- Top of page
- CONFLICT OF INTEREST
Aim: The primary purpose of this questionnaire survey study was to determine the current status of colorectal endoscopic submucosal dissection (ESD) in specialized Japanese referral centers before and after introduction of a government-approved advanced medical treatment system; and, secondly, to determine the current status of colorectal ESD in other Asian specialized referral centers.
Methods: A total of 1321 colorectal ESDs were performed in 11 institutions including two Asian centers outside Japan.
Results: Overall en-bloc resection, curative resection, R0 resection, perforation, delayed bleeding and emergency surgery rates were 95.4%, 89.1%, 87.2%, 2.9%, 2.5% and 0.2%, respectively. Similar clinical results were reported in the two Asian centers.
Conclusion: There were no significant differences with regards to clinical results between the two periods although the perforation rate decreased from 3.3% to 2.4%. In addition, colorectal ESD has become increasingly standardized technically at specialized referral centers not only in Japan, but several other Asian referral institutions as well.
- Top of page
- CONFLICT OF INTEREST
The endoscopic submucosal dissection (ESD) procedure that was initially developed for early gastric cancer facilitates the en-bloc resection of large superficial tumors.1–4 ESD is not widely performed for treating colorectal cancer (CRC) despite its minimal invasiveness because of increased technical difficulty and the higher risk of perforation and resultant peritonitis.4
Due to the widespread acceptance of gastric ESD, however, the number of medical facilities that perform colorectal ESD has been increasing recently and the effectiveness of colorectal ESD has been reported both in Japan5–12 and some Western countries13,14 although most such reports have been limited to a single center, single operator and small patient sample studies.
More recently, colorectal ESDs have only been performed in Japan in accordance with advanced medical treatment system no. 78 approved by the Japanese Ministry of Health, Labor and Welfare in 2009 which distinguished colorectal ESD from gastric and esophageal ESD due to its associated higher complication rates. However, with greater experience and lower complication rates, from April 2012 colorectal ESD will now be permitted under the Japanese Medical Insurance. The indications for colorectal ESD under this system are defined as follows: (i) early CRC of more than 20 mm difficult to treat en-bloc by endoscopic mucosal resection (EMR); and (ii) adenomas with non-lifting sign or residual tumors of more than 10 mm difficult to treat by EMR.
All candidate lesions for ESD should be confirmed as being an intramucosal tumor using magnification colonoscopy15–17 or endoscopic ultrasonography (EUS) before performing the procedure. Only endoscopists certified by the Japanese Gastrointestinal Endoscopy Society should perform ESD in hospitals in which surgeons are available anytime for emergency surgery if necessary. Currently, over 150 institutions have started performing colorectal ESD in accordance with the advanced medical treatment system utilizing recent improvements in ESD-related instruments and devices as well as various other technical innovations.
In our previously reported prospective multicenter study, multivariate analysis revealed that large tumor size of 50 mm or more and a lower experience level performing less than 50 ESDs were independent factors for a significantly increased risk of complications.18
The primary purpose of the present study, therefore, was to determine the current status of colorectal ESD in specialized Japanese referral centers before and after the introduction of the government-approved advanced medical treatment system; and secondly, to determine the current status of colorectal ESD in other Asian specialized centers besides Japan.
- Top of page
- CONFLICT OF INTEREST
A questionnaire survey requested information on colorectal ESDs performed 1 year before and 1 year after the introduction of the advanced medical treatment system, but several of the nine Japanese referral centers provided data for various periods depending on the actual introduction of such a system at those referral centers. In the two other Asian referral centers, the questionnaire survey was based on the most recent 2 years of total colorectal ESD performed without regard to the introduction of the advanced medical treatment system. All 11 participating institutions responded to the questionnaire survey.
ESD was defined as the circumferential marginal resection after submucosal injection of a colorectal lesion using a needle knife for the purpose of performing an en-bloc resection with a knife specifically designed for ESD (Fig. 1).
Figure 1. (a) A 7 cm 0-Is lesion located in the sigmoid colon. (b) Lesion after indigo-carmine dye spraying. (c) Narrow-band imaging with magnification revealed type II or IIIA capillary pattern. (d) Indigo-carmine dye with magnification demonstrated type IIIL and IV pit pattern. (e) Endoscopic submucosal dissection (ESD) performed using short-type ST Hood. (f) During ESD, small muscle defect identified, but external longitudinal muscle layer also observed in this defect. Difficult to evaluate, but this case categorized as perforation. (g) Ulcer bed after en-bloc resection. (h) Four endo-clips were applied to close muscle defect after en-bloc resection. (i) En-bloc resection specimen measured 75 mm × 60 mm. No pain, inflammatory change or fever were observed in this patient. Histology revealed submucosal deep invasion so surgery performed although no residual tumor or lymph node metastasis was detected in this patient.
Download figure to PowerPoint
The main items covered in the survey included total number of ESD patients and lesions; male/female ratio; mean tumor size; macroscopic type; histology; procedure time; en-bloc resection rate; curative resection rate; R0 resection rate; complication rates involving perforations and delayed bleeding; and incidence of cases requiring emergency surgery. Macroscopic type focused on the ratio of non-granular type laterally spreading tumors (LST-NGs).
Definitions of en-bloc, curative and R0 resections
We defined an en-bloc resection as the one-piece resection of an entire lesion as observed endoscopically.9 A curative resection was achieved when both the lateral and vertical margins of the specimen/specimens were free of cancer and there was no submucosal invasion of 1000 µm or more (SM1), lymphatic invasion, vascular involvement or poorly differentiated component.19 An adenoma with an unknown lateral margin was also considered to be curatively resected provided that the specimen/specimens met all the other indicated criteria. An R0 resection was defined as being achieved when both lateral and vertical margins of the specimen were free of tumor irrespective of other histological results. Histological diagnoses were based on the Japanese classification of cancer of the colon and rectum20 and the Vienna classification.21
Definition of complications
A perforation during an ESD procedure was defined as an ‘immediate perforation’ and a ‘delayed perforation’ was defined as having occurred after completion of the ESD procedure. Postoperative bleeding was defined as clinical evidence of bleeding manifested by melena or hematochezia 0–14 days after the procedure that required endoscopic hemostasis.
- Top of page
- CONFLICT OF INTEREST
A total of 1321 colorectal ESD were performed in the 11 participating institutions including two Asian institutions outside Japan. In the recently reported data, overall rates of en-bloc resection, curative resection and R0 resection were 95.4%, 89.1% and 87.2%, respectively. With regard to complications associated with colorectal ESD, i.e. rates of perforation, delayed bleeding and cases requiring emergency surgery were 2.9%, 2.5% and 0.2%, respectively. (Table 1).
Table 1. Summary of endoscopic submucosal dissection (ESD) results from participating centers
|No. of lesions||1321||706||615|
|No. of patients||1281||675||606|
|% of non-granular laterally spreading tumors||38.4%||38.3%||38.5%|
|% of cancer||63.2%||63.7%||62.6%|
|En-bloc resection rate (%)||1260 (95.4%)||657 (94.5%)||603 (96.3%)|
|Curative resection rate† (%)||1177 (89.1%)||616 (88.6%)||561 (89.6%)|
|R0 resection rate (%)||1149 (87.2%)||603 (87.3%)||546 (87.2%)|
|Complications (%)||59 (4.5%)||40 (5.7%)||19 (3.1%)|
|Perforations (%)||38 (2.9%)||23 (3.3%)||15 (2.4%)|
|Postoperative bleeding (%)||33 (2.5%)||18 (2.5%)||15 (2.4%)|
|Emergent surgery cases (%)||3 (0.2%)||2 (0.3%)||1 (0.2%)|
In total, the number of colorectal ESD cases that were included pre- and post-authorization (in accordance with the advanced medical treatment system) were 706 and 615, respectively. There were no significant differences between the pre-authorization and post-authorization periods with regards to clinical results although the en-bloc resection and curative resection rates increased slightly and the perforation rate decreased from 3.3% to 2.4%. Emergency surgery was extremely rare during both periods.
Similarly, there were no significant differences in tumor size, macroscopic type and histology between the two periods.
Tables 2 and 3 indicate the recent clinical results for each participating institution. As indicated, mean tumor size ranged from 25.4–43.0 mm; the ratio of LST-NGs ranged from 23.8–61.0% in the 11 referral centers; and the ratio of cancers varied from 14.3–100%. The en-bloc resection rate ranged from 92.2–100% and the curative resection rate varied from 78.4–100%. As for complications, the perforation rate ranged from 0–8.0%, but the incidence of emergency surgery cases was extremely low at just 0% in Japan and 1.3% in the two Asian institutions.
Table 2. Recent case data from Japanese participating centers
|Basic information||Study periods||2009/10–2011/4||2009/4–2011/3||2010/7–2011/5||2010/8–2011/6||2010/10–2011/4||2010/9–2011/4||2010/10–2011/6||2010/5–2011/5||2010/4–2011/3|| |
|No. of lesions||155||51||21||26||67||111||41||91||103||666|
|No. of cases||152||49||21||26||66||109||41||89||102||655|
|Mean age ± SD, years||66.4 ± 10.3||68.7 ± 8.79||69.3 ± 7.4||65.3 ± 11.5||69.0 ± 9.9||70.0 ± 8.7||67.2 ± 8.4||67.2 ± 12.4||67.0 ± 10.0|| |
|Mean tumor size ± SD, mm||34.3 ± 15.1||35.3 ± 8.05||36.1 ± 22.4||32.8 ± 24.0||30.2 ± 13.8||32.6 ± 16.4||34.2 ± 7.3||37.34 ± 18.31||35.0 ± 14.0|| |
|Mean specimen size ± SD, mm||41.1 ± 13.6||39.1 ± 8.11||38.9 ± 20.6||40.6 ± 24.1||35.9 ± 14.1||42.7 ± 18.1||42.3 ± 8.2||43.89 ± 18.43||35.0 ± 13.0|| |
|Tumor location||Lower rectum (Rb)||26||10||3||5||13||10||2||23||24||116|
|Upper rectum (Ra)||10||9||4||4||8||5||3||19||3||65|
|Macroscopic type||Is+IIa (LST-G); Nodular mixed type||56||17||5||0||34||15||16||47||32||222|
|IIa (LST-G); homogeneous type||36||17||7||4||1||28||0||2||8||103|
|LST-NG; flat elevated||28||12||5||6||14||38||16||24||17||160|
|LST-NG; pseudo-depressed type||32||3||0||1||11||14||9||2||25||97|
|IIa+IIc, IIc, Is+IIc; depressed type||1||1||0||1||4||2||0||0||3||12|
|Is, Isp; elevated type||12||1||2||4||1||11||0||13||7||51|
|Submucosal (SM) tumor||1||0||0||0||0||0||0||3||1||5|
|SM1 cancer (<1000 µm)||19||2||0||1||3||13||3||10||19||70|
|SM2 or deeper cancer (≥1000 µm)||10||3||0||1||6||8||1||11||11||51|
|Clinical results||No. of en-bloc resections (%)||156/166 (94%)||47/51 (92%)||21/21 (100%)||25/26 (96%)||63/67 (94%)||111/111 (100%)||41/41 (100%)||88/91 (97%)||98/103 (95%)||603/626 (96%)|
|No. of curative resections† (%)||152/166 (92%)||40/51 (78%)||21/21 (100%)||25/26 (96%)||56/67 (84%)||103/111 (93%)||40/41 (98%)||75/91 (82%)||89/103 (86%)||561/626 (90%)|
|No. of R0 resections (%)||133/166 (80%)||47/51 (92%)||21/21 (100%)||24/26 (92%)||60/67 (90%)||103/111 (93%)||41/41 (100%)||80/91 (88%)||84/103 (82%)||546/626 (87%)|
|Mean procedure time ± SD, min||95.0 ± 73.4||134.5 ± 75.9||86.7 ± 59.5||–||55.8 ± 37.5||57.9 ± 32.5||45.5 ± 13.0||93.8 ± 63.7||95.0 ± 53.0|| |
|No. of complications (%)||0 (0%)||1 (1.9%)||0 (0%)||1 (3.8%)||2 (3.0%)||3 (2.7%)||0 (0%)||11 (12.1%)||2 (1.9%)||20 (3.0%)|
|No. of perforations (%)||5 (3%)||1 (1.9%)||0 (0%)||1 (3.8%)||1 (1.5%)||3 (2.7%)||0 (0%)||3 (3.3%)||2 (1.9%)||16 (2.4%)|
|No. of postoperative bleeding cases (%)||6 (3.6%)||0 (0%)||0 (0%)||0 (0%)||1 (1.5%)||0 (0%)||0 (0%)||8 (8.8%)||0 (0%)||15 (2.3%)|
|No. of emergent surgery cases (%)||0 (0%)||0 (0%)||0 (0%)||0 (0%)||0 (0%)||0 (0%)||0 (0%)||1 (1.1%)||0 (0%)||1 (0.15%)|
|No. of delayed perforations (%)||0 (0%)||0 (0%)||0 (0%)||0 (0%)||0 (0%)||0 (0%)||0 (0%)||0 (0%)||0 (0%)||0 (0%)|
Table 3. Recent case data from Asian participating centers
|Basic information||Study periods||2009/7–2011/6||2009/1–2010/12|| |
|No. of lesions||153||250||403|
|No. of cases||148||250||398|
|Mean age ± SD||60.2 ± 9.9||50.2 ± 40.1||0|
|Mean tumor size ± SD||25.4 ± 12.6||43.0 ± 12.0|| |
|Mean specimen size ± SD||30.2 ± 12.7|| || |
|Range, mm||12–100|| || |
|Tumor location||Lower rectum (Rb)||1||76||77|
|Upper rectum (Ra)||43||38||81|
|Macroscopic type||Is+IIa (LST-G); nodular mixed type||36||24||60|
|IIa (LST-G); homogeneous type||38||43||81|
|LST-NG; flat elevated||33||52||85|
|LST-NG; pseudo-depressed type||20||26||46|
|IIa+IIc, IIc, Is+IIc; depressed type||1||9||10|
|Is, Isp; elevated type||20||13||33|
|Submucosal (SM) tumor||5||46||51|
|SM1 cancer (<1000 µm)||20||31||51|
|SM2 or deeper cancer (≥1000 µm)||10||16||26|
|Clinical results||No. of en-bloc resections (%)||144/153 (94%)||216/231 (94%)||360/384 (94%)|
|No. of curative resections† (%)||131/153 (86%)||203/243 (84%)||334/396 (84%)|
|No. of R0 resections (%)||137/153 (89%)||186/250 (74%)||323/403 (80%)|
|Mean procedure time ± SD||50.7 ± 38.8||75.3 ± 40.8|| |
|No. of complications (%)||4 (2.6%)||48 (19.2%)||52 (12.9%)|
|No. of perforations (%)||4 (2.6%)||20 (8.0%)||24 (6.0%)|
|No. of postoperative bleeding cases (%)||0 (0%)||16 (6.4%)||16 (4.0%)|
|No. of emergency surgery cases (%)||0 (0%)||7 (2.8%)||7 (1.8%)|
|No. of delayed perforation cases (%)||0.0%||2.0%||1.3%|
- Top of page
- CONFLICT OF INTEREST
Over 150 Japanese institutions have currently been authorized to perform colorectal ESD. From a technical perspective, colorectal ESD has become increasingly standardized at specialized referral centers not only in Japan, but also in other Asian countries including South Korea and China. Although the perforation rate varies quite considerably, the incidence of cases requiring emergency surgery has been extremely low. Even ESDs of large colorectal tumors of more than 2 cm in diameter were safely and successfully performed with an overall 95.4% en-bloc resection rate and an 89.1% curative resection rate with only a 2.9% perforation rate.
In this questionnaire survey study, two Asian institutions outside Japan also participated and, quite surprisingly, 250 colorectal ESDs were performed in the Chinese institution during a recent 1 year period. The perforation rate in this particular institution was the highest in our study at 8.0%, but the mean tumor size of 43.0 mm also was the largest and the colorectal ESD learning curve should also be considered because endoscopists have only started performing ESD in China several years ago.
Many articles have been published so far on colorectal ESD primarily in Japan. We also have previously published several reports including one that clearly demonstrated large tumor size of 50 mm or more was an independent risk factor for complications while a large number of ESD performed at an institution decreased the risk of complications.18
Longer procedure time has been one of the major problems for colorectal ESD compared with conventional EMR and the mean procedure time in this study varied considerably from 45.5–134.5 min. The indications for colorectal ESD are now defined by the advanced medical treatment system as early colorectal cancers of more than 20 mm that are difficult to treat en-bloc by EMR, or adenomas with non-lifting sign or residual tumors of more than 10 mm that are difficult to treat by EMR. The current indications for EMR and ESD are different so the longer procedure time associated with ESD may be considered acceptable now.
Mean tumor size in this study ranged from 25.4–43.0 mm and the ratio of LST-NGs and cancers ranged from 23.8–61.0% and 14.3–100%, respectively. Although the indications for ESD differed from EMR, such ESD indications are still somewhat confusing among institutions particularly in Asian countries outside Japan even after taking into account differences in the histological diagnostic criteria for intramucosal cancer among the various institutions. In addition, the indications for ESD were similar both before and after introduction of the advanced medical treatment system in Japan so all the participating Japanese institutions have been performing colorectal ESD until now, based on indications similar to those of the advanced medical treatment system.
Certainly, one of the primary reported issues concerning colorectal ESD has been the fact that colorectal ESD is only widely accepted and practiced in Japan. It was thought that widespread use of colorectal ESD outside of Japan would be very difficult to achieve, but the two participating Asian institutions reported similarly positive clinical results even in comparison to the nine Japanese specialized referral centers involved in this questionnaire survey study. This could very well have been because two endoscopists at those two institutions previously visited a high-volume ESD center in Japan and observed many colorectal ESD procedures before beginning to perform ESD themselves.
The other reason for the positive results may have been the fact that most of the instruments and devices used in performing ESD such as the insulation-tipped knife (KD-611L; Olympus Medical Systems, Tokyo, Japan), dual knife (KD-650L; Olympus), CO2 insufflation and water jet scope are available at the present time not only in Japan, but in other Asian countries as well. The reality is that a limited number of specialized referral centers in Asia are currently performing a large number of advanced endoscopic treatment procedures. In time, this should lead to improved clinical results even for more difficult procedures like colorectal ESD.
Certain colorectal tumors such as LST-NG lesions of more than 2 cm and granular type LST lesions of more than 3 cm in certain cases in which EMR is not considered feasible should be resected using ESD. We must continue in our efforts, therefore, to promote the widespread acceptance and proper use of colorectal ESD worldwide.
In conclusion, the indications for ESD were similar both before and after introduction of the advanced medical treatment system in Japan. There were no significant differences between the two periods with regards to clinical results although there was a trend towards a decrease in the perforation rate. In addition, colorectal ESD is now regarded as being increasingly standardized technically at specialized referral centers not only in Japan, but also in several other Asian referral centers notably South Korea and China.