Liver Imaging Reporting and Data System Version 2018: Impact on Categorization and Hepatocellular Carcinoma Staging

Abstract

The purpose of this study was to assess the concordance in categorization and radiologic T staging using Liver Imaging Reporting and Data System (LI‐RADS, LR) version 2017 (v2017), version 2018 (v2018), and the Organ Procurement and Transplantation Network (OPTN) criteria. All magnetic resonance imaging and computed tomography reports using a standardized LI‐RADS macro between April 2015 and March 2018 were identified retrospectively. The major features (size, arterial phase hyperenhancement, washout, enhancing capsule, or threshold growth) were extracted from the report for each LR‐3, LR‐4, and LR‐5 observation. Each observation was assigned a new category based on LI‐RADS v2017, v2018, and OPTN criteria. Radiologic T stage was calculated based on the size and number of LR‐5 or OPTN class 5 observations. Categories and T stages assigned by each system were compared descriptively. There were 398 patients (66.6% male; mean age, 63.4 years) with 641 observations (median size, 14 mm) who were included. A total of 73/182 (40.1%) observations categorized LR‐4 by LI‐RADS v2017 were up‐categorized to LR‐5 by LI‐RADS v2018 due to changes in the LR‐5 criteria, and 4/196 (2.0%) observations categorized as LR‐5 by LI‐RADS v2017 were down‐categorized to LR‐4 by LI‐RADS v2018 due to changes in the threshold growth definition. The T stage was higher by LI‐RADS v2018 than LI‐RADS v2017 in 49/398 (12.3%) patients. Compared with the OPTN stage, 12/398 (3.0%) patients were upstaged by LI‐RADS v2017 and 60/398 (15.1%) by LI‐RADS v2018. Of 101 patients, 5 (5.0%) patients with T2 stage based on LI‐RADS v2017 and 10/102 (9.8%) patients with T2 stage based on LI‐RADS v2018 did not meet the T2 criteria based on the OPTN criteria. Of the 98 patients with a T2 stage based on OPTN criteria, 2 (2.0%) had a T stage ≥3 based on LI‐RADS v2017 and 6 (6.1%) had a T stage ≥3 based on LI‐RADS v2018.

LI‐RADS: Review and updates

“…4). However, this affects <1% of all LR‐3 and LR‐4 and only 2% of LR‐5 observations 9 …”

Liver Imaging Reporting and Data System (LI-RADS) v2018: Review of the CT/MRI Diagnostic Categories

“…6 The price of high specificity of LR-5 category is its modest sensitivity, ranging between 50% and 80%. [16][17][18] Nonrim APHE and size 10 mm are required for LR-5 categorization, that is, observations <10 mm or lacking nonrim APHE cannot be categorized LR-5, regardless of any other imaging features. 3 The LR-5 category includes observations >20 mm with 1 additional major feature and 10 to 19 mm observations with 2 additional major features.…”

“…4 The new LR-5 criteria result in upcategorization of approximately 40% of LR-4 observations based on v2017. 16 However, several studies had demonstrated improved sensitivity and accuracy of the v2018 LR-5 category as compared with v2017 for the diagnosis of HCC. [16][17][18] For patients with LR-5 observations, multidisciplinary discussion is needed for staging and optimal treatment.…”

Liver imaging reporting and data system and CT/MRI diagnosis of hepatocellular carcinoma

“…Ninety-four percent of observations in this category are HCC and 97% are malignant [17] . Since HCC can be diagnosed based on imaging features alone and treated without the need for pathologic confirmation, observations in this category have specificity close to 100% with resulting modest sensitivity in the range of 50%-80% [21][22][23] . Size ≥ 10 mm and nonrim APHE are absolute requirements to LR-5 categorization [7,15] .…”