Endoscopy 2010; 42(9): 714-722
DOI: 10.1055/s-0030-1255654
Original article

© Georg Thieme Verlag KG Stuttgart · New York

Retrospective study of technical aspects and complications of endoscopic submucosal dissection for laterally spreading tumors of the colorectum

T.  Toyonaga1 , M.  Man-i1 , T.  Fujita2 , J.  E.  East3 , E.  Nishino4 , W.  Ono5 , Y.  Morita2 , T.  Sanuki2 , M.  Yoshida2 , H.  Kutsumi2 , H.  Inokuchi6 , T.  Azuma2
  • 1Department of Endoscopy, Kobe University Hospital, Kobe, Japan
  • 2Department of Gastroenterology, Kobe University, Kobe, Japan
  • 3Department of Gastroenterology, John Radcliffe Hospital, Oxford, UK
  • 4Department of Pathology, Kishiwada Tokushukai Hospital, Osaka, Japan
  • 5Department of Gastroenterology, Kishiwada Tokushukai Hospital, Osaka, Japan
  • 6Department of Gastroenterology, Hyogo Cancer Center, Kobe, Japan
Further Information

Publication History

submitted 9 September 2009

accepted after revision 14 June 2010

Publication Date:
30 August 2010 (online)

Preview

Background and study aims: Laterally spreading tumors – non granular type (LST-NG) are more often considered candidates for endoscopic submucosal dissection (ESD) than laterally spreading tumors – granular type (LST-G), because of their higher potential for submucosal invasion. However, ESD for LST-NG can be technically difficult. The aim of our study was to compare our ESD results for LST-NG and for LST-G.

Patients and methods: Ninety-nine LST-NG and 169 LST-G measuring 20 mm in size or more were removed by ESD. We retrospectively evaluated the clinicopathological features of the tumors and treatment results (en bloc resection rate, procedure time and speed, rate of use of ancillary devices, and complication and recurrence rates).

Results: Histopathology revealed that there were more submucosally invasive lesions in the LST-NG than in the LST-G group (28 % vs. 9 %; P < 0.0001). The en bloc resection rate, en bloc R0 resection rate, and en bloc curative resection rate of LST-NG were similar to those of LST-G (LST-NG: 99 %, 98 %, and 88 %; LST-G: 99 %, 98 %, and 91 %). In LST-NG, the median procedure time tended to be longer (LST-NG: 69 min; LST-G: 60 min) and the median procedure speed was slower (LST-NG: 0.15 cm2/min; LST-G: 0.25 cm2/min; P < 0.0001). Use of ancillary devices was higher for LST-NG (38 % vs. 15 % for LST-G; P < 0.0001), as was the perforation rate (5.1 % vs. 0.6 % for LST-G; P = 0.027). No recurrence was seen in either group.

Conclusions: ESD was an effective treatment method for both LST-NG and LST-G. However, the degree of technical difficulty appears higher for LST-NG than for LST-G lesions, as shown by the lower dissection speed and higher perforation rate. ESD for LST-NG should probably be performed by those with significant experience of colorectal ESD.

References

T. ToyonagaMD 

Department of Endoscopy
Kobe University Hospital

7-5-1 Kusunoki-cho
Chuo-ku
Kobe
6500-0017 Japan

Fax: +81-78-3826309

Email: toyonaga@med.kobe-u.ac.jp