摘要Background:The superiority of anatomical resection (AR) vs. non-anatomical resection (NAR) in the surgi-cal management of hepatocellular carcinoma (HCC) is debated. ARs are well-defined procedures,whereas the lack of NAR standardization results in heterogeneous outcomes. This study aimed to introduce the SegSubTe classification for NAR detailing the appropriateness of the level of surgical section of the Glis-sonean pedicles feeding the tumor.Methods:A single-center retrospective analysis of pre-and postoperative imaging of consecutive patients treated with NAR for single HCC between 2012 and 2020 was conducted. The quality of surgery was assessed classifying the type of vascular supply and the level of surgical section (segmental,subsegmental or terminal next to the tumor) of vascular pedicles feeding the HCCs;then,the population was divided in "SegSubTe-IN" or "SegSubTe-OUT" groups,and the tumor recurrence and survival were analyzed.Results:Ninety-seven patients who underwent NAR were included;76％ were SegSubTe-IN and 24％ were SegSubTe-OUT. Total disease recurrence,local recurrence and cut-edge recurrence in the SegSubTe-IN vs. SegSubTe-OUT groups were 50％ vs. 83％ ( P=0.006),20％ vs. 52％ ( P=0.003) and 16％ vs. 39％ ( P=0.020),respectively. SegSubTe-OUT odds ratio for local recurrence was 4.1 at univariate regression analysis. One-,three-,and five-year disease-free survival rates in the SegSubTe-IN vs. SegSubTe-OUT groups were 81％,58％ and 35％ vs. 46％,21％ and 11％,respectively ( P<0.001). Conclusions:The SegSubTe classification is a useful tool to stratify and standardize NAR for HCC,aiming at improving long-term oncological outcomes and reducing the heterogeneity of quality of NAR for HCC.