| Objective To investigate the value of multi-parametric MRI combined with clinical risk factors in preoperative prediction of lymphovascular space invasion (LSVI) in rectal cancer. Methods The clinical and imaging data of 31 patients with rectal cancer confirmed by postoperative pathology in Chizhou People's Hospital from April 2022 to August 2023 were selected. All patients underwent conventional MRI, synthetic MRI and IVIM-DWI sequence scanning. According to the postoperative pathological results, the patients were divided into the LVSI positive group (10 cases) and LVSI negative group (21 cases). The clinical data of LVSI positive group and LVSI negative group were analyzed by univariate and multivariate logistic regression to determine the clinical risk factors of LVSI. Synthetic MRI (T1 value, T2 value, PD value) and IVIM-DWI parameters (D value, D* value, f value) were compared between the two groups. The diagnostic efficacy of the prediction model of each quantitative parameter and the prediction model combined with clinical risk factors were assessed using the receiver operating characteristic curve (ROC). Results The T2 value of synthetic MRI, D value and f value of IVIM-DWI were significantly different in LVSI positive and negative group of rectal cancer (P＜0.05). Preoperative CEA（OR=10.818，95%CI:1.391~84.124）and clinical N stage（OR=11.852， 95%CI:1.534~91.552）were independent risk factors for LVSI in rectal cancer (P＜0.05). The AUC of T2 value, D value, f value and the combination of the three values were 0.801, 0.747, 0.766 and 0.807, respectively. The prediction model combined with clinical risk factors had the highest performance (AUC=0.845), with a sensitivity of 78.58% and a specificity of 100%. Conclusion Multi-parametric MRI can effectively preoperatively predict the status of LVSI in rectal cancer. The prediction model combined with clinical risk factors can further improve the prediction efficiency, which is helpful for clinicians to make personalized treatment plans for rectal cancer. |
