The utility of diffusion-weighted imaging in improving the sensitivity of LI-RADS classification of small hepatic observations suspected of malignancy

Autor: Khaled Mohamed Altaher, Talaat Fathy, Ahmed Mohamed El-Maghraby, Taghreed M. Azmy, Eman H. Abdelbary, Mai E.M. Khamis, Ahmed Mohamed Alsowey, Heba F. Taha, Mohammad Abd Alkhalik Basha, Hanan A. Hassan, Heba A. Mohamed, Rania Refaat, Heba F. Tantawy, Ahmed A El Sammak, Amr Shaaban Hanafy, Mohamed Radwan, Mohamed I. Amin, Mohamad Zakarya AlAzzazy, Ahmad Abdullah Alnaggar, Mostafa Mohamad Assy, Faten Fawzy Mohammad, Rania M. Almolla, Ahmed M. Alaa Eldin, Sameh Saber
Rok vydání: 2019
Předmět:
Zdroj: Abdominal Radiology. 44:1773-1784
ISSN: 2366-0058
2366-004X
DOI: 10.1007/s00261-018-01887-z
Popis: We investigated the added value of diffusion-weighted imaging (DWI)/apparent diffusion coefficient (ADC) in the categorization of small hepatic observation (≤ 20 mm) detected in patients with chronic liver disease in reference to LI-RADS (liver imaging reporting and data system) classification system. We prospectively evaluated 165 patients with chronic liver disease with small hepatic observations (≤ 20 mm) which were previously categorized as LI-RADS grade 3–5 on dynamic contrast-enhanced CT (DCE-CT). All patients were submitted to a functional MRI including DCE and DWI. Using LI-RADS v2017, two radiologists independently evaluated the observations and assigned a LI-RADS category to each observation using DCE-MRI alone and combined DCE-MRI and DWI/ADC. In the combined technique, the radiologists assigned a LI-RADS category based on a modified LI-RADS criteria in which restricted diffusion on DWI was considered a major feature of HCC. We evaluated the inter-reader agreement with Kappa statistics and compared the diagnostic performance of the LI-RADS with two imaging techniques by Fisher’s exact test using histopathology as the reference standard. Combined technique in LI-RADS yielded better sensitivities (reader 1, 97% [65/67]; reader 2, 95.5% [64/67]) for HCC diagnosis than DCE-MRI alone (reader 1, 80.6% [54/67], p = 0.005; reader 2, 83.6% [56/67], p = 0.04). The specificities were insignificantly lower in combined technique (reader 1, 88.4% [107/121]; reader 2, 77.7% [94/121]) than in DCE-MRI alone (reader 1, 90.9% [110/121], p = 0.67; reader 2, 79.3% [96/121], p = 0.88). The inter-reader agreement of the LI-RADS scores between combined technique and DCE-MRI was good (κ = 0.765). The use of DWI/ADC as an additional major criterion, improved the sensitivity of LI-RADS in the diagnosis of HCC while keeping high specificity.
Databáze: OpenAIRE
Externí odkaz: