Real-world impact of implementing lumen-apposing metal stents for pancreatic fluid collections: a nationwide Japanese study

Abstract

Background Lumen-apposing metal stents (LAMSs) are increasingly used for the endoscopic management of pancreatic fluid collections (PFCs), including walled-off necrosis (WON) and pancreatic pseudocysts (PCs). Objective To evaluate the nationwide impact of implementing an LAMS (approved in Japan in 2018) on clinical outcomes in real-world settings. Design Using a nationwide inpatient database, we identified 5885 patients who underwent endoscopic ultrasound (EUS)-guided treatment for PFCs at 550 hospitals between 2010 and 2023. We examined the association between treatment period (pre-LAMS vs post-LAMS implementation) and clinical outcomes. Multivariable logistic and linear regression analyses were performed to estimate ORs and coefficients, respectively. Findings were validated using a multi-institutional clinical cohort (n=618) from specialty hospitals with more detailed clinical parameters with data collection during the same period. Results In the pre-LAMS period, 3787 cases were treated exclusively with plastic stents. After 2018, LAMS use stabilised at around 50% among 2098 cases treated between 2019 and 2023. Compared with the pre-LAMS period, patients treated during the post-LAMS period had a higher risk of bleeding with LAMS (adjusted OR, 1.81; 95% CI: 1.55 to 2.13), but not with plastic stents (adjusted OR, 0.96; 95% CI: 0.74 to 1.26). Rates of rescue surgery, in-hospital mortality and length of stay did not differ significantly (p>0.12). The post-LAMS period was associated with increased total costs (adjusted coefficient, US$2813; 95% CI: 1503 to 4122). In analyses stratified by PFC types (WON vs PC), the association of the post-LAMS period with bleeding risk appeared to be stronger for PCs than WON (pinteraction=0.015) and WON patients had a shorter length of stay and lower total costs in the post-LAMS period (pinteraction<0.001). In the validation cohort, the elevated bleeding risk with LAMS was confirmed (adjusted OR, 1.59; 95% CI: 0.86 to 2.95, vs pre-LAMS period), though not statistically significant. Conclusions The implementation of LAMS was not associated with improved key clinical outcomes but was linked to a higher risk of bleeding and increased healthcare costs. These findings do not support the routine use of LAMS for EUS-guided treatment of all PFCs in standard clinical practice but may restrict it to WON cases with clear indications. The datasets derived from the DPC database and analysed during the current study are not publicly available due to contracts with the hospitals providing data to the DPC database. The de-identified data from the WONDERFUL cohort and analytical methods used in the current study will be available from the corresponding author upon reasonable request.