Clin Lab. 2026 Feb 1;72(2). doi: 10.7754/Clin.Lab.2025.250324.
ABSTRACT
BACKGROUND: This study aimed to investigate the clinical features, coagulation, and risk factors of deep vein thrombosis (DVT) in patients with pelvic tumor and to construct a prediction model for postoperative DVT events.
METHODS: Clinical data of 161 patients with pelvic tumors (preoperative DVT group n = 22, non-DVT group n = 139; postoperative DVT group n = 35, NDVT group n = 125; and one case of postoperative pulmonary thrombosis was excluded) were retrospectively analyzed. Age, BMI, disease type, FIGO stage, and coagulation parameters (prothrombin time, PT; activated partial thromboplastin time, APTT; fibrinogen, FIB; D-dimer, D-D; plasminogen activator inhibitor-1, PAI-1) were compared. The key variables were screened using principal component analysis. The prediction model for postoperative DVT was built through logistic regression, and its efficacy was tested using a ROC curve.
RESULTS: PT, D-D, and PAI-1 were significantly higher in the preoperative DVT group than in the non-DVT group (p < 0.001), and APTT was significantly shorter (p = 0.002). The postoperative DVT group was characterized by advanced age (p = 0.032), a higher proportion of ovarian and endometrial cancers, a greater percentage of advanced FIGO stages (p = 0.002), longer postoperative bedtime of more than 72 hours (p = 0.028), and higher levels of PT, FIB, D-D, and PAI-1 (p < 0.001). Principal component analysis showed age and D-D as the main contributing factors. The logistic regression model showed that age (OR = 1.02, p = 0.05), elevated D-D (OR = 1.02, p = 0.001), FIGO stages III and IV (OR = 3.60, p = 0.048), absence of thrombolytic prophylaxis in the postoperative period (OR = 2.85, p = 0.049), and the presence of adjuvant therapy in the postoperative period (OR = 1.02, p = 0.038) were independent risk factors for postoperative DVT, and the AUC of the model reached 0.865 (p < 0.001).
CONCLUSIONS: Age, preoperative DVT, D-D level, and tumor stage are independent predictors of postoperative DVT in pelvic tumors. The constructed prediction model has high clinical value.
PMID:41670501 | DOI:10.7754/Clin.Lab.2025.250324