A Clinical Study of Transoral Pharyngectomies to Treat Superficial Hypopharyngeal Cancers

Abstract

Background

For years, it has been a major interest for surgeons and oncologists to develop a novel technique to detect hypopharyngeal cancers at an early stage and to treat the lesions in a less invasive manner. The advent of the narrow band imaging system combined endoscopy and various endoscopic approaches shed light on the new era of the minimum invasive management of superficial cancers in hypopharyngeal regions.

Methods

Three endoscopic approaches, endoscopic mucosal resection, endoscopic submucosal dissection and endoscopic laryngopharyngeal surgery, were chronologically introduced at our institute. In this study, we focused on the clinical outcomes, advantages and limitations of each procedure.

Results

A total of 30 patients (42 procedures) received transoral pharyngectomies between June 2006 and May 2012. Tracheotomy was performed in 9 of 42 (21%) patients. Three patients developed local recurrence and were subsequently controlled by additional transoral pharyngectomies. The 2- and 5-year local control rates were 92.8 and 83.5%. The 2- and 5-year overall survival rates were 82.1%.

Conclusions

Endoscopic mucosal resection, endoscopic submucosal dissection and endoscopic laryngopharyngeal surgery-transoral pharyngectomies are useful procedures for treating superficial hypopharyngeal cancers. Endoscopic mucosal resection manifested the least invasiveness and may be beneficial for resecting small superficial lesions, endoscopic submucosal dissection may be advantageous for patients with difficult laryngopharyngeal exposure and endoscopic laryngopharyngeal surgery has shown the optimal effectiveness and minimal complications and can be applied to most of the hypopharyngeal sub-sites. All three procedures require a high level of technical skill and close collaboration between otolaryngologists and gastroenterologists.

endoscopic mucosal resection, endoscopic submucosal dissection, endoscopic laryngopharyngeal surgery, narrow band imaging

INTRODUCTION

Hypopharyngeal cancer is a head and neck cancer with poor prognosis. Two clinical features, (i) a lack of noticeable symptoms until advanced stages and (ii) tendency to metastasize through lymphatic drainage, lead to this result. For years, it has been a major interest for surgeons and oncologists to develop a novel technique to detect hypopharyngeal cancers at an early stage and to treat the lesions in a less invasive manner.

The narrow band imaging (NBI) system is an innovative optical technique that enhances the diagnostic capability of gastrointestinal endoscopy by illuminating the intraepithelial papillary capillary loop using narrow and wide filters in a red-green-blue sequential illumination technique (Olympus Optical Co., Tokyo, Japan) (1). The advent of the NBI system has shed light on the clear detection of superficial cancers in the upper gastrointestinal tract. The NBI system showed particular usefulness in the pharyngeal region where iodine staining is not applicable (2). Through NBI combined gastrointestinal endoscopy, superficial hypopharyngeal cancers have been increasingly detected in the esophageal cancer patient group (3). Metachronous superficial pharyngeal cancers also have become easily diagnosed in the patient group of head and neck cancers treated with chemoradiotherapy (4).

To manage the increased number of patients with superficial hypopharyngeal cancers, various endoscopic approaches have been developed. Some techniques have diverted applications from the gastrointestinal field (5,6) and some have been recently developed to treat pharyngeal lesions (7–10). Among these transoral pharyngectomies, we have chronologically introduced endoscopic mucosal resection (EMR), endoscopic submucosal dissection (ESD) and endoscopic laryngopharyngeal surgery (ELPS). In this study, we focused on the clinical outcomes, advantages and limitations of each procedure.

PATIENTS AND METHODS

We conducted a thorough clinical analysis of patients who underwent transoral pharyngectomies for superficial hypopharyngeal cancers between June 2006 and May 2012. Three endoscopic approaches, EMR, ESD and ELPS, were chronologically introduced (Fig. 1). All approaches were performed under general anesthesia through a suspension of Satou's curved laryngoscope (Nagashima Medical Co., Tokyo, Japan). By watching the images on a high-definition monitor, evaluation and resection were done with close collaboration between otolaryngologists and gastroenterologists (Fig. 2).

Figure 1.

Three types of transoral pharyngectomies used at our institute (EMR, endoscopic mucosal resection; ESD, endoscopic submucosal dissection; ELPS, endoscopic laryngopharyngeal surgery).

Figure 2.

Satou's curved laryngoscope and curved needle knife used for ELPS. In ELPS, resection was carried out manually by otolaryngologists with the endoscopic assistance of gastroenterologists.

In each procedure, gastroenterologists performed a thorough evaluation of the extent of the superficial lesions using an optical magnifying endoscope (Olympus GIF-H260Z Magnifying Scope) combined with the NBI system (Olympus CV-260SL processor and light source). A 1.5% Lugol solution was used for iodine staining to reconfirm the extent of superficial lesions.

In EMR, after injecting indigo carmine-epinephrine saline into the submucosal layer for lifting, the lesion was suctioned and snared through a cap-assisted EMR. In this procedure, gastroenterologists performed both evaluation and resection under the assistance of otolaryngologists. In ESD, after submucosal injection and lifting, the lesion was marked by a flex needle knife, followed by en bloc resection with an undermining dissecting maneuver using an IT knife (flex needle knife with a ceramic ball at the top to avoid perforation). The flex needle and IT knives (Olympus Optical Co.) were manipulated in turn through the fiberscope channel by gastroenterologists under the assistance of otolaryngologists. In ELPS, after submucosal injection and lifting, the lesion was resected manually using a curved needle knife (Fig. 2, Olympus Optical Co.) and curved or Steiner's forceps; resection was carried out by otolaryngologists with the endoscopic assistance of gastroenterologists. Argon plasma coagulation (APC, ERBE Co. Marietta, USA), a monopolar indirect-contact coagulator using ionized argon gas, was used to vaporize the surgical wound with close margins and for hemostasis as well.

In this study, clinical and pathological data, surgical complications, local control and overall survival rates were analyzed. With respect to the Kaplan–Meier estimation of local control and overall survival rates, local recurrence from the same hypopharyngeal sub-site and death from any cause were treated as an event. Any metachronous lesion in the same patient from a different hypopharyngeal sub-site was treated as a new lesion. For the Kaplan–Meier estimation, we truncated the data at a follow-up duration of 2 and 5 years.

H–E stained pathological specimens were evaluated based on the guidelines for clinical and pathological studies on carcinoma of the esophagus, which consist of mucosal pathologic gradings of (i) low-grade intraepithelial neoplasia (LGIN: intramucosal atypical findings are limited to the lower half of the epithelium), (ii) high-grade intraepithelial neoplasia (HGIN: intramucosal atypical findings spread to the upper half of the epithelium) (11,12). HGIN includes so-called carcinoma in situ. Positive margin means the specimen has squamous cell carcinoma or HGIN at the lateral margin. All pathological specimens were reviewed by board-certified pathologists and then double checked by the senior pathologist (the third author).

RESULTS

A total of 30 patients (42 procedures) received transoral pharyngectomies for superficial hypopharyngeal cancers between June 2006 and May 2012. The clinical details of the 30 patients are tabulated in Table 1. The chronological treatment trend of transoral pharyngectomies is tabulated in Table 2.

Table 1.

Clinical details of the 30 patients investigated

Patients
Numbers	30
Procedures	42
Age	55–80 (avg 67.2)
Male/female	29/1
Procedure
EMR	21/42 (50%)
ESD	10/42 (24%)
ELPS	8/42 (19%)
APC	27/42 (64%)
Tracheotomy	9/42 (21%)
Pathology
Margin positive	23/35 (66%)
Submucosal extension	20/35 (57%)
Time (min)
Procedure	15–195 (avg 86)
Anesthesia	40–280 (avg 130)
Duration (day)
Resume oral diet	1–36 (avg 5.6)
Admission	7–45 (avg 15.8)

Table 2.

Treatment trends of transoral pharyngectomy procedures

YEAR	EMR	ESD	ELPS	APC	OP TIME	TRAC
2006	○			○	*********	○
	○				******
	○			○	*********	○
2007	○				*****
	○			○	********	○
	○			○	*****
		○			*******	○
	○				****
2008	○			○	*********	○
		○			***
		○			*****
		○		○	******
	○			○	*****
	○			○	**
2009	○			○	****
	○				**
				○	***
		○			*********	○
	○			○	****
2010		○			******
	○			○	***
		○			*********	○
	○			○	****
	○			○	**
	○			○	******	○
				○	**
	○			○	**
	○			○	****
	○			○	**
2011	○			○	**
			○		***
				○	*
		○	○		*****
		○	○		***
		○	○	○	*******
			○	○	****
2012				○	*
			○	○	****
			○		**
				○	**
			○		********	○
				○	*

OP TIME, procedure time (*=20 min); TRAC, tracheotomy.

There were 29 male and 1 female patient with a mean age of 67.2 years (age range from 55 to 80 years). Of 30 patients, 15 and 2 had treatment histories for esophageal and hypopharyngeal cancers, respectively. Primary lesions involved the hypopharyngeal sub-sites at the pyriform sinus, posterior wall, and post-cricoid in 22, 11 and 3 patients, respectively; 37 lesions were confined to one sub-site and five over two sub-sites. Among the 42 procedures, EMR, ESD and ELPS were employed for 21, 10 and 8 patients, respectively; three patients received ESD + ELPS procedures simultaneously. Among the 30 patients, 24 and 6 received single or more than two procedures, respectively; an additional procedure was required for three cases with local recurrence and three cases with metachronous hypopharyngeal cancers. APC was applied in 27 of 42 (64%) patients. EMR was incorporated at the early stage followed by introduction of ESD and recently ELPS has become the main procedure carried out (Table 2). Among the 21 patients who received EMR, 11, 6 and 4 patients had one, two and three split resections, respectively.

The average times for general anesthesia and procedure per case were 130 and 86 min; the average procedure times for EMR, ESD and ELPS were 88, 125 and 83 min, respectively. A steroid (Betamethasone) was instilled intra and post operatively in 38 of 42 (90%) procedures.

Tracheotomy was performed in nine of 42 (21%) patients; the main cause was laryngopharyngeal edema (six patients), bleeding (one), emphysema (one), and transient bilateral glottic paralyses (one). Laryngopharyngeal edema occurred due to the procedures of extended resection (involved more than two hyopharyngeal sub-sites) (Fig. 3), mechanical irritation, iodine reaction and over injection of saline (50 ml) in two, two, one and one patient, respectively. Tracheotomy was done before extubation and after re-intubation in seven and two patients, respectively. In the ESD procedure, extensive emphysema at the neck and superior mediastinum regions developed through a mucosal fistula created at the lateral wall of the pyriform sinus; emaciated (150 cm, 32 kg) and post-irradiated status affected the occurrence of this complication (Fig. 4). Emphysema and fistula healed 2 months after the procedure.

Figure 3.

Larygopharyngeal edema after the ELPS procedure required tracheotomy. (a) Superficial lesions involved two sub-sites (the left pyriform sinus and post cricoid) of the hypopharynx. (b) Larynx after resection showing extensive edema at the supraglottis.

Figure 4.

Subcutaneous emphysema incurred after ESD of a superficial lesion at the left pyriform sinus. (a) Iodine unstained lesion before ESD (arrows), (b) surgical wound after ESD, (c) computed tomography scan from the day after ESD showing extensive emphysema.

Among 42 procedures, 35 surgical specimens were available for pathological evaluation. Squamous cell carcinoma, HGIN and LGIN were diagnosed in 20, 13, and two cases, respectively. Among 35 surgical specimens, 23 (66%) were diagnosed as having positive lateral margins. Positive margins were identified in 15 of 21 (71%), 5 of 9 (44%) and 3 of 5 (60%) specimens operated by EMR, ESD and ELPS procedures, respectively.

Patients resumed an oral diet at an average of 5.6 days (range from 1 to 36 days). The average admission days were 15.8 (range from 7 to 45 days). Among all patients, no one demonstrated deterioration of swallowing function after the procedures. Mild-to-moderate pharyngeal strictures were observed in the cases with repeated pharyngectomies, but no obvious swallowing problem was observed in these patients.

Three patients developed local recurrence (one after EMR and two after ESD procedures) and were subsequently controlled by additional transoral pharyngectomies. None of the patients lost their larynges as a result of the uncontrolled pharyngectomy lesions. The 2- and 5-year local control rates were 92.8 and 83.5% and the median follow-up duration was 24 months with a range between 5 and 72 months (Fig. 5).

Figure 5.

The 2- and 5-year local control and overall survival rates of the 30 patients analyzed.

Three patients died of distant metastases of esophageal cancer and one died of neck disease from metachronous hypopharyngeal cancer unrelated to the pharyngectomy lesion. No patient died of local recurrence related to the transoral pharyngectomies. The 2- and 5-year overall survival rates were 82.1% and the median follow-up duration for surviving patients was 28 months with a range between 6 and 80 months (Fig. 5).

DISCUSSION

The NBI system is a novel illumination technique developed for endoscopy in which the spectral bandwidth of the filtered light is narrowed. Capillaries on the mucosal surface can be seen most clearly at 415 nm, the wave-length corresponding to the hemoglobin absorption band (1). An irregular microvascular proliferation is known to appear as an early event in carcinogenesis. Under NBI-combined magnifying endoscopy, microvascular proliferation can easily be identified, and thus facilitate the detection of superficial cancers (2).

Superficial hypopharyngeal cancers have been increasingly identified during follow-up NBI endoscopic screenings in the esophageal cancer patient group. Among 30 patients who received transoral pharyngectomies, 15 had been treated for and 5 were later diagnosed with esophageal cancer. Overall, two of three pharyngectomized patients were associated with synchronous or metachronous esophageal cancer, presumably related to the field cancerization phenomenon. Careful attention must be paid to esophageal cancer patients who are current smokers and drinkers and who are showing multiple Lugol-voiding lesions in the esophagus (3).

To manage the increased number of superficial cancers in the upper gastrointestinal tract and to meet the need for less invasive treatment, various endoscopic approaches have been developed. It was considered impossible to perform extensive mucosal resection using endoscopy until Inoue and Endo first reported an endoscopic esophageal mucosal resection (EMR) using a transparent tube in 1990 (13). After reviewing 380 cases of EMR in the gastrointestinal tract, Inoue et al. concluded that this approach is the ideal modality for managing superficial cancers (5).

We introduced EMR to treat superficial hypopharyngeal cancers in 2006. Initially, the entire EMR procedure took >120 min, and the patients frequently underwent tracheotomy because of laryngopharyngeal edema. However, as we gained experience, the procedure time was reduced and tracheotomy seldom performed. Although EMR is a fast and a less invasive approach, we have noticed some shortcomings, such as inabilities and limitations to control the resection margin, to perform en bloc resection for large lesions, and to suck and draw mucosal lesions into the cap hood because of the laryngeal cartilage beneath the superficial lesions.

To reinforce the shortcomings of EMR, ESD has been developed in which submucosal dissection is carried out using a flex needle and IT knives (14). In a report comparing ESD and EMR for early mesopharyngeal and hypopharyngeal cancers, en bloc and negative margin resection rates were superior in ESD, but the procedure time was shorter in EMR; there were two postoperative subcutaneous emphysemas in ESD versus none in EMR (15). In our study, we observed similar results as ESD showed a lower rate for positive resection margins and a longer procedure time than EMR. Extensive emphysema at the neck and superior mediastinum regions was observed in one ESD patient and the emaciated status of this particular patient in part affected the occurrence of this complication.

Among 10 ESD patients, three required tracheotomy; in one patient, over injection of saline (50 ml) was the main cause of laryngopharyngeal edema. In the esophageal and colon regions, 20–100 ml saline injection during the EMR or the ESD procedure is recommended for lifting to avoid perforation (5). In the laryngopharyngeal region, based on our experience, saline injection for mucosal lifting should not exceed 20 ml to avoid airway compromise.

Different from EMR and ESD, ELPS is a unique concept in which resection was achieved manually by otolaryngologists using a curved needle knife and forceps under endoscopic assistance of gastroenterologists. In Satou's experience of 113 patients, only two developed local recurrence and the overall survival was 87.2% (7). In our experience with eight ELPS procedures since 2011, no patient developed local recurrence and only one patient received tracheotomy. We are very satisfied with the clinical achievement and limited complications of this procedure. Further experience and instrumental refinements are needed to attain optimal effects.

In a pathological review of surgical specimens, a relatively high incidence of positive margins throughout all procedures was observed. Positive margins, however, did not significantly affect local control (median follow-up 24 months) or overall survival (median follow-up 28 months). Several factors, such as, (i) the relatively lower grade lesion at the margin (for example, squamous cell carcinoma at the center of the lesion and LGIN at the margin), (ii) the possibility of a processing artifact and (iii) application of APC vaporization to the surgical wound with a close margin, may have an overall positive effect on the outcome. Nonetheless, when encountering positive pathological margins in transoral pharyngectomies, the details of the findings, its clinical implications and the need for adjuvant treatments need to be explained to, and discussed with, the patients.

EMR, ESD and ELPS transoral pharyngectomies are useful procedures for treating superficial hypopharyngeal cancers. EMR manifested the least invasiveness and may be beneficial for resecting small superficial lesions, ESD may be advantageous for patients with difficult laryngopharyngeal exposure, and ELPS have shown the optimal effectiveness and minimal complications and can be applied to most of the hypopharyngeal sub-sites. All three procedures require a high level of technical skill and close collaboration between otolaryngologists and gastroenterologists.

Funding

This study was supported by a Grant-in-Aid for Scientific Research (C) from the Ministry of Education, Culture, Sports, Science, and Technology of Japan (#.24592608: 2012–15).

Conflict of interest statement

None declared.