Early non-contrast CT morphology at emergency admission in acute pancreatitis: real-world associations with clinical course.

purposeTo evaluate the clinical associations of admission non-contrast CT morphology in acute pancreatitis within a real-world emergency workflow. MATERIALS AND

methodsThis retrospective observational cohort study included 264 consecutive adult patients admitted with acute pancreatitis to two surgical centers between 2019 and 2024. Patients were categorized according to the first imaging modality obtained at admission into an ultrasound-first (US-first) or computed tomography-first (CT-first) pathway. Baseline characteristics and in-hospital outcomes were compared between pathways. In the CT-first subgroup, all examinations were performed without intravenous contrast, and morphologic severity was assessed using the Balthazar classification. Associations between CT morphology and clinical outcomes were evaluated using univariable analyses.

resultsOf the 264 patients, 143 (54.2%) were managed within a US-first pathway and 121 (45.8%) underwent CT as the initial imaging modality. Baseline demographic and etiologic characteristics were comparable between pathways. Patients in the CT-first pathway demonstrated numerically higher rates of adverse clinical outcomes at admission, including a longer length of hospital stay (median 8 vs. 6 days; p = 0.01) and numerically higher rates of severe acute pancreatitis and in-hospital mortality. Within the CT-first cohort, non-contrast CT morphology demonstrated heterogeneous inflammatory severity. Higher Balthazar grades were associated with stepwise numerical increases in rates of severe disease, complications, and length of hospital stay. When dichotomized, advanced morphologic severity (Balthazar grades D-E) showed higher odds of adverse outcomes compared with grades A-C, although these associations did not reach statistical significance.

conclusionIn routine emergency practice, selection of ultrasound-first or CT-first imaging pathways appears largely driven by triage and organizational factors rather than predefined imaging strategies. In patients undergoing non-contrast CT at admission, higher Balthazar grades demonstrated consistent numerical gradients toward more severe clinical courses; however, these associations did not reach statistical significance. Early non-contrast CT morphology should therefore be interpreted as contextual inflammatory assessment rather than a standalone prognostic tool.