Tracheomalacia after thyroidectomy for retrosternal goitres requiring sternotomy– a myth or reality?

Abstract

Introduction

Tracheomalacia after thyroidectomy is not well understood. Reports on tracheomalacia are conflicting, with some suggesting a high rate and other large cohorts in which no tracheomalacia is reported. The aim of our study was to assess the incidence and factors associated with tracheomalacia after thyroidectomy in patients with retrosternal goitres requiring sternotomy at a high-volume tertiary care referral centre.

Methods

A longitudinal cohort study was conducted from January 2011 to December 2019. All adult patients who underwent thyroidectomy with sternotomy were included. Tracheomalacia was considered when tracheal rings were soft compared with other parts (proximal or distal) of the trachea and required either tracheostomy or resection with anastomosis. The decision to perform a tracheostomy or to administer continuous or bilevel positive airway pressure postoperatively was made depending on the degree of tracheomalacia. Logistic regression analysis was used to assess factors associated with tracheomalacia.

Results

We evaluated 40 patients who underwent thyroidectomy with sternotomy. The mean age of our cohort was 48.7 ± 11.3 years and the population was predominantly female (67.5%). One patient required tracheal resection with anastomosis, and two patients required tracheostomy. Multivariable logistic regression analysis did not reveal any patient- or thyroid-related factor significantly associated with the development of tracheomalacia in our cohort.

Conclusions

The incidence of tracheomalacia after thyroidectomy with sternotomy appears to be very low. However, the occurrence of tracheomalacia after thyroidectomy in cases of large goitre is possible and hence worrisome.

Introduction

Retrosternal goitres have posed a challenge to surgeons for a long time. They are most commonly defined as a thyroid mass with 50% or more lying below the thoracic inlet.¹ Another definition is a thyroid mass that extends 3cm or more below the suprasternal notch when the neck is hyperextended. The incidence of retrosternal extension ranges from 1% to 20% of goitres depending on the definition used.² Retrosternal goitres most commonly present with symptoms related to airway compression, including dyspnoea, dysphagia, sleep disturbance and hoarseness.³ There is a consensus that surgical removal is the desired management option for retrosternal goitres and most can be operated upon using a cervical approach; a minority of patients may require a partial or full sternotomy. However, surgical removal of retrosternal goitres may lead to various complications such as hypoparathyroidism and permanent nerve injury, and notably may exacerbate pre-existing tracheomalacia.⁴

Tracheomalacia is defined as an inability of the cartilaginous framework of the trachea to maintain airway patency when reduction of the cross-sectional area of the trachea is greater than 50%.⁵ Tracheomalacia can be due to either congenital immaturity of the tracheal cartilage, ie primary tracheomalacia, or previously normal cartilage that undergoes degeneration due to acquired factors such as external compression or a chronic inflammatory response, ie secondary tracheomalacia.⁶ Long-standing compression by a large goitre considerably weakens the tracheal rings and leads to reduced tracheal support. This increases the chances of collapse after thyroidectomy, resulting in a possibly detrimental postoperative airway obstruction.⁷

The current gold standard for diagnosis is bronchoscopy, which allows direct visualisation of the severity and extent of airway collapse.⁸ The severity of airway malacia is conventionally graded by the degree of airway collapse: normal, less than 50%; mild, 50–75%; moderate, 75–90%; and severe, 91–100%.⁹ However, when forming a diagnosis it is necessary to exclude other causes of airway obstruction (oedema, secretions, recurrent nerve paralysis and haematoma).¹⁰ Because of the rarity of tracheomalacia, its management has not been standardised. Suggested management techniques are diverse and include tracheostomy, prolonged endotracheal intubation, tracheopexy, internal stent and external splinting with Marlex mesh.¹¹

In addition to being supposedly rare, tracheomalacia is poorly understood. The incidence of tracheomalacia post thyroidectomy has been reported at between 0.8% to 5.8% in studies conducted at various centres.¹² Tracheal compression is likely to be a causative variable in the development of post-thyroidectomy tracheomalacia but this is not completely established.¹³ The aim of our study was to assess the incidence and factors associated with tracheomalacia after thyroidectomy in patients with retrosternal goitres requiring sternotomy at a high-volume tertiary care referral centre.

Methods

A cross-sectional study (data collected retrospectively) was conducted from January 2011 to December 2019. The study received institutional review board exemption (2018-0385-519). Data were collected on patient demographics and treatment-related variables. All adult patients who underwent thyroidectomy with sternotomy were included in the study. All patients underwent total thyroidectomy for either compressive symptoms or malignancy suspected on fine-needle aspiration cytology. Patients with missing data and those who did not give informed consent were excluded.

The cervical component of the thyroid was operated on by one otolaryngologist/head and neck surgeon. Sternotomy, along with dissection of the thoracic component of the thyroid gland, was performed by a single experienced cardiothoracic surgeon. All patients were admitted under the cardiothoracic service. The decision to keep patients in the intensive care unit (ICU) for observation immediately post surgery was taken by the cardiothoracic surgeon. Both surgeons had more than 15 years’ experience. Chest tubes were inserted after sternotomy and subsequently removed depending on respiratory parameters and patient progress.

Tracheomalacia was considered when tracheal rings were soft compared with other parts (proximal or distal) of the trachea and required either tracheostomy or resection with anastomosis. Tracheomalacia was observed after complete removal of the thyroid gland. The decision to perform tracheostomy or to administer continuous positive airway pressure (CPAP) or bilevel positive airway pressure (BiPAP) postoperatively was based on the degree of tracheomalacia. If the trachea collapsed completely on removal of the endotracheal tube leading to stridor, tracheostomy was performed. If there was slight softening of trachea, the patient was put on CPAP/BiPAP postoperatively and was monitored closely in the ICU. Furthermore, we assessed vocal cord mobility at the time of extubation. Patients with any cord paresis or paralysis were not included in our study as stridor was not secondary to tracheal collapse.

Continuous data are presented as means and standard deviation (SD). Categorical data are presented as frequency and percentage. We assessed age, gender, duration of thyroid swelling, body mass index, histology (benign/malignant) and sternotomy (full/partial or mini sternotomy) for association with tracheomalacia using logistic regression analysis. A p-value < 0.05 was taken as significant. Stata version 12 was used for data analysis.

Results

Our longitudinal cohort study evaluated 40 patients for the presence of tracheomalacia post thyroidectomy performed for retrosternal goitre requiring sternotomy. Mean patient age was 48.7 ± 11.3 years and the population was predominantly female (67.5%). Table 1 describes the characteristics of our study population. Mean weight of our study population was 77.2 ± 21.62kg. Intraoperatively, all of our study population underwent sternotomy, either partial (60.0%) or full (40.0%). Tracheal resection and anastomosis were essential in only one patient, and two patients required tracheostomy (Table 2).

Table 1 .

Characteristics of study population

Age, years	48.73 ± 11.259
Gender, n (%)
Male	13 (32.5)
Female	27 (67.5)
Weight (kg)	77.167 ± 21.6150

Table 2 .

Intraoperative characteristics

	n (%)
Tracheal resection and anastomosis
Yes	1 (2.5)
No	39 (97.5)
Sternotomy
Full	16 (40.0)
Partial	24 (60.0)
Tracheostomy
Yes	2 (5.0)
No	38 (95.0)

Histology following the procedure showed: 70.0% benign thyroid, 15.0% papillary, 2.5% follicular and 2.5% anaplastic (Table 3). Mean length of hospital stay was 7.10 ± 4.1 days, and 47.5% of patients also had an ICU stay (Table 4). The requirement for a tracheostomy or administration of CPAP or BiPAP was based on the degree of tracheomalacia. It was found that a tracheostomy was required in only 2 of 40 patients. BiPAP/CPAP was required by 10.0% of our patients, whereas the need did not arise for the majority of patients (90%) (Table 4).

Table 3 .

Histology report

	n (%)
Benign thyroid	28 (70.0)
Follicular	1 (2.5)
Papillary	6 (15.0)
Anaplastic	1 (2.5)
Other	4 (10.0)

Table 4 .

Postoperative characteristics

Duration of hospital stay, days	7.10 ± 4.043
Postoperative ICU stay, n (%)
Yes	19 (47.5)
No	21 (52.5)
Postoperative use of BiPAP/CPAP, n (%)
Yes	4 (10.0)
No	36 (90.0)
Use of other respiratory device, n (%)
Nebuliser	2 (5.0)
Oxygen dependence	1 (2.5)
No	37 (92.5)

ICU = intensive care unit; BiPAP = bilevel positive airway pressure; CPAP = continuous positive airway pressure

Multivariable logistic regression analysis did not reveal any patient- or thyroid-related factor significantly associated with the development of tracheomalacia in our cohort (Table 5).

Table 5 .

Multivariable logistic regression analysis of factors associated with tracheomalacia in patients undergone thyroidectomy with sternotomy

Factors	Coefficient	Standard error	95% confidence interval	p-value*
Age (years)	0.02	0.04	−0.06 to 0.11	0.59
Gender
Female	1	–	–	–
Male	−0.41	1.20	−2.77 to 1.96	0.73
BMI	0.08	0.07	−0.06 to 0.21	0.27
Duration of swelling (months)	0.003	0.01	−0.01 to 0.02	0.62
Histology
Benign Malignant	−0.27	1.21	−2.64 to 2.09	0.82
Sternotomy
No	1	–	–	–
Yes	0.45	1.06	−1.62 to 2.52	0.67

*A p-value < 0.05 is significant

Discussion

Long-standing large goitres usually have retrosternal extension causing tracheal compression and making thyroidectomy the only possible treatment. Post-thyroidectomy tracheomalacia is a serious but rare complication. The incidence of tracheomalacia post thyroidectomy is reported to be between 0.8% and 5.8%.¹² In our study, 3 of 40 (7.5%) patients had tracheomalacia post thyroidectomy. Tracheal resection and anastomosis was required in only one patient (2.5%) and tracheostomy post thyroidectomy was required in two (5%). Logistic regression analysis of our data collected over a period 8 years and encompassing 40 patients, found no significant association between undergoing thyroidectomy with sternotomy and development of tracheomalacia post thyroidectomy.

A number of studies have assessed the risk of tracheomalacia as a post-thyroidectomy respiratory complication. In comparison with our study, Agarwal et al (2007) found a 3.1% incidence of post-thyroidectomy tracheomalacia, which was equivalent to 28 of 900 thyroidectomies performed.⁷ As one treatment option, tracheostomy was performed in 26 patients, whereas 2 patients were put on prolonged intubation.⁷ Tracheostomy leads to fibrosis around a soft trachea, providing the patient with early recovery from tracheomalacia.⁷ It was therefore concluded that tracheomalacia is a complication likely to be seen in a prevailing goitre. This study also identified the possible criteria for making an early diagnosis of tracheomalacia and put forward various management strategies, of which tracheostomy was deemed to be the best option.⁷ Similarly, the presence of tracheomalacia as a complication post thyroidectomy was found in a study conducted by Chauhan and colleagues that reviewed 199 cases of retrosternal goitre treated with thyroidectomy over a period of 14 years. Tracheomalacia was found to be present in 1% of the patients.¹⁴ In 2004, Bennett and colleagues reviewed 12 studies with a total of 1,969 patients and found that 0.3% had tracheomalacia and 0.9% required tracheostomy.¹⁵ All six tracheomalacia cases were reported from a single prospective cohort study of 103 patients from Sudan. Factors common among the target population were identified; it was found that these patients underwent thyroidectomy for large goitres, had neck circumferences of more than 40cm and had a goitre duration of more than 5 years.¹⁵ These conclusions were consistent with the results of our study.

A retrospective review at a UK tertiary referral centre by Findley and colleagues also explored the risk of post-thyroidectomy tracheomalacia in patients with tracheal compression. This study aimed to stratify and quantify the factors leading to an underlying degree of tracheal compression by retrosternal extension of the thyroid.¹³ A total of 334 patients (79% female, 21% male) underwent thyroid surgery between January 2008 and July 2010. The study found no incidence of post-thyroidectomy tracheomalacia (95% CI 0.0–4.8%).¹³ Furthermore, in a series of 200 large cervical and substernal goitres treated at Massachusetts Eye and Ear Infirmary and Massachusetts General Hospital, Randolph and colleagues found not one case of tracheomalacia, even in the setting of chronic significant tracheal deviation, compression and remodelling with massive and recurrent goiters.¹⁶ Most recently, in a retrospective study that assessed 1,236 patients,¹⁷ Valizadeh and colleagues found that, for data collected between 2007 and 2017, if the diagnosis of tracheomalacia is performed only clinically, most post-thyroidectomy respiratory problems such as hoarseness will be mistakenly reported as tracheomalacia; by contrast, bronchoscopy, the gold standard for diagnosis, showed hoarseness to be due to bilateral recurrent laryngeal nerve palsy.¹⁷ These studies revealed no occurrence of tracheomalacia post thyroidectomy, findings which are different from ours.

A retrospective observational study of a rare outcome has its own set of limitations. Small sample size was the most significant limitation in our study. Also, because this was an adult-based study, cases of thyroidectomy from the paediatric population were excluded, making the sample size even smaller. Owing to the retrospective design of this study, we could not evaluate other risk factors such as degree of tracheal compression preoperatively and degree of retrosternal extension because most patients had their radiological investigations done at other centres. Similarly, neuromonitoring is not usually undertaken at our centre, and so information was not present for all patients in this regard. With very few patients with malignancy, data analysis through regression was not possible as had we segregated pathology. Also, importantly, our operational definition of tracheomalacia could vary because to date there is no definitive grading system for tracheomalacia.

Although limited by the small sample size, our study finding has high external validity because retrosternal goitre is encountered almost wherever thyroidectomies are performed. Similarly, our study has high internal validity because the same otolaryngologist and cardiothoracic surgeon operated on all patients. Our study highlights a very controversial, yet important aspect of treating patients with retrosternal goitres.

Conclusion

The incidence of tracheomalacia after thyroidectomy with sternotomy appears to be very low. However, the occurrence of tracheomalacia after thyroidectomy in cases of large goitre is possible and hence worrisome, although with the necessary precautions long-term complications and mortality can be avoided. Further prospective cohort studies are recommended to investigate the incidence of tracheomalacia and its risk factors in patients with large thyroids with retrosternal extension.