Cumulative intra-abdominal pressure exposure and dynamic trajectories in ICU-admitted patients reveal prognostic determinants of severe acute pancreatitis

Abstract

Intra-abdominal pressure (IAP) critically drives organ failure progression in severe acute pancreatitis (SAP). However, traditional static IAP monitoring inadequately captures dynamic injury evolution. This study aimed to assess the impact of cumulative IAP exposure (CumIAP) and dynamic trajectories on the prognosis of SAP. This retrospective cohort study analyzed 1,008 ICU-admitted SAP patients from the Jiangxi cohort (2005–2023) and 83 from MIMIC-IV (2008–2019). CumIAP was quantified via time-weighted integration of serial IAP measurements. Multivariate Cox regression models and restricted cubic splines (RCS) were established to analyze the dose–response relationships between CumIAP and death, infectious pancreatic necrosis (IPN), and persistent multiple organ failure (PMOF). Mediation analysis evaluated CumIAP’s role in albumin (ALB)- and acute necrotic collection (ANC)-associated outcomes. Latent class growth mixture model (LCGMM) was employed to identify dynamic IAP trajectory subtypes, and the associations between each trajectory group and poor prognosis were analyzed. Over a median inpatient follow-up of 17 days in the Jiangxi cohort, 200 (19.8%) patients died in hospital, while 208 (20.6%) and 329 (32.6%) developed IPN and PMOF, respectively. Regression analysis revealed that for each standard deviation increase in CumIAP, the risks of in-hospital death and PMOF increased by 37% and 86%, respectively, and there was a U-shaped association with the risk of IPN (P for nonlinearity = 0.004). Mediation analysis showed that CumIAP mediated 24.26% and 33.76% of the associations between ALB, ANC, and the risk of in-hospital death, respectively. Three IAP trajectories were identified by LCGMM: the high-pressure rapid decline group (HRD-T1), the low-pressure gradual decline group (LGD-T2), and the low-pressure progressive increase group (LPI-T3). Among them, compared with HRD-T1 and LGD-T2, the subjects in the LPI-T3 group had a significantly increased risk of adverse clinical outcomes. This is the first study to revealed that CumIAP is linearly positively correlated with death and PMOF, while exhibits a U-shaped relationship with IPN. Notably, patients with low baseline IAP and a rising trajectory exhibited worse outcomes than those with high baseline IAP and a declining trend. This is the first study to introduce CumIAP as a quantitative metric to assess the early IAP burden and its prognostic significance in SAP, overcoming the limitations of static IAP monitoring. RCS and regression analyses revealed a positive linear association between CumIAP and in hospital mortality/PMOF, and a U-shaped relationship with IPN, suggesting the possible presence of a risk threshold. LCGMM identified three IAP trajectory patterns; the "low baseline with rising trend" group (LPI-T3) had the worst outcomes, underscoring the value of dynamic over static IAP monitoring. Mediation analysis showed that CumIAP partly mediated the effects of hypoalbuminemia and ANC on outcomes, suggesting that IAP might represent an important intermediary pathway and a therapeutic target in SAP.