检验医学与临床2026，Vol.23Issue(3)：358-365,8.DOI:10.3969/j.issn.1672-9455.2026.03.012
基于ADA、GR、VE、Hb、RBC的G6PD缺乏症列线图诊断模型构建及其诊断效能验证
Development and validation of a nomogram diagnostic model for G6PD deficiency based on ADA,GR,VE,Hb and RBC
摘要
Abstract
Objective To construct a glucose-6-phosphate dehydrogenase(G6PD)deficiency diagnostic model based on serum adenosine deaminase(ADA),glutathione reductase(GR),vitamin E(VE)and hemo-globin(Hb)and red blood cell count(RBC)to verify its efficacy.Methods A total of 521 patients who un-derwent G6PD screening at Maoming Maternal and Child Health Hospital from September 1,2022 to Decem-ber 31,2024 were selected as the research subjects.According to the diagnostic criteria for G6PD deficiency in the"Expert consensus on screening,diagnosis and treatment of glucose-6-phosphate dehydrogenase deficiency in newborns",the research subjects were divided into the case group(243 cases)and the control group(278 cases).The dataset was randomly divided into the training set and the validation set using stratified sampling with a ratio of 7∶3.Baseline data of all research subjects were collected,and the levels of ADA,GR and VE were detected.LASSO regression analysis was used to screen variables,and then multivariable Logistic regres-sion analysis was used to analyze the influencing factors of G6PD deficiency.A diagnostic model was construc-ted.The receiver operating characteristic(ROC)curve,calibration curve and decision curve were drawn to verify the efficacy of the model.Results The levels of total bilirubin(TBIL),direct bilirubin(DBIL),indirect bilirubin(IBIL),ADA and GR in the case group of the training set were higher than those in the control group,while the levels of VE,Hb and RBC were lower than those in the control group,and the differences were statistically significant(P<0.05).Through LASSO regression,IBIL,DBIL,VE,ADA,GR,Hb,RBC were selected as core variables.The results of multivariable Logistic regression analysis showed that elevated ADA and GR levels were independent risk factors for G6PD deficiency(P<0.05),and elevated RBC,VE,Hb levels were independent protective factors for G6PD deficiency(P<0.05).The expression of the constructed diagnostic model was Logit(P)=2.89-0.08XVE+0.85XADA+0.92XGR-0.73XHb-1.47XRBC.ROC curve a-nalysis showed that the area under the curve(AUC)of the G6PD deficiency diagnostic model for G6PD defi-ciency diagnosis in the training set was 0.98,and the AUC in the validation set was 0.99.The Hosmer-Leme-show test results were χ2 training set=4.12,Ptraining set=0.763,χ2validation set=3.85,Pvalidation set=0.802.The diagnostic probability was highly consistent with the actual observation value.The decision curve analysis of the training set and validation set showed that the model had clinical net benefits within the threshold range of 0.00 to 1.00.Conclusion VE deficiency,elevated ADA and GR levels,decreased Hb level and RBC are closely related to G6PD deficiency.The constructed G6PD deficiency diagnostic model based on this study has better stability and diagnostic efficacy,providing a new tool for clinical early screening.关键词
葡萄糖-6-磷酸脱氢酶缺乏症/腺苷脱氨酶/谷胱甘肽还原酶/维生素E/诊断模型Key words
glucose-6-phosphate dehydrogenase deficiency/adenosine deaminase/glutathione reduc-tase/vitamin E/diagnostic model分类
医药卫生引用本文复制引用
吕碧绿,罗文英..基于ADA、GR、VE、Hb、RBC的G6PD缺乏症列线图诊断模型构建及其诊断效能验证[J].检验医学与临床,2026,23(3):358-365,8.基金项目
广东省自然科学基金面上项目(2025A1515012793) （2025A1515012793）
广东省钟南山医学基金会科研资助项目(ZNSXS-20250017) （ZNSXS-20250017）
广东省湛江市非资助科技攻关专题项目(2024B01359). （2024B01359）
