Find best premium and Free Joomla templates at

Aorto-Tracheal Fistula Following Tracheo-Bronchial Stenting

C*. Cîrstoveanu , DC*. Gheorghe , A**. Cornea , R*. Tãbãcaru, A*. Diaconescu , M*. Dogaru, L*. Ilie
Bucharest, Romania



* Children’s Hospital “Maria Sklodowska Curie, Bucharest, Bd.C.Brancoveanu 20,

Neonatal Intensive Care Unit: This email address is being protected from spambots. You need JavaScript enabled to view it. ;

ENT department: This email address is being protected from spambots. You need JavaScript enabled to view it.

**Cardio-vascular Institute “C.C. Iliescu” Bucharest



Tracheomalacia is not an uncommon disorder but it usually disappears spontaneously as the child grows up, with no need for medical intervention. Severe cases still need respiratory support and sometimes surgery. Tracheal stenting is an alternative for severe cases that do not respond to other treatment modalities. Still, using of these devices can impose the risk of serious complications like the one we present in this paper. An aorto-tracheal fistula and a tracheo-esophageal fistula were observed after using a Dumon type prosthesis for tracheomalacia of the lower trachea and main left bronchus.

Key-words: tracheomalacia, tracheal stent, pediatric airway, aorto-tracheal fistula, tracheo-esophageal fistula


Tracheomalacia is a relatively rare disorder that consists of abnormal movements of the tracheal walls during respiratory cycle, determining the inspiratory obstruction of the airway. Its incidence, although difficult to determine, is about 1 in 2100 [1]. Some people proposed to name this abnormality as tracheal diskinetic movement (Couvreur, 1980). Anatomically, the collapsing of the trachea can be of different degrees and severe enough that the tracheal walls touch each other. These events occur most prominently when the child is crying or sustaining some kind of effort.

The causes for tracheomalacia vary from congenital to aquired disease. The most frequent seen is the former. There are variable anatomic reasons for tracheomalacia development. Table 1 presents the most common etiology for this disease.

Intrinsec TM Primitive
associated with other malformations
TM by extrinsec compression cardiovascular malformations
thoracic tumours
bone abnormalities of the thorax
TM after disease or surgery after prolonged intubation
after tracheostomy
sequelae after bacterial tracheobronchitis

The clinical signs of TM include: dispneea, biphasic stridor, wheezing, apneic spells with cianosis during feeding or in certain position, difficulty with clearing airway secretions [7]. The disease is not always easily recognised on clinical basis [1]. Sometimes, the disease becomes obvious from birth or is elicited after a tracheal intubation for some (eventually unrelated) reason.

The treatment of TM is not always necesary due to the varying degrees of the tracheal collapse. In time, the tracheal cartilages become more rigid and the mecanism responsible for the simptoms disappears. This fenomenon is evident at about 18-24 months of age, when, ussualy, the child does not need respiratory support any more. The methods used to deal with severe tracheomalacia varied from mask ventilation with positive airway pressure to stenting of the airway, vascular (to correct existent anomalies) or thoracic surgery [2].

Endoluminal prostheses have been introduced into clinical practice during the last ten years (Filler, 1995). The experience is scarce and different types of stents have been used to accommodate different clinical situations. We present our first experience with the Dumon type stent. Unfortunately, our case developed an aorto-tracheal fistula that lead to the death of the patient, after 4 weeks of succesful stenting. We describe the case in detail and discuss different hypotheses that could account for the evolution observed.

aorto traheal 1

Photo 1

aorto traheal 2

Photo 2

aorto traheal 3

Photo 3

aorto traheal 4

Photo 4

Case report

A child aged 7 weeks is admited to a tertiary hospital care center for repeated bouts of apnea, with respiratory distress and biphasic stridor, occuring mainly when feeding. He was born by caesarean section, after an uneventful pregnancy. His weight at birth was 2700 grams and the Apgar score was 8. The history revealed feeding difficulties from the 3rd day of life, accompanied by stridor. The child was discharged, but when the patient was 4 weeks old he presented a major apnea event, treated in hospital with intubation and respiratory support. The apneic incidents repeated oftenly since then, leading to readmitance in hospital for further investigations.

The clinical examinations performed by the pediatric and ENT department established the diagnosis of gastro-esophageal reflux and laringomalacia. Still, despite adequate medication and respiratory support, the child continued to have apnea spells making his feeding inefficient and general status not very well preserved. He was transferred to our Pediatric and Neonatal Intensive Care Department at the age of 11 weeks for complete investigations. The clinical examination upon admittance showed moderate respiratory distress with: polipnea, vesicular murmur and chest sounds decreased on the left thorax, expiratory phased increased, bilateral bronchial and alveolar rales, heart rate 140 bpm, systolic murmur III/VI on the entire precordial chest, moderate decreased in tonus and reactive status. A few hours after the admittance the child presented with an episode of respiratory arrest resuscitated with mask ventilation. The patient was refered to the ENT service for airway endoscopy.

The laryngoscopic examination of the child in our ENT department revealed no sign of laryngomalacia (Photo 1) but instead a grade I subglotic stenosis (Photo 2) and a moderate severe tracheomalacia affecting mainly the distal trachea and the main left bronchus (Photo 3). We considered that having a tracheostomy in place will relieve the negative inspiratory pressure on the tracheal walls and we perfomed the surgery the same day. After 24 hours with no problems, the child sustains 2 more respiratory arrests, apparentely without an obvious reason. The fibroscopy perfomed through the tracheostomy tube showed an important edema developed distal to the canula tip, making difficult to visualise the bronchial orifices. In this situation we changed the canula with an endotracheal tube, introduced so as to by-pass the distal tracheal edema, reaching just above the bifurcation of the trachea. Even in such instance, the left bronchial opening was closed to about 1/6 - 1/8 of its normal dimension (Photo 4).

The difference between the opened lumen and the “closed tip” in the position of the endotracheal tube was about 5mm, so it was difficult to maintain it in proper position. The patient remained mechanically ventilated for about 3 months. During this time he has been sedated with Midasolam and Fentanyl in order to adequately tolerate the tracheostomy tube and not to disturb its position. Nutrition was maintained by parenteral fluids through central lines and partially by nasogastric tube. Due to repeated occasionally bouts of oxygen desaturation occurred from birth, he developed generalized convulsions after admission in our unit, treated successfully with diazepam and valproic acid. A complete imaging diagnostic algorithm was performed, during this stage of the disease. The endoscopy of the esophagus ruled out a tracheo-esophageal fistula. The CT scan of the thorax revealed a vascular malformation, as a possible cause of the malacic trachea and left bronchus.

aorto traheal 5

Photo 5

aorto traheal 6

Photo 6

aorto traheal 7

Photo 7

aorto traheal 8

Photo 8

To be more specific, the scan using contrast material showed at T4 level a double aortic arch producing a vascular ring which incorporated lower trachea and esophagus.The left part of the arch in the anterolateral position had no permeability. That anatomical disposition created an important compression on the anterior wall of the trachea (Photo 5 and 6). Three days after the CT scan, a cardiac angiography was performed showing normal position and distribution of the pulmonary artery. No evidence of double aortic arch was found and the main left bronchus seemed to be compressed by a short, large ductus closed on the pulmonary side. The next therapeutic step was the vascular surgery that confirmed and corrected the malformations shown by the imagistic methods. The anterior aortic arch, with the emergence of the innominate artery was found and Photo 8 sectioned on its left side. The anterior arch was pulled away from the trachea in order to relieve any possible compression upon the latter. The ductus was sectioned to decrease the tension that could impede upon the tracheo-esophageal structures.

The bronchoscopy performed one week after the vascular surgery found no modifications of the anatomy of the airways: trachea and the left bronchus had the same obstruction observed before. Due to longstanding intubation using the endotracheal tube, the tracheal walls presented important granulation tissue that was contributing to the obstruction of the airway as long as the tube was withdrawn from its position. Because, due to the design of the endotracheal tube and the narrowing of the left bronchus, the ventilation of the left lung was poor (both endoscopically and clinically), we decided to use a tracheobronchial stent for better tolerance and enhanced airway patency. We chose the Dumon type stent of Y shape, for stenting simultaneously the trachea and the left main bronchus. The stent was introduced by means of the tracheostomy site. The dimensions of the tube looked well adapted to the diameter of the trachea (Photo 7), endoscopically. Only the right arm of the prosthesis seemed too large to fit the normal anatomy of the bronchus.

As stated by the manufacturer of the stent, that was not an indication to withdraw the tube but to leave it in place so it could self-deploy in its normal shape, after 24 hours. We followed such indications. The evolution of the child was favourable during the next few days, the respiratory support becoming unnecessary after 48 hours. We kept the tracheostomy in place for safety reasons. The fiberscopic examination of the stent, 48 hours after its deployment showed a normal appearance. The pulmonary ventilation seemed very good, bilaterally. The chest X-ray showed a well-positioned stent (Photo 8). Two weeks after the insertion of the stent, the child presents difficulty feeding, regurgitations and milk exteriorisation through the tracheostomy. A new CT scan failed to show a tracheoesophageal fistula; aspiration pneumonia was suspected, due to a specific radiographic image seen in the upper lobe of the right lung. Two days later, he presented an important hemorrhage (40 ml) coming out from the tracheostomy, apparently induced by coughing.

aorto traheal 9

Photo 9

aorto traheal 10

Photo 10

The bleeding stoped after blood aspiration and locally administered vasoconstricting agents (adrenaline). The endoscopy performed the next day found no signs of a clear bleeding point, so the hypothesis of bronchial wall erosion from the stent was issued. The child was sedated and mechanica- lly ventilated for the next 5 days, without any incidents. At 3 weeks after the stent insertion, at the age of 25 weeks, after small bleeding episodes during daytime, the patient had a massive haemorrhage coming through the tracheostomy; the stent was then removed and replaced with a cuffed endotracheal tube. The bleeding continued after a short delay and ultimately led to the death of the child during the night, in spite of aggressive intensive care and blood transfusion (2000 ml)! The post-mortem examination of the child revealed: double aortic arch, the main one passing on the right side of the trachea and esophagus, an aorto-tracheal fistula (located on the antero-lateral aspect of the main right!!! bronchus) (Photo 9) and a tracheoesophageal fistula on the posterior aspect of the tracheal fork (Photo 10). 


Tracheomalacia is an uncommon disorder. Its etiology is not always due to the intrinsic structure of the trachea but also to diseases found in surrounding structures. There are a large number of different therapeutical approaches presented in the literature. None of the methods described has gained widespread acceptance because the lack of randomised-control studies to prove their efficacy [3]. Usually, the specialist encounters a diversity of clinical situations that need medical and/or surgical therapy taylored to the case itself. Most of the patients do not need any intervention but medical follow-up and, eventuallly, respiratory support in times of detress.

One the methods used during the last ten years to treat this condition is tracheal stenting with appropriate protheses [4]. It is useful when other methods like aortopexy have failed to relieve pressure exerted on the trachea by a vascular anomaly. There are many types of stents, made from various materials and with different framework structure [5]. In our reported case, we began the therapeutic approach with aortopexy, because we believed the vascular ring was responsible for the small lumen of the distal trachea and left bronchus. Still, the procedure was unable to make a difference in terms of the internal diameter of the airway. We considered stenting of the trachea an interesting option. We chose the Dumon type because of its availability at small diameters and having the possibility to stent simultaneously both the trachea and the bronchi. The dimensions of the prosthesis were decided according to the dimensions of the airway evaluated endoscopically and were the same as the endotracheal tube used for previous calibrating the lumen.

After 2 weeks of stenting, the tracheal wall developed a tracheo-esophageal fistula and an aorto-tracheal fistula, the latter leading to the death of the patient. We thought that inadequate ratio between the size of the prosthesis and those of the airways was the main cause for the occurrence of such an event. Looking again at the design of the Dumon stent we can observe a design caracteristic that could account for the complications reported: a fork larger than the tracheal part of the stent (Photo 11). That could induce pressure necrotizing lesions upon the tracheal wall. The size of such a prosthesis has to be balanced between 2 goals: to achive a reasonable diameter of the airways, for adequate ventilation and to be tight in place in order to prevent movement. Dislogement of the prosthesis could also lead to obstruction of the airway and death [6]. The experience with tracheal protheses is not very large. The small number of cases makes difficult taking decisions about them. Corelation between the type of prosthesis and the rate of succes is still to find out. The advantage of different types is more or less evident.

aorto traheal 11

Photo 11


Traheomalacia is sometimes a difficult-totreat disease. The methods used include vetilatory support, surgery and tracheo-bronchial stenting. The tracheo-bronchial stents are generally well tolerated. Complications can still occur if the design of the stent is not well fitted to the size of the airway. Choosing the rigth one can be a challenging decision. Aorto-tracheal fistula and tracheo-esophageal fistula can occur, due to necrotizing lesions of the tracheal wall. Better designs of the stents that can adapt their size and shape to the inner aspect of the airway are a future goal of research. 




  1. Boogaard, R., et al., Tracheomalacia and bronchomalacia in children: incidence and patient characteristics. Chest, 2005. 128(5): p. 3391-7.
  2. Vinograd, I., et al., Airway obstruction in neonates and children: surgical treatment. J Cardiovasc Surg (Torino), 1994. 35(6 Suppl 1): p. 712.
  3. Masters, I.B. and A.B. Chang, Interventions for primary (intrinsic) tracheomalacia in children. Cochrane Database Syst Rev, 2005(4): p. CD005304.
  4. Valerie, E.P., et al., A decade of using intraluminal tracheal/bronchial stents in the management of tracheomalacia and/or bronchomalacia: is it better than aortopexy? J Pediatr Surg, 2005. 40(6): p. 904-7; discussion 907.
  5. Miyazawa, T. and K. Arita, Airway stenting in Japan. Respirology, 1998. 3(4): p. 229-34.
  6. Fayon, M., et al., French experience of silicone tracheobronchial stenting in children. Pediatr Pulmonol, 2005. 39(1): p. 21-7.
  7. MacDonald, M.G., et al., Tracheomalacia. Avery’s Neonatology; Pathophisiology & Managemengt of the newborn, 2005: p.1104