期刊文献+

原发性骨质疏松症预测模型的建立与应用

The establishment and application of a predictive model for primary osteoporosis
下载PDF
导出
摘要 目的构建原发性骨质疏松症(POP)的风险评分(RPOPS)模型,评价其预测POP的价值。方法选择2021年5月至2023年10月初次骨质疏松筛查者249例,根据骨密度(BMD)分为对照组(n=120)和骨质疏松组(n=129),并收集一般资料和外周血骨质疏松相关参数,包括中性粒细胞计数(N),淋巴细胞计数(L),中性粒细胞/淋巴细胞比值(NLR),血小板计数(P),血小板/淋巴细胞比值(PLR)和红细胞分布宽度(RDW),通过非参数检验、Spearman相关分析和多因素Logistic逐步回归分析预测POP的独立风险因素,构建POP的RPOPS预测模型,并绘制受试者操作特征曲线(ROC),评价RPOPS预测模型的诊断效能。结果两组除RDW、PLR比较差异无统计学意义(P>0.05),其余各指标比较差异均有统计学意义(P<0.001);相关性分析显示,N与NLR(r=0.585,P<0.001),L与NLR(r=-0.594,P<0.001);Logistic回归分析:年龄增长、BMI降低、PLT降低、NLR升高、t-PINP升高、性别是POP发生的独立危险因素(P<0.05);并建立模型RPOPS=-1.658+0.080×年龄+0.921×性别-0.220×BMI-0.008×PLT+1.053×NLR+0.045×t-PINP;HosmerLemeshow检验模型的拟合优度良好(P=0.53),RPOPS模型截断值为0.04时,其曲线下面积为0.905,预测敏感度和特异度分别为81.40%和84.20%。结论RPOPS模型的构建合理,简单易行,适宜基层医院预测和筛查POP。 Objective To construct a Risk Score(RPOPS)model for Primary Osteoporosis(POP)and to evaluate its predictive value for POP.Methods A retrospective analysis was conducted on 249 subjects from the initial osteoporosis screening population in our hospital,from May 2021 to October 2023.They were divided into the normal group(n=120)and the osteoporosis group(n=129)based on BMD T-score.Their general data and peripheral blood osteoporosis related parameters including neutrophil count(N),lymphocyte count(L),neutrophil/lymphocyte ratio(NLR),platelet count(P),platelet/lymphocyte ratio(PLR),and red blood cell distribution width(RDW)were collected.The differences between the groups were compared through Nonparametric Testing,Spearman Correlation Analysis is used to analyze the correlation between various parameters.Binary Multivariate Logistic Stepwise Regression Analysis was used to predict independent high-risk factors of POP.A RPOPS prediction model for evaluating POP was established,and its receiver operating characteristic(ROC)curve was plotted to evaluate the diagnostic effectiveness of the RSPOPS prediction model.Results In 294 patients,there was significant difference between groups(All P value<0.001),except for RDW and PLR(All P>0.05).The correlation analysis showed that N and NLR(r=0.585,P=0.000),L and NLR(r=-0.594,P=0.000).It was found that age,BMI,PLT,NLR,t-PINP,and gender were independent risk factors for the occurrence of POP by stepwise Logistic Regression Analysis(P<0.05 for all).A model RSPOPS=-1.658+0.080×Age+0.921×Gender-0.220×BMI-0.008×PLT+1.053×NLR+0.045×t-PINP was established.The goodness of fit of the Hosmer Lemeshow test model was good(P=0.530).When the ideal cutoff value of the RPOPS model was 0.04,the area under the curve(AUC)was 0.905,and the predicted sensitivity and specificity were 81.4%and 84.2%,respectively.Conclusion The RPOPS model is reasonably constructed.It is simple and feasible,and suitable for predicting and screening POPs in grassroots hospitals.
作者 郑雅琴 李子军 姜新华 赵燕飞 Zheng Yaqin
出处 《浙江临床医学》 2024年第4期503-505,共3页 Zhejiang Clinical Medical Journal
基金 浙江省基础公益研究计划探索项目(Y20H200001) 浙江省龙泉市重点科技计划项目(财政补助类)(2023KJCZ-014)。
关键词 外周血相关参数 原发性骨质疏松 模型构建 预测 Peripheral blood parameters Primary osteoporosis(POP) Model building Prediction
  • 相关文献

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部