DOI: https://doi.org/10.21203/rs.3.rs-1961646/v1
Background: The prophylactic application of clips for large pedunculated colorectal polyp resection may prevent postpolypectomy bleeding (PPB) but carries the risk of tumor remnants. This study assessed the effectiveness and safety of underwater endoscopic mucosal resection (UEMR) without clips before resecting large pedunculated polyps.
Methods: Pedunculated polyps with a head diameter of 10-20 mm, a stalk diameter of ≥ 5 mm and a stalk length of ≥ 10 mm were eligible for inclusion and removed by UEMR without prophylactically clipping stem. The primary outcome was the rate of PPB, which included immediate PPB (IPPB) and delayed PPB (DPPB). The secondary outcomes included the rate of en bloc resection, complete resection, R0 resection and other adverse events.
Results: Totally, 28 patients with 32 polyps were included in this study. The head diameter was 14.6 ± 3.5 mm. After polypectomy, two cases (6.3%) of nonactive spurting IPPB were observed, and no DPPB occurred during the follow-up period. All pedunculated polyps were successfully resected and achieved en bloc resection, complete resection and R0 resection. None of the polyps showed evidence of other severe adverse events.
Conclusion: UEMR without the application of prophylactic clips has a low postpolypectomy bleeding rate and low risk of residual tumor and other severe adverse events. Therefore, it might be suitable and safe for resection of 10-20 mm pedunculated polyps.
Trial registration: The study was registered on 08/04/2021 at ClinicalTrials.gov (NCT04837690).
Endoscopic mucosal resection (EMR) is often used as a proven and safe method for resecting benign or malignant lesions in the colon [1, 2]. However, for larger pedunculated polyps (head size ≥ 10 mm), the occurrence of postpolypectomy bleeding (PPB) is common after conventional endoscopic mucosal resection (CEMR), which may be related to the thick feeding vessels in the long stalk and thermal injury [3, 4].
The application of prophylactic clips before snare polypectomy has been shown to reduce PPB and immediate postpolypectomy bleeding (IPPB) for large pedunculated polyps [5–7]. However, the existence of endoclips hinders resection near the base of the stem, which might lead to a high risk of tumor remnants [8]. Therefore, endoscopists should not focus solely on bleeding as an adverse outcome both during and after removal of pedunculated polyps.
Underwater endoscopic mucosal resection (UEMR) was first reported in 2012 as a novel method of filling the intestinal cavity with water to carry the lesion away from the submucosal layer, and at the same time, its heat-sink effect can reduce thermal injury [9–11]. Previous studies have demonstrated that UEMR is an effective and safe alternative to CEMR for the resection of colorectal polyps [12–14], even for large lesions (≥ 10 mm), UEMR shows a significant increase in en bloc resection and R0 resection rates and a reduction in recurrence rate [14, 15]. The total postpolypectomy bleeding rate of UEMR was approximately 3–4%, and the delayed bleeding rate was approximately 2% [12, 16], which was lower than CEMR [17].
To the best of our knowledge, no study of UEMR for resection of large pedunculated polyps has been undertaken. Based on the advantages of UEMR, we speculated that UMER would reduce the occurrence of intraoperative or postoperative PPB and simultaneously improve the en bloc resection, complete resection and R0 resection rates for polyps. Therefore, we designed this prospective pilot study to assessed the effectiveness and safety of resecting pedunculated polyps with a head diameter of 10–20 mm without prophylactic clips when applying UEMR.
This is a prospective, pilot study conducted from April 2021 to March 2022 in Ningbo First Hospital (Ningbo, Zhejiang, China). Patients aged 18–75 years with pedunculated colorectal polyps 10–20 mm in head diameter, ≥ 5 mm in stalk diameter, and ≥ 10 mm in stalk length were eligible for inclusion. Patients with the following conditions were excluded: (1) pregnancy or lactation; (2) severe coagulopathy; (3) taking anticoagulant or antiplatelet medications within 7 days; (4) inflammatory bowel disease; (5) familial polyposis; (6) polyps with previous evidence showing signs of deep submucosal invasion; and (7) unwillingness to provide written informed consent.
The study protocol was approved by the medical ethics committee of Ningbo First Hospital. Written informed consent was obtained from all enrolled patients. In addition, the study was registered on 08/04/2021 at ClinicalTrials.gov (NCT04837690).
The included patients received a 3-L split-dose bowel preparation regimen. All endoscopic polypectomies were performed using high-definition video colonoscopes (OlympusCF-HQ290I/CF-H290I/CF-HQ290ZI, Olympus, Japan). In addition, all endoscopic procedures were recorded by a video. Each endoscopic procedure was performed with full access to the ileocecal valve, and the process of endoscopic resection was initiated during the withdrawal period. The characteristics of targeted polyps were recorded, including location and size. The polyp size was measured by comparison to a predetermined width of opened rotatable endoclip (ROCC-D-26-195-C, Microtech, China, maximum width 10 mm).
All of the UEMR procedures were performed by a single endoscopist who was proficient in UEMR and carried out research related to UEMR [13]. The resection process was as follows. When reaching the pedunculated polyps, the colorectal lumen was filled with 500 to 1000 ml sterile water by a flushing pump (OFP2, Olympus, Japan). No preoperative measures, such as submucosal injection or prophylactic clips, were applied to prevent bleeding. The round snare (JHY-SD-23-230-30-A1, JiuHong, China) was opened and positioned to possibly include the normal mucosa of the base and then tightened. Subsequent resection was performed with an electrosurgical generator (VIO200D, ERBE Elektromedizin GmbH, Tübingen, Germany), and pure forced coagulation current (output limit 40 W, effect 2) was initially tried for resecting the polyps. If resection was difficult, a combination of Endo-cut Q mode (effect 4, interval 6, length 1) and forced coagulation current could be used for polypectomy.
After the polypectomy, polypectomy sites were closed with endoscopic clips (ROCC-D-26-195-C, Microtech, China and/or M00522610, Resolution clip, Boston Scientific, USA) and/or nylon endoloops (Loop-20, LeClamp, China) according to the operator’s preference. For the occurrence of IPPB, the wound was flushed with water to expose the bleeding vessels after rapid aspiration of the original water in the intestinal lumen. Then, after electrocoagulation to stop the bleeding, the wound was finally closed with a combination of clips and nylon endoloop.
The primary endpoint was the rate of PPB, including immediate PPB (IPPB) and delayed PPB (DPPB). IPPB was defined as intraprocedural bleeding that occurred immediately after polypectomy. IPPB was classified as grade 1 and grade 2 according to a previous study. Grade 1 IPPB was defined as continuous oozing for 1 minute, and Grade 2 IPPB was defined as active blood spurting [5].
All patients were followed by telephone at 14 and 30 days after polypectomy to monitor the incidence of adverse events. The researchers gave the patients a phone number to contact in the event of an emergency during follow-up period. DPPB was defined as any symptoms of gastrointestinal bleeding, such as hematochezia, within 30 days after polypectomy and was classified as minor or significant DPPB according to the severity of bleeding. Minor DPPB was defined as self-limited hematochezia and hemoglobin loss of < 2 g/dL that did not require endoscopic hemostasis. Significant DPPB was defined as massive hematochezia and/or hemoglobin loss of > 2 g/dL, blood pressure drop > 20 mmHg, or increased pulse rate > 20% that required endoscopic hemostasis.
The secondary outcomes included resection time, discomfort during resection, and rates of en bloc resection, complete resection and R0 resection, perforation and postpolypectomy syndrome. Resection time was defined as the time between the start of snare extending from the endoscopic lumen and completion of the colonoscopic polyp resection. Discomfort during resection included abdominal pain, bloating, nausea, vomiting and other symptoms. Complete resection was defined as complete endoscopic lesion removal with no remaining visible adenoma. R0 resection was defined as a complete resection of a polyp with tumor-free lateral and vertical margins pathologically [18]. All resected polyps were sent for histopathological examination to determine the pathological type and whether the margin was negative.
There were no previous data about PPB of pedunculated polyps for UEMR, thus we referred to the data of other types of polyp resection without prophylactic clips. We used a noninferiority sample size calculation for a single group rate, assuming a postpolypectomy bleeding rate of 7.9% for UEMR and 15.5% for CEMR when resecting 10–20 mm colorectal polyp [19]. Statistical power was taken as 80% with an α-risk of 0.05. The noninferiority margins between the CEMR and UEMR groups were set to 10% for comparative analysis. In addition, we calculated a dropout rate of 10%, and the final required sample size was 32 cases.
Categorical variables were presented as percentages and continuous variables were presented as the median and mean ± standard deviation (SD). Statistical analysis was performed using SPSS statistic version 26.0 for Windows.
A total of 28 patients were included in this trial, and 32 large pedunculated colorectal polyps were treated with UEMR. All patients completed a 30-day follow-up visit. The baseline characteristics of the participating patients are listed in Table 1. This study included 22 males and 6 females with a mean age of 60.4 ± 9.5 (mean ± SD). Table 2 shows the characteristics of the included pedunculated polyps. The head diameter was 14.6 ± 3.5 mm (mean ± SD), the stalk length was 12.3 ± 3.1 mm (mean ± SD) and the stalk diameter was 11.5 ± 3.5 mm (mean ± SD). Most polyps were detected in the sigmoid colon (65.6%), and the most histology of polyps was tubular adenoma (43.8%). One polyp was pathologically diagnosed as tubular adenoma with intramucosal canceration foci in the head.
| Variable | |
|---|---|
| Male, n (%) | 22 (68.8) |
| Age, years, mean ± SD | 60.4 ± 9.5 |
| BMI, kg/m2, mean ± SD | 23.1 ± 2.3 |
| Drinker, n (%) | 14 (50.0) |
| Smoker, n (%) | 11 (39.3) |
| Previous abdominal or pelvic surgical, n (%) | 6 (21.4) |
| Hypertension, n (%) | 15 (53.6) |
| Diabetes mellitus, n (%) | 6 (21.4) |
| SD, standard deviation. | |
| Variable | |
|---|---|
| Head diameter, mm, mean ± SD | 14.6 ± 3.5 |
| Stalk length, mm, mean ± SD | 12.3 ± 3.1 |
| Stalk diameter, mm, mean ± SD | 11.5 ± 3.5 |
| Polyp location, n (%) | |
| Right colon | 4 (12.5) |
| Transverse colon | 3 (9.4) |
| Descending colon | 1 (3.1) |
| Sigmoid colon | 21 (65.6) |
| Rectum | 3 (9.4) |
| Histopathology, n (%) | |
| Tubular | 14 (43.8) |
| Tubulovillous | 10 (31.3) |
| Villous | 0 (0.0) |
| Serrated | 2 (6.3) |
| Hyperplastic | 3 (9.4) |
| Inflammatory | 2 (6.3) |
| Cancer, n (%) | 1 (3.1) |
| SD, standard deviation. | |
Every pedunculated polyp was resected successfully and achieved en bloc resection, complete resection and R0 resection (Table 3). The resection time was 50.2 ± 40.2 s (mean ± SD). Pure forced coagulation mode was conducted for 59.4% to remove the polyps, and a combination of two modes was conducted for the rest. In total, the resection was closed by 3.4 ± 1.7 hemoclips (mean ± SD) and/or 0.4 ± 0.5 endoloops (mean ± SD).
| Variable | |
|---|---|
| Resection time, s, mean ± SD | 50.2 ± 40.2 |
| Resection mode, n (%) | |
| Pure forced coagulation mode | 19/32 (59.4) |
| Pure Endo-cut Q mode | 0/32 (0.0) |
| Combination of two modes | 13/32 (40.6) |
| No. of postpolypectomy clips applied, n, mean ± SD | 3.4 ± 1.7 |
| No. of postpolypectomy endoloops applied, n, mean ± SD | 0.4 ± 0.5 |
| Discomfort during resection, n (%) | |
| None | 21/28 (75.0) |
| Abdominal pain | 7/28 (25.0) |
| Bloating | 8/28 (28.6) |
| Other | 0/28 (0.0) |
| Technical success, n (%) | 32/32 (100.0) |
| Complete resection, n (%) | 32/32 (100.0) |
| En bloc, n (%) | 32/32 (100.0) |
| Overall PPB, n (%) | 2/32 (6.3) |
| IPPB | 2/32 (6.3) |
| Grade 1 | 2/32 (6.3) |
| Grade 2 | 0/32 (0.0) |
| DPPB | 0/32 (0.0) |
| R0 resection, n (%) | 32/32 (100.0) |
| Perforation, n (%) | 0/32 (0.0) |
| Postpolypectomy syndrome, n (%) | 0/32 (0.0) |
| SD, standard deviation; PPB, postpolypectomy bleeding; IPPB, immediate post- | |
| polypectomy bleeding; DPPB, delayed postpolypectomy bleeding. | |
Regarding the incidence of adverse events, two cases (6.3%) of IPPB (both Grade 1) occurred after resection, but no indication of hemorrhagic shock, such as hemoglobin and blood pressure drop, was seen. Finally, both were successfully controlled by endoscopic hemostasis (Table 4). None of the polyps showed evidence of perforation or postpolypectomy syndrome. In addition, no DPPB was reported during the follow-up period.
| Variable | Polyp 1 | Polyp 2 |
|---|---|---|
| Location | Sigmoid colon | Sigmoid colon |
| Head diameter, mm | 10 | 12 |
| Stalk length, mm | 10 | 12 |
| Stalk diameter, mm | 12 | 13 |
| Resection mode | Combination of two modes | Pure forced coagulation |
| Histopathology | Hyperplastic | Serrated |
| Grade of IPPB | Grade 1 | Grade 1 |
| Endoscopic therapy | ||
| Clips, n | 3 | 4 |
| Endoloops, n | 1 | 1 |
| IPPB, immediate postpolypectomy bleeding. | ||
A high rate of PPB has been reported after large pedunculated polypectomy, and even though the application of prophylactic clips could reduce the incidence of PPB, there might be a risk of tumor remnants at the resection margin. Therefore, we first attempted to conduct UEMR to remove 10–20 mm pedunculated polyps of head size, and we found that UEMR had the advantages of low PPB risk (including IPPB and DPPB) as well as high en bloc resection, complete resection and R0 resection rates. In addition, no other serious adverse events were observed in the study.
In our study, 2 cases of slightly immediate bleeding (Grade 1, 6.3%) were observed, and both achieved successful control. As reported by Jae et al. [7], the IPPB rate was 9.0% after the application of prophylactically clipping the stem before CEMR for large pedunculated polyps (head size ≥ 10 mm), which was significantly lower than that of 31.4% without pre-polypectomy clipping. A study conducted by Gweon et al. [5] showed the similar results that the IPPB rate in the clip application group was 2.5%, and that in the non-clip application group was 10.9%. Our research did not use prophylactic clip closure, and the bleeding rate was obviously lower than that of CEMR without clips. The IPPB rate reported in our study seems to approximate what they reported in the clip application group. We considered that UEMR without prophylactic clips was potentially superior to CEMR without prophylactic clips and could also achieve the same effect of prophylactic clip placement. Generally, for pedunculated polyps with the existence of thick nourishing vessels in the stem, the application of the UEMR combined with coagulation mode could better block blood flow and reduce the occurrence of postpolypectomy bleeding even if clips are not used before resection [20, 21]. We mainly used the coagulation-only mode (59.4%) to remove polyps, which may be partly related to the low PPB rate.
Delayed bleeding is more significant than immediate bleeding because delayed hemorrhage occurs more insidiously and often during non-hospitalization, which may require repeated colonoscopy or transfusion [22–24]. However, the effect of prophylactic clips for large pedunculated polyps on DPPB is still controversial [25, 26]. The U.S. Multi-Society Task Force on Colorectal Cancer guidelines recommend the application of prophylactic mechanical methods of the stalk, which can reduce DPPB [27], but a previous study [25] including 1147 polyps confirmed that the use of prophylactic hemoclips prior to resection (OR, 4.1; 95% CI, 1.3–13) was significantly associated with a risk of delayed bleeding for pedunculated polyps of head size larger than 20 mm. For 10–20 mm polyps, there were no studies showing that prophylactic clips could reduce DPPB. Our study included only pedunculated polyps of 10–20 mm, and none of them experienced delayed bleeding. This was a satisfactory result, even though it might be related to the low sample size. Nevertheless, we could try to explain it as follows. Electricity leaking to the hemoclip, resulting in insufficient coagulation of the blood vessel and lumen wall injury [25, 26], may lead to delayed hemorrhage. We postulated that the heat-sink effect of water immersion in UEMR would prevent thermal injury and reduce vascular damage to reduce the incidence of PPB, including IPPB and DPPB [9, 11].
Though The application of prophylactic clips could reduce the incidence of PPB [5, 7], Douglas K [8] pointed out that multiple adverse events, including bleeding, thermal injury and positive resection margins, should be synchronously considered in pedunculated polyp polypectomy. The application of prophylactic clips may make the resection difficult to get close to the polyp base, thus leading to positive resection margins. Low en bloc resection, complete resection and R0 resection rates are strongly associated with tumor recurrence [28, 29]. However, previous studies only assessed the effect of prophylactic clips for postpolypectomy bleeding while ignoring the risk of tumor residue [5, 7, 30]. The prophylactic placement of clips could hinder maximizing the distance between a cancer that may be present in the head of a pedunculated polyp and the resection line [8]. The UEMR as a method reported firstly in 2012 showed high complete resection and en bloc resection rates for large nonpedunculated colorectal polyps [19, 31], but the effect of UEMR on pedunculated polyps has not been reported yet. In our research, no mechanical approaches were used to clamp the stem after the bowel lumen was filled with water, and resection was performed as close to the base as possible. Notably, we found that all pedunculated polyps (100%) achieved en bloc resection, complete resection and R0 resection. During UEMR, the mucosa could better "float" [9] and the resection range was not restricted by clips, making the resection much more complete to reducing the risk of tumor remnants.
Perforation and postpolypectomy syndrome did not occur among the 32 polyps, which could owe to the fact that both adverse events were rare. Theoretically, the heat-sink effect and the floating mucosa away from the submucosal layer can reduce the occurrence of these adverse events. To our knowledge, this was the first prospective study to evaluate the efficacy of UEMR in resecting large pedunculated polyps and showed that UEMR has the potential to reduce the incidence of PPB and tumor residue simultaneously as well as other severe adverse events.
We acknowledge that there were some limitations in our study. First, this was a single-arm study without control groups. However, no serious PPB was observed in any of the 32 cases after polypectomy, and all cases achieved R0 resection and en bloc resection, which showed the potential advantages of UEMR in resecting pedunculated polyps. The randomized control trials with larger sample sizes are needed to further demonstrate the safety and efficacy of UEMR. Second, all wounds were closed after polypectomy, which did not affect the incidence of IPPB, but this might be a confounding factor in assessing the effect of UEMR on DPPB. It is difficult to distinguish whether the lack of DPPB in this study was due to "water", post-polypectomy clipping or the combined effect of both. Third, the included patients had no long-term follow-up data to monitor tumor recurrence, but these patients were scheduled for outpatient follow-up 3–6 months after polyp resection. In cases of recurrence, endoscopic therapy or surgical intervention was performed.
In conclusion, UEMR without the application of prophylactic clips showed a low postpolypectomy bleeding rate and low risk of tumor residual and other severe adverse events. Therefore, it might be suitable and safe for resection of 10–20 mm pedunculated polyps. However, randomized clinical trials are needed to further compare its efficacy and safety with other pedunculated polypectomy methods.
1. EMR: endoscopic mucosal resection;
2. UEMR: underwater endoscopic mucosal resection;
3. CEMR: conventional endoscopic mucosal resection;
4. PPB: postpolypectomy bleeding;
5. IPPB: immediate postpolypectomy bleeding;
6. DPPB: delayed postpolypectomy bleeding;
7. SD: standard deviation.
Ethics approval and consent to participate
The study protocol was approved by the medical ethics committee of Ningbo First Hospital. The study was registered on 08/04/2021 at ClinicalTrials.gov (NCT04837690). Written informed consent was obtained from all enrolled patients.
Authors’ contributions
Xueqin Chen and Lei Xu designed and created the study; Hui Gao, Shuhao Zheng, Cenqin Liu collected clinical data; Hui Gao and Xin Yuan participated in data analysis and interpretation; Hui Gao and Jiarong Xie wrote a draft of the manuscript; All authors read and approved the final manuscript.
Availability of data and materials
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.
Disclosure Statement
The authors declare no conflicts of interest.
Financial support
This study was supported by the Medical Health Science and Technology Project of Zhejiang Provincial Health Commission (No. 2022KY315 to Lei Xu).