Does a combined CT and MRI protocol enhance the diagnostic efficacy of LI-RADS in the categorization of hepatic observations? A prospective comparative study

• Published in: European radiology 2018-06, Vol.28 (6), p.2592-2603
• Main Authors: Basha, Mohammad Abd Alkhalik, AlAzzazy, Mohamad Zakarya, Ahmed, Ayman F., Yousef, Hala Y., Shehata, Samar Mohamad, El Sammak, Dena Abd El Aziz, Fathy, Talaat, Obaya, Ahmed Ali, Abdelbary, Eman H.
• Format: Article
• Language: English
• Subjects: Accuracy; Adult; Aged; Carcinoma, Hepatocellular - diagnosis; Classification; Comparative studies; Computed tomography; Diagnostic Radiology; Diagnostic systems; Female; Hepatobiliary-Pancreas; Hepatocellular carcinoma; Histopathology; Humans; Imaging; Internal Medicine; Interventional Radiology; Liver; Liver cancer; Liver Cirrhosis - diagnosis; Liver Neoplasms - diagnosis; Magnetic resonance imaging; Magnetic Resonance Imaging - methods; Male; Medical diagnosis; Medicine; Medicine & Public Health; Middle Aged; Multimodal Imaging - methods; Neuroradiology; Patients; Prospective Studies; Radiology; Sensitivity; Sensitivity and Specificity; Tomography, X-Ray Computed - methods; Ultrasonic imaging; Ultrasonic methods; Ultrasonic testing; Ultrasonography; Ultrasound
• ISSN: 0938-7994; 1432-1084
• DOI: 10.1007/s00330-017-5232-y

Objective To compare diagnostic performance and agreement between CT, MRI and combined CT/MRI in reference to LI-RADS classification system to categorize hepatic observations detected in hepatic patients during screening ultrasound. Methods 240 patients with 296 liver observations detected during ultrasound surveillance underwent hepatic CT and MRI examinations, histopathology, and clinical and radiological follow-up. Using LI-RADS v2014, six radiologists evaluated the observations independently and assigned a LI-RADS category to each observation using CT, MRI and combined CT/MRI. Results Combined CT and MRI in LI-RADS yielded better accuracy (91.29 %), sensitivity (90.71 %) and specificity (92.31 %) for hepatocellular carcinoma (HCC) diagnosis than using MRI or CT alone; accuracy, sensitivity and specificity decreased to 85.37 %, 86.34 %, and 83.65 %, respectively, for MRI and 67.6 %, 54.10 % and 91.35 %, respectively, for CT. The intraclass agreement of the LI-RADS scores between CT, MRI and combined CT/MRI was excellent (κ=0.9624 (95 % CI: 0.9318–0.9806)). Conclusion CT and MRI are complementary to each other. Combined CT/MRI enabled a more precise determination of LI-RADS category of hepatic observations; however, due to the expense and minor increase in accuracy, the combined methodology should only be utilized in cases of suspected HCC. Key Points • Hepatic observation may be categorized differently depending on the imaging modality used. • We compared LI-RADS categorization between CT, MRI and combined CT/MRI. • MRI produces higher accuracy and sensitivity, while CT produces higher specificity. • Combining CT and MRI improves LIRADS categorization reports. • Considering additional cost, combined methodology could be restricted to challenging cases.