摘要
目的:分析纤维蛋白原(FIB)与白蛋白(ALB)比值(FAR)对脑梗死患者非计划再入院的影响,并分析列线图模型的预测价值。方法:选取2020年10月—2022年5月安徽中医药大学附属太和中医院诊断和治疗的脑梗死患者200例,分析FAR对脑梗死患者1年内非计划再入院的预测价值,根据影响因素建立风险模型,并对模型的预测效能进行验证。通过Cox和Kaplan-Meier生存回归进一步分析FAR对脑梗死患者非计划再入院的影响。结果:FAR对脑梗死患者非计划再入院的预测曲线下面积(AUC)为0.732,最佳截断值为0.127,灵敏度为76.92%,特异度为60.10%。单因素和多因素分析显示:年龄、男性、糖尿病、低密度脂蛋白、FAR≥0.127均为脑梗死患者非计划再入院的风险因素,OR值分别为1.189、3.319、4.067、1.363、4.588。经验证脑梗死患者非计划再入院列线图模型一致性指数为0.841(95%CI 0.778~0.936),预测AUC为0.841(95%CI 0.775~0.908),灵敏度为78.85%,特异度为72.30%。Kaplan-Meier生存分析显示,FAR≥0.127的患者再入院率高于FAR<0.127患者,差异有统计学意义(P<0.05)。Cox生存分析显示,纠正混杂因素前后FAR均为脑梗死患者非计划再入院风险因素(HR>1,P<0.05)。结论:FAR为脑梗死患者非计划再入院的影响因素,建立的列线图模型有较好的预测效能。
Objective:To analyze the effect of fibrinogen(FIB)-to-albumin(ALB)ratio(FAR)on unplanned readmission in patients with cerebral infarction,and to construct a column chart risk prediction model.Methods:200 cerebral infarction patients diagnosed and treated in Taihe Hospital of Anhui University of Traditional Chinese Medicine between October 2020 and May 2022 were selected to analyze the predictive value of FAR on unplanned readmission of cerebral infarction patients within one year.A column chart risk prediction model was constructed based on influencing factors,and the predictive performance of the model was validated.The impact of FAR on unplanned readmission of patients with cerebral infarction through Cox and Kaplan-Meier survival regression were further analyzed.Results:The predictive area under the curve(AUC)of FAR for unplanned readmission of patients with cerebral infarction was 0.732,the optimal cutoff value was 0.127,the sensitivity and specificity were 76.92%and 60.10%respectively.Univariate and multivariate analysis showed that age,male,diabetes,low density lipoprotein and FAR≥0.127 were risk factors for unplanned readmission of patients with cerebral infarction,and the OR values were 1.189.3.319,4.067,1.363 and 4.588,respectively.Based on this,a column chart risk prediction model for unplanned readmission of patients with cerebral infarction was constructed.After verification,the consistency index of the model was 0.841(95%CI 0.778-0.936),the predicted AUC was 0.841(95%CI 0.775-0.908),the sensitivity was 78.85%.and the specificity was 72.30%.Kaplan-Meier survival analysis showed that the readmission rate of patients with FAR≥0.127 was higher than that of patients with FAR<0.127.and the difference was statistically significant(P<0.05).Cox survival analysis showed that FAR was a risk factor for unplanned readmission in patients with cerebral infarction before and after correcting for confounding factors(HR>1,P<o.05).Conclusion:FAR is an influencing factor for unplanned readmission of patients with cerebral infarction,and the constructed risk prediction model has good predictive performance.
作者
李新宇
田方圆
陈艳华
曹科峰
LI Xinyu;TIAN Fangyuan;CHEN Yanhua;CAO Kefeng(Department of Laboratory,Taihe Hospital of Anhui University of Traditional Chinese Medicine,Fuyang,Anhui,236600,China)
出处
《临床血液学杂志》
2025年第2期148-153,共6页
Journal of Clinical Hematology
关键词
脑梗死
纤维蛋白原
白蛋白
非计划再入院
列线图模型
cerebral infarction
fibrinogen
albumin
unplanned readmission
column chart model
作者简介
通信作者:李新宇.E-mail:648282039@qq.com。
