Introduction
Head and neck cancers (HNCs) represent a substantial portion of the global malignancy burden, with tumours arising from non-laryngeal sites, including the oral cavity, oropharynx, nasopharynx and salivary glands, forming a significant clinical subgroup.Reference Barsouk, Aluru, Rawla, Saginala and Barsouk1, Reference Li, Agrawal, Fakhry, Fakhry and D’Souza2 The management of these tumours, particularly when locally advanced, necessitates complex treatment modalities encompassing extensive surgical resection, sometimes followed by microvascular free flap reconstruction, and often utilising definitive or adjuvant chemoradiotherapy (CRT).Reference Li, Agrawal, Fakhry, Fakhry and D’Souza2 The critical therapeutic challenge lies in achieving complete oncologic clearance while simultaneously preserving vital functions, foremost among them the integrity of the airway.Reference Li, Agrawal, Fakhry, Fakhry and D’Souza2
Airway compromise in this patient population is multifactorial. Crucially, it may manifest acutely due to the sheer mechanical obstruction caused by the primary tumour bulk, or peri-operatively resulting from anticipated soft tissue oedema or haematoma following extensive surgical procedures.Reference Barsouk, Aluru, Rawla, Saginala and Barsouk1 Furthermore, a significant subset of complications arises long after treatment completion as delayed sequelae of high-dose radiation therapy (RT), leading to fibrosis and stenosis.Reference Akazawa, Shimizu, Fujino and Kato3 Given these complexities, surgical tracheostomy (ST), defined as the creation of an opening into the trachea, remains a fundamental procedure in the multidisciplinary management of non-laryngeal HNC patients.Reference Li, Agrawal, Fakhry, Fakhry and D’Souza2
Historically, prophylactic tracheostomy was considered the standard of care for virtually all complex head and neck resections, particularly those requiring microvascular free flap reconstruction of the upper aerodigestive tract.Reference Kuo, Lin and Hsieh4 However, contemporary evidence and evolving surgical practices have driven a significant paradigm shift toward selective tracheostomy.Reference Mohamedbhai, Ali, Dimasi and Kalavrezos5 This shift is rooted in the recognition that tracheostomy is an invasive intervention associated with inherent morbidity, including prolonged hospitalisation, delayed initiation of oral intake, increased risk of infection and negative impacts on patient quality of life (QoL).Reference Kuo, Lin and Hsieh4, Reference Mohamedbhai, Ali, Dimasi and Kalavrezos5 The aim is now to employ objective criteria to accurately stratify patients and minimise the performance of procedures where the downsides outweigh the prophylactic benefits.Reference Kuo, Lin and Hsieh4
The relationship between non-laryngeal HNCs and tracheostomy is increasingly being questioned, with various scoring systems proposed for risk stratification.Reference Mohamedbhai, Ali, Dimasi and Kalavrezos5
This retrospective study aimed to analyse the experience of a tertiary oncology centre by describing the clinical characteristics, oncological treatments, indications, timing and underlying airway mechanisms of surgical tracheostomy in patients with non-laryngeal HNCs. Given the heterogeneity of tumour subsites and the limited cohort size, the study was designed to provide an exploratory, hypothesis-generating temporal–mechanistic perspective rather than a validated predictive model.
Material and methods
Study design and patient selection
All procedures performed in this study were in accordance with the ethical standards of the institutional and national research committee and with the Declaration of Helsinki (1975, revised 2008). Ethical approval for this retrospective study was obtained from the Institutional Ethics Committee (approval number: 2022-09/147; date: 22 September 2022).
This was a retrospective, observational study conducted at the Department of Otorhinolaryngology and Head and Neck Surgery of a tertiary health centre specialising in oncological cases. For this study, approval was obtained from the institutional ethics committee. Data for all cases who underwent surgical tracheostomy by the surgical team between January 2020 and July 2022 were obtained from the hospital information management system.
Cases were included if the intervention was related to HNC. Patients diagnosed with primary laryngeal and hypopharyngeal cancer were systematically excluded from the study.
Data collection and statistical analysis
The following data points were collected retrospectively for each case: age, gender, American Society of Anesthesiologists (ASA) classification score, oncological diagnosis, oncological treatment (surgery, radiotherapy or chemotherapy/chemoradiotherapy), temporal relationship of tracheostomy with the applied treatment, tracheostomy indication, emergency or planned intervention and the anatomical/physiological change requiring surgical tracheostomy. The timing of tracheostomy was categorised into four groups to standardise temporal analysis: pre-treatment, during treatment, early post-treatment (<30 days after completion of treatment) and late (>30 days after treatment).
Tracheostomy indications were further classified according to the predominant underlying airway mechanism documented in the clinical record: (1) tumour-related airway obstruction, (2) treatment-related oedema, (3) bleeding-related airway compromise, (4) fibrosis-related airway limitation and (5) functional airway impairment, including vocal cord paralysis or glottic stenosis. In patients with overlapping mechanisms, such as concurrent oedema, tumour burden, fibrosis or glottic dysfunction, classification was based on the mechanism judged to be the principal driver of airway compromise at the time of tracheostomy. For subgroup analysis, tumour-related obstruction, treatment-related oedema and bleeding-related airway compromise were considered acute mechanisms, whereas fibrosis-related airway limitation and functional glottic impairment were considered chronic mechanisms.
Variables are expressed as numbers (n) and percentages (%). Descriptive statistics are expressed as mean (± standard deviation), median (minimum–maximum) and percentage values. The relationship among diagnosis, treatment type, T stage, N stage and timing of tracheostomy was analysed using the chi-squared test. The statistical significance level was accepted as p less than 0.05. Additionally, subgroup analyses were performed based on anatomical tumour location, treatment modality and timing categories to explore potential clinical patterns. Given the limited sample size, analyses were interpreted primarily in terms of clinical trends rather than definitive statistical associations.
Results and analysis
Demographics and tumour characteristics
A total of 83 patients with HNC underwent surgical tracheostomy during the study period, of whom 29 required the procedure for non-laryngeal HNC. The mean age was 55.9 (±10.0) years, with a slight female predominance.
Nasopharyngeal cancer was the most common single subsite (n = 6; 20.7 per cent), whereas oral cavity tumours constituted the largest anatomical group when tongue, buccal mucosa, floor of mouth, lip and gingival arch tumours were considered together (n = 12; 41.4 per cent) (Table 1).
Distribution of the primary tumour according to anatomical subsites in surgical tracheostomy cases associated with non-laryngeal head and neck cancer
Table 1 Long description
Tumour localisation is listed with counts, n, and percentages, percent, for 29 cases. Tongue and thyroid each have 4 cases, 13.8 percent. Buccal mucosa has 3 cases, 10.3 percent. Floor of mouth, tongue root, lip, and oropharyngeal wall each have 2 cases, 6.9 percent. Gingival arch, neck malignant mesenchymal tumour, parotid gland, tonsil, and total each have 1 case, 3.4 percent. The total row reports 29 cases and 100.0 percent.
Airway-related clinical outcomes and RT-related parameters
Airway-related clinical outcomes and RT-related parameters are summarised in Table 2. At the last follow-up, 10 patients (34.5 per cent) had been decannulated, whereas 19 patients (65.5 per cent) remained tracheostomy-dependent or intubated. Long-term tracheostomy dependence, defined as dependence lasting more than six months, was observed in 19 patients (65.5 per cent). Intensive care unit (ICU) admission was recorded in 18 patients (62.1 per cent), and the median hospital stay was 10 days, with a range of 1–72 days.
Airway-related clinical outcomes and radiotherapy-related parameters after surgical tracheostomy
Table 2 Long description
Outcome variables are reported with columns for n and percent, covering airway status, duration of tracheostomy dependence, ICU admission, hospital stay, additional procedures, complications, emergency intervention, and radiotherapy parameters. Estimated duration of tracheostomy dependence was under 3 months for 5 patients, 17.2 percent, 3 to 6 months for 5 patients, 17.2 percent, and over 6 months or long term dependence for 19 patients, 65.5 percent. ICU admission occurred in 18 patients, 62.1 percent, and no ICU admission in 11 patients, 37.9 percent. Median hospital stay was 10 days, with a range from 1 to 72 days. Additional airway related procedures were required for 5 patients, 17.2 percent, and not required for 24 patients, 82.8 percent. Tracheostomy related complications were recorded for 2 patients, 6.9 percent, and not recorded for 27 patients, 93.1 percent. Emergency airway intervention was required for 7 patients, 24.1 percent, and not required for 22 patients, 75.9 percent. Radiotherapy dose had a median of 66 Gy and a range from 54 to 70 Gy. Radiotherapy treatment field or target region counts and percentages were: tumour bed or surgical bed only 3, 10.3 percent; tumour bed or surgical bed with ipsilateral neck lymphatics 9, 31.1 percent; nasopharynx with bilateral neck lymphatics 6, 20.7 percent; neck directed radiotherapy field bilateral 9, 31.1 percent; recurrent tumour bed or recurrent mass bed 1, 3.4 percent; radiotherapy not administered 1, 3.4 percent.
Gy = Gray; ICU = intensive care unit; RT = radiation therapy.
Additional airway-related procedures were required in five patients (17.2 per cent). Tracheostomy-related complications were recorded in two patients (6.9 per cent), and emergency airway intervention was required in seven patients (24.1 per cent). Among RT-related parameters, the median radiation dose was 66 Gy, with a range of 54–70 Gy. The most frequent radiation fields were tumour or surgical bed with ipsilateral neck lymphatics and bilateral neck-directed irradiation, each observed in nine patients (31.1 per cent).
Treatment modalities and timing of tracheostomy
CRT was the most frequently applied treatment modality (Table 3). Tracheostomy timing was categorised as pre-treatment, during treatment, early post-treatment and late. Most procedures were performed during or shortly after treatment, particularly in patients receiving CRT. Late tracheostomies were less frequent and were mainly associated with delayed treatment-related airway changes.
Oncological treatments of the cases
Table 3 Long description
Oncological treatments are listed with counts in the column, n, and percentages in the column, percent. Surgery plus radiotherapy has 8 cases at 27.6 percent. Surgery plus chemoradiotherapy has 7 cases at 24.1 percent. Surgery with primary focus has 1 case at 3.4 percent. The total row reports 29 cases at 100 percent.
When tracheostomy indications were evaluated in relation to timing, distinct clinical patterns emerged. Early interventions were predominantly associated with acute airway compromise due to tumour burden or treatment-related oedema, whereas late tracheostomies were more frequently linked to fibrosis-related airway limitation and functional airway impairment. These relationships are summarised in Table 4.
Relationship between tracheostomy timing and predominant underlying mechanisms of airway compromise
Table 4 Long description
Counts of airway compromise mechanisms are listed by tracheostomy timing, with columns for tumour-related obstruction, oedema, fibrosis, functional glottic impairment or stenosis, bleeding, and a total. Pre-treatment totals 2 cases, all tumour-related obstruction. During treatment totals 11 cases: tumour-related obstruction 5, oedema 4, functional glottic impairment or stenosis 1, bleeding 1, and fibrosis 0. Early post-treatment, less than 30 days, totals 11 cases: oedema 5, tumour-related obstruction 3, fibrosis 1, functional glottic impairment or stenosis 1, and bleeding 1. Late post-treatment, more than 30 days, totals 5 cases: functional glottic impairment or stenosis 3, fibrosis 2, and zero for tumour-related obstruction, oedema, and bleeding. Column totals are tumour-related obstruction 10, oedema 9, fibrosis 3, functional glottic impairment or stenosis 5, bleeding 2, for an overall total of 29.
Tracheostomy indications and pathological conditions
Respiratory distress was the most common indication for tracheostomy. Other indications included airway protection due to anticipated treatment-related oedema, recurrent bleeding and prolonged intubation (Table 5).
Surgical tracheostomy indications of the cases
Table 5 Long description
Indications are listed with counts in the n column and percentages in the percent column, totaling 29 cases. Respiratory tract safety due to possible operation or RT related oedema has 9 cases at 31.0 percent. Respiratory tract safety due to recurrent bleeding has 2 cases at 6.9 percent. Maintaining respiratory tract during arrest intervention has 2 cases at 6.9 percent. Prolonged intubation and neck anatomy change has 1 case at 3.4 percent. The total row reports 29 cases at 100 percent.
RT = radiation therapy.
Relationship among diagnosis, treatment type, tumour stage and timing of tracheostomy
No statistically significant association was found between tracheostomy timing and diagnosis, T stage, N stage or treatment modality when analysed individually (p > 0.05). When underlying mechanisms were grouped according to the predominant clinical mechanism, a significant association with tracheostomy timing emerged. Acute mechanisms included tumour-related obstruction, treatment-related oedema and bleeding-related airway compromise, whereas chronic mechanisms included fibrosis-related airway limitation and functional glottic impairment. Early tracheostomies were predominantly associated with acute mechanisms, whereas late tracheostomies were exclusively related to chronic airway impairment (Fisher’s exact test; p < 0.01).
These findings suggest that tracheostomy timing is more closely related to dynamic airway conditions and treatment-related changes rather than static tumour characteristics. In particular, fibrosis and glottic dysfunction were predominantly observed in patients with a history of RT or CRT, supporting a strong association between late airway compromise and treatment-related tissue effects.
Discussion
Existing literature on surgical tracheostomy in non-laryngeal HNC remains limited, often focusing on risk scoring systems for elective procedures or single anatomical regions.Reference Mohamedbhai, Ali, Dimasi and Kalavrezos5–Reference Janik, Brkic, Grasl, Königswieser, Franz and Erovic7 While some prior studies highlight the importance of the oral cavity and oropharynx,Reference Li, Agrawal, Fakhry, Fakhry and D’Souza2 the present series found that, although oral cavity cancers constituted the largest regional group, nasopharyngeal cancer was the single most common subsite requiring a surgical airway (20.7 per cent of cases). Unlike previous studies focusing primarily on elective tracheostomy or specific anatomical regions, the present study emphasises the temporal evolution and underlying mechanisms of airway compromise, providing a more clinically dynamic perspective. However, this interpretation should be made cautiously because non-laryngeal HNCs represent a heterogeneous clinical group. Tumours of the oral cavity, oropharynx, nasopharynx, thyroid, salivary glands and other cervical sites differ substantially in tumour biology, anatomical proximity to the airway, treatment modality, radiation exposure, surgical extent and expected functional sequelae. Accordingly, the timing and mechanism of airway compromise may differ between subsites, and the proposed temporal–mechanistic framework should not be interpreted as uniformly applicable to all non-laryngeal HNCs.
An exploratory observation of this study is the distinction between early and late patterns of airway compromise in non-laryngeal HNCs. Early tracheostomy was predominantly associated with acute mechanisms, including tumour burden, treatment-related oedema and bleeding-related airway compromise, whereas late tracheostomy was mainly associated with fibrosis, functional glottic impairment and delayed structural changes following RT or CRT. This temporal–mechanistic distinction may help clinicians conceptualise airway risk across different phases of treatment; however, it should be interpreted cautiously because mechanisms may overlap in clinical practice and the present classification was based on the predominant mechanism recorded at the time of intervention.
A key finding from this tertiary centre experience is the strong association between tracheostomy necessity and primary definitive or adjuvant non-surgical treatment. In our series, the majority of interventions were related to CRT (44.8 per cent of cases) and RT (total of 96.6 per cent when combining RT, CRT and surgery+RT). Furthermore, tracheostomy was performed during or after oncological treatment in 27 of 29 patients, highlighting the substantial impact of treatment and treatment-related airway changes on airway patency. This contrasts with earlier studies that often highlighted surgical tracheostomy primarily due to mass obstruction or oedema associated with planned surgical interventions.Reference Gupta, Mandlik, Patel, Patel, Shah and Vijay6 This finding further supports the notion that airway compromise in non-laryngeal HNC is often a consequence of treatment-related changes rather than primary tumour characteristics alone.
Our findings underscore the significance of late-onset complications, which are less commonly tracked in operative tracheostomy series. Patients who achieve long-term survival may experience severe respiratory tract disorders in the late period after definitive treatment.Reference Akazawa, Shimizu, Fujino and Kato3, Reference Tennant, Cash, Bumpous and Potts8 The pathological conditions observed in our series support this: Late-stage fibrosis.Fibrosis in the neck (10.3 per cent) and insufficient mouth opening (6.9 per cent) were observed as late-term side effects after RT/CRT. These changes, known as radiation-induced fibrosis (RIF), are recognised independent predictors of a difficult airway, necessitating careful planning, sometimes requiring specialised techniques like awake tracheostomy.Reference Akazawa, Shimizu, Fujino and Kato3, Reference Tennant, Cash, Bumpous and Potts8 Glottic stenosis: The presence of stenosis at the glottic level (paralysis/stenosis) in five patients (17.2 per cent) was a critical finding, even though laryngeal primaries were excluded. As reported in the literature, if the radiation field included the larynx during treatment for adjacent tumours, patients are at risk of developing laryngotracheal stenosis or functional impairment.Reference Queenan, Trivedi, Bertoni, Siddiqui, Yap and Tibbetts9 For instance, one case of nasopharyngeal cancer treated with CRT developed vocal fold paralysis, demonstrating that functional impairment can occur in areas remote from the primary tumour site.Reference Akazawa, Shimizu, Fujino and Kato3 These observations highlight that late airway complications are not rare events but represent a clinically relevant subset that requires structured long-term surveillance, particularly in patients treated with CRT. From a clinical perspective, this temporal–mechanistic framework may assist clinicians in anticipating airway compromise at different stages of treatment and may support more individualised airway management strategies.
Although no statistically significant associations were found when variables were analysed individually, grouping mechanisms into acute and chronic categories revealed a significant relationship with tracheostomy timing. This suggests that conventional variable-based analyses may fail to capture clinically meaningful patterns, particularly in small and heterogeneous cohorts.
One of the clinically important findings of the present study was the occurrence of glottic dysfunction and vocal fold impairment despite the exclusion of primary laryngeal and hypopharyngeal cancers. Functional glottic impairment or stenosis was observed in five patients (17.2 per cent), indicating that clinically relevant laryngeal-level airway compromise may develop even when the primary tumour is located outside the larynx. This finding is particularly relevant in patients treated with RT or CRT, as radiation fields for non-laryngeal tumours may still include or affect laryngeal structures, recurrent laryngeal nerve pathways, pharyngeal constrictors and adjacent soft tissues. Delayed fibrosis, impaired laryngeal mobility, neuromuscular dysfunction, chronic oedema and progressive glottic narrowing may contribute to respiratory symptoms, aspiration risk, dysphonia and eventual airway compromise.Reference Akazawa, Shimizu, Fujino and Kato3, Reference Queenan, Trivedi, Bertoni, Siddiqui, Yap and Tibbetts9
These complications may be under-recognised during routine oncological follow-up because surveillance is often focused on tumour control, recurrence, swallowing and nutritional status rather than dynamic airway function. In long-term survivors of non-laryngeal HNC, new-onset dyspnoea, stridor, voice change, aspiration, reduced exercise tolerance or unexplained respiratory deterioration, therefore, should prompt detailed laryngeal assessment. Flexible laryngoscopy, evaluation of vocal fold mobility, airway calibre assessment and multidisciplinary input from otorhinolaryngology and speech–language therapy may help identify evolving glottic dysfunction before emergency airway intervention becomes necessary.Reference Burns, Ward, Hill, Kularatna, Byrnes and Kenny10 In this context, glottic dysfunction should be considered an important survivorship issue rather than an incidental airway finding.
Consistent with the clinical profile of this cohort, respiratory distress and desaturation was the most common indication for tracheostomy, recorded in 15 patients (51.8 per cent). The most frequent predominant mechanisms were tumour-related obstruction in 10 patients (34.5 per cent) and treatment-related oedema in nine patients (31.0 per cent). These findings indicate that acute airway compromise in non-laryngeal HNCs is often related to dynamic airway narrowing rather than tumour stage alone. Previous studies have similarly emphasised the importance of objective airway risk assessment and early identification of patients who may require tracheostomy in head and neck malignancies.Reference Gupta, Mandlik, Patel, Patel, Shah and Vijay6
RT-containing treatment was administered in 28 patients (96.6 per cent), including definitive CRT in 13 patients (44.8 per cent), surgery followed by RT in eight patients (27.6 per cent) and surgery combined with CRT in seven patients (24.1 per cent). This high proportion of RT exposure provides important clinical context for interpreting late airway complications such as fibrosis-related airway limitation and functional glottic impairment. Functional glottic impairment or stenosis was observed in five patients (17.2 per cent), despite the exclusion of primary laryngeal and hypopharyngeal cancers.
In the present study, no statistically significant association was found between tracheostomy timing and diagnosis, treatment modality, T stage or N stage when these variables were analysed individually. This suggests that the timing of tracheostomy in non-laryngeal HNCs may be influenced more by evolving airway conditions, airway obstruction risk and peri-operative or post-treatment airway management needs than by static baseline oncological variables alone.Reference Nagarkar, Kokane, Wagh, Kulkarni, Roy and Tandale11 Existing scoring systems, such as TRACHY and CASST, provide useful structured approaches for peri-operative airway assessment and elective tracheostomy decision-making in HNC surgery.Reference Mohamedbhai, Ali, Dimasi and Kalavrezos5, Reference Gupta, Mandlik, Patel, Patel, Shah and Vijay6 However, their applicability to the present cohort may be limited for several reasons. First, these systems primarily address pre-operative or peri-operative airway risk, whereas most patients in our cohort required tracheostomy during or after oncological treatment. Second, delayed treatment-related mechanisms, including RT-induced fibrosis, glottic dysfunction, vocal fold impairment and progressive soft-tissue narrowing, may evolve after treatment completion and are not easily captured by a single pre-treatment score.Reference Akazawa, Shimizu, Fujino and Kato3, Reference Queenan, Trivedi, Bertoni, Siddiqui, Yap and Tibbetts9 Third, non-laryngeal HNCs comprise anatomically and biologically heterogeneous subsites, and airway compromise may occur through different mechanisms despite similar tumour stage or treatment category.Reference Barsouk, Aluru, Rawla, Saginala and Barsouk1, Reference Li, Agrawal, Fakhry, Fakhry and D’Souza2 Therefore, although predictive scores remain useful for surgical planning, they should be complemented by longitudinal airway surveillance, particularly in patients receiving RT or CRT.
The role of tracheostomy should also be interpreted within the broader context of function-preserving treatment strategies for non-laryngeal HNCs. In oropharyngeal cancer, transoral robotic surgery has been increasingly used as a minimally invasive approach aiming to achieve oncological control while reducing treatment-related functional morbidity, particularly regarding swallowing and speech outcomes.Reference Guarino, Chiari, Cordeschi, D’Alessio, Ingelido and Motta12 However, even in function-preserving paradigms, airway management remains clinically relevant because post-operative oedema, bleeding risk, previous irradiation, flap reconstruction or combined-modality treatment may still necessitate temporary or prolonged airway protection. Therefore, tracheostomy decisions should be integrated with functional outcome assessment rather than considered solely as an oncological or anatomical consequence.
Similarly, minimally invasive and function-preserving oncological treatments, such as electrochemotherapy, are increasingly discussed for selected oral cavity and oropharyngeal cancers.Reference Chiari, Filippini, Fermi, Presutti, Donadio Caporale and Bertino13 Although electrochemotherapy may offer local tumour control with a tissue-sparing intent in carefully selected patients, airway compromise may still occur because of tumour location, treatment-related oedema, bleeding or prior multimodal therapy. In this setting, tracheostomy should be viewed not only as an emergency airway procedure but also as part of comprehensive functional and supportive care planning in selected high-risk patients.
This study has several limitations. First, its retrospective design and limited sample size substantially restrict statistical power and preclude definitive causal inferences. Therefore, the proposed timing- and mechanism-based framework should be interpreted as exploratory and hypothesis-generating rather than as a clinically validated predictive model. Second, the cohort was heterogeneous, including multiple non-laryngeal tumour subsites with distinct biological behaviours, anatomical patterns of airway compromise and treatment pathways. This heterogeneity may have influenced both the indication and timing of tracheostomy and should be considered when interpreting the temporal–mechanistic classification. Third, although mechanisms of airway compromise were grouped for analytical purposes, this approach may oversimplify the clinical reality, as some patients may present with overlapping contributors such as tumour burden, oedema, fibrosis, bleeding and glottic dysfunction. In such cases, classification was based on the predominant mechanism documented in the clinical record. Despite these limitations, the study provides real-world tertiary centre data on an underexplored patient group and may help generate hypotheses for future prospective studies with larger, more homogeneous cohorts.
• Airway compromise in head and neck cancer may result from tumour obstruction, peri-treatment oedema, bleeding or delayed radiotherapy-related fibrosis and stenosis
• Surgical tracheostomy remains an important component of airway management, although current practice has shifted from routine prophylactic procedures towards selective risk-based intervention
• Existing tracheostomy scoring systems mainly focus on elective surgical settings and may be less informative for delayed or treatment-related airway compromise
• In this cohort of non-laryngeal head and neck cancers, tracheostomy was most commonly required during or after chemoradiotherapy rather than being determined solely by tumour site or stage
• Early tracheostomy was mainly associated with acute mechanisms such as tumour-related obstruction, treatment-related oedema and bleeding-related airway compromise
• Late tracheostomy was linked to chronic mechanisms, particularly fibrosis and functional glottic impairment after radiotherapy or chemoradiotherapy
• A timing- and mechanism-based framework may help clinicians recognise airway risk across different phases of oncological treatment, but should be interpreted as exploratory
Conclusion
Surgical tracheostomy in non-laryngeal HNCs appears to be driven not only by tumour site or stage but also by dynamic treatment-related airway changes. In this retrospective tertiary centre cohort, early airway compromise was mainly associated with tumour burden or treatment-related oedema, whereas late airway compromise was more commonly related to fibrosis and functional airway impairment after RT or CRT.
Given the limited sample size and heterogeneous tumour subsites, the proposed timing- and mechanism-based approach should be regarded as exploratory and hypothesis-generating. Nevertheless, it may provide a clinically useful perspective for recognising airway risk across different phases of oncological treatment. Long-term multidisciplinary surveillance remains essential to anticipate delayed airway deterioration and reduce the need for emergency tracheostomy.
Acknowledgments
None.
Author contributions
Arzubetül Duran: Study design and article planning, final editing and critical revision; Selçuk Erol: Study design and article planning, data collection and analysis, manuscript writing.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Data availability
The datasets generated and/or analysed during the current study are available from the corresponding author on reasonable request, provided that participant confidentiality is not compromised.
Ethical approval
All procedures performed in this study were in accordance with the ethical standards of the institutional and national research committee and with the Declaration of Helsinki (1975, revised 2008). Ethical approval for this retrospective study was obtained from the Dr. Abdurrahman Yurtaslan Ankara Oncology Training and Research Hospital Institutional Ethics Committee (Approval number: 2022-09/147; Date: 22 September 2022).
Consent for publication
All authors have read and approved the final manuscript, agree on the order of authorship and confirm that no other individuals meet the criteria for authorship. The corresponding author is responsible for all communication with the journal and ensures that all authors approve the final version of the manuscript.
Competing interests
The authors declare that they have no competing interests.
