Endoscopic full-thickness resection using full-thickness resection device for treatment of upper and lower gastrointestinal tract lesions—the first New Zealand study

Endoscopic resection has revolutionised the management of gastrointestinal (GI) tract lesions, offering a minimally invasive alternative to traditional surgical approaches with reduced morbidity and shorter recovery times. Previously, we have shown conventional endoscopic techniques such as endoscopic mucosal resection and endoscopic submucosal dissection (ESD) to be efficacious and safe in the treatment of mucosal and subepithelial lesions.^1,2 However, application may be limited by factors including difficult location, depth of invasion and significant fibrosis.

The full-thickness resection device (FTRD® system, Ovesco Endoscopy AG, Tuebingen, Germany) represents an advanced endoscopic tool designed to overcome some of these limitations to potentially reduce the need for more invasive surgery by enabling resection of GI tract lesions that are not possible to remove with traditional endoscopic techniques. This over-the-scope device combines a grasper, snare and large clip, designed to enable the resection to include all layers including the muscularis propria. The device is equipped with a specialised clip to immediately close the inevitable perforation that occurs during the resection of the muscular layer, effectively sealing the site to minimise complications. Common indications for FTRD include scarred or non-lifting polyps, subepithelial lesions, early cancers, lesions located in difficult anatomic sites such as the appendiceal orifice and diagnostic resection for suspected Hirschsprung’s disease or amyloidosis where analysis of the muscularis propria is required.³ The safety and efficacy of FTRD has been demonstrated in multiple retrospective and prospective studies;^4–8 however, no studies have evaluated this in New Zealand.

A schematic presentation of endoscopic full-thickness resection (EFTR) of a lesion at appendiceal orifice using FTRD is illustrated in Figure 1.

View Figure 1, Table 1–2.

Method

We performed a retrospective study including consecutive adult (>18 years) patients who had an endoscopic resection of upper and lower GI tract lesions using the FTRD at two tertiary centres, Middlemore Hospital and North Shore Hospital in Auckland, New Zealand. Three endoscopists performed the procedures: two at Middlemore Hospital and one at North Shore Hospital. Data from all consecutive procedures performed between 1 January 2017 and 30 April 2023 were included. Written informed consent was obtained from all patients. All endoscopists completed a mandatory in-person training course for safe use of the device. After completion of FTRD, patient and procedural characteristics as well as histology and post-procedural complications up to 30 days were collated. The study protocol was reviewed and approved by the Auckland Health Research Ethics Committee (AHREC) and was granted locality approval at each study centre.

Outcomes

The primary endpoint was to assess technical success of the procedure. This was defined as achieving three key criteria: 1) successfully reaching the target lesion with the FTRD, 2) accurately applying the FTRD clip to the lesion, and 3) achieving immediate and complete resection of the lesion.

Complete resection was determined by the endoscopist’s assessment and was defined as the lack of visible residual lesion following the use of the FTRD.

Secondary end points of this study included:

1.       Histological margin: R0 resection was characterised as histologically complete removal of the lesion with negative resection margins both lateral and deep. R1 resection was defined as microscopic evidence of residual pathology at the margins of resection. Histological examination of the removed lesions was conducted on-site at each study centre.

2.       Immediate and delayed procedure-related complications up to 30 days: Immediate complications were adverse events that occurred during the FTRD procedure until completion. Any complications that occurred after the procedure up to 30 days was classified as delayed. Minor immediate bleeding was described as bleeding necessitating haemostasis via endoscopic means only. Major immediate or delayed bleeding was defined as bleeding that required repeat endoscopy for haemostasis or bleeding that warranted further interventions such as transfusion of blood or blood products and/or surgical treatment. Minor delayed bleeding was bleeding that only required observation and spontaneously self-resolved, entailing no additional intervention. Other complications including infection, perforation or need for surgery were collected.

Data sources and statistical analysis

Patient, procedure and histological data were collected online from a digital service provider (Clinical Portal). Data were then de-identified and analysed using SAS software (SAS Institute, Cary, North Carolina). Binomial calculation with Wald confidence intervals was conducted to calculate the rate of technical success of the FTRD. The technical success was further explored by gender, age group, indication groups and endoscopic lesion size. For categorical variables, Fisher’s exact test was used to assess associations between groups. The two-sided t-Test was utilised to compare the success rate of the study with the 90% benchmark.

Results

Between January 2017 and April 2023, 51 patients underwent EFTR using the FTRD (see Table 1). Patients had a mean age of 63 years, and the majority were female (65%). The mean lesion size was 15mm (range 4–30mm). Forty-six lesions (90%) were located in the colon, including 20 at the appendiceal orifice. Five lesions (10%) were in the upper GI tract. The mean diameter of specimens was 17mm (standard deviation [SD] 7.3, range 6–45mm). Most cases were performed for lesion resection, with 8% of cases performed for diagnostic purposes. Histopathological results are shown in Table 2.

Technical success

Technical success was achieved in 44/51 patients (86%). Sub-group analysis was performed to evaluate the technical success rate by gender, age group, indication and lesion size, without any significant differences noted (p<0.05).

R0 status for neoplastic lesions was achieved in 31/38 patients (82%). Five incompletely resected lesions from the colon were managed successfully with repeat endoscopy. One patient had rectal adenocarcinoma that was positive at radial margin. This patient successfully completed long-course chemoradiotherapy with 50Gy in 25 fractions with capecitabine, without evidence of disease recurrence at 12-month follow-up.

A high-grade dysplastic lesion in the gastric antrum was positive at radial margin. The patient has opted for further attempts at endoscopic resection with ESD technique.

Procedure-related adverse events

The rate of procedure-related adverse events within 30 days was 12%. Three patients developed appendicitis (two managed conservatively, one surgically) following appendiceal orifice lesion resection. One patient developed delayed bleeding requiring endoscopic management and blood transfusion. Technical complications occurred in two cases: one snare entrapment causing perforation and requiring surgical intervention and one clip entrapment requiring endoscopic retrieval.

Discussion

This is the first study to describe the efficacy and safety profile of the FTRD for EFTR of both upper and lower GI lesions in the New Zealand setting.

We report a technical success rate of 86%, aligning closely with existing literature where success rates typically range from 80% to 90%.^3–5 These lesions are among the most difficult to remove, where other endoscopic techniques are not possible to employ, and traditionally these lesions would have been sent for surgical management.

Lesion characteristics may play perhaps the most significant role in determining technical success. This series includes the complete learning curve for this procedure, and we have since modified local protocols regarding case selection, including requirement for biopsy confirmation of pathology (as opposed to occasional cases referred with endoscopic optical biopsy).

We achieved an R0 resection rate of 82%. The ability to achieve clear histological margins is critical in achieving cure. This rate is comparable to other FTRD studies reporting rates between 70% and 85%.^3–5 Risk factors for incomplete resection include lesion size >3cm, right-sided colonic lesions and tissue fibrosis leading to difficulty with lesion retraction.⁹

Adverse events occurred in 12% of cases, similar to rates in comparably sized retrospective studies¹⁰ and recent meta-analyses.^4,5 Appendicitis occurred in 3/20 lesions (15%) at the appendiceal orifice, similar to previously reported rates of 17%.¹⁰ The high rates of appendicitis have raised concerns about the safety of FTRD for this indication; however, most cases of appendicitis were managed conservatively, with only one requiring surgical intervention in our cohort. All cases were given intravenous antibiotics during the procedure, and for 3–5 days orally thereafter. All potential cases in our centres are discussed with the patient in-clinic for a thorough consent process, where a laparoscopic surgery is offered as an alternative. In our experience, anecdotally, there is an approximate 50% split between modalities chosen.

Major bleeding occurred in one patient, who had a lesion located in the stomach, requiring blood transfusion and repeat endoscopy for haemostasis. EFTR using the gastroduodenal FTRD is a more novel development, with recent meta-analysis showing this technique to be feasible, safe and efficacious.⁴ The RESET trial¹¹ reported minor peri-procedural bleeding that was managed endoscopically in 31% of the patients undergoing EFTR for gastric subepithelial lesions. However, in a larger cohort of 56 patients, Hajifathalian et al. reported similar major delayed bleeding rates to our cohort.¹⁰ Most bleeding events were managed endoscopically without the need for surgical intervention. We observed no bleeding with FTRD in the colorectal procedures and this is in accordance with existing studies that report higher bleeding rates in the upper compared with lower GI tract.⁹ This may be due to rich blood supply and a thicker gastric wall, resulting in suboptimal compressive effect of the over-the-scope clip.

Other potential adverse events reported in larger cohorts that we did not observe include post-polypectomy syndrome and enterocolonic fistula formation.³

Technical complications, which occurred in two cases—snare and clip entrapment—with rare device failures, have been reported in other series.¹² Ovesco have since modified the FTRD system to prevent the likelihood of this occurrence, which occurred very early in our experience, and we have not had any further recent episodes. It does also emphasise the importance of appropriate training and advanced endoscopic skills to navigate these challenges.

Strengths of this study include the enrolment of consecutive patients, which limits selection bias. This is a real-world experience that includes the complete learning curve in terms of lesion resection and technical use of this device. Limitations include its retrospective nature.

In conclusion, we report the first case series in New Zealand of the FTRD for management of GI lesions. Our findings are comparable to international evidence. This work highlights the potential of the FTRD as a valuable tool in the interventional endoscopist’s arsenal as a modality to potentially avoid surgical resection. It does also, however, underscore the need for careful patient selection, individualised treatment planning and consent. Further research is required to assess long-term outcomes of the lesions managed with the FTRD and also further refinement of lesion and patient assessment to delineate the role of the FTRD in GI lesion management.