Analysis of CT and MRI imaging features of ectopic pancreas.

BACKGROUND: Ectopic pancreas (EP) is a rare congenital anomaly that is often difficult to diagnose preoperatively due to its nonspecific clinical manifestations and imaging overlap with other gastrointestinal subepithelial lesions. Several characteristic imaging findings have been reported; however, the quantitative imaging features of EP remain insufficiently investigated. Therefore, this study aimed to characterize the clinical and imaging manifestations of EP, summarize its computed tomography (CT) features, quantitatively compare CT attenuation values between EP and native pancreas across different imaging phases, and explore phase-specific quantitative differences that may provide additional imaging information for EP characterization. Available magnetic resonance imaging (MRI) findings were also analyzed descriptively for supplementary illustration. METHODS: The clinical and imaging data of 17 patients with surgically confirmed EP were retrospectively reviewed. All patients underwent non-contrast CT, 13 underwent triphasic contrast-enhanced CT, and two underwent non-contrast MRI. The imaging characteristics of benign-appearing lesions representing EP tissue were evaluated, and exploratory qualitative and quantitative comparisons were performed between the EP lesions and the native pancreatic tissue. RESULTS: EP was most frequently located in the duodenum (52.9%, 9/17) and stomach (35.3%, 6/17), typically presenting as solitary ovoid lesions (52.9%, 9/17) with ill-defined margins (64.7%, 11/17) and a long diameter-to-short diameter ratio >1.4 in 64.7% (11/17) of cases. The following three imaging signs were observed in the 13 patients who underwent contrast-enhanced CT: the duct sign (5/13 cases, 38.5%), the umbilical sign (1/13 cases, 7.7%), and the perilesional fat line sign (1/13 cases, 7.7%). On the non-contrast CT and arterial-phase images, EP showed lower attenuation than the native pancreas [non-contrast: 45.18±2.91 vs. 48.59±2.90 Hounsfield units (HU), corrected P=0.002; arterial phase: 76.38±15.12 vs. 95.46±7.34 HU, corrected P=0.002]; however, no statistically significant differences were observed during the venous or delayed phases. In the exploratory analysis, arterial-phase CT attenuation showed the largest attenuation difference and the highest area under the curve (AUC) among the evaluated phases in this cohort. CONCLUSIONS: EP demonstrates characteristic morphological features and phase-specific quantitative CT attenuation patterns. In this small cohort, arterial-phase CT attenuation showed the greatest quantitative difference relative to the native pancreas among the evaluated phases and may provide supportive information for EP identification when interpreted alongside specific CT signs. These results should be interpreted with caution.