髋部骨折围手术期下肢深静脉血栓的相关因素

Factors related to deep vein thrombosis of lower extremity in perioperative period of hip fractures

摘要:

［目的］分析老年髋部骨折围手术期下肢深静脉血栓（deep vein thrombosis, DVT）的危险因素，建立预测模型，指导临床预判 DVT 发生风险。［方法］回顾性分析本院 2015 年 8 月—2021 年 9 月收治的老年髋部骨折共 384 例患者临床资料。按是否发生血栓，分为 DVT 组和非 DVT 组，采用单因素和多因素分析得出下肢 DVT 的危险因素。建立预测模型，对模型效能进行评定。［结果］384 例患者中，围术期共发生下肢 DVT 67 例（17.5%）。单项因素比较，两组年龄、性别构成、骨折类型、骨折侧别、ASA 分级、受伤至入院时间、是否口服抗凝、抗栓药物、合并内科疾病、手术方式、麻醉方式以及入院时 Hb、 HCT、WBC、PLT、Alb、纤维蛋白原、D-D、PT、PO2 的差异均无统计学意义（P<0.05）。与非 DVT 组相比，DVT 组患者术前等待时间更长、合并症数量≥3 种、输血比例及合并骨折占比更高、手术时间更长、术中出血量更多、APTT 更长，差异均有统计学意义 （P<0.05）。多因素二元逻辑回归分析显示：围术期输血 （OR=2.565，P<0.001）、合并症数量≥3 种 （OR=1.880，P= 0.031）、术前等待时间长 （OR=1.110，P<0.001） 是围手术期下肢 DVT 发生的独立危险因素。基于逻辑回归结果所得的下肢 DVT 预测方程准确度为 82.8%，临界值为 0.319，敏感度为 71.6%，特异度为 60.3%；ROC 曲线下面积为 0.705。［结论］ 术前等待时间长、围术期输血、合并≥3 种内科疾病是围手术期下肢 DVT 发生的独立危险因素。DVT 预测模型有助于临床判断血栓发生风险并提前预防。

Abstract:

[Objective] To explore the risk factors of deep vein thrombosis (DVT) in the perioperative period of hip fractures in the el- derly, and to establish a prediction model to guide clinical prediction of DVT. [Methods] A retrospective study was conducted on 384 elder- ly patients who received surgical treatment for hip fractures in our hospital from August 2015 to September 2021. According to the occur- rence of thrombosis, the patients were divided into DVT group and non-DVT group. Univariate and multivariate analysis were performed to identify the risk factors of DVT in lower extremities, and establish the prediction model and evaluate the efficiency of the model. [Results] Among 384 patients, 67 cases (17.5%) developed lower extremity DVT in the perioperative period. In term of univariate comparison, there were no significant differences in terms of age, sex, fracture type, fracture side, ASA grade, time elapsed between the injury and surgery, whether oral anticoagulant, antithrombotic drugs, underlying medical disease, surgical procedure and anesthesia, as well as Hb, HCT, WBC, PLT, fibrinogen, D-dimer, PT and PO2 measured on admission between the two groups (P＞0.05) . However, the DVT group had sig- nificantly longer waiting time before surgery, ≥3 types of comorbidities, higher proportion of blood transfusion and combined fractures, lon- ger operation time, more intraoperative blood loss, and longer APTT than the non-DVT group (P＜0.05) . As results of multivariate binary lo- gistic regression analysis, perioperative blood transfusion (OR=2.565, P＜0.001) , number of comorbidities ≥3 (OR=1.880, P=0.031) , longer preoperative waiting time (OR=1.110, P＜0.001) were the independent risk factors for perioperative DVT of lower extremity. The accuracy of lower extremity DVT prediction equation based on logistic regression results was of 82.8%, with the critical cutoff value of 0.319, the sensi- tivity of 71.6%, the specificity of 60.3%, and the area under the ROC curve of 0.705. [Conclusion] Longer preoperative waiting time, peri- operative blood transfusion and ≥3 kinds of underling medical diseases are the independent risk factors for perioperative DVT of lower ex- tremities. In addition, this DVT prediction model is helpful to judge the risk of thrombosis and prevent it in advance.