Letter to the editor
暂无分享,去创建一个
We have read with great interest the recent article by Fowler et al. [1], which discussed new diagnostic criteria of LI-RADS v2017 focusing on the LI-RADS category M (LR-M). As the authors explained, using LR-M category may help obtain near-perfect specificity of LR-5 category for hepatocellular carcinoma (HCC), especially when deceased donor liver transplantation (DDLT) is a primary treatment option for HCC [2]. Currently, the LR-M category may indicate non-HCC malignancy, frequent HCC, or occasional benign lesion [1, 3]. In our experiences, we have found that the frequency of HCC among observations in LR-M category to be considerably high (approximately 3/4, unpublished data), while none of them were benign. Although further research is needed to validate our data, most of the LRM observations undergo surgical resection or loco-regional treatment without biopsy at our institution due to the high probability of HCC. In countries where DDLT is not a major initial treatment approach for HCC, it is more important to confidently differentiate malignant categories (LR-5 or LR-M) from other categories, such as LR-4. Therefore, we sometimes group the two categories together and report them as ‘‘LR-5M,’’ which represents definite malignancy with the possibility of non-HCC malignancy. Of course, definite HCC is categorized as LR-5. LR-M is currently an early branch point before evaluation of major features for LR-3, LR-4, or LR-5 categories. However, here we suggest dividing LR-M into LR-4M (probably malignant but not specific for HCC) and LR-5M (definitely malignant but not specific for HCC), and instead remove it from the early diagnostic step. This way, treatment strategy for the categories does not change, and important additional information can still be effectively delivered. In the United States, both LR-5M and LR-4M are regarded as LR-M and managed as before; if management strategy for LR-5M does not change according to actual diagnosis (HCC or non-HCC malignancy), as in our practice, the definitely malignant nature of the observation can be explained accurately by radiologists and understood more easily by clinicians. Furthermore, clinicians can also clearly and intuitively understand the certainty level of malignant diagnoses: probably malignant [LR-4 (probable HCC) or LR-4M (probable malignancy, not necessarily HCC)], or definitely malignant [LR-5 (definite HCC) or LR-5M (definite malignancy, not necessarily HCC)]. Last but not least, the significant regional differences in HCC diagnosis and treatment can also be considered in the LI-RADS. Users from countries other than the U.S. tend to manage a great number of HCC patients, and are willing to implement this system in their practices. Therefore, we propose to incorporate the current LRM category into LR-4 and LR-5 using the suffix ‘‘M,’’ as LR-4M or LR-M categories, to indicate the probability of non-HCC malignancy in each category.