1784. Modified Threshold Growth Using Growth Rate in LI-RADS Improves the Diagnosis of Hepatocellular Carcinoma
Authors* Denotes Presenting Author
Boryeong Jeong *;
Asan Medical Center
Sang Hyun Choi;
Asan Medical Center
Objective:
The 2018 version of the Liver Reporting and Data System (LI-RADS) simplified the definition of threshold growth (TG) to =50% size increase within a mass in =6 months. Although this achieved simplicity and concordance with definitions by the Organ Procurement and Transplantation Network, the growth rate is not completely considered, especially in the case of insufficient follow-up. We evaluated and compared the diagnostic performance of LI-RADS with modified TG (applying growth rate) for hepatocellular carcinomas (HCCs).
Materials and Methods:
Patients who underwent gadoxetic acid-enhanced MRI and subsequent resection, biopsy, or transplantation for focal solid lesions (3.0 cm) within 1 month between January 2016 and December 2020 were retrospectively evaluated. Three readers measured the size of each lesion at both prior CT or MRI and index MRI, and the growth rate was defined as the percentage change in lesion size per month. In the modified TG, TG was considered when the following growth rate criteria were present: (a) = 10%/month (TG-10%), (b) = 20%/month (TG-20%), or (c) = 30%/month (TG-30%). The sensitivity and specificity of LI-RADS category 5 for diagnosing HCC were calculated and compared between LI-RADS v2018 and LI-RADS with modified TG (TG-10%, TG-20%, or TG-30%) using generalized estimation equations.
Results:
In a total of 508 lesions from 370 patients, 69 lesions (13.6%) showed LI-RADS v2018 TG. The frequency of modified TG-10%, TG-20%, and TG-30% was 37.0%, 17.5%, and 7.1%, respectively. The use of TG-10% as a major feature significantly increased sensitivity (79.0% vs. 72.5%, p < 0.001) with the same specificity (96.6% vs. 96.6%, p > 0.999). Compared with LI-RADS v2018, LI-RADS with TG-20% had similar sensitivity (73.1% vs. 72.5%, p = 0.563) and that with TG-30% had lower sensitivity (70.1% vs. 72.5%, p = 0.020).
Conclusion:
The frequency of modified TG-10% was higher than that of TG in LI-RADS v2018. Using a modified TG of10%/month as a major feature can be clinically useful by improving the sensitivity of LI-RADS for diagnosing HCC = 3.0 cm.
