| 赵晓勇,刘弘扬,任原.基于 Caprini血栓风险评分及 D-二聚体构建老年髋部骨折术后深静脉血栓形成的风险预测模型[J].安徽医药,2024,28(12):2452-2457. |
| 基于 Caprini血栓风险评分及 D-二聚体构建老年髋部骨折术后深静脉血栓形成的风险预测模型 |
| Risk prediction model for deep vein thrombosis after hip fracture surgery in the elderly based on Caprini thrombosis risk score and D-dimer |
| |
| DOI：10.3969/j.issn.1009-6469.2024.12.024 |
| 中文关键词: 髋骨折 老年人 术后深静脉血栓形成 Caprini血栓风险评分 D-二聚体 预测模型 |
| 英文关键词: Hip fractures Elderly Postoperative deep vein thrombosis Caprini thrombosis risk score D-dimer Prediction model |
| 基金项目:河北省医学科学课题研究计划( 20221737) |
|
| 摘要点击次数: 513 |
| 全文下载次数: 252 |
| 中文摘要: |
| 目的基于 Caprini血栓风险( CRS)评分及 D-二聚体( D-D)构建老年髋部骨折术后深静脉血栓形成( DVT)的风险预测模型。方法选取 2019年 8月至 2022年 7月唐山市第二医院收治的 206例拟接受髋部骨折手术的老年病人，依据术后有无发生 DVT分为 DVT组( 52例)和对照组( 154例)，比较两组 CRS、D-D及基线资料。采用二元 logistic回归模型分析老年髋部骨折术后 DVT的危险因素，建立风险预测模型，并对模型进行验证和效能评估。结果老年髋部骨折术后 DVT发生率为 25.24%。 DVT组合并糖尿病、受伤至入院时间 >24 h、转子间骨折、腰硬联合麻醉、术后卧床时间 >5 d病人占比及手术时间、总蛋白、酐、凝血酶原时间( PT)、活化部分凝血活酶时间( APTT)、 D-D［(1.64±0.13)mg/L比( 1.17±0.10)mg/L］、 CRS评分［(8.44±1.65)分肌比( 4.26±1.10)分］均高于对照组( P<0.05)。二元 logistic回归分析显示，受伤至入院时间 >24 h(OR=2.04)、术后卧床时间 >5 d(OR=4.54)、合并糖尿病( OR=2.31)、转子间骨折( OR=2.81)、 D-D升高( OR=3.81)、 CRS评分升高( OR=5.33)是老年髋部骨折病人术后发生 DVT的独立危险因素( P<0.05)； Bootstrap法内部验证显示，风险模型列线图的区分度良好［C-index指数 =0.81， 95%CI:(0.74，0.89)］；校准曲线显示模型拟合度好( Hosmer-Lemeshow χ2=0.30，P=0.126)。受试者操作特征曲线( ROC曲线)结果显示，列线图总分最佳截断值为 175.5分，灵敏度、特异度、约登指数分别为 84.61%、93.42%、0.78，曲线下面积为 0.88［95%CI:(0.82，0.92)P<0.05］且列线图预测模型优于 D-D和 CRS评分单独的预测效能( P<0.05)。结论基于 CRS评分及 D-D构建老年髋部骨折术，后 DVT风，险列线图模型区分度、拟合度好，能够直观地预测 DVT发生风险。 |
| 英文摘要: |
| Objective To construct a risk prediction model for deep vein thrombosis (DVT) after hip fracture surgery in the elderlybased on the Caprini thrombosis risk score (CRS) and D-dimer (D-D).Methods A total of 206 elderly patients admitted to The SecondHospital of Tangshan for hip fracture surgery from August 2019 to July 2022 were selected and assigned into a DVT group (n=52) and control group (n=154) based on the presence or absence of DVT after surgery, and the CRS, D-D and baseline information of the two groups were compared. A binary logistic regression model was used to analyse the risk factors for postoperative DVT in elderly patientswith hip fracture. A risk prediction model was established, being validated and assessed for the efficacy.Results The incidence of DVT after hip fracture in the elderly was 25.24%. The proportion of patients with diabetes mellitus, time from injury to hospital admis-sion > 24 hours, intertrochanteric fracture, combined lumbar and epidural anesthesia, postoperative bed time > 5 days, and operationtime, total protein, creatinine, prothrombin time (PT), activated partial thromboplastin time (APTT), D-D [(1.64±0.13) mg/L vs. (1.17± 0.10) mg/L] and CRS score [(8.44±1.65) vs. (4.26±1.10)] in DVT group were higher than those in the control group (P<0.05). Binary lo-gistic regression analysis results showed that time from injury to admission >24 h (OR=2.04), postoperative time in bed >5 d (OR=4.54), combined diabetes (OR=2.31), intertrochanteric fracture (OR=2.81), elevated D-D(OR=3.81), and elevated CRS score (OR=5.33) were independent risk factors for the development of DVT in postoperative elderly patients with hip fracture (P<0.05). Internal validation of the Bootstrap method showed good discrimination of the risk model column line plot [C-index=0.81, 95% CI: (0.74, 0.89)]. Calibration curve showed good model fit (Hosmer-Lemeshow χ2=0.30, P=0.126). The results of the receiver operating characteristic curve plotted showed that the best cut-off value for the total column line plot score was 175.5, with sensitivity, specificity and Jorden index of 84.61%, 93.42% and 0.78 respectively, and the area under the curve of 0.88 [95% CI: (0.82, 0.92), P<0.05], and that the column line plot prediction model outperformed the predictive efficacy of the D-D and CRS scores alone (P< 0.05).Conclusion DVT risk profile model constructed based on CRS score and D-D is good in differentiation and fit, which can intuitively predict the risk of DVT. |
|
查看全文
查看/发表评论 下载PDF阅读器 |
| 关闭 |
|
|
|
