RESUMEN
Objective: To investigate the correlation between dynamic enhanced energy spectral CT parameters and Ki-67 high expression in hepatocellular carcinoma (HCC). Methods: A cross-sectional study. This retrospective case-control study analyzed the clinical data of 101 patients with pathologically confirmed HCC in Xiamen Hospital of Traditional Chinese Medicine and Zhongshan Hospital of Xiamen University from December 2017 to March 2023. These 101 patients included 84 males and 17 females, and the age[M(Q1, Q3)] was 59.0(49.0,66.0)years. These patients were divided into two groups according to the immunohistochemical Ki-67 expression levels in tumor tissues: the high expression group (Ki-67%>20%, n=59) and the low expression group (Ki-67%≤20%, n=42).CT values on 70 keV and 140 keV monochromatic energy images (HU70 keV-a, HU140 keV-a, HU70 keV-p, HU140 keV-p) and water density (Dwater-a, Dwater-p) were measured in arterial phase and portal vein phase, and the difference of HU70 keV, HU140 keV, Dwater values between portal vein and arterial phase (ΔHU70 keV, ΔHU140 keV, ΔDwater), as well as ratio of HU70 keV, HU140 keV, Dwatervalues between portal vein and arterial phase (HU70 keVratio, HU140 keVratio, Dwaterratio) were calculated. Spearman correlation analysis was used to analyze the correlation between the CT spectral parameters and Ki-67%. Multivariate logistic regression model was used to determine the factors associated with high expression of Ki-67. The receiver operating characteristics (ROC) curves were used to indicate the efficacy of dynamic enhanced spectral CT in evaluating Ki-67 high expression in HCC. Results: The high Ki-67 expression group revealed higher alpha fetal protein levels, larger tumor diameter and more irregular tumor shape compared with the low Ki-67 expression group,and the differences were statistically significant (all P<0.05). Spearman correlation analysis showed that the HU140 keV-p, Dwater-p, HU70 keV ratio, HU140 keV ratio, Dwater ratio, ΔHU70 keV, ΔHU140 keV, ΔDwater were positively correlated with Ki-67 positivity rate (r:0.31-0.50, all P<0.05). The spectral CT parameters (HU70 keV-p, HU140 keV-p, Dwater-p, HU70 keV ratio, HU140 keV ratio, Dwater ratio, ΔHU70 keV, ΔHU140 keV, ΔDwater) in high Ki-67 expression group were significantly higher than those in low Ki-67 expression group (all P<0.05). Multivariate logistic regression model shows that Dwater-p(OR=1.16, 95%CI: 1.05-1.29, P=0.005), ΔHU140 keV(OR=1.39, 95%CI: 1.20-1.62, P<0.001) and irregular tumor morphology (OR=5.25, 95%CI: 1.61-17.12, P=0.006) were correlative factors for high Ki-67 expression. The HU140 keV ratio and ΔHU140 keV alone evaluated the highest AUC of high Ki-67 high expression in HCC, which were 0.82 (95%CI: 0.74-0.90), the sensitivity were 61.0%, and the specificity were 88.1% and 85.7%. The combined analysis of Dwater-p, ΔHU140 keV and irregular tumor morphology had an increased AUC of 0.88 (95%CI: 0.81-0.95) in assessment high Ki-67 expression, with the sensitivity of 84.7% and the specificity of 78.6%. Conclusions: Dynamic enhanced spectral CT parameters were positively correlated with the Ki-67 expression in HCC. Spectral CT provides a non-invasive method to evaluate the proliferation status of HCC cells, and the efficiency could be improved by multi-parameter analysis combining spectral CT parameters and morphologic features.