A new nomogram for predicting the postoperative venous thromboembolism in patients with spinal metastasis undergoing operation: A retrospective study.

Abstract

Background: Venous thromboembolism (VTE), including deep vein thrombosis (DVT) and pulmonary embolism (PE), represents a severe clinical illness. Cancer patients exhibit a VTE risk substantially increased, ranging from four to seven times that of non-cancer patients. Our study aimed to ascertain the occurrence and risk determinants of VTE in spinal metastasis patients undergoing operation, as well as to formulate a predictive model.

Methods: Examining 166 cancer patients treated surgically for spinal metastases and 100 non-cancer patients treated surgically for spinal diseases between 2022 and 2025 at our institution in this study. Data on patient characteristics and clinical parameters were collected for comparison between VTE and non-VTE cases in cancer populations to identify risk factors. Subsequently, an independent risk factor-driven nomogram was constructed, and its predictive performance was assessed.

Results: The authors identified 166 cancer patients, 43 (27.20 %) were diagnosed with a DVT postoperatively, 2 (1.20 %) were diagnosed with PE. Among the 100 non-cancer patients, only 2 (2 %) were diagnosed with DVT postoperatively. In the multivariate logistic regression model, nine independent risk factors for VTE in cancer patients were identified. These included preoperative albumin level (OR = 0.889, 95 % CI 0.823-0.962, P = 0.003) and one-day postoperatively (OR = 1.261, 95 % CI 1.062-1.498, P = 0.008), the preoperative D-dimer level (OR = 1, 95 % CI 1.00-1.00, P = 0.032) and three-day postoperatively (OR = 1, 95 % CI 1.00-1.001, P = 0.005), the PLT level one-day postoperatively (OR = 1.016, 95 % CI 1.005-1.027, P = 0.004) and three-day postoperatively (OR = 0.989, 95 % CI 0.980-0.999, P = 0.023), the preoperative hemoglobin level (OR = 1.050, 95 % CI 1.001-1.102, P = 0.044) and one-day postoperatively (OR = 0.901, 95 % CI 0.833-0.974, P = 0.009), the fibrinogen level three-day postoperatively (OR = 2.011, 95 % CI 1.125-3.595, P = 0.018). The constructed Nomogram model exhibited high accuracy in predicting VTE in spinal metastasis patients undergoing operation, with external validation confirming its predictive performance. Based on the receiver operating characteristic (ROC) curves, the nomogram showed the areas under the curves (AUCs) of the model were 0.819.

Conclusion: Following oncological spine surgery, this study confirmed the relevance of nine biomarkers in predicting VTE in cancer patients. The model devised by our team exhibits significant potential to support clinicians in clinical decision-making and patient risk assessment. Validating these biomarkers is crucial for future research.