Prognostic value of admission D-dimer in acute pancreatitis: association with disease severity and clinical outcomes in a retrospective cohort study

Early risk stratification in acute pancreatitis (AP) is crucial for improving patient outcomes and guiding timely clinical management. D-dimer has emerged as a potential biomarker reflecting activation of coagulation and fibrinolysis and may assist in early risk stratification of patients with acute pancreatitis. However, its prognostic performance remains incompletely validated across different populations. This retrospective cohort study included 200 adult patients diagnosed with acute pancreatitis between March 2023 and March 2024 at a tertiary care hospital. Disease severity was classified using the Revised Atlanta Classification. Admission serum D-dimer was measured within the first six hours after hospital presentation. Admission D-dimer levels were analyzed in relation to clinical severity, BISAP and Ranson scores, ICU admission, length of hospital stay, and in-hospital mortality. Statistical analyses included non-parametric tests, correlation analysis, logistic regression, and ROC curve analysis to evaluate predictive performance. Two hundred consecutive patients, with a mean age of 58.1 ± 17.5 years and 54% male patients, were included in this study. Severity distribution showed 68.5% mild, 17% moderately severe, and 14.5% severe AP. Overall mortality was 7% ( n = 14), and 56% ( n = 112) required ICU admission. Admission D-dimer levels increased significantly with disease severity, with median values of 1534 ng/mL (mild), 1908.5 ng/mL (moderately severe), and 2960 ng/mL (severe). Non-survivors had markedly higher levels than survivors (3776.5 vs. 1663 ng/mL, p < 0.001). D-dimer showed significant positive correlations with BISAP (ρ = 0.6068), Ranson scores (ρ = 0.3425), and hospital length of stay. ROC analysis demonstrated good discriminatory ability for severe AP (AUC = 0.778; cutoff 2035 ng/mL) and good discriminatory performance for mortality prediction (AUC = 0.912; cutoff 2726 ng/mL, sensitivity 85.7%, specificity 80.1%). However, performance for predicting ICU admission was limited (AUC = 0.579). Multivariable analysis demonstrated that each 1-ng/mL increase in admission D-dimer level was independently associated with higher odds of severe AP (OR = 1.001, p < 0.001). Admission D-dimer is a simple and rapid biomarker associated with disease severity and mortality in acute pancreatitis. While it should not replace established scoring systems, it can serve as a valuable adjunct for early risk stratification, particularly for identifying high-risk patients. Prospective multicenter studies are needed to validate its clinical utility and define standardized cutoff values.