Using a customizable L stent for laryngotracheal stenosis: Patterns of success and failure

Abstract

Background

A recent innovation in the management of tracheostomy‐dependent patients with laryngotracheal stenosis (LTS) has been the customizable laryngotracheal stent, made by dividing a T‐tube in half (“L stent”). These L stents can be placed into the subglottis as well as the posterior glottis to prevent rapid re‐stenosis following surgery, and they are easily removed in the office. We sought to review our experience with these stents and analyze patterns of success and failure.

Methods

Retrospective chart review was performed of all tracheostomy‐dependent patients who underwent L stent placement from January 2019 through February 2024. Clinical and demographic data were reviewed, with decannulation as the primary endpoint. Patients were assessed as having comorbid conditions (diabetes, coronary artery disease, morbid obesity, prior radiation) or not.

Results

A total of 16 patients underwent stent placement. Six patients had posterior glottic stenosis (PGS), six had subglottic stenosis (SGS), and four had multilevel stenosis. Of these 16 patients, 9 were ultimately decannulated (56.2%). Three of the successes were PGS (33.3%), four were SGS (44.4%), and two had both SGS and PGS (22.2%). Patients who were decannulated were statistically less likely to have comorbid conditions than patients who failed decannulation (p = 0.0350).

Conclusion

For tracheostomy‐dependent patients who have failed first‐line therapy, endoscopic surgery paired with L stent placement offers a reasonable alternative to open airway reconstruction or more destructive arytenoidectomy or cordotomy procedures. As with most airway surgeries, patient selection is important, as the presence of multilevel stenosis and/or comorbid conditions can predict a higher failure rate.

INTRODUCTION

Adult laryngotracheal stenosis (LTS) is a complex disorder with various etiologies, including prolonged intubation, prior tracheostomy or other surgeries, idiopathic stenosis, autoimmune disease, neoplasm, and trauma. Therapeutic management can be broadly classified into endoscopic and open airway procedures, and while many patients respond well to endoscopic management alone, some cases of recalcitrant disease do not. Typically, endoscopic procedures are exhausted before open airway procedures such as cricotracheal resection (CTR) and tracheal resection (TR) due to the increased surgical risk of open procedures. For this reason, adjuvant treatments such as serial in‐office intralesional steroid injections (SILSI) have been developed to increase surgical intervals and induce stabilization or regression of stenosis. ¹ , ² , ³ , ⁴

While SILSI has improved outcomes for many patients, some tracheostomy‐dependent cases of LTS are still resistant to dilation, so alternative surgical approaches are continually being sought. T‐tubes, so named due to their distinctive shape, have long been used for management of LTS and can provide some unique benefits as alternatives to traditional tracheostomy tubes, such as continuous capping, softer silicone material, and stenting a stenotic segment on a semipermanent basis. ⁵ , ⁶ These T‐tubes can also be modified by bisecting them in half symmetrically— creating the “L stent.” The L stent can then be placed into the subglottis, and optionally the posterior glottis, for a prolonged period in tracheotomized patients, maintaining a patent airway and preventing rapid re‐stenosis following dilation. Unlike traditional metallic stents, these L stents can be easily removed in the office without anesthesia.

We sought to review our outcomes and complications with L stents in the management of LTS. We hypothesized that patients' medical comorbidities affecting wound healing, such as diabetes mellitus, coronary artery disease, and obesity, would have a lower rate of decannulation than patients without comorbidities.

METHODS

This research was approved by the University of Miami Human Subjects Research Office Internal Review Board (#20180634). The electronic medical records at the University of Miami Health System were reviewed between January 2019 to February 2024. Included subjects were patients 18 years or older who underwent L stent placement with prior tracheotomy at the time of surgery. All patients had undergone at least one prior endoscopic dilation that did not lead to decannulation. The degree and type of stenosis was diagnosed endoscopically at the time of surgery. Patients with SGS with both membranous scar as well as cartilaginous framework deformities were included in the sample. After a case was identified, demographic and comorbidity information of the patient including age, sex, body mass index (BMI), etiology of stenosis, and medical comorbidities including diabetes mellitus (DM), end‐stage renal disease (ESRD), coronary artery disease, recurrent respiratory papillomatosis, morbid obesity (MO), and prior radiation to the neck were included. The type of LTS was also recorded as posterior glottic stenosis (PGS), subglottic stenosis (SGS), or a combination. Primary endpoint was successful tracheostomy tube decannulation.

Statistical analysis

Two‐sample t‐tests were used to compare the average time of stent removal, average age at time of surgery, and average BMI. Fisher's exact test was used to compare categorical variables between patients who were not and who were decannulated, including gender, age, airway pathology, ethnicity, race, and etiology of stenosis. Fisher's exact test was used to compare the comorbidities between both groups. A p value <0.05 was considered significant in all tests.

Surgical technique and L stent preparation

After performing radial cuts into the stenosis with either a KTP or 445‐nm blue laser under microlaryngoscopy, the airway was dilated using an airway balloon system. When the stenosis was primarily cartilaginous in nature, this was done in a fashion similar to that described by Nouraei et al. ⁷ The L stents were created by modifying either standard or long‐sized 12‐mm T‐tubes (E Benson Hood Laboratories). Standard length T‐tubes were generally used for stenting the subglottis only, while longer tubes were necessary to reach the posterior glottis. First, the T‐tube was cut in half down the middle of the external limb (Figure 1). This created two l‐shaped pieces, and the longer (distal) half was kept. The distal limb of the T‐tube was then trimmed to the appropriate length to span the stenotic area in the subglottis, which was determined during the operative rigid bronchoscopy. For cases of PGS, the posterior portion of the superior limb was also trimmed and tapered to ensure stent contact to the posterior commissure and vocal processes while minimizing contact with the membranous vocal folds, after the technique described by Zeitels et al. ⁸ (Figures 2, 3, 4, 5).

Figure 1.

Preparation of L stent from T tube.

Figure 2.

Posterior glottic scar with granulation tissue.

Figure 3.

Post laser excision of scar band, dilation, and stent placement spanning the posterior commissure.

Figure 4.

Stent placement in the subglottis.

Figure 5.

Stable airway, 8 months post stent removal.

The L stents were then inserted through the existing tracheostoma under laryngoscopic guidance. A hemostat clamp was used to insert the upper limb of the stents and they were advanced into the airway lumen atraumatically using microlaryngeal forceps to pull the stents into position, with the external limb remaining outside of the tracheostoma. Proper placement was confirmed endoscopically. No sutures were required for this procedure, as the size and right angle of the L stent kept the stent snugly in position. At the end of the procedure, once patients resumed spontaneous breathing, a new tracheostomy tube was inserted just below the stent. The stent was left in place for approximately 2 months before removal in the office. For 1 week before removal and 1 week afterwards, patients were treated with a course of oral corticosteroids to decrease inflammation and granulation tissue.

RESULTS

A total of 16 patients underwent stent placement (eight males and eight females). A second surgery was attempted in three patients after these individuals developed re‐stenosis after L stent removal. The average age of patients at the time of surgery was 49.65 years. Six patients had PGS alone, six had SGS alone, and four had multilevel stenosis (Table 1). Of 16 patients, 9 were ultimately decannulated (56.3%). The most common etiology for stenosis among both decannulated and non‐decannulated patients was prolonged intubation. The rate of comorbidities was significantly lower in the decannulated group compared to the non‐decannulated group (p = 0.0350, Table 1). In this study, no single specific medical condition or demographic was significantly likely to predict lower likelihood of decannulation.

Table 1.

Clinicodemographic features of patients who underwent a customizable L stent for laryngotracheal stenosis.

Characteristics	Decannulated (n = 9, 56.2%)	Non‐decannulated (n = 7, 43.7%)	p Value
Sex			1.0
Male	4 (44.4%)	4 (57.1%)
Female	5 (55.6%)	3 (42.9%)
Race
White	0	1 (14.3%)
African American	4 (44.4%)	4 (57.1%)
Hispanic	5 (55.6%)	2 (28.6%)
Other	0	0
Airway pathology
Posterior glottic stenosis	3 (33.3%)	3 (42.8%)
Subglottic stenosis	4 (44.4%)	2 (28.5%)
Multilevel stenosis	2 (22.2%)	2 (28.5%)
Average time of stent removal (days)	62.3	56.0
Average age of surgery (years)	54.8	44.5
Average BMI (kg/m2)	33.52	35.94	0.1904
Etiology of stenosis			0.2256
Iatrogenic	6	6	0.1969
XRT	1	0
Unknown	1	1
Other (autoimmune, idiopathic, neoplastic, congenital)	1	0
Patients
With comorbidities	1	5
Without comorbidities	8	2
Comorbidities			0.0350
Diabetes miletus	1	2
End‐stage renal disease	0	1
Coronary artery disease	0	1
Recurrent respiratory papillomatosis	0	1
Morbid obesity	0	3	0.0625
Radiation therapy to neck	0	1	0.4375
Serial intralesional steroid injections after stent removal
Yes	3 (33.3%)	2 (28.6%)
No	6 (66.6%)	5 (71.1%)
Number of complications	0	2

Abbreviations: BMI, body mass index; XRT, radiation therapy to the neck.

Among the decannulated group, five were female (56%) and four were male (44%). Fifty‐six percent were identified as Hispanic whites, while 44% were non‐Hispanic Black. Patients in this subset underwent surgery at a slightly older age compared to the non‐decannulated group, with an average age of 55 years versus 45 years, respectively, but this difference was not significant (p = 0.23). Within this successful cohort, three patients presented with PGS, four with SGS, and two with multilevel disease. Only one patient in this group had a medical comorbidity, specifically DM. The average duration for stent retention before removal in this group was 62.26 days, and no postoperative complications were reported.

In the non‐decannulated group, most patients had PGS (42.8%), while the rest had either SGS or multilevel stenosis. The average age at surgery was 44.5 years and the average time for stent removal was 56 days. Among the seven patients in this group, five had existing medical comorbidities, with MO and DM the two most common. Furthermore, there were two postoperative complications in this group. One patient developed tracheitis requiring antibiotics and early stent removal. The other patient had migration of the L stent, requiring uncomplicated operative removal. After this event, the external limb was always sutured on itself to prevent inadvertent entry into the tracheostoma, and no further stent migration events occurred. It is important to highlight that in the current L stent technique, the tracheostomy tube sits underneath the L stent as an extra safety measure. No deaths were reported. Of note, no statistically significant difference was seen between the groups in patients treated with postoperative SILSI.

DISCUSSION

The management of LTS presents a formidable challenge in upper airway surgery. With the increased frequency and duration of intubations in recent years, LTS has emerged as a significant complication, substantially impairing the quality of life for patients dependent on tracheostomy. Presently, no universally successful treatment approach exists for LTS. To our knowledge, this is the first study to examine patterns of success and failure in LTS patients undergoing endoscopic treatment with customizable L stent placement. While endoscopic dilation is commonly employed as an initial management strategy for LTS and for maintaining airway patency, it may not yield optimal outcomes for all patients, particularly those with advanced disease. ⁹ Surgical resection and anastomosis has been reported in multiple case series as offering a high success rate for definitive treatment of LTS, although with associated risks and potentially fatal complications. ¹⁰ , ¹¹ , ¹² For tracheostomy‐dependent patients, the L stent presents a viable alternative, providing an additional option between endoscopic and open resection which is generally safe, effective, and well tolerated.

In this study, we identified a risk profile distinguishing patients who may benefit from endoscopic dilation with L stent placement versus those more likely to fail, with a decannulation rate of 56.2% in our cohort. Among decannulated patients, only one had comorbid conditions compared to five in the non‐decannulated group (p = 0.0350). Although a greater proportion of patients in the non‐decannulated group had medical comorbidities, no specific condition was indicative of a lower decannulation risk. Additionally, no statistically significant differences were found between the two groups regarding age at surgery, duration of stent placement, or BMI (Table 1).

Associations between medical comorbidities such as DM, ESRD, obesity, and airway stenosis have been consistently observed in other large series, aligning with our findings. ¹³ , ¹⁴ , ¹⁵ , ¹⁶ These comorbidities were chosen for this study due to their association with poor wound healing. In a recent multicenter study assessing risk factors for complications and failures following CTR, systemic comorbidities emerged as the sole independent predictor of postsurgical complications, correlating with prolonged hospitalization and the need for additional surgeries (p = 0.043). ¹² Another study by Halum et al. demonstrated a higher likelihood of airway stenosis in obese patients intubated with large endotracheal tubes compared to those intubated with smaller tubes. ¹⁷ While our study did not yield statistically significant results regarding BMI, this was likely due to the small sample size, and we suspect the association between obesity and poor outcomes after LTS surgery likely persists.

The COVID‐19 pandemic has also exacerbated LTS cases, primarily due to prolonged intubation, increased tracheostomy rates, the utilization of large endotracheal tubes, and possibly the use of percutaneous dilatational tracheostomies. ¹⁸ , ¹⁹ These interventions could have resulted in unintended complications leading to long‐term changes in voice perception and various airway issues, such as trap‐door deformities in the trachea, injuries near the vocal folds, and multilevel airway injuries that often do not respond to the standard dilation technique. ²⁰ , ²¹ , ²² , ²³ The L stent represents a potential solution for patients with existing tracheostomy tubes who have failed endoscopic treatment or are unsuitable candidates for open airway reconstruction. Various T‐tube insertion techniques for airway stenosis have been described in previous studies. ⁵ , ⁶ , ²⁰ , ²¹ , ²² , ²⁴ Generally, T‐tubes offer adaptability, with various lengths and diameters available and the ability to be precisely modified under endoscopic guidance. In our study, only two patients experienced adverse reactions necessitating L stent removal, with no deaths or major complications reported.

Our study is subject to several limitations. Firstly, it is important to note that this investigation is preliminary in nature and is underpowered due to a small cohort of patients. Subsequent larger‐scale studies are imperative to substantiate and extend the findings presented herein. Secondly, the potential for selection bias is acknowledged, given the absence of a standardized protocol guiding patient selection for endoscopic dilation with stent placement. The decision‐making process regarding this procedure predominantly relied on clinical judgment or prior endoscopic failures. Additionally, while the Cotton–Myer grading system is widely utilized for assessing SGS, its subjective nature has been documented, and previous research has demonstrated poor reliability. Consequently, no scoring system was employed in our study to grade SGS severity. ²³ Future investigations should incorporate more objective measures, such as spirometry, to compare stenosis severity and progress across patient groups.

CONCLUSION

Endoscopic dilation with a modified L stent offers a potential option for tracheostomy‐dependent patients with more advanced LTS stenosis who have failed standard endoscopic dilation. Endoscopic dilation with L stent placement offers shorter operative time to open neck reconstruction and is associated with lower risk of postoperative complications. Patients with more advanced LTS and minimal or no comorbidities may benefit from L stent placement as a way to achieve decannulation. Additional larger studies tailored at identifying specific predisposing risk factors could help both patient and surgeons in the preoperative planning and help set realistic expectations.

AUTHOR CONTRIBUTIONS

David E. Rosow, Nicolas J. Casellas, and Etyan Keidar were involved in the conception of the study. David E. Rosow and Nicolas J. Casellas completed data collection and analyzed the data. Nicolas J. Casellas and Etyan Keidar drafted the manuscript. David E. Rosow and Mursalin M. Anis edited, reviewed, and approved the final version of the manuscript.

CONFLICT OF INTEREST STATEMENT

Professor Mursalin M. Anis is a member of the World Journal of Otorhinolaryngology‐Head & Neck Surgery (WJOHNS) editorial board and is not involved in the peer review process of this article. Other authors declare no conflict of interest.

ETHICS STATEMENT

Not applicable.

Casellas NJ, Keidar E, Anis MM, Rosow DE. Using a customizable L stent for laryngotracheal stenosis: patterns of success and failure. World J Otorhinolaryngol Head Neck Surg. 2025;11:462‐467. 10.1002/wjo2.204

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available on request from the corresponding author.