Circumferential Laryngotracheal Resection in Thyroid Cancer: Experience and Outcome in a Single Center

Abstract

There is limited experience of laryngotracheal resection in patients with differentiated thyroid carcinoma (DTC). The aim of this study was to report our experience of circumferential laryngotracheal resection in DTC and its long-term outcome. In this retrospective study, 10 patients of locally invasive DTC who underwent circumferential laryngotracheal resection between January 2000 and December 2015 were included. Clinicopathologic profile and follow up was noted. Mean age of the cohort was 50.1 ± 7.8 years (M:F = 1: 2.3). Papillary carcinoma was the commonest pathology (60%) followed by follicular carcinoma (20%), 10% each had Hurthle cell and poorly differentiated thyroid carcinoma. Sixty percent patients presented with recurrent or persistent disease and 20% with distant metastases. Vocal cord palsy was observed in 30%. Fifty percent patients underwent tracheal resection with end to end anastomosis and remaining laryngotracheal resection. Carotid artery resection and anastomosis was performed in one patient. Shin stage IV invasion was observed in 80% and stage III in remaining. There was no perioperative mortality. All patients received adjuvant radioiodine therapy and 40% external beam radiotherapy. Median follow up was 48 months. One patient who earlier had tracheal resection underwent total laryngectomy due to recurrent disease after 24 months. Forty percent patients developed distant metastases during follow-up. Mean survival was 77.8 months (CI = 63.0–92.5). Five-year overall survival was 60% and was significantly high in those without distant metastases (p = 0.006). The outcome of circumferential larygotracheal resections for DTC is excellent in terms of local disease control and long term survival.

Introduction

Differentiated thyroid carcinoma (DTC) patients have excellent outcome with 10 year overall survival (OS) rate close to 90% following surgical resection [1, 2]. Extrathyroidal invasion is an independent prognostic factor and patients with extrathyroidal invasion have significantly worse OS than those without invasion [1–5]. The incidence of extrathyroidal invasion varies between 10 and 34% [1–3]. Among those having extrathyroidal invasion the incidence of strap muscle invasion is the highest and great cervical vessel invasion the least. The incidence of tracheal and laryngeal involvement ranges between 37–55% and 4–12% respectively [5]. Airway obstruction is the cause of death in 33–47% of DTC patients, emphasizing the importance of local control of disease [5–7].

Due to the unique anatomy and vital functions the invasion of larygotracheal tree poses complex management challenge [5–7]. The extent of surgery has been debated and recommendations vary from tangential shave resection to circumferential sleeve resection [1, 5–9]. The intent of surgery in such patients could be curative, palliative, improvement in quality of life, prevention of recurrence and last but not least prolongation of survival. Complete resection of the local disease if feasible, could achieve all the aforementioned goals, but these operations are challenging and call for a fine balanced judgement between patient safety and expected outcome [5, 7–9]. Some studies have demonstrated excellent outcome following largyngotracheal resection in experienced hand, but the experience is still limited [4–10]. The incidence of locally invasive DTC is high in developing nations like ours [11]. In this study, we share our experience of laryngotracheal resection in DTC and its long-term outcome.

Material and Methods

This is retrospective analysis of prospectively maintained data. The study was approved by department research committee. All patients with locally advanced DTC with airway invasion who underwent circumferential resection between January 2000 and December 2015 were included in the study. Patients who underwent shave resection, window resection and those who did not consent for surgery were excluded from the study. The patients’ clinical details, imaging findings, surgery records, adjuvant therapy and follow up details were retrieved from patient files and hospital information system. Ten patients met the inclusion criteria.

Surgical Protocol

All DTC patients with symptoms and signs and/ suspicion of local invasion and those presenting with recurrent or residual disease undergo contrast enhanced tomography (CECT) of neck and thorax to assess extent of invasion (Fig. 1). Direct laryngoscopy is performed in all to assess recurrent laryngeal nerve (RLN) function and/or mucosal involvement. Larygotracheoscopy and/or oesopgagoscopy is performed if indicated. All procedures are performed under general anesthesia preferably with endotracheal intubation (ETT) using video laryngoscopy. Preoperative tracheostomy is performed if ETT attempt fails. Nasogastric tube is inserted for intra operative identification of esophagus. Collar incision or apron incision is used as per need. Thyroidectomy and neck dissection if indicated are performed first. For those requiring complete laryngectomy, larynx is dissected out from its surrounding after division of muscular attachment. Trachea is divided below the lower limit of tumor and freed from all the sides and permanent fish mouth tracheostomy is made. Pharyngeal opening is closed in two layers.

Fig. 1.

Coronal reconstruction of CECT image of a papillary thyroid carcinoma patient having Shin Stage IV invasion into trachea, showing tumor invasion through left lateral wall of trachea resulting in almost complete obliteration of tracheal lumen

For tracheal resection followed by end to end anastomosis, the trachea is carefully dissected from all sides keeping the dissection minimal so as not to interfere with the vascular supply, preserving the integrity of RLN nerve and parathyroids. Suprahyoid release and/or mediastinal mobilization is performed if indicated to facilitate tension free anastomosis. Once surgeon is ready for resection, the anesthetist is asked to withdraw the ETT under direct vision, and leave it in situ beyond point of proximal resection margin. After incising the trachea at the inferior resection margin a long tracheostomy tube is inserted and ventilation is maintained via it. Two stay sutures are put on the lateral sites of inferior end and tracheal resection is performed in oblique fashion. Continuity is restored by first approximating the membranous trachea followed by reinsertion of ETT beyond distal cut margin after withdrawing tracheostomy tube and then cartilaginous portion of trachea is approximated. Single layer closure with interrupted sutures with 3–0 vicryl is performed and knots are placed outside the tracheal lumen. Suction drains are placed and once incision is closed the flexion of neck is ensured by placing one or two loupe Prolene sutures between the submental and upper thoracic region. These sutures are removed 5–7 days after surgery. Frozen section is not routinely used to assess the resection margins. Shin classification was used to report the depth of tracheal invasion [3].

Adjuvant therapy and Follow-up Protocol

All these patients are referred for radioiodine (RAI) scan and therapy (RAIT) if indicated. They also received TSH suppressive dose of levothyroxine with intent to keep serum TSH levels < 0.1 mU/L. Patients with R1 and R2 resection are referred for adjuvant external beam radiotherapy (EBRT) and final decision regarding such therapy is taken on case to case basis. Wherever indicated, EBRT is usually administered after first dose of RAIT. Our protocol is to follow these patients at 6 monthly intervals with serum thyroglobulin (Tg) estimation, high-resolution neck ultrasonography (USG) and/or RAI scan. RAIT is repeated if tumor is still RAI avid. FDG-PET is performed when Tg is high and RAI scan fail to localize the disease. Other imaging and endoscopy examinations are performed as indicated. Targeted therapy (sorafenib) is reserved for patients having progressive, symptomatic disease not amenable to RAIT.

Statistical Analysis

The statistical analysis was performed using SPSS software (version 22; SPSS, Inc., Chicago, IL, USA). Categorical variable were expressed as frequencies and continuous variables as mean with standard deviation (SD). The Kaplan–Meier method was used to estimate OS. Two tailed p-value of < 0.05 was considered significant for all the tests.

Results

Ten patients were operated during the study period, their mean age was 50.1 ± 7.8 years (32–58 years), 40% of patients were 55 years or more in age and women outnumbered men. Sixty percent of patients presented with recurrent or residual disease whereas remaining had primary presentation. Seventy percent had symptoms suggestive of local invasion including dyspnea-60%, hoarseness-60%, hemoptysis-20% and dysphagia-20%. Thyromegaly was noted in 60% patients. Cervical lymph node involvement and distant metastases were observed in 10% and 20% patients respectively. All patients had CECT and laryngotracheoscopy examination prior to surgery. More than 50% occlusion of lumen was seen in 60% patients and 30% showed unilateral vocal cord palsy. One patient with right carotid artery invasion had angiography to test patency of circle of Willis.

All patients underwent total thyroidectomy including 60% completion total thyroidectomy. Lymph node dissection was performed in 70% patients, 60% had CCLND performed whereas 10% had bilateral SLND performed in addition to CCLND and 10% had only SLND. Circumferential tracheal resection with end to end anastomosis was done in 50% patients. Partial laryngeal drop and mediastinal mobilization was required in 3 and 1 patients respectively. On two occasions ETT attempt failed and tracheostomy had to be performed prior to surgery and one patient had emergency tracheostomy performed prior to the planning of definitive surgery. The median number of tracheal ring excised was 4 (3 rings-1, 4–6 rings-3, and > 6 rings-1 patient). Total laryngectomy with permanent tracheostomy was performed in remaining 50% patients including 30% of those who in addition had up to four tracheal rings excised (Fig. 2). Apart from airway invasion, other structures involved were—strap muscle = 60%, RLN = 40%, pharynx = 20%, oesophagus = 10%, internal jugular vein = 40% and common carotid artery = 10%. One patient each of PTC and FTC underwent thrombectomy of great cervical veins and one had part of carotid artery excised followed by end to end anastomosis. Sixty percent of patients were extubated immediately after surgery 10% within 24 h, 10% within 48 h and 20% after 48 h.

Fig. 2.

Gross specimen of the same patients wherein en block circumferential tracheal resection along with total thyroidectomy and central compartment lymph node dissection was performed

There was no perioperative mortality. Postoperative morbidity included hemorrhage and chyle leak in one patient each and minor pharyngocutaneous fistula in two. Patient with post-operative hemorrage needed re-exploration to secure a bleeder from strap muscle, rest all the patients were managed conservatively. Papillary thyroid carcinoma was the commonest histology type (60%), followed by follicular thyroid carcinoma in 20%. Ten percent each had Hurthle cell carcinoma and poorly differentiated thyroid carcinoma. Resected margins were free of tumor in 70% and microscopic involvement of superior margin on epithelial side was observed in 30%. Shin Stage IV invasion was observed in 80% and Shin III in remaining. The details of patients are summarized in Table 1.

Table 1.

Summary of clinicopathologic profile and outcome

Sl no	Age/ Gender	Mode of presentation	Histologya	Structures involvedb	Distant metastases	Procedurec	Adjuvant therapyd EBRT/RAIT	Follow-up (in months)	Status
1	45/F	Recurrent	PTC	T, L, SM, RLN	Yes/Synchronous Pul + Skeletal	CT + CCLND + TL	RAIT	12	Dead
2	57/M	Primary	PTC	T, SM, RLN	–	TT + CCLND + TR + EEA* Tracheostomy	RAIT	36	Dead
3	50/F	Recurrent	HCC	T, L, P	Yes/Metachronous Skeletal	CT + TL	RAIT + EBRT Sorafenib + ZA	48	Dead
4	52/M	Recurrent	PTC	T, L, E,	Yes/ Metachronous Pulmonary	TL	RAIT	90	Alive with asymptomatic pulmonary metastases
5	56/F	Recurrent	FTC	T	–	CT + CCLND + TR + EEA TL after 24 months	RAIT + EBRT after second surgery	88	Alive with biochemical disease
6	55/F	Primary	PTC	T, SM, RLN	–	TT + CCLND + B/L LND + TR + EEA	RAIT	77	Alive without disease
7	58/M	Recurrent	PTC	L, SM, IJV CCA	Yes, Synchromous Pulmonary	TL + CCA resection + EEA	RAIT/ Pallliative EBRT Prior to surgery	36	Dead
8	45/F	Primary	PTC	T, SM, RLN, IJV, CCA	-	TT + CCLND + TR + EEA + Thrombectomy of Rt Brachiocephalic vein	RAIT	76	Alive with biochemical disease
9	51/F	Primary	FTC	T, SM, IJV, CCA	Yes/Metachronous Pulmonary	TT + CCLND + TR + EEA + Thrombectomy of Rt Brachiocephalic vein	RAIT	48	Alive with asymptomatic pulmonary metastases
10	32/F	Recurrent	PDTC	T, L, P, IJV	Yes/Metachronous Skeletal	TT + CCLND + Jugular vein resection + TL	RAIT + EBRT Sorafenib + ZA	48	Alive with skeletal metastases

^aHistology: PTC—papillary thyroid carcinoma, FTC—follicular thyroid carcinoma, PDTC—poorly differentiated thyroid, MTC—medullary thyroid carcinoma

^bSructures: T—trachea, L—larynx, SM—strap muscle RLN—recurrent laryngeal nerve, P—pharynx, E—esophagus, IJV—internal jugular vein, CCA—common carotid artery

^cSurgical procedure: TT—total thyroidectomy, CT—completion thyroidectomy, CCLND—central compartment lymph node dissection, LND—lateral neck dissection

TR—tracheal resection, TL—total laryngectomy, EEA—end to end anastomosis

^dAdjuvant therapy: EBRT—external beam ratiotherapy, RAIT—radioiodine therapy

Whole body RAI scanning was performed in all patients (no uptake- 2, thyroid bed and regional uptake- 5, distant metastases- 2). All patients received at least one dose of RAIT, with a mean dose of 107.5 ± 29 mCi. EBRT was administered to 40% (locoregional- 3, palliative- 1). Out of three patients with positive resection margin 2 received EBRT and one didn’t comply with advice. One patient who initially didn’t consent for laryngectomy received EBRT in prior to surgery and later underwent surgery due to continuing hemoptysis. Another patient received palliative therapy for painful skeletal metastases. All patients received suppressive dose of levothyroxine and tolerated it well.

Median follow up was 48 (12–90) months. Sixty percent patients developed disease progression, 20% suffered from both loco-regional recurrence and distant metastases, whereas 20% each developed only distant metastases and local recurrence. Median duration of local disease progression was 18 months and half of these patients had successful excision of local recurrence. Forty percent patients died during follow-up, all those who had initial presentation with distant metastases and 50% of those who developed distant metastases in follow up. Distant metastatic disease was the cause of death in the two-third of patient and in one-third non-resectable local recurrence contributed to death. Fifty percent patients were alive with persistent disease, 20% with biochemical disease (raised Tg levels) and 30% with structural disease. Ten percent were alive free of disease until last follow-up visit. One patient who earlier had tracheal resection and refused to receive EBRT underwent total laryngectomy due to recurrent disease 24 months after the first surgery. She was alive and symptom free with only biochemical persistence of disease 50 months after second surgery. Mean survival was 77.8 months (CI = 63.0–92.5). Five-year overall survival was 60% and was significantly high (p = 0.006) in those without distant metastases (Fig. 3).

Fig. 3.

Kaplan–Meier curve depicting overall survival of the cohort

Discussion

The current study shows that in a select group of patients circumferential tracheal and laryngeal resection can be safely performed and results in excellent symptom palliation, long-term local disease control and improved survival.

Majority of the patients in the current study were symptomatic which is consistent with the finding of other studies which show that a larger proportion of patients undergoing laryngotracheal resection are likely to be symptomatic [7, 8, 10]. CT scan was the most commonly used imaging modality in our study. Needless to state that accurate pre-operative imaging is essential for optimal surgical planning of these cases. But, there seems to be no perfect imaging modality which could accurately assess the depth and extent of invasion. Both CT and MRI have their own limitations and while ultrasonography is considered good by some centers, it remained highly operator dependent [12, 13]. Therefore various centers seem to rely on their local expertise and standard of care. But majority of surgeons agree on the value of preoperative endoscopy wherein apart from vocal cord function assessment, subtle features of mucosal invasion like mucosal erythema and oedema could be noticed [14]. It is a well-established fact that submucosal extent of invasive is likely to be greater than the exterior spread and exact delineation of which helps in optimal planning of surgery [3, 15, 16]. Hence, a preoperative direct laryngoscopic examination should be performed in all the patients planned for laryngotracheal resection.

These patients often present with distressing symptoms and hence one of the most pressing aims of surgery remains symptomatic palliation and improvement in quality of life. Resections involving laryngotracheal tree are complex procedure associated with high morbidity. Hence, decision to go ahead with these operations need due deliberations. Benefits should outweigh the morbidity and preservation of vital functions involving swallowing, phonation and respiration should be prime consideration [1, 9]. Due to inherent complexities of situation various surgical options have been offered ranging from less intense tangential shave resections to slightly extensive window resections and the most extensive circumferential resection [8, 14, 17–19]. Although technically less demanding the utility of tangential shave resection is limited to Shin stage I invasive DTC where a local control rates of 95% could be achieved. But for tumors having deeper invasion results of shaving procedures are significantly worse than resections in terms of local control and OS [8, 19]. Furthermore, salvage procedures for recurrence after shave resection are associated with higher morbidity and lower OS than primary resections [2, 10, 20]. The concerns of depth of invasion can be addressed by window resection [21]. But these procedures could only be offered to the patients who have less than one-fourth of tracheal circumference and less than one-third of cricoids cartilage invasion by tumor [14, 18, 22].

Circumferential resections are preferred option in patients having large circumferential involvement and also those with tumor invasion stage of Shin II and beyond [8, 9]. The goal is to achieve complete resection of the local disease in order to achieve improved OS and best palliation [18]. Presence of one functional RLN is must for sleeve resections and anastomosis. Though earlier studies reported significant mortality (5–9%) and morbidity associated with these procedure, the outcomes have improved with better understanding of operative technique and supportive care [2, 14, 20, 22]. The current evidence points towards long-term effectiveness of these operations including contribution to quality of life [22, 23]. But some of the early and late postoperative complications including dysphagia, anastomotic dehiscence, tracheal or stomal stenosis could still pose management challenge [7, 22]. The key to prevent all these is attention to operative technique ensuring tension free repair with preservation of vascular supply. More than 50% of patients undergoing complete suprahyoid release suffer from prolonged dysphagia [24]. Extent of tumor involvement determines the need for suprahyoid release and mediastinal mobilization. Median number of tracheal rings excised in the current study was four. We performed partial suprahyoid release in three patients and none developed post operative dysphagia.

Patients with airway invasion are at high risk of recurrence and death due to malignancy [3–6]. One of the major achievements of laryngotracheal resection in DTC patients is that those succumbing to the disease in follow-up are spared of distressing local symptoms [7].Five and 10 year- OS of more than 70% and 60% respectively have been reported after successful resection [5, 6, 20, 23]. The OS in the current study is at par with other study and majority of those dying didn’t die of local disease. The role of adjuvant EBRT in invasive DTC is still evolving but some studies report that it contributes to DFS in patients with residual disease following surgery [25]. The number of patients receiving EBRT was low in the current study therefore we can’t draw any meaningful conclusion on the role of EBRT in these patients.

There are few limitations of this study. There was no head on comparison with shave resection and window resection. The number of patients enrolled in this study is small. But that’s how the things stand in real world as majority of patients with invasive DTC don’t consent for laryngotracheal resections fearing loss of voice or other complication, similarly many don’t consent for EBRT fearing associated morbidity and affordability of intensity modulated radiotherapy is limited in our country.

In conclusion, laryngotracheal resection operations need precise planning involving a multidisciplinary team. Complete excision of the tumor with negative margins should be the goal in order to achieve long term local control of the disease. With refinement in technique, circumferential resections, though technically demanding procedure can be safely performed; and provide excellent outcome in terms of local disease control and OS.

Author Contributions

AM, RAB: study conception and design; VS, RAB, AM, PKP, SKM: acquisition of data; VS, RAB, AM, PKP, SKM: analysis and interpretation of data; VS, RAB: drafting of manuscript; VS, RAB, AM, PKP, SKM: critical revision of manuscript.

Funding

No funding was received from any source.

Compliance with Ethical Standards

Conflict of interest

The authors declare that they have no conflict of interest.

Consent to Participate

This is a retrospective review of prospectively kept data.

Consent for Publication

Approved by all concerned authorities.

Data Availability and Material

Data is Institute property and can be provided on specific request.

Ethical Approval

The study was approved by the department research committee.