Unsuccessful colonic endoscopic full-thickness resection – failure of device or operator?

Since the introduction of the full-thickness resection device (FTRD; Ovesco Endoscopy) in 2014, endoscopic full-thickness resection (EFTR) has quickly entered clinical routine and gained broad acceptance among interventional endoscopists [11] [22]. One reason for its rapid clinical implementation is probably the “simplicity” of the clip-based nonexposure technique; however, every endoscopist performing EFTR knows that the procedure is not always as simple as is suggested in videos and animal training models. Although we currently have a substantial amount of data on resection completeness and adverse events (AEs), specific studies classifying “technical failures” have not been published, so far. I would therefore congratulate Gibiino and colleagues on their study [33], and I acknowledge their effort; however, the methodology, the proposed definitions, and the conclusions deserve further discussion.

The authors define technical failure as an “inability to achieve en bloc resection after successful application of the clip.” The definition is somehow contradictory to the proposed “type II failure,” which the authors define as unsuccessful clip release. Clip misdeployment is reported in 24 cases, corresponding to 3.2% of the total cohort of 750 patients. This rate seems very high. With the latest generations of the device, it is usually very easy to release the clip, especially for endoscopists who are experienced with application of over-the-scope (OTS) clips. In the Dutch FTRD registry (n = 346), three such events (0.9%) were reported and the rate was even lower (n = 1; 0.08%) in the German registry [22] [44].

"Future studies should aim to define the minimum requirements of training and case volume for EFTR. I also agree with the authors that a consensus definition of failures and related AEs of EFTR is desirable."

Misdeployment can result in intraprocedural perforation, when the resection is carried out despite insufficient clip release. The authors have reported eight intraprocedural perforations, but it is unclear if these were related to insufficient clip release. As discussed in several studies, such perforations can be avoided when the endoscopist visually confirms clip release (white ring on the outside of the cap) before resection. Furthermore, it remains unclear how these resections were “rescued” by the endoscopists.

The most frequent type of failure was “snare noncutting,” which was seen in 41 cases (5.5% of the total cohort) [33]. Indeed, this may be considered as true failure of the device, as it is known that snares were prone to disruption or sometimes failed to close, especially when using the first generation of the device [55]. In 2017, the first prospective multicenter study on colorectal EFTR was published by our group [66]. In the total cohort of 181 patients, resection was reported to be “difficult” in 28 procedures (15.4%). Of these, 13 cases (7.1%) were due to snare dysfunction, but 12/13 resections could be completed with a conventional snare. The snare that is part of the device underwent substantial modifications in 2016 and 2022. Since then, snare dysfunction should rarely occur. In the German FTRD registry, including 1178 procedures between 2025 and 2019, snare dysfunctions were reported in only five cases (0.4%) [22]. The authors of the current study state that about half of the events were “not device related” and it remains unclear what this means. In my experience, application of high frequency current before full closure can lead to insufficient cutting and too firm closure of the snare by the assistant can indeed lead to disruption of the wire, especially in scarred lesions.

The authors defined type III failure as an inability to entrap the lesion completely into the cap. In my view, this is not a device-related or technical failure. Rather, as discussed in other studies, it strongly depends on lesion characteristics (scarring, presence of advanced malignancy, size, and location). Unfortunately, the study lacks details on the lesion characteristics, such as lesion size. It is generally accepted that scarred lesions selected for standalone EFTR should not exceed a diameter of 20 mm [66]. It is also important to note that a significant proportion in the reported cohort of technical failures were adenocarcinomas (22.6%). Past studies have indicated that, to achieve optimal R0 resection, lesion size for T1 tumors should not exceed 15 mm (to 20 mm) [77] [88]. It therefore remains unclear whether inappropriate selection of lesions may have biased the results of the current study.

Two major weaknesses of this retrospective study were the selection process and the data analysis: neither was well defined. A retrospective description of failures requires highly detailed databases or at least thorough reporting (or recording) of all procedural steps. At least in my daily routine, I would have difficulties remembering all of the details of procedural difficulties from cases 3–4 years ago and there might have been some subjective bias in the analysis of procedures. It would also have been very interesting to see baseline and procedural data for the whole cohort of 750 patients in order to compare the reported subcohort with those with successful procedures.

The authors deemed “technical failures” as “device dysfunction.” The reported total failure rate of 10.2%, at least at first glance, seems unacceptably high for a medical device or a new procedure. But do those failures really reflect device dysfunction and/or immaturity of the EFTR technique? In the supplementary material, the authors state that 17.6% of participants had no prior specific EFTR training and 35.3% did not undergo hands-on training on animal models. As mentioned, EFTR might be easier to learn compared with other techniques, such as ESD, but this does not mean that it is an easy procedure. At least a full day of hands-on and theoretical training is strongly recommended for all FTRD users, independent of their experience in other resection techniques. Furthermore, supervision by an experienced EFTR operator during the first interventions is necessary (as is the case for every procedure in interventional endoscopy).

The possible lack of experience in EFTR is reflected by the quite low case volumes of most of the participating centers in the study. Nine centers had <30 cases in 4 years and centers with >30 cases per 4 years were considered to be “high volume.” Although there is no evidence that case volume correlates with the outcomes of EFTR, this is likely to affect procedural success and safety. Again, it would have been very interesting to see results on the R0 rate and AEs, not only for the cases of technical failure but also among the whole cohort of patients.

Despite all of this criticism, the classification and detailed description of “technical failures” is very important, not only in terms of resection success but also in terms of safety. The rate of AEs in this study is almost three times higher than that reported in a comprehensive analysis of the German and Dutch registry [99]. This suggests that technical failure is strongly associated with the occurrence of AEs and a better understanding of the underlying mechanisms is necessary, also in terms of teaching and training.

What we can learn from the study by Gibiino and colleagues is that failures have multiple reasons and are not necessarily a failure of the device or the procedure itself. A Tour de France cyclist needs a reliable and fast bike. But ultimately, strategy, training, experience, and interplay with the team will determine success. In interventional endoscopy, we need to carefully select patients and we must know the limitations and pitfalls of our techniques and devices. Furthermore, we need to ensure that we receive sufficient training and supervision for complex procedures. Future studies should aim to define the minimum requirements of training and case volume for EFTR. I also agree with the authors that a consensus definition of failures and related AEs of EFTR is desirable. This ultimately may help to further improve the outcomes of EFTR.

Publication History

Article published online:
09 July 2024

• References

• 1 Schmidt A, Damm M, Caca K. Endoscopic full-thickness resection using a novel over-the-scope device. Gastroenterology 2014; 147: 740-742.e2
• 2 Meier B, Stritzke B, Kuellmer A. et al. Efficacy and safety of endoscopic full-thickness resection in the colorectum: results from the German Colonic FTRD Registry. Am J Gastroenterol 2020; 115: 1998-2006
• 3 Gibiino G, Binda C, Papparella LG. et al. Technical failure during colorectal endoscopic full thickness resection: the “through thick and thin” study. Endoscopy 2024; 2024
• 4 Zwager LW, Bastiaansen BAJ, Bronzwaer MES. et al. Endoscopic full-thickness resection (eFTR) of colorectal lesions: results from the Dutch colorectal eFTR registry. Endoscopy 2020; 52: 1014-1023
• 5 Schmidt A, Bauerfeind P, Gubler C. et al. Endoscopic full-thickness resection in the colorectum with a novel over-the-scope device: first experience. Endoscopy 2015; 47: 719-725
• 6 Schmidt A, Beyna T, Schumacher B. et al. Colonoscopic full-thickness resection using an over-the-scope device: a prospective multicentre study in various indications. Gut 2018; 67: 1280-1289
• 7 Zwager LW, Bastiaansen B, van der Spek B. et al. Endoscopic full-thickness resection of T1 colorectal cancers: a retrospective analysis from a multicenter Dutch eFTR registry. Endoscopy 2022; 54: 475-485
• 8 Kuellmer A, Mueller J, Caca K. et al. Endoscopic full-thickness resection for early colorectal cancer. Gastrointest Endosc 2019; 89: 1180-1189.e1
• 9 Zwager LW, Mueller J, Stritzke B. et al. Adverse events of endoscopic full-thickness resection: results from the German and Dutch nationwide colorectal FTRD registry. Gastrointest Endosc 2023; 97: 780-789.e4