Submucosal fibrosis in large colorectal serrated lesions in cases receiving endoscopic submucosal dissection

Abstract

Background:

Submucosal fibrosis is a known risk factor for adverse outcomes in colorectal endoscopic submucosal dissection (ESD). However, evidence regarding the presence and impact of fibrosis in colorectal serrated lesions (CSLs) remains limited.

Objectives:

This study aimed to evaluate the association between CSLs and submucosal fibrosis, and to assess the impact of fibrosis presence on therapeutic outcomes of colorectal ESD.

Design:

Retrospective cohort study.

Methods:

We retrospectively reviewed consecutive colorectal ESD cases performed between 2020 and 2024. Only lesions ⩾20 mm were included; cases were classified as CSLs and adenoma + T1 cancer according to histological diagnosis. CSLs included sessile serrated lesions (SSL), traditional serrated adenomas, SSL with dysplasia, and unclassified serrated adenomas. Patient/lesion characteristics, presence of fibrosis, severe fibrosis, and ESD outcomes were assessed. Multivariate analyses, including CSLs’ histology, were performed to identify factors associated with the presence of submucosal fibrosis and severe fibrosis.

Results:

A total of 445 colorectal ESD cases were included, comprising 72 CSLs and 373 adenoma + T1 cancer. CSLs were significantly associated with a lower presence of fibrosis (34.7% and 48.5%, p = 0.04) and severe fibrosis (6.9% vs 18.2%; p = 0.03) compared to adenoma + T1 cancer. In multivariate analysis, CSLs’ histology (odds ratio (OR): 0.58; p = 0.04), lesion size ⩾40 mm (OR: 2.03; p < 0.01), and rectal location (OR: 0.40; p < 0.01) were significantly related to fibrosis presence. Lesion size ⩾40 mm (OR: 2.45; p < 0.01) and polypoid morphology (OR: 3.42; p < 0.01) were significantly related to severe fibrosis.

Conclusion:

CSLs’ histology was negatively correlated with submucosal fibrosis in colorectal ESD.

Keywords

colorectal cancer colorectal polyps endoscopic submucosal dissection serrated lesions submucosal fibrosis

Introduction

Colorectal serrated lesions (CSLs) are increasingly recognized due to their rising incidence and contribution to colorectal cancer development, accounting for up to 25% of them.^1,2 Their subtle endoscopic appearance could lead to missed detections, contributing to the occurrence of interval cancers.^3,4 The challenging nature of these lesions highlights the critical need for improved diagnostic recognition and effective management strategies.

In recent years, significant advances have been made regarding the diagnosis and treatment of CSLs. First, the World Health Organization (WHO) introduced an updated classification system and diagnostic criteria in 2019, defining CSLs as sessile serrated lesions (SSL), traditional serrated adenomas (TSA), sessile serrated lesions with dysplasia (SSLD), and unclassified serrated adenomas.⁵ Second, recent guidelines and consensus statements have proposed new treatment strategies for these lesions.^6,7 Although piecemeal resection is considered acceptable for large CSLs, these lesions may harbor unexpected dysplasia, thereby increasing the risk of colorectal neoplasia.⁸ In addition, larger CSLs may be associated with a higher incidence of advanced metachronous neoplasias.⁹ Therefore, advanced resection techniques are recommended, particularly for lesions with complex morphology, large size, difficult anatomical locations, non-lifting signs, or suspected superficial submucosal invasion, to ensure an accurate histological diagnosis. In such cases, endoscopic submucosal dissection (ESD) offers a valuable approach for achieving en bloc resection, particularly in technically demanding CSLs.^10,11 ESD also enables precise histopathological assessment and is associated with high rates of complete resection and low recurrence rates, making it a key strategy for selected lesions. Consequently, its use is increasingly expanding across Western countries.¹²

Despite its advantages, ESD is technically challenging and is associated with potential risks such as perioperative and postoperative bleeding, perforation, and prolonged procedure times. Factors such as lesion size, location, and the presence of submucosal fibrosis can increase procedural difficulty and the risk of complications.^13,14 Submucosal fibrosis poses challenges by obscuring the appropriate submucosal layer for dissection and increasing the likelihood of incomplete resection or perioperative perforation.¹⁵ However, there is limited evidence exploring the relationship between large CSLs, submucosal fibrosis, and the technical aspects of ESD for these lesions.¹¹ This study aimed to evaluate the association between CSLs and therapeutic results of ESD, including the status of fibrosis, and to explore its effects on colorectal ESD outcomes.

Methods

This retrospective cohort study analyzed data from consecutive colorectal ESD procedures performed at Kyoto Prefectural University of Medicine between January 2020 and November 2024. From the overall cohort, only lesions measuring 20 mm or larger were included. Although ESD has been performed at our institution since 2006, this time frame was selected to ensure alignment with the WHO histological classification introduced in 2019.¹⁶ The indications for ESD followed Japanese guidelines and included lesions ⩾20 mm in size that were considered unsuitable for en bloc resection using endoscopic mucosal resection (EMR).¹⁷ Especially, the indications of ESD for serrated lesions were as follows: (1) SSL larger than 30 mm, due to the risk of requiring multiple-piece EMR (defined as resection in ⩾5 fragments); or (2) lesions larger than 20 mm suspected to be SSLD or TSA based on magnified endoscopic assessment. Endoscopic diagnosis of the lesion was performed by magnification with narrow band imaging, blue laser/light imaging, and pit pattern observation in all cases.^18–21 Cases were classified as CSLs and adenoma or T1 cancer according to histological diagnosis. CSLs included SSL, TSA, SSLD, and unclassified serrated adenomas.¹⁶ Patients with a history of inflammatory bowel disease, prior biopsy of the lesion, or previous endoscopic resection attempt were also included. Patient/lesion characteristics, fibrosis presence, and ESD outcomes were assessed. Lesions <20 mm or those with histology other than serrated lesions, adenomas, or T1 cancer were excluded. In addition, cases with missing data on lesion size or procedure time were excluded from the final analysis (Figure 1).

Figure 1.

Study flow/endoscopic submucosal dissection.

This study was reported in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement.²² (The STROBE checklist is available online in the Supplemental Material).

Study outcomes

The primary outcome of this study was to evaluate the frequency of submucosal fibrosis and severe fibrosis in large CSLs of ⩾20 mm compared to adenomas and T1 cancers undergoing ESD. In addition, therapeutic outcomes of ESD for CSLs were assessed, including procedure time, en bloc resection, R0 resection, and adverse events such as intraprocedural perforation, delayed bleeding, and delayed perforation. Secondary outcomes included identifying independent predictors of any submucosal fibrosis, severe fibrosis, and prolonged procedure time in colorectal ESD. Prolonged procedure time was defined as greater than 90 min, based on our previously published experience.²³

Pretreatment and equipment for ESD

All patients followed a low-residue diet the day before the procedure. Bowel preparation included 10 mL of sodium picosulfate hydrate at 9:00 PM. On the day of ESD, patients received either 1.0 L of a polyethylene glycol-ascorbic acid solution (MoviPrep; Ajinomoto Pharmaceutical Co., Tokyo, Japan) or 480 ml of an oral sulfate-based preparation (SULPREP; Fuji Pharma Co., Ltd., Tokyo, Japan) for colonic cleansing.

Colorectal ESD was performed using EC-L600ZP, EC-L600ZP7, or EC-760V endoscopes (Fujifilm Medical Co., Tokyo, Japan), all with water-jet and CO₂ insufflation systems (JW-2/3, GW-100; Fujifilm Medical Co.) according to our previous report.²³ A transparent cap, Elastic Touch (F-25; TOP Corporation, Tokyo, Japan) or short ST hood (DH-29CR; Fujifilm Medical Co.) was used based on lesion and fibrosis characteristics. A lens cleaning agent (Cleash; Fujifilm Medical Co.) was applied to prevent fogging. A viscous injection solution of 0.2%–0.4% sodium hyaluronate (MucoUp; Boston Scientific Japan Co., Tokyo, Japan) or 0.3%–0.6% sodium alginate (Liftal K; Kaigen Pharma, Tokyo, Japan) mixed with 0.2% indigo carmine (final concentration 0.06%) was used. Injection was performed with a 25-G needle (ImpactFlow; TOP Corporation).

High-frequency electrosurgical units (VIO300D and VIO3; Erbe Elektromedizin, Tübingen, Germany) were used in combination with various knives selected according to the endoscopist’s preference. Scissor-type knives included the Clutch Cutter (Fujifilm Medical Co.) and SB Knife Jr2 (Sumitomo Bakelite Co., Tokyo, Japan), while needle-type knives such as the ORISE™ ProKnife (Boston Scientific, Marlborough, MA, USA) and the Flush Knife BT-S (Fujifilm Medical Co.) were also employed to facilitate efficient mucosal incision and submucosal dissection.²³ Electrosurgical settings were standardized: forced coagulation (6.0) for marking, Endocut I (Effect 1, Duration 4, Interval 1) for mucosal incision, and either forced coagulation (3.0) or Endocut I for submucosal dissection. Hemostasis was achieved using soft coagulation (3.0) for major bleeding and forced coagulation (3.0) for minor bleeding.

Procedural steps were standardized and included lesion marking, full circumferential mucosal incision, trimming, and submucosal dissection (Figure 2). When needed, traction was applied using either an S-O clip (Zeon Medical, Tokyo, Japan) or a SureClip Traction Band (Micro-Tech Co., Nanjing, China), as per operator preference. Following ESD, full or partial closure of the ESD defect was performed at the discretion of the endoscopist using traditional reopenable clips, the MANTIS Closure Device (Boston Scientific), or SureClip Traction Band (Micro-Tech Co.), depending on defect size, location, and procedural complexity.²⁴

Figure 2.

An SSL treated by ESD. (a) WLI of a flat elevated lesion (0–IIa) 30 mm in the right-sided colon. (b) Blue light imaging highlighting dilated glands and clear margin of the lesion. (c) Lesion marking prior ESD. (d) Mucosal incision. (e) Traction device (S–O clip) was deployed. (f) Submucosal dissection without fibrosis, (g) ESD defect, (h) histology: sessile serrated lesion with negative margins.

Definitions

The lesion’s location was divided into right-sided colon (cecum to transverse colon), left-sided colon (descending colon to sigmoid colon), and rectum. Lesion morphology was classified as polypoid or non-polypoid based on the Paris classification.²⁵ Lesion size was calculated from the resected specimen. Submucosal fibrosis was assessed endoscopically and classified according to the criteria proposed by Matsumoto et al.,²⁶ as follows: F0, indicating no fibrosis and characterized by a blue transparent submucosal layer; F1, representing mild fibrosis, appearing as a white web-like structure within the blue submucosal layer; and F2, indicating severe fibrosis, visualized as a dense, white, muscular-like structure without a blue transparent layer.

Adverse events were defined according to our previous paper.²³ Intraprocedural perforation was defined as a visible defect in the muscularis propria with exposure of fatty tissue or adjacent organs, confirmed during the ESD procedure. Delayed bleeding was defined as the development of post-procedural bleeding requiring endoscopic hemostasis or a decrease in hemoglobin of more than 2 g/dL within 30 days after ESD. Delayed perforation was defined as the presence of free air detected by computed tomography within 14 days after ESD, not associated with an intraprocedural perforation.

The histopathological diagnosis of CSLs, adenoma, and T1 cancer was according to the WHO classification. An R0 resection was defined as the histological absence of dysplastic or neoplastic cells at both the lateral and deep margins.²⁷

Statistical analysis

Categorical variables were compared using the Chi-squared or Fisher’s exact test, as appropriate. Continuous variables were analyzed using the Student’s t test or Mann–Whitney U test, based on data distribution. The confidence interval (CI) was 95% and statistical significance was set at p < 0.05. A logistic regression model for the presence of fibrosis, severe fibrosis, and procedural time was conducted, including variables with p < 0.05. For the multivariate analysis, lesion size was categorized as <40 or ⩾40 mm, based on prior studies that identified this threshold as a significant cutoff influencing outcomes and treatment strategy decisions.²⁸ RStudio software (version 2023.12.0+369; Posit Software, PBC, Boston, MA, USA) was used for all statistical analyses.

Results

Among 445 colorectal ESD procedures, the mean age of patients was 69.0 ± 10.8 years, and 248 cases (55.7%) were male (Table 1). Lesions were located in the right-sided colon in 273 cases (61.3%); the left-sided colon in 104 cases (23.3%); and the rectum in 68 cases (15.4%). Non-polypoid morphology was observed in 80.7% of cases. The mean lesion size was 33.5 ± 16.0 mm. Submucosal fibrosis was present in 46.3% of lesions, and severe fibrosis was identified in 16.4%. The mean procedural time was 64.8 ± 42.0 min. En bloc resection was achieved in 99.6% of cases, and R0 resection was confirmed in 91.9%. Histologically, 72 lesions (16.2%) were classified as CSLs, and 373 (83.8%) as adenomas + T1 cancers.

Table 1.

Baseline characteristics and outcomes of colorectal ESD patients.

Patient characteristics	n = 445
Age, mean ± SD	69.0 ± 10.8
Sex, male/female, n (%)	248 (55.7)/197 (44.3)
Use of antithrombotic, n (%)	58 (13.0)
Use of anticoagulant, n (%)	34 (7.6)
Location, right-sided/left-sided/rectum, n (%)	273 (61.3)/104 (23.3)/68 (15.4)
Lesion size, mm, mean ± SD (range)	33.5 ± 16.0 (20–140)
Morphology, polypoid/non-polypoid, n (%)	86 (19.3)/359 (80.7)
Presence of fibrosis, n (%)	206 (46.3)
Severe fibrosis, n (%)	73 (16.4)
Intraprocedural perforation, n (%)	3 (0.7)
ESD procedural time, min, mean ± SD (range)	64.8 ± 42.0 (12–293)
Long procedure time (⩾90 min), n (%)	73 (16.4)
En bloc resection/piecemeal, n (%)	443 (99.6)/2 (0.4)
R0 resection, n (%)	409 (91.9)
Histology, n (%) Serrated lesions/adenoma + T1 cancer	72 (16.2)/373 (83.8)
Histology of serrated lesions, n (%) SSL/SSLD/TSA/unclassified	50 (69.4)/8 (11.1)/8 (11.1)/6 (8.4)
Histology of adenoma + T1 cancer, n (%) low-grade/high-grade/T1	102 (27.4)/215 (57.6)/56 (15.0)
Delayed bleeding, n (%)	5 (1.1)
Delayed perforation, n (%)	9 (2.0)

High-grade: high-grade adenoma. Left-sided: descending colon to sigmoid colon. Low-grade: low-grade adenoma. Right-sided: cecum to transverse colon.

ESD, endoscopy submucosal dissection; SD, standard deviation; SSL, sessile serrated lesions; SSLD, SSL with dysplasia; TSA, traditional serrated adenoma.

Serrated lesion subtypes were analyzed (Table 2). SSLs were the most common (n = 50), followed by SSLD (n = 8), TSA (n = 8), and unclassified lesions (n = 6). SSLDs had the highest rate of severe fibrosis (25%) and longer procedure times (⩾90 min in 37.5%) and were associated with a lower R0 resection rate (75%). TSAs more frequently exhibited polypoid morphology (75%). Overall, complications were infrequent among serrated lesions. A comparative analysis between serrated lesions and adenoma + T1 cancers was conducted (Table 3). Serrated lesions were significantly more frequent in females (61.1% vs 41.0%, p < 0.01). CSLs were located as follows: right-sided colon, 53 (73.6%); left-sided colon, 12 (16.7%); and rectum, 7 (9.7%). Non-polypoid morphology was significantly more common in CSLs (95.8% vs 77.7%, p < 0.01). The presence of submucosal fibrosis was significantly lower in CSLs (34.7% vs 48.5%, p = 0.04), and severe fibrosis was also significantly less frequent in the CSLs group (6.9% vs 18.2%, p = 0.03). The mean procedure time was significantly shorter in the CSLs group (53.0 ± 25.6 vs 67.1 ± 44.2 min, p < 0.01).

Table 2.

Colorectal serrated lesions characteristics and ESD outcomes.

Patient characteristics	Total serrated lesions(n = 72)	SSL(n = 50)	SSLD(n = 8)	TSA(n = 8)	Unclassified(n = 6)
Age, mean ± SD	67.3 ± 11.1	66.9 ± 10.9	73.2 ± 9.2	64 ± 13.1	67.5 ± 11.8
Sex, male/female, n (%)	28 (38.9)/44 (61.1)	19 (38)/31 (62)	3 (37.5)/5 (62.5)	3 (37.5)/5 (62.5)	3 (50)/3 (50)
Use of antithrombotic, n (%)	6 (8.3)	6 (12)	0 (0)	0 (0)	0 (0)
Use of anticoagulant, n (%)	3 (4.2)	2 (4)	0 (0)	0 (0)	1 (16.7)
Location, n (%) right-sided/left-sided/rectum	53 (73.6)/12 (16.7)/7 (9.7)	45 (90)/4 (8)/1 (2)	7 (87.5)/1 (12.5)/0 (0)	5 (62.5)/3 (37.5)	1 (16.7)/4 (66.6)/1 (16.7)
Lesion size, mm, mean ± SD (range)	30.2 ± 10.5 (20–70)	29.5 ± 8 (20–55)	35.1 ± 19 (20–65)	32.1 ± 15.7 (20–70)	26.8 ± 5.2 (20–32)
Lesion ⩾40 mm, n (%)	9 (12.5)	5 (10)	3 (37.5)	1 (12.5)	0 (0)
Morphology, polypoid/non-polypoid, n (%)	3 (4.2)/69 (95.8)	0 (0)/50 (100)	0 (0)/8 (100)	2 (25)/6 (75)	1 (16.7)/5 (83.3)
Presence of fibrosis, n (%)	25 (34.7)	20 (40)	3 (37.5)	1 (12.5)	1 (16.7)
Severe fibrosis, n (%)	5 (6.9)	3 (6)	2 (25)	0 (0)	0 (0)
Intraprocedural perforation, n (%)	1 (1.4)	1 (2)	0 (0)	0 (0)	0 (0)
ESD procedural time, min, mean ± SD (range)	53.0 ± 25.6 (18–138)	50.7 ± 22.9 (21–138)	67.4 ± 36.3 (18–114)	56.3 ± 31.1 (19–110)	48 ± 23.9 (21–79)
ESD procedure time ⩾90 min, n (%)	7 (9.7)	3 (6)	3 (37.5)	1 (12.5)	0 (0)
En bloc resection/piecemeal, n (%)	71 (98.6)/1 (1.4)	49 (98)/1 (2)	8 (100)/0 (0)	8 (100)/0 (0)	6 (100)/0 (0)
R0 resection, n (%)	65 (90.2)	46 (92)	6 (75)	7 (87.5)	6 (100)
Delayed bleeding, n (%)	0 (0.0)	0 (0)	0 (0)	0 (0)	0 (0)
Delayed perforation, n (%)	1 (1.4)	1 (2)	0 (0)	0 (0)	0 (0)

Right-sided: cecum to transverse colon. Left-sided: descending colon to sigmoid colon.

ESD, endoscopy submucosal dissection; SD, standard deviation; SSL, sessile serrated lesion; SSLD, SSL with dysplasia; TSA, traditional serrated adenomas.

Table 3.

Comparison of baseline characteristics and ESD outcomes by histology.

Patient characteristics	Serrated lesions(n = 72)	Adenoma + T1 cancer(n = 373)	p Value
Age, mean ± SD	67.3 ± 11.1	69.6 ± 10.7	0.12
Sex, male/female, n (%)	28 (38.9)/44 (61.1)	220 (59.0)/153 (41.0)	<0.01
Use of antithrombotic, n (%)	6 (8.3)	52 (13.9)	0.27
Use of anticoagulant, n (%)	3 (4.2)	31 (8.3)	0.33
Location, n (%) right-sided/left-sided/rectum	53 (73.6)/12 (16.7)/7 (9.7)	220 (59)/92 (24.7)/61 (16.4)	0.06
Lesion size, mm, mean ± SD (range)	30.2 ± 10.5 (20–70)	34.2 ± 16.9 (20–140)	0.05
Lesion ⩾40 mm, n (%)	9 (12.5)	103 (27.6)	0.01
Morphology, polypoid/non-polypoid, n (%)	3 (4.2)/69 (95.8)	83 (22.3)/290 (77.7)	<0.01
Presence of fibrosis, n (%)	25 (34.7)	181 (48.5)	0.04
Severe fibrosis, n (%)	5 (6.9)	68 (18.2)	0.03
Intraprocedural perforation, n (%)	1 (1.4)	2 (0.5)	0.40
ESD procedural time, min, mean ± SD (range)	53.0 ± 25.6 (18–138)	67.1 ± 44.2 (18–293)	<0.01
ESD procedure time ⩾90 min, n (%)	7 (9.7)	66 (17.7)	0.13
En bloc resection/piecemeal, n (%)	71 (98.6)/1 (1.4)	372 (99.7)/1 (0.3)	0.29
R0 resection, n (%)	65 (90.2)	344 (92.2)	0.74
Delayed bleeding, n (%)	0 (0.0)	5 (1.3)	1.00
Delayed perforation, n (%)	1 (1.4)	8 (2.1)	1.00

Right-sided: cecum to transverse colon. Left-sided: descending colon to sigmoid colon.

ESD, endoscopy submucosal dissection; SD, standard deviation.

Multivariate analysis identified serrated histology as an independent negative predictor of submucosal fibrosis (odds ratio (OR): 0.58; 95% CI: 0.33–0.99; p = 0.04; Table 4). Rectal location was also significantly associated with a lower risk of fibrosis (OR: 0.40; 95% CI: 0.22–0.70; p < 0.01). In contrast, lesions ⩾40 mm were significantly associated with an increased risk of submucosal fibrosis (OR: 2.03; 95% CI: 1.30–3.20; p < 0.01).

Table 4.

Risk factors for the presence of fibrosis in colorectal ESD.

Patient characteristics	Univariate analysis			Multivariate analysis
Patient characteristics	With fibrosis(n = 206)	Without fibrosis(n = 239)	p Value	OR	95% CI	p Value
Age, mean ± SD	68.8 ± 11.5	69.5 ± 10.2	0.74
Sex, male/female, n (%)	117 (57.8)/89 (43.2)	131 (54.8)/108 (45.2)	0.74
Use of antithrombotic, n (%)	29 (14.1)	29 (12.1)	0.64
Use of anticoagulant, n (%)	18 (8.74)	16 (6.69)	0.62
Location, colon/rectum, n (%)	185 (89.8)/21 (10.2)	192 (80.3)/47 (19.7)	<0.01	0.40	0.22–0.70	<0.01
Lesion size, mm, mean ± SD (range)	35.6 ± 16.1	31.8 ± 15.8	<0.01	^a
Lesion <40 mm/⩾40 mm, n (%)	140 (68)/66 (32)	193 (80.8)/46 (19.2)	<0.01	2.03	1.30–3.20	<0.01
Morphology, polypoid/non-polypoid, n (%)	48 (23.3)/158 (76.7)	38 (15.9)/201 (84.1)	0.06
Histology, n (%)CSLs/Adenoma + T1 cancer	25 (12.1)/181 (87.9)	47 (19.7)/192 (80.3)	0.04	0.58	0.33–0.99	0.04

Not included in the multivariate analysis.

CSLs, colorectal serrated lesions; ESD, endoscopy submucosal dissection; OR, odds ratio; SD, standard deviation.

In the multivariate analysis, lesion size ⩾40 mm (OR: 2.45; 95% CI: 1.40–4.27; p < 0.01) and polypoid morphology (OR: 3.42; 95% CI: 1.92–6.06; p < 0.01) were associated with an increased risk of severe fibrosis (Table 5).

Table 5.

Risk factors for severe fibrosis in colorectal ESD.

Patient characteristics	Univariate analysis			Multivariate analysis
Patient characteristics	With severe fibrosis(n = 73)	Without severe fibrosis (n = 372)	p Value	OR	95% CI	p Value
Age, mean ± SD	67.8 ± 12	69.5 ± 10.6	0.28
Sex, male/female, n (%)	48 (65.8)/25 (34.2)	200 (53.8)/172 (46.2)	0.07
Use of antithrombotic, n (%)	7 (9.6)	51 (13.7)	0.44
Use of anticoagulant, n (%)	4 (5.5)	30 (8.1)	0.60
Location, colon/rectum, n (%)	64 (87.7)/9 (12.3)	313 (84.1)/59 (15.9)	0.55
Lesion size, mm, mean ± SD (range)	38.5 ± 17.6	32.6 ± 15.6	<0.01	^a
Lesion <40 mm/⩾40 mm, n (%)	44 (60.3)/29 (39.7)	289 (77.7)/83 (22.3)	<0.01	2.45	1.40–4.27	<0.01
Morphology, polypoid/non-polypoid, n (%)	28 (38.4)/45 (61.6)	58 (15.6)/314 (84.4)	<0.01	3.42	1.92–6.06	<0.01
Histology, n (%)CSLs/Adenoma + T1 cancer	5 (6.8)/68 (93.2)	67 (18.0)/305 (82.0)	0.02	0.49	0.16–1.20	0.15

Not included in the multivariate analysis.

CSLs, colorectal serrated lesions; ESD, endoscopy submucosal dissection; OR, odds ratio; SD, standard deviation.

Prolonged ESD procedure time of 90 mins was significantly related with lesion size ⩾40 mm (OR: 25.00; 95% CI: 12.50–54.40; p < 0.01), the presence of fibrosis (OR: 3.73; 95% CI: 1.70–8.47; p < 0.01), and severe fibrosis (OR: 5.14; 95% CI: 2.25–12.40; p < 0.01).

A descriptive analysis was performed on the five serrated lesions that presented with severe fibrosis (Table 6). Among these cases, 60% were female, and four were located in the cecum, with one in the transverse colon. Lesion sizes ranged from 20 to 65 mm, and R0 resection was achieved in only 60% of cases. Notably, 40% of lesions exhibited dysplasia.

Table 6.

Details of five serrated cases with severe fibrosis.

Case	Age	Sex	AT	AC	Location	Size(mm)	Morphology	Intraprocedural perforation	ESDPT (min)	En bloc resection	R0 resection	Histology	Delayed bleeding	Delayed perforation
1	71	F	N	N	C	20	0–IIa	N	112	Y	N	SSLD	N	N
2	41	M	N	N	C	25	0–IIa	N	37	Y	Y	SSL	N	N
3	67	M	N	N	C	65	0–IIa	N	96	Y	Y	SSLD	N	N
4	41	F	N	N	T	20	0–IIa	N	48	Y	Y	SSL	N	N
5	61	F	N	N	C	30	0–IIa	N	65	Y	N	SSL	N	N

AC, anticoagulant; AT, antithrombotic; C, cecum; ESD, endoscopic submucosal dissection; F, female; M, male; min, minutes; mm, millimeters; N, no; PT, procedural time; SSL, sessile serrated lesion; SSLD, SSL with dysplasia; T, transverse colon; Y, yes.

Discussion

Prior to the update of the WHO classification in 2019, a study published in 2016 compared large serrated and adenomatous histologies in 2,000 colorectal lesions of ⩾20 mm resected by EMR. It reported a higher proportion of female patients (58%) and a significantly lower rate of fibrosis in CSLs compared to adenomatous lesions (12% vs 26%; p < 0.01). However, the severity of fibrosis was not evaluated.²⁹ A more recent comparative study, in 2020, analyzing 41 CSLs and 226 non-serrated lesions, resected by ESD, also found a female predominance among CSLs (63%).¹¹ However, it did not observe significant differences in fibrosis rates (2.4% vs 6.6%, p = 0.61) or procedural time (52 vs 55 min, p = 0.17) between groups.¹¹ Despite these experiences, evidence on the specific impact of serrated histology on ESD characteristics and outcomes remains limited. In our study, which followed the updated WHO classification and included 72 CSLs, both fibrosis and severe fibrosis were less frequently observed in CSLs compared to adenomas + T1 cancers. In addition, ESD procedure time was shorter for CSLs compared to adenomas + T1 cancers, likely due to the lower prevalence of fibrosis and a smaller proportion of lesions ⩾40 mm. Further large-scale studies are warranted to validate these findings.

Regarding the risk factor for fibrosis, the largest previously published series by Pellise et al.²⁹ in 2016, which included 246 large serrated lesions, resected by EMR, reported significantly lower fibrosis rates in CSLs compared to adenomas. A less fibrotic submucosal layer in serrated lesions has previously been proposed in the context of EMR,³⁰ based on favorable outcomes regarding complete resection and low recurrence rates reported with this technique.^31,32 However, this finding has not been consistently validated in ESD studies, where the submucosal layer is more thoroughly evaluated.^11,27 Regarding other established predictors of fibrosis, our findings align with previous reports showing that rectal location is associated with a lower prevalence of fibrosis compared to colonic lesions,^33,34 while larger lesion size remains a significant risk factor.^35,36 Polypoid morphology was not a significant predictor of fibrosis. Similar results were reported in a previous study, which identified surface granularity as an independent predictor, while laterally spreading tumor (LST) type and polypoid morphology were not significant factors.²⁷ The predictive value of morphology has varied across studies. Chiba et al.³³ associated cecal location, preoperative biopsy, straddling folds, and non-granular pseudo-depressed LSTs with mild fibrosis, whereas Sferrazza et al.³⁶ identified lesion morphology and size as key predictors of fibrosis risk. In the current study, CSLs remained significantly associated with a lower risk of fibrosis. In addition, colonic location and large lesions were other risk factors for fibrosis. The definition of fibrosis was different in each study and seemed a little subjective. Further analysis of risk factors for fibrosis should be performed using an objective definition.

With respect to severe fibrosis, although serrated histology showed a trend toward lower rates of severe fibrosis, this association did not reach statistical significance in the multivariate analysis (OR: 0.49, 95% CI: 0.16–1.20, p = 0.15), likely due to the limited number of CSL cases presenting severe fibrosis in our cohort. Our data confirmed a stronger association with larger lesions (OR: 2.45, 95% CI: 1.40–4.27, p < 0.01) and polypoid morphology (OR: 3.42, 95% CI: 1.92–6.06, p < 0.01). Similar findings were shown by a large series of 524 colorectal ESD cases, where size ⩾40 mm (OR: 2.22, 95% CI: 1.23–4.02, p = 0.008) and polypoid lesions (OR: 4.43, 95% CI: 2.17–9.06, p = 0.001) were significantly associated with severe fibrosis.³⁷ Both variables have been identified as predictive factors in previously published models.^13,36,38

Although univariate analysis suggested that CSLs’ histology was less frequently associated with prolonged procedures, this finding was not retained in the multivariate model. This may be attributed to the low number of serrated lesions exceeding 90 min of procedural time, potentially limiting the statistical power to detect an independent association. In contrast, lesion size emerged as the strongest predictor of procedural time, consistent with prior time estimation models in which size remains a key variable.^39,40 In line with existing evidence, the presence of any submucosal fibrosis⁴¹ and, as expected, severe fibrosis were also significantly associated with prolonged ESD duration.^37,42 However, procedural time should be interpreted cautiously, as it reflects a multifactorial combination of patient, lesion, and operator-related factors, not lesion behavior alone.¹⁵ As highlighted by Jacques et al.,⁴³ it must be viewed in the broader context of overall case complexity, with histology being just one of several contributing elements.

The clinical relevance of the findings on fibrosis and severe fibrosis in this study lies in their potential application to procedural planning and predictive modeling. As interest in anticipating submucosal fibrosis grows, our data suggest that histological subtype, specifically serrated histology, may serve as a valuable variable in fibrosis prediction tools.³⁶ Furthermore, the lower fibrosis burden associated with CSLs may support the confident use of less invasive resection techniques, such as cold snare polypectomy or conventional EMR, in selected cases. This may be particularly advantageous in settings where ESD is debated, or in nonexpert centers, where concerns about fibrosis-related complications often limit the resection of large lesions. Given the reduced fibrosis risk in CSLs, these techniques could be considered safer and more feasible alternatives in appropriate clinical scenarios.

To adopt histology as a reliable predictor of submucosal fibrosis across diverse clinical settings, including both high-volume expert centers and community-based nonexpert institutions, meticulous endoscopic assessment is essential. This includes the use of image-enhanced endoscopy techniques and validated classification systems such as JNET¹⁸ (with magnification) and WASP⁴⁴ (without magnification) to accurately identify CSLs and assess dysplasia risk.

Several limitations must be acknowledged. As a retrospective, single-center study, this analysis is inherently subject to selection bias and missing data, which represent key methodological constraints. In addition, we were unable to determine whether lesions had been previously biopsied, a known risk factor for submucosal fibrosis,⁴¹ though not necessarily associated with impaired outcomes.⁴⁵ Another limitation is the reliance on endoscopic fibrosis assessment, which could be subjective, as histological confirmation was not performed. Although the method for differentiating the status of fibrosis is commonly used, there is a lack of histological validation of it. Future prospective validation is expected. Finally, an important consideration is the external validity and applicability of our findings.

Conclusion

This study investigated serrated histology as a potential predictor of reduced submucosal fibrosis in colorectal ESD. Our findings demonstrated that CSLs were associated with a lower prevalence of fibrosis. These results may aid in therapeutic decision-making. Lesions suspected to be CSLs without features of carcinoma could be safely resected using EMR instead of ESD. Conversely, ESD should be considered for lesions ⩾40 mm or those with polypoid morphology, as these characteristics were associated with a higher risk of severe submucosal fibrosis.

Supplemental Material

sj-doc-1-tag-10.1177_17562848251360097 – Supplemental material for Submucosal fibrosis in large colorectal serrated lesions in cases receiving endoscopic submucosal dissection

Supplemental material, sj-doc-1-tag-10.1177_17562848251360097 for Submucosal fibrosis in large colorectal serrated lesions in cases receiving endoscopic submucosal dissection by Erik Manriquez-Alegria, Naohisa Yoshida, Reo Kobayashi, Naoto Iwai, Ken Inoue, Osamu Dohi, Lucas Cardoso and Hideyuki Konishi in Therapeutic Advances in Gastroenterology

Footnotes

Acknowledgements

We thank all members of the Department of Molecular Gastroenterology and Hepatology, Kyoto Prefectural University of Medicine for their help with this study.

Declarations

ORCID iDs

Erik Manriquez-Alegria

Naohisa Yoshida

Supplemental material

Supplemental material for this article is available online.

References

Crockett

Nagtegaal

. Terminology, molecular features, epidemiology, and management of serrated colorectal neoplasia. Gastroenterology 2019; 157: 949–966.e4.

Anderson

Hisey

Mackenzie

, et al. Clinically significant serrated polyp detection rates and risk for postcolonoscopy colorectal cancer: data from the New Hampshire Colonoscopy Registry. Gastrointest Endosc 2022; 96: 310–317.

East

Atkin

Bateman

, et al. British Society of Gastroenterology position statement on serrated polyps in the colon and rectum. Gut 2017; 66: 1181–1196.

Trovato

Turshudzhyan

Tadros

. Serrated lesions: a challenging enemy. World J Gastroenterol 2021; 27: 5625–5629.

Kim

Kang

. Evolving pathologic concepts of serrated lesions of the colorectum. J Pathol Transl Med 2020; 54: 276–289.

Ferlitsch

Hassan

Bisschops

, et al. Colorectal polypectomy and endoscopic mucosal resection: European Society of Gastrointestinal Endoscopy (ESGE) Guideline—Update 2024. Endoscopy 2024; 56: 516–545.

Djinbachian

Rex

Chiu

H-M

, et al. International consensus on the management of large (⩾20 mm) colorectal laterally spreading tumors: World Endoscopy Organization Delphi study. Dig Endosc 2024; 36: 1253–1268.

De Jong

Nagtegaal

Vos

, et al. Increased colorectal neoplasia risk in patients with inflammatory bowel disease and serrated polyps with dysplasia. Dig Dis Sci 2022; 67: 5647–5656.

Medawar

Djinbachian

Taghiakbari

, et al. Large serrated polyps indicate a greater risk of advanced metachronous colorectal neoplasia than high-grade adenomas. Endosc Int Open 2023; 11: E849–E858.

10.

Repici

Hassan

De Paula Pessoa

, et al. Efficacy and safety of endoscopic submucosal dissection for colorectal neoplasia: a systematic review. Endoscopy 2012; 44: 137–150.

11.

Kuroki

Endo

Iwahashi

, et al. Acceptability of endoscopic submucosal dissection for sessile serrated lesions: comparison with non-sessile serrated lesions. Endosc Int Open 2020; 8: E1832–E1839.

12.

Draganov

Aihara

Karasik

, et al. Endoscopic submucosal dissection in North America: a large prospective multicenter study. Gastroenterology 2021; 160: 2317–2327.e2.

13.

Lee

Kim

Sung

I-K

, et al. Effect of submucosal fibrosis on endoscopic submucosal dissection of colorectal tumors: pathologic review of 173 cases. J Gastroenterol Hepatol 2015; 30: 872–878.

14.

Inada

Yoshida

Kugai

, et al. Prediction and treatment of difficult cases in colorectal endoscopic submucosal dissection. Gastroenterol Res Pract 2013; 2013: 523084.

15.

H-H

Jung

Y-W

Jin

B-C

, et al. Predictive factors associated with technical difficulty in colorectal endoscopic submucosal dissection: a Honam Association for the Study of Intestinal Disease (HASID) multicenter study. Medicine (Baltimore) 2024; 103: e37936.

16.

Nagtegaal

Odze

Klimstra

, et al. The 2019 WHO classification of tumours of the digestive system. Histopathology 2020; 76: 182–188.

17.

Tanaka

Kashida

Saito

, et al. Japan Gastroenterological Endoscopy Society guidelines for colorectal endoscopic submucosal dissection/endoscopic mucosal resection. Dig Endosc 2020; 32: 219–239.

18.

Sano

Tanaka

Kudo

S-E

, et al. Narrow-band imaging (NBI) magnifying endoscopic classification of colorectal tumors proposed by the Japan NBI Expert Team. Dig Endosc 2016; 28: 526–533.

19.

Yoshida

Hisabe

Inada

, et al. The ability of a novel blue laser imaging system for the diagnosis of invasion depth of colorectal neoplasms. J Gastroenterol 2014; 49: 73–80.

20.

Kudo

Tamura

Nakajima

, et al. Diagnosis of colorectal tumorous lesions by magnifying endoscopy. Gastrointest Endosc 1996; 44: 8–14.

21.

Kobayashi

Yoshida

Morinaga

, et al. The comparison of diagnostic ability between blue laser/light imaging and narrowband imaging for sessile serrated lesions with or without dysplasia. Gastroenterol Res Pract 2024; 2024: 2672289.

22.

Vandenbroucke

von Elm

Altman

, et al. Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): explanation and elaboration. PLoS Med 2007; 4: e297.

23.

Shen

Yoshida

Kobayashi

, et al. Efficacy of two kinds of scissor-type knives for colorectal endoscopic submucosal dissection: a retrospective comparative study. Dig Dis Sci 2024; 69: 4214–4223.

24.

Yoshida

Hirose

Dohi

, et al. A novel reopenable clip with sharp claw for complete closure of mucosal defects after colorectal endoscopic submucosal dissection. Endoscopy 2025; 57: 354–360.

25.

Participants in the Paris Workshop. The Paris endoscopic classification of superficial neoplastic lesions: esophagus, stomach, and colon: November 30 to December 1, 2002. Gastrointest Endosc 2003; 58: S3–43.

26.

Matsumoto

Tanaka

Oba

, et al. Outcome of endoscopic submucosal dissection for colorectal tumors accompanied by fibrosis. Scand J Gastroenterol 2010; 45: 1329–1337.

27.

Chow

CWS

Fung

TLD

Chan

, et al. Endoscopic submucosal dissection for colorectal polyps: outcome determining factors. Surg Endosc 2022; 37: 1293–1302.

28.

Nakajima

Saito

Tanaka

, et al. Current status of endoscopic resection strategy for large, early colorectal neoplasia in Japan. Surg Endosc 2013; 27: 3262–3270.

29.

Pellise

Burgess

Tutticci

, et al. Endoscopic mucosal resection for large serrated lesions in comparison with adenomas: a prospective multicentre study of 2000 lesions. Gut 2017; 66: 644–653.

30.

Gupta

East

. Optimal endoscopic treatment and surveillance of serrated polyps. Gut Liver 2020; 14: 423–429.

31.

Rao

Soetikno

Raju

, et al. Large sessile serrated polyps can be safely and effectively removed by endoscopic mucosal resection. Clin Gastroenterol Hepatol 2016; 14: 568–574.

32.

Rex

Vemulapalli

Rex

. Recurrence rates after EMR of large sessile serrated polyps. Gastrointest Endosc 2015; 82: 538–541.

33.

Chiba

Tachikawa

Arimoto

, et al. Predictive factors of mild and severe fibrosis in colorectal endoscopic submucosal dissection. Dig Dis Sci 2020; 65: 232–242.

34.

Cecinato

Lisotti

Azzolini

, et al. Left colonic localization, non-granular morphology, and pit pattern independently predict submucosal fibrosis of colorectal neoplasms before endoscopic submucosal dissection. Endoscopy 2023; 55(S02): S45.

35.

Chiba

Ohata

Hayashi

, et al. Outcomes in colorectal endoscopic submucosal dissection for large protruded lesions: a retrospective multicenter study. Endosc Int Open 2024; 12: E757–E763.

36.

Sferrazza

Maida

Calabrese

, et al. The derivation and external validation of a fibrosis risk model for colorectal tumours undergoing endoscopic submucosal dissection. J Clin Med 2024; 13(15): 4517.

37.

Kaosombatwattana

Yamamura

Limsrivilai

, et al. Preoperative endoscopic predictors of severe submucosal fibrosis in colorectal tumors undergoing endoscopic submucosal dissection. Endosc Int Open 2019; 7: E421–E430.

38.

Matsumoto

Uehara

Mashima

. Construction of a preoperative scoring system to predict the difficulty level of colorectal endoscopic submucosal dissection. PLoS One 2019; 14: e0219096.

39.

Dang

Dekkers

Steyerberg

, et al. Predicting procedure duration of colorectal endoscopic submucosal dissection at Western endoscopy centers. Endosc Int Open 2023; 11: E724–E732.

40.

Shi

E-P

, et al. Prediction of technically difficult endoscopic submucosal dissection for large superficial colorectal tumors: a novel clinical score model. Gastrointest Endosc 2021; 94: 133–144.e3.

41.

Kuroha

Shiga

Kanazawa

, et al. Factors associated with fibrosis during colorectal endoscopic submucosal dissection: does pretreatment biopsy potentially elicit submucosal fibrosis and affect endoscopic submucosal dissection outcomes? Digestion 2021; 102: 590–598.

42.

Jung

Hwangbo

Cho

, et al. Is colorectal endoscopic submucosal dissection safe and effective for 15–19-mm tumors? Int J Colorectal Dis 2023; 38: 206.

43.

Jacques

Legros

Wallenhorst

, et al. Prediction of technically difficult, colorectal, endoscopic submucosal dissection: is the procedural time a good endpoint? Gastrointest Endosc 2021; 93: 1435–1436.

44.

IJspeert

JEG

Bastiaansen

BAJ

van Leerdam

, et al. Development and validation of the WASP classification system for optical diagnosis of adenomas, hyperplastic polyps and sessile serrated adenomas/polyps. Gut 2016; 65: 963–970.

45.

Fukunaga

Nagami

Shiba

, et al. Impact of preoperative biopsy sampling on severe submucosal fibrosis on endoscopic submucosal dissection for colorectal laterally spreading tumors: a propensity score analysis. Gastrointest Endosc 2019; 89: 470–478.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.08 MB