The QNI (Quadrant–Necrosis–Infection) classification system has emerged as a structured framework for characterizing walled-off necrosis (WON). However, its performance in predicting clinical severity, procedural complexity, and Endoscopic Retrograde Cholangiopancreatography (ERCP)- related pancreatic duct pathology during endoscopic management of pancreatic pseudocysts and WON in real-world settings remains underexplored. This study evaluates outcomes among low and high-risk QNI groups undergoing endoscopic drainage.

In this retrospective analysis of 300 patients who underwent endoscopic drainage for pancreatic pseudocysts/WON, subjects were stratified into low-risk (n=232) and high-risk (n=68) categories based on QNI criteria. Clinical characteristics, procedural interventions, ERCP requirement and findings, and outcomes were compared using standard statistical methods, with significance set at P<0.05.

Baseline age and sex distribution were comparable between groups. High-risk QNI patients demonstrated significantly greater disease severity, with higher rates of sepsis (41.2% vs 18.5%, P<0.001) and complications (55.9% vs 28.0%, P<0.001). High-risk group was characterised by increased need for necrosectomy (98.5% vs 37.5%, P<0.001), more necrosectomy sessions (median 3 vs 1, P<0.001), and higher rates of nasocystic drainage (67.6% vs 22.8%, P<0.001). Irrigation—especially with saline or betadine–hydrogen peroxide—was also significantly more frequent in the high-risk cohort. Metal stent use predominated in both groups but was more common in high-risk patients (98.5% vs 90.5%, P=0.04). Among high-risk subjects, additional radiologic or surgical interventions were required in 10 patients(14.7%)- 7 patients(10.4%) required percutaneous drainage, 3 patients(4.4%) required percutaneous endoscopic drainage. In high risk subjects, The median length of hospital stay did not differ based on requirement of irrigation (P>0.05) and it was highest among those irrigated with betadine and hydrogen peroxide (21.0 days) and least among those who did not require any irrigation (13.0 days). ERCP requirement was similar between low- and high-risk groups (61.6% vs 51.5%, P>0.05), as were pancreatic duct stenting rates (58.7% vs 60.0%, P>0.05). Across both groups, pancreatic duct leak was the most common ERCP finding (47.6% low risk; 57.1% high risk), followed by duct disconnection (30.1% vs 28.6%). Strictures were infrequent, and normal ducts were observed in a minority. Among high-risk subjects, most did not require additional radiologic or surgical interventions (85.3%). Hospital stay duration did not significantly differ by irrigation modality (P>0.05).

The high-risk QNI phenotype was associated with greater clinical severity, higher rates of sepsis, and markedly increased hospital stay. ERCP findings—dominated by ductal leaks and disconnections—were comparable between risk groups, suggesting ductal pathology may be independent of QNI stratification. These findings support the utility of QNI scoring in predicting severity and resource utilization and reinforce QNI as a meaningful classification tool for guiding endoscopic management strategies and future comparative effectiveness studies in pancreatic pseudocyst and WON therapy.