Indexed In : Science Citation Index Expanded(SCIE), MEDLINE,
Pubmed/Pubmed Central, Elsevier Bibliographic, Google Scholar,
Databases(Scopus & Embase), KCI, KoreaMed, DOAJ
Gut and Liver is an international journal of gastroenterology, focusing on the gastrointestinal tract, liver, biliary tree, pancreas, motility, and neurogastroenterology. Gut atnd Liver delivers up-to-date, authoritative papers on both clinical and research-based topics in gastroenterology. The Journal publishes original articles, case reports, brief communications, letters to the editor and invited review articles in the field of gastroenterology. The Journal is operated by internationally renowned editorial boards and designed to provide a global opportunity to promote academic developments in the field of gastroenterology and hepatology. +MORE
Yong Chan Lee |
Professor of Medicine Director, Gastrointestinal Research Laboratory Veterans Affairs Medical Center, Univ. California San Francisco San Francisco, USA |
Jong Pil Im | Seoul National University College of Medicine, Seoul, Korea |
Robert S. Bresalier | University of Texas M. D. Anderson Cancer Center, Houston, USA |
Steven H. Itzkowitz | Mount Sinai Medical Center, NY, USA |
All papers submitted to Gut and Liver are reviewed by the editorial team before being sent out for an external peer review to rule out papers that have low priority, insufficient originality, scientific flaws, or the absence of a message of importance to the readers of the Journal. A decision about these papers will usually be made within two or three weeks.
The remaining articles are usually sent to two reviewers. It would be very helpful if you could suggest a selection of reviewers and include their contact details. We may not always use the reviewers you recommend, but suggesting reviewers will make our reviewer database much richer; in the end, everyone will benefit. We reserve the right to return manuscripts in which no reviewers are suggested.
The final responsibility for the decision to accept or reject lies with the editors. In many cases, papers may be rejected despite favorable reviews because of editorial policy or a lack of space. The editor retains the right to determine publication priorities, the style of the paper, and to request, if necessary, that the material submitted be shortened for publication.
Myeongseok Koh1 , Jong Yoon Lee1
, Song-Hee Han2
, Seong Woo Jeon3
, Su Jin Kim4
, Joo Young Cho5
, Seong Hwan Kim5
, Jae Young Jang6
, Gwang Ho Baik7
, Jin Seok Jang1
Correspondence to: Jin Seok Jang
ORCID https://orcid.org/0000-0001-8067-4598
E-mail jsjang@dau.ac.kr
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
Gut Liver 2023;17(2):234-242. https://doi.org/10.5009/gnl220025
Published online November 1, 2022, Published date March 15, 2023
Copyright © Gut and Liver.
Background/Aims: Endoscopic submucosal dissection is a widely used treatment for gastric epithelial neoplasms. Accurate delineation of the horizontal margins is necessary for the complete resection of gastric epithelial neoplasms. Recently, image-enhanced endoscopy has been used to evaluate horizontal margins of gastric epithelial neoplasms. The aim of this study was to investigate whether I-SCAN-optical enhancement (I-SCAN-OE) is superior to chromoendoscopy in evaluating the horizontal margin of gastric epithelial neoplasms.
Methods: This was a multicenter, prospective, and randomized trial. The participants were divided into two groups: I-SCAN-OE and chromoendoscopy. For both groups, we first evaluated the horizontal margins of early gastric cancer or high-grade dysplasia using white-light imaging, and then evaluated, the horizontal margins using I-SCAN-OE or chromoendoscopy. We devised a unique scoring method based on the pathological results obtained after endoscopic submucosal dissection to accurately evaluate the horizontal margins of gastric epithelial neoplasms. The delineation scores of both groups were compared, as were the ratios of positive/negative horizontal margins.
Results: In total, 124 patients were evaluated for gastric epithelial neoplasms, of whom 112 were enrolled in the study. A total of 112 patients participated in the study, and 56 were assigned to each group (1:1). There was no statistically significant difference in the delineation scores between the groups (chromoendoscopy, 7.80±1.94; I-SCAN-OE, 8.23±2.24; p=0.342).
Conclusions: I-SCAN-OE did not show superiority over chromoendoscopy in delineating horizontal margins of gastric epithelial neoplasms.
Keywords: Chromoendoscopy, Gastric cancer, Endoscopic submucosal dissection, Gastric neoplasm, Image-enhanced endoscopy
Endoscopic submucosal dissection (ESD) is a widely accepted treatment for early gastric cancer (EGC) or gastric adenoma. Accurate delineation of the horizontal margins is necessary for the complete resection of gastric epithelial neoplasms.1,2 Chromoendoscopy, which has been used in combination with indigo carmine and acetic acid, is considered a standard technique for delineation. This technique improves the diagnostic value of delineating the horizontal margins of EGCs or gastric adenomas.3-5 However, recently, image-enhanced endoscopy has replaced chromoendoscopy and is being used to evaluate the horizontal margins of EGCs or gastric adenomas.
Image-enhanced endoscopy has been developed and widely used clinically in recent years, including narrow-band imaging (NBI; Olympus, Tokyo, Japan), blue laser imaging (BLI; Fujifilm, Tokyo, Japan), and I-SCAN (Pentax, Tokyo, Japan). NBI is an optical digital imaging technique wherein two selected wavelengths (415±30 nm and 540±30 nm) enhance the structural aspects of the surface of the mucosa and existing vessels,6 indicating its utility in the diagnosis of EGC and delineation.7-10 BLI includes two types of lasers with wavelengths of 410 and 450 nm, similar to NBI, used in the diagnosis of EGC and the evaluation of margins.11 I-SCAN involves contrast enhancement, surface enhancement, and tone enhancement, and optical enhancement (OE) has recently been added since 2016. The I-SCAN-OE function employs band-limited light to achieve high overall transmittance by connecting the peaks of the hemoglobin absorption spectrum (415, 540, and 570 nm), thus creating a continuous wavelength spectrum.12,13 This is similar to the previously introduced NBI or BLI, but since it was recently introduced, there have been few studies on this technique. Therefore, this study aimed to determine whether I-SCAN-OE is superior to chromoendoscopy in evaluating the horizontal margins of EGCs or gastric adenomas.
This was a multicenter, prospective, randomized trial. The participants were divided into two groups: I-SCAN-OE and chromoendoscopy. For both groups, we first evaluated the horizontal margins of the gastric epithelial neoplasms with white-light imaging, and then evaluated, the horizontal margins of gastric epithelial neoplasms for each group using I-SCAN-OE or chromoendoscopy (Fig. 1). We devised a unique scoring method based on pathological results obtained after ESD to accurately evaluate the horizontal margins of gastric epithelial neoplasms. The delineation scores of the I-SCAN-OE and chromoendoscopy groups and the ratios of horizontal margin resection were compared. This study was conducted in three Korean institutions that are tertiary hospitals with extensive ESD experience.
Patients aged ≥20 years who underwent ESD for the treatment of category 4 (mucosal high-grade dysplasia or intraepithelial carcinoma) disease according to revised Vienna classification confirmed by forceps biopsy before ESD were included.14 The exclusion criteria were as follows: (1) refusal to provide informed consent; (2) category 1 or 2 confirmed in the specimen after ESD; (3) remnant stomach; (4) low cardiopulmonary function with high risk of ESD; and (5) withdrawal of consent after study enrollment. After enrollment, the participants were allocated to either the I-SCAN-OE or chromoendoscopy group in a 1:1 ratio using a simple randomization method.
We evaluated whether the horizontal margins were visually well delineated before the procedure. The chromoendoscopy group was evaluated using white-light imaging, and the confidence level was evaluated again after spraying with 20 mL of 0.2% indigo carmine solution. The I-SCAN-OE group was evaluated using white-light imaging, and the confidence level was evaluated using I-SCAN-OE mode.
Before performing ESD, the horizontal margins were evaluated in each group, and a dot was marked 5 mm outside the margin. As the thick tip of the dual knife (Olympus Medical Corp., Tokyo, Japan) used during the procedure was 5 mm in size, the marking dot was positioned using this (Fig. 2). The tumor was then cut along the marked dot (Fig. 3). After resection, the resected specimen was marked with margins of lateral and vertical sides with different colors and sectioned perpendicularly at 2 mm intervals to examine the exact margin state and the distance of the safety margin. Pathological evaluation measured the length of the nearest resection margin with a pen on the hematoxylin and eosin-stained slide of the cancerous lesion (Fig. 4). The lateral margin of the pathological specimen was evaluated as proximal, distal, anterior, or posterior and scored. If the horizontal margin was close to 5 mm (4–6 mm), it was considered “Excellent” and scored 3 points. If the horizontal margin was ≥2 mm, <4 mm or >6 mm, and ≤8 mm, it was judged as “Good” and scored 2 points; and if the horizontal margin was >0 mm, <2 mm or >8 mm, and ≤10 mm), it was judged as “Fair” and scored 1 point. If the margin was more than 10 mm away or the lateral margin was in contact with the resected specimen, it was judged as “Poor” and scored 0. Each of the four margins was assigned a score from 0 to 12, and the sum of the scores was called the delineation score (Fig. 5). Accurate delineation was defined as a total score of 8 points or more and 1 point or more obtained from each of the four margins. Inaccurate delineation was defined as a total score of 7 points or less, or including 0 points, even on one of the four margins. If the neoplasm was resected completely, including the horizontal margins, it was defined as curative resection; if the margins were not completely resected after ESD, incomplete resection was considered.
Upper gastrointestinal endoscopes with an I-SCAN-OE function were used in this study. All endoscopists used an EPK-i 7010 video processor with a Pentax EG-10-Zi endoscope (Hoya Corp., Tokyo, Japan). The endoscopists who performed the procedure in this study were all gastroenterologists, and they had performed endoscopy for at least 5 years at a gastrointestinal endoscopy training institution under the Korean Society of Gastrointestinal Endoscopy, with at least 4 years and at least 100 cases. Four endoscopists from three institutions participated in this study. Two were classified as less experienced endoscopists with less than 10 years of experience in ESD. The other two were endoscopists with more than 10 years of experience.
Before evaluation with chromoendoscopy or I-SCAN-OE, the endoscopists determined whether the horizontal margins were well demarcated using only white-light imaging and evaluated the confidence level. The confidence level was defined as whether the boundary was clearly demarcated visually by the endoscopist to determine the boundary of the lesion.
The primary endpoint was the delineation score according to group. The secondary endpoints were positive lateral margins according to groups, and the clinical factors associated with accurate delineation and positive lateral margins were determined.
The participants were divided into two groups and assigned in a 1:1 ratio to the chromoendoscopy and I-SCAN-OE groups. Continuous data are presented as mean±standard deviation and were analyzed using the Student t-test. Categorical data were analyzed using the Pearson chi-square or Fisher exact tests. A logistic regression model was used for multivariate analysis to assess the associations between clinical factors and accurate delineation of gastric epithelial neoplasms. Variables found to be significantly associated with the univariate analysis were used as co-variables in the multivariate analysis. All analyses were performed using SPSS software (version 26.0; IBM Corp., Armonk, NY, USA). Statistical significance was set at p<0.05.
The primary endpoint of this study was the delineation score. We used a unique method for evaluating the horizontal margins of gastric epithelial neoplasms, which has not been previously reported. Therefore, it was not possible to measure sample size by proposing a hypothesis based on previous studies. We analyzed 10 samples at the beginning of the study and confirmed that the delineation score was 8.5 (standard deviation, 1.6). The difference between the scores of the two groups was defined as superiority and inferiority of more than 10%, and the sample sizes to verify this were measured at an alpha level of 0.05, and a beta level of 0.80. In a
Our research protocol was approved by the Ethics Committee and was in accordance with international agreements (World Medical Association Declaration of Helsinki: Ethical principles for medical research involving human subjects). Informed consent was obtained from all participants. The study protocol was reviewed and approved by the Institutional Review Board of Dong-A University College of Medicine (IRB number: DAUHIRB-19-063) on March 27, 2019. This trial was registered with the Clinical Trials Registry (Clinical Research Information Service in South Korea, CRIS) (number KCT0003819). All authors had access to the study data and reviewed and approved the final manuscript. Research data supporting this publication are available from https://cris.nih.go.kr/cris/search/detailSearch.do/19003.
To assign participants, 124 patients were evaluated for EGC and HGD, of which 112 patients were enrolled in the study, from March 27, 2019, to March 31, 2021. A total of 112 patients participated in the study, and 56 were assigned to each group. EGC or HGD was confirmed by forceps biopsy before the procedure, but four patients were excluded because they showed negative or indefinite results. Eight patients were excluded due to withdrawal of consent (Fig. 6). There were no significant differences in demographic and clinical characteristics between the two groups (Table 1).
Table 1 Demographic and Clinical Characteristics of the Participants According to Group
Characteristics | Chromoendoscopy (n=56) | I-SCAN-OE (n=56) | p-value |
---|---|---|---|
Age, yr | 66.00±8.56 | 67.93±9.59 | 0.264 |
Sex | 0.154 | ||
Male | 48 (85.71) | 42 (75.00) | |
Female | 8 (14.29) | 14 (25.00) | |
Tumor size, mm | 12.92±7.60 | 15.09±8.48 | 0.157 |
Procedure time, min | 18.61±9.52 | 17.88±10.24 | 0.696 |
Location | 0.209 | ||
Antrum | 37 (66.07) | 43 (76.79) | |
Body and fundus | 19 (33.93) | 13 (23.21) | |
Macroscopic type | 0.958 | ||
Elevated | 21 (37.50) | 23 (41.07) | |
Flat | 10 (17.86) | 8 (14.29) | |
Depressed | 20 (35.71) | 20 (35.71) | |
Mixed | 5 (8.93) | 5 (8.93) | |
1.000 | |||
Positive | 19 (33.93) | 19 (33.93) | |
Negative | 37 (66.07) | 37 (66.07) | |
Final pathology | 0.238 | ||
HGD | 12 (21.43) | 6 (10.71) | |
EGC (differentiated) | 42 (75.00) | 49 (87.50) | |
EGC (undifferentiated) | 2 (3.57) | 1 (1.79) | |
Invasion depth | 0.768 | ||
HGD and T1a | 50 (89.29) | 49 (87.50) | |
T1b and deeper | 6 (10.71) | 7 (12.50) | |
Hospital | 0.845 | ||
A | 40 (71.43) | 38 (67.86) | |
B | 10 (17.86) | 10 (17.86) | |
C | 6 (10.71) | 8 (14.29) | |
Endoscopists | 0.324 | ||
Experienced | 48 (85.71) | 44 (78.57) | |
Less experienced | 8 (14.29) | 12 (21.43) | |
0.647 | |||
Yes | 54 (96.43) | 53 (94.64) | |
Piecemeal resection | 2 (3.57) | 3 (5.36) | |
Anticoagulation or antiplatelet use | 0.357 | ||
Yes | 10 (17.86) | 14 (25.00) | |
No | 46 (82.14) | 42 (75.00) |
Data are presented as mean±SD or number (%).
I-SCAN-OE, I-SCAN-optical enhancement; HGD, high-grade dysplasia; EGC, early gastric cancer.
Before evaluation with chromoendoscopy or I-SCAN-OE, there was no statistically significant difference between the groups (chromoendoscopy, 55.36%; I-SCAN-OE, 48.21%; p=0.449). When the margins were evaluated with chromoendoscopy or I-SCAN-OE, and the confidence level was evaluated again, it was found to have increased in the I-SCAN-OE group, but this was not statistically significant (chromoendoscopy, 55.36%; I-SCAN-OE, 71.43%; p=0.078) (Table 2).
Table 2 Confidence Level of Delineation as Assessed before Endoscopic Submucosal Dissection According to Group
Confidence level | Chromoendoscopy (n=56) | I-SCAN-OE (n=56) | p-value |
---|---|---|---|
Before enhancing | 0.449 | ||
Low | 25 (44.64) | 29 (51.79) | |
High | 31 (55.36) | 27 (48.21) | |
After enhancing | 0.078 | ||
Low | 25 (44.64) | 16 (28.57) | |
High | 31 (55.36) | 40 (71.43) |
Data are presented as the number (%).
I-SCAN-OE, I-SCAN-optical enhancement.
There was no statistically significant difference in the delineation score between the groups (chromoendoscopy, 7.80±1.94; I-SCAN-OE, 8.23±2.24; p=0.342). The ratio of the negative lateral margin was 91.07%, which was the same in both groups. The ratio of accurate delineation was higher in the I-SCAN-OE group than in the chromoendoscopy group; however, the difference was not statistically significant (chromoendoscopy, 50.00%; I-SCAN-OE, 60.71%; p=0.254) (Table 3).
Table 3 Delineation Score, Curative Resection Rate and Accuracy According to Group
Variable | Chromoendoscopy (n=56) | I-SCAN-OE (n=56) | p-value |
---|---|---|---|
Delineation score, mean±SD | 7.80±1.94 | 8.23±2.24 | 0.342 |
Negative lateral margin, % | 91.07 | 91.07 | 1.000 |
Yes:No | 51:5 | 51:5 | |
Accuracy, % | 50.00 | 60.71 | 0.254 |
Accurate:Inaccurate | 28:28 | 34:22 |
I-SCAN-OE, I-SCAN-optical enhancement.
There were 62 accurate and 50 inaccurate delineations. Among the clinical features, tumor size (accurate delineation, 12.42±7.18 mm; inaccurate delineation, 15.69±8.81 mm; p=0.020), procedure time (accurate delineation, 15.69±8.67 minutes; inaccurate delineation, 21.40±10.27 minutes; p=0.002), and confidence level (accurate delineation, 74.19%; inaccurate delineation, 50.00%; p=0.008) showed statistically significant differences between accurate and inaccurate delineations. Accurate delineation was associated with smaller tumor size, shorter procedure time, and higher confidence level than inaccurate delineation (Table 4).
Table 4 Demographic and Clinical Characteristics According to Accuracy
Characteristics | Accurate delineation (n=62) | Inaccurate delineation (n=50) | p-value |
---|---|---|---|
Group | 0.254 | ||
Chromoendoscopy | 28 (45.16) | 28 (56.00) | |
I-SCAN-OE | 34 (54.84) | 22 (44.00) | |
Age, yr | 65.90±8.67 | 68.28±9.53 | 0.170 |
Sex | 0.694 | ||
Male | 49 (79.03) | 41 (82.00) | |
Female | 13 (20.97) | 9 (18.00) | |
Tumor size, mm | 12.42±7.18 | 15.69±8.81 | 0.020* |
Procedure time, min | 15.69±8.67 | 21.40±10.27 | 0.002* |
Location | 0.118 | ||
Antrum | 48 (77.42) | 32 (64.00) | |
Body and fundus | 14 (22.58) | 18 (36.00) | |
Macroscopic type | 0.261 | ||
Elevated | 26 (41.94) | 18 (36.00) | |
Flat | 7 (11.29) | 11 (22.00) | |
Depressed | 25 (40.32) | 15 (30.00) | |
Mixed | 4 (6.45) | 6 (12.00) | |
0.414 | |||
Positive | 19 (30.65) | 19 (38.00) | |
Negative | 43 (69.35) | 31 (62.00) | |
Final pathology | 0.058 | ||
HGD | 7 (11.29) | 11 (22.00) | |
EGC (differentiated) | 55 (88.70) | 36 (72.00) | |
EGC (undifferentiated) | 3 (6.00) | ||
Invasion depth | 0.058 | ||
HGD and T1a | 58 (93.55) | 41 (82.00) | |
T1b and deeper | 4 (6.45) | 9 (18.00) | |
Confidence level (after enhancing) | 0.008* | ||
Low | 16 (25.80) | 25 (50.00) | |
High | 46 (74.19) | 25 (50.00) | |
Endoscopists | 0.043* | ||
Experienced | 55 (88.71) | 37 (74.00) | |
Less experienced | 7 (11.29) | 13 (26.00) | |
0.170 | |||
Yes | 61 (98.39) | 46 (92.00) | |
Piecemeal resection | 1 (1.61) | 4 (8.00) | |
Anticoagulation or antiplatelet use | 0.128 | ||
Yes | 10 (16.13) | 14 (28.00) | |
No | 52 (83.87) | 36 (72.00) |
Data are presented as number (%) or mean±SD.
I-SCAN-OE, I-SCAN-optical enhancement; HGD, high-grade dysplasia; EGC, early gastric cancer.
*p<0.05.
Table 1 shows that the procedure time, tumor size, macroscopic features, and confidence level were statistically significant factors related to positive lateral margins in the univariate logistic regression analysis. Long procedure duration, large and flat tumors, and low confidence level were related to positive lateral margins. Confidence level was identified as a statistically significant factor when these factors were analyzed using multivariate analysis as covariates (odds ratio [OR], 0.221; p=0.044) (Table 5). Procedure time (OR, 1.082; p=0.005), tumor size (OR, 1.092; p=0.024), confidence level (OR, 0.214; p=0.033), and endoscopist experience were statistically significant factors related to accurate delineation in univariate logistic regression analysis. A short procedure time, small tumor size, high confidence level, and extensive endoscopy experience were strongly associated with accurate delineation (Table 6). When these factors were analyzed by multivariate analysis as covariates, confidence level and endoscopist experience were identified as statistically significant factors.
Table 5 Univariate and Multivariate Analyses for Factors Related to Positive Lateral Margin
Factor | Univariate analysis | Multivariate analysis | |||
---|---|---|---|---|---|
OR (95% CI) | p-value | OR (95% CI) | p-value | ||
Procedure time | 1.082 (1.025–1.142) | 0.005* | 1.055 (0.990–1.124) | 0.097 | |
Tumor size | 1.092 (1.014–1.175) | 0.024* | 1.082 (0.989–1.183) | 0.084 | |
Macroscopic | 0.043* | 0.080 | |||
Elevated/depressed | 1 | 1 | |||
Flat | 4.190 (1.049–16.743) | 4.220 (0.842–21.152) | |||
Confidence level (after enhancing) | 0.033* | 0.044* | |||
Low | 1 | 1 | |||
High | 0.214 (0.052–0.881) | 0.221 (0.044–1.110) |
OR, odds ratio; CI, confidence interval.
*p<0.05.
Table 6 Univariate and Multivariate Analyses for Factors Related to Accurate Delineation
Factor | Univariate analysis | Multivariate analysis | |||
---|---|---|---|---|---|
OR (95% CI) | p-value | OR (95% CI) | p-value | ||
Procedure time, min | 0.935 (0.893–0.979) | 0.004* | 0.956 (0.912–1.003) | 0.064 | |
Tumor size | 0.944 (0.899–0.993) | 0.024* | 0.969 (0.917–1.024) | 0.261 | |
Confidence level (after enhancing) | 0.009* | 0.006* | |||
Low | 1 | 1 | |||
High | 2.875 (1.299–6.362) | 3.543 (1.431–8.774) | |||
Endoscopists | 0.049* | 0.023* | |||
Experienced | 1 | 1 | |||
Less experienced | 0.362 (0.132–0.993) | 0.265 (0.084–0.833) |
OR, odds ratio; CI, confidence interval.
*p<0.05.
Chromoendoscopy aims to increase the rate of visualization and detection of neoplastic lesions and has been applied throughout the gastrointestinal tract.3 In contrast, conventional white-light endoscopy has a miss rate of up to 25% for gastrointestinal pathology, specifically in the context of small and flat lesions.15 Although chromoendoscopy helps in the early diagnosis of neoplastic lesions, it takes time to disperse and aspirate, and has the disadvantage of difficulty in identifying the lesion if inflammation is severe.16
This was a multicenter, prospective, randomized trial aimed at determining whether I-SCAN-OE is superior to chromoendoscopy in delineating gastric epithelial neoplasms according to delineation score. Several previous studies have used NBI to evaluate the horizontal margin of EGCs. The accuracy of diagnosis of the cancer margin in EGC in addition to magnifying NBI was high when the margins were unclear using chromoendoscopy alone.9 in the comparison of chromoendoscopy and NBI, NBI has been found to be superior to the former,17,18 but other studies have shown both to be clinically similar.10 Another image-enhanced endoscopy, BLI, used in the diagnosis of EGC and the evaluation of margins, have been reported to have clinical utility similar to that of NBI.11,19,20
Our study comparatively analyzed the efficacy of I-SCAN-OE and chromoendoscopy. I-SCAN-OE uses white-light imaging components in addition to NBI at wavelengths of 415, 540, and 570 nm. Only one previous study has used I-SCAN-OE to confirm the diagnostic accuracy of cancer demarcation in EGC. Nagao
In the present study, the primary endpoint was delineation score. We devised this method to evaluate the diagnostic accuracy of demarcation of gastric epithelial neoplasms. There was no statistically significant difference between I-SCAN-OE (8.23±2.24) and chromoendoscopy (7.80±1.94). In addition, the ratio of positive lateral margins was 8.93%, which was similar in both groups. Thus, our data did not reveal any differences between I-SCAN-OE and chromoendoscopy in terms of their ability to accurately delineate the horizontal margins of gastric epithelial neoplasms. The accurate delineation group had a smaller tumor size, shorter procedure time, higher confidence level in pre-procedure delineation evaluation, and endoscopists with more experience than the inaccurate delineation group. The other factors were not statistically significant. Multivariate analysis indicated that a high confidence level was significantly associated with negative lateral margins and accurate delineation. The high confidence of endoscopists that the margins were well delineated before ESD was reflected in the pathologic results after the procedure. Hence, the delineation score reflects the demarcation accuracy relatively accurately.
This study had some limitations. First, we compared the two groups without the use of a magnifying endoscope. Because the gastric lumen is wide and the light source of NBI is weak, the use of NBI without magnifying endoscopy when observing the microsurface structure of the stomach may result in poor diagnostic accuracy due to the dark resolution.22 Therefore, NBI is usually observed using magnification. Images obtained by I-SCAN-OE are brighter than those obtained by NBI. I-SCAN-OE has the ability to determine gastric epithelial neoplasm demarcation without magnification.21 However, the application of a magnifying endoscope could have influenced the results. Second, we targeted four margins of the lesions. The superiority or inferiority of I-SCAN-OE over chromoendoscopy for the entire margin remains unknown. Despite these limitations, this study is the first to directly compare I-SCAN-OE and chromoendoscopy in evaluating the horizontal margins of gastric epithelial neoplasms, and the accuracy of gastric epithelial neoplasm demarcation was scored using a unique method. Although we targeted the four edges of the lesion, if the pathologist judged that there were residual lesions on any margins, it was judged as incomplete resection regardless of the delineation score. Third, the confidence level of lesion boundary evaluation was defined as a subjective index for the judgment of the endoscopist. However, when performing ESD, there is no alternative method because the decision of the boundary is determined by the endoscopist performing the procedure. Fourth, the endoscopists were unable to blind the two groups owing to the nature of the study. However, he declared that he did not participate in data collection or statistical analysis and did not affect the research results.
In conclusion, this study did not reveal the superiority of I-SCAN-OE over chromoendoscopy for delineating the margins of gastric epithelial neoplasms. However, because these results show that I-SCAN-OE and chromoendoscopy have clinically equivalent diagnostic capabilities, they support the use of I-SCAN-OE instead of chromoendoscopy to describe the edges of gastric epithelial neoplasms.
This study was supported by the Dong-A University Research Fund.
No potential conflict of interest relevant to this article was reported.
Study concept and design: J.S.J. Data acquisition: all authors. Data analysis and interpretation: J.Y.L. Drafting of the manuscript: M.K. Critical revision of the manuscript for important intellectual content: M.K. Statistical analysis: J.Y.L. Obtained funding: J.S.J. Administrative, technical, or material support; study supervision: J.S.J. Approval of final manuscript: all authors.
Gut and Liver 2023; 17(2): 234-242
Published online March 15, 2023 https://doi.org/10.5009/gnl220025
Copyright © Gut and Liver.
Myeongseok Koh1 , Jong Yoon Lee1
, Song-Hee Han2
, Seong Woo Jeon3
, Su Jin Kim4
, Joo Young Cho5
, Seong Hwan Kim5
, Jae Young Jang6
, Gwang Ho Baik7
, Jin Seok Jang1
Departments of 1Internal Medicine and 2Pathology, Dong-A University College of Medicine, Busan, 3Department of Internal Medicine, Kyungpook National University Chilgok Hospital, School of Medicine, Kyungpook National University, Daegu, 4Department of Internal Medicine, Pusan National University Yangsan Hospital, Pusan National University College of Medicine, Yangsan, 5Department of Internal Medicine, CHA Gangnam Medical Center, CHA University, 6Department of Internal Medicine, College of Medicine, Kyung Hee University, Seoul, and 7Department of Internal Medicine, Hallym University College of Medicine, Chuncheon, Korea
Correspondence to:Jin Seok Jang
ORCID https://orcid.org/0000-0001-8067-4598
E-mail jsjang@dau.ac.kr
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
Background/Aims: Endoscopic submucosal dissection is a widely used treatment for gastric epithelial neoplasms. Accurate delineation of the horizontal margins is necessary for the complete resection of gastric epithelial neoplasms. Recently, image-enhanced endoscopy has been used to evaluate horizontal margins of gastric epithelial neoplasms. The aim of this study was to investigate whether I-SCAN-optical enhancement (I-SCAN-OE) is superior to chromoendoscopy in evaluating the horizontal margin of gastric epithelial neoplasms.
Methods: This was a multicenter, prospective, and randomized trial. The participants were divided into two groups: I-SCAN-OE and chromoendoscopy. For both groups, we first evaluated the horizontal margins of early gastric cancer or high-grade dysplasia using white-light imaging, and then evaluated, the horizontal margins using I-SCAN-OE or chromoendoscopy. We devised a unique scoring method based on the pathological results obtained after endoscopic submucosal dissection to accurately evaluate the horizontal margins of gastric epithelial neoplasms. The delineation scores of both groups were compared, as were the ratios of positive/negative horizontal margins.
Results: In total, 124 patients were evaluated for gastric epithelial neoplasms, of whom 112 were enrolled in the study. A total of 112 patients participated in the study, and 56 were assigned to each group (1:1). There was no statistically significant difference in the delineation scores between the groups (chromoendoscopy, 7.80±1.94; I-SCAN-OE, 8.23±2.24; p=0.342).
Conclusions: I-SCAN-OE did not show superiority over chromoendoscopy in delineating horizontal margins of gastric epithelial neoplasms.
Keywords: Chromoendoscopy, Gastric cancer, Endoscopic submucosal dissection, Gastric neoplasm, Image-enhanced endoscopy
Endoscopic submucosal dissection (ESD) is a widely accepted treatment for early gastric cancer (EGC) or gastric adenoma. Accurate delineation of the horizontal margins is necessary for the complete resection of gastric epithelial neoplasms.1,2 Chromoendoscopy, which has been used in combination with indigo carmine and acetic acid, is considered a standard technique for delineation. This technique improves the diagnostic value of delineating the horizontal margins of EGCs or gastric adenomas.3-5 However, recently, image-enhanced endoscopy has replaced chromoendoscopy and is being used to evaluate the horizontal margins of EGCs or gastric adenomas.
Image-enhanced endoscopy has been developed and widely used clinically in recent years, including narrow-band imaging (NBI; Olympus, Tokyo, Japan), blue laser imaging (BLI; Fujifilm, Tokyo, Japan), and I-SCAN (Pentax, Tokyo, Japan). NBI is an optical digital imaging technique wherein two selected wavelengths (415±30 nm and 540±30 nm) enhance the structural aspects of the surface of the mucosa and existing vessels,6 indicating its utility in the diagnosis of EGC and delineation.7-10 BLI includes two types of lasers with wavelengths of 410 and 450 nm, similar to NBI, used in the diagnosis of EGC and the evaluation of margins.11 I-SCAN involves contrast enhancement, surface enhancement, and tone enhancement, and optical enhancement (OE) has recently been added since 2016. The I-SCAN-OE function employs band-limited light to achieve high overall transmittance by connecting the peaks of the hemoglobin absorption spectrum (415, 540, and 570 nm), thus creating a continuous wavelength spectrum.12,13 This is similar to the previously introduced NBI or BLI, but since it was recently introduced, there have been few studies on this technique. Therefore, this study aimed to determine whether I-SCAN-OE is superior to chromoendoscopy in evaluating the horizontal margins of EGCs or gastric adenomas.
This was a multicenter, prospective, randomized trial. The participants were divided into two groups: I-SCAN-OE and chromoendoscopy. For both groups, we first evaluated the horizontal margins of the gastric epithelial neoplasms with white-light imaging, and then evaluated, the horizontal margins of gastric epithelial neoplasms for each group using I-SCAN-OE or chromoendoscopy (Fig. 1). We devised a unique scoring method based on pathological results obtained after ESD to accurately evaluate the horizontal margins of gastric epithelial neoplasms. The delineation scores of the I-SCAN-OE and chromoendoscopy groups and the ratios of horizontal margin resection were compared. This study was conducted in three Korean institutions that are tertiary hospitals with extensive ESD experience.
Patients aged ≥20 years who underwent ESD for the treatment of category 4 (mucosal high-grade dysplasia or intraepithelial carcinoma) disease according to revised Vienna classification confirmed by forceps biopsy before ESD were included.14 The exclusion criteria were as follows: (1) refusal to provide informed consent; (2) category 1 or 2 confirmed in the specimen after ESD; (3) remnant stomach; (4) low cardiopulmonary function with high risk of ESD; and (5) withdrawal of consent after study enrollment. After enrollment, the participants were allocated to either the I-SCAN-OE or chromoendoscopy group in a 1:1 ratio using a simple randomization method.
We evaluated whether the horizontal margins were visually well delineated before the procedure. The chromoendoscopy group was evaluated using white-light imaging, and the confidence level was evaluated again after spraying with 20 mL of 0.2% indigo carmine solution. The I-SCAN-OE group was evaluated using white-light imaging, and the confidence level was evaluated using I-SCAN-OE mode.
Before performing ESD, the horizontal margins were evaluated in each group, and a dot was marked 5 mm outside the margin. As the thick tip of the dual knife (Olympus Medical Corp., Tokyo, Japan) used during the procedure was 5 mm in size, the marking dot was positioned using this (Fig. 2). The tumor was then cut along the marked dot (Fig. 3). After resection, the resected specimen was marked with margins of lateral and vertical sides with different colors and sectioned perpendicularly at 2 mm intervals to examine the exact margin state and the distance of the safety margin. Pathological evaluation measured the length of the nearest resection margin with a pen on the hematoxylin and eosin-stained slide of the cancerous lesion (Fig. 4). The lateral margin of the pathological specimen was evaluated as proximal, distal, anterior, or posterior and scored. If the horizontal margin was close to 5 mm (4–6 mm), it was considered “Excellent” and scored 3 points. If the horizontal margin was ≥2 mm, <4 mm or >6 mm, and ≤8 mm, it was judged as “Good” and scored 2 points; and if the horizontal margin was >0 mm, <2 mm or >8 mm, and ≤10 mm), it was judged as “Fair” and scored 1 point. If the margin was more than 10 mm away or the lateral margin was in contact with the resected specimen, it was judged as “Poor” and scored 0. Each of the four margins was assigned a score from 0 to 12, and the sum of the scores was called the delineation score (Fig. 5). Accurate delineation was defined as a total score of 8 points or more and 1 point or more obtained from each of the four margins. Inaccurate delineation was defined as a total score of 7 points or less, or including 0 points, even on one of the four margins. If the neoplasm was resected completely, including the horizontal margins, it was defined as curative resection; if the margins were not completely resected after ESD, incomplete resection was considered.
Upper gastrointestinal endoscopes with an I-SCAN-OE function were used in this study. All endoscopists used an EPK-i 7010 video processor with a Pentax EG-10-Zi endoscope (Hoya Corp., Tokyo, Japan). The endoscopists who performed the procedure in this study were all gastroenterologists, and they had performed endoscopy for at least 5 years at a gastrointestinal endoscopy training institution under the Korean Society of Gastrointestinal Endoscopy, with at least 4 years and at least 100 cases. Four endoscopists from three institutions participated in this study. Two were classified as less experienced endoscopists with less than 10 years of experience in ESD. The other two were endoscopists with more than 10 years of experience.
Before evaluation with chromoendoscopy or I-SCAN-OE, the endoscopists determined whether the horizontal margins were well demarcated using only white-light imaging and evaluated the confidence level. The confidence level was defined as whether the boundary was clearly demarcated visually by the endoscopist to determine the boundary of the lesion.
The primary endpoint was the delineation score according to group. The secondary endpoints were positive lateral margins according to groups, and the clinical factors associated with accurate delineation and positive lateral margins were determined.
The participants were divided into two groups and assigned in a 1:1 ratio to the chromoendoscopy and I-SCAN-OE groups. Continuous data are presented as mean±standard deviation and were analyzed using the Student t-test. Categorical data were analyzed using the Pearson chi-square or Fisher exact tests. A logistic regression model was used for multivariate analysis to assess the associations between clinical factors and accurate delineation of gastric epithelial neoplasms. Variables found to be significantly associated with the univariate analysis were used as co-variables in the multivariate analysis. All analyses were performed using SPSS software (version 26.0; IBM Corp., Armonk, NY, USA). Statistical significance was set at p<0.05.
The primary endpoint of this study was the delineation score. We used a unique method for evaluating the horizontal margins of gastric epithelial neoplasms, which has not been previously reported. Therefore, it was not possible to measure sample size by proposing a hypothesis based on previous studies. We analyzed 10 samples at the beginning of the study and confirmed that the delineation score was 8.5 (standard deviation, 1.6). The difference between the scores of the two groups was defined as superiority and inferiority of more than 10%, and the sample sizes to verify this were measured at an alpha level of 0.05, and a beta level of 0.80. In a
Our research protocol was approved by the Ethics Committee and was in accordance with international agreements (World Medical Association Declaration of Helsinki: Ethical principles for medical research involving human subjects). Informed consent was obtained from all participants. The study protocol was reviewed and approved by the Institutional Review Board of Dong-A University College of Medicine (IRB number: DAUHIRB-19-063) on March 27, 2019. This trial was registered with the Clinical Trials Registry (Clinical Research Information Service in South Korea, CRIS) (number KCT0003819). All authors had access to the study data and reviewed and approved the final manuscript. Research data supporting this publication are available from https://cris.nih.go.kr/cris/search/detailSearch.do/19003.
To assign participants, 124 patients were evaluated for EGC and HGD, of which 112 patients were enrolled in the study, from March 27, 2019, to March 31, 2021. A total of 112 patients participated in the study, and 56 were assigned to each group. EGC or HGD was confirmed by forceps biopsy before the procedure, but four patients were excluded because they showed negative or indefinite results. Eight patients were excluded due to withdrawal of consent (Fig. 6). There were no significant differences in demographic and clinical characteristics between the two groups (Table 1).
Table 1 . Demographic and Clinical Characteristics of the Participants According to Group.
Characteristics | Chromoendoscopy (n=56) | I-SCAN-OE (n=56) | p-value |
---|---|---|---|
Age, yr | 66.00±8.56 | 67.93±9.59 | 0.264 |
Sex | 0.154 | ||
Male | 48 (85.71) | 42 (75.00) | |
Female | 8 (14.29) | 14 (25.00) | |
Tumor size, mm | 12.92±7.60 | 15.09±8.48 | 0.157 |
Procedure time, min | 18.61±9.52 | 17.88±10.24 | 0.696 |
Location | 0.209 | ||
Antrum | 37 (66.07) | 43 (76.79) | |
Body and fundus | 19 (33.93) | 13 (23.21) | |
Macroscopic type | 0.958 | ||
Elevated | 21 (37.50) | 23 (41.07) | |
Flat | 10 (17.86) | 8 (14.29) | |
Depressed | 20 (35.71) | 20 (35.71) | |
Mixed | 5 (8.93) | 5 (8.93) | |
1.000 | |||
Positive | 19 (33.93) | 19 (33.93) | |
Negative | 37 (66.07) | 37 (66.07) | |
Final pathology | 0.238 | ||
HGD | 12 (21.43) | 6 (10.71) | |
EGC (differentiated) | 42 (75.00) | 49 (87.50) | |
EGC (undifferentiated) | 2 (3.57) | 1 (1.79) | |
Invasion depth | 0.768 | ||
HGD and T1a | 50 (89.29) | 49 (87.50) | |
T1b and deeper | 6 (10.71) | 7 (12.50) | |
Hospital | 0.845 | ||
A | 40 (71.43) | 38 (67.86) | |
B | 10 (17.86) | 10 (17.86) | |
C | 6 (10.71) | 8 (14.29) | |
Endoscopists | 0.324 | ||
Experienced | 48 (85.71) | 44 (78.57) | |
Less experienced | 8 (14.29) | 12 (21.43) | |
0.647 | |||
Yes | 54 (96.43) | 53 (94.64) | |
Piecemeal resection | 2 (3.57) | 3 (5.36) | |
Anticoagulation or antiplatelet use | 0.357 | ||
Yes | 10 (17.86) | 14 (25.00) | |
No | 46 (82.14) | 42 (75.00) |
Data are presented as mean±SD or number (%)..
I-SCAN-OE, I-SCAN-optical enhancement; HGD, high-grade dysplasia; EGC, early gastric cancer..
Before evaluation with chromoendoscopy or I-SCAN-OE, there was no statistically significant difference between the groups (chromoendoscopy, 55.36%; I-SCAN-OE, 48.21%; p=0.449). When the margins were evaluated with chromoendoscopy or I-SCAN-OE, and the confidence level was evaluated again, it was found to have increased in the I-SCAN-OE group, but this was not statistically significant (chromoendoscopy, 55.36%; I-SCAN-OE, 71.43%; p=0.078) (Table 2).
Table 2 . Confidence Level of Delineation as Assessed before Endoscopic Submucosal Dissection According to Group.
Confidence level | Chromoendoscopy (n=56) | I-SCAN-OE (n=56) | p-value |
---|---|---|---|
Before enhancing | 0.449 | ||
Low | 25 (44.64) | 29 (51.79) | |
High | 31 (55.36) | 27 (48.21) | |
After enhancing | 0.078 | ||
Low | 25 (44.64) | 16 (28.57) | |
High | 31 (55.36) | 40 (71.43) |
Data are presented as the number (%)..
I-SCAN-OE, I-SCAN-optical enhancement..
There was no statistically significant difference in the delineation score between the groups (chromoendoscopy, 7.80±1.94; I-SCAN-OE, 8.23±2.24; p=0.342). The ratio of the negative lateral margin was 91.07%, which was the same in both groups. The ratio of accurate delineation was higher in the I-SCAN-OE group than in the chromoendoscopy group; however, the difference was not statistically significant (chromoendoscopy, 50.00%; I-SCAN-OE, 60.71%; p=0.254) (Table 3).
Table 3 . Delineation Score, Curative Resection Rate and Accuracy According to Group.
Variable | Chromoendoscopy (n=56) | I-SCAN-OE (n=56) | p-value |
---|---|---|---|
Delineation score, mean±SD | 7.80±1.94 | 8.23±2.24 | 0.342 |
Negative lateral margin, % | 91.07 | 91.07 | 1.000 |
Yes:No | 51:5 | 51:5 | |
Accuracy, % | 50.00 | 60.71 | 0.254 |
Accurate:Inaccurate | 28:28 | 34:22 |
I-SCAN-OE, I-SCAN-optical enhancement..
There were 62 accurate and 50 inaccurate delineations. Among the clinical features, tumor size (accurate delineation, 12.42±7.18 mm; inaccurate delineation, 15.69±8.81 mm; p=0.020), procedure time (accurate delineation, 15.69±8.67 minutes; inaccurate delineation, 21.40±10.27 minutes; p=0.002), and confidence level (accurate delineation, 74.19%; inaccurate delineation, 50.00%; p=0.008) showed statistically significant differences between accurate and inaccurate delineations. Accurate delineation was associated with smaller tumor size, shorter procedure time, and higher confidence level than inaccurate delineation (Table 4).
Table 4 . Demographic and Clinical Characteristics According to Accuracy.
Characteristics | Accurate delineation (n=62) | Inaccurate delineation (n=50) | p-value |
---|---|---|---|
Group | 0.254 | ||
Chromoendoscopy | 28 (45.16) | 28 (56.00) | |
I-SCAN-OE | 34 (54.84) | 22 (44.00) | |
Age, yr | 65.90±8.67 | 68.28±9.53 | 0.170 |
Sex | 0.694 | ||
Male | 49 (79.03) | 41 (82.00) | |
Female | 13 (20.97) | 9 (18.00) | |
Tumor size, mm | 12.42±7.18 | 15.69±8.81 | 0.020* |
Procedure time, min | 15.69±8.67 | 21.40±10.27 | 0.002* |
Location | 0.118 | ||
Antrum | 48 (77.42) | 32 (64.00) | |
Body and fundus | 14 (22.58) | 18 (36.00) | |
Macroscopic type | 0.261 | ||
Elevated | 26 (41.94) | 18 (36.00) | |
Flat | 7 (11.29) | 11 (22.00) | |
Depressed | 25 (40.32) | 15 (30.00) | |
Mixed | 4 (6.45) | 6 (12.00) | |
0.414 | |||
Positive | 19 (30.65) | 19 (38.00) | |
Negative | 43 (69.35) | 31 (62.00) | |
Final pathology | 0.058 | ||
HGD | 7 (11.29) | 11 (22.00) | |
EGC (differentiated) | 55 (88.70) | 36 (72.00) | |
EGC (undifferentiated) | 3 (6.00) | ||
Invasion depth | 0.058 | ||
HGD and T1a | 58 (93.55) | 41 (82.00) | |
T1b and deeper | 4 (6.45) | 9 (18.00) | |
Confidence level (after enhancing) | 0.008* | ||
Low | 16 (25.80) | 25 (50.00) | |
High | 46 (74.19) | 25 (50.00) | |
Endoscopists | 0.043* | ||
Experienced | 55 (88.71) | 37 (74.00) | |
Less experienced | 7 (11.29) | 13 (26.00) | |
0.170 | |||
Yes | 61 (98.39) | 46 (92.00) | |
Piecemeal resection | 1 (1.61) | 4 (8.00) | |
Anticoagulation or antiplatelet use | 0.128 | ||
Yes | 10 (16.13) | 14 (28.00) | |
No | 52 (83.87) | 36 (72.00) |
Data are presented as number (%) or mean±SD..
I-SCAN-OE, I-SCAN-optical enhancement; HGD, high-grade dysplasia; EGC, early gastric cancer..
*p<0.05..
Table 1 shows that the procedure time, tumor size, macroscopic features, and confidence level were statistically significant factors related to positive lateral margins in the univariate logistic regression analysis. Long procedure duration, large and flat tumors, and low confidence level were related to positive lateral margins. Confidence level was identified as a statistically significant factor when these factors were analyzed using multivariate analysis as covariates (odds ratio [OR], 0.221; p=0.044) (Table 5). Procedure time (OR, 1.082; p=0.005), tumor size (OR, 1.092; p=0.024), confidence level (OR, 0.214; p=0.033), and endoscopist experience were statistically significant factors related to accurate delineation in univariate logistic regression analysis. A short procedure time, small tumor size, high confidence level, and extensive endoscopy experience were strongly associated with accurate delineation (Table 6). When these factors were analyzed by multivariate analysis as covariates, confidence level and endoscopist experience were identified as statistically significant factors.
Table 5 . Univariate and Multivariate Analyses for Factors Related to Positive Lateral Margin.
Factor | Univariate analysis | Multivariate analysis | |||
---|---|---|---|---|---|
OR (95% CI) | p-value | OR (95% CI) | p-value | ||
Procedure time | 1.082 (1.025–1.142) | 0.005* | 1.055 (0.990–1.124) | 0.097 | |
Tumor size | 1.092 (1.014–1.175) | 0.024* | 1.082 (0.989–1.183) | 0.084 | |
Macroscopic | 0.043* | 0.080 | |||
Elevated/depressed | 1 | 1 | |||
Flat | 4.190 (1.049–16.743) | 4.220 (0.842–21.152) | |||
Confidence level (after enhancing) | 0.033* | 0.044* | |||
Low | 1 | 1 | |||
High | 0.214 (0.052–0.881) | 0.221 (0.044–1.110) |
OR, odds ratio; CI, confidence interval..
*p<0.05..
Table 6 . Univariate and Multivariate Analyses for Factors Related to Accurate Delineation.
Factor | Univariate analysis | Multivariate analysis | |||
---|---|---|---|---|---|
OR (95% CI) | p-value | OR (95% CI) | p-value | ||
Procedure time, min | 0.935 (0.893–0.979) | 0.004* | 0.956 (0.912–1.003) | 0.064 | |
Tumor size | 0.944 (0.899–0.993) | 0.024* | 0.969 (0.917–1.024) | 0.261 | |
Confidence level (after enhancing) | 0.009* | 0.006* | |||
Low | 1 | 1 | |||
High | 2.875 (1.299–6.362) | 3.543 (1.431–8.774) | |||
Endoscopists | 0.049* | 0.023* | |||
Experienced | 1 | 1 | |||
Less experienced | 0.362 (0.132–0.993) | 0.265 (0.084–0.833) |
OR, odds ratio; CI, confidence interval..
*p<0.05..
Chromoendoscopy aims to increase the rate of visualization and detection of neoplastic lesions and has been applied throughout the gastrointestinal tract.3 In contrast, conventional white-light endoscopy has a miss rate of up to 25% for gastrointestinal pathology, specifically in the context of small and flat lesions.15 Although chromoendoscopy helps in the early diagnosis of neoplastic lesions, it takes time to disperse and aspirate, and has the disadvantage of difficulty in identifying the lesion if inflammation is severe.16
This was a multicenter, prospective, randomized trial aimed at determining whether I-SCAN-OE is superior to chromoendoscopy in delineating gastric epithelial neoplasms according to delineation score. Several previous studies have used NBI to evaluate the horizontal margin of EGCs. The accuracy of diagnosis of the cancer margin in EGC in addition to magnifying NBI was high when the margins were unclear using chromoendoscopy alone.9 in the comparison of chromoendoscopy and NBI, NBI has been found to be superior to the former,17,18 but other studies have shown both to be clinically similar.10 Another image-enhanced endoscopy, BLI, used in the diagnosis of EGC and the evaluation of margins, have been reported to have clinical utility similar to that of NBI.11,19,20
Our study comparatively analyzed the efficacy of I-SCAN-OE and chromoendoscopy. I-SCAN-OE uses white-light imaging components in addition to NBI at wavelengths of 415, 540, and 570 nm. Only one previous study has used I-SCAN-OE to confirm the diagnostic accuracy of cancer demarcation in EGC. Nagao
In the present study, the primary endpoint was delineation score. We devised this method to evaluate the diagnostic accuracy of demarcation of gastric epithelial neoplasms. There was no statistically significant difference between I-SCAN-OE (8.23±2.24) and chromoendoscopy (7.80±1.94). In addition, the ratio of positive lateral margins was 8.93%, which was similar in both groups. Thus, our data did not reveal any differences between I-SCAN-OE and chromoendoscopy in terms of their ability to accurately delineate the horizontal margins of gastric epithelial neoplasms. The accurate delineation group had a smaller tumor size, shorter procedure time, higher confidence level in pre-procedure delineation evaluation, and endoscopists with more experience than the inaccurate delineation group. The other factors were not statistically significant. Multivariate analysis indicated that a high confidence level was significantly associated with negative lateral margins and accurate delineation. The high confidence of endoscopists that the margins were well delineated before ESD was reflected in the pathologic results after the procedure. Hence, the delineation score reflects the demarcation accuracy relatively accurately.
This study had some limitations. First, we compared the two groups without the use of a magnifying endoscope. Because the gastric lumen is wide and the light source of NBI is weak, the use of NBI without magnifying endoscopy when observing the microsurface structure of the stomach may result in poor diagnostic accuracy due to the dark resolution.22 Therefore, NBI is usually observed using magnification. Images obtained by I-SCAN-OE are brighter than those obtained by NBI. I-SCAN-OE has the ability to determine gastric epithelial neoplasm demarcation without magnification.21 However, the application of a magnifying endoscope could have influenced the results. Second, we targeted four margins of the lesions. The superiority or inferiority of I-SCAN-OE over chromoendoscopy for the entire margin remains unknown. Despite these limitations, this study is the first to directly compare I-SCAN-OE and chromoendoscopy in evaluating the horizontal margins of gastric epithelial neoplasms, and the accuracy of gastric epithelial neoplasm demarcation was scored using a unique method. Although we targeted the four edges of the lesion, if the pathologist judged that there were residual lesions on any margins, it was judged as incomplete resection regardless of the delineation score. Third, the confidence level of lesion boundary evaluation was defined as a subjective index for the judgment of the endoscopist. However, when performing ESD, there is no alternative method because the decision of the boundary is determined by the endoscopist performing the procedure. Fourth, the endoscopists were unable to blind the two groups owing to the nature of the study. However, he declared that he did not participate in data collection or statistical analysis and did not affect the research results.
In conclusion, this study did not reveal the superiority of I-SCAN-OE over chromoendoscopy for delineating the margins of gastric epithelial neoplasms. However, because these results show that I-SCAN-OE and chromoendoscopy have clinically equivalent diagnostic capabilities, they support the use of I-SCAN-OE instead of chromoendoscopy to describe the edges of gastric epithelial neoplasms.
This study was supported by the Dong-A University Research Fund.
No potential conflict of interest relevant to this article was reported.
Study concept and design: J.S.J. Data acquisition: all authors. Data analysis and interpretation: J.Y.L. Drafting of the manuscript: M.K. Critical revision of the manuscript for important intellectual content: M.K. Statistical analysis: J.Y.L. Obtained funding: J.S.J. Administrative, technical, or material support; study supervision: J.S.J. Approval of final manuscript: all authors.
Table 1 Demographic and Clinical Characteristics of the Participants According to Group
Characteristics | Chromoendoscopy (n=56) | I-SCAN-OE (n=56) | p-value |
---|---|---|---|
Age, yr | 66.00±8.56 | 67.93±9.59 | 0.264 |
Sex | 0.154 | ||
Male | 48 (85.71) | 42 (75.00) | |
Female | 8 (14.29) | 14 (25.00) | |
Tumor size, mm | 12.92±7.60 | 15.09±8.48 | 0.157 |
Procedure time, min | 18.61±9.52 | 17.88±10.24 | 0.696 |
Location | 0.209 | ||
Antrum | 37 (66.07) | 43 (76.79) | |
Body and fundus | 19 (33.93) | 13 (23.21) | |
Macroscopic type | 0.958 | ||
Elevated | 21 (37.50) | 23 (41.07) | |
Flat | 10 (17.86) | 8 (14.29) | |
Depressed | 20 (35.71) | 20 (35.71) | |
Mixed | 5 (8.93) | 5 (8.93) | |
1.000 | |||
Positive | 19 (33.93) | 19 (33.93) | |
Negative | 37 (66.07) | 37 (66.07) | |
Final pathology | 0.238 | ||
HGD | 12 (21.43) | 6 (10.71) | |
EGC (differentiated) | 42 (75.00) | 49 (87.50) | |
EGC (undifferentiated) | 2 (3.57) | 1 (1.79) | |
Invasion depth | 0.768 | ||
HGD and T1a | 50 (89.29) | 49 (87.50) | |
T1b and deeper | 6 (10.71) | 7 (12.50) | |
Hospital | 0.845 | ||
A | 40 (71.43) | 38 (67.86) | |
B | 10 (17.86) | 10 (17.86) | |
C | 6 (10.71) | 8 (14.29) | |
Endoscopists | 0.324 | ||
Experienced | 48 (85.71) | 44 (78.57) | |
Less experienced | 8 (14.29) | 12 (21.43) | |
0.647 | |||
Yes | 54 (96.43) | 53 (94.64) | |
Piecemeal resection | 2 (3.57) | 3 (5.36) | |
Anticoagulation or antiplatelet use | 0.357 | ||
Yes | 10 (17.86) | 14 (25.00) | |
No | 46 (82.14) | 42 (75.00) |
Data are presented as mean±SD or number (%).
I-SCAN-OE, I-SCAN-optical enhancement; HGD, high-grade dysplasia; EGC, early gastric cancer.
Table 2 Confidence Level of Delineation as Assessed before Endoscopic Submucosal Dissection According to Group
Confidence level | Chromoendoscopy (n=56) | I-SCAN-OE (n=56) | p-value |
---|---|---|---|
Before enhancing | 0.449 | ||
Low | 25 (44.64) | 29 (51.79) | |
High | 31 (55.36) | 27 (48.21) | |
After enhancing | 0.078 | ||
Low | 25 (44.64) | 16 (28.57) | |
High | 31 (55.36) | 40 (71.43) |
Data are presented as the number (%).
I-SCAN-OE, I-SCAN-optical enhancement.
Table 3 Delineation Score, Curative Resection Rate and Accuracy According to Group
Variable | Chromoendoscopy (n=56) | I-SCAN-OE (n=56) | p-value |
---|---|---|---|
Delineation score, mean±SD | 7.80±1.94 | 8.23±2.24 | 0.342 |
Negative lateral margin, % | 91.07 | 91.07 | 1.000 |
Yes:No | 51:5 | 51:5 | |
Accuracy, % | 50.00 | 60.71 | 0.254 |
Accurate:Inaccurate | 28:28 | 34:22 |
I-SCAN-OE, I-SCAN-optical enhancement.
Table 4 Demographic and Clinical Characteristics According to Accuracy
Characteristics | Accurate delineation (n=62) | Inaccurate delineation (n=50) | p-value |
---|---|---|---|
Group | 0.254 | ||
Chromoendoscopy | 28 (45.16) | 28 (56.00) | |
I-SCAN-OE | 34 (54.84) | 22 (44.00) | |
Age, yr | 65.90±8.67 | 68.28±9.53 | 0.170 |
Sex | 0.694 | ||
Male | 49 (79.03) | 41 (82.00) | |
Female | 13 (20.97) | 9 (18.00) | |
Tumor size, mm | 12.42±7.18 | 15.69±8.81 | 0.020* |
Procedure time, min | 15.69±8.67 | 21.40±10.27 | 0.002* |
Location | 0.118 | ||
Antrum | 48 (77.42) | 32 (64.00) | |
Body and fundus | 14 (22.58) | 18 (36.00) | |
Macroscopic type | 0.261 | ||
Elevated | 26 (41.94) | 18 (36.00) | |
Flat | 7 (11.29) | 11 (22.00) | |
Depressed | 25 (40.32) | 15 (30.00) | |
Mixed | 4 (6.45) | 6 (12.00) | |
0.414 | |||
Positive | 19 (30.65) | 19 (38.00) | |
Negative | 43 (69.35) | 31 (62.00) | |
Final pathology | 0.058 | ||
HGD | 7 (11.29) | 11 (22.00) | |
EGC (differentiated) | 55 (88.70) | 36 (72.00) | |
EGC (undifferentiated) | 3 (6.00) | ||
Invasion depth | 0.058 | ||
HGD and T1a | 58 (93.55) | 41 (82.00) | |
T1b and deeper | 4 (6.45) | 9 (18.00) | |
Confidence level (after enhancing) | 0.008* | ||
Low | 16 (25.80) | 25 (50.00) | |
High | 46 (74.19) | 25 (50.00) | |
Endoscopists | 0.043* | ||
Experienced | 55 (88.71) | 37 (74.00) | |
Less experienced | 7 (11.29) | 13 (26.00) | |
0.170 | |||
Yes | 61 (98.39) | 46 (92.00) | |
Piecemeal resection | 1 (1.61) | 4 (8.00) | |
Anticoagulation or antiplatelet use | 0.128 | ||
Yes | 10 (16.13) | 14 (28.00) | |
No | 52 (83.87) | 36 (72.00) |
Data are presented as number (%) or mean±SD.
I-SCAN-OE, I-SCAN-optical enhancement; HGD, high-grade dysplasia; EGC, early gastric cancer.
*p<0.05.
Table 5 Univariate and Multivariate Analyses for Factors Related to Positive Lateral Margin
Factor | Univariate analysis | Multivariate analysis | |||
---|---|---|---|---|---|
OR (95% CI) | p-value | OR (95% CI) | p-value | ||
Procedure time | 1.082 (1.025–1.142) | 0.005* | 1.055 (0.990–1.124) | 0.097 | |
Tumor size | 1.092 (1.014–1.175) | 0.024* | 1.082 (0.989–1.183) | 0.084 | |
Macroscopic | 0.043* | 0.080 | |||
Elevated/depressed | 1 | 1 | |||
Flat | 4.190 (1.049–16.743) | 4.220 (0.842–21.152) | |||
Confidence level (after enhancing) | 0.033* | 0.044* | |||
Low | 1 | 1 | |||
High | 0.214 (0.052–0.881) | 0.221 (0.044–1.110) |
OR, odds ratio; CI, confidence interval.
*p<0.05.
Table 6 Univariate and Multivariate Analyses for Factors Related to Accurate Delineation
Factor | Univariate analysis | Multivariate analysis | |||
---|---|---|---|---|---|
OR (95% CI) | p-value | OR (95% CI) | p-value | ||
Procedure time, min | 0.935 (0.893–0.979) | 0.004* | 0.956 (0.912–1.003) | 0.064 | |
Tumor size | 0.944 (0.899–0.993) | 0.024* | 0.969 (0.917–1.024) | 0.261 | |
Confidence level (after enhancing) | 0.009* | 0.006* | |||
Low | 1 | 1 | |||
High | 2.875 (1.299–6.362) | 3.543 (1.431–8.774) | |||
Endoscopists | 0.049* | 0.023* | |||
Experienced | 1 | 1 | |||
Less experienced | 0.362 (0.132–0.993) | 0.265 (0.084–0.833) |
OR, odds ratio; CI, confidence interval.
*p<0.05.