基于Meta分析构建ICU获得性衰弱风险预测模型

Construction of a risk prediction model for ICU acquired weakness-based on Meta-analysis

目的 构建并验证ICU获得性衰弱风险预测模型,为早期筛查ICU获得性衰弱危险人群提供有效工具。方法 Meta分析ICU获得性衰弱的危险因素,以各危险因素的综合危险度的自然对数为模型的β系数,以某时段ICU获得性衰弱发病率与未发病率比值的自然对数为模型的α系数,建立预测模型。收集137例ICU患者资料,分析模型的预测性能。结果 ICU获得性衰弱预测模型为logit(P)=-0.781+0.039年龄+0.077APACHE+0.940休克+0.122机械通气时长+0.971脓毒血症+0.756神经肌肉阻滞剂+1.311血管活性药物+1.308皮质类固醇+0.747性别+0.626高血糖+1.973肾脏替代治疗;模型ROC曲线下面积为0.752,95%CI(0.672,0.833),概率P函数的临界值为0.998,敏感度为0.930,特异度为0.489。结论 基于Meta分析建立的ICU获得性衰弱风险预测模型可作为早期预测ICU获得性衰弱危险人群的工具。

Objective To construct and validate an ICU acquired weakness(ICUAW) risk prediction model,and to provide an effective tool for early screening of ICUAW-risk population.Methods Meta-analysis was conducted to calculate the risk factors of ICUAW,the natural logarithm of the comprehensive risk of each risk factor was taken as the β coefficient,and the natural logarithm of the ratio of ICUAW incidence rate to non-incidence rate during a certain period was taken as the α coefficient to construct a risk prediction model of ICUAW.Then the data of 137 ICU patients were collected to evaluate the predictive performance of the model.Results The model of ICUAW was logit(P)=-0.781+0.039 age+0.077 APACHE+0.940 shock+0.122 duration of mechanical ventilation+0.971 sepsis+0.756 neuromuscular blocker+1.311 vasoactive drugs+1.308 corticosteroids+0.747 gender+0.626 hyperglycemia+1.973 renal replacement therapy.The area under the receiver operating characteristic curve of the model was 0.752,95%CI was(0.672,0.833).The cut-off value of probability function(P) was 0.998,the sensitivity was 0.930,and specificity was 0.489.Conclusion The ICUAW risk prediction model based on Meta-analysis can be used as a tool for early prediction of ICUAW risk population.