Long-term results of intensive care patients with post-intubation tracheal stenosis: 7 years follow-up

Abstract

Background/aim

Benign tracheal stenosis (TS) is a potentially life-threatening condition caused by iatrogenic events resulting from endotracheal intubation and tracheostomy. The study aimed to determine the clinical characteristics, management, and long-term outcomes of patients with simple or complex post-intubation tracheal stenosis (PITS) who were admitted to the intensive care unit (ICU).

Materials and methods

Retrospective analysis of patients admitted to the ICU between June 2016 and June 2022 and diagnosed with PITS were performed until June 2023.

Results

76 patients (mean age 54.9 ± 17.3%, 63.2% male, 61.8% complex stenosis) admitted to the ICU were analysed. The diagnosis of PITS was made by fiberoptic bronchoscopy (FOB) in 53.9% and computed tomography (CT) in 46.1%. The most common location of stenosis was the upper trachea area (57.9%, n = 44). The median stenosis length of the TS was 1.9 cm (median 2 cm for complex stenosis, 1 cm for simple stenosis, p < 0.001). Complex stenosis was associated with longer intubation time and multiple intubation history (p = 0.011, p = 0.028). At the same; complex stenosis was associated with prolonged and high stenosis percentage (p < 0.001). While dilatation and cryotherapy were sufficient for treatment in all patients with simple stenosis, the stent was applied to 34% patients with complex stenosis. Recurrence was higher in complex stenosis after bronchoscopic procedure (complex stenosis 95.7%, simple stenosis 17.2%). 19 (25%) patients required surgical operation, and all of these patients had complex stenosis (p < 0.001). Median follow-up was 33 months for simple stenosis and 34 months for complex stenosis. ICU length of stay (LOS) was longer in patients with complex stenosis. There was no statistically significant difference between TS classes in terms of ICU mortality (p = 0.466), 1-year mortality (p = 0.951), and mortality during follow-up (p = 0.608).

Conclusion

PITS is a challenging situation in ICU patients. In patients with a preliminary diagnosis of PITS, bronchoscopy should be performed without delay, and it should be known that interventional bronchoscopy is not only a diagnostic tool but an effective strategy in treatment management after the diagnosis is finalised. The advantage of interventional bronchoscopy is that it alleviates symptoms that necessitate hospitalisation in the ICU and eliminates the need for ICU of patients.

Background

Tracheal stenosis (TS) is a rare but life-threatening condition. Benign tracheobronchial stenosis, which refers to abnormal narrowing of the trachea and bronchi, usually results from post-intubation tracheal stenosis (PITS), previous tracheotomy, tuberculosis, granulomatosis with polyangiitis, and idiopathic strictures. The main symptoms of TS include stridor, dyspnea, and wheezing [1]. Although there are no precise results regarding the incidence of TS, acquired TS is generally associated with PITS, with a reported incidence of 0.6–21% [2]. High cuff pressure is widely recognised as the primary factor in the development of TS, alongside the patient’s clinical condition [3, 4]. However, the absence of effective cuff pressure monitoring in clinical practice exacerbates this issue [5].

Depending on the severity of the stenosis in TS, patients may show conditions ranging from asymptomatic to respiratory failure [2]. The clinical condition of TS patients may deteriorate rapidly and may require intensive care treatment. Because of TS and its accompanying comorbidities, the management of these patients in the intensive care unit (ICU) can be a very challenging process. In addition to ICU follow-up, many interventional chest disease procedures (balloon dilatation, cryotherapy, laser cauterization and placement of a stent) are applied to these patients. PITS management is still controversial, and there has yet to be an international consensus on the best treatment option [6]. Surgical resection of the affected trachea followed by anastomosis is considered the definitive treatment for this condition [2, 6, 7]. Although surgery is still considered the first treatment option, especially in complex strictures [6, 8], interventional bronchoscopy has shown promising results in selected patients with simple stenosis [9]. In fact, some authors have suggested bronchoscopic treatment as an alternative to surgery in patients who are unsuitable and awaiting surgery [6, 10, 11]. In addition, recent data have demonstrated the bronchoscopic approach’s critical role in managing benign TS and suggest that after accurate stenosis classification and careful patient selection, this treatment modality may represent a definitive strategy in most cases [8, 12, 13]. Although various approaches have been developed for treating TS, surgical resection and tracheal reconstruction remain the most effective methods for complex tracheal stenosis [7].

Comprehensive clinical evaluation and a multidisciplinary approach can positively affect the outcome of these patients and shorten the length of stay (LOS) in the ICU while enabling patients to continue their lives with less invasive procedures without the need for surgical treatment. In this way, while improving patients’ quality of life, effective use of health services can be ensured. This retrospective cohort study aims to evaluate the treatment, management, long-term results, and survival analyses in patients admitted to the ICU with the diagnosis of simple or complex PITS between 2016 and 2022. The study also highlights the role of bronchoscopic interventions in treatment management and the practical and effective use of health services.

Materials and methods

Study design

This retrospective cohort study was started after the ethics committee approval was obtained with the decision of the ethics committee of Yıldırım Beyazıt University Yenimahalle Training and Research Hospital Clinical Research Ethics Committee, dated 11.10.2023 and numbered E-2023-52. This study was registered with the ClinicalTrials.gov international protocol registration and results system under registration number NCT06121024 (07/11/2023). The study complies with the Declaration of Helsinki.

Participants

Between 2016 and 2022, a total of 128 adult patients aged 18 years and older were admitted to the ICU due to PITS. 37 patients due to malignancy, 8 patients due to idiopathic etiology, and 7 patients with missing data were excluded from the study. Data from 76 patients were analyzed. Our tertiary care centres are referral centres especially for complex benign TS.

Patients’ age, gender, comorbidities, reason of TS development, LOS in the ICU, TS-related symptoms, grade of TS, localisation of TS, treatment of TS, complications after TS treatment, hospitalisation in ICU mortality over time, one-year mortality, and one-year recurrence were recorded. Intubation time and/or tracheostomy time were also recorded.

Interventions and clinical definitions

In our ICU, the diagnosis of PITS was made by thoracic and cervical computed tomography (CT) or bronchoscopy. Patients underwent a flexible bronchoscopy (FOB) to diagnose and evaluate features of stenosis, which are the location of the stenosis, its vertical extension, its severity, and stenosis classification as simple or complex. Simple PITS was defined as lesions with a vertical extension < 1 cm (short segment) with endoluminal occlusion, without tracheomalacia or loss of cartilage support; complex ones were defined as lesions ≥ 1 cm and varying degrees of cartilage involvement or circumferential contractile scar or TS associated with malacia and inflammation [6, 12]. The classification defined by Myer et al. using the outer diameters of the endotracheal tubes was used to compare the diameter of the tracheal lumen with the bronchoscope. Up to 50% of the lumen is defined as Grade I, between 51% and 70% as Grade II, and above 70% as Grade III. An airway with no visible lumen was defined as grade IV [14]. The location of laryngotracheal stenosis was classified according to Galluccio et al. as follows: subglottic area (just beneath the vocal cords, but within the cricoid cartilage, also referred to as laryngostenosis), upper tracheal area (generally at the internal stoma site), mid/lower tracheal area, and two or more areas [8].

Following stenosis evaluation, all patients (intensive care patient with symptoms) underwent interventional rigid bronchoscopy. Patients with simple PITS received mechanical dilatation, while cryotherapy and/or laser were used for eligible patients [9]. Bronchoscopic treatment was performed on patients with complex PITS. If dilatation was ineffective (obstruction ≥ 50% after the procedure), a stent was placed as described by Dumon [15]. If obstruction was still present after stenting, a tracheostomy was performed to ensure a safe airway. After the procedure, all complex PITS patients were evaluated for surgical resection through consultation with a surgeon. The treatment strategy of our hospital for patients with suspected PITS is summarised in Fig. 1.

Fig. 1

The treatment strategy of our hospital for patients with suspected PITS

Patients who were discharged from the ICU and did not relapse one year after the last interventional procedure were also considered “treatment successful”. It was considered as a long-term recurrence if any additional intervention was required later on [8].

Outcomes

The study’s primary outcomes were the discharge rate from the ICU and the mean LOS in the ICU following interventional bronchoscopic treatment for patients diagnosed with simple or complex TS in the ICU. The study’s secondary outcome was the rate of recurrence of TS within one year, while the tertiary outcome was the rate of mortality within the same period.

Statistical analysis

SPSS 23.0 package program was used to analyse the collected data statistically. Categorical variables were defined as number (n) and percentage (%). Continuous variables were defined as mean, standard deviation, and minimum (min) – maximum (max). The conformity of the variables to the normal distribution was examined by Kolmogorov-Smirnov/Shapiro-Wilk Tests. Chi-square Test and Fisher’s Exact Test were used to compare categorical variables. Independent Student’s t-test was used in groups with normal distribution, and Mann Whitney U Test was used in groups that did not comply with normal distribution. A p < 0.05 level was considered statistically significant. Kaplan-Meier estimation was used to compare the survival functions between groups.

Results

The study involved 76 patients who were monitored in the ICU for PITS from June 2016 to June 2022. Table 1 shows the clinical features of the patients.

Table 1 Clinical characteristics of the patients

PITS was diagnosed in 48 (53.9%) patients by FOB (46.1%) and in 28 patients by CT. The most common medical condition at first intubation of patients was respiratory causes (32.9%). The most common symptom at admission to the ICU was dyspnea. The most common history of patients was arterial hypertension (21.0%).

Patients with TS were classified as follows, 38.2% (n = 29) simple and 61.8% (n = 47) complex. PITS features are shown in Table 2. The median intubation time of the patients was 20.5 days (min-max, 3-120). Complex stenosis was associated with longer intubation time compared to simple (32.4 vs. 17.9 days, p = 0.011). While 30 (39.5%) of the patients had a single intubation history, 46 (60.5%) had a history of multiple intubation. Multiple intubation history was statistically significant and higher in patients with complex stenosis (70.2% vs. 44.8%, p = 0.028). There was a history of tracheostomy in 16 (21.0%) patients. Although complex stenosis was more common in previously tracheostomised patients, it was not statistically significant (21.3% vs. 20.7%, p = 0.951).

Table 2 Post-intubation tracheal stenosis features

The most common location of stenosis was the upper tracheal area (57.9%, n = 44), and there was no statistically significant difference between stenosis classifications (p = 0.110).

The median stenosis length of the tracheal stenoses was 1.9 cm (range, 0.5–5.5 cm). There was a statistically significant difference (p < 0.001) with a median stenosis length of 2 cm (range, 1.0–5.5 cm) for complex stenosis and 1 cm (range, 0.5–3.0 cm) for simple stenosis. The length of the stenosis was longer in complex stenosis.

There was a statistically significant difference between simple and complex stenosis in terms of percent degree of stenosis (p < 0.001). The percent degree of stenosis was higher in patients with complex stenosis.

In our ICU, FOB was performed on all patients who were thought to have PITS in the preliminary diagnosis, and dilatation was applied to all patients when the diagnosis was confirmed during bronchoscopy (As shown in Fig. 2). The treatment methods applied to the patients are summarised in Table 3. There was a statistically significant difference between stenosis classes regarding treatment management, and patients with complex stenosis required more invasive treatment (p = < 0.001). Dilation was sufficient in treating stenosis in only 2 (2.6%) patients, and the stenosis class was simple in these patients. While dilatation and cryotherapy were sufficient for treatment in all patients with simple stenosis, 16 (34%) patients with complex stenosis required a stent for treatment (As shown in Fig. 3).

Table 3 Treatment strategies of the patients

Fig. 2

Dilation + cryotherapy application with rigid bronchoscopy in the patient (a post-intubation tracheal stenosis in bronchoscopy, b after dilation and cryotherapy with rigid bronchoscopy)

Fig. 3

Dilation + cryotherapy + stent application with rigid bronchoscopy in the patient with complex stenosis (a post-intubation tracheal stenosis in bronchoscopy, Grade III stenosis, b after dilation and cryotherapy with rigid bronchoscopy, c silicone stent placed to maintain airway patency)

In 5 (17.2%) patients with simple stenosis and 45 (95.7%) patients with complex stenosis, restenosis developed after the first procedure and repeated bronchoscopic treatment procedures were required. There was a statistically significant difference between the classes regarding recurrence, which was higher in the class with complex stenosis (p <0.001). Considering the patients who developed recurrent stenosis, the median time for recurrence of stenosis after the first procedure was 20 days. A total of 19 (25%) patients required surgical operation, and all of these patients had complex stenosis (p < 0.001). Only one patient and his relatives did not accept the surgical procedure, and this patient died in the 44th month of follow-up.

The patient’s LOS in the ICU, their mortality during the intensive care follow-up period, their one-year mortality, the mean follow-up period and the mortality data during the total follow-up period are summarised in Table 4. The median follow-up period of patients with simple stenosis was 33 months (range, 0.9–84 months), and the median follow-up period of patients with complex stenosis was 34 months (range, 0.6–81 months), and there was no significant difference between stenosis classes. The LOS in the ICU was statistically significantly higher in patients with complex stenosis, and the median was four days (range, 2–35 days) in patients with simple stenosis and 11 days (range, 4–35 days) in patients with complex stenosis (p<. 001). There was no statistically significant difference between TS classes in terms of ICU mortality (p = 0.466), 1-year mortality (p = 0.951) and mortality during follow-up (p = 0.608). No significant difference was found between the groups when survival analysis was performed based on stenosis classes (p = 0.705, see Fig. 4).

Table 4 Follow-up data of the patients

Fig. 4

Comparison of mean survival time (month) across tracheal stenosis classification using Kaplan-Meier estimation

Considering the success of the treatment, 23 (79.3%) of the patients with simple stenosis and 2 (4.3%) of the patients with complex stenosis did not have recurrence and mortality within one year, and the treatment was considered successful. There was a statistically significant difference between TS classes regarding treatment success (p < 0.001).

A subgroup analysis was conducted to compare patients who underwent surgical procedures after interventional bronchoscopic procedures with those who did not, in patients with complex stenosis. The results are summarised in Table 5. The median LOS in the ICU was 14 days (range, 4–35 days) for patients who did not undergo surgery and 9 days (range, 4–35 days) for those who did undergo surgery (p = 0.177). There was no statistically significant difference between the surgical and non-surgical groups regarding ICU mortality (p = 0.137), follow-up mortality (p = 0.104), and survival function (p = 0.051).

Table 5 Follow-up data of patients with complex tracheal stenosis (subgroup analysis)

Discussion

Benign TS is a rare but potentially fatal condition that significantly impacts quality of life. Currently, there are no international guidelines for managing PITS because the effectiveness of bronchoscopic or surgical treatment compared to endoluminal treatment modalities mostly depends on the experience of different referral centres [6]. Most centres consider surgical treatment the gold standard for PITS [6, 7, 16]. However, there is evidence that interventional bronchoscopy may have a role in the treatment of PITS not only in inoperable patients or as a bridge to surgery [10, 11] but also as a definitive therapeutic role in selected patients with an accurate stenosis assessment [8, 12, 13]. In the present study, we aimed to evaluate the role of interventional bronchoscopy in treating patients with PITS in the ICU and to show the efficient use of intensive care beds, which are valuable and costly, with effective treatment management.

It has been suggested that only classified simple stenosis can be treated using laser and mechanical dilation to treat TS, and cure rates range from 60 to 95% [8, 17]. It has been reported that the recurrence rate after dilatation and laser treatment in complex TS is around 90% [18]. Galluccio et al. [8] reported a 69% success rate for complex TS with endoluminal therapy and suggested a multidisciplinary approach for patients with complex stenosis. Cavaliere et al. [12] reported a 100% success rate for simple stenosis and a 22% success rate for complex stenosis. They emphasised the importance of appropriate bronchoscopic, clinical, and radiological evaluation to improve the outcome of endoscopic therapy. All of the studies in the literature were on outpatients with TS. Although there is no study in the literature on ICU patients, our study is the first in the literature in this respect. In our study, 53.9% (41 patients) of the patients were diagnosed by bronchoscopy, and 46.1% (35 patients) were diagnosed by CT. In our study, a complete cure was achieved with a single bronchoscopic procedure in 79.3% (23 patients) of patients with simple stenosis, and no recurrence was observed in the long-term follow-up of the patients. In addition, in 2 (6.9%) patients with simple stenosis, only dilatation with bronchoscopy was sufficient. No surgical procedure was required in any of the patients with simple stenosis. Recurrence was observed in 95.7% of patients with complex stenosis after the bronchoscopic procedure. 2 patients (4.3%) with complex stenosis were treated with only bronchoscopic intervention, and no recurrence was observed within one year. Although the success of our patients is low compared to the literature, we attribute this to the fact that our patients are ICU patients, and their comorbidity and risks are higher than other patients. In addition, 89.5% (68 patients) of patients who were found to have TS in the ICU through the bronchoscopic procedure were discharged from the ICU, and 32.8% of the patients (25 patients, 23 simple stenosis – 2 complex stenoses) reported that their treatment was successful; as a result, we would like to state that there was no recurrence or mortality within one year.

Complex stenoses with cartilage involvement require treatment with a stent. In patients with complex stenosis, stenting or any other intervention may prolong the length of a stenosis, especially in patients with high subglottic stenosis [19, 20]. Brichet et al. [6] reported a 17.6% success rate for complex TS treated with endotracheal stenting. In our study, stents were required in 16 (34%) patients with complex stenosis. All patients who underwent stents had recurrence within the first year and required bronchoscopic procedures. A surgical procedure was required in 50% of the patients with stent implantation. Two patients (12.5%) who underwent stenting died during ICU follow-up. Both patients who died in the ICU follow-up were palliative care patients with a history of trauma and followed up with a tracheotomy home-type mechanical ventilator. Four of the 14 patients with complex stenosis who were discharged from the hospital and stented died within a year. Among the patients with complex stenosis and stenting, two patients (12.5%) were treated for one year with only bronchoscopic treatment without the need for surgery. In this respect, it can be said that our success rate is 12.5% in our patients who are diagnosed with PITS in the ICU and require bronchoscopic procedures and stent treatment. No increase in the length of stent-related stenosis was observed in any of our patients. We attribute our low success rate to our patients being ICU patients, and their comorbidities were high. 14 of 16 patients with complex stenosis who underwent stenting were discharged from the ICU with the application of a bronchoscopic procedure, and the median LOS in the ICU was 8.5 (range 4 to 20 days) days. We think that with the application of bronchoscopic treatment, the LOS in the ICU of patients with severe complex stenosis was shortened, and the efficiency of health care increased.

Surgical treatment continues to be accepted as the standard gold treatment for PITS in most centres [6, 7, 16]. Recent studies support the fact that interventional bronchoscopy can be an alternative [8, 10, 11, 13]. Treating comorbidities is necessary to optimise surgical conditions, and patients diagnosed with PITS have high comorbidities. As a result, interventional bronchoscopy may be the only way to relieve symptoms. In addition, surgical procedures cannot be performed with high subglottic stenosis and long stenosis length (> 4–6 cm) [21]. However, even in severe cases of complex TS, advances in techniques such as airway management and extracorporeal membrane oxygenation (ECMO) may provide a wider range of surgical indications [22]. The incidence of postoperative restenosis varies between 8.1% and 24% [23,24,25]. In the present study, only 25% of the patients required a surgical procedure. All of these patients had complex stenosis. For patients with complex stenosis, there was no significant difference in the LOS in the ICU, ICU mortality, and mortality during follow-up between those who underwent surgical procedures and those who did not. When analyzing survival data, patients who underwent surgical procedures had a mean survival of 64 months, while those who did not had a mean survival of 41 months, but the difference was not statistically significant. Restenosis developed within the first year in 3 (15.7%) of the patients who underwent surgery and required bronchoscopic procedures. There was no mortality in the first year after the surgical procedure. The median follow-up period of patients who underwent surgery was 39 months (range, 13–81 months). During the follow-up period, five patients (26.3%) who underwent surgical procedures died, and all were after one year. Three of the five patients who died were patients who developed restenosis within the first year. The decision regarding the treatment of patients with complex PITS should be made by an experienced team. For symptomatic and unstable patients requiring intensive care hospitalisation, an interventional bronchoscopic procedure performed before surgery may stabilise both the patient’s follow-up and the surgical procedure. Interventional bronchoscopic procedures can be highly valuable in managing patients with complex PITS, provided that the right treatment is chosen and an experienced team is involved. However, while exciting developments in the field of interventional pulmonology are promising, it is important to acknowledge that surgical intervention remains the gold standard in specialized cases.

Present study has some limitations. Firstly, this cohort study population is a specially selected cohort of ICU patients. Therefore, it is not a representative sample of treatment approaches for all patients with TS in a population. Secondly, this study was retrospective and had inherent limitations that may limit its ability to highlight causal associations with relevant diagnoses, treatments or drug administration. However, the information can still inform clinical decision-making and guide future more rigorous studies. Further research is needed in this subject. Third, our hospital treats mainly lung and thoracic surgery patients and is a referral center for complex TS. Therefore, our rate of complex stenosis was higher and these results could not reflect the general population.

Conclusion

Our study data show the importance, efficacy, and safe procedure of interventional bronchoscopy in managing PITS in ICU patients. The main advantage of interventional bronchoscopy is that it alleviates symptoms that necessitate hospitalisation in intensive care and obviates the need for intensive care. For patients with TS requiring intensive care, interventional bronchoscopic procedures are sufficient for simple PITS. However, a multidisciplinary approach is required for treatment selection in patients with complex PITS. For patients with complex PITS, interventional bronchoscopic procedures are effective in treating symptoms that require intensive care admission. However, surgical evaluation should be conducted, and patients should be thoroughly analyzed to determine the necessity of surgical intervention.

Interventional bronchoscopic procedures are also less invasive and can provide treatment for patients with significant comorbidities or longer stenosis. Regarding the efficient and effective use of health services, the pre-diagnosis of PITS should not be forgotten in ICU patients. At the same time, bronchoscopy should not be delayed in patients with a preliminary diagnosis of PITS, and it should be known that bronchoscopy is not only a diagnostic tool but also a treatment tool. More comprehensive prospective studies are needed on ICU patients on this subject.