Impact of margin thermal ablation after cold-forceps avulsion with snare-tip soft coagulation for nonlifting large nonpedunculated colorectal polyps

Abstract

Background

Nonlifting large nonpedunculated colorectal polyps (NL-LNPCPs) account for 15% of LNPCPs and are effectively managed by endoscopic mucosal resection (EMR) with adjunctive cold-forceps avulsion with adjuvant snare-tip soft coagulation (CAST). Recurrence rates >10% at surveillance colonoscopy are however a significant limitation. We aimed to compare the outcomes of CAST plus margin thermal ablation (MTA) versus CAST alone for NL-LNPCPs.

Methods

Prospective observational data on consecutive patients with NL-LNPCPs treated by EMR and CAST at a single tertiary center were retrospectively evaluated. Two cohorts were established: the pre-MTA period (January 2012–June 2017) and the MTA period (July 2017–October 2023). The primary outcome was the residual/recurrent adenoma (RRA) rate at first surveillance colonoscopy (SC1). Secondary outcomes included the RRA rate at SC2 and adverse events.

Results

Over 142 months, 300 patients underwent EMR and CAST for LNPCP: 103 lesions pre-MTA and 197 with MTA. At SC1 and SC2, recurrence was lower in the MTA cohort compared with the pre-MTA cohort (5.0% vs. 18.8% and 0.8% vs. 10.0%, respectively; both P<0.001). Adverse events were similar between the two cohorts for deep mural injury types III–V (pre-MTA 2.9% vs. MTA 5.6%; P=0.29) and delayed bleeding (pre-MTA 8.7% vs. MTA 7.1%; P=0.49). On multivariate analysis, MTA was the only variable independently associated with a reduced likelihood of recurrence (odds ratio 0.20, 95%CI 0.07–0.54; P=0.001).

Conclusions

For NL-LNPCPs, MTA in combination with CAST is safe and effective and reduces recurrence at SC1 in comparison with CAST alone.

 

MissingLinkText

 

MissingLinkText

Introduction

Endoscopic mucosal resection (EMR) is the treatment of choice for large (≥20 mm) nonpedunculated colorectal polyps (LNPCPs) within international consensus guidelines [1] [2]. Consistent and robust evidence demonstrates EMR to be a safe and effective resection modality for the majority of LNPCPs [3] [4] [5] [6] [7]. Moreover, EMR is safer, more cost-effective, and less resource-intensive than endoscopic submucosal dissection (ESD) or colorectal surgery [8] [9] [10].

Post-EMR residual or recurrent adenoma (RRA) was historically a major limitation of the technique, with rates exceeding 15% [11] [12]. The introduction of margin thermal ablation (MTA) using snare-tip soft coagulation has been transformative for EMR practice, significantly reducing the post-EMR RRA rate at the first surveillance colonoscopy (SC1) to <2% in a multicenter Australian cohort [13]. Data from other centers have shown rates between 4% and 7% [14] [15].

Nonlifting LNPCPs (NL-LNPCPs) account for approximately 15% of all LNPCPs [16]. These lesions are characterized by submucosal fibrosis, either due to their native biology or as a result of previous manipulation. Submucosal fibrosis hinders endoscopic resection by preventing the expansion of the submucosal plane, making snare capture challenging and increasing the risk of deep mural injury (DMI) [16].

While surgery was traditionally required for NL-LNPCPs, safe and effective endoscopic treatments have subsequently emerged. Cold-forceps avulsion with adjuvant snare-tip soft coagulation (CAST) avoids the need for surgery in all cases, with perforations occurring in only 3%; however, endoscopic recurrence remains a significant limitation, with RRA at SC1 in >15% of cases [16]. Similarly to conventional EMR, MTA is now routinely incorporated into the CAST technique, but whether this translates into reduced recurrence rates is currently unknown.

This study aimed to compare the outcomes of CAST with MTA versus CAST alone (performed in the pre-MTA era) for NL-LNPCPs

 

MissingLinkText

Methods

Study design

This manuscript was written in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines [17].

A retrospective analysis was conducted on prospectively collected data from consecutively enrolled patients at a single center from January 2012 to October 2023. Data were entered into the Australian Colonic Endoscopic (ACE) Resection Database, a prospective observational study of patients referred for endoscopic resection of LNPCPs. Institutional review board approval was obtained from the Western Sydney local health district human research ethics committee. Written informed consent was obtained from all included patients.

The inclusion criteria were patients with NL-LNPCPs, defined as lesions with submucosal fibrosis, who were treated by EMR plus CAST. At our institution, CAST is the gold-standard technique for LNPCPs that are deemed resectable by EMR if technical difficulty arises owing to nonlifting characteristics in the absence of optically diagnosed deep submucosal invasive cancer (SMIC). The presence of fibrosis alone was not an indication for ESD, which was selectively performed on LNPCPs that were suspicious for superficial submucosal cancer based on morphology, location, or their surface pit and vascular pattern. LNPCPs requiring ESD were excluded from the analysis.

Excluded LNPCPs included: those with incomplete MTA, sessile serrated lesions, and inflammatory bowel disease (IBD)-related dysplasia ([Fig. 1]). The optimal resection modality for IBD-related dysplasia remains unclear, whereas serrated lesions have been conclusively demonstrated to be safely and effectively resected with cold snare techniques [18].

Fig. 1 Flow of patients through the study. CAST, cold-forceps avulsion with snare-tip soft coagulation; CSP, cold snare polypectomy; EMR, endoscopic mucosal resection; ER, endoscopic resection; ESD, endoscopic submucosal dissection; IBD, inflammatory bowel disease; LNPCP, large nonpedunculated colorectal polyp; MTA, margin thermal ablation; NL-LNPCP, nonlifting LNPCP; SC1, first surveillance colonoscopy; SC2, second surveillance colonoscopy; SMIC, submucosal invasive cancer; SSL, sessile serrated lesion.

Patients with post-EMR SMIC histology or referred to surgery for unfavorable histology were excluded from the follow-up.

Two cohorts were established: patients treated in the pre-MTA period (January 2012–June 2017) and those treated in the MTA period (July 2017–October 2023).

 

MissingLinkText

EMR and the CAST technique

Endoscopic procedures were performed by either a study investigator (an accredited gastroenterologist with advanced training at a tertiary referral practice in colorectal endoscopic resection) or a senior interventional endoscopy fellow under supervision. Study investigators were consistent across the entire study period.

Standard split-dose bowel preparation was administered 24 hours preprocedure and discontinuation of antiplatelet and anticoagulation agents was conducted in accordance with consensus guidelines [19] [20]. Conscious sedation was provided with a combination of intravenous propofol, midazolam, and fentanyl. Colonoscopy was performed using Olympus 180 and 190 series variable-stiffness colonoscopes (180/190 PCF/CF; Olympus, Tokyo, Japan) during both periods, with carbon dioxide insufflation. Optical assessment of the lesion was carried out using high definition white-light endoscopy (HD-WLE) and narrow-band imaging (NBI). Near-focus technology is available only on the 190 PCF/CF series and was used at the operator's discretion.

The resection of NL-LNPCPs consisted of two parts. Initially, a standard inject and resect EMR technique was employed, as previously described [21]. After a submucosal cushion had been created with an injection of succinylated gelatin (Gelofusine; B. Braun, Bella Vista, Australia) mixed with 0.4% indigo carmine and 1:100000 epinephrine, snare excision was systematically performed. A microprocessor-controlled generator (ERBE VIO; ERBE, Tübingen, Germany), using Endocut Q, effect 3, was used for snare resection to remove all snareable tissue. The nonlifting area was then isolated and treated with the CAST technique. As previously described, systematic tissue avulsion was performed with cold-forceps avulsion (Radial Jaw biopsy forceps; Boston Scientific, Massachusetts, USA). Once all visible tissue had been completely avulsed, all cases received controlled thermal ablation with snare-tip soft coagulation (ERBE VIO; soft coagulation, 80 W, effect 4) to the exposed submucosa at the avulsion site, using the extended tip (1–2 mm) of the same snare that was used for EMR. Only the CAST-treated area was ablated [16].

Resection defects were inspected, extensively photodocumented, and classified according to the Sydney classification of DMI [22]. Type II–V defects were closed with through-the-scope (TTS) clips. For procedures performed prior to 2017, defect images were reviewed and reclassified according to the corresponding DMI score.

For the MTA cohort, ablation of the defect margins with snare-tip soft coagulation was performed [13]. The tip of the snare was systematically applied to the defect margin using a light-touch technique, aiming to create a rim of at least 2–3 mm of completely ablated tissue ([Fig. 2] and [Fig. 3]).

Fig. 2 Endoscopic images during endoscopic mucosal resection (EMR) and cold-forceps avulsion with snare-tip soft coagulation (CAST) of a 35-mm previously attempted polyp in the transverse colon showing: a optical assessment of the lesion using high definition white-light endoscopy; b the appearance after submucosal injection of succinylated gelatin mixed with indigo carmine and epinephrine; c the appearance after snare excision of the lifted regions; d cold-forceps avulsion of all visible nonlifting adenoma; e a through-the-scope (TTS) clip placed onto the brownish-white eschar after the application of soft coagulation, with the defect classified as type II according to the Sydney classification of deep mural injury; f the final appearance after thermal ablation of the margins using snare-tip soft coagulation and closure of the defect following treatment with CAST using TTS clips.

Fig. 3 Endoscopic images during endoscopic mucosal resection (EMR) and cold-forceps avulsion with snare-tip soft coagulation (CAST) of a 25-mm previous manipulated polyp at the hepatic flexure showing: a,b lesion assessment using: a high definition white-light endoscopy; b narrow-band imaging; c–f the appearance of: c the lifting area after submucosal injection; d the lesion after snare excision of the lifted polyp; e CAST of the nonlifting adenoma; f the defect after closure using clips.

Post-procedure, patients were observed for 4 hours and then discharged home if well. A clear fluid diet was advised until the next morning. Histological samples were assessed by specialist gastrointestinal pathologists.

 

MissingLinkText

Follow-up

Patients were contacted by a study coordinator 2 weeks post-procedure, and a structured telephone interview was conducted to identify post-procedural adverse events. SC1 was scheduled at 6 months after the index resection. The scar was assessed using HD-WLE, NBI and near focus, when available.

Initially routine biopsies were taken from the EMR scar. However, after the development of the standardized imaging protocol for the evaluation of scars in 2017 [23], biopsies of bland EMR scars were not consistently performed. In the presence of a bland EMR scar at SC1, SC2 was scheduled for 18 months after the index resection.

 

MissingLinkText

Data collection and outcomes

Patient baseline demographics (age and biological gender) were recorded in addition to pre-procedural data on previous biopsy or resection attempts, as well as Spot tattoo marking. LNPCP characteristics were captured including size, surface granularity, Paris classification, Kudo pit pattern, location, and histopathological diagnosis. Complex location was defined as LNPCPs located at colonic flexures or involving the ileocecal valve, anorectal junction, or appendiceal orifice.

Intraprocedural data included intraprocedural perforation and procedure duration. Intraprocedural perforation was defined as Sydney DMI classification grade III–V [22].

Post-procedural data included technical success and adverse events (AEs), including clinically significant post-EMR bleeding or delayed perforation. Technical success was defined as complete excision of all visible adenomatous tissue. Clinically significant post-EMR bleeding was any bleeding that occurred within 14 days of resection that required emergency department admission, hospitalization, or intervention.

The primary outcome was the RRA rate at SC1 for NL-LNPCPs treated with CAST and adjuvant MTA compared with CAST procedures with thermal therapy limited to the defect base. The early RRA rate was assessed with and without NL-LPNCPs at the ileocecal valve, whose treatment is known to be associated with higher rates of RRA, even in the era of new advances in EMR [24].

Secondary outcomes included the rates of RRA at SC2, technical success, and AEs, including post-procedural pain, intraprocedural perforation, clinically significant post-EMR bleeding, and delayed perforation. RRA at the surveillance colonoscopies (SC1 and SC2) was based on endoscopic and histopathological findings.

 

MissingLinkText

Statistical analysis

SPSS software, version 29.0 (IBM, Armonk, New York, USA) was used for data analysis. Continuous variables were summarized using median and interquartile range (IQR) and tested with the Shapiro–Wilk test. Categorical variables were summarized as frequencies and percentages. To test for association between categorical variables, the Pearson chi-squared or the Fisher’s exact tests were used, where appropriate. For continuous variables, the Mann–Whitney U test was used. For baseline comparison, a P value <0.05 was considered as statistically significant.

To exclude selection bias, the baseline characteristics of the patients in the pre-MTA and MTA cohorts with endoscopic follow-up (SC1 and SC2) were compared with patients who had dropped out by the respective follow-ups. A P value <0.05 was considered statistically significant.

The 95%CIs were constructed for proportions. The bootstrap method was used for continuous variables, and the Wilson method was applied for categorical variables.

To explore the independent association of the treatment groups with the primary outcome, multivariable adjusted regression models were implemented using a stepwise approach. Potential confounders were selected iteratively based on their significance in univariate analysis. Measures of association were expressed as odds ratios (ORs) or adjusted ORs (adjORs), along with their 95%CIs.

 

MissingLinkText

 

MissingLinkText

Results

From January 2012 to October 2023, 2639 LNPCPs were referred for endoscopic resection. EMR was attempted in 2239 LNPCPs. Of these, 314 NL-LNPCPs in 314 patients (14%) underwent EMR and CAST. After exclusion criteria had been applied, 300 NL-LNPCPs were identified: 103 NL-LNPCPs in the pre-MTA period and 197 after the implementation of MTA ([Fig. 1]).

Patient and lesion characteristics

Patient and lesion characteristics are summarized in [Table 1]. Patient age was similar between the two groups (pre-MTA 74 years vs. MTA 70 years; P=0.05), as was the size of NL-LNPCPs (pre-MTA 30 mm vs. MTA 35 mm; P=0.30).

Table 1 Patient and lesion characteristics for the included patients who underwent endoscopic mucosal resection (EMR) and cold-forceps avulsion with adjuvant snare-tip soft coagulation (CAST) for a nonlifting large nonpedunculated colorectal polyp.

	Pre-MTA cohort n = 103	MTA cohort n = 197	P value
IQR, interquartile range; MTA, margin thermal ablation.
Age, median (IQR) [95%CI], years	74 (66–78) [73, 74, 75]	70 (62–76) [69, 70, 71]	0.05
Sex, female, n (%) [95%CI]	50 (48.5) [39.1–58.1]	77 (39.1) [32.5–46.0]	0.12
Polyp size, median (IQR) [95%CI], mm	30 (20–40) {27–32]	35 (25–40) [34, 35, 36]	0.30
Previous biopsy, n (%) [95%CI]	38 (36.9) [28.2–46.5]	110 (55.8) [48.9–62.6]	0.02
Previous EMR attempt, n (%) [95%CI]	39 (37.9) [29.1–47.5]	96 (48.7) [41.8–55.7]	0.07
Previous Spot marking within 10 mm, n (%) [95%CI]	24 (23.3) [16.2–32.3]	36 (18.3) [13.5–24.3]	0.68
Complex location, n (%) [95%CI]	27 (26.2) [18.7–35.5]	51 (25.9) [20.3–32.4]	0.39
Paris classification, n (%) [95%CI]			0.10
Is	2 (1.9) [0.5–6.8]	10 (5.1) [2.8–9.1]
IIa	73 (70.9) [61.5–78.8]	131 (66.5) [59.6–72.7]
IIb	8 (7.8) [4.0–14.6]	7 (3.6) [1.7–7.2]
IIa + Is	16 (15.5) [9.8–23.8]	45 (22.8) [17.5–29.2]
Other	4 (3.9) [1.5–9.6]	4 (2.0) [0.8–5.1]
Morphology, n (%) [95%CI]			0.01
Granular	43 (41.7) [32.7–51.4]	123 (62.4) [55.5–68.9]
Nongranular	52 (50.5) [41.0–59.9]	62 (31.5) [25.4–38.3]
Mixed	6 (5.9) [2.7–12.1]	10 (5.1) [2.8–9.1]
Unable to be classified	2 (1.9) [0.5–6.8]	2 (1.0) [0.3–3.6]

The percentage of LNPCPs that had undergone previous biopsy was higher in the MTA cohort (pre-MTA 36.9% vs. MTA 55.8%; P = 0.02). LNPCPs were predominately Paris classification 0-IIa (pre-MTA 70.9% vs. MTA 66.5%; P=0.10), with a higher proportion of granular lesions in the MTA group (pre-MTA 41.7% vs. MTA 62.4%, P=0.01).

 

MissingLinkText

Endoscopic and histological data

Difficult lesion access was similar between the two cohorts (pre-MTA 32% vs. MTA 37%; P=0.43). Rates of histological low grade dysplasia (pre-MTA 77.7% vs. MTA 78.7%; P=0.95) and SMIC (pre-MTA 5.8% vs. MTA 5.6%; P=0.95) were comparable between the groups ([Table 2]).

Table 2 Endoscopic and histological data for the included patients.

	Pre-MTA cohort n = 103	MTA cohort n = 197	P value
DMI, deep mural injury; HGD, high grade dysplasia; IQR interquartile range; LGD, low grade dysplasia; MTA, margin thermal ablation.
Difficult to reach and/or position endoscope for access, n (%) [95%CI]	33 (32.0) [23.8–41.6]	73 (37.0) [30.6–44.0]	0.43
Submucosal fibrosis, n (%) [95%CI]	103 (100) [96.4–100]	197 (100) [98.1–100]	1.00
DMI type II, n (%) [95%CI]	16 (15.5) [9.8–23.8]	87 (44.2) [37.4–51.1]	<0.001
Clip closure of endoscopic resection defect, n (%) [95%CI]	22 (21.4) [14.5–30.2]	123 (62.4) [55.5–68.9]	<0.001
Procedure duration, median (IQR) [95%CI], minutes	35 (25–50) [33, 34, 35, 36, 37]	60 (40–72.5) [58, 59, 60, 61, 62]	<0.001
Histopathology, n (%) [95%CI]			0.02
Tubular adenoma	40 (38.9) [30.0–48.5]	52 (26.4) [20.7–33.0]
Tubulovillous adenoma	57 (55.3) [45.7–64.6]	133 (67.5) [60.7–73.7]
Dysplasia, n (%) [95%CI]			0.95
LGD	80 (77.7) [68.7–84.6]	155 (78.7) [72.4–83.8]
HGD	17 (16.5) [10.6–24.9]	30 (15.2) [10.9–20.9]
Invasive cancer, n (%) [95%CI]	6 (5.8) [2.7–12.1]	11 (5.6) [3.1–9.7]	0.95

In the MTA group, the rate of DMI type II was higher (pre-MTA 15.5% vs. MTA 44.2%; P<0.001), with a consequently higher proportion of defect clip closure (pre-MTA 21.4% vs. MTA 62.4%; P<0.001). The procedure duration was higher in the MTA cohort (pre-MTA 35 [IQR 25–50] minutes vs. MTA 60 [IQR 40–72.5]; P<0.001).

 

MissingLinkText

Surveillance outcomes and RRA

First surveillance colonoscopy (SC1)

There were 87 patients (89.7%) from the pre-MTA group and 159 patients (85.5%) from the MTA group who underwent SC1 at a median 5.0 months and 6.0 months, respectively ([Fig. 1]). The recurrence rate was higher in the pre-MTA group compared with the MTA group (16/87 [18.4%] vs. 8/159 [5.0%], respectively; P<0.001). This result remained statistically significant after applying the Bonferroni correction ([Table 3]). Excluding ICV lesions, the RRA rate at SC1 in the MTA cohort was significantly lower than in the pre-MTA cohort (2.8% vs. 18.8%; P<0.001).

Table 3 Surveillance outcomes and adverse event rates.

	Pre-MTA cohort n = 103	MTA cohort n = 197	P value
DMI, deep mural injury; ICV, ileocecal valve; MTA, margin thermal ablation; SC1, first surveillance colonoscopy; SC2, second surveillance colonoscopy.
Recurrence at SC1, n (%) [95%CI]	16/87 (18.4) [11.6–27.8]	8/159 (5.0) [2.6–9.6]	<0.001
Recurrence at SC1, excluding at ICV, n (%) [95%CI]	15/80 (18.8) [11.7–28.7]	4/141 (2.8) [1.1–7.1]	<0.001
Recurrence at SC2, n (%) [95%CI]	7/70 (10.0) [5.9–21.0]	1/125 (0.8) [0.1–4.4]	<0.001
Significant pain requiring post-procedure analgesia, n (%) [95%CI]	4 (3.8) [1.5–9.6]	9 (4.6) [2.4–8.5]	0.82
Intraprocedural perforation (DMI III–V), n (%) [95%CI]	3 (2.9) [1.0–8.2]	11 (5.6) [3.1–9.7]	0.29
Delayed bleeding, n (%) [95%CI]	9 (8.7) [4.7–15.8]	14 (7.1) [4.3–11.6]	0.49
Delayed perforation, n (%) [95%CI]	0 (0) [0–3.6]	0 (0) [0–1.9]	>0.99
Technical success, n (%) [95%CI]	103 (100) [96.4–100]	197 (100) [98.1–100]	>0.99

 

MissingLinkText

Second surveillance colonoscopy (SC2)

There were 70 (72.2%) of the pre-MTA group and 125 (67.2%) of the MTA group who underwent SC2 at median time of 20 months ([Fig. 1]). The recurrence rate was higher in the pre-MTA group (7/70 [10.0%] vs. 1/125 [0.8%], respectively; P<0.001), which remained statistically significant after applying the Bonferroni correction. The only late RRA in the MTA cohort was a resistant RRA that had been previously detected and treated at SC1.

 

MissingLinkText

 

MissingLinkText

Safety

Technical success was achieved in all of the treated NL-LNPCPs (300/300; 100%). The rates of significant pain requiring post-procedure analgesia were similar between the groups (pre-MTA 4 [3.8%] vs. MTA 9 [4.6%]; P=0.82).

DMI types III–V and post-procedural bleeding were similar between the two cohorts (pre-MTA 2.9% vs. MTA 5.6% [P=0.29] and 8.7% vs. 7.1% [P=0.49], respectively). All intraprocedural perforations were managed with endoscopic clips, except for one resection defect in the distal rectum in the MTA cohort, which was treated conservatively. Clinically significant post-EMR bleeding was similar between the cohorts (pre-MTA 8.7% vs. MTA 7.1%; P=0.49). Of the 23 cases with clinically significant post-EMR bleeding, 10 required a recheck endoscopy, while 13 were managed conservatively. All patients were discharged without long-term sequelae. No cases of delayed perforation occurred ([Table 3]).

 

MissingLinkText

Predictive factors of early RRA

On univariate logistic regression analysis, lesion size (OR 1.02, 95%CI 1.00–1.05; P=0.03) was associated with RRA at SC1. The use of MTA with snare-tip soft coagulation (OR 0.23, 95%CI 0.01–0.57; P=0.002) was inversely associated with RRA. On multivariate analysis, MTA with snare-tip soft coagulation (adjOR 0.20, 95%CI 0.07–0.54; P=0.001) was the only independent predictor of reduced likelihood of RRA at SC1 (Table 1s, see online-only Supplementary material).

 

MissingLinkText

Missing data analysis

In the pre-MTA cohort, 10 patients had missing SC1 follow-up, and 17 had missing SC2 follow-up. In the MTA cohort, 27 patients had missing SC1 follow-up, and 34 had missing SC2 follow-up.

Baseline demographic, LNPCP, and procedural characteristics were compared between the cohorts for cases with both complete and incomplete follow up (Tables 2s–5s). In the cohort that predated snare-tip soft coagulation, patients without SC2 follow-up had a higher rate of type II defects compared with those with follow-up (6 [35.3%] vs. 7 [10.0%], respectively; P=0.01). No other differences were observed.

 

MissingLinkText

 

MissingLinkText

Discussion

CAST is a simple, safe, and effective technique that has been proven to successfully treat NL-LNPCPs [16]. Importantly, CAST does not require preplanning, making it the ideal rescue therapy for EMR when nonlifting areas are identified during resection.

A key aspect of the technique is the removal of well-lifted normal and/or adenomatous tissue surrounding the nonlifting area with high quality EMR before attempting CAST of the nonlifting polyp. This helps delineate the scarred area, which is usually outlined by the underlying fibrosis, which appears whiter than the normal submucosal layer. Because the lateral margins are free and the scarred area is only attached underneath, it can be excised surprisingly easily by avulsion with biopsy forceps. A methodical approach should be taken to ensure complete removal of all visible nonlifting polyp tissue.

Snare-tip soft coagulation is then applied to the avulsion site, using the tip of the same snare used for EMR. The technique involves a gentle but effective touch of the snare until all potentially visible tissue in the avulsion bed has been denatured. This generally results in a brownish white eschar over the site. Generally, the avulsion bed is bloody; it is advisable to start ablating the most superior bleeding site, which can be identified according to the “fan principle” in the opposite direction to where the fluid is pooling and blood spreading towards.

In the preliminary experience involving 101 NL-LNPCPs (38 previously attempted nonlifting lesions and 63 naïve nonlifting lesions [median size 27.5 mm]) treated between January 2012 and December 2016, CAST was successful in removing nonlifting tissue in all cases (101/101; 100%); however, the recurrence rate at SC1 was over 15% [16].

This study has compared the outcomes of CAST for the treatment of NL-LNPCPs between an historical pre-MTA cohort (January 2012–June 2017) and those treated with conventional therapy with adjuvant MTA (June 2017–October 2023). CAST achieved 100% technical success in both cohorts and, in the MTA era, led to a significantly lower rate of RRA at SC1 (5% vs. 19%; P<0.001). Moreover, multivariate analysis showed that MTA was the only independent factor associated with RRA reduction at SC1 (OR 0.20, 95%CI 0.07–0.54; P=0.001).

In the MTA cohort, early RRA was successfully treated endoscopically with CAST in 7/8 cases at SC1. The only lesion not eradicated at SC1 was a 30-mm rectal polyp referred to our center after two previous treatments, making resection very challenging at the index EMR. No de novo late RRAs were observed after a clear SC1 in the MTA cohort, which has crucial implications for surveillance schedules, as the surveillance intervals could likely be safely extended. We have recently demonstrated this in a large cohort of conventional LNPCPs [25].

We acknowledge that comparing the two cohorts may be susceptible to “time effect” bias. In the MTA cohort, factors such as advancements in endoscopy technology and the learning curve related to the technique could have contributed to the reduction in RRAs.

Recent data have however provided robust evidence on this matter. In a recently published study, we evaluated the impact of MTA on the RRA rate after the resection of ≥40-mm LNPCPs [26]. We defined three periods: “pre-MTA” (before the introduction of MTA; July 2008–June 2012); “MTA adoption” (training MTA phase; July 2012–June 2017); and “standardized MTA” (after the introduction of MTA into practice; July 2017–July 2023). While the overall RRA rate was significantly lower in the standardized MTA phase compared with the pre-MTA and MTA-adoption phases (2.1% vs. 13.5% and 12.6%, respectively; P<0.001), the RRA rate was consistent across the three periods where MTA was not applied (P=0.20). This indicates the stability of the EMR technique's effectiveness over time and suggests that the pivotal innovation in reducing RRA is MTA, rather than an improvement in the EMR technique. Additionally, in this study, the use of MTA eliminated the relevance of size as a risk factor for RRA.

The impact of the evolution of the CAST technique on RRA rate in the MTA cohort cannot be entirely excluded. The MTA cohort had a higher rate of type II defects compared with the pre-MTA cohort (45.7% vs. 15.5%), highlighting the evolution of the treatment and the use of more aggressive ablation to ensure effectiveness. Although initially less common, in the past 2 years, it has become our practice to create type II defects after CAST ablation for polyp treatment and close them with clips. This is further evidenced by the higher rate of clip closure in the MTA cohort (62.4% vs. 21.4%).

The MTA cohort demonstrated a higher rate of previously biopsied lesions (55.8% vs. 36.9%; P=0.02). This reflects the endoscopic practice of referring physicians rather than what is routinely performed within our center. Furthermore, LNPCPs in the MTA cohort had a greater proportion of granular morphology (62.4% vs. 41.7%; P=0.01), aligning with the findings of previous NL-LNPCP studies [27].

In the MTA cohort, the rate of significant intraprocedural DMI was higher compared with the pre-MTA cohort (5.6% vs. 2.9%), although this difference was not statistically significant. This can be explained by the increasingly aggressive ablation technique and technical confidence in DMI clip closure. The MTA group had a higher number of lesions previously subjected to EMR and with difficult endoscopic access, which, even if not statistically significant, may have influenced the results. Additionally, the increasing confidence of endoscopists with the Sydney classification in the contemporary era likely contributed to better identification of events. All intraprocedural perforations were successfully managed conservatively or with endoscopic clips. There were no cases of delayed perforation.

Similarly, the rates of clinically significant post-EMR bleeding were similar between the two cohorts. We recognize that routine prophylactic defect closure in the right colon, introduced after 2022 [28], may have influenced these findings.

We found a longer procedural time in the MTA cohort compared with the pre-MTA cohort. This can be explained by various factors. Among these are the refinement of the technique, which includes MTA and more decisive CAST treatment, as mentioned above. The higher number of challenging lesions in the MTA era may also have contributed. Additionally, in the pre-MTA era, when CAST was first introduced, procedures were almost exclusively performed by senior endoscopists. In contrast, in the MTA era, when CAST had been established as the mainstay technique, procedures were mostly carried out independently by fellows under supervision, which added additional time. Unfortunately, we do not have specific granular data for this. Clipping may also have added time.

The strengths of this study include the large cohort of NL-LNPCPs treated with CAST and EMR, as well as the prospective data collection and strong compliance with surveillance. However, several limitations bear consideration. The improvement in the CAST technique over time may have influenced the outcomes. Differences in some baseline characteristics between the two cohorts must also be acknowledged, including the rate of previously biopsied polyps and polyp morphology. While these variables neither make the cohorts fundamentally different nor affect the outcomes, based on our experience and the literature produced by our group, it should be noted they also did not demonstrate any influence on early RRA in the regression modelling. The only identifiable factor associated with a reduction in RRA was the implementation of MTA with CAST.

In conclusion, our study shows that, in a large cohort of 300 NL-LNPCPs, CAST was technically successful in all cases. The addition of MTA reduced the RRA rate at SC1 to 5%. If the scar is clean at SC1, late recurrence is rare. These data position EMR with CAST and MTA as the ideal tool for the treatment of NL-LNPCPs. It is simple, cost-effective, and a seamless extension of the routine EMR technique without the need for extensive preprocedural planning.

 

MissingLinkText

 

MissingLinkText

Conflict of Interest

M.J. Bourke has received research support from Olympus, Cook Medical, and Boston Scientific. N.G. Burgess, O. Cronin, J.L. Gauci, S. Gupta, C. Kerrison, B. Lam, E.Y. Lee, F.V. Mandarino,,R. Medas, T. O’Sullivan, V. Perananthan, D.J. Tate, and A. Whitfield declare that they have no conflict of interest.

• References

• 1 Ferlitsch M, Hassan C, Bisschops R. et al. Colorectal polypectomy and endoscopic mucosal resection: European Society of Gastrointestinal Endoscopy (ESGE) Guideline – Update 2024. Endoscopy 2024; 56: 516-545
• 2 Kaltenbach T, Anderson JC, Burke CA. et al. Endoscopic removal of colorectal lesions - Recommendations by the US Multi-Society Task Force on Colorectal Cancer. Gastroenterology 2020; 158: 1095-1129
• 3 Tate DJ, Desomer L, Awadie H. et al. EMR of laterally spreading lesions around or involving the appendiceal orifice: technique, risk factors for failure, and outcomes of a tertiary referral cohort (with video). Gastrointest Endosc 2018; 87: 1279-1288 e2
• 4 Nanda KS, Tutticci N, Burgess NG. et al. Endoscopic mucosal resection of laterally spreading lesions involving the ileocecal valve: technique, risk factors for failure, and outcomes. Endoscopy 2015; 47: 710-718
• 5 Shahidi N, Sidhu M, Vosko S. et al. Endoscopic mucosal resection is effective for laterally spreading lesions at the anorectal junction. Gut 2020; 69: 673-680
• 6 Tutticci N, Klein A, Sonson R. et al. Endoscopic resection of subtotal or completely circumferential laterally spreading colonic adenomas: technique, caveats, and outcomes. Endoscopy 2016; 48: 465-471
• 7 Mandarino FV, Danese S, Uraoka T. et al. Precision endoscopy in colorectal polyps' characterization and planning of endoscopic therapy. Dig Endosc 2024; 36: 761-777
• 8 Bahin FF, Heitman SJ, Rasouli KN. et al. Wide-field endoscopic mucosal resection versus endoscopic submucosal dissection for laterally spreading colorectal lesions: a cost-effectiveness analysis. Gut 2018; 67: 1965-1973
• 9 Ahlenstiel G, Hourigan LF, Brown G. et al. Actual endoscopic versus predicted surgical mortality for treatment of advanced mucosal neoplasia of the colon. Gastrointest Endosc 2014; 80: 668-676
• 10 Jayanna M, Burgess NG, Singh R. et al. Cost analysis of endoscopic mucosal resection vs surgery for large laterally spreading colorectal lesions. Clin Gastroenterol Hepatol 2016; 14: 271-278 e82
• 11 Moss A, Williams SJ, Hourigan LF. et al. Long-term adenoma recurrence following wide-field endoscopic mucosal resection (WF-EMR) for advanced colonic mucosal neoplasia is infrequent: results and risk factors in 1000 cases from the Australian Colonic EMR (ACE) study. Gut 2015; 64: 57-65
• 12 Hassan C, Repici A, Sharma P. et al. Efficacy and safety of endoscopic resection of large colorectal polyps: a systematic review and meta-analysis. Gut 2016; 65: 806-820
• 13 Sidhu M, Shahidi N, Gupta S. et al. Outcomes of thermal ablation of the mucosal defect margin after endoscopic mucosal resection: a prospective, international, multicenter trial of 1000 large nonpedunculated colorectal polyps. Gastroenterology 2021; 161: 163-170 e3
• 14 Rex DK, Haber GB, Khashab M. et al. Snare tip soft coagulation vs argon plasma coagulation vs no margin treatment after large nonpedunculated colorectal polyp resection: a randomized trial. Clin Gastroenterol Hepatol 2024; 22: 552-561 e4
• 15 Abu Arisha M, Scapa E, Wishahi E. et al. Impact of margin ablation after EMR of large nonpedunculated colonic polyps in routine clinical practice. Gastrointest Endosc 2023; 97: 559-567
• 16 Tate DJ, Bahin FF, Desomer L. et al. Cold-forceps avulsion with adjuvant snare-tip soft coagulation (CAST) is an effective and safe strategy for the management of non-lifting large laterally spreading colonic lesions. Endoscopy 2018; 50: 52-62
• 17 von Elm E, Altman DG, Egger M. et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. J Clin Epidemiol 2008; 61: 344-349
• 18 van Hattem WA, Shahidi N, Vosko S. et al. Piecemeal cold snare polypectomy versus conventional endoscopic mucosal resection for large sessile serrated lesions: a retrospective comparison across two successive periods. Gut 2021; 70: 1691-1697
• 19 Veitch AM, Vanbiervliet G, Gershlick AH. et al. Endoscopy in patients on antiplatelet or anticoagulant therapy, including direct oral anticoagulants: British Society of Gastroenterology (BSG) and European Society of Gastrointestinal Endoscopy (ESGE) guidelines [published correction appears in Endoscopy 2016; 48: c1]. Endoscopy 2016; 48: 385-402
• 20 Veitch AM, Radaelli F, Alikhan R. et al. Endoscopy in patients on antiplatelet or anticoagulant therapy: British Society of Gastroenterology (BSG) and European Society of Gastrointestinal Endoscopy (ESGE) guideline update. Endoscopy 2021; 53: 947-969
• 21 Klein A, Bourke MJ. How to perform high-quality endoscopic mucosal resection during colonoscopy. Gastroenterology 2017; 152: 466-471
• 22 Burgess NG, Bassan MS, McLeod D. et al. Deep mural injury and perforation after colonic endoscopic mucosal resection: a new classification and analysis of risk factors. Gut 2017; 66: 1779-1789
• 23 Desomer L, Tutticci N, Tate DJ. et al. A standardized imaging protocol is accurate in detecting recurrence after EMR. Gastrointest Endosc 2017; 85: 518-526
• 24 Vosko S, Gupta S, Shahidi N. et al. Impact of technical innovations in EMR in the treatment of large nonpedunculated polyps involving the ileocecal valve (with video). Gastrointest Endosc 2021; 94: 959-968 e2
• 25 O'Sullivan T, Mandarino FV, Gauci JL. et al. Impact of margin thermal ablation after endoscopic mucosal resection of large (≥20 mm) non-pedunculated colonic polyps on long-term recurrence. Gut 2024; 74: 67-74
• 26 Gauci J, Mandarino FV, Kerrison C. et al. Margin thermal ablation eliminates size as a risk factor for recurrence after piecemeal endoscopic mucosal resection. Gut 2025; 2025
• 27 Shahidi N, Vosko S, Gupta S. et al. Previously attempted large nonpedunculated colorectal polyps are effectively managed by endoscopic mucosal resection. Am J Gastroenterol 2021; 116: 958-966
• 28 Gupta S, Sidhu M, Shahidi N. et al. Effect of prophylactic endoscopic clip placement on clinically significant post-endoscopic mucosal resection bleeding in the right colon: a single-centre, randomised controlled trial [published correction appears in Lancet Gastroenterol Hepatol 2022; 7: e2]. Lancet Gastroenterol Hepatol 2022; 7: 152-160

Publication History

Received: 30 June 2024

Accepted after revision: 07 February 2025

Accepted Manuscript online:
07 February 2025

Article published online:
21 March 2025