Spectral CT Imaging in the Differential Diagnosis of Small Bowel Adenocarcinoma From Primary Small Intestinal Lymphoma.

Purpose: To investigate the value of dual-energy spectral computed tomography (CT) imaging in the differential diagnosis of small bowel adenocarcinoma (SBA) from primary small intestinal lymphoma (PSIL).
Materials and Methods: We retrospectively analyzed the images of 27 SBA cases and 15 PSIL cases. These patients underwent spectral CT imaging in the arterial phase (AP) and venous phase (VP). CT attenuation values of tumors at different energy levels were measured to generate spectral attenuation curve and to calculate curve slope (λ _HU ). Iodine concentration (IC) in tumors at AP and VP were measured and normalized to that of aorta as normalized iodine concentration (NIC). Independent samples t test was used to analyze the spectral CT parameters; Receiver operating characteristic curves were generated to evaluate the diagnostic efficacy of each parameter.
Results: There were significant differences between SBA and PSIL in IC (2.09 ± 0.71 vs 1.33 ± 0.15 mg/ml), NIC (0.20 ± 0.06 vs 0.13 ± 0.02) and slope (λ _HU ) (2.78 ± 1.06 vs 1.86 ± 0.30) in AP and (1.86 ± 0.68 vs 1.37 ± 0.18 mg/ml for IC; 0.47 ± 0.13 vs 0.33 ± 0.02 for NIC and 2.00 ± 0.56 vs 1.50 ± 0.26 for λ _HU ) in VP (all p < 0.05). For the CT value measurement, there were significant differences between SBA and PSIL in the 40-60keV energy range (p < 0.05), but not in the 70-140keV range (p > 0.05). Using 1.38 mg/ml as a threshold value for iodine concentration at AP, one could obtain the area-under-curve of 0.93 for receiver operating characteristic study and sensitivity of 94% and specificity of 85% for differentiating SBA from PSIL. The sensitivity and specificity values were significantly higher than the respective values of 62% and 60% with the conventional CT numbers at 70keV.
Conclusion: Quantitative parameters obtained in spectral CT, especially iodine concentration in AP, provide high accuracy for differentiating SBA from PSIL.
(Copyright © 2019. Published by Elsevier Inc.)