A Predictive Model for Tumor Invasion of the Inferior Vena Cava Wall Using Multimodal Imaging in Patients with Renal Cell Carcinoma and Inferior Vena Cava Tumor Thrombus

Purpose. Developed a preoperative prediction model based on multimodality imaging to evaluate the probability of inferior vena cava (IVC) vascular wall invasion due to tumor infiltration. Materials and Methods. We retrospectively analyzed the clinical data of 110 patients with renal cell carcinoma (RCC) with level I-IV tumor thrombus who underwent radical nephrectomy and IVC thrombectomy between January 2014 and April 2019. The patients were categorized into two groups: 86 patients were used to establish the imaging model, and the data validation was conducted in 24 patients. We measured the imaging parameters and used logistic regression to evaluate the uni- and multivariable associations of the clinical and radiographic features of IVC resection and established an image prediction model to assess the probability of IVC vascular wall invasion. Results. In all of the patients, 46.5% (40/86) had IVC vascular wall invasion. The residual IVC blood flow (OR 0.170 [0.047-0.611]; P = 0.007 ), maximum coronal IVC diameter in mm (OR 1.203 [1.065-1.360]; P = 0.003 ), and presence of bland thrombus (OR 3.216 [0.870-11.887]; P = 0.080 ) were independent risk factors of IVC vascular wall invasion. We predicted vascular wall invasion if the probability was >42% as calculated by: Ln Pre / 1 − pre = 0.185 × maximum cornal IVC diameter + 1.168 × bland thrombus – 1.770 × residual IVC blood flow – 5.857 . To predict IVC vascular wall invasion, a rate of 76/86 (88.4%) was consistent with the actual treatment, and in the validation patients, 21/26 (80.8%) was consistent with the actual treatment. Conclusions. Our model of multimodal imaging associated with IVC vascular wall invasion may be used for preoperative evaluation and prediction of the probability of partial or segmental IVC resection.