长效抗凝血类灭鼠剂中毒患者出现凝血功能障碍的风险预测模型
DOI:
作者:
作者单位:

作者简介:

通讯作者:

基金项目:


Analysis of clinical characteristics of long-acting anticoagulant rodenticide poisoning and predictive model of coagulation dysfunction risk
Author:
Affiliation:

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
    摘要:

    目的 探讨长效抗凝血类灭鼠剂(LAARs)中毒后的临床特征及对凝血功能的影响,建立凝血功能障碍风险预测模型,预测LAARs中毒后发生凝血功能障碍的风险,以指导临床诊疗。 方法 选取2016年6月~2021年5月西南医科大学附属医院收治的临床诊断为LAARs中毒的患者100例,根据入院时有无凝血功能障碍分为凝血功能障碍(56例)组和无凝血功能障碍组(40例),比较两组患者的一般人口学特征、临床症状、实验室检查指标、治疗措施,选出具有统计学意义或具有重要临床意义的变量,进一步纳入多因素Logistic回归模型以建立LAARs患者出现凝血功能障碍的风险预测模型。 结果 96例中出现凝血功能障碍的潜伏期为(8.94±4.37)d,出血症状以血尿最多共32例,实验室检查凝血因子Ⅱ、Ⅶ、Ⅸ、Ⅹ值明显低于参考值;抗凝血治疗3 d后,PT、APTT、TT、纤维蛋白原(FIB)、INR值较治疗前均有不同程度降低,凝血酶原活动度(PTA)值较治疗前升高。凝血功能障碍组NEU-R%、SCR、CRP、cTNT、PT、APTT、FIB、INR、输新鲜冰冻血浆量等指标较无凝血功能障碍组高,凝血因子Ⅱ、Ⅶ、Ⅸ、Ⅹ、PTA、Hb值较无凝血功能障碍组低,差异有统计学意义(均P<0.05)。LAARs中毒后潜伏期较长,临床症状主要表现为全身不同组织器官出血,以血尿最常见;实验室检查主要表现为 PT、APTT明显延长,凝血因子Ⅱ、Ⅶ、Ⅸ、Ⅹ活性降低;使用维生素K1治疗有效。用5-折交叉验证考察3个多因素Logistic回模型的备选方案,最终选定预测精度最高的模型,其中包含鼠药类型、CRP、SCR、NEU-R%、WBC、VK1剂量、用药天数等7个变量,进一步采用随机森林模型对该模型中各变量对于凝血功能障碍风险预测的重要性进行排序,排序结果为CRP>NEU-R%>用药天数>WBC>VK1剂量>SCR>鼠药类型(P<0.05)。 结论 被选中的预测精度最高的模型中包含鼠药类型、CRP、SCR、NEU-R%、WBC、VK1剂量、用药天数等7个变量,其中CRP值是LAARs中毒凝血功能障碍风险预测模型中最重要的预测变量。

    Abstract:

    Objective To investigate the clinical characteristics and effects of long-acting anticoagulant rodenticides (LAARs) poisoning on coagulation function, and establish a risk prediction model for coagulation dysfunction, so as to predict the risk of coagulation dysfunction after LAARs poisoning, and guide clinical diagnosis and treatment. Methods One hundred patients with clinical diagnosis of LAARs intoxication admitted to The Affiliated Hospital of Southwest Medical University from June 2016 and May 2021 were selected, and 96 cases were included after screening by inclusion and exclusion criteria, divided into two groups according to the presence or absence of coagulation dysfunction on admission. The general demographic characteristics, clinical symptoms, laboratory test indexes, and treatment measures of the two groups were compared, and statistically significance or clinically significant variables were selected and further incorporated into a multifactorial logistic regression model to establish a risk prediction model for the development of coagulation dysfunction in patients with LAARs. Results The incubation period of coagulopathy was (8.94±4) 37 days in 96 cases, and the highest number of bleeding symptoms was hematuria in 32 cases. The values of coagulation factors Ⅱ Ⅶ, Ⅸ and Ⅹ in laboratory tests were significantly lower than the reference values. After 3 days of anticoagulation treatment, the values of PT, APTT, TT, fibrinogen (FIB) and INR were decreased to varying degrees, while the value of prothrombin activity (PTA) was increased. Blood coagulation dysfunction group NEU-R%, SCR, CRP, cTNT, PT, APTT, FIB, INR, lost fresh frozen plasma volume indicators such as less high blood coagulation dysfunction group, clotting factor Ⅱ, Ⅶ, Ⅸ, Ⅹ, PTA, Hb value was lower than those of no blood coagulation dysfunction group, the difference was statistically significant (P<0.05). The incubation period after LAARs poisoning is long, and the main clinical symptoms are bleeding in different tissues and organs of the whole body, and hematuria is the most common. Laboratory examination showed that PT and APTT were significantly prolonged and the activities of coagulation factors Ⅱ, Ⅶ, Ⅸ and Ⅹ were decreased. Treatment with vitamin K1 is effective. With five-fold cross-validation study for more than three factor Logistic alternatives of back model, finally selected highest precision model, including type of rat poison, CRP, SCR, NEU-R%, the WBC and VK1 seven variables, such as dosage, medication days Further USES the random forest model for each variable in the model for the importance of the blood coagulation dysfunction risk prediction for sorting, sorting results for CRP> NEU-R %> medication days> the WBC> VK1 dose SCR> type of rat poison (P<0.05). Conclusion The model with the highest prediction accuracy included 7 variables including rat drug type, CRP, SCR, NEU-R%, WBC, dose of VK1 and medication days, among which CRP value was the most important predictor of LAARs poisoning cocoagulant dysfunction risk prediction model.

    参考文献
    相似文献
    引证文献
引用本文
分享
文章指标
  • 点击次数:
  • 下载次数:
历史
  • 收稿日期:
  • 最后修改日期:
  • 录用日期:
  • 在线发布日期: 2022-10-20
您是第位访问者
网站版权所有:《西部医学》编辑部     蜀ICP备18038379号-4
地址:四川省成都市武侯区小天竺街75号财富国际18F-1号    邮政编码:610041
电话:028-85570072/85588403 本网站支持 IPv6    E-mail:xbyxqk@163.com
技术支持:北京勤云科技发展有限公司