首页|期刊导航|汕头大学医学院学报|基于临床及实验室数据构建肺结核患者合并感染的辅助诊断模型

基于临床及实验室数据构建肺结核患者合并感染的辅助诊断模型OA

Construction of an auxiliary diagnostic model for co-infection in pulmonary tuberculosis patients based on clinical and laboratory data

中文摘要英文摘要

目的:基于临床常规实验室数据,构建并初步评价肺结核患者合并感染的辅助诊断模型.方法:选取2022年1月至2025年3月于惠州市第一人民医院就诊的152例肺结核合并感染患者(感染组)及152例单纯肺结核患者(未感染组).收集患者一般资料及实验室指标,按7∶3比例随机分为训练集(214例)和验证集(90例).采用LASSO回归及多因素logistic回归筛选训练集中与合并感染相关的独立因素,并构建风险评分模型.采用受试者工作特征(ROC)曲线、校准曲线及决策曲线分别评估模型的区分能力、校准度及临床适用性.结果:与未感染组比较,感染组患者糖尿病史比例更高、空洞型结核占比更高、肺结核病程更长,降钙素原(PCT)、中性粒细胞与淋巴细胞比值(NLR)、血小板与淋巴细胞比值(PLR)、C反应蛋白与白蛋白比值(CAR)更高,淋巴细胞与单核细胞比值(LMR)更低(P均<0.05).多因素logistic回归分析显示,糖尿病史(OR=7.62,95%CI:2.24~25.90)、肺结核病程长(OR=1.48,95%CI:1.22~1.79)、PCT高(OR=3.52,95%CI:1.56~7.92)、NLR高(OR=1.41,95%CI:1.21~1.64)、LMR低(OR=0.16,95%CI:0.08~0.33)、CAR高(OR=2.97,95%CI:1.55~5.71)是肺结核患者合并感染的独立相关因素(P均<0.05).训练集和验证集ROC曲线下面积分别为0.874(95%CI:0.835~0.914)和0.833(95%CI:0.779~0.887),校准曲线拟合良好(P均>0.05),决策曲线表明模型在较宽阈值范围内具有临床净获益.结论:糖尿病史、肺结核病程、PCT、NLR、CAR、LMR是肺结核患者合并感染的独立相关因素.基于上述指标构建的模型具有良好的辅助诊断价值,可在病原学结果回报前为临床决策提供量化参考.

Objective:To construct and preliminarily evaluate an auxiliary diagnostic model for co-infection in pulmonary tuberculosis(PTB)patients based on routine clinical laboratory data.Methods:A total of 152 PTB patients with co-infection(infection group)and 152 PTB patients without co-infection(non-infection group)treated at Huizhou First Hospital from January 2022 to March 2025 were enrolled.General clinical data and laboratory parameters were collected.Patients were randomly divided into a training set(n=214)and a validation set(n=90)at a ratio of 7∶3.LASSO regression and multivariate logistic regression were used to identify independent factors associated with co-infection in the training set,and a risk-scoring model was constructed.The discriminative ability,calibration,and clinical applicability of the model were evaluated using receiver operating characteristic(ROC)curves,calibration curves,and decision curves,respectively.Results:Compared with the non-infection group,the infection group had a higher proportion of patients with diabetes history and cavitary tuberculosis,a longer duration of PTB,and higher levels of procalcitonin(PCT),neutrophil-to-lymphocyte ratio(NLR),platelet-to-lymphocyte ratio(PLR),and C-reactive protein-to-albumin ratio(CAR),while the lymphocyte-to-monocyte ratio(LMR)was lower(all P<0.05).Multivariate logistic regression analysis showed that diabetes history(OR=7.62,95%CI:2.24-25.90),longer PTB duration(OR=1.48,95%CI:1.22-1.79),elevated PCT(OR=3.52,95%CI:1.56-7.92),elevated NLR(OR=1.41,95%CI:1.21-1.64),decreased LMR(OR=0.16,95%CI:0.08-0.33),and elevated CAR(OR=2.97,95%CI:1.55-5.71)were independently associated with co-infection in PTB patients(all P<0.05).The area under the ROC curve of the model was 0.874(95%CI:0.835-0.914)in the training set and 0.833(95%CI:0.779-0.887)in the validation set.Calibration curves showed good fit(all P>0.05),and decision curves indicated that the model provided clinical net benefit across a wide range of threshold probabilities.Conclusion:Diabetes history,PTB duration,PCT,NLR,CAR,and LMR are independently associated with co-infection in PTB patients.The model constructed based on the above indicators demonstrates good auxiliary diagnostic value and can provide quantitative decision support for clinicians before pathogen test results are available.

邓宇伟;孔萍;程知音;罗清兰;吴显劲

汕头大学医学院,广东 汕头 515041||惠州市第一人民医院医学检验科,广东 惠州 516003惠州市第一人民医院医学检验科,广东 惠州 516003惠州市第一人民医院呼吸与危重症医学科,广东 惠州 516003惠州市第一人民医院医学检验科,广东 惠州 516003汕头大学医学院,广东 汕头 515041||惠州市中心人民医院检验中心,广东 惠州 516001

医药卫生

肺结核合并感染相关因素辅助诊断模型炎症-免疫相关指标

pulmonary tuberculosisco-infectionassociated factorsauxiliary diagnostic modelinflammation-immunity-related indicators

《汕头大学医学院学报》 2026 (3)

150-155,6

惠州市医疗卫生领域科技计划(2024CZ010099)

10.13401/j.cnki.jsumc.2026.03.005

评论