Optimal endoscopic resection method based on vertical margin distance for small rectal neuroendocrine tumors: Propensity score-matched study

Keywords

Endoscopy Lower GI Tract - Polyps / adenomas / ... - Endoscopic resection (polypectomy, ESD, EMRc, ...) - CRC screening

Introduction

Rectal neuroendocrine tumors (NETs) are increasingly detected during screening colonoscopy [1]. Although these tumors are typically indolent and low-grade, they possess metastasis potential, necessitating prompt and accurate treatment [2] [3]. Current guidelines and consensus uniformly endorse endoscopic resection as first-line therapy for rectal NETs measuring less than 10 mm in diameter [4] [5] [6]. A recent multicenter prospective study in Japan demonstrated that endoscopic mucosal resection with ligation (EMR-L) and endoscopic submucosal dissection (ESD) are the predominant techniques used [7]. However, lack of a defined optimal endoscopic treatment technique in these guidelines highlights the need for comparative effectiveness research to determine the best approach.

Rectal NETs, which originate in the mucosal layer and have a tendency for submucosal invasion, necessitate strict vertical margin control to minimize incomplete resection and local recurrence risks. Although recent meta-analyses report comparable R0 resection and positive vertical margin rates between EMR-L and ESD [8], this consensus is challenged by an earlier meta-analysis highlighting EMR-L's superior R0 achievement [9]. Furthermore, ESD exhibits marked procedural heterogeneity, evidenced by widely variable R0 rates (54%-100%) across studies (Supplementary Table 1) [10] [11] [12] [13] [14] [15] [16] [17]. Crucially, current evidence predominantly stems from small-scale retrospective cohorts with inconsistent baseline controls, rendering the debate on EMR-L versus ESD for histological completeness scientifically inconclusive.

Theoretically, securing extended vertical resection margins in endoscopic management of rectal NETs could enhance histological completeness by ensuring deeper submucosal clearance. However, current evidence remains paradoxically limited and contradictory, with only three underpowered studies directly comparing EMR-L versus ESD on this critical parameter, demonstrating irreconcilable outcomes [12] [16] [17]. We propose that establishing a quantifiable vertical margin threshold could serve as a surrogate biomarker for assessing the technical adequacy of resection depth—a hypothesis that requires validation through rigorously controlled studies. To clarify the ongoing debate about the superiority of EMR-L versus ESD, particularly in terms of vertical margin optimization, this multicenter cohort study employed propensity score matching (PSM) to balance differences in baseline characteristics.

Patients and methods

Study design and eligibility criteria

This dual-center retrospective cohort study was conducted at two tertiary hospitals in China, Xijing Hospital and the First Hospital of Nanchang, from January 2010 to November 2024. The study population consisted of a prospectively collected, consecutive series of patients diagnosed histologically with rectal NETs measuring up to 10 mm in diameter. Exclusion criteria for this study were as follows: rectal NETs with a diameter greater than 10 mm; cases managed with conventional EMR or ESD using a traction device; patients requiring rescue therapy following previous endoscopic treatment; specimens unsuitable for pathological margin assessment; and presence of lymph node or distant metastases. Ethical approval for the study was granted by the Institutional Review Board of Xijing Hospital and the First Hospital of Nanchang.

Interventions

All endoscopic procedures were performed by five experienced endoscopists, each having completed a minimum of 300 ESD procedures. ESD procedures were performed in strict adherence to standardized recommendations, without use of traction devices as traction might improve margin adequacy [18]. For EMR-L, the procedure was performed using an endoscope equipped with a band ligation device (Sumitomo Bakelite Co. Ltd., Tokyo, Japan) [19]. The submucosal injection solution used during ESD and EMR-L procedures was normal saline combined with indigo carmine. These procedures were performed under intravenous anesthesia using Olympus endoscopy systems, including the EVIS LUCERA CV-260, EVIS LUCERA ELITE CV-290, and EVIS X1 models.

Currently, there are no standardized protocols or specific guidelines for selecting the resection method. Primary considerations at our center include lesion characteristics such as size, morphology, depth of invasion, and location, as well as clinical factors like operative time and operator experience and discretion.

Pathological assessment

Pathological evaluation was performed by experienced pathologists at each participating institution. The resected specimens were carefully spread on a specimen plate and evaluated by expert pathologists, followed by a detailed histological examination. Immunohistochemical staining was used to confirm NETs diagnosis and grading. Lymphovascular involvement was mainly diagnosed by HE staining; immunohistochemistry was not routinely performed. The vertical margin distance was defined as the shortest distance from the nearest tumor front to the vertical resection margin [20]. A margin was classified as positive if tumor cells were present at the resection edges.

Endpoints

The primary endpoint was the rate of histologically curative (R0) resection, defined as en bloc removal of a targeted lesion with both horizontal and vertical tumor-free margins. Secondary endpoints included the proportion of cases with adequate vertical margin, rate of positive vertical margins, procedure time, complications, and length of hospital stay.

There is currently insufficient evidence establishing a clear association between vertical margin distance and patient prognosis. Given lack of a universal standard, this study exploratively set the threshold for adequate vertical margin distance at the 25th percentile (115 μm) to distinguish between adequate and inadequate vertical margins ([Fig. 1]). When vertical margin positivity was identified, the vertical margin distance was recorded as zero. Procedure time was measured from initiation of submucosal injection to completion of resection. Complications included perforation and post-procedure bleeding, which may occur either during the procedure or in a delayed manner post-procedure [21]. Perforation is identified by a complete defect in the mural tissue [22]. Delayed bleeding is characterized by a decrease in hemoglobin of more than 2 g/dL or confirmation of significant hematochezia following endoscopic treatment.

Fig. 1 a,b A 57-year-old female with a 5-mm rectal neuroendocrine tumor, located 5 cm from the anal verge, underwent endoscopic mucosal resection with ligation (EMR-L). The postoperative vertical margin was 1040 μm, meeting the adequate margin standard. c,d A 50-year-old female with a 4-mm rectal neuroendocrine tumor, located 8 cm from the anal verge, underwent endoscopic submucosal dissection (ESD). The postoperative vertical margin was 90 μm, failing to meet the adequate margin standard.

Statistical analysis

Continuous variables were assessed for normality using the Shapiro-Wilk test, supplemented by visual inspection of Q-Q plots and histograms. Variables with a Shapiro-Wilk test P ≥ 0.05 were considered normally distributed and presented as mean ± standard deviation (SD). Conversely, variables with P < 0.05 were classified as non-normally distributed and reported as median with interquartile range (IQR). PSM was employed to control for confounding variables between treatment groups. Propensity scores were calculated using a multivariable logistic regression model, with treatment assignment as the outcome variable and the following covariates included: age, tumor size, involvement of the muscularis propria, central depression, and biopsy status. A 1:1 nearest-neighbor matching method with a 1% caliper was applied based on pre-exposure characteristics. The balance of covariates between groups was evaluated using the standardized mean difference (SMD), with a smaller SMD indicating reduced between-group variation in covariates. Binary variables were analyzed using logistic regression models, whereas generalized linear models with a gamma distribution and a log link function were employed to evaluate continuous endpoints. Complication outcomes involving a count of zero were calculated using Fisher's exact test. Subgroup analyses on vertical margin distance were conducted both before and after PSM. All statistical analyses were performed using R software (version 4.2.2).

Results

Patient and lesion characteristics

A total of 204 rectal NETs from 186 patients were included in the study ([Fig. 2]). Baseline characteristics of patients are summarized in [Table 1]. Mean (SD) age of the patients was 50.7 years (10.4), with 106 (57.0%) being male. Among these patients, 15 (7.6%) presented with at least two lesions. There were 56 patients (30.1%) in the EMR-L group and 130 patients (69.9%) in the ESD group.

Fig. 2 Flowchart of the study design. EMR-L, endoscopic mucosal resection with ligation; ESD, endoscopic submucosal dissection; NET, neuroendocrine tumor.

Table 1 Baseline characteristics and follow-up information (n = 186).

Variables	Values
EMR-L, endoscopic mucosal resection with ligation; ESD, endoscopic submucosal dissection; IQR, interquartile range; rNETs, rectal neuroendocrine tumors; SD, standard deviation; WBC, white blood cell. Normal distribution is presented as mean ± SD; non-normal distribution as median (IQR). Increase in WBC count was defined as postoperative WBC count minus preoperative WBC count. For patients with multiple lesions: if any lesion is non-R0, the patient is classified as non-R0; if any lesion is G2, the patient is considered to have a G2 lesion; if any lesion has a perforation or bleeding, the patient is considered to have a perforation or bleeding.
Age (years), mean ± SD	50.7 ± 10.4
Gender, n (%)
Female	80 (43.0)
Male	106 (57.0)
Comorbidities, n (%)	58 (31.2)
Hypertension	29 (15.6)
Diabetes	11 (5.9)
Heart disease	11 (5.9)
Others	28 (15.1)
Family or personal history of cancer, n (%)	14 (7.5)
Number of rNETs per patients, n (%)
1	172 (92.5)
2	10 (5.4)
3	4 (2.2)
Endoscopic resection (ER) modality*, n (%)
EMR-L	56 (30.1)
ESD	130 (69.9)
Pathologic results
R0	172 (92.5)
WHO grade 1, n (%)	170 (91.4)
WHO grade 2, n (%)	16 (8.6)
Lymphovascular invasion, n (%)	1 (0.5)
Complications	36 (19.4)
Perforation, n (%)	17 (9.1)
Bleeding, n (%)	3 (1.6)
Increase in WBC count after ER (×109), median (IQR)	0.6 (-0.3, 1.4)
Duration of hospitalization (days), median (IQR)	5.0 (4.0, 6.0)
Follow-up
Follow-up time (months), median (IQR)	64.4 (32.2, 91.3)
Metachronous cancerous lesions, n (%)	2 (1.1)
Recurrence or metastasis, n (%)	0 (0)
Death related/unrelated to rNETs, n (%)	0 (0)/1 (0.5)
Lost to follow-up, n (%)	15 (8.1)

For the 204 lesions, after PSM, 122 matched lesions remained, with 61 in the ESD group and 61 in the EMR-L group ([Table 2]). Median follow-up duration was 64.4 months (IQR 32.2–91.3), with 15 patients lost to follow-up. During this period, two patients (1.1%) were diagnosed with metachronous rectal NETs, both of whom were treated with EMR-L. In addition, one patient (0.5%) died from causes unrelated to rectal NETs, with none of the deaths being rectal NET-related.

Table 2 Baseline characteristics of all lesions before and after PSM.

Variables	Before PSM			After PSM
	EMR-L (n = 61)	ESD (n = 143)	SMD	EMR-L (n = 61)	ESD (n = 61)	SMD
EMR-L, endoscopic mucosal resection with ligation; ESD, endoscopic submucosal dissection; MP, muscularis propria; NA, not applicable; PSM, propensity score matching; SD, standard deviation; SMD, standardized mean difference.
Age, mean ± SD	51.6 ± 10.1	50.5 ± 10.2	0.109	51.6 ± 10.1	50.5 ± 9.1	0.109
Tumor size category, n (%)
≤ 5 mm	46 (75.4)	69 (48.3)	0.631	46 (75.4)	46 (75.4)	0.000
> 5 mm	15 (24.6)	74 (51.8)	-0.631	15 (24.6)	15 (24.6)	0.000
MP involved, n (%)
No	60 (98.4)	139 (97.2)	0.091	60 (98.4)	60 (98.4)	0.000
Yes	1 (1.6)	4 (2.8)	-0.091	1 (1.6)	1 (1.6)	0.000
Central depression, n (%)
No	56 (91.8)	120 (83.9)	0.288	56 (91.8)	56 (91.8)	0.000
Yes	5 (8.2)	23 (16.1)	-0.288	5 (8.2)	5 (8.2)	0.000
Biopsy, n (%)
No	52 (85.3)	116 (81.1)	0.116	52 (85.3)	53 (86.9)	-0.046
Yes	9 (14.8)	27 (18.9)	-0.116	9 (14.8)	8 (13.1)	0.046

R0 resection rate

Following PSM, R0 resection was achieved in 91.0% of the 122 lesions (n=111). The EMR-L group demonstrated a significantly higher rate of R0 resections compared with the ESD group, with rates of 98.4% versus 83.6% (OR 11.8, 95% CI 1.5–95), consistent with results observed before PSM ([Table 3]).

Table 3 Study endpoints before and after propensity score matching.

Outcome	Phase of PSM	No. (%)/median (IQR)		OR (95% CI)/coefficient estimate	P value
		EMR-L	ESD
CI, confidence interval; ESD, endoscopic submucosal dissection; EMR-L, endoscopic mucosal resection with ligation; IQR, interquartile range; NA, not applicable; OR, odds ratio; Q1, 25th percentile (115 μm). The ESD group serves as the reference.
R0	Before	60 (98.4)	125 (87.4)	8.6 (1.1, 66.2)	0.038
	After	60 (98.4)	51 (83.6)	11.8 (1.5, 95.0)	0.021
Free vertical margin > Q1	Before	49 (80.3)	90 (62.94)	2.40 (1.2, 4.9)	0.016
	After	49 (80.3)	38 (62.3)	2.5 (1.1, 5.6)	0.030
Vertical margin distance, μm	Before	240.0 (126.0, 368.0)	177.0 (68.0, 431.5)	-21.2 (-123.3, 81.0)	0.685
	After	240.0 (126.0, 368.0)	150.0 (59.0, 404.0)	-15.6 (-137.3, 106.2)	0.803
Positive vertical margin,	Before	1 (1.6)	16 (11.2)	0.1 (0.02, 1.02)	0.052
	After	1 (1.6)	8 (13.1)	0.1 (0.01, 0.9)	0.041
Procedure time, min	Before	5.0 (4.0, 7.0)	20.0 (13.0, 30.0)	-17.3 (-21.1, -13.5)	<.001
	After	5.0 (4.0, 7.0)	19.0 (14.0, 28.0)	-17.2 (-21.0, -13.4)	<.001
Complications	Before	5 (8.2)	34 (23.8)	0.3 (0.1, 0.8)	0.014
	After	5 (8.2)	18 (29.5)	0.2 (0.07, 0.6)	0.005
Perforation	Before	2 (3.3)	15 (10.5)	0.3 (0.1, 1.3)	0.107
	After	2 (3.3)	10 (16.4)	0.2 (0.04, 0.8)	0.028
Bleeding	Before	3 (4.9)	0 (0.0)	NA	0.026
	After	3 (4.9)	0 (0.0)	NA	0.242
Duration of hospitalization, days	Before	5.0 (3.0, 5.0)	5.0 (4.0, 6.0)	-0.6 (-1.1, -0.03)	0.041
	After	5.0 (3.0, 5.0)	5.0 (3.0, 6.0)	-0.3 (-0.96, 0.3)	0.348

Vertical margin distance between groups

Median vertical margin distance for the 223 lesions, excluding 17 patients with positive vertical margins, was 220 μm (IQR 115.3 to 428.8 μm). Analyses conducted both before and after PSM revealed that the EMR-L group had a significantly higher proportion of lesions with a vertical margin distance exceeding 115 μm compared with the ESD group ([Table 3]). Specifically, after PSM, the EMR-L group showed a proportion of 80.3%, whereas the ESD group had 62.3% (OR 2.5, 95% CI 1.1–5.6) ([Table 3]). However, when vertical margins were analyzed as continuous variables, no significant differences were observed between the EMR-L and ESD groups ([Table 3]).

In terms of lesions with a positive vertical margin, the after-PSM results indicated that the EMR-L group had a significantly lower proportion compared with the ESD group (1.6% vs. 13.1%; OR 0.1, 95% CI 0.01–0.9) ([Table 3]). Subgroup analysis of adequate vertical margin distances, both before and after PSM, consistently suggests that lesions with a tumor size ≤ 5mm, without prior biopsy, and without central depression are more likely to benefit from EMR-L ([Fig. 3]).

Fig. 3 Subgroup analysis Forrest plot of adequate vertical margin distance before and after propensity score matching (PSM). CI, confidence interval; EMR-L, endoscopic mucosal resection with ligation; ESD, endoscopic submucosal dissection; OR, odds ratio. The ESD group is used as the reference group.

Other treatment outcomes

Overall median procedure time was 9.0 minutes (IQR 5.0–19.0 minutes). Notably, the ESD group required significantly more time than the EMR-L group, with median procedure durations of 19.0 minutes (IQR 14.0 to 28.0 minutes) compared with 5.0 minutes (IQR 4.0 to 7.0 minutes) for the EMR-L group after PSM, which also aligns with the before-PSM findings ([Table 3]).

After-PSM analysis revealed that the EMR-L group experienced a significantly lower incidence of overall complications compared with the ESD group (8.2% vs. 29.5%, OR 0.2, 95% CI 0.07–0.6, P = 0.005) ([Table 3]). Similarly, the perforation rate was reduced in the EMR-L group (3.3% vs. 16.4%, OR 0.2, 95% CI 0.04–0.8, P = 0.028) ([Table 3]). Although bleeding was observed in 4.9% of the EMR-L group, none occurred in the ESD group after PSM, with the difference not reaching statistical significance (P = 0.242) ([Table 3]). These complications were effectively managed with subsequent endoscopic interventions, routinely closing mucosal defects using clips. After PSM, median length of hospital stay was comparable between the EMR-L and ESD groups, both at 5.0 days, with IQRs of 3.0 to 5.0 and 3.0 to 6.0 days, respectively ([Table 3]).

Discussion

Our study compared EMR-L to ESD on the aspect of adequate vertical margin for the first time for treatment of rectal NETs smaller than 10 mm. The results demonstrate that EMR-L achieves a greater proportion of adequate vertical margin distances and a lower rate of positive vertical margins, resulting in a higher R0 resection rate compared with ESD. Conversely, the ESD group experiences significantly longer operation times and a higher incidence of perforation.

Rectal NETs predominantly develop in the submucosal layer, making it challenging to achieve clear vertical margins at the deepest point of invasion. Nonetheless, securing an R0 resection with an adequate vertical margin is essential for effective treatment. Few studies have explored how different endoscopic resection techniques affect achievement of suitable vertical margin distances in rectal NETs. Three previous studies have compared vertical margin distances between EMR-L and ESD, treating these as continuous variables [12] [16] [17]. The results have been inconsistent, showing significant variability in both range and SD (Supplementary Table 1). Clinically, the key is not the effect size of the vertical margin but ensuring it exceeds a specific threshold to improve prognosis. In early colorectal cancer, adequate vertical margins (typically > 500 μm) after complete endoscopic resection are associated with reduced risk of metastatic recurrence [23]. Therefore, examining the proportion of cases achieving adequate vertical margins is clinically significant. Our study is the first to compare this proportion across different treatments using PSM. Although the optimal cutoff for a vertical margin remains uncertain, our study suggests a 25% interquartile range (115 μm), which needs further validation through additional research.

Our study demonstrates that EMR-L achieves higher R0 resection rates and lower positive vertical margin rates compared with ESD, aligning with previous research and meta-analyses [9]. EMR-L is particularly effective for small rectal NETs with distinct margins, because it involves lifting the mucosa, suctioning, ligating, and excising the nodule, thus ensuring a deeper vertical resection margin [24]. This explains why EMR-L is most suitable for lesions ≤ 5 mm, without prior biopsy and central depression, aligning with recent large multicenter studies [1] [7]. This approach also reduces risk of perforation, a noted concern with ESD. A recent study highlighted that 64.6% of lesions involving use of biopsy forceps resulted in residual rectal NETs. In addition, scarring from biopsies can impede subsequent endoscopic treatments; therefore, avoiding biopsies for rectal NETs is advisable whenever possible [25]. Furthermore, EMR-L is straightforward and safe, even for beginners without ESD experience, because it requires no more technical skills than conventional EMR. Its simplicity also improves time efficiency compared with ESD, as confirmed by our study and others [8] [9].ESD was previously thought to achieve higher resection rates, facilitating en bloc removal of larger lesions. However, our study found that the ESD group had a higher risk of perforation rate compared with the EMR-L group. In ESD, achieving adequate vertical margins necessitates precise dissection near the muscularis propria to avoid positive margins and enhance R0 resection rates [6] [20]. This requirement explains the higher incidence of perforations, despite efforts to maximize vertical margin distance. A key limitation of ESD is lack of sufficient traction during dissection, which can lead to inadequate vertical margins. To counter this, we explored traction-assisted ESD using a dental floss loop, enabling deeper vertical dissection [26]. In addition, ESD procedures generally take more time, as confirmed by our study and others [8] [9].

This study has several limitations. First, being retrospective in nature, the study inherently carries a risk of selection bias. To mitigate this risk, we employed a prospective data collection approach using a consecutive cohort of patients with blinded pathological analysis. Second, although we utilized PSM to minimize the impact of selection bias, it was not feasible to balance for unmeasured confounders between groups, particularly regarding the selection process for the resection method.

This limitation may affect the comparability and generalizability of the results. Third, in the subgroup analysis, certain groups had limited sample sizes, such as those with tumor size > 5 mm, muscularis propria involvement, central depression, and biopsy history. This limitation hinders further analysis of the benefits of EMR-L for these subgroups. Fourth, due to the absence of established cutoff values, we adopted the 25th percentile as an exploratory threshold; although this approach may reduce bias to some extent, it cannot completely eliminate it. Fifth, EMR-L generally involves a shorter interval between submucosal injection and specimen fixation, and residual injection fluid can affect specimen flattening during fixation, which may influence the measured vertical margin distance. Finally, a significant limitation is that nearly all of the studies were conducted in Asian countries. Therefore, caution should be exercised when generalizing these findings to Western populations.

Conclusions

EMR-L offers significant advantages over ESD in terms of efficacy, safety, and procedure efficiency, particularly for lesions ≤ 5 mm, without prior biopsy and central depression. EMR-L consistently achieves a higher rate of R0 resections, likely due to better adequate vertical resection margins and a reduced incidence of positive margins. In addition, the procedure requires significantly less time compared with ESD while minimizing risk of perforation. Further research is needed to evaluate the comparative benefits of EMR-L versus traction-assisted ESD.

• References

• 1 Sekiguchi M, Hotta K, Takeuchi Y. et al. Characteristics of colorectal neuroendocrine tumors in patients prospectively enrolled in a Japanese multicenter study: a first report from the C-NET STUDY. J Gastroenterol 2022; 57: 547-558
• 2 Dasari A, Shen C, Halperin D. et al. Trends in the incidence, prevalence, and survival outcomes in patients with neuroendocrine tumors in the United States. JAMA Oncol 2017; 3: 1335-1342
• 3 Ngamruengphong S, Kamal A, Akshintala V. et al. Prevalence of metastasis and survival of 788 patients with T1 rectal carcinoid tumors. Gastrointest Endosc 2019; 89: 602-606
• 4 Ramage JK, De Herder WW, Delle Fave G. et al. ENETS Consensus Guidelines Update for Colorectal Neuroendocrine Neoplasms. Neuroendocrinology 2016; 103: 139-143
• 5 Tanaka S, Saitoh Y, Matsuda T. et al. Evidence-based clinical practice guidelines for management of colorectal polyps. J Gastroenterol 2021; 56: 323-335
• 6 Deprez PH, Moons LMG, O'Toole D. et al. Endoscopic management of subepithelial lesions including neuroendocrine neoplasms: European Society of Gastrointestinal Endoscopy (ESGE) Guideline. Endoscopy 2022; 54: 412-429
• 7 Ito S, Hotta K, Sekiguchi M. et al. Short-term outcomes of endoscopic resection for colorectal neuroendocrine tumors: Japanese multicenter prospective C-NET STUDY. Dig Endosc 2024; 36: 942-951
• 8 Chen J, Ye J, Zheng X. et al. Endoscopic treatments for rectal neuroendocrine tumors: a systematic review and network meta-analysis. J Gastrointest Surg 2024; 28: 301-308
• 9 Pan J, Zhang X, Shi Y. et al. Endoscopic mucosal resection with suction vs. endoscopic submucosal dissection for small rectal neuroendocrine tumors: a meta-analysis. Scand J Gastroenterol 2018; 53: 1139-1145
• 10 Niimi K, Goto O, Fujishiro M. et al. Endoscopic mucosal resection with a ligation device or endoscopic submucosal dissection for rectal carcinoid tumors: an analysis of 24 consecutive cases. Dig Endosc 2012; 24: 443-447
• 11 Choi CW, Kang DH, Kim HW. et al. Comparison of endoscopic resection therapies for rectal carcinoid tumor: endoscopic submucosal dissection versus endoscopic mucosal resection using band ligation. J Clin Gastroenterol 2013; 47: 432-436
• 12 Kim KM, Eo SJ, Shim SG. et al. Treatment outcomes according to endoscopic treatment modalities for rectal carcinoid tumors. Clin Res Hepatol Gastroenterol 2013; 37: 275-282
• 13 Bang BW, Park JS, Kim HK. et al. Endoscopic resection for small rectal neuroendocrine tumors: comparison of endoscopic submucosal resection with band ligation and endoscopic submucosal dissection. Gastroenterol Res Pract 2016; 2016: 6198927
• 14 Kaneko H, Hirasawa K, Koh R. et al. Treatment outcomes of endoscopic resection for rectal carcinoid tumors: an analysis of the resectability and long-term results from 46 consecutive cases. Scand J Gastroenterol 2016; 51: 1489-1494
• 15 Ebi M, Nakagawa S, Yamaguchi Y. et al. Endoscopic submucosal resection with an endoscopic variceal ligation device for the treatment of rectal neuroendocrine tumors. Int J Colorectal Dis 2018; 33: 1703-1708
• 16 Lim HK, Lee SJ, Baek DH. et al. Resectability of rectal neuroendocrine tumors using endoscopic mucosal resection with a ligation band device and endoscopic submucosal dissection. Gastroenterol Res Pract 2019; 2019: 8425157
• 17 Kamigaichi Y, Yamashita K, Oka S. et al. Clinical outcomes of endoscopic resection for rectal neuroendocrine tumors: Advantages of endoscopic submucosal resection with a ligation device compared to conventional EMR and ESD. DEN Open 2022; 2: e35
• 18 Libanio D, Pimentel-Nunes P, Bastiaansen B. et al. Endoscopic submucosal dissection techniques and technology: European Society of Gastrointestinal Endoscopy (ESGE) Technical Review. Endoscopy 2023; 55: 361-389
• 19 Takada K, Imai K, Yamada T. et al. Efficacy of endoscopic submucosal resection with a ligation device for small rectal neuroendocrine tumor: study protocol of a multicenter open-label randomized control trial (BANDIT trial). BMC Gastroenterol 2024; 24: 69
• 20 Wallenhorst T, Masgnaux LJ, Grimaldi J. et al. Obtaining a free vertical margin is challenging in endoscopic submucosal dissection of a rectal neuroendocrine tumor: use of adaptive traction to improve exposure in a child. Endoscopy 2023; 55: E763-E764
• 21 Tanaka S, Kashida H, Saito Y. et al. Japan Gastroenterological Endoscopy Society guidelines for colorectal endoscopic submucosal dissection/endoscopic mucosal resection. Dig Endosc 2020; 32: 219-239
• 22 Tahara T, Shijimaya T, Nishimon S. et al. Injury to the muscle layer and risk of non-cardiac chest pain after endoscopic submucosal dissection for esophageal cancer. J Gastrointestin Liver Dis 2024; 33: 25-29
• 23 Nishimura T, Oka S, Kamigaichi Y. et al. Vertical tumor margin of endoscopic resection for T1 colorectal carcinoma affects the prognosis of patients undergoing additional surgery. Surg Endosc 2022; 36: 5970-5978
• 24 Park SJ. Tips and tricks for better endoscopic treatment of colorectal tumors: usefulness of cap and band in colorectal endoscopic mucosal resection. Clin Endosc 2013; 46: 492-494
• 25 Cheminel L, Lupu A, Wallenhorst T. et al. Systematic resection of the visible scar after incomplete endoscopic resection of rectal neuroendocrine tumors. Am J Gastroenterol 2024; 119: 378-381
• 26 Liu J, Fang N. Traction by dental floss loop for adequate submucosal dissection depth in a rectal neuroendocrine tumor. Endoscopy 2023; 55: E326-E327