Endoscopic Mucosal Resection: An Overview of the Value of Different Techniques

Two articles in this issue of Endoscopy advocate the use of endoscopic mucosal resection (EMR), in a new treatment technique for intramucosal gastric tumors using an insulated-tip diathermic knife [1], and a novel approach to endoscopic colorectal mucosal resection using a three-channel outer tube and multiple forceps in an experimental assessment [2]. EMR, which is a recently introduced therapeutic method for resection of flat or depressed lesions of the gastrointestinal tract, has come to play an increasingly important role in the treatment of early cancer in the gastrointestinal tract. Most importantly, long-term studies which demonstrated EMR outcomes similar to those of surgery have led to the acceptance of EMR as a standard procedure [3].

Why would endoscopists consider EMR in the first place? If a lesion is to be removed endoscopically, it could be removed by either ablation or resection. Ablation may be technically easier and faster than resection, but it is usually less desirable because a specimen cannot be obtained for histological confirmation. Which lesions are suited for EMR? This procedure is potentially curative in early cancers of the gastrointestinal tract with a negligible risk of lymphatic metastasis. It has been shown in the surgical literature that the risk of lymphatic invasion increases with progressive infiltration of the mucosa and submucosa. The risk is almost nil when invasion is limited to the mucosa, very low (around 10 %) if the muscularis mucosa is invaded, and increases to rates as high as 40 % when the submucosa is invaded, depending on whether the superficial (sm1) or deeper (sm2) portions are involved [4]. Based on these findings, Japanese authors have recommended that stage m1 and m2 early cancers represent an absolute indication for mucosal resection, while m3 and sm1 cancers should represent a relative indication. A retrospective study of the 1312 solitary early gastric cancers resected at the National Cancer Center Hospital between 1962 and 1991 revealed that in mucosal cancers the incidence of metastases was extremely low at 0.64 % (3/462) in intestinal type, regardless of size, though several lesions of the diffuse type still showed metastases despite their small size [5]. In general, lymph node metastasis of early gastric cancer showed a rate of 0 % to 3 % in mucosal cancer and 9 % to 19 % in cancer invading the submucosa [6].

Various factors which should be considered prior to the performance of EMR, other than infiltrative depth of lesion, have been suggested, such as the endoscopic appearance of early cancers in the gastrointestinal tract, their size, and their differentiation grade on histological examination. Many endoscopists have established various criteria for application of EMR. Generally, the absolute indications for early gastric cancer have been those accepted by the Japanese Gastroenterological Endoscopy Society: (i) elevated-type intramucosal cancer less than 20 mm in size; (ii) depressed-type mucosal cancer without ulceration, less than 10 mm in size; and (iii) intestinal-type adenocarcinoma. In the esophagus, the criteria for EMR are as follows: superficial esophageal cancer of type IIa, IIb, or IIc, not more than 2 cm in size, and intraepithelial cancer or m1 cancer (cancer with minimal invasion into the muscularis mucosa). In a large survey of more than 2000 patients who had surgery, endoscopic treatment, or both, the incidences of lymph node metastases were 0 % and 3 % in patients with m1 and m2 stage cancers, respectively [3]. The role of EMR in the treatment of early cancer of the esophagus is not yet clear because of the limited number of reports published to date. In the colon, the criteria for EMR are as follows: well differentiated adenocarcinoma of type Is, IIb, IIa, or IIc, not more than 1 cm in size, m cancer (cancer confined to the mucosa) or smIa cancer (cancer with minimal invasion to submucosa), and without any invasion to lymphatic channels or vessels [7] [8] [9] [10] .

There is a general consensus in the specification of the indications for EMR, although follow-up procedures and management of incomplete resections and recurrent cancers are still subject to differing opinions. According to data from The Cancer Institute Hospital in Tokyo, which involves more than 10 000 resected gastric cancer cases, IIc-type mucosal cancer of less than 10 mm is associated with no lymph node involvement, while IIc-type mucosal cancer of less than 20 mm has lymph node involvement in 0.4 % of cases. No lymph node metastasis is found with IIa-type mucosal cancer of less than 20 mm and IIa-type submucosal cancer of less than 20 mm also has no lymph node involvement. However, submucosal cancer of less than 30 mm shows lymph node involvement in 28.6 % of cases [11]. The differentiation in the depth tumor infiltration between mucosal and submucosal cancer in IIa-type elevated lesion is relatively problematic. In poorly differentiated carcinoma some risk of lymph-node metastasis exists even in small lesions, therefore those lesions should be generally excluded from the application of mucosal resection.

With regard to EMR indications concerning the size of the lesion, some trials are now underway using piecemeal resection. When piecemeal resection is carried out, the residual portion between the first and second resected areas must not be left because it could easily develop local recurrence. Divided resection is gradually gaining general support in Japan, although it is still controversial. Amano et al. recommended an extended indication for mucosal resection in early gastric cancer as follows: all well-differentiated adenocarcinomas, mucosal cancers less than 30 mm in size without ulcer or ulcer scars; mucosal cancer less than 20 mm in size with ulcer or ulcer scar; sm1 cancer less than 20 mm in size without ulcer or ulcer scar; and poorly differentiated cancer less than 10 mm in size [4]. The rate of cure in 69 cases was statistically similar to that in patients with absolute indications under the conventional criteria (98 %). These new extended indications should be applied in clinical cases on a prospective study basis and restricted only to patients who are not candidates for surgery.

In recent years, progress has been made in the development of endoscopic resection techniques for early cancers in the gastrointestinal tract. These techniques could be broadly divided into three categories: techniques without suction, techniques with suction, and modified techniques (Table [1])[1] [2] [8] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] . These recent diverse techniques have varied advantages and disadvantages, depending on the size and location of lesions and the complexity of procedures. In 1973 Dehle first described submucosal injection with physiological saline as a technique for raising sessile polyps in the colon to facilitate snare excision (injection and cut technique) [12]. Submucosal injection may also improve the safety of EMR by providing a buffer zone for resection; the injected saline separates the lesion in the mucosa/submucosa from the underlying muscularis propria and thereby minimizes the risk of perforation. Endoscopic resection with local injection of highly concentrated saline and epinephrine (ERHSE), which has been devised by Hirao et al., follows principles similar to those of strip biopsy [14]. However in ERHSE, after marking around the target lesion, 15 - 40 ml of a hypertonic saline-epinephrine solution (40 ml of 50 % hypertonic glucose solution plus 1 ml of 0.1 % epinephrine plus indigocarmine) are injected, instead of the saline solution used in strip biopsy. The mucosa surrounding the target lesion is cut with a needle knife, using electrical current and marking, and snared for removal of the lesion. The inject, precut and cut technique permit the resection of larger lesions, but the risk of perforation by the needle knife is higher than that in strip biopsy.

To completely remove large lesions in a single session and to decrease the frequency of perforation, Hosokawa has developed a new EMR procedure using an insulated-tip (IT) electrosurgical knife which has a ceramic bulb at the tip of the incising needle to prevent electric leakage toward the deeper layers of the stomach [33]. In this issue of Endoscopy the report by Ohkuwa et al. describes the excellent results of this new technique [1]. The circumferential incision with the IT knife can overcome the difficulties regarding histological diagnosis of the depth of invasion and/or the tumor margin in the resected specimen. The authors believe that EMR using the IT knife is effective for en bloc resection of larger flat lesions between 11 and 20 mm, with less perforation than with the conventional inject, precut, and cut technique. The IT-knife alone cannot overcome the limitations of EMR or IT-EMR with regard to size of lesions, for those located in the upper part of the stomach, since these require attention to bleeding and a long procedure time.

In 1976 Martin et al. described the lift and cut technique for biopsy of submucosal lesions, using a double-channel endoscope [15]. Extending this technique to EMR, Tada et al. described „strip biopsy” using a double-channel endoscope to completely excise an early cancer [13]. The obvious drawback of the technique is that it requires a double-channel endoscope to enable both lifting and cutting. Furthermore, mucosectomy using the double-channel instrument may be difficult when the angle of access is tangential, such as along the lesser curvature of the body or the posterior wall of the antrum. To overcome this problem, Takechi et al. have described an alternative approach using two narrow diameter endoscopes inserted alongside each other, whereby one endoscope is used to lift the lesion while the other is used to resect the mucosa at its base [34]. EMR using the esophageal endoscopic mucosal resection tube (EEMRT), which has been developed by Makuuchi et al., is a technique which enables resection of the lesion by using a long transparent overtube designed to induce intraluminal negative pressure by endoscopic suction [22]. The overtube contains a small accessory channel alongside the tube for the insertion of the snare. After the lesion is sucked into the overtube, the snare is placed around the lesion and the pseudotumor is excised at its base. This technique has the advantage of enabling resection of larger lesions, but has the limitation that it cannot be used in sites other than the esophagus. This technique requires acquisition and insertion of the accessory device. Insertion of the long tube, 18 mm in diameter with an obliquely cut tip, is unpleasant for the patient and can be difficult.

Colorectal EMR has limitations, both anatomically and technically, when it is done with the conventional snare wire or double-channel endoscope. In practice, the intestinal mucosa is so soft and snaring is so easy that the double-channel endoscope is not necessary for colonic strip biopsy, and colonic depressed-type lesions larger than 10 mm are not ideal candidates for curative EMR because of the high risk of perforation and the difficulty in accessing lesions with the instruments. In this issue of Endoscopy, Mukai et al. describe a new experimental method for overcoming the above mentioned problems in colorectal EMR [2]. They employed a three-channel outer tube with three forceps and a small caliber colonoscope with a needle-type precutting knife. This new three-channel outer tube was based originally on the Makuuchi EEMR tube, extended from one channel to three channels for more safety, convenience, and complete en bloc resection of the lesion. This method has some advantages in that colonoscope re-insertion and tissue retrieval is easier and large en bloc mucosal resection is possible, but a special small caliber colonoscope is needed and insertion of the special forceps through the long channel tube is difficult, in addition to procedures which are complex and problematic.

Other authors, Fleischer et al. and Inoue et al. have described the use of the transparent cap adapter included in the Stiegmann-Goff endoscopic ligator kit to perform EMR (EMRL) using the suck and cut technique [19] [27] [35] . After submucosal injection, the lesion is sucked into the cap and transected at its base with a mini-snare inserted through the working channel. This technique is easier than others, but has the limitation that it requires three insertions of the endoscope to resect a lesion larger than about 1.5 cm in diameter. Endoscopic mucosal resection with a cap (EMRC) and EMRL have been reported to be easier to carry out than the inject and cut technique, when the lesions are located in the lesser curvature or posterior wall of the stomach [6] [36] . The EMRC technique devised by Inoue et al. is a simpler and easier refinement of EMR methods which use specialized caps [24]. Inoue et al. have popularized the technique in which a specially designed cap is affixed to the distal end of the endoscope. A special snare is looped into the lip of the cap, into which the pathologic lesion is suctioned, and then snaring and cutting could be performed without difficulty. Inoue emphasized that the EMRC procedure has three advantages. Firstly, it is simple. The clinician performs the procedure using a standard diagnostic endoscope. Theoretically, clinicians can carry out EMRC without additional training, because the technique is based on the widely used technique of endoscopic band ligation, and they can retrieve the specimen easily by placing it in the cap. Secondly, EMRC has wide applicability. The procedure can be performed in any part of the gastrointestinal tract, including areas where endoscopic visualization is difficult (the lesser curvature of the upper third of the stomach or the cardial area). Thirdly, this treatment is repeatable [7]. A larger lesion can be removed using a larger cap and specimens up to 30 mm in diameter have been recovered. However, this technique has the disadvantage that piecemeal resection is needed to deal with lesions larger than 1.5 cm in diameter. Complete excision may not be possible using the cap attachment, either because the lesion does not fit into the cap, or suction is inadequate for drawing it completely into the cap.

Endoscopic double-snare polypectomy (EDSP) using a double-channel endoscope, described by Takekoshi et al., has indications limited to protruding lesions [17]. In this method, two snares are inserted, one in each channel, with one passing through the loop of the other; the first snare is used to grasp and lift up the lesion, and it is strangulated using the second snare, for complete resection. Inatsuchi et al. reported that conventional EMR is complicated by the major problem of difficulty in defining the oral and anal sides of the resected specimen. To overcome this difficulty, Inatsuchi and colleagues have developed the method of EMR with four-point fixation. With this method the area of resection is defined by short colored clips, thereby facilitating snare cautery [20]. Soehendra et al. have used a specialized 0.4 mm monofilament stainless steel wire snare to entrap the diseased mucosa without using a specialized tube or cap [28]. In this procedure, the polypectomy snare is placed around the tumor and is pressed firmly against the mucosal surface. Recently, the present author and colleagues have produced a newly designed snare with injection needle in a single catheter for use in EMR with injection and snaring of the lesion immediately before the disappearance of pseudotumor [21]. This method may be considered to be a relatively simple and easy technique for EMR. However, further clinical evaluation and more technical improvements are needed.

Several combined laparoscopic and gastroscopic excision methods have been developed, such as percutaneous transgastric wall endoscopic mucosal resection (PT-EMR) and direct puncture endoscopic mucosal resection (DP-EMR). With these methods, the gastric mucosa containing the lesion is retracted by grasping forcepses which are introduced percutaneously and through the gastric wall; the lesion is then cut off by a snare introduced through a gastroscope. Recently, Ono and colleagues described percutaneous traction-assisted EMR (PTA-EMR). This method raises up a clip attached to the edge of the lesion, by means of a thin retractor through the abdominal wall, then the lesion is resected using the usual IT knife method. It can provide a countertraction to the lesion as well as surgical mucosectomy and it can be carried out without resect anesthesia. Therefore, this procedure can easily resect large lesions over 3 cm in size and lesions with ulcer scars [29] [30] .

Which method is best for which lesion? What are the technical limits for complete resection of cancers using the above methods? Until recently, there have been no comparative studies and standardization of EMR with regard to lesion characteristics (size, location, etc.). Also, what methods are available for EMR of early cancer in the gastrointestinal tract? As mentioned above, various EMR methods are applicable to all early cancers in the gastrointestinal tract. However, the most important points are the endoscopist’s experience and the advantages and disadvantages of the various EMR methods. Definite conclusions regarding complications can not be drawn from comparison of the different techniques used in the reports reviewed [37].

The limitations of EMR concern diagnosis before EMR, the location of lesions (areas posing technical difficulties with endoscopy), technical problems, indications for EMR (size, undifferentiated intramural carcinoma), and the problem of reliable histopathological interpretation of the surgical specimen [37]. Kodama & Kakegawa reviewed the histopathological characteristics and optimal treatment modalities for superficial esophageal cancer [39]. The survival rates of patients who had m1 or m2 disease and who had been treated using EMR were no different from those of patients who had been treated by surgery. The 5-year survival rate of patients who had m3 cancer was approximately 88 %. Kojima et al. reviewed the reported outcomes of EMR performed in early gastric cancer, using a Medline search up to 1998. They found a total of 1832 cases reported from 12 major Japanese institutions [6]. The most commonly used EMR technique was inject and cut (1515 cases), followed by EMRL and EMRC. En bloc resection was carried out in 75 % of cases, and complete resection was achieved in 74 % of cases. The highest complete resection rate (98 %) was obtained when EMRC was used. The present author has evaluated the efficacy of EMR in a total of 67 patients. Complete resection was achieved in 88 % (59/67). The results of EMR in large colorectal polyps and early cancers also have been favorable. Kudo reported no local recurrence or metastases after performing EMR in more than 6000 patients with mucosal and early submucosal colorectal cancer [40]. Binmoeller et al. reported that only 19 (16 %) of 117 patients who had adenomatous polyps more than 3 cm in size required further endoscopic treatment for recurrent adenoma during a follow-up of at least 6 months [41].

In conclusion, EMR is considered to be a useful procedure, contributing towards not only correct diagnosis of lesions that cannot be diagnosed by conventional forceps biopsy, but also to curative treatment of flat or depressed malignant mucosal lesions in the gastrointestinal tract. Recent advances in EMR techniques are truly remarkable. These developments allow minimally invasive treatment of diseases that would otherwise require major surgery. The new techniques will continue to overcome the limitations of EMR and its use in our practice will also continue to expand.

Table 1Classification of techniques of endoscopic mucosal resection (EMR) Techniques without suction Injection and cut (strip biopsy) Dehle et al. 1973 12, Tada et al. 1993 13 Endoscopic resection with local injection hypertonic saline and epinephrine of (ERHSE, injection, precut and cut method) Hirao et al. 1988 14 Injection, lifting, and cutting Martin et al. 1976 15, Takekoshi et al. 1994 16 Endoscopic double-snare polypectomy witha double-channel endoscope (EDSP) Takekoshi et al. 1986 17 Injection with hyaluronic acid, precutting and cutting Yamamoto et al. 1999 18 Lift and cut method using an overtube (endoscopic mucosal resection with tube, (EMRT)) Inoue & Endo 1990 19 EMR with four-point fixation Inatsuchi & Tanaka 1996 20) EMR using snare with injection needle in single catheter Shim et al. 1999 21 Techniques with suction Esophageal endoscopic mucosal resection (EEMR) or Makuuchi tube method Makuuchi 1996 22 Negative pressure endoscopic esophageal mucosectomy (NP-EMR) or modified Makuuchi tube method Kawano et al. 1991 23 Endoscopic mucosal resection with cap (EMRC) Inoue et al. 1990 24 Endoscopic mucosal resection with ligation (EMRL) Shim 1999 8, Akiyama et al. 1997 25, Chaves et al. 1996 26, Fleischer et al. 1996 27 Simple suction Soehendra et al. 1997 28 Modified techniques Percutaneous transgastric wall endoscopic mucosal resection (PT-EMR) Murai et al. 1995 29 Direct punture endoscopic mucosal resection (DP-EMR) Chibai et al. 1997 30 Percutaneous traction-assisted EMR (PTA-EMR) Ono et al. 2000 31 Endoscopic mucosal resection with three-channel outer tube and multiple forceps Mukai et al. 2001 2 Endoscopic mucosal resection with insulated-tip diathermic knife (IT-EMR) Ohkuwa et al. 2001 1 Modified from Inoue H, Tani M, Nagai K, et al. Treatment of esophageal and gastric tumors. Endoscopy 1999; 31: 47, with permission.

References

• 1 Ohkuwa M, Hosokawa K, Boku N, et al. New endoscopic treatment for intramucosal gastric tumors using an insulated-tip diathermic knife.  Endoscopy. 2001;  32 221-226
• 2 Mukai M, Iwase H, Himeno R, et al. A novel approach to endoscopic colorectal mucosal resection using a three-channel outer tube and multiple forceps: an experimental assessment.  Endoscopy. 2001;  32 253-256
• 3 Soelikno R M, Inoue H, Chang K J. Endoscopic mucosal resection.  Gastrointest Endosc Clin N Am. 2000;  10 595-617
• 4 Amano Y, Ishihara S, Amano K, et al. An assesment of local curability of endoscopic surgery in early gastric cancer without satisfaction of current therapeutic indications.  Endoscopy. 1998;  30 548-552
• 5 Yoshida S. Endoscopic diagnosis and treatment of early cancer in the alimentary tract.  Digestion. 1998;  59 502-508
• 6 Kojima T, Parra-Blanco A, Takahashi H, et al. Outcome of endoscopic mucosal resection for early gastric cancer: review of the Japanese literature.  Gastrointest Endosc. 1998;  48 550-554
• 7 Inoue H. Endoscopic mucosal resection for entire gastrointestinal mucosal lesions.  Clin Perspect Gastroenterol. 2000;  July/August 177-185
• 8 Shim C S. Endoscopic mucosal resection. International Gastric Cancer Congress. 1999 Seoul, Korea; April 27-30 1999: 49
• 9 Lambert R. Endoscopic treatment of esophagogastric tumors.  Endoscopy. 1998;  30 80-93
• 10 Fleischer D. Endoscopic mucosal: (not) made in the USA (so commonly). A dissection of the definition, technique, use, and controversies.  Gastrointest Endosc. 2000;  52 440-444
• 11 Nakajima T. Tabular analysis of 10 000 cases of gastric cancer in CIH.  Jpn J Cancer Chemother. 1994;  21 1813-1818
• 12 Dehle P, Largiader F, Jenny S, et al. A method for endoscopic electroresection of sessile colonic polyps.  Endoscopy. 1973;  5 38-40
• 13 Tada M, Murakami A, Karita M, et al. Endoscopic resection of early gastric cancer.  Endoscopy. 1993;  25 445-450
• 14 Hirao M, Masuda K, Asanuma T, et al. Endoscopic resection of early gastric cancer and other tumors with local injection of hypertonic saline-epinephrine.  Gastrointest Endosc. 1988;  34 264-269
• 15 Martin T R, Onstad G R, Silvis S E, et al. Lift and cut biopsy technique for submucosal sampling.  Gastrointest Endosc. 1976;  23 29-30
• 16 Takekoshi K, Baba Y, Ota H, et al. Endoscopic resection of early gastric carcinoma: Results of a retrospective analysis of 308 cases.  Endoscopy. 1994;  26 352-358
• 17 Takekoshi T, Takagi K, Fujii A, et al. The treatment of early gastric cancer by endoscopic double snare polypectomy (EDSP).  Gan No Rinsho Jpn J Cancer Clin. 1986;  32 1185-1190
• 18 Yamamoto T, Yube T, Isoda N, et al. A novel method of endoscopic mucosal resection using sodium hyaluronate.  Gastrointest Endosc. 1999;  50 251
• 19 Inoue H, Endo M. Endoscopic esophageal mucosal resection using a transparent tube.  Surg Endosc. 1990;  4 198-201
• 20 Inatsuchi S, Tanaka M. Safer and more reliable endoscopic mucosal resection by the four-point fixation method in the treatment of early gastric resection.  Endosc Digest. 1996;  8 499-507
• 21 Shim C S, et al. Trial study on endoscopic mucosal resection using a newly designed snare with injected needle catheter: preliminary report. 1999 Gastrointestinal Endoscopy Congress; Seoul, Korea. 
• 22 Makuuchi H. Endoscopic mucosal resection for early esophageal cancer.  Dig Endosc. 1996;  8 175-179
• 23 Kawano T, Miyake S, Yasuno M, et al. A new technique for endoscopic esophageal mucosectomy using a transparent overtube with intraluminal negative pressure (NP-EEM).  Dig Endosc. 1991;  3 159-167
• 24 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
• 25 Akiyama M, Ota M, Nakajima H, et al. Endoscopic mucosal resection of gastric neoplasms using a ligating device.  Gastrointest Endosc. 1997;  45 182-186
• 26 Chaves D M, Sakai P, Mester M, et al. A new endoscopic technique for the resection of flat polypoid lesions.  Gastrointest Endosc. 1996;  40 224-226
• 27 Fleischer D E, Dawsey S, Tio T L, et al. Tissue band ligation followed by snare resection (band and snare): a new technique for tissue acquisition in the esophagus.  Gastrointest Endosc. 1996;  44 68-72
• 28 Soehendra N, Binmoeller K F, Bohnacker S, et al. Endoscopic snare mucosectomy in the esophagus without any additional equipment: a simple technique for resection of flat early cancer.  Endoscopy. 1997;  29 380-383
• 29 Murai R, Mitsumori Y, Ando H, et al. Percutaneous transgastric wall endoscopic mucosal resection (PTEMR) for early gastric cancer.  Operation. 1995;  49 1343-1348
• 30 Chibai M, Fujisaki J. Direct puncture endoscopic mucosal resection (DP-EMR) for early gastric cancer. In: Makuuchi H, editor. Endoscopic mucosal resection for early esophageal and gastric cancer: a challenge to its limitation. Japan Medical Center Tokyo; 1997: 171-177
• 31 Ono H, Gotoda T, Kondo H. The limitation and application of IT knife on difficult cases of strip biopsy: PTA-EMR.  Gastroenterological Endoscopy (Japanese abstract). 2000;  42 1544
• 32 Inoue H, Tani M, Nagai K, et al. Treatment of esophageal and gastric tumors.  Endoscopy. 1999;  31 47-55
• 33 Hosokawa K, Yoshida S. Recent advances in endoscopic mucosal resection for early gastric cancer [in Japanese with English abstract].  Jpn J Cancer Chemother. 1998;  25 483
• 34 Takechi K, Mihara M, Saito Y, et al. A modified technique for endoscopic mucosal resection of small early gastric carcinomas.  Endoscopy. 1992;  24 215-217
• 35 Shim C S. Endoscopic mucosal resection.  J Korean Med Sci. 1996;  11 456-466
• 36 Kida M, Tanabe S, Watanabe M, et al. Staging of gastric cancer with endoscopic ultrasonography and endoscopic mucosal resection.  Endoscopy. 1998;  30 Suppl 1 A64
• 37 Makuuchi H, Kise Y, Shimada H, et al. Endoscopic mucosal resection for early gastric cancer.  Semin Surg Oncol. 1999;  17 (2) 108-116
• 38 Al-Kawas F H, Gostout C J, Ginsberg G G, et al. Endoscopic mucosal resection/ablation.  ASGE Clinical Symposium 1998
• 39 Kodama M, Kakegawa T. Treatment of superficial cancer of the esophagus: a summary of response to a questionnaire on superficial cancer of the esophagus in Japan.  Surgery. 1998;  123 432-439
• 40 Kudo S. Endoscopic mucosal resection of flat and depressed types of early colorectal cancer.  Endoscopy. 1993;  25 455
• 41 Binmoeller K F, Bohnacker S, Seifert H, et al. Endoscopic snare excision of “giant” colorectal polyps.  Gastrointest Endosc. 1996;  43 183-188