文章摘要
魏港花,徐秉轩,穆清爽.血清 25-羟维生素 D与社区获得性肺炎的相关性及对死亡风险预测价值分析[J].安徽医药,2025,29(8):1536-1541.
血清 25-羟维生素 D与社区获得性肺炎的相关性及对死亡风险预测价值分析
Correlation between serum 25-hydroxyvitamin D and community-acquired pneumonia and its predictive value for mortality risk
  
DOI:10.3969/j.issn.1009-6469.2025.08.010
中文关键词: 社区获得性肺炎  维生素 D  logistic回归模型  胸腔积液  死亡风险  感染风险
英文关键词: Community-acquired pneumonia  Vitamin D  Logistic regression model  Pleural effusion  Mortality risk  Infec-tion risk
基金项目:新疆维吾尔自治区自然科学基金项目( 2021D01C371);医药卫生“天山英才”高层次人才培养计划( TSYC202301B062)
作者单位E-mail
魏港花 新疆医科大学第二附属医院干部一科,新疆维吾尔自治区乌鲁木齐 830063  
徐秉轩 新疆医科大学第二附属医院干部一科,新疆维吾尔自治区乌鲁木齐 830063  
穆清爽 新疆医科大学第二附属医院干部一科,新疆维吾尔自治区乌鲁木齐 830063 shootingrains@163.com 
摘要点击次数: 1071
全文下载次数: 267
中文摘要:
      目的探究血清 25-羟维生素 D[( 25(-OH)D)]与社区获得性肺炎( CAP)的相关性及对死亡风险预测价值分析。方法收集 2022年 12月至 2023年 8月于新疆医科大学第二附属医院就诊的 100例社区获得性肺炎病人临床资料作为病例组,并根据社区获得性肺炎严重指数( PSI)评估死亡风险,并将其分为中高危组( PSI评分 ≥91分,共 20例)、低危组( PSI评分 <91分,80例)同时选取同时期于该院行健康体检者 100例作为健康组,并收集其一般资料及血清 25-(OH)D检测结果。对比健康组共及病例组,(OH)D水平分析血清 25-采用 logistic回归分析探讨 CAP病人死亡风险的影的 25-(OH)D与 CAP感染风险的相关性,响因素并构建死亡风险预测模型,分析模型的预测价值。结果病例组血清 25(-OH)D水平( 15.35±5.30)μg/L明显低于健康(18.68±7.98)μg/L(t=3.48,P=0.001)。以血清 25-(OH)D构建 CAP病人死亡风险预测模型的受试者操作特征曲线( ROC曲线)组下面积为 0.70[95%CI:(0.57,0.82),P=0.007],总预测正确率为 79.0%,Hosmer-Lemeshow检验统计量为 5.58(P=0.695),提示该模型拟合效果良好,但预测价值有限;以性别( X1)、年龄( X2)、胸腔积液( X5)、 25-(OH)D(X10)构建的 logistic回归模型为 Log(P2)=6.437+0.269X1 . 2.291X2+4.292X5 . 0.435X10(χ2=65.14,P<0.001),其曲线下面积为 0.97[95%CI:(0.95,1.00), P<0.001],总预测正确率为 91.0%,Hosmer-Lemeshow检验统计量为 3.65(P=0.887)提示多因素死亡风险模型的预测价值高、拟合效果好。结论血清 25-且 25-年龄、胸腔积液构建的多因素 logistic回归模别(OH)D与 CAP的感染呈负相关,(OH)D联合性型预测价值高、拟合效果好,可为多因素联合预测 CAP病人死亡风险提供参考依据。
英文摘要:
      Objective To explore the correlation between serum 25-hydroxyvitamin D [25-(OH) D] and community-acquired pneu-monia (CAP) and its predictive value for mortality risk.Methods Clinical data of 100 patients with CAP treated in the Second Affiliat-ed Hospital of Xinjiang Medical University from December 2022 to August 2023 were collected as the case group. The risk of death wasevaluated according to the Pneumonia severity index (PSI), and the patients were assigned into medium and high risk group (PSI score≥91, a total of 20 cases) and low risk group (PSI score<91, a total of 80 cases). At the same time, 100 patients who underwent physical ex-amination in our hospital at the same time were selected as the health group, and their general data and serum 25-(OH) D detection re-sults were collected. The serum 25-(OH) D levels of the health group and the case group were compared in order to analyze the correla-tion between the serum 25-(OH) D levels and the risk of CAP infection. Logistic regression analysis was used to explore the factors af-fecting the mortality risk in CAP patients, and the model was constructed to analyze the predictive value of the model.Results The se-rum 25-(OH) D level in the case group (15.35±5.30) μg/L was significantly lower than that in the health group (18.68±7.98) μg/L, (t= 3.48, P=0.001). The area under ROC curve (AUC) of the mortality risk prediction model based on serum 25-(OH) D was 0.70 [95%CI: (0.57, 0.82), P=0.007]. The total prediction accuracy was 79.0%, and the Hosmer-Lemeshow test statistic was 5.58 (P=0.695), which in-dicated that the model had good fitting effect but limited prediction value. The logistic regression model based on gender (X1), age (X2), pleural effusion (X5) and serum 25-(OH) D (X10) was Log(P2) =6.437+0.269X1 . 2.291X2+4.292X5 . 0.435X10(χ2=65.14, P<0.001). The AUC of the prediction model was 0.97 [95%CI: (0.95, 1.00), P<0.001]. The total prediction accuracy was 79.0%, and the Hosmer-Lem-eshow test statistic was 3.65 (P=0.887), which suggested that the multi-factor mortality risk model had high prediction value and good fitting effect.Conclusion Serum 25-(OH) D is inversely associated with the risk of CAP infection. In addition, the multi-factor logistic regression model constructed by 25-(OH) D combined with gender, age and pleural effusion shows high predictive value and good fit-ting effect, which can provide a reference for multi-factor combined prediction of death risk in CAP patients.
查看全文   查看/发表评论  下载PDF阅读器
关闭

分享按钮