Endoscopic mucosal resection and endoscopic submucosal dissection for en bloc resection of superficial pharyngeal carcinomas

• Introduction
• Patients and methods
• Endoscopic resection procedures
• EMR techniques
• ESD techniques
• Histopathological evaluation and assessment of the therapeutic efficacy
• Follow-up examinations
• Statistical analysis
• Results
• Endoscopic resection results
• Postoperative clinical course
• Complications and short-term follow-up data
• Risk factors for severe laryngeal edema
• Long term outcomes
• Discussion
• References

Background and study aim: Endoscopic mucosal resection (EMR) and endoscopic submucosal dissection (ESD) are being used increasingly to treat superficial oropharyngeal and hypopharyngeal carcinomas. The aim of this study was to clarify whether ESD provided better results than EMR for en bloc and complete resection of superficial pharyngeal carcinomas.

Patients and methods: A total of 76 superficial pharyngeal carcinomas in 59 consecutively treated patients were included. Patients underwent either conventional EMR (using a transparent cap or strip biopsy) (n = 45 lesions) or ESD (n = 31 lesions) between October 2006 and January 2011. The rates of en bloc resection, complete resection (defined as en bloc resection with tumor-free margins), major complications, and local recurrence were evaluated retrospectively as the therapeutic outcomes.

Results: ESD yielded significantly higher rates of both en bloc and complete resection compared with EMR (en bloc 77.4 % [24/31] vs. 37.8 % [17/45], P = 0.0002; complete 54.8 % [17/31] vs. 28.9 % [13/45], P = 0.0379). ESD was more frequently complicated by severe laryngeal edema (4/21 [19.0 %] vs. 1/31 [3.2 %], P = 0.1446) and was also more time-consuming (124.9 ± 65.1 minutes vs. 57.2 ± 69.6 minutes; P = 0.0014). Local recurrence was observed more often after EMR than after ESD (3/45 [6.7 %] vs. 0/31 [0 %]), although this difference did not reach statistical significance (P = 0.2658).

Conclusions: ESD appears to be a superior method of endoscopic resection of superficial pharyngeal carcinomas for achieving both en bloc and complete resection, although these benefits were also associated with a higher incidence of complications and a significantly longer procedure time. Large prospective studies are needed to compare ESD with conventional EMR for superficial pharyngeal carcinomas.

Introduction

It is often difficult for gastrointestinal endoscopists to detect superficial pharyngeal cancers by conventional white light endoscopy because these cancers are associated with few morphological changes [1] [2] and a high level of skill is required for their detection. Recent progress in and development of new endoscopic diagnostic techniques have allowed imaging of the pharynx by magnification endoscopy with narrow-band imaging (ME-NBI) and eliminated the need for conventional iodine staining in the diagnosis of dysplasia and squamous cell carcinoma. ME-NBI has resulted in an increase in the rate of early detection of pharyngeal dysplasia and superficial pharyngeal carcinoma [3] [4] [5]. Previously, because the majority of pharyngeal carcinomas were detected at an advanced stage, treatment by surgical resection was necessary, which causes loss of swallowing and/or speech functions and also results in cosmetic deformities [6] [7] [8]. Chemoradiotherapy is also widely performed for such lesions, but is associated with a high rate of complications, such as oral pain, disorder of the sense of taste, and swallowing dysfunction [9]. Moreover, radiotherapy cannot be repeated for recurrent or metachronous pharyngeal carcinomas after the initial treatment.

Endoscopic resection is considered to be minimally invasive surgery and has recently been applied to the treatment of superficial oropharyngeal and hypopharyngeal carcinomas as an experimental procedure that does not impair the patients’ quality of life or physical functioning [2] [10] [11] [12] [13]. However, the indication of endoscopic resection for the treatment of superficial pharyngeal carcinoma still remains controversial due to the lack of data on the definitive factors predictive of lymph node metastasis and long term outcome, including the oncological safety results from large scale prospective studies.

According to the results from a clinical series of esophageal lesions, each endoscopic mucosal resection (EMR) session allows resection of lesions of only limited size, and large lesions have to be resected in several fragments, resulting in higher rates of incomplete resection and cancer recurrence due to the presence of residual neoplastic tissue [14] [15]. Precise reconstruction of the tumor specimens was found to be difficult after resection in a piecemeal fashion because of alterations in the shapes of the resected tissues, especially in the case of small resected fragments, inability to identify the coagulation points, and/or presence of residual tissue bridges between the resection sites [5] [16] [17]. In contrast to EMR, the technique of endoscopic submucosal dissection (ESD) allows en bloc resection of even large lesions [15] [18] [19]. However, ESD is technically difficult and time-consuming [20]. Until now, the efficacy of ESD for superficial pharyngeal carcinoma has not been compared with that of EMR. In the study reported here, consecutive cases of superficial pharyngeal carcinoma were treated and the comparative efficacy and safety of ESD vs. EMR performed using a transparent cap or strip biopsy [21] were evaluated. The rates of en bloc resection, complete resection (defined as en bloc resection with tumor-free margins), major complications, and local/distant recurrence were compared in an effort to identify the more superior method for the treatment of superficial pharyngeal carcinomas.

Patients and methods

This was a retrospective study and included 76 superficial pharyngeal carcinomas in 59 consecutive patients who were treated at the Cancer Institute Hospital of the Japanese Foundation for Cancer Research in Tokyo, Japan, between October 2006 and January 2011. According to the definition of the Japan Society for Head and Neck Cancer [22], a superficial pharyngeal lesion is one in which the depth of invasion is comparatively shallow and the visualized changes do not indicate advanced cancer. However, as the pharynx has no muscularis mucosa, this somewhat vague definition only suggests that the depth of invasion is limited to the epithelium or just beneath the epithelium, without extending to the underlying muscle layer.

As part of the preoperative procedures, ME-NBI was performed in all patients to define the carcinomatous area. All patients also underwent ultrasonography of the neck and contrast-enhanced computed tomography (CT) of the neck and chest for preoperative detection of distant and/or lymph node metastasis.

The inclusion criteria were: (1) histologically proven high grade dysplasia or squamous cell carcinoma lesion; (2) endoscopic diagnosis of superficial pharyngeal carcinoma or dysplasia; (3) no apparent lymph node or other organ metastasis on ultrasound and/or CT. Patients with large bilateral hypopharyngeal lesions in the postcricoid region and/or aryepiglottis were excluded to avoid potential stenosis of the pharynx as a result of the ulcer scars. All of the data on patients, lesions, and endoscopic resection procedures was stored consecutively and prospectively in a database.

This study was conducted in accordance with the Declaration of Helsinki, and the study protocol was approved by the Ethics Committee of the Cancer Institute Hospital of the Japanese Foundation for Cancer Research.

Endoscopic resection procedures

All patients were informed about the risks and benefits of several treatment options, including endoscopic resection, radiotherapy, and surgery, and all provided written informed consent for the endoscopic treatment procedure. An otolaryngologist examined the patient’s mouth and pharynx before and after the treatment. The endoscopic resection was performed under general anesthesia after intubation of the patient in the supine position following administration of a muscle relaxant to ensure that the procedure was completed safely. The otolaryngologist then elevated the larynx using a laryngoscope. The laryngoscope was inserted until the tip was close to, but not in contact with, the glottis. The intubation tube was balanced on the back of the laryngoscope and fixed onto the laryngeal side. At this point, the endoscope was inserted to check the working space, location of the lesion, and maneuverability of the endoscope; the laryngoscope was then fixed.

Next, the extent of the lesion was confirmed by iodine staining (1.5 % iodine solution). Marking dots were placed to delineate the circumference of the lesion using argon plasma coagulation (APC), which was applied 1 – 2 mm outside the border of the lesion to represent the planned incision line. Endoscopic resection was performed by ESD or EMR, and was completed by total removal of the non-iodine-stained portion of the lesion and/or the marking dots. If any residual tumor was observed on the lateral margin of the artificial ulcer after endoscopic resection, electrocautery of the tumor tissue was performed using APC. Any immediate bleeding was treated by APC, a hemostatic device (FD-410LR; Olympus, Tokyo, Japan), or by compression hemostasis performed transorally by application of a cotton swab soaked in epinephrine – saline solution.

The endoscopic resection procedure was classified as either en bloc or piecemeal, according to the number of resections required. At the end of the endoscopic resection, all visible vessels on the artificial ulcer were coagulated by APC or use of hemostatic forceps, irrespective of the presence/absence of evidence of bleeding.

All endoscopic procedures were performed by the same expert endoscopist (T. T.), whose endoscopy experience up to 2006 consisted of endoscopic resection of over 500 tumors of the esophagus.

EMR techniques

EMR using a cap-fitted endoscope was mainly used from October 2006 to April 2009 for almost all lesions, whereas the strip biopsy method was used when treating protruded lesions. Saline solution containing 0.005 mg/mL epinephrine was injected into the base of the lesion with a needle. For EMR with cap, a forward-viewing endoscope (GIF-260H; Olympus) with a plastic cap (D-206 – 04; Olympus) on its tip was introduced. A small-diameter snare (SD-7P; Olympus) was opened within the plastic cap, and the lesion was aspirated into the plastic cap. The snare was then closed, and forced coagulation or blended current was applied to resect the lesion [21] [23]. A forward-viewing endoscope was also used for the strip biopsy method. After the injection of saline solution containing 0.005 mg/mL epinephrine, the lesion was strangulated using a snare, including the normal surrounding mucosa to some extent, and the tumor was then resected by application of an electrosurgical current [24] [25].

ESD techniques

ESD was mainly used from April 2009 to January 2011. ESD was usually performed using an electrosurgical needle knife (Olympus Optical, Tokyo, Japan), as previously described [19]. A forward-viewing endoscope (GIF-260H) with a plastic cap (D-201 – 11804; Olympus) on its tip was introduced. Saline solution containing 0.005 mg/mL epinephrine was injected into the subepithelium, and the mucosa was incised outside the marking dots using the needle knife, with current applied using an electrical generator. The lesion was dissected with adequate tension applied to the removed layer and additional local injection, as needed, to ensure total removal of the lesion. The dissection line included a subepithelial layer in which the muscular layer was barely visible. Finally, the lesion was removed en bloc.

Histopathological evaluation and assessment of the therapeutic efficacy

All resected specimens were sectioned into 2-mm slices and evaluated by histopathological examination. The macroscopic type was defined as follows (according to the Japanese Classification of Esophageal Cancer): protruded type, 0-I and 0-I + IIa ( + IIb); elevated type, 0-IIa ( + IIb); flat type, 0-IIb; depressed type, 0-IIc or 0-III [26]. For technical and histological assessment of the endoscopic resection, the rates of en bloc and complete resection, presence/absence of lymphovascular invasion, the sizes of the resected specimens, the treatment times, details of the postoperative clinical course, including the changes of the laboratory data, and rates of major complications were compared between the EMR and ESD groups.

Complete resection was defined as resection of the tumor en bloc with tumor-free margins. The depth of invasion was measured from the basal membrane for invasive lesions due to the lack of a muscularis mucosae in the pharynx, which makes it impossible to distinguish mucosal from submucosal cancers. For each resection, the size of the largest resected specimen and the procedure time were estimated by referral to the database or pictures taken during the procedure. The treatment time was defined as the time from the start of elevation of the larynx to removal of the endoscope after treatment. Bleeding related to the procedure was defined as bleeding that necessitated postoperative hemostatic treatment, such as coagulation therapy.

In the analyses of the treatment time, postoperative clinical course, postoperative complications (aspiration pneumonia, transient vocal cord paresis, and postoperative pharyngeal stenosis) and risk factors for severe laryngeal edema, the study utilized only data from 52 endoscopic resection sessions with a single lesion resected per endoscopic resection session. Thus, the following cases were excluded from these analyses: cases in which two or more lesions were resected during the same procedure (i. e. in the case of two lesions resected as one piece, or in cases where each lesion was resected individually, but on the same day: 16 lesions in seven patients), or those in which pharyngeal endoscopic resection and another organ surgery were performed at the same time, including esophageal endoscopic resection (four lesions in four patients), gastric ESD (two lesions in two patients), esophagotomy for advanced esophageal cancer (one lesion in one patient), and local resection of mesopharyngeal carcinoma by transoral surgery (one lesion in one patient). Severe laryngeal edema was defined as the condition in which the vocal cords of the two sides cannot be observed endoscopically due to edema of the aryepiglottic folds, and tracheotomy and/or prolonged overnight intubation is necessary.

Two specialist ear – nose – throat (ENT) pathologists independently reviewed the biopsy and resected specimens. If there was any inconsistency in the histopathological assessment, the final diagnosis was decided by a joint assessment that included a third expert ENT pathologist.

Follow-up examinations

In the Division of Endoscopy at the Cancer Institute Hospital, follow-up investigations after endoscopic resection are scheduled at 3 and 6 months after the procedure and annually thereafter, until the patient’s death. The follow-up program consists of physical examination, including palpation, laboratory blood tests, including a tumor marker test for squamous cell carcinoma, high resolution endoscopy with NBI, laryngoscopy by an otolaryngologist, ultrasound of the neck, and CT of both the neck and chest. Recurrence was diagnosed based on clinical, endoscopic, and/or radiological signs of disease. Local recurrence was diagnosed when a carcinomatous area was detected adjacent to an endoscopic resection scar and the presence of carcinoma cells could be verified histologically in a biopsy specimen. If a new superficial pharyngeal carcinoma was detected within 12 months, it was defined as a synchronous lesion, and if it was detected after 12 months, it was defined as a metachronous lesion [4]. All patients were followed up without additional treatment, on the basis of adequate informed consent, until recurrence. The overall and cause-specific survival times were evaluated from the date of the endoscopic resection to the date of death or the date of last confirmation of survival.

Statistical analysis

The relationships among the clinicopathological variables were evaluated by the chi-squared test or Fisher’s exact test for categorical data, and the Student’s t test or Mann – Whitney U test for numerical data, as appropriate. The overall and cause-specific survival curves were generated and assessed by the Kaplan – Meier analysis, starting from the date of endoscopic resection to the date of death or that of the last confirmation of survival. Independent variables with P < 0.1 (based on univariate analysis) were entered into a multivariate logistic regression model, using the Wald statistic with forward stepwise selection to identify independent predictors of severe laryngeal edema. The statistical significance level was set at P < 0.05. Statistical analyses were performed using the StatView software, version 5.0 (SAS Institute Inc., Cary, North Carolina, USA).

Results

Endoscopic resection results

The clinicopathological characteristics of the patients and characteristics of the lesions for which endoscopic resection was performed are shown in [Table 1] and [Table 2]. Neither the patient nor the lesion characteristics differed significantly between the ESD and EMR groups. The therapeutic outcomes, including the operative details and histopathological findings, are shown in [Table 3]. All resected specimens with biopsy-proven high grade dysplasia were confirmed to be squamous cell carcinomas by histopathological examination. In the lesions undergoing EMR, EMR with cap-fitted endoscope was attempted for almost all lesions (43/45 [95.6 %]), and strip biopsy was used only for two protruded lesions (2/45 [4.4 %]).

The en bloc resection and complete resection rates for ESD were significantly higher than those for EMR (en bloc 77.4 % vs. 37.8 %, P = 0.0002; complete 54.8 % vs. 28.9 %, P = 0.0379). Even for small lesions less than 15 mm in diameter, ESD yielded higher en bloc and complete resection rates than EMR, although the differences were not statistically significant (en bloc 81.8 % [9 /11] vs. 50.0 % [14/28], P = 0.0857; complete 54.5 % [6 /11] vs. 46.4 % [13/28], P = 0.7311) (data not shown). The median number of specimens per resected lesion was 1.0 for those treated by ESD and 2.0 for those treated by EMR (P = 0.0176); 11 lesions treated by EMR (24.4 %) required concurrent treatment by electrocautery with APC for small residual lesions, which was significantly more than the number of patients in the ESD group requiring such additional treatment (0 % [0/0] vs. 24.4 % [11/45]; P = 0.0021). The mean resected specimen size (± SD) was significantly larger in the ESD group compared with the EMR group (29.9 ± 11.1 mm vs. 19.5 ± 5.2 mm, respectively; P < 0.0001), even though there was no statistically significant between-group difference in the mean lesion size (20.3 ± 12.0 mm vs. 15.6 ± 6.5 mm, respectively; P = 0.1027). The treatment time for ESD was significantly longer than that for EMR (124.9 ± 65.1 minutes vs. 57.2 ± 69.6 minutes; P = 0.0014). However, the procedure time tended to shorten as the number of cases accumulated, which suggested a learning curve effect, particularly in the ESD group (data not shown).

There were three local recurrences (6.7 %) after a median follow-up of 6 months (range 5 – 6 months) in the EMR group; all of them were successfully treated by additional ESD. Additional radiotherapy was required in one patient who underwent salvage EMR after transoral local resection by otolaryngologists that resulted in incomplete resection.

Postoperative clinical course

Details of the postoperative clinical courses are described in [Table 4]. The duration of hospitalization was significantly longer in the ESD group (mean 10.9 days, range 4 – 38 days) compared with the EMR group (mean 7.6 days, range 4 – 33 days) (P = 0.0027). Oral food intake could be resumed a mean of 3.6 days after ESD (range 1 – 19 days), which was significantly later than the mean of 1.6 days (range 1 – 14 days) after EMR (P = 0.0115); this was because some patients in the ESD group had to undergo tracheotomy after endoscopic resection or suffered from aspiration pneumonia. There were no statistically significant differences between the ESD and EMR groups in terms of the duration of persistent fever, postoperative maximum body temperature, white blood cell counts or serum C-reactive protein levels on the first postoperative day, or in the duration of anti-inflammatory drugs needed for throat pain.

Complications and short-term follow-up data

As shown in [Table 5], complications included one case of postoperative bleeding that occurred shortly after extubation post-ESD. This was successfully treated by coagulation therapy using an oral approach without the need for transfusion; however, the patient required emergency tracheotomy due to severe laryngeal edema and the possibility of further delayed bleeding. There was one case of micro-perforation, associated with pneumohypoderma of the neck, during the ESD procedure for a lesion that was located in the right pyriform sinus but had spread to the esophageal orifice. It was presumed that the pneumohypoderma was caused by micro-perforation of the deeply dissected and widely exposed muscle layer, although the perforation site could not be visualized endoscopically. The patient recovered well with intravenous antibiotic treatment administered for 2 days and by withholding intake by mouth for 4 days. No case of postoperative pharyngeal stenosis was observed in either group, possibly because lesions with greater than two-thirds of circumferential involvement of the pharyngeal lumen were excluded from the study. Six patients suffered from aspiration pneumonia, which was successfully managed by antibiotic therapy for 1 week. Transient vocal cord paresis was observed after EMR in one case with lesions located in the left pyriform sinus that had spread to the aryepiglottis; the paresis resolved within a few days without any intervention. None of the remaining patients had any major complications or any loss of function in terms of swallowing or speaking, other than throat pain and discomfort, which were relieved by pethidine hydrochloride or nonsteroidal anti-inflammatory drugs.

Risk factors for severe laryngeal edema

ESD was more frequently complicated by severe laryngeal edema compared with EMR (4/21 [19.0 %] vs. 1/31 [3.2 %]; P = 0.1446). [Table 6] summarizes the results of the analyses performed to identify the prognostic factors in the six cases of severe laryngeal edema that required tracheotomy; in all of these patients, the tube was removed within 2 weeks of the procedure. In the lesion-related analysis, one patient undergoing three endoscopic resections in a single session (two lesions resected by EMR and one by ESD) was excluded. A univariate analysis identified large lesion size (P = 0.0061) and long treatment time (P = 0.0099) as statistically significant risk factors for severe laryngeal edema. Long treatment time (≥ 150 minutes) was also extracted by multivariate analysis as an independent variable predictive of severe laryngeal edema (odds ratio 40.7, 95 % confidence interval [CI] 2.5 – 673.7; P = 0.0097).

Long term outcomes

During the study period, one case of right cervical lymph node metastasis, confirmed by ultrasound-guided fine needle aspiration biopsy, was detected during follow-up at 6 months after EMR, and right radical neck dissection was performed for this patient. The lesion was a 20-mm 0-IIb + IIa lesion in the right pyriform sinus, with a microinvasion depth of 1100 μm and lymphovascular involvement. Eight patients died of other co-existing diseases by a median follow-up time of 26.5 months (range 2 – 41 months) and six patients developed metachronous superficial pharyngeal carcinomas, which were successfully removed by repeat endoscopic resection. The 3-year overall and cause-specific survival rates were 79.8 % (95 %CI 65.7 % – 93.9 %) and 100 %, respectively, at a median follow-up time of 26.0 months (range 2 – 50 months).

Discussion

Complete resection of superficial cancers is vitally important because local recurrence appears to arise from small remnants of cancer left on the margin of resected ulcers. Achieving complete resection would minimize the need for frequent endoscopic examinations to check for any local recurrence. In the current study, ESD yielded higher en bloc and complete resection rates than EMR and was found to be suitable for treating pharyngeal carcinomas; however, ESD has several disadvantages, including a higher rate of severe laryngeal edema, longer procedure time, and longer hospital stay.

EMR techniques have proved to be effective for the resection of superficial pharyngeal lesions and have a low complication rate, although the en bloc resection rate seems to be relatively low. Suzuki et al. reported that the en bloc resection rate obtained with EMR of superficial pharyngeal carcinomas was 43 %, which is similar to the results obtained in the current study (37.8 %) [11]. For pharyngeal lesions, it is difficult to perform resection at sites of cartilage protrusions and on the posterior side of the pharynx, because the anatomy prevents optimal fit of the endoscopic cap. Moreover, suction of the targeted area into the cap and strangulation by the endoscopic snare often do not allow a precise resection of the targeted area.

These technical limitations may explain why en bloc EMR was achieved in only 17 of the 45 cases (37.8 %) in the current study. Because of these problems with en bloc resection, EMR is performed piecemeal in most cases, and in this study concurrent APC treatment was necessitated for small residual lesions in 24.4 % (11/45). However, as local recurrence occurred in only 3 of the 32 incomplete resection lesions (9.4 %) over a median follow-up of 26.0 months, we speculate that the addition of concurrent APC treatment may have helped prevent local recurrence.

In contrast, ESD allows the endoscopist to remove the lesion in one piece while keeping the size of the resected specimen to a minimum by enabling circumferential cutting with a small safety margin between the tumor border and the coagulation points that delineate the incision line [27]. In the current study, significantly higher en bloc resection and complete resection rates were obtained with ESD than with EMR ([Table 3]). Iizuka et al. reported that ESD yielded high rates of both en bloc and complete resection of pharyngeal carcinoma (93.8 % [15/16] and 87.5 % [14/16], respectively) [13], which are superior to the results presented here. The main reason for the relatively low rate of complete resection compared with the en bloc resection rate was the difficulty in histological determination of the lateral and/or vertical margins caused by detachment of the mucosa and/or thermal tissue damage due to use of the electrosurgical knife over an extended period of time (50.0 % [8/16]); however, according to naked-eye observations, complete resection was achieved in almost all lesions classified as en bloc resection ([Fig. 1]). In addition, the circumferential incision was made just 1 – 2 mm outside of the lesion border because a concerted effort was made to minimize the resected area in order to prevent the late complications of ESD, such as pharyngeal stenosis, delayed bleeding, and laryngeal edema. However, as a result of the small distance between the tumor border and the incision, inadvertent burning of adjacent tissues and/or detachment of the mucosa could have easily reached the tumor border, in which case a histologically normal specimen margin could not have been maintained.

In this study, the risk factors for severe laryngeal edema, identified by univariate analysis, were a large lesion size and a long treatment time. Of these, long treatment time ( > 150 minutes) was extracted by multivariate analysis as an independent risk factor for severe laryngeal edema. These risk factors may be related to the time duration for which the endoscope is in direct contact with the laryngeal mucosa, the amount of injection solution used, and the degree of thermal tissue damage, although the volume of injection solution was not measured in this study. Moreover, the status of lymphatic and venous flow might affect the edematous condition, especially in the case of postoperative neck dissection for cervical metastasis and/or following radiation therapy. In addition, wide lesions located in the posterior wall of the pharynx may affect the risk of development of bilateral edema of the aryepiglottis, although the differences related to the macroscopic tumor type did not reach statistical significance ([Table 6]).

The treatment time required to perform ESD was significantly longer compared with that for EMR. However, the longer treatment time may be acceptable when interpreted from the point of view of the superior effectiveness of ESD, except perhaps in patients considered to be at high risk because of advanced age or poor cardiopulmonary status. The major complication rates were not significantly different between the treatment groups, although the duration of follow-up in this study was relatively short (median 26.0 months, range 2 – 55 months). ESD was more frequently complicated by severe laryngeal edema than EMR was, although the difference was not statistically significant. Laryngeal edema may involve the aryepiglottic folds bilaterally, making it difficult to remove the intubation tube on the day of the procedure, particularly in patients with widely spread lesions. After appropriately obtaining informed consent before the procedure and diagnosing the condition of edema after the procedure, tracheotomy and/or an artificial respirator should be used without hesitation when removal of the intubation tube is considered difficult because of extensive edema, or when the patient is at high risk of delayed bleeding.

In this study, 11.9 % (17 lesions in seven patients, 7/59) and 10.2 % (seven lesions in six patients, 6/59) of patients had synchronous or metachronous superficial pharyngeal carcinomas, respectively, during a median follow-up period of 26.0 months, respectively. Synchronous lesions may have been present before the original resection but may have gone undetected, presumably because they were too small or because of unpredictable difficulties during endoscopy, such as excessive gag and cough reflexes. Thus, continued surveillance is essential, even after complete resection, because multiple carcinomas occur often according to the “field cancerization” concept [28].

We recognize that this single-center study has limitations, including those inherent to a retrospective design. First, not all superficial pharyngeal carcinomas diagnosed during the study period were treated by endoscopic resection. For example, EMR is more difficult in the posterior pharynx and lesions were treated by transoral surgery or radiation therapy in the earlier period. This may have introduced a selection bias, because lesions in this area may have been selected due to their resectability based on prior experiences with EMR. However, possible selection bias was minimized by including all patients with pharyngeal lesions treated by endoscopic resection who were identified within the database, and the original data had been collected prospectively over the original study period. Second, a possible bias may have been introduced as a result of the differences in the treatment periods: EMR was used in the earlier period, whereas ESD was used mainly in the later period. However, this bias was minimized because the same devices and techniques were used by a single endoscopist who performed endoscopic resection throughout the study period. Therefore, we consider that the different treatment strategies between the earlier and later time treatment periods would have had minimal impact on the results, although it may make the results less applicable to other centers. Moreover, because EMR was performed before ESD, the experience with EMR for lesions in the pharynx likely helped in ESD of lesions in the same region during the later time period. This study, by its nature, examined the performance of ESD after experience had been obtained in performing pharyngeal EMR. A trial to compare the two methods would best have been performed using a randomized design with the two procedures performed concurrently during the same period. A third limitation was the lack of comparison of endoscopic resection with any other treatment modality, especially transoral surgery, such as endoscopic laryngopharyngeal surgery performed by surgeons or otolaryngologists. However, pharyngeal lesions are an indication for endoscopic resection, which warrants evaluation by this type of preliminary study. The fourth limitation was the small number of patients followed up for a relatively short period, and the fact that the follow-up period for patients treated with EMR was longer because this procedure was performed earlier. In this study, the most important factor affected by the duration of follow-up was the local recurrence rate. All of the cases of local recurrence were detected within 6 months of the endoscopic resection by ME-NBI. In the ESD group, only one case was followed up for less than 6 months, but complete resection was achieved in this case; therefore, the possibility of development of local recurrence was remote. Large-scale, prospective, multicenter trials are needed to confirm whether ESD might be the most effective type of surgery for endoscopic resection, and to document with certainty the benefits and oncological safety of ESD for superficial pharyngeal carcinomas, including the frequency of, and risk factors for, lymph node metastasis.

In summary, this study demonstrated that ESD yields significantly higher rates of both en bloc and complete resection of pharyngeal lesions compared with EMR. The longer time required to perform ESD would seem to be acceptable in view of the higher efficacy of this procedure, even though it is associated with a higher risk of tracheotomy and/or prolonged intubation. Further clinical trials with a higher number of cases are needed to confirm the results of this initial comparison of ESD and EMR.

Competing interests: None.

• References

• 1 Erkal HS, Mendenhall WM, Amdur RJ et al. Synchronous and metachronous squamous cell carcinomas of the head and neck mucosal sites. J Clin Oncol 2001; 19: 1358-1362
  PubMedSearch in Google Scholar
• 2 Muto M, Nakane M, Katada C et al. Squamous cell carcinoma in situ at oropharyngeal and hypopharyngeal mucosal sites. Cancer 2004; 101: 1375-1381
  CrossrefPubMedSearch in Google Scholar
• 3 Muto M, Minashi K, Yano T et al. Early detection of superficial squamous cell carcinoma in the head and neck region and esophagus by narrow band imaging: a multicenter randomized controlled trial. J Clin Oncol 2010; 28: 1566-1572
  CrossrefPubMedSearch in Google Scholar
• 4 Muto M, Hironaka S, Nakane M et al. Association of multiple Lugol-voiding lesions with synchronous and metachronous esophageal squamous cell carcinoma in patients with head and neck cancer. Gastrointest Endosc 2002; 56: 517-521
  CrossrefPubMedSearch in Google Scholar
• 5 Nonaka S, Saito Y. Endoscopic diagnosis of pharyngeal carcinoma by NBI. Endoscopy 2008; 40: 347-351
  Thieme ConnectPubMedSearch in Google Scholar
• 6 Kraus DH, Zelefsky MJ, Brock HA et al. Combined surgery and radiation therapy for squamous cell carcinoma of the hypopharynx. Otolaryngol Head Neck Surg 1997; 116: 637-641
  CrossrefPubMedSearch in Google Scholar
• 7 Johansen LV, Grau C, Overgaard J. Hypopharyngeal squamous cell carcinoma – treatment results in 138 consecutively admitted patients. Acta Oncol 2000; 39: 529-536
  CrossrefPubMedSearch in Google Scholar
• 8 Eckel HE, Staar S, Volling P et al. Surgical treatment for hypopharynx carcinoma: feasibility, mortality, and results. Otolaryngol Head Neck Surg 2001; 124: 561-569
  CrossrefPubMedSearch in Google Scholar
• 9 Fang FM, Tsai WL, Chien CY et al. Changing quality of life in patients with advanced head and neck cancer after primary radiotherapy or chemoradiation. Oncology 2005; 68: 405-413
  CrossrefPubMedSearch in Google Scholar
• 10 Muto M, Satake H, Yano T et al. Long-term outcome of transoral organ-preserving pharyngeal endoscopic resection for superficial pharyngeal cancer. Gastrointest Endosc 2011; 74: 477-484
  CrossrefPubMedSearch in Google Scholar
• 11 Suzuki H, Saito Y, Oda I et al. Feasibility of endoscopic mucosal resection for superficial pharyngeal cancer: a minimally invasive treatment. Endoscopy 2010; 42: 1-7
  Thieme ConnectPubMedSearch in Google Scholar
• 12 Shimizu Y, Yamamoto J, Kato M et al. Endoscopic submucosal dissection for treatment of early stage hypopharyngeal carcinoma. Gastrointest Endosc 2006; 64: 255-259 discussion 260-252
  CrossrefPubMedSearch in Google Scholar
• 13 Iizuka H, Okamura S, Onozato Y et al. Endoscopic submucosal dissection for colorectal tumors. Gastroenterol Clin Biol 2009; 33: 1004-1011
  CrossrefPubMedSearch in Google Scholar
• 14 Oda I, Saito D, Tada M et al. A multicenter retrospective study of endoscopic resection for early gastric cancer. Gastric Cancer 2006; 9: 262-270
  CrossrefPubMedSearch in Google Scholar
• 15 Oka S, Tanaka S, Kaneko I et al. Advantage of endoscopic submucosal dissection compared with EMR for early gastric cancer. Gastrointest Endosc 2006; 64: 877-883
  CrossrefPubMedSearch in Google Scholar
• 16 Esaki M, Matsumoto T, Hirakawa K et al. Risk factors for local recurrence of superficial esophageal cancer after treatment by endoscopic mucosal resection. Endoscopy 2007; 39: 41-45
  Thieme ConnectPubMedSearch in Google Scholar
• 17 Ishihara R, Tanaka H, Iishi H et al. Long-term outcome of esophageal mucosal squamous cell carcinoma without lymphovascular involvement after endoscopic resection. Cancer 2008; 112: 2166-2172
  CrossrefPubMedSearch in Google Scholar
• 18 Ono H, Kondo H, Gotoda T et al. Endoscopic mucosal resection for treatment of early gastric cancer. Gut 2001; 48: 225-229
  CrossrefPubMedSearch in Google Scholar
• 19 Oyama T, Tomori A, Hotta K et al. Endoscopic submucosal dissection of early esophageal cancer. Clin Gastroenterol Hepatol 2005; 3: 67-70
  CrossrefPubMedSearch in Google Scholar
• 20 Neuhaus H, Wirths K, Schenk M et al. Randomized controlled study of EMR versus endoscopic submucosal dissection with a water-jet hybrid-knife of esophageal lesions in a porcine model. Gastrointest Endosc 2009; 70: 112-120
  CrossrefPubMedSearch in Google Scholar
• 21 Inoue H, Takeshita K, Hori H et al. Endoscopic mucosal resection with a cap-fitted panendoscope for esophagus, stomach, and colon mucosal lesions. Gastrointest Endosc 1993; 39: 58-62
  CrossrefPubMedSearch in Google Scholar
• 22 Japan Society for Head and Neck Cancer. General rules for clinical studies on head and neck cancer. Tokyo: Kanehara; 2005
  Search in Google Scholar
• 23 Inoue H, Endo M, Takeshita K et al. A new simplified technique of endoscopic esophageal mucosal resection using a cap-fitted panendoscope (EMRC). Surg Endosc 1992; 6: 264-265
  CrossrefPubMedSearch in Google Scholar
• 24 Tada M, Shimada M, Murakami F et al. Development of strip-off biopsy [in Japanese with English abstract]. Gastroenterol Endosc 1984; 26: 833-839
  PubMedSearch in Google Scholar
• 25 Tada M. One piece resection and piece meal resection of early gastric cancer by strip biopsy [in Japanese with English abstract]. Igaku-Shoin 1998; 68-87
  PubMedSearch in Google Scholar
• 26 Japanese Society for Esophageal Disease. Guidelines for clinical and pathological studies on carcinoma of the esophagus. [in Japanese] 10th. edition Tokyo: Kanehara; 2007
  Search in Google Scholar
• 27 Soetikno R, Kaltenbach T, Yeh R et al. Endoscopic mucosal resection for early cancers of the upper gastrointestinal tract. J Clin Oncol 2005; 23: 4490-4498
  CrossrefPubMedSearch in Google Scholar
• 28 Slaughter DP, Southwick HW, Smejkal W. Field cancerization in oral stratified squamous epithelium; clinical implications of multicentric origin. Cancer 1953; 6: 963-968
  CrossrefPubMedSearch in Google Scholar

Corresponding author

• References

• 1 Erkal HS, Mendenhall WM, Amdur RJ et al. Synchronous and metachronous squamous cell carcinomas of the head and neck mucosal sites. J Clin Oncol 2001; 19: 1358-1362
  PubMedSearch in Google Scholar
• 2 Muto M, Nakane M, Katada C et al. Squamous cell carcinoma in situ at oropharyngeal and hypopharyngeal mucosal sites. Cancer 2004; 101: 1375-1381
  CrossrefPubMedSearch in Google Scholar
• 3 Muto M, Minashi K, Yano T et al. Early detection of superficial squamous cell carcinoma in the head and neck region and esophagus by narrow band imaging: a multicenter randomized controlled trial. J Clin Oncol 2010; 28: 1566-1572
  CrossrefPubMedSearch in Google Scholar
• 4 Muto M, Hironaka S, Nakane M et al. Association of multiple Lugol-voiding lesions with synchronous and metachronous esophageal squamous cell carcinoma in patients with head and neck cancer. Gastrointest Endosc 2002; 56: 517-521
  CrossrefPubMedSearch in Google Scholar
• 5 Nonaka S, Saito Y. Endoscopic diagnosis of pharyngeal carcinoma by NBI. Endoscopy 2008; 40: 347-351
  Thieme ConnectPubMedSearch in Google Scholar
• 6 Kraus DH, Zelefsky MJ, Brock HA et al. Combined surgery and radiation therapy for squamous cell carcinoma of the hypopharynx. Otolaryngol Head Neck Surg 1997; 116: 637-641
  CrossrefPubMedSearch in Google Scholar
• 7 Johansen LV, Grau C, Overgaard J. Hypopharyngeal squamous cell carcinoma – treatment results in 138 consecutively admitted patients. Acta Oncol 2000; 39: 529-536
  CrossrefPubMedSearch in Google Scholar
• 8 Eckel HE, Staar S, Volling P et al. Surgical treatment for hypopharynx carcinoma: feasibility, mortality, and results. Otolaryngol Head Neck Surg 2001; 124: 561-569
  CrossrefPubMedSearch in Google Scholar
• 9 Fang FM, Tsai WL, Chien CY et al. Changing quality of life in patients with advanced head and neck cancer after primary radiotherapy or chemoradiation. Oncology 2005; 68: 405-413
  CrossrefPubMedSearch in Google Scholar
• 10 Muto M, Satake H, Yano T et al. Long-term outcome of transoral organ-preserving pharyngeal endoscopic resection for superficial pharyngeal cancer. Gastrointest Endosc 2011; 74: 477-484
  CrossrefPubMedSearch in Google Scholar
• 11 Suzuki H, Saito Y, Oda I et al. Feasibility of endoscopic mucosal resection for superficial pharyngeal cancer: a minimally invasive treatment. Endoscopy 2010; 42: 1-7
  Thieme ConnectPubMedSearch in Google Scholar
• 12 Shimizu Y, Yamamoto J, Kato M et al. Endoscopic submucosal dissection for treatment of early stage hypopharyngeal carcinoma. Gastrointest Endosc 2006; 64: 255-259 discussion 260-252
  CrossrefPubMedSearch in Google Scholar
• 13 Iizuka H, Okamura S, Onozato Y et al. Endoscopic submucosal dissection for colorectal tumors. Gastroenterol Clin Biol 2009; 33: 1004-1011
  CrossrefPubMedSearch in Google Scholar
• 14 Oda I, Saito D, Tada M et al. A multicenter retrospective study of endoscopic resection for early gastric cancer. Gastric Cancer 2006; 9: 262-270
  CrossrefPubMedSearch in Google Scholar
• 15 Oka S, Tanaka S, Kaneko I et al. Advantage of endoscopic submucosal dissection compared with EMR for early gastric cancer. Gastrointest Endosc 2006; 64: 877-883
  CrossrefPubMedSearch in Google Scholar
• 16 Esaki M, Matsumoto T, Hirakawa K et al. Risk factors for local recurrence of superficial esophageal cancer after treatment by endoscopic mucosal resection. Endoscopy 2007; 39: 41-45
  Thieme ConnectPubMedSearch in Google Scholar
• 17 Ishihara R, Tanaka H, Iishi H et al. Long-term outcome of esophageal mucosal squamous cell carcinoma without lymphovascular involvement after endoscopic resection. Cancer 2008; 112: 2166-2172
  CrossrefPubMedSearch in Google Scholar
• 18 Ono H, Kondo H, Gotoda T et al. Endoscopic mucosal resection for treatment of early gastric cancer. Gut 2001; 48: 225-229
  CrossrefPubMedSearch in Google Scholar
• 19 Oyama T, Tomori A, Hotta K et al. Endoscopic submucosal dissection of early esophageal cancer. Clin Gastroenterol Hepatol 2005; 3: 67-70
  CrossrefPubMedSearch in Google Scholar
• 20 Neuhaus H, Wirths K, Schenk M et al. Randomized controlled study of EMR versus endoscopic submucosal dissection with a water-jet hybrid-knife of esophageal lesions in a porcine model. Gastrointest Endosc 2009; 70: 112-120
  CrossrefPubMedSearch in Google Scholar
• 21 Inoue H, Takeshita K, Hori H et al. Endoscopic mucosal resection with a cap-fitted panendoscope for esophagus, stomach, and colon mucosal lesions. Gastrointest Endosc 1993; 39: 58-62
  CrossrefPubMedSearch in Google Scholar
• 22 Japan Society for Head and Neck Cancer. General rules for clinical studies on head and neck cancer. Tokyo: Kanehara; 2005
  Search in Google Scholar
• 23 Inoue H, Endo M, Takeshita K et al. A new simplified technique of endoscopic esophageal mucosal resection using a cap-fitted panendoscope (EMRC). Surg Endosc 1992; 6: 264-265
  CrossrefPubMedSearch in Google Scholar
• 24 Tada M, Shimada M, Murakami F et al. Development of strip-off biopsy [in Japanese with English abstract]. Gastroenterol Endosc 1984; 26: 833-839
  PubMedSearch in Google Scholar
• 25 Tada M. One piece resection and piece meal resection of early gastric cancer by strip biopsy [in Japanese with English abstract]. Igaku-Shoin 1998; 68-87
  PubMedSearch in Google Scholar
• 26 Japanese Society for Esophageal Disease. Guidelines for clinical and pathological studies on carcinoma of the esophagus. [in Japanese] 10th. edition Tokyo: Kanehara; 2007
  Search in Google Scholar
• 27 Soetikno R, Kaltenbach T, Yeh R et al. Endoscopic mucosal resection for early cancers of the upper gastrointestinal tract. J Clin Oncol 2005; 23: 4490-4498
  CrossrefPubMedSearch in Google Scholar
• 28 Slaughter DP, Southwick HW, Smejkal W. Field cancerization in oral stratified squamous epithelium; clinical implications of multicentric origin. Cancer 1953; 6: 963-968
  CrossrefPubMedSearch in Google Scholar