalization procedure.
In the 14 patients in whom autopsy specimens were studied (Table 2⇓), the age at death ranged from 0.1 to 39.8 years (median, 2.9 years). Five had died in the early postoperative period. The clinical notes described airflow limitation and desaturation in 2 patients, with early postoperative bronchoscopy demonstrating marked edema of the right main bronchus and its branches, causing airway obstruction and right upper lung collapse in 1 patient (patient 5) who died of pericardial tamponade 5 weeks later, and signs of submucosal hemorrhage of the left main bronchus and its branches in the autopsy specimen of the other patient (patient 8). Forty-four MAPCAs were identified, and their courses were analyzed. Eight were found to contribute to a peribronchial or peritracheal vascular network (Figure 4⇓, top), and careful dissection of distal branches of these collaterals in 2 specimens revealed the occurrence of extensive intrapulmonary bronchomucosal anastomoses (Figure 4⇓, bottom).
Although unifocalization has expanded the treatment options in patients with pulmonary atresia, ventricular septal defect, and MAPCAs, severe airflow limitation in the early postoperative period can be a life-threatening complication. This study investigated the possible mechanisms. In patients who developed airflow limitation, we demonstrated severe tracheobronchial necrosis, confirming that epithelial ischemia with luminal obstruction rather than bronchospasm formed the basis for the airflow limitation. A review of angiograms and postmortem specimens showed that an intimate relationship exists between some of the MAPCAs and the tracheobronchial tree.
Unifocalization procedures have been developed for the treatment of patients with multiple sources of pulmonary blood supply. The precise surgical technique, in particular, whether a central or lateral approach should be undertaken, remains controversial. In most of our patients, a lateral approach was used that generally involved ligation and mobilization of collaterals. In most patients, this required minimal extrahilar dissection of the collateral arteries, but in some, it was necessary to dissect distally into the lung parenchyma.
Three patients developed severe airflow limitation, a similar proportion to that seen in other series.8 9 Superficially, the clinical appearance was similar to classic bronchospasm, and increased airway reactivity with bronchospasm has been reported in patients with pulmonary atresia and hypoplastic pulmonary arteries.12 However, as previously observed, it was refractory to treatment with bronchodilator therapies in our patients.11 One patient had airflow limitation of such severity that she died after prolonged intensive care. In other reports, the morbidity and mortality for this disease process have been considerable. In one series, it was described in 3 of 4 patients with prolonged ventilatory requirements,8 and in another series, it was the cause of 2 of 3 early postoperative deaths.9
Fiberoptic bronchoscopy demonstrated that in our patients, the airflow limitation was related not to bronchospasm but to severe airway necrosis, suggesting that the nutritive supply to the airway epithelium was compromised by the unifocalization procedure and that other mechanisms for unresponsiveness to bronchodilators, such as lymphatic flow interruption13 or denervation hypersensitivity,14 were therefore unlikely. A review of the preoperative angiograms and careful dissection of a series of postmortem specimens demonstrated that in some patients with pulmonary atresia, ventricular septal defect, and MAPCAs, extensive vascular networks arise from the collateral arteries that provide significant blood flow to the large airways.
In normal subjects, there are multiple small systemic arteries that serve a nutritive role to the lungs. In addition, there are 3 to 4 recognizable bronchial arteries: 2 supplying the left lung and 1 to 2 supplying the right lung.4 It is widely considered that in patients with pulmonary atresia and ventricular septal defect, at least some major aortopulmonary collaterals have an embryological origin from these bronchial arteries.3 5 However, it has been suggested that in these patients, the collaterals have lost their nutritive role to the lungs, being functionally “pulmonary” rather than “bronchial.”5 15 16 This assumption is based on the hypothesis that collateral arteries join the pulmonary arterial circulation at the hila or within the lungs before reaching the respiratory units and proximal to the origin of any nutritive end arteries. However, more recent anatomic descriptions1 6 17 and our own observations in this study bring this view into question and demonstrate that collateral arteries may give branches to the tracheobronchial tree before they anastomose with the peripheral pulmonary circulation. Thus, the major aortopulmonary collaterals not only may be pulmonary in function but also may have an important bronchial role and nutritive component to the tracheobronchial tree. The precise anatomy of the MAPCA needs to be delineated with selective high-resolution angiography, which may require pressurized injections of contrast material into each of the target collaterals to overcome wash-in/washout phenomena and the potentially higher resistance to blood flow to the peribronchial network compared with the lung parenchyma. If such a peribronchial network originating from a specific collateral can be demonstrated, then the subsequent surgical technique should take this information into account to avoid damage to the tracheobronchial arterial supply and to obviate the morbidity and mortality associated with tracheobronchial ischemia.
In summary, our data suggest that careful preoperative and intraoperative attention must be paid to the potential nutritive role of each MAPCA in patients in whom unifocalization is considered. If after the unifocalization procedure such patients develop signs of airflow limitation that is unresponsive to standard bronchodilator treatment, prompt bronchoscopy should be performed.
This work was made possible by a grant from the SPARKS charity (No. 98/BRM/1).
- Copyright © 2000 by American Heart Association
1. Liao P, Edwards W, Julsrud P, et al. Pulmonary blood supply in patients with pulmonary atresia and ventricular septal defect. J Am Coll Cardiol. 1985;6:1343–1350.
Wagenvoort C, Wagenvoort N. Arterial anastomoses, bronchopulmonary arteries, and pulmobronchial arteries in perinatal lungs. Lab Invest. 1967;16:13–24.
Haworth SG, Macartney FJ. Growth and development of pulmonary circulation in pulmonary atresia with ventricular septal defect and major aortopulmonary collateral arteries. Br Heart J. 1980;44:14–24.
Deffebach M, Charan N, Lakshminarayan S, et al. The bronchial circulation: small, but a vital attribute to the lung. Am Rev Respir Dis. 1987;135:463–481.
Rabinovitch M, Herrera-de Leon V, Castaneda A, et al. Growth and development of the pulmonary vascular bed in patients with tetralogy of Fallot with or without pulmonary atresia. Circulation. 1981;64:1234–1249.
Haworth SG, Rees PG, Taylor JF, et al. Pulmonary atresia with ventricular septal defect and major aortopulmonary collateral arteries: effect of systemic pulmonary anastomosis. Br Heart J. 1981;45:133–141.
Sullivan I, Wren C, Star J, et al. Surgical unifocalization in pulmonary atresia and ventricular septal defect: a realistic goal? Circulation. 1988;78(suppl III):III-5–III-13.
Reddy V, Liddicoat J, Hanley F, et al. Midline one-stage complete unifocalization and repair of pulmonary atresia with ventricular septal defect and major aortopulmonary collaterals. J Thorac Cardiovasc Surg. 1995;109:832–845.
Puga F, Leoni F, Julsrud P, et al. Complete repair of pulmonary atresia, ventricular septal defect, and severe peripheral arborization abnormalities of the central pulmonary arteries. J Thorac Cardiovasc Surg. 1989;98:1018–1029.
Oliver W, Murray M, Raimundo H, et al. The use of halothane to treat severe bronchospasm after a unifocalization procedure. J Cardiothorac Vasc Anesth. 1995;9:177–180.
Iyer K, Varma M, Mee R. Use of azygos vein as interposition graft for surgical unifocalization of pulmonary blood supply. Ann Thorac Surg. 1989;48:776–778.
Jedele KB, Michels VV, Puga FJ, et al. Velo-cardio-facial syndrome associated with ventricular septal defect, pulmonary atresia, and hypoplastic pulmonary arteries. J Pediatr. 1992;89:915–919.
Deslauriers J, Aucoin A, Gregoire J. Postpneumonectomy pulmonary edema. Chest Surg Clin N Am. 1998;8:611–631.
Thiagarajah S, Lear E, Azar I, et al. Bronchospasm following interscalene brachial plexus block. Anesthesiology. 1984;61:759–761.
Thiene G, Frescura C, Bini RM, et al. Histology of pulmonary arterial supply in pulmonary atresia with ventricular septal defect. Circulation. 1979;60:1066–1074.
Jefferson K, Rees S, Somerville J. Systemic arterial supply to the lungs in pulmonary atresia and its relation to pulmonary artery development. Br Heart J. 1972;34:418–427.
Luciani GB, Swilley S, Starnes VA. Pulmonary atresia, intact ventricular septum, and major aortopulmonary collaterals: morphogenetic and surgical implications. J Thorac Cardiovasc Surg. 1995;110:853–854.