Risk factors of preoperative deep venous thrombosis in young and middle-aged patients with hip fracture and construction of a Nomogram predictive model

Objective To analyze the risk factors of preoperative deep venous thrombosis (DVT) in young and middle-aged patients with hip fracture, and to construct a Nomogram predictive model.

Methods A total of 408 young and middle-aged patients firstly diagnosed as hip fracture were selected, and they were divided into DVT group with 60 cases and no DVT group with 348 cases according to the result of preoperative dual functional color Doppler ultrasonography (DUS). Clinical materials and blood biochemical indicators were compared between two groups; the multivariate Logistic regression analysis was used to screen risk factors, and a Nomogram model was established.

Results Single factor analysis showed that when compared to the no DVT group, the age was older, the low energy injury increased, the time from injury to DUS and the total hospital stay prolonged, the ratios of patients with total serum protein < 60 g/L, albumin < 35 g/L, lactate dehydrogenase (LDH) ≥250 U/L, the red blood cell (RBC) count < 4.0×10¹²/L, hemoglobin (Hb) < 110 g/L, hematocrit (HCT) ≥50%, platelet count ≥300.0×10⁹/L and D-dimer ≥0.5 mg/L in the DVT group were higher, and all these differences were statistically significant (P < 0.05). Multivariate Logistic regression analysis showed that older age, longer time from injury to DUS, LDH ≥250 U/L, D-dimer ≥0.5 mg/L and HCT ≥50% were the independent risk factors for preoperative DVT (P < 0.05). R software was used to establish the Nomogram model, with a total score of 220 points. The area under thecurve (AUC) of the Nomogram for predicting DVT by receiver operating characteristic (ROC) curve was 0.897, which was significantly larger than that of each single risk factor (P < 0.05). The Hosmer-Lemeshow test showed that the goodness of fit of the Nomogram was good (χ²=1.625, P=0.342). The correction curve showed that the predicted probability of DVT by the Nomogram was consistent with the actual incidence of DVT.

Conclusion Young and middle-aged patients with hip fracture have a certain incidence rate of preoperative DVT, and older age, longer time from injury to DUS, LDH ≥250 U/L, D-dimer ≥0.5 mg/L and HCT ≥50% are the independent risk factors affecting the occurrence of preoperative DVT. The established Nomogram model shows a good application potential for guiding preoperative screen of patients with high-risk DVT.