Introduction

Cold snare polypectomy (CSP) is widely used to bluntly resect polyps because of its high safety profile1,2,3,4,5,6,7. However, CSP is associated with a less effective resection ability compared to that of endoscopic mucosal resection (EMR), which utilizes energization for resection. Occasionally, during CSP, the cold snare can become stuck and fails to resect, resulting in the need to switch to EMR1,5.

When multiple polyps are resected, the CSP snare must be inserted several times through the instrument channel port. Accidental bending of the snare sheath during insertion occasionally occurs, thus decreasing the snare performance; however, it remains unclear whether snare bending reduces the resection ability. Additionally, there is no consensus regarding whether CSP can be continued with a bent snare or whether the bent snare should be replaced with a new snare.

In this study, using currently available CSP snares and prototype snares, we aimed to demonstrate that snare bending reduces the ability to resect polyps (Fig. 1). We recently revealed a strong correlation between the resection ability and stiffness of the sheath and wire spindle8. To elucidate the mechanism underlying the reduced resection ability of bent CSP snares, we investigated whether the changes in the resection ability of the bent snare were related to the stiffness of its sheath and wire spindle.

Fig. 1
figure 1

Prototype snares. The four prototype snares with different characteristics of each component were prepared.

Full size image

Results

When the snares were bent during snare insertion, both the sheath and the wire spindle inside the sheath were bent (Fig. 2). Snare bending significantly decreased the wire spindle stiffness. Additionally, compared with bending the snare once, bending the snare five times further decreased the wire spindle stiffness (Fig. 3A). Specifically, regarding the Exacto cold snare, Captivator cold snare, Micro-Tech cold snare, SnareMaster Plus snare, and hard/hard and hard/soft prototype snares (numbers 1 and 2, respectively), after bending once and after bending five times, their wire spindle stiffness was significantly lower than that of snares that had not been bent (P < 0.001 for all). Moreover, regarding the soft/hard and soft/soft prototype snares (numbers 3 and 4, respectively), after bending five times, the wire spindle stiffness was significantly lower than that of snares that had not been bent (P < 0.001 for all). In contrast, snare bending minimally affected the sheath stiffness (Fig. 3B).

Fig. 2
figure 2

Bent snare. The sheath portion of the snare is sometimes accidentally bent when the snare is inserted through the instrument channel port. When the sheath portion of a snare is bent, the sheath and wire spindle inside the sheath are bent simultaneously.

Full size image
Fig. 3
figure 3

Measurement of wire spindle stiffness of the snares bent one time or five times (A) and measurement of sheath stiffness of the snares bent one time or five times (B). Data are expressed as the mean ± standard error of at least five independent experiments. The prototype snare is denoted as follows: wire spindle (soft or hard) / sheath (soft or hard). *P < 0.001 (versus no bending).

Full size image

Snare bending significantly increased the force required to resect (FRR) a 15-mm-wide human colonic mucosa and decreased the resection ability of commercially available CSP snares. Bending the snare five times significantly increased the FRR and decreased its resection ability compared with those of snares that had been bent only once (Fig. 4). The FRRs of Exacto cold snares increased from 6.5 ± 1.0 N to 20.5 ± 1.8 N and 31.3 ± 1.6 N after bending once and after bending five times, respectively. Furthermore, the FRRs of Captivator cold snares increased from 7.4 ± 1.2 N to 22.1 ± 2.2 N and 30.9 ± 1.7 N, those of Micro-Tech cold snares increased from 12.5 ± 1.2 N to 20.1 ± 1.1 N and 28.2 ± 1.4 N, and those of SnareMaster Plus snares increased from 13.6 ± 1.0 N to 23.6 ± 1.5 N and 33.7 ± 1.4 N after bending once and after bending five times, respectively. The same trend was observed for the four prototype snares, with a more pronounced reduction in resection ability after bending and high stiffness of the wire spindle (numbers 1 and 2) (Fig. 4). The FRRs of the Exacto cold snare, Captivator cold snare, Micro-Tech cold snare, SnareMaster Plus snare, and hard/hard and hard/soft prototype snares (numbers 1 and 2, respectively) after bending once and after bending five times were significantly higher than those of snares that were not bent (P < 0.001 for all). Moreover, the FRRs of soft/hard and soft/soft prototype snares (numbers 3 and 4, respectively) after bending five times were significantly higher than those of snares that were not bent (P < 0.001 for all).

Fig. 4
figure 4

Measurement of force required for resection (FRR) of the snares bent one time or five times. Data are expressed as the mean ± standard error of at least five independent experiments. The prototype snare is denoted as follows: wire spindle (soft or hard) / sheath (soft or hard). *P < 0.001 (versus no bending).

Full size image

Decreased wire spindle stiffness resulting from snare bending reduced resection ability, and a strong correlation was observed between wire spindle stiffness and resection ability after snare bending (correlation coefficient = 0.82; P < 0.001) (Fig. 5A). In contrast, no correlation was observed between sheath stiffness and resection ability after snare bending (correlation coefficient = 0.46; P < 0.001) (Fig. 5B).

Fig. 5
figure 5

Correlation analysis of the force required for resection (FRR) and wire spindle stiffness (A) and correlation analysis of the FRR and sheath stiffness (B). Pearson’s correlation coefficient was used to evaluate the correlations.

Full size image

Snare bending also reduced the completion rate of CSP for pseudo-lesions on the porcine rectal mucosa (Fig. 6 and Supplementary Table S1). The resection completion rates of Exacto cold snares decreased from 100% ± 0.0–80.0% ± 20.0% and 66.7% ± 11.5% after bending once and after bending five times, respectively. The resection completion rates of Captivator cold snares decreased from 100% ± 0.0–86.7% ± 11.5% and 73.3% ± 11.5%, those of Micro-Tech cold snares decreased from 93.3% ± 11.5–80.0% ± 20.0% and 73.3% ± 11.5%, and those of SnareMaster Plus snares decreased from 93.3% ± 11.5–80.0% ± 0.0% and 73.3% ± 11.5% after bending once and after bending five times, respectively. Bending the snare five times significantly decreased the resection completion rates of Exacto cold snares and Captivator cold snares (P = 0.008 and P = 0.016, respectively).

Fig. 6
figure 6

Resection completion rate of cold snare polypectomy using snares bent one time or five times. To evaluate the performance of the bent snares, cold snare polypectomy was performed for a 5-mm-diameter pseudo-lesion using a porcine rectum, and the resection completion rate was calculated. The resection completion rate was defined as the percentage of pseudo-lesions that could be resected with a snare without becoming stuck. Pathological evaluations were not included because pseudo-lesions were targeted. Data are expressed as means ± standard errors. *P < 0.05 (versus no bending). See also Supplementary Table S1 for detailed data.

Full size image

Discussion

When multiple polyps are resected during endoscopic treatment, the CSP snare must be inserted numerous times, thus increasing the risk of accidental bending of the snare sheath, which decreases snare performance. We investigated the changes in the stiffness of bent snare sheaths and wire spindles and their effects on resection ability.

Snare resection ability decreased with an increasing number of bends; bending only once significantly reduced the resection ability of the currently available CSP snares. Therefore, continuation of CSP with a bent snare is not recommended because it may cause an increased incidence of stuck lesions and a reduced resection completion rate. Our evaluation of CSP for pseudo-lesions on the porcine rectal mucosa showed that reduced resection ability may reduce the resection completion rate and increase the incidence of stuck lesions. Sheath stiffness remained almost unchanged after bending; however, the wire spindle stiffness decreased with increased bending. Our previous study revealed that the stiffness of the sheath and that of the wire spindle are factors that determine the resection ability of CSP snares8. The present study revealed that when a CSP snare is bent during endoscopic treatment, its resection ability is reduced because of the resulting reduction in the stiffness of the wire spindle; this resection ability cannot be restored. Because the wire spindle inside the sheath is the main cause of reduced resection ability, if the snare is accidentally bent, then replacement with a new snare should be considered before continuing CSP. Modifying the bent sheath with methods such as reinforcement with tape will not restore the resection ability.

During the analysis of prototype snares, the reduction in resection ability caused by snare bending was more pronounced in snares with wire spindles with high stiffness and high resection ability (numbers 1 and 2); this was also true for the current CSP snares. In contrast, the reductions in wire spindle stiffness and resection ability caused by bending the snare one time were minor for prototype snares with wire spindles with low stiffness (numbers 3 and 4). However, these prototype snares had low resection ability before bending because their wire spindle stiffness was low; therefore, they were not very practical.

Although bending may also reduce the wire spindle stiffness of EMR snares, the reduction in blunt resection ability does not interfere with the procedures or outcomes of EMR, which involves resecting polyps using a current and a snare. Thus, snare bending is a serious concern inherent to CSP, but not to conventional EMR.

This study had some limitations. The resection ability and friction measurement models were generated using colons obtained from forensic autopsy specimens within 12 h of death; therefore, the colons may not have represented the same conditions as those of living individuals. Despite this, these colon models provide the best current results because these measurements cannot be performed in vivo.

In conclusion, when a CSP snare is bent, its resection ability is reduced because of the reduced stiffness of the wire spindle, and this ability cannot be restored. This reduced resection ability may lower the resection completion rate of CSP and increase the incidence of stuck lesions. Therefore, endoscopists should avoid bending CSP snares during treatment. Replacement of the snare is recommended if bending occurs.

Methods

Preparation of snares

We prepared currently available CSP snares, including the Exacto cold snare (US Endoscopy, Mentor, OH, USA), Captivator cold single-use snare (Boston Scientific, Marlborough, MA, USA), Micro-Tech cold snare (Micro-Tech, Nanjing, China), and SnareMaster Plus snare (Olympus, Tokyo, Japan). Prototype snares were also prepared, including hard and soft versions of both the sheath and wire spindle. Four prototype snares were developed by combining these features (Fig. 1). The soft and hard sheaths were made of Teflon and high-density polyethylene, respectively, whereas the wire spindles were made of stainless steel.

Each target snare was bent approximately 120 degrees at a point 10 cm from the sheath tip. The same procedure was repeated when bending the snare multiple times. The bent snares were manually returned to their original shape (i.e., straight) before the start of each evaluation.

Evaluation of the resection ability

The FRR to resect a 15-mm-wide human colonic mucosa was measured as an indicator of resection ability using a previously constructed model8,9. A length of 15 mm corresponds to a circumference of a polyp with a diameter of 4.8 mm, and this assessment reproduced the CSP situation for diminutive colon polyps with a diameter of approximately 4–5 mm. Colon specimens collected after forensic autopsies at the Department of Forensic Medicine at the Kyoto Prefectural University of Medicine were cut open and flattened, and only the mucosal layer was detached. The detached mucosal layer was stretched and excised as 15-mm-wide strips that were resected using a target snare. The force required to push the snare handle during resection was measured using a force gauge (ZTA-500 N; IMADA, Aichi, Japan), and the maximum force was defined as the FRR. The FRR reflected the resection ability of the snare. If the FRR was low, then the resection ability was considered high. Data are expressed as means ± standard errors. Calculations were performed after at least five independent experiments.

Measurement of the wire spindle stiffness and sheath stiffness

The stiffness of the wire spindles and that of the sheaths were assessed using a force tester (MCT-2150; A&D Systems, Tokyo, Japan) following previously established methods8,9,10. The 5-cm wire spindles were subjected to a tensile test by attaching a jig to the wire tip. The load required to stretch a 1-mm wire in the elastic region was calculated based on the data obtained (MSAT-Lite, A&D). This value was used as an index of wire spindle stiffness. To obtain the load value used as an index of sheath stiffness, a titanium rod was passed through the sheath lumen to prevent deflection of the sheath, and a 5-cm sheath was set so that it could be pushed only with an attached jig. A compression test was performed, and the load required to compress a 1-mm sheath in the elastic region was calculated based on the data obtained.

CSP for pseudo-lesions on the porcine rectal mucosa

To evaluate the performance of bent snares, CSP was performed for pseudo-lesions using porcine rectums. Rectums were harvested from 6-month-old domestic hybrid female pigs, frozen immediately, and thawed at room temperature 6 h prior to each experiment11. A pseudo-lesion with a 5-mm diameter was created on the rectal mucosa, and CSP using currently available snares was performed. Two endoscopists (two experts with experience performing ≥ 5,000 colonoscopies) conducted all CSPs and were not informed prior to CSP about the type of snare and whether it was bent. CSP was performed five times, and the resection completion rate was calculated. The resection completion rate was defined as the percentage of pseudo-lesions that could be resected with a snare without becoming stuck. Pathological evaluations were not included because pseudo-lesions were targeted. This analysis was conducted thrice. Data are expressed as means ± standard deviations.

Ethical considerations

The study protocol, including the sample collection procedures, was reviewed and approved by the Institutional Review Board of the Kyoto Prefectural University of Medicine (ERB-C-2573). In addition, all methods were carried out in accordance with relevant guidelines and regulations, and are reported in accordance with ARRIVE guidelines.

Statistical analysis

Data were analyzed using GraphPad Prism 7 (GraphPad Inc., La Jolla, CA, USA). Pearson’s correlation coefficient was used to evaluate the correlation between the FRR and the stiffnesses of the wire spindles and sheaths. Continuous variables were assessed using Student’s t-test. Statistical significance was set at P < 0.05 (two-sided).