| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2003;75:1579-1586
© 2003 The Society of Thoracic Surgeons
a department of Thoracic Surgery , Royal Brompton Hospital, London, United Kingdom,
b department of Paediatric Respiratory Medicine, Royal Brompton Hospital, London, United Kingdom
Accepted for publication November 22, 2002.
* Address reprint requests to Dr Goldstraw, Department of Thoracic Surgery, Royal Brompton Hospital, Sydney St, London SW3 6NP, UK. (Email: p.goldstraw{at}rbh.nthames.nhs.uk).
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and methodsResultsCommentReferences |
|---|
Methods: Seventeen children (10 boys and 7 girls) aged 2 months to 16 years underwent tracheobronchial stenting. Etiology of the tracheobronchial obstruction included external vascular compression (n = 9); tracheobronchial anastomotic strictures after heart-lung/lung transplantation (n = 4); airway compression by malignant mediastinal mass (n = 2), and subglottic/high tracheal stenosis after prolonged intubation with a tracheostomy in situ (n = 2). Indications for airway stenting were failure to wean from ventilator after a mean of 82.5 days (range, 2 to 210) in 8 children; and dyspnea or stridor in the remaining 9 children.
Results: Ten children had a total of 24 uncovered self-expanding metal stents (either Magic Wallstent or Ultraflex Microvasive) and 7 children had silicone stents (2 straight, 3 Y and 2 T tube stents). At follow-up at 1 week to 72 months (median 21), only 8 of 17 (47%) children were alive but all the deaths were secondary to the underlying pathology and not related to tracheobronchial stenting. Six of 8 ventilator-dependent children were extubated after a mean of 5.3 days (range, 2 to 11) after airway stenting. For the 9 children stented for dyspnea, mean Medical Research Council dyspnea score decreased from 3.0 to 1.6 after stenting.
Conclusions: Tracheobronchial stenting in children is only rarely needed and often undertaken in dire circumstances. The procedure has led to significant symptomatic benefit in dyspneic children and has enabled ventilator-dependent children to be extubated. Medium-term outlook after stenting with self-expanding metal stents for vascular compression of the airway is encouraging. The long-term outcome remains uncertain, however, and is ultimately influenced by the underlying disease.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentReferences |
|---|
In children airway stenting presents special problems owing to the small size of the airway with concerns over the possible effects of the stents on the growth of the child’s airway and the long-term performance required of stents in this age group. The range of disease processes leading to tracheobronchial obstruction in the pediatric population is different when compared with that of adults.
In this paper we report our experience with endoscopic tracheobronchial stents in children.
|
Material and methods |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentReferences |
|---|
|
The self-expanding metal stents used were either Magic Wallstent or Ultraflex Microvasive stents (Boston Scientific, Natick, MA). All the stents used in this series were uncovered stents. These stents are made from filaments of a biomedical alloy woven to form a tubular mesh. The self-expanding metal stents are available premounted on a catheter down to size 5F in diameter and maintained in a precontracted state. As the stent expands on being positioned in the airway it shortens, resulting in proximal movement of the distal end of the stent. This movement has to be taken into account when positioning the stent before its release. Once released the inherent radial wall tension holds the stent against the airway and further manipulation is difficult. If the initial self-expanding metal stents failed to cover obstructed segment of the airway then further stents were telescoped within each other to span the narrowed segment. If several self-expanding metal stents were thought to be required at the start of the procedure, they were inserted sequentially beginning distally.
Silicone stents (Hood Laboratories, Pembroke, MA) are available in straight, Y, T, and T-Y configuration and in a range of sizes down to 5 mm. If a silicone stent was to be used the airway was sized to ensure a snug fit through the site of narrowed segment of the airway and cut to the appropriate length after making a bronchoscopic assessment of the length of the obstructed segment of the airway. Carinal obstruction was managed with Y stents or distal tracheal stents. Details of the construction of the silicone stents has been described in an earlier article [1].
Follow-up
Regular chest radiographs, clinical follow-up, and frequent bronchoscopy checks were undertaken. The need for regular bronchoscopy is demanding on resources and personnel with the need for almost an "open access" system for repeat bronchoscopy in the event of clinical deterioration. The first bronchoscopy examination was done at approximately 6 weeks after tracheobronchial stenting and further bronchoscopic examination depended upon the clinical situation.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentReferences |
|---|
Presenting symptoms, investigations, and diagnosis
Eight children, all with external vascular compression of the tracheobronchial tree, could not be weaned from the ventilator. Chest radiographs of these patients either confirmed areas of collapsed lung parenchyma distal to the compressed segment of the airway (Fig 1) or radiographic evidence of air trapping with hyperexpanded, oligemic lung associated with diaphragmatic flattening (Fig 2). Radiographic appearances of the distal lung parenchyma act as a guide in locating and defining the affected obstructed segment of the airway during bronchoscopy. Two patients had tracheostomy performed at other institutions before transfer to our unit to manage their subglottic/high tracheal obstruction, which had resulted from prolonged intubation. The remaining 7 patients presented with dyspnea, mean MRC dyspnea class 3.0, including 2 children who had stridor before tracheobronchial stenting.
|
|
Stents
In the series presented here, uncovered self-expanding metal stents (Magic Wallstent or Ultraflex Microvasive; Boston Scientific) were used in 10 children and silicone stents were used in 7 children (straight silicone stents in 2, Y-bifurcated silicone stents in 3, and T tube silicone stents in 2).
Self-expanding metal stents
Ten children aged 2 months to 16 years were treated with a total of 24 uncovered self-expanding metal stents (1 to 4 stents per child). These self-expanding metal stents ranged from 4 mm to 16 mm in diameter and 15 mm to 40 mm in length. The sites of stent insertion included the trachea (3 stents), left main bronchi (9 stents), right main bronchi (6 stents), and bronchus intermedius (6 stents). Follow-up for children with self-expanding metal stents has ranged from 1 week to 72 months (median, 23 months).
Nine children with congenital cardiac defects had external pulsatile vascular compression of the tracheobronchial tree and were treated with self-expanding metal stents (total of 22 stents). Indication for stenting in these 9 children included tracheobronchial obstruction leading to failure to wean from the ventilator (n = 8) and dyspnea (n = 1). In all these 9 children further surgery could not be undertaken without compromising the cardiovascular status or the children were far too unwell to withstand another major surgical procedure. In 1 child significant anastomotic strictures developed after bilateral lung transplantation for cystic fibrosis and the indication for stenting was anastomotic stricture leading to dyspnea.
All 8 of the ventilator-dependent children had external vascular compression of the tracheobronchial tree either by the pulmonary artery (n = 6) or the aorta (n = 2) and 7 of 8 of these children had undergone congenital cardiac surgery during the same admission and could not be weaned from the ventilator after surgery. These 8 children had been ventilator dependent for a median of 50 days (range, 20 to 210) and could not be weaned from the ventilator. After tracheobronchial stenting 5 of 8 children were extubated after a median 5 days (range, 2 to 11). Four of these children have now been followed up for 72, 56, 41, and 21 months after stenting, having been stented at the age of 2 months, 2 months, 6 months, and 10 years, respectively, and they remain well. The fifth child who was successfully extubated after airway stenting died 2 months later of multiorgan failure secondary to his poor cardiac function. Of the 3 children who failed to be weaned from the ventilator, 1 went on to have a tracheostomy and required bilevel intermittent positive airway pressure ventilation and is well after 23 months; the airway stents remain in situ. The other 2 could not be weaned from the ventilator owing to their poor cardiac function and died 1 and 3 weeks after airway stenting. In each case there was a satisfactory airway on follow-up bronchoscopy.
The ninth child, a 16-year-old boy with absent pulmonary valve syndrome with markedly dilated pulmonary artery compressing the major central airway, was symptomatic with dyspnea. He was treated with two self-expanding metal stents, which provided symptomatic benefit (MRC dyspnea class improving from 3 to 1); however, he died suddenly 3 months after airway stenting.
In the 10th child, a 16-year-old boy with bilateral lung transplantation for cystic fibrosis, anastomotic strictures developed and he was symptomatic with dyspnea. He was treated with two self-expanding metal stents with symptomatic improvement (MRC dyspnea class improving from 4 to2) and he remains well after 33 months.
None of the self-expanding metal stents has been removed.
Only 5 of the 10 children with self-expanding metal stents were alive after a median follow-up of 23 months (range, 1 week to 72 months). All 5 deaths were secondary to the underlying pathology 2 weeks to 14 months after stenting and were unrelated to tracheobronchial obstruction (Table 1).
Silicone stents
Seven children aged 3 months to 14 years were treated with silicone stents. Two children had straight tracheal silicone stents (10 mm); 3 children had Y silicone stents (10 mm to 14 mm) and 2 children with subglottic/high tracheal strictures (after prolonged intubation) and each with a tracheostomy in situ were treated with silicone T tube stents (5 mm and 7 mm).
The underlying etiology of the airway obstruction in children treated with silicone stents included anastomotic stricture after heart lung transplantation (n = 3), external compression of the tracheobronchial tree by mediastinal mass (n = 2), and subglottic/high tracheal strictures after prolonged intubation with a tracheostomy in situ (n = 2).
Of the 3 children with strictures after transplantation 2 were stented with a straight tracheal silicone stent; 1 of these stents was removed after 3 months. One child was stented with a 10-mm Y silicone stent initially, which was changed to a 12-mm Y silicone stent 6 weeks later and stent was removed after a further 6 weeks. There was a marked improvement in symptomatic status after stenting in all 4 children (mean MRC dyspnea class improving from 3.7 to 2.0). All three of these children died, 6, 6, and 14 months after airway stenting; in each case the cause of death was obliterative bronchiolitis. In each case satisfactory large airway patency was confirmed before death.
Two children with anterior mediastinal masses compressing the airway were treated with silicone stents. A child with mediastinal rhabdomyosarcoma presenting with stridor was stented with a silicone Y stent. After stenting the child was free of stridor and was asymptomatic from a respiratory point of view but died 2 months later of metastatic disease. The second child with lymphoma was stented with a silicone Y stent on bronchoscopic appearances of a narrowed carina but the stent displaced early and was removed on day 1.
Both the children with T tubes had repeat bronchoscopies with bougienage and their T tubes were changed serially to a larger size, 5 mm to 6 mm over a 3-month period and 7 mm to 10 mm over a 6-month period. Subsequently the T tubes have been removed after a total of 6 and 9 months of airway stenting, respectively, and both these children are well from a respiratory viewpoint at 24 and 33 months after T tube removal.
After stenting with the silicone stents there was significant symptomatic improvement with mean MRC dyspnea class improving from 3.0 to 1.4 and the 2 children with stridor before stenting were free from stridor after stenting. The duration for which the airway was stented ranged from 1 day to 9 months and in 5 of these children the stents were eventually removed.
After a median follow-up of 6 months (range, 6 to 33), only 3 of 7 children (43%) were alive although none of the deaths were related to the large airway treated with stents. Both children with T tubes went on to have repeat bronchoscopies and bougienage dilatation and their T tube stents were changed serially to a larger one (5 mm to 6 mm and 7 mm to 10 mm). Eventually both the T tubes were removed and both the children are well and asymptomatic. There were 4 deaths 2 to 6 months after stenting secondary to obliterative bronchiolitis in transplanted children or metastatic disease.
Survival after tracheobronchial stenting in children
Of the 17 children treated by tracheobronchial stenting only 8 children (47%) were alive after a median of 21 months (range, 1 week to 72 months). The cause of death is detailed in Table 1 and in each case the cause relates to the underlying pathology. Before death each child had a satisfactory airway on follow-up bronchoscopy and the death was unrelated to tracheobronchial stenting.
Complications
In 1 child aged 2 months at the time of stenting who had a tracheal self-expanding metal stent significant granulation tissue developed and required a further tracheal self-expanding metal stent 4 weeks after the initial stent insertion. The child remains well after 72 months.
One silicone stent was noted to have displaced on a routine postprocedural chest radiograph and was therefore removed. The child was only mildly symptomatic (MRC dyspnea class 2) and was stented on bronchoscopic appearances of a narrowed carina. She turned out to have mediastinal lymphoma, which responded very well to chemotherapy; she did not require any further airway stents.
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentReferences |
|---|
The experience with airway stents in children is limited. Concerns regarding the effect of growth on stent position, long-term effects of the stent on the child’s airway, effectiveness and risk of migration of the stent, and induction of granulation tissue make it difficult to decide upon the ideal stent. Several options are available and stent technology continues to evolve. Stents available include self expanding metal stents that may be covered or uncovered [1, 2, 12, 13], balloon-expandable metal stents [8, 11, 12], silastic stents [1], silicone stents [2], extraluminal stents [14–16], and completely absorbable airway stents made from polyglactin 910 (Vicryl) [17]. Each of the stents in current clinical use has its merits and drawbacks. The ideal stent should be simple to insert, support the airway without causing any adverse effect while in place, and be simple to remove or not require removal. None of the stents currently available meets all these objectives.
All uncovered expandable metal stents (self-expanding or balloon expandable) to some extent tend to become epithelialized over time and subsequently incorporated into the wall of the airway. While this prevents stent migration it means that removal becomes increasingly difficult with time [16]. Therefore self-expanding metal stents should probably be considered as permanent implants. Balloon-expandable metal stents require regular bronchoscopy and balloon expansions [18–19]. Moreover if the child’s airway outgrows the stent diameter then the stent has the potential to migrate [12]. Self-expanding metal stents minimize the risk of migration as the radial tension allows the stent to expand continuously against the airway supporting the obstructed segment of the airway. In the series reported here a total of 24 uncovered self-expanding metal stents were used in 10 children and to date we have not experienced any stent migration or displacement. Furthermore we believe that the self-expanding metal stents are preferable as they retain contact with the airway wall even when one cannot assess the true diameter of the airway, and that is particularly pertinent in children where the unknown factor of growth has to considered. It is likely that there is little to choose between the self-expanding metal stents that we have used in our patients and balloon-expandable metal stents such as Palmaz stents used by others [8, 11]. For the reasons discussed above, however, and based on our large experience with self-expanding metal stents in adults we have preferred the self-expanding metal stents in the pediatric age group too.
Both self-expanding and balloon-expandable metal stents do not expand to any calculated endpoint. It is uncertain as to what effect such stents may have on airway growth. The longest follow-up in our series has been 72 months; there has been no adverse effect on the growth of the airway. At present there are no objective data on the long-term effect of these self-expanding metal stents on the child’s airway and therefore any assessment of this aspect remains speculative. In our series we have children 72, 56, 41, and 21 months after stenting, having been stented at the age of 2 months, 2 months, 6 months, and 10 years, respectively, and they remain well. In all these children to date there has been no clinical indication for further assessment of the airway but this is an area that requires further exploration.
Silicone and silastic stents are easy to insert and remove. There is little to choose between these two stents and they vary only in the composition of their material. Both silicone and silastic stents are ideal for short-term use or where healing and recovery are expected. Silicone and silastic stents may cause excessive granulation tissue at the upper and lower margins, however, causing further airway obstruction. They may also obstruct with inspissated mucus and thus require frequent bronchoscopic examination. Silicone and silastic stents have thicker walls than self-expanding metal stents and hence the interior/exterior diameter of the stent is reduced significantly, making them less than ideal for the small airway particularly in young children. Silicone and silastic stents have a fixed diameter and do not expand. Therefore as the child outgrows the stent, the stent has to be changed frequently for a larger one.
Failure to wean from a mechanical ventilator after cardiac surgery in children has multifactorial causes [20]. Phrenic nerve injury is often a contributing factor and this was the case in 2 children in the series presented here. These 2 children failed to wean from the ventilator despite diaphragmatic placation and required airway stenting in an attempt to wean them from the ventilator.
Causes of airway obstruction include extrinsic compression either by an enlarged pulmonary artery (left to right shunting of blood, absent pulmonary valve syndrome) or by direct compression by the heart and other great vessels [21]. Surgical repair of a severe coarctation or interrupted aortic arch may result in external compression of the tracheobronchial tree by the reconstructed aorta. Dissection of the vessels around the airway leads to edema, which can distort the anatomy of the tracheobronchial tree [7]. The diagnosis and management of severe airway obstruction due to vascular compression leading to ventilator dependency is challenging. In our experience rigid bronchoscopy is the most useful investigation in making an accurate assessment as the ventilation can be suspended while one evaluates any possible dynamic compression by adjacent vascular structure [13].
If the airway obstruction is secondary to dynamic vascular compression, then one would first address the vascular anomaly either by plication/remodelling or reducing the blood flow through it (for example closing the left to right shunt in the event of a major pulmonary artery compressing the airway). Two of the ventilator-dependent children in the series reported here underwent a second surgical procedure to close the shunt to reduce blood flow through the pulmonary arteries compressing the airway. That failed to deal with the compression of the airway issue and both went on to have airway stenting. In all the children with vascular compression of the airway in this series further surgical correction was not felt to be appropriate without compromising the cardiovascular status. Furthermore the ventilator-dependent children were unwell and really clinically not well enough to withstand further major surgery. The role of tracheostomy and prolonged ventilation has already been mentioned. Endoluminal tracheobronchial stenting may provide an alternative to facilitate weaning from the ventilator. Eight of the children in our series were ventilator dependent for a median of 50 days (range, 20 to 210) and all had tracheobronchial stenting with self-expanding metal stents. Five of these children were successfully extubated after a median of 5 days (range, 2 to 11) after completing tracheobronchial stenting.
The long-term outlook after placement of endobronchial self-expanding metal stents for vascular compression of the tracheobronchial tree remains uncertain. Bronchial fistula has been reported in the literature with the use of both Gianturco self-expanding metal stent and Palmaz balloon-expandable metal stent and has occurred as early as 3 weeks after stenting [22]. It is hoped that the design of the new models of stents that have more filaments per unit length will avoid that risk. In the series reported here self-expanding metal stents were used in 9 children with external vascular compression of the airway with medium-term follow-up in 4 children. These 4 children who were stented at the age of 2 months, 2 months, 6 months, and 10 years have now been followed up for 72, 56, 41, and 21 months, respectively, without any adverse effect occurring.
Induction of granulation tissue leading to further endoluminal obstructions remains a major concern. Regular check-up bronchoscopy is required initially after deploying any stent to remove any obstructive granulation tissue. Bronchoscopic examinations are undertaken as indicated clinically. It is our routine practice, however, to undertake bronchoscopy to remove any granulation tissue after about 6 weeks. Thereafter we are guided by the clinical condition of the patient in deciding further bronchoscopic examination. This approach has enabled us to keep the silicone stents satisfactorily patent and allowed the self-expanding metal stents to be epithelialized and incorporated within the airway wall.
The microbiologic flora associated with granulation tissue and stent infections have been well documented. Pseudomonas aeruginosa, Streptococcus viridans, nonhemolytic streptococci, and Staphylococcus aureus have all been identified and therefore antibiotic prophylaxis needs to be tailored appropriately [23]. In our series 1 infant was found to have granulation tissue causing significant narrowing of the airway at the distal end of the stented segment and was treated with another tracheal stent. This particular problem may be prevented with the development of covered metal stents. Polypropylene mesh covering has been shown to have excellent biohistocompatibility compared with other materials such as silicone and polytetrafluoroethylene (PTFE) in mongrel dogs [23] but covered stents will obstruct any airway that they bridge. In our experience the available models are less stable and may become displaced. Experimental work suggests that thermolabile metallic stents may induce less granulation tissue [24].
In our series silicone Y stents were used successfully to treat carinal obstruction in older children. Silicone stents are not appropriate for neonates and infants and self-expanding metal stents are preferred. Carinal obstruction was therefore relieved by distal tracheal stenting or by stenting both the bronchi, allowing the proximal ends of the two stents to span the carina. Others have reported their limited experience with customised carinal Q stents [16].
The long-term outlook for these children remains uncertain as survival was only 47% after a median follow-up of 21 months. It is noteworthy, however, that all the deaths were related to the underlying disease entities and all the children who died had a satisfactory airway after stenting.
Conclusions
Although tracheobronchial stenting in children is not a panacea, it may represent the best of a limited range of treatment options for many sick children in dire circumstances. Tracheobronchial stenting has been valuable for providing symptomatic benefit to dyspneic children and for enabling ventilator-dependent children to be extubated. The long-term outcome remains uncertain but the medium-term outlook is encouraging. The prognosis is ultimately dependent on the underlying cardiac or pulmonary pathology.
|
References |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. L. Anton-Pacheco, D. Cabezali, R. Tejedor, M. Lopez, C. Luna, J. V. Comas, and E. de Miguel The role of airway stenting in pediatric tracheobronchial obstruction Eur J Cardiothorac Surg, June 1, 2008; 33(6): 1069 - 1075. [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 |
