Predictive value of CT-based arterial enhancement fraction in the preoperative lymphovascular invasion of pancreatic ductal adenocarcinoma

To investigate whether arterial enhancement fraction (AEF) from computed tomography (CT) can be used to assess lymphovascular invasion (LVI) status in pancreatic ductal adenocarcinoma (PDAC). A total of 92 patients with histopathologically confirmed PDAC who underwent CT were included and assigned to this retrospective study. The CT attenuation values of CT-derived enhancement parameters between the groups positive and negative for LVI were evaluated. The optimal combination of enhancement parameters for predicting LVI status was investigated by employing binary logistic regression models and evaluated via receiver operating characteristic (ROC) curves. Furthermore, the optimal model was developed into a visual nomogram and evaluated for clinical applicability by using decision curve analysis (DCA). A total of 22 patients positive for LVI and 70 patients negative for LVI were included. The CT values of the arterial phase (Ap), absolute contrast enhancement of the Ap (△Ap), and AEF were significantly higher in the LVI-positive group than the LVI-negative group (all P < 0.05). LVI (+) had the highest predictive efficiency for AEF, with an area under the curve (AUC), accuracy, sensitivity, and specificity of 0.798, 0.717, 0.714, and 0.727, respectively. The nomogram incorporating these three predictors exhibited the best evaluation efficacy with an AUC of 0.846, accuracy of 0.826, sensitivity of 0.871, and specificity of 0.682. Subsequently, the DCA of the nomogram indicated optimal clinical value for predicting LVI status. AEF derived from CT and nomogram have the potential for the preoperative evaluation of LVI status in patients with PDAC.