Introduction
Laryngotracheal stenosis still remains a therapeutic challenge. In the adult population, this condition is commonly caused by mechanical trauma from endotracheal intubation, external trauma, upper respiratory tract infection, systemic disorders (e.g. Wegener's granulomatosis) and neoplasms.
However, in a small number of cases the cause of subglottic stenosis cannot be determined. These patients usually have a long history of progressive shortness of breath, hoarseness and wheezing. The usual macroscopic appearance of the lesion is circumferential, fibrotic stenosis that is limited to the cricoid cartilage and the first one or two tracheal rings, usually occurring in female patients.Reference Benjamin, Jacobson and Eckstein1, Reference Ashiku and Mathisen2 This condition is termed idiopathic laryngotracheal stenosis.
Many different treatment methods have been used for idiopathic laryngotracheal stenosis, ranging from conservative therapies to definitive surgical procedures.Reference Dedo and Catten3–Reference Ashiku, Kuzucu, Grillo, Wright, Wain and Lo6 Conservative treatment involves local corticosteroid injections, airway dilatation and laser ablation. Definitive surgical procedures include laryngotracheal segmental resection and primary anastomosis, which have been considered to have good to excellent long-term results. However, the optimal management strategy in the treatment of idiopathic laryngotracheal stenosis remains controversial. Treatment with laryngotracheal reconstruction has been reported in a few cases.Reference Benjamin, Jacobson and Eckstein1
In this paper, we present our experience in treating five cases of idiopathic laryngotracheal stenosis using laryngotracheal reconstruction, over a five-year period.
Materials and methods
From January 2001 to December 2005, 129 patients with laryngotracheal stenosis were treated at our hospital. Of these patients, five (4 per cent) female patients were identified as having idiopathic laryngotracheal stenosis. Their mean age was 23 years (range, 14 to 34 years). The time interval from onset of symptoms to diagnosis ranged from 10 to 36 months (mean, 21.8 months). These patients' follow-up times ranged from 24 to 76 months (mean, 54.6 months).
The diagnosis of idiopathic laryngotracheal stenosis was made by excluding known causes of subglottic stenosis. All patients were tested for antineutrophil cytoplasmic antibodies to rule out Wegener's granulomatosis. No patients had gastroesophageal reflux disease. No patient gave a history of intubation for general anaesthesia in the two years before onset of symptoms. Flexible fibre-optic laryngoscopy and biopsy were performed to exclude neoplasm, granuloma, infection, vasculitis and other systemic connective tissue disorders as causes. Plain X-rays and computed tomography scanning of the larynx and upper trachea were performed to assess the extent and severity of the lesion. Stenosis was graded according to the Myer-Cotton grading system.Reference Myer, O'Connor and Cotton7
Previous treatment had included laser ablation in one patient, laryngeal enlargement procedures with T-tubes in two patients and the same without T-tubes in one patient. Four patients were tracheostomy-dependent. These treatments had been performed at other institutions.
Laryngotracheal reconstruction with costal cartilage grafting was undertaken. This operation has been well described by Cotton and Zalzal and Cotton.Reference Cotton8, Reference Zalzal and Cotton9 Briefly, a mid-line incision was made anteriorly in the cricoid and upper trachea. A boat-shaped costal cartilage graft was used to expand the airway, placed anteriorly with stenting. A Montgomery T-tube was placed, and remained in situ for 12 to 22 months. We modified the operation by removing any scar tissue and repairing the defect with a free skin graft.
Results and analysis
A summary of patient data is shown in Table I. Of the five patients, three presented with grade three and two with grade two stenosis. The stenosis was almost always located in the region between the upper edge of the cricoid and the lower edge of the second tracheal ring.
Table I Patient data: summary
* Myer-Cotton classification. †Nd:YAG. ‡Nine months then 13 months. Pt = patient; y = years; prev = previous; trache = tracheostomy; mths = months; FU = follow up; F = female; Y = yes; LTR + CCG = laryngotracheal reconstruction with anterior costal cartilage grafting
Four patients were able to be decannulated, with normal respiratory function and good exercise tolerance. One patient failed decannulation. Three patients were decannulated after one procedure and one after two procedures (the failed patient underwent two procedures). The mean time to decannulation was 14.8 months (range, 12 to 22 months).
In the one patient not extubated (patient one), severe scar tissue was present from the subglottis to the second tracheal ring and almost no cricoid or tracheal rings could be identified, because of previous procedures.
For patient three, we used only laryngotracheal reconstruction with scar tissue removal and free skin grafting, instead of costal cartilage grafting, because of the mild degree of the lesion.
Patient two was extubated nine months after surgery but failed decannulation. She required a second laryngotracheal reconstruction procedure, and a T-tube remained in place for a further 13 months, before final decannulation.
Patients three, four and five required only one procedure, and were decannulated after 12 months of stenting.
The most common complication was granulation tissue formation in the trachea and epiglottis. Four patients had granulation tissue growth requiring laser removal (one to three applications).
Discussion
The aetiology of idiopathic laryngotracheal stenosis is still unclear.
Oestrogen is considered to be the major cause, as almost all patients reported in the literature have been female. However, Benjamin et al. and Dedo and Catten found no oestrogen receptors in samples taken from the lesion area in patients with idiopathic laryngotracheal stenosis.Reference Benjamin, Jacobson and Eckstein1, Reference Dedo and Catten3
Another possible cause is long-term gastroesophageal reflux disease. Jindal et al. reported six patients with idiopathic laryngotracheal stenosis who had stabilised and responded to surgical management after evaluation for gastroesophageal reflux disease followed by medical management.Reference Jindal, Milbrath, Shaker, Hogan and Toohill10 However, other authors speculated that if gastroesophageal reflux disease were the underlying cause, one would expect many more male patients, and also recurrence of stenosis in the absence of reflux treatment.Reference Dedo and Catten3, Reference Grillo, Mathisen, Ashiku, Wright and Wain11 In Grillo's series of 73 patients, none developed recurrent stenosis over many years, despite receiving no treatment for gastroesophageal reflux disease. None of our patients had symptoms or signs related to gastroesophageal reflux disease.
Idiopathic laryngotracheal stenosis is a diagnosis of exclusion. Biopsy and histopathological examination are important to exclude known causes such as tuberculosis, histoplasmosis and amyloid disease. Antinuclear cytoplasmic antigen testing is helpful to exclude Wegener's granulomatosis. When all known causes have been excluded, a diagnosis of idiopathic laryngotracheal stenosis can be made. As the signs and symptoms are mild at the onset of the disease, it usually takes a long time to make a final diagnosis. In our series, a mean time of 21.8 months was observed between onset of symptoms and final diagnosis.
There is currently no consensus on the optimal treatment of idiopathic laryngotracheal stenosis.
Benjamin et al. treated 12 patients with endoscopic laser vaporisation, and were successful in maintaining the airway of eight patients with an average of 4.2 laser procedures per patient (range, two to eight).Reference Benjamin, Jacobson and Eckstein1
Dedo and Catten used operative techniques (endoscopic laser submucosal resection and mucosal flap rotation) in 50 patients.Reference Lee and Rutter13 Most patients required continued treatment, with an average number of six procedures each, while 13 patients failed to decannulate.
Some authors consider endoscopic laser-assisted dilatation and scar tissue resection to be the treatments of choice for initial management.Reference Benjamin, Jacobson and Eckstein1, Reference Valdez and Shapshay4, Reference Giudice, Piazza, Foccoli, Toninelli, Cavaliere and Peretti12
In repeated failures, open neck surgery with laryngoplasty or laryngotracheal resection and anastomosis is recommended.
Lee and Rutter recently reported a series of six patients treated by balloon dilatation.Reference Lee and Rutter13 The airway was dilated to 2.0 to 3.5 endotracheal tube diameters larger than its initial width. Four patients were followed for 10 to 30 months, without recurrence.
Grillo and colleagues have advocated definitive treatment with single-stage laryngotracheal resection.Reference Grillo, Mathisen, Ashiku, Wright and Wain11 In their series of 73 patients, all were successfully decannulated; 67 (91 per cent) had good to excellent long-term results in terms of voice and breathing quality, and did not require further intervention.
We used laryngotracheal reconstruction with anterior costal cartilage grafting to treat idiopathic laryngotracheal stenosis. This procedure is a popular and effective method of widening the airway lumen in paediatric cases of laryngotracheal stenosis, but it has not been confirmed as useful in the management of adult, idiopathic laryngotracheal stenosis.Reference Benjamin, Jacobson and Eckstein1, Reference Giudice, Piazza, Foccoli, Toninelli, Cavaliere and Peretti12
Both Cotton and Rethi have emphasised that scar tissue should not be removed during the procedure, since to do so would produce a raw intraluminal surface.Reference Cotton8, Reference Rethi14 Idiopathic laryngotracheal stenosis has a unique histopathological appearance, with bland inflammatory fibrosis of the keloidal type, a thickened lamina propria and intact cartilage.Reference Grillo, Mark, Mathisen and Wain5 Therefore, we chose to resect the scar, in order both to widen the airway lumen and to remove the pathological tissue; we then repaired the defect by free skin grafting.
It is our experience that the Montgomery T-tube should remain in place for a relatively long time, although there is no optimal duration of stenting following laryngotracheal reconstruction. Stern et al. reported a duration of stenting ranging from two weeks to 23 months in paediatric patients, with a mean of 7.4 months.Reference Stern, Willging and Cotton15 In patients with idiopathic laryngotracheal stenosis, a long period of stenting may be needed in order to hold the cartilage and skin grafts in position and to prevent scar contracture. In our series, all patients wore a T-tube for at least 12 months. Patient two failed to decannulate after nine months of stenting, and required a second laryngotracheal reconstruction procedure and stenting for a further 13 months before ultimate decannulation.
• Idiopathic laryngotracheal stenosis has been treated by various surgical techniques; effective management by laryngotracheal reconstruction with anterior costal cartilage grafting has not been confirmed
• Laryngotracheal reconstruction was found to be a relatively safe and effective method, and thus provides an alternative treatment for idiopathic laryngotracheal stenosis
• Removal of scar tissue and repair of the defect with a free skin graft, and a long period of T-tube stenting, may be important for achieving decannulation
For patients with severe scar tissue formation, laryngotracheal reconstruction with cartilage grafting may not be suitable. Our patient one was such a case, failing due to previous, unsuccessful procedures and severe scar formation in the subglottic and upper tracheal regions. This patient eventually refused to receive any further therapy, including further laryngotracheal resection.
Conclusion
Laryngotracheal reconstruction with anterior costal cartilage grafting is a relatively safe and effective method that provides an alternative treatment for idiopathic laryngotracheal stenosis. Removal of scar tissue and repair of the defect by free skin graft, plus a long period of T-tube stenting, may be important for achieving decannulation. Given the small number of patients in this series, a larger study would be needed to further support these conclusions.
