Dual-energy CT in early acute pancreatitis: improved detection using iodine quantification

• DOI: 10.1007/s00330-018-5844-x
• Publication Date: 2018-11-28T09:31:45Z
• Authors: Simon S. Martin; Franziska Trapp; Julian L. Wichmann; Moritz H. Albrecht; Lukas Lenga; James Durden; Christian Booz; Thomas J. Vogl; Tommaso D’Angelo
• Keywords: Male; Contrast Media; Middle Aged; 3. Good health; Diagnosis, Differential; 03 medical and health sciences; Multidetector computed tomography, Pancreas, Pancreatitis, Iodine, Diagnostic imaging, Multidetector CT, Atlanta Classification, Cell Carcinoma, Severity index, Phase, Differentiation, Adenocarcinoma, Optimization, Accuracy, Contrast; 0302 clinical medicine; Pancreatitis; ROC Curve; Acute Disease; Multidetector Computed Tomography; Humans; Female; Pancreas; Iodine; Retrospective Studies

To evaluate the diagnostic performance of a dual-energy computed tomography (DECT)-based technique using iodine quantification and fat fraction analysis for the diagnosis of early acute pancreatitis METHODS: In this retrospective study, 45 patients (35 men and 10 women; mean age, 54.9 ± 14.0 years) with early acute pancreatitis were included. Serum lipase levels and follow-up examinations served as the reference standard. A matched control group (n = 45) was assembled for evaluation of material decomposition values of normal pancreatic parenchyma. Three blinded radiologists independently interpreted all cases on conventional grayscale DECT series. In addition, readers re-evaluated all cases by manually performing region-of-interest (ROI) measurements on pancreatic-phase DECT material density images of the head, body, and tail of each patient's pancreas. Receiver operating characteristic (ROC) curve analysis was performed to estimate the optimal threshold for discriminating between inflammatory and normal pancreas parenchyma.DECT-based iodine density values showed significant differences between inflammatory (1.8 ± 0.3 mg/mL) and normal pancreatic parenchyma (2.7 ± 0.7 mg/mL) (p ≤ 0.01). Fat fraction measurements showed no significant differences (p = 0.08). The optimal iodine density threshold for the diagnosis of acute pancreatitis was 2.1 mg/mL with a sensitivity of 96% and specificity of 77%. Iodine quantification revealed an area under the curve (AUC) of 0.86, significantly higher compared to standard image evaluation of the radiologists (AUC, 0.80; sensitivity, 78%; specificity, 82%) (p < 0.01).DECT using iodine quantification allows for diagnosis of early acute pancreatitis with higher sensitivity compared to standard image evaluation.• Iodine density values showed significant differences between inflammatory and normal pancreatic parenchyma. • DECT using iodine quantification allows for diagnosis of early acute pancreatitis. • An iodine density of ≤ 2.1 mg/mL optimizes the diagnosis of acute pancreatitis.