Construction and evaluation of a risk prediction model for postoperative deep vein thrombosis in gynecological patients.

BACKGROUND: To investigate the predictive role of a risk assessment model constructed using Caprini score combined with D-dimer in gynecological postoperative patients for the occurrence of deep vein thrombosis. METHODS: Patients scheduled for gynecological surgery at our hospital between January 2018 and April 2024 were included. This study included 136 patients, with 35 cases in the DVT group and 101 cases in the non-DVT group. General information, intraoperative parameters, D-dimer levels, Caprini Score, lower extremity Doppler ultrasonography, and intervention methods were collected. Logistic regression analysis and a combined model were employed to analyze the factors influencing the occurrence of DVT. RESULTS: Compared to non-DVT patients, the DVT group had a significantly older age (p = 0.035), higher hypertension prevalence (p = 0.025), and more complex surgeries (p = 0.004). Pre-discharge D-dimer levels and pre-/postoperative Caprini scores were markedly elevated in DVT patients (p < 0.05). Critically, logistic regression identified pre-discharge D-dimer levels (p < 0.001), preoperative Caprini score (p = 0.003), and postoperative Caprini score (p < 0.001) as independent risk factors for DVT. The combined prediction model integrating these factors achieved an AUC of 0.812, demonstrating high discriminative power for postoperative DVT occurrence. CONCLUSION: The predictive value of the DVT prediction model constructed using the Caprini score in combination with D-dimer for the occurrence of DVT is high. The combined predictive model can be further promoted in clinical practice to take appropriate preventive measures to reduce the likelihood of DVT occurrence and to intervene promptly in existing risk factors.