[Study of a nomogram model of gadoxetate disodium-enhanced magnetic resonance imaging for the preoperative diagnosis of proliferative hepatocellular carcinoma and its value].

Objective: To develop and explore the clinical value of a nomogram model for the preoperative diagnosis of proliferative hepatocellular carcinoma (HCC) based on gadoxetate disodium (Gd-EOB-DTPA) enhanced magnetic resonance imaging (MRI). Methods: The preoperative Gd-EOB-DTPA-enhanced MRI data and clinical pathological data of patients with pathologically confirmed proliferative (178 cases) and non-proliferative type HCC (378 cases) from September 2017 to November 2022 were retrospectively collected. The MRI features and clinicopathological features of proliferative and non-proliferative type HCC were evaluated. Multivariate logistic regression analysis was used to determine the independent predictive factors of proliferative-type HCC. The nomogram prediction model was constructed using R software. The receiver operating characteristic curve (ROC) was used to evaluate its diagnostic efficacy. The calibration curve and decision curve analysis (DCA) were drawn to evaluate the calibration performance and clinical application value of the nomogram model. The optimal threshold for distinguishing high-risk from low-risk was determined using the Youden index. The survival prognosis of proliferative and non-proliferative type HCC was analyzed and compared using the Kaplan-Meier survival curve and the log-rank test. The measurement data were analyzed using the independent sample t-test or the Mann-Whitney U test. The count data were compared using the χ² test. Results: There were statistically significant differences in alpha-fetoprotein (AFP) levels (χ²=17.244, P<0.001), tumor morphology (χ²=13.669, P<0.001), intratumoral fatty degeneration (χ²=10.495, P=0.001), abnormal enhancement of peritumoral abnormalities during arterial phase (χ²=37.662, P<0.001), tumor capsule (χ²=23.961, P<0.001), intratumoral necrosis (χ²=77.184,P<0.001), intratumoral hemorrhage (χ²=4.892,P=0.027), peritumoral hypointense in hepatobiliary phase (χ²=47.675,P<0.001), rim arterial phase hyperenhancement (χ²=115.976,P<0.001), intratumoral artery (χ²=15.528,P<0.001) and intravenous tumor thrombus (χ²=10.532,P=0.001) between proliferative and non-proliferative type HCC groups. Multivariate logistic regression analysis showed that AFP>200 μg/L (OR=1.561, P=0.044), no intratumoral fatty degeneration (OR=1.947, P=0.033), intratumoral necrosis (OR=2.084, P=0.003), peritumoral hypointensity in the hepatobiliary phase (OR=2.314, P=0.001), and annular hyperenhancement in the arterial phase (OR=5.557, P<0.001) were independent predictors for preoperative diagnosis of proliferative-type HCC. A nomogram model for preoperative prediction of proliferative type HCC was constructed based on the independent predictors. The area under the ROC curve model for predicting proliferative-type HCC was 0.772 (95%CI: 0.735-0.807), with a sensitivity of 69.1% and a specificity of 75.4%. The calibration curve and DCA curve showed superior calibration performance and clinical applicability of the nomogram model. The Kaplan-Meier curve showed that the recurrence free survival rate after liver resection was significantly lower in patients with proliferative-type HCC than that of non-proliferative-type HCC (P<0.001), and the high-risk group was significantly lower than the low-risk group (P<0.001). Conclusions: The construction of a nomogram model based on Gd-EOB-DTPA-enhanced MRI features combined with AFP >200μg/L can accurately diagnose proliferative-type HCC and predict its preoperative prognosis.