IDDF2024-ABS-0012 Relationships between serum HMGB1 and intestinal mucosal barrier damage in acute pancreatitis

<h3>Background</h3> The objective of this study is to explore the correlation between serum levels of high mobility group protein B1 (HMGB1) and intestinal mucosal barrier impairment in cases of acute pancreatitis (AP). <h3>Methods</h3> In this retrospective study, patients diagnosed with MSAP and SAP between January 2008 and August 2019 from the Medical Information Mart for Intensive Care-IV (MIMIC-IV) database were included. A control group of 56 non-AP patients was also selected during the same time period. Serum levels of high mobility group protein B1 (HMGB1) and diamine oxidase (DAO) were measured using the ELISA method. The endotoxin (LPS) level was determined through tachypleus amebocyte lysate turbidimetry. Enzymological spectrophotometry was employed to assess the serum D-lactic acid level. Pearson correlation analysis was conducted to investigate the associations between serum HMGB1 levels and the intestinal mucosal injury index. Additionally, the diagnostic accuracy of serum HMGB1 in identifying intestinal mucosal injury in AP patients was assessed using the ROC curve analysis. <h3>Results</h3> The levels of D-lactic acid, LPS, and DAO exhibited significant elevations in the observation group compared to the control group at 24, 48, and 72 hours (P&lt;0.05). Furthermore, serum HMGB1 levels demonstrated significant increases in the observation group relative to the control group at 24, 48, and 72 hours (P&lt;0.05). Moreover, the levels of serum HMGB1 in the SAP group demonstrated noteworthy elevations compared to the MSAP and MAP groups at 24, 48, and 72 hours (P&lt;0.05). At 24 hours after symptom onset, serum HMGB1 levels were found to be positively correlated with indicators of intestinal mucosal injury including D-lactic acid, LPS, and DAO (P&lt;0.05). The areas under the curve for serum HMGB1 in diagnosing intestinal mucosal injury in AP patients were 0.85 at 24 hours after symptom onset, with a sensitivity of 87.04% and a specificity of 77.10%. <h3>Conclusions</h3> Implementing this model can assist healthcare professionals in categorizing patients for primary prevention and timely therapeutic intervention, which can enhance prognosis.