Introduction
In paediatric patients, subglottic stenosis is a leading cause of airway obstruction, and it is divided into congenital and acquired categories. Congenital subglottic stenosis is uncommon and usually stems from improper recanalisation of the laryngeal lumen, while acquired subglottic stenosis represents nearly 90 per cent of cases.Reference Rodríguez, Cuestas, Botto, Cocciaglia, Nieto and Zanetta1, Reference Marston and White2 Acquired types can be caused by various factors,Reference Albrecht and Ostrower3 with prolonged endotracheal intubation being the most common.Reference Alshammari, Arafat and Halawani4 The clinical presentation usually involves varying degrees of persistent coughing, stridor, dyspnoea, wheezing and changes in voice quality.
Endoscopic or open surgical procedures can effectively treat subglottic stenosis, but open surgery is typically reserved for more severe cases. Endoscopic techniques, which are less invasive and preferred for less severe lesions, generally lead to quicker recovery and healing than open surgery.Reference Gelbard, Anderson, Berry, Amin, Benninger and Blumin5, Reference Feinstein, Goel, Raghavan, Long, Chhetri and Berke6 Procedures such as cricotracheal resection and laryngotracheoplasty with cartilaginous grafting, which are types of open neck surgery to treat subglottic stenosis, are lengthy operations and have a risk of failure. However, these procedures result in fewer follow-up interventions and provide longer-lasting airway patency.Reference Gelbard, Anderson, Berry, Amin, Benninger and Blumin5, Reference Gelbard, Donovan, Ongkasuwan, Nouraei, Sandhu and Benninger7, Reference Carpenter, Pierce and Smith8 Endoscopic management includes two main approaches: dilatation and scar resection without dilatation.Reference Gelbard, Anderson, Berry, Amin, Benninger and Blumin5, Reference Gelbard, Donovan, Ongkasuwan, Nouraei, Sandhu and Benninger7, Reference Maldonado, Loiselle, Depew, Edell, Ekbom and Malinchoc9 Scar resection or radial incisions are carried out with a carbon dioxide (CO2) laser or a cold knife, while dilatation techniques utilise continuous radial expansion balloons and rigid dilators. Although these methods are less invasive than open-neck surgical procedures, they often lead to recurrent stenosis that requires further interventions.Reference Gelbard, Anderson, Berry, Amin, Benninger and Blumin5, Reference Gelbard, Donovan, Ongkasuwan, Nouraei, Sandhu and Benninger7, Reference Benjamin, Jacobson and Eckstein10 Nevertheless, they generally provide better outcomes regarding swallowing function, voice quality and recovery time compared with open procedures.Reference Dwyer, Qiabi, Fortin, Inculet, Nichols and MacNeil11
Different outcomes regarding endoscopic management of subglottic stenosis in children have been reported in the literature. Our study sought to thoroughly assess the existing evidence on endoscopic treatment approaches for paediatric subglottic stenosis.
Methods
The methodology used in the study was aligned with the Cochrane Handbook guidelines and adhered to Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) recommendations.Reference Higgins, Thomas, Chandler, Cumpston, Li and Page12, Reference Page, McKenzie, Bossuyt, Boutron, Hoffmann and Mulrow13
Search process
The search covered the Medline/PubMed, Scopus, Web of Science and Cochrane Library databases, encompassing all available records from their inception through August 2024. We employed a comprehensive search strategy that combined keywords related to endoscopic procedures such as laser, dilatation and cold knife with terms specific to subglottic stenosis. The detailed search strategy was as follows: (‘subglottic stenosis’ or ‘proximal tracheal stenosis’ or ‘idiopathic subglottic stenosis’) and (endoscopic or ‘endoscopic balloon dilatation’ or ‘balloon dilatation’ or ‘cold knife’ or laser or endoscope or ‘rigid dilatation’ or ‘CO2 laser’ or ‘YAG laser’).
Eligibility criteria and outcome definitions
We included studies that evaluated any endoscopic intervention for managing subglottic stenosis in children, without restrictions on the type of endoscope used. The primary outcome was the success rate, defined as improvement of stenosis evidenced by relief of stridor, avoidance of tracheotomy, immediate clinical and symptomatic improvement, decannulation, relief from dyspnoea or sustained airway patency without additional interventions. Secondary endpoints included recurrence rate, defined by the need for more than one procedure during follow up, failure to respond to dilatation, recurrence of stenosis and dyspnoea requiring further surgery. Additional secondary outcomes included the total number of interventions per patient during follow up, decannulation rate and complications. Literature reviews, conference abstracts, non-peer-reviewed studies, studies including adults, case reports and those lacking data on endoscopic procedures were excluded.
Screening procedure
Duplicates were eliminated using EndNote software (Clarivate Analytics, PA, USA). References were then reviewed in two stages: initial screening of titles and abstracts, followed by full-text review of selected articles to determine eligibility. The Rayyan web platformReference Ouzzani, Hammady, Fedorowicz and Elmagarmid14 was used to assist in the selection process.
Data gathering
A uniform data extraction template was used to systematically collect information from each included study. This encompassed bibliographic data (author and publication year), study design, geographical setting, mean age of participants, number of males, total number of participants, type of endoscopy, adjunctive drug use, inclusion criteria related to subglottic stenosis, outcomes measured and key findings.
Quality assessment
The Newcastle–Ottawa Scale was used to evaluate the quality of observational studies.Reference Wells, Shea, O’Connell, Peterson, Welch and Losos15 The Newcastle–Ottawa Scale assesses key domains including cohort selection and outcome measurement. Owing to the predominantly single-arm observational design of the included studies, the comparability domain was not applicable. Quality ratings were therefore categorised as follows: scores of 1–2 indicated poor quality, 3–4 indicated fair quality and 5–6 indicated good quality.
Data analysis
Dichotomous data were combined into proportion estimates with 95 per cent confidence intervals (CIs). Continuous outcomes were reported as means with 95 per cent CI. A random-effects model was used to pool results. A p value below 0.05 was considered statistically significant. Heterogeneity was considered significant if the chi-square p value was below 0.05 or if the I Reference Marston and White2 value exceeded 50 per cent. All statistical analyses were performed using OpenMeta-Analyst software.Reference Wallace, Dahabreh, Trikalinos, Lau, Trow and Schmid16
Results
Literature search results
The initial literature search identified 2685 records. Using EndNote to remove duplicates, we narrowed the dataset to 1835 records for screening. Through comprehensive full-text review, we identified 43 studiesReference Alshammari, Arafat and Halawani4, Reference Hu, He, He and Jiao17–Reference Koufman, Thompson and Kohut58 that met the inclusion criteria. Of these, 38 studies were included in the meta-analysis, while five were incorporated solely into the systematic review.Reference Yafit, Cavel, Ungar, Wasserzug, Oestreicher-Kedem and Shklovsky-Kordi26, Reference Lee, Irace and Rahbar31, Reference Bitar, Al Barazi and Barakeh32, Reference Depierraz, Ravussin, Brossard and Monnier53, Reference Hebra, Powell, Smith and Othersen56 The PRISMA flow diagram summarises the study selection process (Figure 1).
Preferred Reporting Items for Systematic Reviews and Meta-Analysis flow chart of study selection.
Study characteristics
The 43 included studiesReference Alshammari, Arafat and Halawani4, Reference Hu, He, He and Jiao17–Reference Koufman, Thompson and Kohut58 involved a total of 1088 patients: 28 retrospective cohorts, 6 prospective cohorts, 8 retrospective case series and 1 prospective case series. The studies were conducted in several geographical regions, including China, India, Brazil, Italy, Argentina, Turkey, the UK and the USA. Multiple endoscopic procedures were performed across the studies, involving laser techniques (CO2 and potassium titanyl phosphate (KTP) and dilatation techniques (rigid, balloon and cold knife). The main adjunctive drugs used during endoscopic procedures were mitomycin C and corticosteroids. The mean follow-up duration ranged from 2 to 50.5 months. The study populations comprised only children, with mean ages ranging from 2 to 110 months. A detailed summary of included studies is presented in Table 1.
Summary and baseline characteristics of included studies
SD = standard deviation; SGS = subglottic stenosis; PPI = ; EBD = endoscopic balloon dilatation; CO2 = carbon dioxide; NSAID = non-steroidal anti-inflammatory drugs; MLB = microlaryngobronchoscopy; NR = not reported; LTP = laryngotracheoplasty; KTP = potassium titanyl phosphate
Quality assessment
Of the 43 observational studies evaluated using the Newcastle–Ottawa Scale,Reference Wells, Shea, O’Connell, Peterson, Welch and Losos15 15 were classified as good quality, 18 as fair quality and 10 as poor quality. The primary concerns in lower-quality studies related to representativeness of the exposed cohort and adequacy of follow up. Detailed information on the risk of bias for each observational study is provided in Supplementary Table 1.
Study endpoints
Success rate
Overall, 560 of 713 patients experienced successful endoscopic procedures, with a pooled success rate of 79.2 per cent (95 per cent CI 73.7–84.7 per cent, p < 0.001). The pooled studies showed significant heterogeneity (I Reference Marston and White2 = 84.11 per cent, p < 0.001) (Figure 2).
Forest plot of pooled success rate of endoscopic procedures.
By specific procedure, CO2 laser achieved an estimated success rate of 79.9 per cent (6 studies), KTP laser achieved 100 per cent (1 study), balloon dilatation achieved 79.9 per cent (17 studies), rigid dilatation achieved 82.4 per cent (5 studies) and cold knife achieved 47.2 per cent (2 studies). Combining cold knife with balloon dilatation resulted in a success rate of 83.3 per cent (1 study), while combined CO2 laser and rigid dilatation showed a success rate of 82.8 per cent (1 study) (Figure 2).
Recurrence rate
Recurrence was reported in 310 of 699 patients following endoscopic procedures, with a pooled recurrence rate of 36.8 per cent (95 per cent CI 22–50 per cent, p < 0.001). Pooled studies exhibited significant heterogeneity (I Reference Marston and White2 = 95.36 per cent, p < 0.001) (Figure 3).
Forest plot of pooled recurrence rate after endoscopic procedures.
By procedure, CO2 laser was associated with a pooled recurrence rate of 45.3 per cent (3 studies), rigid dilatation with 20.7 per cent (6 studies), balloon dilatation with 42.2 per cent (13 studies) and cold knife with 30 per cent (2 studies). Combined CO2 laser and balloon dilatation demonstrated a recurrence rate of 71.4 per cent (1 study), combined cold knife and balloon dilatation demonstrated a recurrence rate of 16.7 per cent (1 study) and combined CO2 laser and rigid dilatation demonstrated a recurrence rate of 17.2 per cent (1 study) (Figure 3).
Decannulation rate
Decannulation occurred in 171 of 245 tracheostomised patients, with a pooled decannulation rate of 60.4 per cent (95 per cent CI 47.7–73.1 per cent, p < 0.001). Heterogeneity was significant (I Reference Marston and White2 = 92.18 per cent, p < 0.001) (Figure 4).
Forest plot of pooled decannulation rate after endoscopic treatment.
By endoscopic procedure, balloon dilatation had a decannulation rate of 60 per cent (7 studies), CO2 laser had a decannulation rate of 65.9 per cent (4 studies), rigid dilatation had a decannulation rate of 53 per cent (3 studies), cold knife had a decannulation rate of 50 per cent (4 studies) and KTP laser had a decannulation rate of 100 per cent (1 study) (Figure 4).
Complications
Only 27 of 537 patients experienced complications, yielding a pooled complication rate of 3.02 per cent (95 per cent CI 1.6–4.9 per cent, p = 0.008). Heterogeneity was low (I Reference Marston and White2 = 19.9 per cent, p = 0.207) (Figure 5).
Forest plot of complication rates following endoscopic procedures.
By procedure, balloon dilatation had a complication rate of 3.9 per cent (15 studies) and rigid dilatation had a complication rate of 2.8 per cent (3 studies) (Figure 5). In one study using CO2 laser alone, no complications were reported during follow up.Reference Mirabile, Serio, Baggi and Couloigner48
Total number of interventions
The pooled analysis from 10 studies showed a mean of 2.15 interventions per patient during follow up (95 per cent CI 1.84–2.46, p < 0.001), with significant heterogeneity (I Reference Marston and White2 = 82.56 per cent, p < 0.001) (Figure 6).
Forest plot of total number of interventions per patient during follow up.
By procedure, balloon dilatation was associated with a mean of 2.07 interventions and rigid dilatation with 2.5 interventions over follow up (Figure 6).
Discussion
This systematic review and meta-analysis evaluated the efficacy and safety of various endoscopic procedures for managing subglottic stenosis in children and included 1088 patients from 43 studies. The pooled success rate for endoscopic treatment was 79.2 per cent. CO2 laser achieved a success rate of 79.9 per cent across 6 studies, while KTP laser achieved 100 per cent success in a single study. Among dilatation techniques, rigid dilatation had a success rate of 82.4 per cent (5 studies) and balloon dilatation had a success rate of 79.9 per cent (17 studies). By contrast, cold knife procedures alone had a lower success rate of 47.2 per cent (2 studies).
Subglottic stenosis is difficult to manage regardless of aetiology, and traditional approaches often involve open surgical interventions such as laryngotracheal reconstruction or posterior cricoid split with tracheal resection.Reference Cotton59, Reference Sandu and Monnier60 Although open procedures achieve high decannulation rates,Reference Monnier, George, Monod and Lang61 they carry the risk of significant complications.Reference Agrawal, Black and Morrison62 Endoscopic procedures avoid many of these complications but show variable outcomes in the literature.Reference Hirshoren and Eliashar63, Reference Myer, O’Connor and Cotton64 Endoscopic management includes CO2 laser, KTP laser, rigid dilatation, balloon dilatation, cold knife procedures and combined laser–dilatation techniques.Reference Alshammari, Arafat and Halawani4, Reference Hu, He, He and Jiao17–Reference Koufman, Thompson and Kohut58 There is no universally accepted consensus on optimal management and a definitive algorithm has not been established. This study represents the most comprehensive and up-to-date evaluation of the efficacy and safety of available endoscopic treatments for paediatric subglottic stenosis.
Our pooled analysis showed that 560 of 713 patients had successful endoscopic procedures, with a cumulative success rate of 79.2 per cent. Balloon dilatation had a success rate of 79.9 per cent across 17 studies. In the series by Maresh et al.,Reference Maresh, Preciado, O’Connell and Zalzal37 94 endoscopic balloon dilatations were performed in 44 patients, with 47 endoscopic balloon dilatation procedures delivered as primary treatment in 27 patients; 14 of these were managed successfully with endoscopic balloon dilatation alone. Outcomes were influenced by stenosis characteristics, particularly severity as assessed by the Cotton–Myer grading system.Reference Cotton59 Among patients successfully treated with endoscopic balloon dilatation, 10 had Cotton–Myer grade II and 4 had grade III subglottic stenosis. Despite initial success, 13 patients required laryngotracheoplasty owing to persistent symptoms after endoscopic balloon dilatation. All patients in the endoscopic balloon dilatation failure group had grade III stenosis.
Successful endoscopic balloon dilatation was more frequent in patients with grade I or II stenosis (n = 10) than in those with severe (grade III or IV) stenosis (n = 4). No patients with grade I or II stenosis were in the endoscopic balloon dilatation failure group, whereas all patients with grade III or IV stenosis experienced endoscopic balloon dilatation failure (p < 0.001).Reference Maresh, Preciado, O’Connell and Zalzal37 None of the patients with associated syndromes, congenital subglottic stenosis or multilevel airway disease had successful initial endoscopic balloon dilatation.Reference Maresh, Preciado, O’Connell and Zalzal37 Failed endoscopic balloon dilatation was more common in preterm in-patients (62 per cent vs 43 per cent) and in patients with more co-morbidities (85 per cent vs 43 per cent) compared with those who had successful outcomes.Reference Maresh, Preciado, O’Connell and Zalzal37 However, severe stenosis was the only independent predictor of failure in multivariate analysis. Earlier studies have similarly suggested that endoscopic balloon dilatation is more effective for lower-grade stenosis.Reference Hu, He, He and Jiao17, Reference Lang and Brietzke65
The success of endoscopic balloon dilatation may also differ between acute and chronic subglottic stenosis. Hu et al.Reference Hu, He, He and Jiao17 reported an endoscopic balloon dilatation success rate of 86.7 per cent in acute lesions and 61.1 per cent in chronic lesions, although the difference was not statistically significant. Maunsell and AvelinoReference Maunsell and Avelino38 found endoscopic balloon dilatation to be completely successful (100 per cent) in acute subglottic stenosis, but the success rate for chronic subglottic stenosis was only 32 per cent. Some studies have suggested that outcomes are less favourable in chronic cases.Reference Durden and Sobol51 Maunsell and AvelinoReference Maunsell and Avelino38 proposed that the duration of stenosis and scar characteristics may influence outcomes. Whigham et al.Reference Whigham, Howell, Choi, Peña, Zalzal and Preciado43 reported that endoscopic balloon dilatation failures were related to co-morbid airway conditions but did not find associations with stenosis grade, age or scar characteristics. In contrast, Maunsell and AvelinoReference Maunsell and Avelino38 observed better outcomes in younger patients, those with milder stenosis and those with thin or laminar scars. Whigham et al.Reference Whigham, Howell, Choi, Peña, Zalzal and Preciado43 also found that absence of a prior tracheostomy was associated with procedural success, possibly reflecting better overall health in these children. Predictors of favourable outcomes therefore include acute stenosis, younger age at dilatation, absence of prior tracheostomy and lower-grade stenosis.Reference Maunsell and Avelino38 Early intervention may also reduce the need for multiple procedures.Reference Romero Manteola, Patiño González, Ravetta, Defago and Tessi19, Reference Tuzuner, Bas, Jafarov, Bahcecitapar and Aydin66
Our study showed that CO2 laser had a success rate of 79.9 per cent across 6 studies. Pulsed CO2 laser technology allows rapid tissue vaporisation while limiting thermal damage, scarring and charring.Reference Monnier and Monnier67, Reference Moseley, Oswal, Oswal and Remacle68 Hseu et al.Reference Hseu, Benninger, Haffey and Lorenz69 reported that CO2 laser surgery was not associated with worse outcomes or increased dilatation requirements, supporting its broader use for radial incisions and excision of stenotic tissue. Alshammari et al.Reference Alshammari, Arafat and Halawani4 demonstrated successful endoscopic removal of congenital subglottic stenosis using CO2 laser alone, avoiding mechanical dilatation and achieving a 75 per cent success rate (6 of 8 patients) in high-grade circumferential subglottic stenosis.
Rigid dilatation achieved an 82.4 per cent success rate across 5 studies. Although balloon dilatation has been widely reported,Reference Hu, He, He and Jiao17, Reference Romero Manteola, Patiño González, Ravetta, Defago and Tessi19–Reference Şahin S, Yılmaz A, Ö and Yılmaz21 fewer data exist for rigid or semi-rigid dilators.Reference Bakthavachalam and McClay49, Reference Maksoud-Filho, Gonçalves, Cardoso and Tannuri50 Balloon dilatation success rates range between 60 and 70 per cent,Reference Whigham, Howell, Choi, Peña, Zalzal and Preciado43, Reference Durden and Sobol51 whereas rigid dilatation has reported success rates up to 100 per cent in selected series.Reference Whigham, Howell, Choi, Peña, Zalzal and Preciado43, Reference Durden and Sobol51 Higher success rates in rigid dilatation cohorts may reflect a greater proportion of grade I stenosis and repeated interventions.Reference Maksoud-Filho, Gonçalves, Cardoso and Tannuri50 In the series by Romero Manteola et al.,Reference Romero Manteola, Patiño González, Ravetta, Defago and Tessi19 rigid dilatation produced favourable results in 82 per cent of patients with grades I–III subglottic stenosis, and age and stenosis grade were not significant predictors of success. However, longer intubation duration was associated with failure, possibly because prolonged intubation results in more extensive and deeper lesions and more pronounced cartilaginous changes.Reference Romero Manteola, Patiño González, Ravetta, Defago and Tessi19 Differences in success rates across studies may also reflect varying definitions of success.
Recurrence occurred in 310 of 699 patients, with a pooled recurrence rate of 36 per cent. Balloon dilatation had a recurrence rate of 42 per cent. As noted, stenosis severity is a key determinant of failure and recurrence. Maunsell and AvelinoReference Maunsell and Avelino38 reported endoscopic balloon dilatation failure in 25 of 37 patients and linked this to more severe stenosis. Triglia et al.Reference Triglia, Nicollas and Roman52 observed that airway widening and decannulation were achieved in all but one case, a child with grade IV subglottic stenosis. Whigham et al.Reference Whigham, Howell, Choi, Peña, Zalzal and Preciado43 suggested that failure of balloon dilatation might be related to pre-existing airway co-morbidities. Other studies have associated poor outcomes with scar contracture, severe infections and higher stenosis grade.Reference Avelino, Maunsell and Jubé Wastowski34 Recurrence rates of 40–70 per cent have been reported with endoscopic management.Reference Hseu, Benninger, Haffey and Lorenz69 Triglia et al.Reference Triglia, Nicollas and Roman52 recommended limiting laser resection to grade I subglottic stenosis less than 5 mm in length; in their series, post-operative improvement decreased from 92 per cent in grade I stenosis to 46 per cent in grade II and 13 per cent in grade III disease. Endoscopic approaches thus perform less well than open surgery in more severe subglottic stenosis.Reference Monnier and Monnier67
In our pooled analysis, rigid dilatation had a recurrence rate of 20 per cent. Romero Manteola et al.Reference Romero Manteola, Patiño González, Ravetta, Defago and Tessi19 identified prolonged intubation and prior tracheostomy as predictors of poor outcome. More than half of patients who had tracheostomy before endoscopic treatment did not respond.Reference Romero Manteola, Patiño González, Ravetta, Defago and Tessi19 Although some earlier studies linked co-morbidities to poorer outcomes,Reference Whigham, Howell, Choi, Peña, Zalzal and Preciado43 Romero Manteola et al.Reference Romero Manteola, Patiño González, Ravetta, Defago and Tessi19 observed that patients who failed treatment actually had fewer co-morbidities and the difference was not statistically significant. The presence of additional airway pathology should therefore not be considered an absolute contraindication to dilatation.
Our pooled analysis showed a decannulation rate of 60.4 per cent after endoscopic treatment, with balloon dilatation achieving similar rates. CO2 laser treatment has been associated with even higher decannulation rates. Triglia et al.Reference Triglia, Nicollas and Roman52 reported a 97 per cent decannulation rate using CO2 laser, with successful airway widening in almost all patients.
Complication rates were low. Our pooled complication rate was 3.0 per cent. Choo et al.Reference Choo, Tan and Balakrishnan47 reported three complications after rigid dilatation, including two deaths from non-subglottic stenosis causes (aspiration pneumonia and bronchopneumonia) and one death related to failure of decannulation. Hautefort et al.Reference Hautefort, Teissier, Viala and Van Den Abbeele42 reported five deaths among 44 children treated with balloon dilatation, all attributed to severe underlying conditions rather than airway complications.
The pooled mean number of interventions was 2.15 per patient. Maresh et al.Reference Maresh, Preciado, O’Connell and Zalzal37 reported that among children in whom endoscopic balloon dilatation prevented laryngotracheoplasty, the number of interventions ranged from one to five, with a mean of 1.9 dilatations per patient.
This is the first meta-analysis to evaluate the efficacy and safety of different endoscopic treatments for paediatric subglottic stenosis, thereby addressing a significant gap in the literature. Strengths of the study include its comprehensive and up-to-date nature, the inclusion of studies from diverse geographical regions and detailed subgroup analyses of individual procedures. However, limitations exist. First, the observational design of the included studies may introduce selection bias. Second, definitions of success and recurrence varied across studies, contributing to heterogeneity. Third, many outcomes showed substantial statistical heterogeneity, likely reflecting differences in patient characteristics, endoscopic techniques, outcome definitions, disease aetiology (idiopathic vs acquired) and stenosis severity. Finally, some procedures, particularly combined approaches, were reported in only one or two studies, limiting generalisability.
Overall, our findings provide robust evidence that several endoscopic techniques can effectively manage paediatric subglottic stenosis, but recurrence remains common and multiple interventions are often required. Treatment decisions should therefore be individualised based on patient factors, stenosis characteristics and local expertise, ideally within a multidisciplinary airway team.
• Subglottic stenosis is a major cause of paediatric airway obstruction and can be managed with endoscopic or open surgical approaches.
• Endoscopic techniques, including laser and dilatation procedures, are less invasive but show variable success and recurrence rates
• There is no consensus on the optimal endoscopic strategy for paediatric subglottic stenosis
• This is the most comprehensive and up-to-date systematic review and meta-analysis of endoscopic treatments for paediatric subglottic stenosis, including 43 studies and 1088 children
• Combined and modality-specific success and recurrence rates are quantified for carbon dioxide laser, potassium titanyl phosphate laser, balloon dilatation, rigid dilatation, cold knife and combined techniques
• The review highlights that, although endoscopic interventions achieve high initial success, recurrence is common, underscoring the need for careful patient selection, long-term follow up and individualised treatment planning
Conclusion
This systematic review and meta-analysis highlight the complexity of managing subglottic stenosis in children. Endoscopic treatments such as CO2 laser and balloon dilatation demonstrate high initial success rates and favourable decannulation outcomes, but notable recurrence rates and the need for repeated procedures temper their overall effectiveness. Nevertheless, endoscopic approaches remain valuable, less invasive options for many children with subglottic stenosis. Long-term follow up, careful patient selection and standardised outcome definitions are essential, and future prospective studies comparing different endoscopic strategies are needed to refine management algorithms.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S0022215125104179.
Data availability statement
The datasets were generated from open-source databases. All data are available within the manuscript and its supplemental files.
Competing interests
None declared
