Thyroid-split tracheostomy: short- and long-term outcomes| ACTA Otorhinolaryngologica Italica

Abstract

Objective. Compare long-term outcomes of thyroid-split and standard thyroid-retraction tracheostomy in a large cohort.
Methods. The healthcare database of a university-affiliated hospital was searched for past patients over 18 years of age from all of the hospital’s wards on whom an ear, nose, and throat specialist performed a tracheostomy in the operating room between 2010 and 2020. Clinical data were extracted from the hospital and outpatient medical records. Life-threatening and non-life-threatening intra-operative and early and late post-operative adverse events in patients who underwent split-thyroid tracheostomy were compared with those who underwent standard tracheostomy.
Results. There was no significant difference in intra-operative and early post-operative complications, hospitalisation length, or early reoperation and death rates between the 140 (28%) thyroid-split tracheostomy patients and the 354 (72%) standard tracheostomy patients, although the thyroid-split group had more non-decannulated patients and a longer operative time.
Conclusions. Thyroid-split tracheostomy is safe and feasible. Compared to the standard procedure, it provides better exposure and a similar rate of complications, although its decannulation success rate is lower.

Introduction

Tracheostomy was first performed in 100 BC, making it among the oldest surgical procedures ¹. Although translaryngeal intubation is the preferred initial artificial airway in mechanically ventilated patients, early tracheostomy accelerates the weaning process and reduces aspiration and ventilator-acquired pneumonia ². Thus, tracheostomy is one of the most commonly performed surgical procedures in critically-ill patients who require airway protection or long-term mechanical ventilation ^3,4. Tracheostomy also has established benefits for patients who need prolonged invasive mechanical ventilation, including weaning from relaxation and sedation drugs, facilitating pulmonary hygiene, decreasing dead space, enhancing weaning from mechanical ventilation, decreasing subglottic stenosis and shortening the intensive care unit (ICU) stay ⁵.

The indications for tracheostomy, the complex underlying conditions of the patients, and the involvement of the airway increase the tendency for complications. Early complications occurring within 30 days of tracheostomy include bleeding, stomal infection, injuries to nearby structures, subcutaneous emphysema, pneumothorax, tracheostomy tube obstruction and tube dislodgement. Late complications, occurring after 30 days, include tracheal stenosis or malacia, followed by trachea-oesophageal fistula, trachea-innominate fistula and tracheocutaneous fistula (TCF) ⁶. Complication rates reported in the literature range from 6% to 66% depending on the study design, duration of follow-up and specific definition of complications, with mortality rates up to 5% ^1,7,8. However, more recent studies have reported infrequent complications, ranging from minor to life-threatening ⁹.

Traditionally, tracheostomy has been performed as an open surgical procedure in the operating room (OR). However, percutaneous tracheostomy has become more common in recent decades, primarily performed at bedside in the ICU. As a result, many different methods have arisen over time, although their superiority in terms of complication rates and outcomes remains unclear ¹⁰. One controversy involves the approach to the thyroid isthmus, which can be preserved by retracting it superiorly (standard tracheostomy) or splitting it for better exposure of the trachea ¹¹. Studies comparing the two have reported similar complication rates ^11,12, but included only small patient cohorts with a short follow-up of six months.

This study compared the outcomes of thyroid-split and standard thyroid-retraction tracheostomy in a large cohort over a long period of follow-up.

Materials and methods

Design and setting

The healthcare database of a university-affiliated hospital was retrospectively searched for patients from all of the hospital’s wards, including the ICU, aged over 18 years on whom an ear, nose, and throat specialist (ENT) performed a tracheostomy for any indication in the OR between 2010 and 2020. Possible tracheostomy indications are prolonged intubation, protective tracheostomy, airway obstruction by a tumour or oedema, bilateral vocal cord paralysis, glottis or subglottis stenosis, and trauma (Tab. I). Patients were divided into two groups according to the type of tracheostomy they underwent: thyroid-split or standard.

Data collection

Clinical data were extracted from hospital and outpatient medical records. The patient data included demographics, clinical history, laboratory tests and date of death, and were collected from each patient’s computerised medical records. Hospitalisation data included antibiotic treatment, length of hospitalisation, and short-term post-operative complications (within 30 days after surgery) and were collected from the patient’s computerised admission-transfer-discharge (ATD) records. The tracheostomy data included the method, date, intra-operative complications and operative time obtained from surgical reports. Patients were followed until death or the end of the research period (June 2021). Follow-up data, including long-term complications (occurring more than 30 days after surgery) and cause of death, were obtained from follow-up meeting reports or computerised medical records.

Surgical approach

After evaluating the clinical course and prognosis, critical care consultants and ENT surgeons decided to perform a tracheostomy jointly. The choices of technique, operator, and equipment were based primarily on local resources and expertise, but patient factors, including body habitus, adequacy of neck extension and laryngoscopy grade, were also considered. A senior otorhinolaryngologist and an ENT resident performed the tracheostomy by either retraction or thyroid split. The intra-operative haemostasis technique involved high-energy electrocautery with the LigaSureTM (Valleylab, Boulder, CO) or Rex^® Harmonic Scalpel System (Ethicon EndoSurgery, Cincinnati, OH). The Bjork flap or tracheal flap technique is not common in our institution. Finally, the cannula flanges were fixed with stitches to the skin in each procedure. The tracheostomy multidisciplinary team managed de-cannulation and performed it in the ENT ward.

Outcomes

The study’s primary outcomes were intra-operative and direct post-operative life-threatening adverse events (AEs). Intra-operative life-threatening AEs were defined as significant vascular injuries or excessive bleeding (determined by the need for blood transfusion or referred to in the surgical report as severe intra-operative bleeding) or a misplaced airway (paratracheal insertion of the tracheostomy tube). Direct post-operative severe, life-threatening AEs were defined as excessive bleeding (determined by the need for blood transfusion or an additional surgical procedure) and/or cannula dislodgement. Premature mortality was defined as death within 30 days of the tracheostomy.

The secondary outcomes of this study were: (i) intra-operative, post-operative and late non-life-threatening events (non-LTEs); (ii) duration of the procedure; and (iii) 7-day survival. Intra-operative non-LTEs were defined as transient hypotension or technical difficulties. Among the direct post-operative non-LTEs were peristomal infection (defined as a body temperature above 38°C or local peristomal swelling, redness and secretions), tracheal and oesophageal damage as evaluated by either endoscopy or imaging, and aspiration during the post-operative hospitalisation. The late non-LTEs were tracheal stenosis, tracheal malacia, TCF, granuloma development, re-operation within a month of the surgery and hypothyroidism [the first thyroid-stimulating hormone (TSH) test after the tracheostomy]. Our 7-day survival analysis was based on all-cause mortality.

Post-operative complications based on the Clavien-Dindo Classification System

Post-operative complications were further categorised using the Clavien-Dindo Classification System (Appendix 1) ¹³. Two authors (NT and ES) independently assigned a Clavien-Dindo grade to each post-operative complication. Any disagreements were deliberated and decided jointly. Complication grade frequency was defined as the number of times the complication grade occurred divided by the sum of the post-operative complications. In this analysis, each patient could have multiple complications. However, only the highest complication grade for each patient was considered when comparing post-operative complication grades between tracheostomy techniques.

Statistical analysis

Statistical analyses were performed using SPSS version 25 (IBM Corp. Armonk, NY). Continuous variables with a normal distribution are presented as means and standard deviations and compared between groups with t-tests. Continuous variables with a non-normal distribution are presented as medians and ranges and analysed using the Mann-Whitney and Wilcoxon signed-rank tests for dependent samples. Categorical variables are presented as percentages and analysed using the chi-square or Fisher’s exact test. Multiple comparisons were adjusted using the Bonferroni correction. Survival was calculated from the surgery date to the date of death or last follow-up. Univariable survival models and curves were generated using the Kaplan-Meier method and compared using a log-rank test. All the tests were two-sided, and the threshold for statistical significance was set at a p-value of < 0.05.

Results

Study population and demographics

The cohort consisted of 494 patients, of whom 140 (28%) had thyroid-split tracheostomy and 354 (72%) had standard tracheostomy. Their characteristics are shown in Table I. There were no significant differences in demographics and comorbidities except for a higher prevalence of chronic obstructive pulmonary disease in the standard-tracheostomy group (23.1% vs 10.1%, p < 0.001). However, indications for tracheostomy differed between the groups. The standard-tracheostomy group had higher rates of prolonged intubation (47.5% vs 28.3%, p = 0.001), and the thyroid-split tracheostomy group had higher rates of the protective airway (26.8% vs 15.5%, p = 0.002) and airway obstruction by oedema (15.9% vs 9.9%, p = 0.04). Furthermore, we performed a subanalysis of 198 patients who underwent tracheostomy due to prolonged intubation to assess differences in the indications between the groups (Tab. II). A significant difference was shown in the indications for prolonged intubation (p = 0.042), with a higher prevalence of reduced consciousness in the thyroid-split group (33.3% vs 10.1%, p < 0.001).

Complications

Nine (2%) emergent tracheal dislodgements occurred: six in the thyroid-split group and three in the standard-tracheostomy group. The mean time of occurrence of these dislodgements from surgery was on post-operative Day 5, between 3 p.m. and 7 a.m. Six cannulas were relocated to the patient’s bedside, and one went into the OR. Two patients with tracheal dislodgment could not be relocated and died due to airway obstruction and hypoxaemia. Of them, one had undergone a thyroid-split tracheostomy and the other had undergone a standard tracheostomy.

The early and late complications are shown in Tables III and IV, respectively. The rates of early post-operative complications of bleeding, pneumothorax, infection, canula dislocation, aspiration, tracheal damage and oesophageal damage did not differ between groups. However, the operative time of the thyroid-split tracheostomy group was significantly longer than that of the standard-tracheostomy group (median, 54 min vs 40 min, respectively, p < 0.001).

The late complications of tracheal stenosis, tracheomalacia, TCF, granuloma formation, and all-cause mortality (41% in thyroid slit group vs 48.9% in standard method group) did not differ between groups. However, the percentage of patients who were not decannulated was higher in the thyroid-split tracheostomy group (57.1% vs 44.4%, p = 0.039).

The Clavien-Dindo Classification System for post-operative complications

According to the Clavien-Dindo Classification System (Appendix 1), one or more post-operative complications occurred in 285 (57.7%) patients, 40.78% of which were classified as Grade 1; 20.61%, Grade 2; 11.06%, Grade 3a; 24.95%, Grade 3b; and 2.6%, Grade 5. No complication was classified as Grade 4. According to the Clavien-Dindo Classification, the severity of complications did not differ significantly between groups (p = 0.340), as shown in Table V. Likewise, when complications were classified into low or high grades (defined as Grade 3 and above), there was no difference between the tracheostomy technique groups (p = 0.059).

Survival

Our 7-day survival evaluation showed a significant difference between groups (Fig. 1). The thyroid-split group had a survival rate of 98.6% after 9 days compared to 94.6% in the standard-tracheostomy group (p = 0.05).

Subpopulations

We performed subpopulation analyses to determine if specific patient groups were more prone to tracheostomy complications. The complication rates and lengths of hospitalisation of the 254 patients older than 65 years did not differ significantly according to the type of tracheostomy performed. These findings were similar to those in the subanalysis of 102 obese patients (BMI > 30). However, in the subanalysis on indications for tracheostomy, among the 82 patients who underwent the procedure because of prolonged intubation and on whom data on planned elective de-cannulation was available, those who had a thyroid-split tracheostomy had a lower de-cannulation rate than those who had a standard tracheostomy (57.6% vs 87%, respectively, p = 0.018).

Discussion

This single-centre retrospective cohort study compared standard and thyroid-split tracheostomy in 494 patients (Fig. 2). To our knowledge, few studies have compared these techniques even though they are performed daily in many ENT surgical procedures. Despite decades of experience, the preferred tracheostomy technique in critically-ill patients is still debatable, and newer, less invasive procedures that often do not require an OR are presented yearly at intensive care symposia, with potential differences in outcomes, complication rates and costs ^5,14. Furthermore, in this study, participants were seemingly homogenous in terms of demographics, medical history and overall mortality, allowing for comparison of the effects of standard vs thyroid-split tracheotomy.

In a prospective tracheostomy study, Calhoun et al. ¹² compared electrocautery division to clamp-and-tie. They found that the former had a shorter operative time and the same complication rate. However, in a more extensive study of 1,143 patients by Janik et al. ¹¹ that compared isthmus preservation with suture ligation of the thyroid isthmus, no significant difference in complication rates between the two groups was reported. Our study refined the outcome analysis by differentiating early and late post-operative complications and including many different and specific difficulties. We also analysed both life-threatening and non-life-threatening complications. Overall, the complication profile was similar for the thyroid-split and standard techniques, consistent with earlier studies.

Cannula dislodgement is a life-threatening tracheostomy complication associated with 50% of airway-related deaths ¹⁶. Its rates in the literature range from 0.35% to 15% ^17,18. Various methods of decreasing the rate of emergency cannula dislodgement have been described, including sutured neckties ¹⁹, using specific fresh tracheostomy guidelines ²⁰, and establishing airway rapid response teams ²¹. In our study, dislodgement rates did not significantly differ between the thyroid-split and standard tracheostomy groups. Moreover, we evaluated several variables, including anoxic brain damage, dislodgement-related death and bedside recannulation. However, dislodgement is fortunately rare, with only 0.2% of our cohort experiencing it. Given the devastating potential of this complication, however, thyroid split can offer potential advantages over the standard tracheostomy technique, as it allows for a better view of the stoma and easier insertion. In our cohort, surgeons chose the thyroid-split technique over the standard-tracheostomy technique for patients when short-term de-cannulation was not anticipated, but the prognosis was deemed favourable, such as decreased consciousness (33% vs 10%, p < 0.001). The surgeons’ choice of the thyroid-split over standard technique in this study can be attributed to the potential advantage of the thyroid-split technique due to its more stable tract for potential dislodgements. The devastating potential of emergent dislodgement and the rarity of this phenomenon require evaluation in further studies using a caregiver-based survey that will assess the ease of recannulation in different tracheostomy techniques.

At the same time, the thyroid-split technique has several theoretical disadvantages. First, manipulating the highly vascularised thyroid gland may cause intra-operative bleeding, requiring more extensive exposure and blood transfusion with an overall prolonged operative procedure ²². However, our study showed no increase in intra-operative or post-operative bleeding, probably due to our meticulous intra-operative haemostasis technique during thyroid-split. Second, during thyroid split, exposing the medial margins of the thyroid to an unsterilised environment near the stoma may increase the peri-operative infection rate ¹⁰. However, we did not observe an increased post-operative infection rate in the thyroid-split group compared to the standard-tracheostomy group. This finding may be explained by a shift in both thyroid lobes posteriorly during the procedure, away from the stoma, thereby lessening the potential for respiratory pathogen contamination. Moreover, the high-energy split technique could also have contributed to the local sterilisation of thyroid tissue and decreased bacterial biofilm degradation, as demonstrated in a previous study ²³.

When we evaluated planned de-cannulation, we assumed that an intact thyroid lobe, together with subcutaneous tissue and strap muscles, can serve as a soft-tissue barrier between the trachea and the skin, which could lead to a higher likelihood of de-cannulation success and a decreased risk of a permanent TCF. Furthermore, in our study, the de-cannulation success rate was 42.9% in the thyroid-split group and 55.6% in the standard-tracheostomy group (p = 0.039). Known predictors of elective tracheostomy de-cannulation include older age, BMI < 22 kg/m², respiratory indication for tracheostomy and neurologic disease ^24-28. However, no evidence has been reported that the location of the thyroid isthmus relative to the location of the stoma is a predictor of de-cannulation success. Therefore, we assumed that the factors contributing to the lower de-cannulation rates were, first, the indications for tracheostomy, which showed that the thyroid split was performed for patients as a protective measure against airway obstruction. Thus, physicians may be reluctant to try early de-cannulation due to airway compromise and opt to practice a more conservative de-cannulation approach. Moreover, the thyroid isthmus can be a vascularised flap between the trachea and the skin after de-cannulation, offering better wound healing and successful de-cannulation.

Thyroid split was associated with a longer surgical duration than standard tracheostomy (median of 54 vs 40 min). This can be attributed to the need to expose the thyroid isthmus, split it by electrocautery and maintain meticulous haemostasis. Moreover, in our tertiary centre, residents mostly perform thyroid split to increase tracheal visibility and thus improve manual and surgical manipulation of the thyroid gland (for future surgeries). However, this habit might increase surgical time.

Overall, our results are consistent with those in previous publications, suggesting that thyroid-split tracheostomy is as safe as standard thyroid-retraction tracheostomy and can be applied according to the surgeon’s preference and need for better exposure. Indeed, when addressing the thyroid-split tracheostomy indications, our study showed that surgeons choose the thyroid-split technique when addressing protective and airway obstruction (Tab. I), with the probable advantage of easier tracheal exposure in emergency cannula dislodgement in patients with complicated orotracheal intubation. Finally, the potential disadvantage of the thyroid-split technique was prolonged surgical time and decreased elective de-cannulation.

Due to the study’s retrospective design, we did not have information regarding other factors that might influence the decision on tracheostomy technique, such as thyroid gland volume or neck circumference. However, further prospective studies evaluating both anthropometric measures and gland volume may shed light on the potential advantages of the thyroid-split technique for specific subpopulations, including patients with a short neck, increased neck circumference and large retrosternal thyroid gland.

This study has several limitations. Its retrospective design with manual data extraction from medical records is inherently subject to selection and information bias. However, we focused on a large cohort, detailed early and late complication profiles, and extended the follow-up period. Furthermore, our choice of patients suitable for tracheostomy might have been influenced by the patient’s medical history, demographics and indications. It is noteworthy that patients hospitalised in the ICU and undergoing tracheostomy have a high mortality rate due to their primary condition, regardless of the tracheostomy procedure itself. Thus, their follow-up duration may be shorter, and the frequency of long-term post-operative complications is lower. Missing data could also have influenced the results. In this study, data on the surgical duration and the intra-operative complications were limited due to gaps in the surgical reports of the different members of the anesthesia and surgical teams. Finally, as stated earlier, tracheostomy is one of the most common surgical techniques and has a high safety profile and a low complication rate ^5,9,16. Therefore, even in our relatively large cohort, a low rate of complications could be expected, with low power for statistical analysis. A multicentre study can overcome this problem.

Conclusions

Thyroid split is a safe and feasible tracheostomy technique. Its rate of complications is similar to that of the standard procedure. It was the surgeon’s technique of choice for tracheostomy to address imminent airway obstruction, probably due to the better tracheal exposure associated with it. The abovementioned advantages should outweigh the disadvantages of thyroid-split tracheostomy of longer operational length and decreased elective de-cannulation rates.

Conflict of interest statement

The authors declare no conflict of interest.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Author contributions

NT, EN: conceptualisation; NT, ES, EY: methodology; NT: project administration; NT, ES: writing – original draft; NT, ES, EY, NK, IZ, ER, OW, GB, EN: writing – review and editing; ES, NK, IZ: data curation; ES: statistical analysis; GB, EN: supervision.

Ethical consideration

This study was approved by th Institutional Ethics Board (protocol number RMC 0651-2020).

The Institutional Review Board, waived the need for informed consent due to the retrospective nature of this research.

Figures and tables

Figure 1.7-day survival (Kaplan-Meier).

Figure 2.Thyroid split tracheostomy technique showing (A) before the thyroid split and (B) after the thyroid split, exposing the trachea above and below the cartilage window location (From De Leyn, et al., 2007, mod. Prof. Dr. P. De Leyn has permitted the author to use the figure in date January 20, 2023) ¹⁵.

References

1. Johnson-Obaseki S, Veljkovic A. Complication rates of open surgical versus percutaneous tracheostomy in critically ill patients. Laryngoscope. 2016; 126:2459-2467. DOI
2. Adly A, Youssef TA, El-Begermy MM. Timing of tracheostomy in patients with prolonged endotracheal intubation: a systematic review. Eur Arch Otorhinolaryngol. 2018; 275:679-690. DOI
3. Wood DE. Tracheostomy. Chest Surg Clin N Am. 1996; 6:749-764.
4. Freeman BD, Kennedy C, Robertson TE. Tracheostomy protocol: Experience with development and potential utility. Crit Care Med. 2008; 36:1742-1748. DOI
5. Cheung NH, Napolitano LM. Tracheostomy: epidemiology, indications, timing, technique, and outcomes. Respir Care. 2014; 59:895-915. DOI
6. Fernandez-Bussy S, Mahajan B, Folch E. Tracheostomy tube placement. J Bronchol Interv Pulmonol. 2015; 22:357-364. DOI
7. Chew JY, Cantrell RW. Tracheostomy: complications and their management. Arch Otolaryngol. 1972; 96:538-545.
8. Hazard PB, Garrett HE, Adams JW. Bedside percutaneous tracheostomy: experience with 55 elective procedures. Ann Thorac Surg. 1988; 46:63-67.
9. Bathula SS, Srikantha L, Patrick T. Immediate post-operative complications in adult tracheostomy. Cureus. 2020; 12:E12228. DOI
10. Brass P, Hellmich M, Ladra A. Percutaneous techniques versus surgical techniques for tracheostomy. Cochrane Database Syst Rev. 2016; 7:CD008045. DOI
11. Janik S, Kliman J, Hacker P. Preserving the thyroidal isthmus during low tracheostomy with creation of a Björk flap. Laryngoscope. 2018; 128:2783-2789. DOI
12. Calhoun KH, Weiss RL, Scott B. Management of the thyroid isthmus in tracheostomy: a prospective and retrospective study. Otolaryngol Head Neck Surg. 1994; 111:450-452. DOI
13. Dindo D, Demartines N, Clavien PA. Classification of surgical complications: A new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg. 2004; 240:205-213. DOI
14. Divisi D, Stati G, De Vico A. Is percutaneous tracheostomy the best method in the management of patients with prolonged mechanical ventilation?. Respir Med Case Rep. 2015; 16:69-70. DOI
15. De Leyn P, Bedert L, Delcroix M. Tracheotomy: clinical review and guidelines. Eur J Cardiothorac Surg. 2007; 32:412-421. DOI
16. McGrath BA, Bates L, Atkinson D. Multidisciplinary guidelines for the management of tracheostomy and laryngectomy airway emergencies. Anaesthesia. 2012; 67:1025-1041. DOI
17. Friedman Y, Fildes J, Mizock B. Comparison of percutaneous and surgical tracheostomies. Chest. 1996; 110:480-485.
18. Halum SL, Ting JY, Plowman EK. A multi-institutional analysis of tracheotomy complications. Laryngoscope. 2012; 122:38-45. DOI
19. Fine KE, Wi MS, Kovalev V. Comparing the tracheostomy dislodgement and complication rate of non-sutured neck tie to skin sutured neck tie fixation. Am J Otolaryngol. 2021; 42:102791. DOI
20. Hood K, Lewis B, Bowens DC. Reducing fresh tracheostomy decannulations following implementation of a fresh tracheostomy guideline. Crit Care Nurs Clin North Am. 29:131-141. DOI
21. Rassekh CH, Zhao J, Martin ND. Tracheostomy complications as a trigger for an airway rapid response: analysis and quality improvement considerations. Otolaryngol Head Neck Surg. 2015; 153:921-926. DOI
22. Bontempo LJ, Manning SL. Tracheostomy emergencies. Emerg Med Clin North Am. 2019; 37:109-119. DOI
23. Xiao B, Zou Z, Bhandari J. Exposure to diode laser (810nm) affects the bacterial adherence and biofilm formation in a E. faecalis biofilm model. Photodiagnosis Photodyn Ther. 2020; 31:101772. DOI
24. Park C, Ko RE, Jung J. Prediction of successful de-cannulation of tracheostomised patients in medical intensive care units. Respir Res. 2021; 22:131. DOI
25. Cankar Dal H. Difficult airway management and emergency tracheostomy in a patient with giant goiter presenting with respiratory arrest: a case report. Exp Ther Med. 2022; 24:499. DOI
26. Osinaike B, Ogunsiji A, Joseph O. Challenging airway management in a patient with retrosternal goiter presenting in respiratory distress. Niger J Surg. 2021; 27:66-70. DOI
27. Kim SM, Kim HJ. Successful advancement of endotracheal tube with combined fiberoptic bronchoscopy and videolaryngoscopy in a patient with a huge goiter. SAGE Open Med Case Rep. 2020; 8:2050313X20923232. DOI
28. Omokanye HK, Dunmade AD, Segun-Busari S. Accidental decannulation of tracheostomy tubes - Case Series. J West Afr Coll Surg. 6:108-118.