血常规、生化及其复合指标对重症药疹的预测价值分析

Predictive value of blood cell analysis, biochemical indicators and its composite indexes for severe drug eruption

摘要:

目的 探讨血常规、生化及其复合指标对重症药疹（SDE）预测的临床价值。方法 选取2017年1月至2022年12月孝感中心医院299例药疹患者作为研究对象，通过比较SDE和非SDE患者的一般情况（年龄、性别等）和实验室指标（血常规、生化指标），计算红细胞分布宽度与白蛋白比（RAR）、预后营养指数（PNI）和中性粒细胞与淋巴细胞比（NLR）等复合指标。通过比较分析，选取差异有统计学意义的指标进行多因素二元Logistic回归分析，挑选出影响SDE发生的危险因素，使用受试者工作特征（ROC）曲线分析血常规、生化及其复合指标对SDE的临床预测价值。结果 SDE组既往高血压病史、吸烟比例高于非SDE组（P<0.05），住院天数大于非SDE（P<0.05），而性别、糖尿病、饮酒史差异无统计学意义（P>0.05）。SDE组患者NLR、RAR、丙氨酸氨基转移酶、天冬氨酸氨基转移酶、腺苷脱氨酶、总胆红素、直接胆红素、γ谷氨酰基转移酶、肌酐大于非SDE组（P<0.05），血小板、淋巴细胞数、白蛋白、PNI、前白蛋白、碱性磷酸酶、钠、氯、钙则小于非SDE组（P<0.05）。多因素二元Logistic回归分析表明RAR是SDE的危险因素。ROC曲线分析示NLR、RAR、腺苷脱氨酶以及血小板预测SDE的AUC分别为0.689、0.732、0.659及0.253，RAR对SDE具有良好的预测价值。结论 RAR对SDE患者预测有指导意义。

Abstract:

Objective To investigate the clinical value of blood cell analysis, biochemical indicators and its composite indexes in the prediction of severe drug eruption (SDE).Methods A total of 299 patients with drug eruption treated in Xiaogan Central Hospital from January 2017 to December 2022 were selected as the research subjects, and composite indexes such as red blood cell distribution width to albumin ratio (RAR), prognostic nutrition index (PNI) and the neutrophil to lymphocyte ratio (NLR) were calculated by comparing the general conditions of SDE and non-SDE patients, including age and gender, laboratory indicators (blood cell analysis and biochemical indicators). In comparative analysis, multivariate binary logistic regression analysis was performed by selecting statistically significant indicators, the risk factors affecting the occurrence of SDE were selected, and the receiver operating characteristic (ROC) curve was used to analyze the clinical predictive value of blood cell analysis, biochemical indicators and composite indexes on SDE.Results In general, the proportion of previous hypertension history and smoking in SDE group was higher than that in non-SDE group (P<0.05), and the length of hospital stay of SDE group was greater than that of non-SDE group (P<0.05), while the differences in gender, diabetes and alcohol history were not statistically significant (P>0.05). Among the laboratory test indexes, NLR, RAR, alanine aminotransferase, aspartate aminotransferase, adenosine deaminase (ADA), total bilirubin, direct bilirubin, γ glutamyl transferase, creatinine in SDE group were greater than those in non-SDE group (P<0.05), platelet (PLT), lymphocyte number, albumin, PNI, prealbumin, alkaline phosphatase, sodium, chloride, calcium in SDE group were smaller than those in non-SDE group (P<0.05), and the differences in other laboratory indicators were not statistically significant (P>0.05). Multivariate binary logistic regression analysis showed that RAR is a risk factor for SDE. ROC curve analysis showed that the area under the curve (AUC) of NLR, RAR, ADA and PLT in predicting SDE were 0.689, 0.732, 0.659 and 0.253, respectively, and RAR had a good predictive value for SDE.Conclusion RAR has guiding significance for the prediction of patients with SDE.