PurposeTo investigate the usefulness of dual energy spectral computed tomographic (DESCT) imaging parameters in diagnosing hepatic alveolar echinococcosis (AE) disease by comparing with traditional CT. Materials and methodsIn this prospective study, 35 consecutive patients (22 men, 13 women; mean age, 42.6 years) suffering from hepatic echinococcosis (HE) (HE cases were confirmed by surgery, pathology or multiple imaging methods) underwent abdominal DESCT scan, consisting of arterial phase (AP), portal venous phase (PVP) and delayed phase (DP), in dual-energy (80 kV/140 kV) mode. 15 patients had AE; 20 patients were pathologically confirmed cystic echinococcosis (CE). The CT patterns of the AE lesions were classified into solid, pseudocystic, or mixed type. The size, location, invasion findings were recorded. The serial function images (ie, iodine-based material-decomposition CT images, contrast-to-noise ratio images, monochromatic images, spectral curve images) from DESCT data were processed with a software algorithm designed to calculate optimal CNR value, quantitative iodine concentration (QIC), curve slope for evaluation of HE lesions. Sensitivity and specificity were compared between the qualitative and quantitative studies for DESCT and traditional CT in HE. The two-sample test and ROC curve analyses were performed to compare quantitative parameters between AE and CE. ResultsA total of 18 AE lesions (6 solid, 2 pseudocystic, and 10 mixed) with average 87.6 ± 50.7 mm lesion size were assessed. DECT detected more contrast enhanced lesions (71.6%) than traditional CT (50.9). There was a significant difference in the QICs between various layers of the HAE in each of the three scan phases (P < 0.001). The QICs in the marginal zone were significantly higher than that in solid or cystic layers and liver parenchyma. No significant difference was found among the various CT patterns of hepatic AE. 65 keV is the optimal monochromatic image with corresponding contrast noise radio (CNR) 10.24 ± 2.62. The spectrum curve showed “inversed saddle pattern” for HAE lesion. The differences of slopes between solid, margin zone and normal liver parenchyma of HAE lesion were of statistical significances in the region of 40–60 keV and 60–70 keV (P < 0.05). There were statistical significances in slopes of spectrum curve between AE and CE with different keV regions. The spectrum curve of CT was in gradual falling type. ROC showed the areas under the curves (AUC) for k values to differentiate between AE and CE were 1 in region of 40–60 keV and 0.847 in region of 70–140 keV. ConclusionsDESCT could provide more information by using monochromatic images, iodine-based material decomposition images and the quantitative analysis of IC than traditional CT for the diagnosis of HE.