PREDICTING THE NEED FOR STEP-UP AFTER EUS-GUIDED DRAINAGE OF PERIPANCREATIC FLUID COLLECTIONS, INCLUDING QNI SCORE VALIDATION: A PROSPECTIVE COHORT STUDY

Background:
Despite EUS-guided drainage (EUS-FCD) is the preferred modality for management of peripancreatic fluid collections (PFCs), factors predicting the need for step-up to additional procedures (including radiology, surgery or additional endoscopies) are retrospectively explored in studies restricted to Lumen Apposing Metal Stents (LAMS) or excluding pseudocysts and post-surgical collections, which might require a step-up as well.

Methods:
All consecutive candidates to EUS-FCD with either LAMS or Double-pigtail Plastic Stents (DPPS) between 2020-2023 were included in a Prospective Registry of Therapeutic EUS (PROTECT). The role of baseline clinical and morphological factors, including the Quadrant / Necrosis / Infection (QNI) score, in predicting the need for any step-up to additional procedures was evaluated by stepwise logistic regression.

Results:
Twenty-three post-surgical PFCs, 19 pseudocysts and 18 walled-off necrosis [WOPN]) were included, 42% treated with LAMS and 58% with DPPS based on morphology. Clinical success was 92.9%, but 38% of patients required a step-up to necrosectomy (26.7%) or additional procedures. Patients requiring step-up versus those sufficing drainage had a higher number of quadrants involved (OR=3.6 [1.5-8.4]), a more frequent paracolic gutter extension (OR=6 [1.4-26]), a higher necrosis content (>30%, OR=7.2 [1.8-28.4] ; >60% OR=25.2 [4.2-153]) and were in the high-risk group according to the QNI score, with no difference in the diameter and rate of infection. At multivariate analysis, being in the high-risk QNI group was the only independent predictor of step-up (OR=9.8 [2.9-32.9]). The score however was not able to stratify in analyses restricted to WOPN only or LAMS only. High-risk QNI patients presented also a longer hospital stay (17.5 [8-45] vs 4 [2-6.5], p=0,001), without any difference in ICU stay and nutritional support.

Conclusions:
In this prospective series of consecutive non-selective PFCs undergoing EUS-FCD, QNI score was the only independent predictor of any need for step-up to additional procedures, and strongly correlated to the EUS-FCD allocation to LAMS versus DPPS, unravelling a potential additional utility.