CLINICAL AND PERFORMANCE OUTCOMES OF COLORECTAL SUBMUCOSAL DISSECTIONS USING A NOVEL ADVANCED ENERGY PLATFORM: A SINGLE PROSPECTIVE COHORT OBSERVATIONAL STUDY

Introduction
This single centre prospective observational study aims to measure clinical and
performance outcomes using the Speedboat 10fr Inject device/SBI, and the
Speedboat-assisted Endoscopic Submucosal Dissection/S-ESD modality when
removing complex colorectal lesions.

Methods and Aims
The primary outcome was rate for en-bloc and conversion to piecemeal resection. Other clinical outcomes: Peri-endoscopic complications, unplanned surgery and surveillance were recorded. Polyp features (size, surface, location, morphology, and submucosal fibrosis), technical and procedural specifications, and dissection time and speed were also collected. Univariate and multivariate analyses were used to identify factors affecting performance in the whole cohort and across 3
different periods with equally distributed number of cases to assess the
performance curve.

Results
From a total of 200 cases (33 operational months), 179 were endoscopically resected. 21/200 cases were abandoned: 16 for muscle retraction sign and 5 for medical non-endoscopic emergencies. Mean polyp surface was 18.32cm2, mean long axis length was 6.5cm (biggest polyp axes of 15x13cm) and mean procedure duration was 106.35 minutes. Out of 179 resected polyps, en-bloc resection was achieved in 92.2% (n=165/179) with a curative rate of 82% (n=147/179) including 22 malignant polyps. Conversion to piecemeal resection was 7.8% (n=14/179) including 4 cases of device failure. Delayed bleeding occurred in 3 patients out of 165 (1.82%). No colorectal perforation and unplanned emergency surgery were recorded. No mortality was noted. One 5mm benign recurrence (0.6%) in the largest complex rectal polyp covering 95% of the circumference was noted. Overall mean dissection speed was 9.98 cm2/hr (SD 5.86). Multivariate analysis showed that key energy modalities of the S-ESD function efficiently to tackle complex features of colorectal lesions (large surface, mild and severe fibrosis, and incidence of bleeding). Under these clinical conditions (same operator, designated device, and modality) and these variables, polyp surface and presence of severe fibrosis were found to account for 70% of variance in dissection duration and 30% of variance in dissection speed. Larger polyps were removed faster in period 3 (12.05cm2/hr) compared to period 1 (7.97cm2/hr)/(p.002) with lower incidence of intraprocedural bleeding.

Conclusions
This single centre prospective cohort study demonstrates that the SBI is a safe
and effective device with minimal complications. The S-ESD appears to be an
acceptable modality to remove complex colorectal polyps in a timely manner even in larger polyp surface and higher degree of submucosal fibrosis. Over time,
proficiency with the S-ESD modality resulted in quicker dissection as polyp surface increased across similar severity of submucosal fibrosis.