论文部分内容阅读
目的建立艾滋病(AIDS)合并马尔尼菲青霉菌病(PSM)的临床诊断评分体系,并对其运用价值进行初步评价。方法采用回顾性研究的方法,对柳州市人民医院2010年4月至2015年5月309例AIDS住院患者的临床和实验室资料,通过SPSS 19.0软件随机化分组(建模组240例和验模组69例)。选择两组间有统计学差异的变量进行Logistic回归分析,筛选出对PSM诊断有影响的独立变量。根据各变量的β回归系数设立相应分值,并建立PSM的临床诊断评分体系(PSMCDS),用受试者工作特征曲线(ROC曲线)评价PSMCDS的预测价值,最后用验模组的临床资料对PSMCDS进行验证。结果共有5个因素被筛选入模型中,分别为外周或腹腔淋巴结肿大、肝肿大、脾肿大、AST>1×ULN且AST>ALT、CD4+T淋巴细胞(简称CD4细胞)计数的自然对数,建立了由这5个变量组成的PSMCDS,分别为外周或腹腔淋巴结肿大(10分)、肝脏肿大(10分)、脾脏肿大(10分)、AST>1×ULN且AST>ALT(达标20分)、CD4细胞计数>25个/!L≤50个/!L(5分)及≤25/!L(10分)。取分值≥35.0时,用于诊断PSM的曲线下面积(AUC)为0.871[95%可信区间(CI):0.824~0.918(P<0.05)],灵敏度、特异性、阳性预测值、阴性预测值分别为73.9%、87.4%、90.6%和66.4%。验模组验证结果:AUC为0.838(95%CI:0.746~0.930)(P<0.05)。结论 5个变量建立的PSM临床诊断评分体系简单易行,在PSM流行地区,用于评估CD4细胞计数<200个/!L的AIDS病人是否合并PSM有较好的临床诊断价值。
Objective To establish a clinical diagnostic scoring system for AIDS combined with Penicillium marneffei (PSM) and to evaluate its application value. Methods A retrospective study was conducted to analyze the clinical and laboratory data of 309 AIDS patients admitted to the People’s Hospital of Liuzhou between April 2010 and May 2015 by SPSS 19.0 software (group 240 and model Group 69 cases). Logistic regression analysis was used to select the statistically significant variables between the two groups to screen out independent variables that had an impact on the diagnosis of PSM. According to the β regression coefficient of each variable, the corresponding score was established, and the PSMCDS was established. The predictive value of PSMCDS was evaluated by the receiver operating characteristic curve (ROC curve). Finally, the clinical data of the model group PSMCDS to verify. Results A total of 5 factors were screened into the model, which were peripheral or abdominal lymphadenopathy, hepatomegaly, splenomegaly, AST> 1 × ULN, AST> ALT, count of CD4 + T lymphocytes Natural Logarithm, a PSMCDS consisting of these 5 variables was established, with peripheral or abdominal lymphadenopathy (10 points), enlarged liver (10 points), enlarged spleen (10 points), AST> 1 × ULN and AST> ALT (up to 20 points), CD4 count> 25 / L≤50 / L (5 points) and ≤25 / L (10 points). The area under the curve (AUC) for the diagnosis of PSM was 0.871 [95% confidence interval (CI): 0.824 to 0.918 (P <0.05)] with a score of 35.0 or higher, sensitivity, specificity, positive predictive value, The predicted values are 73.9%, 87.4%, 90.6% and 66.4% respectively. The validation results of the test module: AUC was 0.838 (95% CI: 0.746-0.930) (P <0.05). Conclusions The PSM clinical diagnostic scoring system established by five variables is simple and easy. In the endemic areas of PSM, AIDS patients with CD4 counts <200 cells / L have better clinical diagnostic value in combination with PSM.