Digestive Endoscopy 2014; ••: ••–••
doi: 10.1111/den.12232
Original Article
Endoscopic resection yields reliable outcomes for small rectal neuroendocrine tumors Joon Han Jeon,1 Dae Young Cheung,1 Seong Jin Lee,1 Hyun Jin Kim,1 Hye Kang Kim,1 Hyung Jun Cho,1 In Kyu Lee,2 Jin Il Kim,1 Soo-Heon Park1 and Jae Kwang Kim1 Departments of 1Internal Medicine and 2Surgery, the Catholic University of Korea College of Medicine, Seoul, Korea Background and Aim: We analyzed the characteristics of gastrointestinal neuroendocrine tumors and examined the outcomes and safety of modalities in rectal neuroendocrine tumors.
group (18.5%) (P = 0.003). No local tumor recurrence was observed in all patients, regardless of the procedure, during the median follow-up period of 21.5 ± 13.5 months.
Methods: Between 2007 and 2011, a total of 91 patients with gastrointestinal neuroendocrine tumors were retrospectively reviewed in terms of the characteristics of tumors.
Conclusions: ESD achieved a higher complete resection rate than EMR and comparable to TEM. Tumor recurrence was not observed in the endoscopic resection and TEM groups, regardless of the completeness of resection. Small neuroendocrine tumors of the gastrointestinal tract can be managed reliably with both endoscopic resection and TEM.
Results: Sixty-six patients had rectal neuroendocrine tumors and underwent endoscopic mucosal resection (EMR, n = 29), endoscopic submucosal dissection (ESD, n = 23), or transanal endoscopic microsurgery (TEM, n = 14). The complete resection rate was higher in the ESD group (82.7%) and in the TEM group (100%) compared to the EMR group (65.5%) (P < 0.046). The complication rate was higher in the ESD group (47.8%) than in the EMR
Key words: endoscopic mucosal resection (EMR), endoscopic submucosal dissection (ESD), neuroendocrine tumor, transanal endoscopic microsurgery
INTRODUCTION
METHODS
G
Patients and lesions
ASTROINTESTINAL TRACT NEUROENDOCRINE tumors (GI-NET) account for approximately two-thirds of all neuroendocrine tumors.1,2 In Asia, in particular, rectal neuroendocrine tumors have a relatively high incidence, accounting for 20–55% of all GI-NET.1 Various techniques for resection of rectal neuroendocrine tumors have been reported and include endoscopic mucosal resection (EMR), endoscopic submucosal dissection (ESD), and transanal endoscopic microsurgery (TEM).3–6 To date, the best approach is still being debated. ESD has shown efficacy for complete resection of early gastric cancers, especially in Japan and Korea, and is thought to be the best technique for these lesions.7–9 ESD has also been reported to be efficacious in small rectal neuroendocrine tumors without lymphovascular metastasis.10–12 TEM can also be an effective surgical option for complete removal of neuroendocrine tumors.13
bs_bs_banner
Corresponding: Dae Young Cheung, Department of Internal Medicine, the Catholic University of Korea College of Medicine, Yeongdeungpo-gu, 63-ro 10, Seoul 150-713, Korea. Email: [email protected] Received 3 September 2013; accepted 16 December 2013.
F
ROM FEBRUARY 2007 to December 2011, we enrolled all cases diagnosed with gastrointestinal neuroendocrine tumor at St Mary’s Hospital, Seoul, Korea, which was established 80 years ago and is a well-known tertiary academic hospital. From the electronic medical records database, we identified 91 cases of neuroendocrine tumor arising from the gastrointestinal tract. Nineteen patients had upper GI-NET, one had colon NET and one had appendix NET and all 21 cases were excluded from the present study. Of the remaining 70 cases, four were excluded because they refused further treatment and were lost to follow up. Finally, 66 cases with rectal NET were included in the present analysis. Patients with rectal NET were evaluated microscopically and radiologically to define the differentiation, grade and stage of tumor. Immunohistochemical staining with neuronspecific enolase and synaptophysin was done to support the diagnosis. Mitotic count Ki-67 index was used.14 Abdominal and pelvic computed tomography (CT) scan was done in all patients to evaluate the presence of distant metastasis and lymph node involvement. Fortunately, all 66 patients were
© 2014 The Authors Digestive Endoscopy © 2014 Japan Gastroenterological Endoscopy Society
1
2
J. H. Jeon et al.
proven to have well-differentiated grade 1 tumor without metastasis or lymph node involvement. Endoscopic ultrasonography (EUS) was carried out preoperatively to measure the exact tumor size and to evaluate submucosal space and proper muscle invasion. Twenty nine patients were treated with EMR, 23 patients with ESD and 14 patients with TEM. Clinical information was reviewed retrospectively to extract characteristics of tumors and associated treatment. All NET were confirmed by histological evaluation of endoscopic biopsy specimens or postoperative specimens. The present study was approved by the institutional board of Catholic Medical Center, Seoul, Korea (IRB No. SC12RISI0211).
Endoscopic procedures Indications for each procedure (conventional EMR or EMR with precut, ESD, or TEM) were at the discretion of the attending endoscopist for small rectal NET ≤ 20 mm. Rectal NET > 20 mm were treated by surgery. Endoscopic procedures were carried out by two expert endoscopists (J.I.K and D.Y.C) and five endoscopists in clinical fellowship (J.H.J., S.J.L., H.J.K., H.K.K., H.J.C.). Conventional EMR (C-EMR) or precut EMR (EMR-P) was carried out using a conventional single-channel Olympus endoscope (CF-H260AL; Olympus Optical Co., Tokyo, Japan) and a 25 mm snare (SD-9U-1; Olympus Optical Co.). EMR with precut is defined as a circumferential submucosal incision around the tumor using an endoscopic knife and followed by snaring resection as for the conventional EMR technique.15 ESD was done with a dual knife (KD-650U; Olympus Optical Co.) or IT-II knife (KD611-L; Olympus Optical Co.) A two-channel endoscope (GIF-2TQ260N; Olympus Optical Co.) with a hood and a high-frequency generator (VIO 300D; ERBE, Tübingen, Germany) were used. Marking dots were made around the circumference of the lesion. Submucosal injection was then carried out around the tumor to lift it off the muscle layer. Glycerol with epinephrine (1:1000) was used as the submucosal injection solution. Thereafter, an incision of the mucosa outside the marking dots was done. The tumor was separated from the surrounding normal mucosa. The submucosal connective tissue was gradually dissected from the muscle layer by using either EndoCut I or forced coagulation mode.
Operative procedure for TEM Transanal endoscopic microsurgery was carried out under general or epidural anesthesia, with the patient placed in a suitable position depending on the location of the tumor. The scheduled resection area was circled with marking dots. Full-thickness resection down to the perirectal fat was
Digestive Endoscopy 2014; ••: ••–••
carried out in all procedures. Resection sites were then closed with a continuous running stitch with an absorbable suture. TEM have been carried out by one colorectal surgeon (I.K.L.) since January 2009.
Evaluation of endoscopic findings Endoscopic reports included the location, size, gross appearance and initial impression of the tumors. Gross appearance or impressions of lesions were decided by the attending endoscopists. Complete endoscopic resection of the lesion was considered if the lesion was resected in an en bloc fashion and an endoscopic view showed a clear submucosal layer with no remaining lesion. En bloc resection was defined as resection of the entire lesion in a single piece.
Histopathological evaluation Resected specimens were evaluated histologically using light microscopy to determine histological type, depth of invasion, lateral and vertical resection margin involvement, and lymphovascular invasion. Tumor differentiation was evaluated morphologically under a microscope with hematoxylineosin staining and grade was determined by mitotic count and Ki-67 index. Complete resection was defined histologically when the resected specimen had an intervening normal tissue margin of ≥1.0 mm in width both laterally and vertically. When the safety margin was
doi: 10.1111/den.12232
Original Article
Endoscopic resection yields reliable outcomes for small rectal neuroendocrine tumors Joon Han Jeon,1 Dae Young Cheung,1 Seong Jin Lee,1 Hyun Jin Kim,1 Hye Kang Kim,1 Hyung Jun Cho,1 In Kyu Lee,2 Jin Il Kim,1 Soo-Heon Park1 and Jae Kwang Kim1 Departments of 1Internal Medicine and 2Surgery, the Catholic University of Korea College of Medicine, Seoul, Korea Background and Aim: We analyzed the characteristics of gastrointestinal neuroendocrine tumors and examined the outcomes and safety of modalities in rectal neuroendocrine tumors.
group (18.5%) (P = 0.003). No local tumor recurrence was observed in all patients, regardless of the procedure, during the median follow-up period of 21.5 ± 13.5 months.
Methods: Between 2007 and 2011, a total of 91 patients with gastrointestinal neuroendocrine tumors were retrospectively reviewed in terms of the characteristics of tumors.
Conclusions: ESD achieved a higher complete resection rate than EMR and comparable to TEM. Tumor recurrence was not observed in the endoscopic resection and TEM groups, regardless of the completeness of resection. Small neuroendocrine tumors of the gastrointestinal tract can be managed reliably with both endoscopic resection and TEM.
Results: Sixty-six patients had rectal neuroendocrine tumors and underwent endoscopic mucosal resection (EMR, n = 29), endoscopic submucosal dissection (ESD, n = 23), or transanal endoscopic microsurgery (TEM, n = 14). The complete resection rate was higher in the ESD group (82.7%) and in the TEM group (100%) compared to the EMR group (65.5%) (P < 0.046). The complication rate was higher in the ESD group (47.8%) than in the EMR
Key words: endoscopic mucosal resection (EMR), endoscopic submucosal dissection (ESD), neuroendocrine tumor, transanal endoscopic microsurgery
INTRODUCTION
METHODS
G
Patients and lesions
ASTROINTESTINAL TRACT NEUROENDOCRINE tumors (GI-NET) account for approximately two-thirds of all neuroendocrine tumors.1,2 In Asia, in particular, rectal neuroendocrine tumors have a relatively high incidence, accounting for 20–55% of all GI-NET.1 Various techniques for resection of rectal neuroendocrine tumors have been reported and include endoscopic mucosal resection (EMR), endoscopic submucosal dissection (ESD), and transanal endoscopic microsurgery (TEM).3–6 To date, the best approach is still being debated. ESD has shown efficacy for complete resection of early gastric cancers, especially in Japan and Korea, and is thought to be the best technique for these lesions.7–9 ESD has also been reported to be efficacious in small rectal neuroendocrine tumors without lymphovascular metastasis.10–12 TEM can also be an effective surgical option for complete removal of neuroendocrine tumors.13
bs_bs_banner
Corresponding: Dae Young Cheung, Department of Internal Medicine, the Catholic University of Korea College of Medicine, Yeongdeungpo-gu, 63-ro 10, Seoul 150-713, Korea. Email: [email protected] Received 3 September 2013; accepted 16 December 2013.
F
ROM FEBRUARY 2007 to December 2011, we enrolled all cases diagnosed with gastrointestinal neuroendocrine tumor at St Mary’s Hospital, Seoul, Korea, which was established 80 years ago and is a well-known tertiary academic hospital. From the electronic medical records database, we identified 91 cases of neuroendocrine tumor arising from the gastrointestinal tract. Nineteen patients had upper GI-NET, one had colon NET and one had appendix NET and all 21 cases were excluded from the present study. Of the remaining 70 cases, four were excluded because they refused further treatment and were lost to follow up. Finally, 66 cases with rectal NET were included in the present analysis. Patients with rectal NET were evaluated microscopically and radiologically to define the differentiation, grade and stage of tumor. Immunohistochemical staining with neuronspecific enolase and synaptophysin was done to support the diagnosis. Mitotic count Ki-67 index was used.14 Abdominal and pelvic computed tomography (CT) scan was done in all patients to evaluate the presence of distant metastasis and lymph node involvement. Fortunately, all 66 patients were
© 2014 The Authors Digestive Endoscopy © 2014 Japan Gastroenterological Endoscopy Society
1
2
J. H. Jeon et al.
proven to have well-differentiated grade 1 tumor without metastasis or lymph node involvement. Endoscopic ultrasonography (EUS) was carried out preoperatively to measure the exact tumor size and to evaluate submucosal space and proper muscle invasion. Twenty nine patients were treated with EMR, 23 patients with ESD and 14 patients with TEM. Clinical information was reviewed retrospectively to extract characteristics of tumors and associated treatment. All NET were confirmed by histological evaluation of endoscopic biopsy specimens or postoperative specimens. The present study was approved by the institutional board of Catholic Medical Center, Seoul, Korea (IRB No. SC12RISI0211).
Endoscopic procedures Indications for each procedure (conventional EMR or EMR with precut, ESD, or TEM) were at the discretion of the attending endoscopist for small rectal NET ≤ 20 mm. Rectal NET > 20 mm were treated by surgery. Endoscopic procedures were carried out by two expert endoscopists (J.I.K and D.Y.C) and five endoscopists in clinical fellowship (J.H.J., S.J.L., H.J.K., H.K.K., H.J.C.). Conventional EMR (C-EMR) or precut EMR (EMR-P) was carried out using a conventional single-channel Olympus endoscope (CF-H260AL; Olympus Optical Co., Tokyo, Japan) and a 25 mm snare (SD-9U-1; Olympus Optical Co.). EMR with precut is defined as a circumferential submucosal incision around the tumor using an endoscopic knife and followed by snaring resection as for the conventional EMR technique.15 ESD was done with a dual knife (KD-650U; Olympus Optical Co.) or IT-II knife (KD611-L; Olympus Optical Co.) A two-channel endoscope (GIF-2TQ260N; Olympus Optical Co.) with a hood and a high-frequency generator (VIO 300D; ERBE, Tübingen, Germany) were used. Marking dots were made around the circumference of the lesion. Submucosal injection was then carried out around the tumor to lift it off the muscle layer. Glycerol with epinephrine (1:1000) was used as the submucosal injection solution. Thereafter, an incision of the mucosa outside the marking dots was done. The tumor was separated from the surrounding normal mucosa. The submucosal connective tissue was gradually dissected from the muscle layer by using either EndoCut I or forced coagulation mode.
Operative procedure for TEM Transanal endoscopic microsurgery was carried out under general or epidural anesthesia, with the patient placed in a suitable position depending on the location of the tumor. The scheduled resection area was circled with marking dots. Full-thickness resection down to the perirectal fat was
Digestive Endoscopy 2014; ••: ••–••
carried out in all procedures. Resection sites were then closed with a continuous running stitch with an absorbable suture. TEM have been carried out by one colorectal surgeon (I.K.L.) since January 2009.
Evaluation of endoscopic findings Endoscopic reports included the location, size, gross appearance and initial impression of the tumors. Gross appearance or impressions of lesions were decided by the attending endoscopists. Complete endoscopic resection of the lesion was considered if the lesion was resected in an en bloc fashion and an endoscopic view showed a clear submucosal layer with no remaining lesion. En bloc resection was defined as resection of the entire lesion in a single piece.
Histopathological evaluation Resected specimens were evaluated histologically using light microscopy to determine histological type, depth of invasion, lateral and vertical resection margin involvement, and lymphovascular invasion. Tumor differentiation was evaluated morphologically under a microscope with hematoxylineosin staining and grade was determined by mitotic count and Ki-67 index. Complete resection was defined histologically when the resected specimen had an intervening normal tissue margin of ≥1.0 mm in width both laterally and vertically. When the safety margin was
![[Endoscopic submucosal resection of small neuroendocrine rectal tumors].](/assets/img/hold.png)
