Standardised training for endoscopic mucosal resection of large non-pedunculated colorectal polyps to reduce recurrence (*STAR-LNPCP study): a multicentre cluster randomised trial
Lonne W T Meulen,Roel M M Bogie,Peter D Siersema,Bjorn Winkens,Marije S Vlug,Frank H J Wolfhagen,Martine Baven-Pronk,Michael van der Voorn,Matthijs P Schwartz,Lauran Vogelaar,Wouter H de Vos tot Nederveen Cappel,Tom C J Seerden,Wouter L Hazen,Ruud W M Schrauwen,Lorenza Alvarez Herrero,Ramon-Michel M Schreuder,Annick B van Nunen,Esther Stoop,Gijs J de Bruin,Philip Bos,Willem A Marsman,Edith Kuiper,Marc de Bièvre,Yasser A Alderlieste,Robert Roomer,John Groen,Marloes Bargeman,Monique E van Leerdam,Linda Roberts-Bos,Femke Boersma,Karsten Thurnau,Roland S de Vries,Jos M Ramaker,Frank P Vleggaar,Rogier J de Ridder,María Pellisé,Michael J Bourke,Ad A M Masclee,Leon M G Moons
DOI: https://doi.org/10.1136/gutjnl-2023-330020
IF: 24.5
2024-01-12
Gut
Abstract:Objective Endoscopic mucosal resection (EMR) is the preferred treatment for non-invasive large (≥20 mm) non-pedunculated colorectal polyps (LNPCPs) but is associated with an early recurrence rate of up to 30%. We evaluated whether standardised EMR training could reduce recurrence rates in Dutch community hospitals. Design In this multicentre cluster randomised trial, 59 endoscopists from 30 hospitals were randomly assigned to the intervention group (e-learning and 2-day training including hands-on session) or control group. From April 2019 to August 2021, all consecutive EMR-treated LNPCPs were included. Primary endpoint was recurrence rate after 6 months. Results A total of 1412 LNPCPs were included; 699 in the intervention group and 713 in the control group (median size 30 mm vs 30 mm, 45% vs 52% size, morphology, site and access (SMSA) score IV, 64% vs 64% proximal location). Recurrence rates were lower in the intervention group compared with controls (13% vs 25%, OR 0.43; 95% CI 0.23 to 0.78; p=0.005) with similar complication rates (8% vs 9%, OR 0.93; 95% CI 0.64 to 1.36; p=0.720). Recurrences were more often unifocal in the intervention group (92% vs 76%; p=0.006). In sensitivity analysis, the benefit of the intervention on recurrence rate was only observed in the 20–40 mm LNPCPs (5% vs 20% in 20–29 mm, p=0.001; 10% vs 21% in 30–39 mm, p=0.013) but less evident in ≥40 mm LNPCPs (24% vs 31%; p=0.151). In a post hoc analysis, the training effect was maintained in the study group, while in the control group the recurrence rate remained high. Conclusion A compact standardised EMR training for LNPCPs significantly reduced recurrences in community hospitals. This strongly argues for a national dedicated training programme for endoscopists performing EMR of ≥20 mm LNPCPs. Interestingly, in sensitivity analysis, this benefit was limited for LNPCPs ≥40 mm. Trial registration number NTR7477.
gastroenterology & hepatology