Saline-pocket endoscopic submucosal dissection for superficial colorectal neoplasms: a randomized controlled trial (with video)

ORIGINAL ARTICLE: Clinical Endoscopy

Saline-pocket endoscopic submucosal dissection for superficial colorectal neoplasms: a randomized controlled trial (with video) Hideaki Harada, MD,1 Ryotaro Nakahara, MD,1 Daisuke Murakami, MD,1 Satoshi Suehiro, MD,1 Tetsuro Ujihara, MD,1 Ryota Sagami, MD,1 Yasushi Katsuyama, MD,1 Kenji Hayasaka, MD,1 Yuji Amano, MD, PhD2 Chiba, Japan

GRAPHICAL ABSTRACT

Background and Aims: Colorectal endoscopic submucosal dissection (ESD) is a time-consuming procedure because of the technical difficulty. The newly developed saline-pocket ESD (SP-ESD) provides a clearer view and better traction of the submucosal layer compared with the standard ESD with gas insufflation (S-ESD). This study aimed to prospectively compare the efficacy and safety between S-ESD and SP-ESD in patients with superficial colorectal neoplasms (SCNs). Methods: From April 2017 to November 2018, 95 patients with SCNs 20 mm in diameter were prospectively and randomly enrolled. Four patients were excluded because of an incomplete ESD procedure. Patients were finally allocated to 2 groups, S-ESD with 45 patients and SP-ESD with 46 patients. The primary outcome was dissection speed. Secondary outcomes were ESD procedure time, en bloc and complete resection rates, perforation rate, and adverse effects. Results: Median dissection speed was significantly faster in the SP-ESD than the S-ESD group (20.1 mm2/min [range, 17.3-28.1] vs 16.3 mm2/min [range, 11.4-19.8]; P < .001). Median procedure time was significantly shorter in the SP-ESD than the S-ESD group (29.5 minutes [range, 22.3-44] vs 41 minutes [range, 31-55]; P < .001). The en bloc and complete resection rates were 100% in both groups. No perforations occurred among patients. The volume of saline solution used in the SP-ESD group was significantly greater than that in the S-ESD group (200 mL [range, 120-250] vs 150 mL [range, 100-200]; P Z .016). Conclusions: SP-ESD improved dissection speed and procedure time compared with S-ESD. SP-ESD may be an alternative method for resection of SCNs. (Clinical trial registration number: UMIN 000026317.) (Gastrointest Endosc 2019;90:278-87.)

(footnotes appear on last page of article)

Endoscopic submucosal dissection (ESD) has been rapidly accepted as a common treatment for superficial colorectal neoplasms (SCNs).1 The standard process of ESD is performed in an air view using gas insufflation,

such as carbon dioxide2,3; however, ESD with gas insufflation distends the lumen and creates fumes from diathermy, which sometimes interferes with the endoscopic view and hides the dissection line in the submucosal layer. In

278 GASTROINTESTINAL ENDOSCOPY Volume 90, No. 2 : 2019

www.giejournal.org

Harada et al

contrast, endoscopic resection in water immersion has advantages compared with the gas insufflation method, such as the absorption of fumes, image enhancement in the near field, attenuation of glare, and the floating effect of the submucosa.4-10 The use of water immersion for ESD (underwater ESD) contributes to a clearer endoscopic view and more precise recognition of the dissection line. In general, colorectal ESD has various risks, such as perforation and fever (post-ESD electrocoagulation syndrome),11–14 and additional devices and methods are required for a safer colorectal ESD procedure. Previously, we reported the efficacy and safety of an underwater method with a submucosal pocket creation (saline-pocket [SP]-ESD).15 SP-ESD for gastric neoplasms results in a shorter procedure time and a faster dissection speed than standard (S)-ESD, which is performed with the gas insufflation method. We believe that SP-ESD is more suitable for narrow portions of the GI tract, such as the colorectal lumen, because a submucosal pocket can be easily created under the target lesion because of the tangential endoscopic view. SP-ESD is safely performed, even in the thin colorectal wall, which is considered to be one of the main technical difficulties for colorectal ESD. Therefore, we planned a prospective and randomized clinical trial of SP-ESD for SCNs to clarify the efficacy of this method for colorectal ESD.

Saline-pocket ESD for colorectal neoplasms

monitored. To avoid bias of ESD skills among endoscopists, all procedures were performed by a single endoscopist (H.H.) who has performed >500 colorectal ESDs. All procedures were assisted by 2 endoscopy technicians (R. F. or Y. I.). A single-channel endoscope (PCF-Q260AZI; Olympus, Tokyo, Japan) with an attached transparent tip hood (ST hood, DH-29CR; Fujifilm, Tokyo, Japan) was used in this study. A FlushKnife BT (DK2620J; Fujifilm) was used as an electrosurgical knife for mucosal incisions and submucosal dissection. All procedures were performed using an electrosurgical unit (VIO300D; Erbe, Türbingen, Germany) with carbon dioxide insufflation. Sodium hyaluronate (.4%, MucoUp; Johnson and Johnson, Tokyo, Japan) as a submucosal injectate was used to lift the submucosa off the muscle layer. During the submucosal dissection, hemostatic forceps (Coagrasper, FD-411QR; Olympus) were used to achieve hemostasis for vascular bleeding and precoagulation prophylaxis for visible vessels.

S-ESD procedure

METHODS

The initial injection was administered in the submucosal layer on the distal aspect of the lesion. A distal hemicircumferential incision was made using the electrosurgical knife. Then, submucosal dissection was performed from the distal-to-proximal aspect with carbon dioxide insufflation while making a submucosal flap. If the submucosal injection was insufficient, a local injection was administered to avoid perforation.

Study population and design

SP-ESD procedure

This study was a prospective, randomized, clinical trial conducted in a single center. Patients who met the eligibility criteria were enrolled from April 2017 to November 2018 at the New Tokyo Hospital. The inclusion criteria were as follows: patients with SCNs (adenomas or adenocarcinomas) with laterally spreading tumors (LSTs) 20 mm in diameter with a depth of invasion limited to the mucosa and submucosa by endoscopic estimation. Lesions were classified into the following 2 types according to the Paris classification16: granular LST (LST-G) and nongranular LST (LST-NG) of neoplastic lesions. Exclusion criteria were lesions >50 mm in diameter, macroscopic type lesions with protruded type, polyposis, inflammatory disease, local recurrence after endoscopic resection, patients <20 years of age, and patients with severe general adverse events. Study participants were randomly assigned to the S-ESD group or the SP-ESD group (Fig. 1). The study was approved by the Institutional Review Board of New Tokyo Hospital (Institutional Review Board no. NTH 0111) and was registered in the University Hospital Medical Network Clinical Trials Registry (UMIN 000026317). Written informed consent was obtained from all participants.

While patients were under conscious sedation, they were treated with intravenous diazepam, and vital signs were

The original SP-ESD concept proposed that the ESD procedure was carried out underneath the target lesion via a locally created water pool in the submucosal pocket. The SP-ESD procedure for a case with LST-NG located in the transverse colon is shown in Figure 2A and (Video 1, available online at www.giejournal.org). First, a proximal submucosal injection with a small volume of a sodium hyaluronate mixed with indigo carmine was administered as a landmark of the end of the submucosal dissection in the submucosal pocket. Next, after the distal submucosal injection without indigo carmine, an initial incision was made 10 to 20 mm apart from the distal edge of the lesion (Fig. 2B). Some dissections were performed to create a submucosal pocket. The tip of the attached transparent hood was inserted into the submucosal pocket while injecting saline solution (.9% sodium chloride) using the water jet function, and dissection of the submucosal layer was initiated in the saline pool (Fig. 2C). Dissection was performed continuously toward the proximal aspect in the saline pool until reaching the submucosal marking with indigo carmine (Fig. 2D). Finally, the remaining mucosa and submucosa were cut to remove the lesion (Fig. 2E and F). During the submucosal dissection in the saline pool, carbon dioxide insufflation was switched off. Hemostatic coagulation was routinely performed using hemostatic

www.giejournal.org

Volume 90, No. 2 : 2019 GASTROINTESTINAL ENDOSCOPY 279

Procedures and endoscopic materials

Saline-pocket ESD for colorectal neoplasms

Harada et al

Superficial colorectal neoplasms of LST-G or LST-NG ≥20 mm (n = 100)

Excluded (n = 5) Postponed treatment (n = 2) Declined to participate (n = 3)

Randomization in the study (n = 95)

S-ESD group (n = 48)

SP-ESD group (n = 47)

Failure of ESD (n = 3) Rescue by snaring (n = 2) Severe fibrosis, F3 (n = 1)

Failure of ESD (n = 1) Rescue by snaring (n = 1)

Analyzed (n = 45)

Analyzed (n = 46)

Figure 1. Study flowchart showing patient inclusion and randomization. ESD, Endoscopic submucosal dissection; LST-G, laterally spreading tumor, granular type; LST-NG, laterally spreading tumor, nongranular type; S-ESD, standard endoscopic submucosal dissection; SP-ESD, saline-pocket endoscopic submucosal dissection.

The primary outcome was dissection speed (mm2/min), which was defined as the calculated ratio of the resected

specimen area (mm2) divided by the procedure time (minutes). The resected specimen area was calculated using the ellipse formula area (mm2) Z longest length (mm) / 2  shortest length (mm) / 2  3.14. Secondary outcomes included ESD procedure time, rate of en bloc resection, rate of complete resection, incidence of adverse effects (perforation, delayed bleeding, and fever and abdominal pain), and quantity of saline solution. ESD procedure time was defined as the time from the submucosal injection to the removal of the tumor. En bloc resection was defined as resection in a single piece. Complete resection was defined as tumor resection in a single piece with negative lateral and vertical margins. Adverse effects included delayed perforation, delayed bleeding, and fever or abdominal pain. Delayed perforation was defined as the presence of free air in the peritoneal or retroperitoneal space detected by abdominal radiography or abdominopelvic CT after the completion of colorectal ESD. Delayed bleeding was defined as overt hematochezia or melena arising from the resection site after 6 hours after completion of the colorectal ESD. Abdominal pain was defined as sustained,

280 GASTROINTESTINAL ENDOSCOPY Volume 90, No. 2 : 2019

www.giejournal.org

forceps or an electrosurgical knife under saline and carbon dioxide insufflation. All lesions were resected in an antegrade (not retrograde) direction. The water jet function setting was the same as that used for gastric ESD.15

Randomization Randomization of participants was performed using Excel 2013 (Microsoft, Redmond, Wash) by the assistant coordinator. Patients were randomly assigned to the SESD or SP-ESD group at a 1:1 allocation ratio. Furthermore, randomization was also done with stratification in the tumor locations (right side, left side, and rectum) and macroscopic types (LST-G or LST-NG). The right side of the colon was defined as proximal to the splenic flexure. The left side of the colon was defined as the descending colon to the sigmoid colon. The 2 endoscopy technicians were also randomly assigned to the 2 arms.

Outcome and definitions

Harada et al

Saline-pocket ESD for colorectal neoplasms

Figure 2. Saline-pocket endoscopic submucosal dissection. A, Laterally spreading tumor, nongranular type located in the transverse colon. B, An initial incision was made from the distal aspect of the lesion. C and D, Submucosal dissection was performed in the submucosal pocket while injecting saline solution. E, The remaining mucosa and submucosa were cut to remove the lesion. F, After resection of the tumor.

spontaneous pain of regional rebound tenderness but without a frank perforation. Fever was defined as a body temperature of 37.5 C that developed between 4 hours and 3 days.17 Submucosal fibrosis was classified into the following 3 groups based on fibrosis grade: F0, no fibrosis (appearing as a blue transparent submucosal layer); F1, mild fibrosis (appearing as a white web-like structure in the blue submucosal layer); and F2, severe fibrosis (appearing as a

white structure without a blue transparent submucosal layer).18 The lesions that were graded as F1 and F2 were classified into the fibrosis group in this study. Submucosal fatty tissue was also classified into the following 3 groups based on fatty tissue grade: grade 0, no fatty tissue; grade 1, mild-to-moderate fatty tissue manifesting as a speckle of coagulated tissue after dissection and coagulation; and grade 2, severe fatty tissue manifesting as sheets of coagulated tissue.19

www.giejournal.org

Volume 90, No. 2 : 2019 GASTROINTESTINAL ENDOSCOPY 281

Saline-pocket ESD for colorectal neoplasms

Harada et al

TABLE 1. Univariate analysis of baseline characteristics between the S-ESD and SP-ESD groups by intention-to-treat analysis

Median age, y (IQR) Sex, male

S-ESD group (n [ 48)

SP-ESD group (n [ 47)

P value

71 (61-80)

73 (64.5-76)

.864

28 (58.3)

30 (63.8)

.675

22.7  3.9

22.9  3.8

.765

Hypertension

24 (50.0)

18 (38.3)

.304

Diabetes mellitus

9 (18.8)

5 (10.6)

.386

BMI, kg/m2, mean  SD Comorbidity

Chronic kidney disease

3 (6.3)

1 (2.1)

.617

13 (27.1)

16 (34.0)

.509

Right side

25 (52.1)

25 (53.2)

Left side

17 (35.4)

17 (36.2)

Rectum

6 (12.5)

5 (10.6)

Antithrombotic agents

>.99

Tumor location

>.99

Macroscopic type LST-G

24 (50.0)

23 (48.9)

LST-NG

24 (50.0)

24 (51.1)

25 (21.8-29.3)

25 (21-29)

23 (47.9)

24 (51.1)

Median tumor size, mm (IQR)

.979

Endoscopy technician R. F. Y. I. Technical success

.838

25 (52.1)

23 (48.9)

45 (93.8)

46 (97.9)

.617

Values are n (%) unless otherwise defined. BMI, Body mass index; IQR, interquartile range; LST-G, laterally spreading tumor, granular type; LST-NG, laterally spreading tumor, nongranular type; SD, standard deviation; S-ESD, standard endoscopic submucosal dissection; SP-ESD, saline-pocket endoscopic submucosal dissection.

The quantity of saline solution was measured by the volume of water jet function during ESD in each group. Furthermore, we evaluated the technical success rate with intention-to-treat analysis. Technical success was defined as completion of the ESD procedure without withdrawal of ESD or the need for an additional snaring technique.

Statistical analysis In this study we performed an intention-to-treat analysis for all randomized patients and per-protocol analysis for all randomized patients, except patients in whom the ESD procedures were incomplete. In patients with an incomplete ESD procedure because of severe fibrosis or an additional snaring technique, neither the dissection speed nor the procedure time was precisely calculated. Therefore, baseline characteristics and technical success rate were evaluated using the intention-to-treat analysis. The clinical outcomes were evaluated using perprotocol analysis. Categorical variables are expressed as proportions and were analyzed using the Fisher exact test or c2 test. Continuous variables are expressed as means with standard deviations or medians with ranges and were analyzed using the Student t test or Mann-Whitney U test. A P < .05 282 GASTROINTESTINAL ENDOSCOPY Volume 90, No. 2 : 2019

was considered statistically significant. All data analyses were performed using SPSS (version 24.0; IBM Corp, Armonk, NY).

Sample size calculation Sample size calculation was based on statistical analysis of the dissection speed of previous colorectal ESD procedures performed at our hospital from April 2014 to March 2017. The average dissection speed was 15.1 mm2/min (standard deviation, 10.1). We estimated that the average dissection speed using SP-ESD would be 21.1 mm2/min based on a pilot study and the previous report.15 Therefore, a sample size of 45 patients for each group was needed, considering a power of 80% (2-sided) and an alpha error of 5%. We anticipated a 10% dropout, and thus enrolled 50 patients in each group.

RESULTS Characteristics of the enrolled patients Based on the sample size analysis, 100 patients were enrolled to be randomized to the S-ESD and SP-ESD groups. Five patients were excluded because of postponed treatment for other medical problems (n Z 2) or they www.giejournal.org

Harada et al

Saline-pocket ESD for colorectal neoplasms

TABLE 2. Univariate analysis of clinical outcomes between the S-ESD and SP-ESD groups by per-protocol analysis S-ESD group (n [ 45)

SP-ESD group (n [ 46)

<.001

Pathologic findings Adenoma

35 (77.8)

19 (41.3)

Intramucosal carcinoma

10 (22.2)

23 (50.0)

Carcinoma with submucosal invasion Median specimen size, mm (IQR)

0 (0)

4 (8.7)

34 (29-38)

32.5 (27.3-38)

Fibrosis 16 (35.6)

12 (26.1)

Fibrosis 1

22 (48.9)

26 (56.5)

Fibrosis 2

7 (15.6)

8 (17.4)

Fatty tissue

.651

Grade 0

27 (60.0)

29 (63.0)

Grade 1

13 (28.9)

10 (21.7)

Grade 2

5 (11.1)

7 (15.2)

En bloc resection

45 (100)

46 (100)

Complete resection

45 (100)

46 (100)

Median dissection speed, mm2/min (IQR)

.732 .634

Fibrosis 0

Median procedure time, min (IQR)

P value

41 (31-55)

29.5 (22.3-44)

<.001

16.3 (11.4-19.8)

20.1 (17.3-28.1)

<.001

4 (8.9)

4 (8.7)

>.99

Adverse effects Perforation Delayed bleeding

0

0

3 (6.7)

3 (6.5)

Fever or abdominal pain

1 (2.2)

1 (2.2)

Median water quantity, mL (IQR)

150 (100-200)

200 (120-250)

.016

Values are n (%) unless otherwise defined. IQR, Interquartile range; S-ESD, standard endoscopic submucosal dissection; SP-ESD, saline-pocket endoscopic submucosal dissection.

declined to participate (n Z 3). Randomization was conducted for 95 patients (48 in the S-ESD group and 47 in the SP-ESD group; Fig. 1). Univariate analysis of the baseline characteristics in patients between the S-ESD and SP-ESD groups by intention-to-treat analysis is shown in Table 1. There were no significant differences between the 2 groups with respect to age, gender, body mass index, comorbidity (hypertension, diabetes mellitus, and chronic kidney disease), antithrombotic agents, tumor location, macroscopic type, tumor size, and endoscopy technician. The technical success rate was 97.9% (46/47) in the SP-ESD group and 93.8% (45/48) in the S-ESD group (P Z .617).

The ESD procedure was not completed in the S-ESD group for technical reasons in 1 patient because of severe fibrosis and in 2 patients because of the need for an additional snaring technique. The ESD procedure was not completed in 1 patient in the SP-ESD group because of the need for an additional snaring technique. Therefore, 91 patients underwent per-protocol analysis in the study

(45 patients in the S-ESD group and 46 patients in the SP-ESD group). Univariate analysis of the clinical outcomes in patients between the S-ESD and SP-ESD groups by perprotocol analysis is shown in Table 2. There were no significant differences between the 2 groups with respect to specimen size, fibrosis, fatty tissue, and adverse effects. With respect to the pathologic findings, the percentages of adenomas, intramucosal carcinomas, and carcinomas with submucosal invasion were 77.8%, 22.2%, and 0% in the S-ESD group and 41.3%, 50.0%, and 8.7% in the SP-ESD group, respectively. The dissection speed in the SP-ESD group was significantly faster than that in the S-ESD group (20.1 mm2/min [range, 17.3-28.1] vs 16.3 mm2/min [range, 11.4-19.8]; P < .001). The procedure time in the SP-ESD group was significantly shorter than that in the S-ESD group (29.5 minutes [range, 22.3-44] vs 41 minutes [range, 31-55]; P < .001). En bloc and complete resection rates in both groups were 100%. No perforations occurred in either group. The volume of saline solution used in the SP-ESD group was significantly greater than that in the S-ESD group (200 mL [range, 120250] vs 150 mL [range, 100-200]; P Z .016).

www.giejournal.org

Volume 90, No. 2 : 2019 GASTROINTESTINAL ENDOSCOPY 283

Study outcomes

Saline-pocket ESD for colorectal neoplasms

Harada et al

TABLE 3. Subgroup analysis of therapeutic outcomes related to morphology between the S-ESD and SP-ESD groups by per-protocol analysis LST-G (n [ 46) S-ESD group (n [ 23)

SP-ESD group (n [ 23)

45 (29-60)

26 (23-41.5)

Median dissection speed, mm /min (IQR)

19.3 (13.5-23.1)

Median water quantity, mL (IQR)

150 (100-200)

Median procedure time, min (IQR) 2

LST-NG (n [ 45)

P value

S-ESD group (n [ 22)

SP-ESD group (n [ 23)

P value

.018

40 (35.5-54)

30 (22.5-44.5)

.058

21.6 (17.8-29.8)

.102

14.4 (9.4-17.6)

19.3 (16.1-21.7)

.001

200 (150-245)

.037

150 (100-200)

180 (100-250)

.269

IQR, Interquartile range; LST-G, laterally spreading tumor, granular type; LST-NG, laterally spreading tumor, nongranular type; S-ESD, standard endoscopic submucosal dissection; SP-ESD, saline-pocket endoscopic submucosal dissection.

TABLE 4. Subgroup analysis of therapeutic outcomes related to fibrosis grade between the S-ESD and SP-ESD groups by per-protocol analysis Fibrosis 0 (n [ 28) SP-ESD group (n [ 12)

36.5 (25.8-54)

Fibrosis 1 (n [ 48)

P value

S-ESD group (n [ 22)

SP-ESD group (n [ 26)

24.5 (15-31.8)

.048

39 (36.3-48.8)

25 (22.3-38.5)

Median dissection speed, 19.4 (16.4-22.4) 21.6 (18.2-31) mm2/min (IQR)

.296

14.2 (10.6-18.1) 19.1 (16.7-28.6)

Median water quantity, mL (IQR)

.033

150 (100-200) 160 (100-200)

Median procedure time, min (IQR)

S-ESD group (n [ 16)

100 (92.5-177.5) 190 (137.5-220)

Fibrosis 2 (n [ 15)

P value

S-ESD group (n [ 7)

SP-ESD group (n [ 8)

P value

.005

60 (54-60.5)

45.5 (42-51.8)

.324

.001

11.4 (10.5-17.5) 20 (17.7-21.3)

.021

.385

200 (185-237.5) 350 (300-402.5)

.016

IQR, Interquartile range; S-ESD, standard endoscopic submucosal dissection; SP-ESD, saline-pocket endoscopic submucosal dissection.

Subgroup analysis Subgroup analysis of therapeutic outcomes related to morphology between the S-ESD and SP-ESD groups by per-protocol analysis is shown in Table 3. The procedure time for LST-G in the SP-ESD group was significantly shorter than that in the S-ESD group (26 minutes [range, 23-41.5] vs 45 minutes [range, 29-60]; P Z .018). The volume of saline solution used in the SP-ESD group was significantly greater than that in the S-ESD group (200 mL [range, 150-245] vs 150 mL [range, 100-200]; P Z .037). In contrast, the dissection speed for LST-NG in the SPESD group was significantly faster than that in the S-ESD group (19.3 mm2/min [range, 16.1-21.7] vs 14.4 mm2/min [range, 9.4-17.6]; P Z .001). Subgroup analysis of therapeutic outcomes related to fibrosis grade between the S-ESD and SP-ESD groups by per-protocol analysis is shown in Table 4. The dissection speed with fibrosis 1 and 2 in the SP-ESD group was significantly faster than that in the S-ESD group (19.1 mm2/min [range, 16.7-28.6] vs 14.2 mm2/min [range, 10.6-18.1; P Z .001] and 20 mm2/min [range, 17.7-21.3] vs 11.4 mm2/min [range, 10.5-17.5; P Z .021]). Although the procedure time with fibrosis 1 in the SP-ESD group was significantly shorter than that in the S-ESD group (25 minutes [range, 22.3-38.5] vs 39 minutes [range, 36.3-48.8]; P Z .005), the procedure time with fibrosis 2 in the SP-ESD group was not significantly

shorter than that in the S-ESD group (45.5 minutes [range, 42-51.8] vs 60 minutes [range, 54-60.5]; P Z .324). The volume of saline solution with fibrosis 2 used in the SP-ESD group was significantly greater than that in the S-ESD group (350 mL [range, 300-402.5] vs 200 mL [range, 185-237.5]; P Z .016). The procedure time without fibrosis in the SP-ESD group was significantly shorter than that in the S-ESD group (24.5 minutes [range, 15-31.8] vs 36.5 minutes [range, 25.8-54]; P Z .048). The volume of saline solution without fibrosis used in the SP-ESD group was significantly greater than that in the S-ESD group (190 mL [range, 137.5-220] vs 100 mL [range, 92.5-177.5]; P Z .033). Subgroup analysis of therapeutic outcomes related to tumor locations between the S-ESD and SP-ESD groups by per-protocol analysis is shown in Table 5. On the right side of the colon, the dissection speed and the procedure time were significantly different (18.8 mm2/ min [range, 15.5-27.8] vs 14.3 mm2/min [range, 11.3-19; P Z .008] and 31 minutes [range, 24-39 min] vs 39 minutes [range, 33-57; P Z .033]). On the left side of the colon, the dissection speed in the SP-ESD group was significantly faster than that in the S-ESD group (21.1 mm2/min [range, 17.5-29.1] vs 15.8 mm2/min [range, 10.5-19.6]; P Z .011). In contrast, the dissection speed and procedure time for the rectum in the 2 groups were not significantly different (P > .99 and P Z .329).

284 GASTROINTESTINAL ENDOSCOPY Volume 90, No. 2 : 2019

www.giejournal.org

Harada et al

Saline-pocket ESD for colorectal neoplasms

TABLE 5. Subgroup analysis of therapeutic outcomes related to tumor locations between the S-ESD and SP-ESD groups by per-protocol analysis Right side (n [ 48)

Left side (n [ 32)

S-ESD group (n [ 23)

SP-ESD group (n [ 25)

P value

39 (33-57)

31 (24-39)

.033

14.3 (11.3-19) 18.8 (15.5-27.8) Median dissection speed, mm2/min (IQR) Median water quantity, mL (IQR)

Median procedure time, min (IQR)

150 (100-200) 200 (150-230)

S-ESD group (n [ 16)

SP-ESD group (n [ 16)

Rectum (n [ 11)

P value

S-ESD group (n [ 6)

SP-ESD group (n [ 5)

P value

41.5 (26.8-52.5) 25.5 (14.8-46.5)

.097

57 (40.5-63)

30 (22-58)

.329

.008

15.8 (10.5-19.6) 21.1 (17.5-29.1)

.011

21.5 (15.9-30.4) 21.7 (19.9-21.9)

>.99

.138

125 (100-197.5) 160 (100-297.5)

.147

200 (200-200)

.766

210 (200-250)

IQR, Interquartile range; S-ESD, standard endoscopic submucosal dissection; SP-ESD, saline-pocket endoscopic submucosal dissection.

This study determined the efficacy and safety of SP-ESD for SCNs, although we had previously clarified the efficacy and safety of SP-ESD for gastric neoplasms.15 Theoretically, SP-ESD is more suitable for colorectal than gastric neoplasms because the tangential line, which is advantageous for making the submucosal pocket, can be easily secured to a narrower space. This randomized study showed that SP-ESD is an effective and safe treatment for SCNs. The dissection speed and procedure time in the SP-ESD group were superior to those in the S-ESD group. No major adverse effects, including perforations, occurred in the SP-ESD group. Additionally, the dissection speed in the right-sided and left-sided colon was faster in the SP-ESD group than the S-ESD group. This result suggests that SP-ESD is more effective in the proximal colon, where the space is narrower than the rectum. Dissection of the submucosal layer was very stable in the pocket because the tip of the endoscope was fixed to insert into the pocket (pocket-creation method]).20–24 Therefore, regardless of arterial pulsations and respiratory fluctuation, the pocket-creation method facilitates accurate and safe dissection to avoid perforation and predict bleeding. Similarly, SP-ESD allows the tip of the endoscope to be fixed in the submucosal pocket and provides for a precise dissection line because of water immersion. Furthermore, in SP-ESD, water immersion provides floating effects that enable a wider space for the dissection line in the submucosa, thus resulting in better traction. SPESD also provides image enhancement (approximately 1.3fold) in a water environment because of the refractive index of water. Therefore, SP-ESD facilitates accurate dissection because the endoscopic view in water immersion is more visible. Accordingly, in the SP-ESD group the dissection speed and procedure time were superior to those in the S-ESD group in spite of the larger amount of fibrosis and/or submucosal invasion. Furthermore, in cases with LST-NG and fibrosis, the dissection speed in the SP-ESD group was faster than that in the S-ESD group. Thus, we believe that SP-ESD contributes

to lesions with fibrosis. The dissection line can be more precisely recognized by water immersion in spite of the restricted space via fibrosis in the pocket. The endoscopic view becomes cloudy in the gas insufflation method because of fatty tissue of the submucosal layer and coagulation of submucosal tissue.19 It is timeconsuming to clean the cloudiness of the endoscopic lens by injecting water through the water jet function of the electrosurgical knife and the endoscopic channel or by directly wiping the endoscopic lens. Furthermore, the cloudiness of the endoscopic lens causes an inaccurate submucosal dissection, resulting in possible adverse effects, such as bleeding and perforation. In contrast, SP-ESD provides a clearer view to absorb the fumes from diathermy with an electrosurgical knife and to wash out the lipid from fatty tissue. Although there were enriched fatty tissues in the ascending and transverse colon in this study, especially in the ileocecal valve, by using the water jet function in SP-ESD it was possible to wash out oil droplets in the water immersion, which resulted in a clearer view to avoid the cloudiness of the endoscopic view. As a result of these novel benefits, the SP-ESD method provided an easier and safer procedure for colorectal ESD. In this study we used saline solution for water immersion, which was sufficiently feasible and safe to dissect the submucosa. Despott and Murino25 reported that saline solution is suitable for underwater endoscopic resection as ESD; however, air bubbles arise from an electrosurgical knife with electrocautery because of the air contained in the submucosa and steaming of saline solution. Air bubbles must be removed using the water jet function because air bubbles impair the endoscopic view. In urologic procedures, sorbitol is used for transurethral resection. The quantity of air bubbles differs based on type of solution; however, sorbitol can be the cause of hyponatremia, thus resulting in transurethral resection syndrome.26–28 In gastroenterologic procedures, although sorbitol and mannitol have been used to cleanse the colon, there have been a few reports involving combustion when using cautery because of the mixture of air and

www.giejournal.org

Volume 90, No. 2 : 2019 GASTROINTESTINAL ENDOSCOPY 285

DISCUSSION

Saline-pocket ESD for colorectal neoplasms

hydrogen produced by bacterial fermentation.29–31 Further studies involving various solutions are warranted. This study had limitations. First, ESD procedures were performed by 1 expert ESD endoscopist at a single center to avoid the bias of interoperator variance. Therefore, a similar evaluation by endoscopy trainees and an estimation of the learning curve for the SP-ESD method should be conducted in a corollary study. Second, this study was designed to compare the efficacy and feasibility between SP-ESD and S-ESD but not to compare between SP-ESD and pocket-creation methods. In the near future, we plan to compare SP-ESD and pocket-creation methods. Finally, SP-ESD might not be suitable for lesions in the cecum because of the perpendicular condition. The tip of the endoscope is difficult to insert into the submucosal pocket for lesions in the cecum adjacent to the appendiceal orifice because the angle of the tip of the endoscope is perpendicular to the colonic wall. Therefore, some improvements regarding SP-ESD may be needed in a future study. In conclusion, SP-ESD is an effective and safe treatment for SCNs because SP-ESD provides a clearer field of view and better traction in the submucosal layer than S-ESD. SP-ESD may serve as an alternative method for resection of SCNs. ACKNOWLEDGEMENT This work was supported by The Japanese Foundation for Research and Promotion of Endoscopy (JFE) Grant (18-026). REFERENCES 1. Tanaka S, Kashida H, Saito Y, et al. JGES guidelines for colorectal endoscopic submucosal dissection/endoscopic mucosal resection. Dig Endosc 2015;27:417-34. 2. Saito Y, Uraoka T, Matsuda T, et al. A pilot study to assess the safety and efficacy of carbon dioxide insufflation during colorectal endoscopic submucosal dissection with the patient under conscious sedation. Gastrointest Endosc 2007;65:537-42. 3. Nonaka S, Saito Y, Takisawa H, et al. Safety of carbon dioxide insufflation for upper gastrointestinal tract endoscopic treatment of patients under deep sedation. Surg Endosc 2010;24:1638-45. 4. Binmoeller KF, Weilert F, Shah J, et al. “Underwater” EMR without submucosal injection for large sessile colorectal polyps (with video). Gastrointest Endosc 2012;75:1086-91. 5. Binmoeller KF, Shah JN, Bhat YM, et al. “Underwater” EMR of sporadic laterally spreading nonampullary duodenal adenomas (with video). Gastrointest Endosc 2013;78:496-502. 6. Amato A, Radaelli F, Spinzi G, et al. Underwater endoscopic mucosal resection: The third way for en bloc resection of colonic lesions? United Eur Gastroenterol J 2016;4:595-8. 7. Kim HG, Thosan N, Banerjee S, et al. Underwater endoscopic mucosal resection for recurrences after previous piecemeal resection of colorectal polyps (with video). Gastrointest Endosc 2014;80:1094-102. 8. Binmoeller KF, Hamerski CM, Shah JN, et al. Attempted underwater en bloc resection for large (2-4 cm) colorectal laterally spreading tumors (with video). Gastrointest Endosc 2015;81:713-8.

286 GASTROINTESTINAL ENDOSCOPY Volume 90, No. 2 : 2019

Harada et al 9. Binmoeller KF, Hamerski CM, Shah JN, et al. Underwater EMR of adenomas of the appendiceal orifice (with video). Gastrointest Endosc 2016;83:638-42. 10. Curcio G, Granata A, Ligresti D, et al. Underwater colorectal EMR: remodeling endoscopic mucosal resection. Gastrointest Endosc 2015;81:1238-42. 11. Saito Y, Uraoka T, Yamaguchi Y, et al. A prospective, multicenter study of 1111 colorectal endoscopic submucosal dissections (with video). Gastrointest Endosc 2010;72:1217-25. 12. Isomoto H, Nishiyama H, Yamaguchi N, et al. Clinicopathological factors associated with clinical outcomes of endoscopic submucosal dissection for colorectal epithelial neoplasms. Endoscopy 2009;41: 679-83. 13. Hayashi N, Tanaka S, Nishiyama S, et al. Predictors of incomplete resection and perforation associated with endoscopic submucosal dissection for colorectal tumors. Gastrointest Endosc 2014;79: 427-35. 14. Sato K, Ito S, Kitagawa T, et al. Factors affecting the technical difficulty and clinical outcome of endoscopic submucosal dissection for colorectal tumors. Surg Endosc 2014;28:2959-65. 15. Harada H, Murakami D, Suehiro S, et al. Water-pocket endoscopic submucosal dissection for superficial gastric neoplasms (with video). Gastrointest Endosc 2018;88:253-60. 16. Kudo S, Lambert R, Allen J, et al. Nonpolypoid neoplastic lesions of the colorectal mucosa. Gastrointest Endosc 2008;68:S3-47. 17. Harada H, Suehiro S, Murakami D, et al. Clinical impact of prophylactic clip closure of mucosal defects after colorectal endoscopic submucosal dissection. Endosc Int Open 2017;05:E1165-71. 18. Matsumoto A, Tanaka S, Oba S, et al. Outcome of endoscopic submucosal dissection for colorectal tumors accompanied by fibrosis. Scand J Gastroenterol 2010;45:1329-37. 19. Yoshida N, Naito Y, Hirose R, et al. Risk of lens cloudiness during colorectal endoscopic submucosal dissection and ability of a novel lens cleaner to maintain and restore endoscopic view. Dig Endosc 2015;27:609-17. 20. Hayashi Y, Sunada K, Takahashi H, et al. Pocket-creation method of endoscopic submucosal dissection to achieve en bloc resection of giant colorectal subpedunculated neoplastic lesions. Endoscopy 2014;46:E421-2. 21. Hayashi Y, Miura Y, Yamamoto H. Pocket-creation method for the safe, reliable, and efficient endoscopic submucosal dissection of colorectal lateral spreading tumors. Dig Endosc 2015;27:534-5. 22. Sakamoto H, Hayashi Y, Miura Y, et al. Pocket-creation method facilitates endoscopic submucosal dissection of colorectal laterally spreading tumors, non-granular type. Endosc Int Open 2017;5:E123-9. 23. Kanamori A, Nakano M, Kondo M, et al. Clinical effectiveness of the pocket-creation method for colorectal endoscopic submucosal dissection. Endosc Int Open 2017;5:E1299-305. 24. Yoshida N, Naito Y, Yasuda R, et al. The efficacy of the pocket-creation method for cases with severe fibrosis in colorectal endoscopic submucosal dissection. Endosc Int Open 2018;6: E975-83. 25. Despott EJ, Murino A. Saline-immersion therapeutic endoscopy (SITE): an evolution of underwater endoscopic lesion resection. Dig Liver Dis 2017;49:1376. 26. Mebust WK, Holtgrewe HL, Cockett AT, et al. Transurethral prostatectomy: immediate and postoperative complications. A cooperative study of 13 participating institutions evaluating 3,885 patients. J Urol 2002;167:999-1003. 27. Michielsen DPJ, Debacker T, De Boe V, et al. Bipolar transurethral resection in saline-an alternative surgical treatment for bladder outlet obstruction? J Urol 2007;178:2035-9. 28. Gravenstein D. Transurethral resection of the prostate (TURP) syndrome: a review of the pathophysiology and management. Anesth Analg 1997;84:438-46. 29. Bigard MA, Gaucher P, Lassalle C. Fatal colonic explosion during colonoscopic polypectomy. Gastroenterology 1979;77:1307-10.

www.giejournal.org

Harada et al 30. La Brooy SJ, Avgerinos A, Fendick CL, et al. Potentially explosive colonic concentrations of hydrogen after bowel preparation with mannitol. Lancet 1981;1:634-6. 31. Josemanders DFGM, Spillenaar Bilgen EJ, van Sorge AA, et al. Colonic explosion during endoscopic polypectomy: avoidable complication or bad luck? Endoscopy 2006;38:943-4.

Abbreviations: ESD, endoscopic submucosal dissection; LST, laterally spreading tumor; LST-G, laterally spreading tumor, granular type; LST-NG, laterally spreading tumor, nongranular type; S-ESD, standard endoscopic submucosal dissection; SP-ESD, saline-pocket endoscopic submucosal dissection; SCN, superficial colorectal neoplasm. DISCLOSURE: All authors disclosed no financial relationships relevant to this publication. Research support for this study was provided by a grant from the Japanese Foundation for Research and Promotion of Endoscopy to H. Harada.

Saline-pocket ESD for colorectal neoplasms

This video can be viewed directly from the GIE website or by using the QR code and your mobile device. Download a free QR code scanner by searching “QR Scanner” in your mobile device’s app store. Copyright ª 2019 by the American Society for Gastrointestinal Endoscopy 0016-5107/$36.00 https://doi.org/10.1016/j.gie.2019.03.023 Received January 30, 2019. Accepted March 15, 2019. Current affiliations: Department of Gastroenterology (1), Department of Endoscopy (2), New Tokyo Hospital, Chiba, Japan. Reprint requests: Hideaki Harada, MD, Department of Gastroenterology, New Tokyo Hospital, 1271 Wanagaya, Matsudo Chiba 270-2232, Japan.

Audio and Podcasts of article abstracts published in Gastrointestinal Endoscopy are now available online. Recordings are performed by Deborah Bowman, MFA, ELS, Senior Managing Editor of Clinical Publications. Scan the QR code or log on to www.giejournal.org to listen to recordings from the current issue.

www.giejournal.org

Volume 90, No. 2 : 2019 GASTROINTESTINAL ENDOSCOPY 287