Stenosis after colorectal endoscopic submucosal dissection: When to expect, how to manage?

Endoscopic Submucosal Dissection for Resections Larger than 10 cm: Outcomes from a Portuguese Center

Background

Endoscopic submucosal dissection (ESD) is a minimally invasive technique for en bloc resection of superficial neoplastic lesions, independent of their size. However, for giant gastrointestinal superficial neoplasia, the risk of invasive cancer is higher, and ESD is typically challenging. Despite the increasing literature on giant resections, data on their efficacy and safety are still lacking.

Objective

The aim of this study was to describe ESD outcomes from a Portuguese center, compare them with other international studies, and analyze the possible risk factors influencing outcomes.

Methods

We conducted a retrospective single-center review using a prospectively collected database, including patients with rectal ESD resections larger than 10 cm, between January 2016 and December 2021. Clinical, procedural, and pathological data were collected and analyzed. Revision of the literature for comparison with international results was done through PubMed. Data were analyzed and statistical analysis performed, using Microsoft Excel and SPSS, to identify significant risk factors.

Results

The study included 15 rectal resections, with a mean diameter of 140.9 mm (range 105-270), corresponding to lesions of 125.9 mm (87-238). The overall en bloc resection rate was 100% (n = 15). According to ESGE criteria, procedure was considered curative in 53.3% (n = 8), non-curative with high risk in 13.3% (n = 2), and local-risk recurrence in 33.3% (n = 5). Adverse events occurred in 26.7% (n = 4): 1 minor perforation and 3 stenosis, most endoscopically managed. For non-curative resections with local-risk recurrence, surveillance without adjuvant therapy was performed in all cases. For high-risk non-curative resections, surgery was performed in 1 patient and adjuvant chemoradiation therapy in another. Follow-up (mean 16 months) demonstrated a recurrence rate of 0%. Statistical analysis revealed resection size ≥20 cm as a risk factor for perforation (p value 0.067), and involvement of ≥90% of the circumference and procedural time ≥4 h as risk factors for stenosis (p value 0.029 and 0.009, respectively).

Conclusions

Although challenging, ESD for giant lesions seems effective and safe, with a still relevant rate of complications, which were mostly endoscopically treated. Rigorous characterization of lesions is crucial to predict and avoid complications or the need for therapy escalation.

Prolonged ischemia of the ileum and colon after surgical mucosectomy explains contraction and failure of "mucus free" bladder augmentation

INTRODUCTION

Mucus production by the intestinal segment used in bladder augmentation results in long term concerns especially stones and UTI. Bladder augmentation with demucosalized intestinal flap is a potential promising approach for mucus-free bladder augmentation, however the contraction of the flap remains a major concern. Mucosectomy has been shown to result in abrupt and immediate cessation of microcirculation in the ileum. However, assessment of microcirculation shortly after mucosectomy may miss a gradual recovery of micro-circulation over a longer period of time. Previous studies have not assessed the colon response to mucosectomy.

OBJECTIVE

Our aim was to assess the effect of mucosectomy on the microcirculation of the colon and ileum beyond the known warm ischemia time.

STUDY DESIGN

Ileum and colon segments were detubularised and mucosectomy was performed in (n = 8) anesthetised minipigs. Group A: sero-musculo-submucosal flaps were created with removal of the mucosa and preserving the submucosal layer Group B: sero-muscular flaps were created with the removal of submucosal-mucosal layer. The Microvascular Flow Index (MFI), the velocity of the circulating red blood cells (RBCV) was measured using Intravital Dark Field (IDF) side stream videomicroscopy (Cytoscan Braedius, The Netherlands) after mucosectomy, for up to 180 min.

RESULTS

Both the MFI and RBCV showed an abrupt reduction of microcirculation, on both surfaces of the remaining intestinal flap, in the ileum as well as in the colon. Slightly better values were seen in Group A of the colon, but even these values remain far below the preoperative (control) results. Some, tendency of recovery of the microcirculation was noted after 60-90 min, but this remained significantly lower than the preoperative control values at 180 min.

CONCLUSION

Both the ileal and the colonic flap remains in severe ischemia after mucosectomy beyond the warm ischemia time.

DISCUSSION

This study shows that surgical mucosectomy compromises vascular integrity of the intestinal flaps used for bladder augmentation. Partial recovery which occurs within the warm ischemia time is not significant enough to avoid fibrosis therefore flap shrinkage may be inevitable with this technique.

LIMITATION

The gastrointestinal structure of the porcine model is not the same exactly as the human gastrointestinal system. However, although not an exact match it is the closest, readily available animal model to the human gastrointestinal system.

A novel technique of endoscopic submucosal dissection for circumferential ileocecal valve adenomas with terminal ileum involvement: the "doughnut resection" (with videos)

BACKGROUND

Ileocecal valve (ICV) lesions are difficult to resect endoscopically and patients are often referred for laparoscopic colectomy. ICV involvement has been shown to be related to technical failure and tumor recurrence after endoscopic mucosal resection (EMR) and represents a challenge for endoscopic submucosal dissection (ESD). Few publications have focused specifically on endoscopic management of ICV lesions.

METHODS

We developed a novel ESD technique, the "doughnut resection," for circumferential ICV adenomas with terminal ileum involvement. Two circumferential mucosal incisions are performed, one in the ileum and the other in the cecum, followed by submucosal dissection of the disk of tissue between the two incisions around a guiding stent placed across the valve that helps guide the dissection as it crosses the valve orifice. The lesion is removed en bloc in the shape of a "doughnut" with two concentric assessable lateral margins. The underwater ESD technique and a gastroscope were used to facilitate the resection.

RESULTS

Seven patients received the doughnut resection. The median patient age was 67 years. All patients had prior biopsy and three had prior endoscopic resection (1-6 times). The median specimen diameter was 4.5 cm (range 3-8). All resections were en bloc and R0. There was no perforation, delayed bleeding, or other clinically significant adverse events. After a median follow-up of 21 months (range 12-32), there was no tumor recurrence.

CONCLUSION

The "doughnut resection" is a feasible, safe, and effective method to remove circumferential ICV lesions endoscopically even for patients with multiple prior tumor manipulations.