Predictive value of inflammatory markers for disease severity in patients with acute pancreatitis in pregnancy: a logistic regression analysis

This study aimed to investigate the associations of inflammatory markers—namely the Neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), and lymphocyte-to-monocyte ratio (LMR)—with disease severity in patients with acute pancreatitis in pregnancy (APIP), identify risk factors and evaluate the predictive value of these markers for disease severity. Patients diagnosed with APIP at Shengjing Hospital, China Medical University, between January 2020 and December 2025 were enrolled. They were classified into severe and non-severe groups (the latter including mild and moderate cases) based on disease severity. Demographic characteristics and laboratory parameters—including age, parity, gravidity, gestational age, disease type, white blood cell count (WBC), neutrophil percentage, lymphocyte percentage, neutrophil count (NEU), lymphocyte count (LYM), platelet count (PLT), and monocyte count (MON)—were collected for retrospective analysis. NLR, PLR, and LMR were calculated. Correlations between these markers and disease severity were analyzed. Logistic regression was performed to identify risk factors for disease severity and receiver operating characteristic (ROC) curves were constructed to assess diagnostic performance. Significant differences between the severe and non-severe groups were observed in disease type, WBC, neutrophil percentage, lymphocyte percentage, NEU, LYM, NLR, PLR, and LMR ( P < 0.05). NLR and PLR showed moderate positive correlations with disease severity ( r = 0.544, r = 0.412, P < 0.05), whereas LMR showed a moderate negative correlation. Multivariate logistic regression identified NLR as an independent risk factor for disease severity. ROC analysis yielded area under the curve (AUC) values for predicting disease severity of 0.816 for NLR, 0.740 for PLR, 0.743 for LMR, and 0.825 for the combination of all three markers. The corresponding sensitivities were 0.779, 0.597, 0.610, and 0.519, and specificities were 0.721, 0.869, 0.803, and 0.984 ( P < 0.05). NLR, PLR, and LMR are associated with disease severity in APIP, and elevated NLR was independently associated with severe APIP. For clinical application, we propose a two-step strategy: NLR alone may serve as a candidate adjunctive screening tool for early identification of severe APIP in emergency settings (higher sensitivity), whereas the combined model (NLR + PLR+LMR) may be reserved for exploratory stratification, given its superior AUC and high specificity. The combined model’s lower sensitivity should be noted when used for screening purposes.