Introduction

Endoscopic submucosal dissection (ESD) enables the en bloc ressection of superficial lesions of the gastrointestinal tract. Initially developed in the East, the implementation of the technique has been more difficult in the West, mainly due to the lower incidence of gastric neoplasia.

Aims & Methods

In our study, we aimed to evaluate the clinical outcomes since the introduction of the ESD technique in two distinct centers and describe the learning curve of 2 endoscopists that performed the procedures. This is a prospective analysis composed of patients that underwent rectal and gastric ESD between 2019 and 2024. We evaluated en bloc and R0 resection rates and recurrence rates. Complications were defined as adverse events related to the procedure that required reintervention or prolongation of the hospital stay. The area of the resected lesion was calculated in cm² (assuming an eliptical shape) and the speed of dissection in cm²/hour. Using cumulative sum analysis, we identified the procedure in which a significative increase in the speed of dissection was observed and compared outcomes in the two time periods defined by this cut-off.

Results

We included a total sample of 100 patients (59% male, age 69.9±5.4 years).
Gastric ESD was conducted in 56 patients (57% male, age 59.3±2.9 years). En bloc ressection rate was 98.2% (n=55) and R0 resection was 91.1% (n=51). 10.7% of the procedures had complications (1 perforation, 2 lesions of the muscular layer and 3 post-procedural bleedings). The mean area of the resected specimens was 7.8 ± 5.0 cm². The mean dissection speed was 8.2 ± 5.2 cm²/hour. The mean hospital stay was 1.3 ± 0.3 days.
There was a significant increase in proficiency by the procedure number 27. When comparing the two groups, there was a tendency for an increase in R0 resection (88.9% vs 93.0%, p=0.294) and a decrease in the recurrence rate (12% vs 5%, p=0.194) in the later time period. There was a significant reduction in the complication rate (18.5% vs 3.4%, p<0.001).
Rectal ESD was performed in 44 patients (61.4% male, age 71.5±3.1 years). En bloc resection was 97.7% (n=43) and R0 resection was 86.4% (n=38). The complication rate was 38.6% (9 post-procedural bleedings; 8 post-coagulation syndromes), none of which required surgical treatment. The mean area of the resected specimens was 13.4 ± 4.6 cm² and mean dissection speed was 6.5 ± 4.1 cm²/hour. The mean hospital stay was 2.2 ± 0.7 days.
There was a significant increase in proficiency by the procedure number 33. The R0 resection rate was similar between groups (85%vs82%, p =0.34), with a tendency to a decrease in complication rate (36%vs27%, p=0.29). A total of 9 adenocarcinomas were resected, with endoscopic cure in 88.9% cases (n=8).

Conclusion

We believe that the evaluation of the results of this intervention allows not only to assure that the quality criteria are being met but also to aid in the creation of learning programs for trainees. The learning curve associated with gastric ESD was steeper than that of rectal ESD. It appears that rectal ESD requires a larger number of procedures until the quality criteria proposed by the European Society of Gastrointestinal Endoscopy are met.

Disclosure

No conflict of interest.

Introduction

The learning curve for EUS-guided Gastroenterostomy (EUS-GE) has been assessed through retrospective studies involving few elite senior endoscopists (SE), using procedural time as the main surrogate outcome. This study aims to evaluate the training course of junior endoscopists (JE) approaching EUS-GE using the WEST technique, considering both technical efficiency and clinically relevant patient outcomes.

Aims & Methods

From a prospective single-centre registry of EUS-GE (PROTECT, ClinicalTrials.gov NCT04813055) using the Wireless Simplified Technique (WEST), the learning curve of 3 JE, experienced in pancreatobiliary endoscopy (PBE) was analysed to identify inflection points (“knots”), applying adaptive regression and cumulative sum control chart (CUSUM) on several continuous and dichotomous variables. EUS-GE performed by a SE and those on post-surgical anatomy were excluded.

Results

From a prospective database of 165 EUS-GE, 100 were performed by a JE with previous EUS-GE training (JE1) and 26 by two naïve JEs.
In JE1’s curve, neither procedural nor fluoroscopy time showed any improvement while radiation dose (mean Air Kerma: 392 ± 317 Gy) significantly decreased (F-test p=.0005) after 57 (adaptive regression) and 72 interventions (CUSUM), stabilizing thereafter. The need for supervision/backup from a SE showed a significant knot after 32 procedures (p<.0001). Technical success (100%), clinical success (95.0%), adverse events (AEs; 6.0%) and moderate/severe AEs (4.0%) remained constant through the learning curve, clustered according to the aforementioned knots. Conversion from EUS-GE to enteral stenting was rare (7.2%) and non-significantly decreased over time.
Despite requiring significantly higher fluoroscopy time (234 vs. 187 seconds, p=.009) and procedural time (55 vs. 43 minutes, p=.002), naïve endoscopists achieved efficacy and safety outcomes comparable to the remaining cohort.

Conclusion

From the analysis of a prospective learning curve, performing ≈ 30 EUS-GE seems required for independent practice, and ≈ 70 to stabilize the fluoroscopy usage. EUS-GE training of naïve PBE endoscopists in high-volume institutions, using a standardized technique, does not compromise clinical and safety standards.

Introduction

Endoscopic submucosal dissection (ESD) is widely implemented in Asia. Experience from the western world is still limited, so the evidence on the description of a learning curve (LC) is sparse.

Aims & Methods

The aim of this study is to evaluate the LC for ESD of gastric lesions using accepted proficiency benchmarks (PB).
Retrospective analysis with a prospectively maintained database of all patients undergoing ESD at a Western tertiary center from June 2016 to December 2022.
Inclusion criteria: ESD of gastric lesions performed by a single operator.
Primary endpoint: a LC to estimate the number of ESDs required to achieve PB (>90% for en bloc resection, >80% for histologic margin-negative (R0) resection and resection speeds (RS) >9cm2/hr), using the CUSUM method.

Results

Of 477 ESD performed, a total of 235 ESD met the inclusion criteria.
The en bloc resection rate was 100%. The R0 resection rate was 97% so no statistically significant differences were found, since it was always within PB.
Using the CUSUM method, the LC was divided into learning periods [Phase I - learning phase (cases 1-48); Phase II - adaptation phase (49-72), Phase III - consolidation (from case 73 on).
Globally, after reaching the level of 9cm2/h and after resection 73, the cumulative moving average remains above this speed for all following resections.
Several locations reach moving average proficiency levels at different number of interventions (antrum: from the beginning, body: 134th).

Conclusion

The PB regarding the en bloc and R0 resection were achieved from the beginning.
The PB regarding de RS was achieved from the beginning for the antrum.
The PB regarding de RS, globally, has a clinically significant first learning inflection at ESD number 49 and at ESD number 73 achieved sustained PB for RS.

Disclosure

There are no conflict of interest.

Introduction

Endoscopic submucosal dissection (ESD) is recommended as a treatment modality for early gastric cancer because it allows organ-sparing, curative en-bloc resection, and complete histopathological evaluation. Although many endoscopists are interested in learning how to perform ESD, the procedure requires a high level of expertise and is technically challenging, especially for less experienced endoscopists.

Aims & Methods

This study aimed to investigate the single-operator learning curve and clinical outcomes of consecutive gastric ESDs. Eight hundred-eleven gastric ESDs were performed between March 2016 and December 2022. The learning curve for ESD was evaluated using the cumulative sum (CUSUM) method to analyze the number of ESDs required for achieving proficiency.

Results

Of the enrolled patients, 310 (40.9%) cases were diagnosed with early gastric cancer. Overall en-bloc resection rates, R0 resection, and curability is 98.6%, 96.3%, and 93.7%, respectively. The median ESD speed was 18.1 cm2/h (IQR 12.2 – 25.3). ESD speed was slower than average in the upper third (9.9 cm2/h), presence of submucosal fibrosis (11.0 cm2/h) and the presence of subcutaneous fat (11.1 cm2/h), and faster than 20 cm2/h in the lower third (20.2 cm2/h) and lesion sized > 3 cm (23.6 cm2/h). According to the existing literature, the proficiency benchmark is defined as en-bloc resection >90%, R0 resection and >80%, and resection speed >9 cm2/h. However, since the first case satisfies all the existing benchmarks, we should establish a new definition of speed among benchmarks. In the changes of resection speed, we defined 16.5 cm2/h as a benchmark for ESD speed, which is the median value of the rapidly changing slope, and the CUSUM curve revealed that 156 cases were needed to achieve proficiency.

Conclusion

Due to the development of endoscopic equipment, a higher level of ESD proficiency benchmark is required. One hundred sixty-five cases were required to attain a higher technical level of successful performance in gastric ESD.

Introduction

The outcome of ESD in early gastrointestinal neoplasia in Japan is well established. However, in the west, ESD practice is variable given different disease prevalence in UGI tract and also the volume and experience of each centre.

Aims & Methods

We aim to evaluate ESD practice and outcomes in the UK and hence, multi centre registry was set up. This is a prospective observational study for patients undergoing ESD for early GI neoplasia. All technical and outcomes data was prospectively collected onto REDCap.

Results

There were 451 UGI ESD cases recruited between August 2016 to August 2023 from 6 large tertiary referral centres in the UK. Mean age of the patients was 70.8 years (range: 24-94 years) with male preponderance (318 males vs. 133 females). Majority were in oesophagus: 282 (62.53%), stomach: 159 (35.25%) and duodenum: 10 (2.22%).
Of 282 oesophageal ESDs, 277 (98.23%) were technically successful with en-bloc resection rate of 96%. There were 3 intraprocedural complications (2 oesophageal perforations and 1 oral laceration from overtube). All of them were managed endoscopically. There were 29 (10.28%) post ESD strictures requiring endoscopic dilatations. 3 (1.06%) delayed bleeding were noted and 3 were readmitted for decompensation of underlying co-morbidities.
There were 204/282 (72.34%) Barrett’s cases with average length of C3 M5 and mean size 34.5mm. 34 cases (16.67%) had previous ESD or ablation at the same site. Post ESD resection showed deep submucosal (SM) invasion in 21 patients (deep R0 achieved in 42.9% and lateral R0 in 76.2%), 23 pT1bSM1 (deep R0 in 87.0% and lateral R0 in 87%), 107 pT1a (deep R0 in 89.7% and lateral R0 in 85.1%), 31 high grade dysplasia (HGD) (R0 in 97%) and 9 low grade dysplasia (LGD) (R0 in 78%) as shown in Table 1.
There were 44 squamous cell carcinoma (SCC), 26 gastro-oesophageal junctional neoplasia and 8 others (submucosal lesion and granular cell tumours). Of 44 SCC, there were 8 pT1b (5 achieved deep R0, 7 achieved lateral R0), 10 pT1a (all achieved both deep and lateral R0), 22 HGD (81.8% achieved R0), 3 LGD (all achieved R0) as shown in Table 1.
There were 160 gastric ESDs with completion rate of 96.25% and en-bloc rate of 95.5%. There were 8 (5%) intra-procedural perforation which required endoscopic clippings. There were 7 (4.36%) delayed bleeds which were managed endoscopically. Of 160 cases, 138 were epithelial neoplasia, 11 were submucosal lesions while 4 were large hyperplastic polyps. As shown in Table 1, of 138 epithelial neoplasia, 10 had deep SM invasion (7 achieved deep R0, 8 lateral R0), 14 pT1bSM1 (7 achieved deep R0, 10 lateral R0), 50 pT1a (90% achieved deep R0, 80% lateral R0), 35 HGD (94.3% R0) and 27 LGD (77.8% R0).

Conclusion

Unlike in the East, most ESDs in our study were performed in Barrett’s oesophagus rather than the stomach. Our data demonstrates the feasibility and safety of UGI ESD in Western setting with high en-bloc rate and low complication rates. We found that within indication of ESD, complete resection (R0) rate is high but lesions with deep submucosal invasion have poor R0 rate.

Disclosure

Professor Bhandari has received research grants or is the advisory board for Fujifilm, Boston, Olympus, Pentax, 3-D matrix, NEC (Japan), Medtronic.

Introduction

Despite undeniable advantages as higher en-bloc and curative resection rates, better carcinologic outcomes and lower recurrence rates, endoscopic submucosal dissection (ESD) is associated with more adverse events, than piece-meal endoscopic mucosal resection, limiting its spread in the West. However, risk factors in western patients and clinical implications of perforation are not well known.

Aims & Methods

This is a prospective cohort including all colorectal ESD performed in 12 French centers and 1 Belgian center between September 2019 and September 2022, with additional retrospective analysis of patients with per-procedural or delayed perforations.
The objectives of this study were to analyze prevalence, risk factors and clinical outcomes of perforation following colorectal ESD.

Results

Per-procedural perforations (PPP) (n=354) or delayed perforation (DB) (n=20) occurred in 374/4025 procedures, with a mean ±SD lesion size of 54 ±26 mm, and median procedure time of 57 min (IQR 34-90).
In the perforation group, lesions were larger (60 ±32 mm vs 53 ±25 mm, p= 0,001), with lower resection speed (25 ±21 mm2/min vs 35 ±24 mm2/min) and presented more severe fibrosis (55,5% vs 26,2%, p<0,0001). En-bloc (87,2% vs 95,8%), R0 (74,0% vs 89,0%) and curative resection (71,2% vs 85,2%) rates were lower in the perforation group (all p<0,0001).
Non-granular laterally spreading tumors (LST) was a risk factor of both PPD (p=0,036) and DP (p<0,0001). Female sex was associated with PPP (p=0,019) but not with DP (p=0,714).
PPP were more frequent in intermediate-volume centers compared to low and high volume (p<0,0001), but there was no difference regarding the risk of DP (p=0,949).
Poor manoeuvrability was associated with PPP (p<0,0001) but not with DP (p=0,351).
PPP were particularly frequent in the splenic (21/102 lesions, 19,0%) and hepatic (30/184 lesions, 16,3%) flexures and in the appendix (20/118, 16,9%), with statistical significance (p<0,0001). All DP occurred in the colon but there was no statistical difference between the different segments.
Prophylactic closure of the scar did not prevent DP (OR 1,22 95%CI 0,41-1,22).
By calculation of area under the curve (AUC), a threshold of ³ 70 mm was identified as predictive for immediate perforation (p<0,0001).
In multivariate analysis, lesion ³ 70 mm (OR 2,17 95%CI 1,50-3,15), speed of resection (OR 0,98 95%CI 0,97-0,99), female sex (OR 1,30 95%CI 1,01-1,68), poor maneuverability (OR 1,53 95%CI 1,15-2,03) and severe fibrosis (OR 3,2 95%CI 2,30-4,4,43) were predictive of PPP.
Regarding clinical burden, DP required surgery in most of the cases (17/20, 85%) whereas PPP were treated conservatively (13/354 emergent surgeries, 3,7%) (p<0,0001). Stoma was performed in 37,5% of cases. Among rectal perforations, no closure was done in 23% of cases without the need for surgical or radiological interventions.
In colonic PPP, the need for peritoneal exsufflation or the type of closure (TTS clips vs OTS clips) were not associated with surgery. CRP and white cell blood levels on day 1 after ESD were not different between groups but fever was associated with surgery (OR 9,41 95%CI 2,44-36,29).

Conclusion

In this large study, risk factors for per procedural perforations could be identified and conservative treatment appears effective. Delayed perforations are rare and cannot be predicted but require surgery in most of the cases. This could help the patient and the physician to choose the best treatment with balance of benefices and risks.