【摘要】 目的：探讨经补肾生血法与益气养血法为主治疗的再生障碍性贫血（AA）患者T淋巴细胞亚群变化特点，构建疗效预测的多因素评估模型。方法：回顾性收集一项前瞻、多中心、随机对照试验中患者的临床资料，包括治疗分组、性别、合并症、年龄、病程、HGB、PLT、WBC、ANC和各项T淋巴细胞亚群等指标。根据《血液病诊断及疗效标准》对治疗6个月后的AA患者进行疗效评价，以“有效”“无效”作为因变量，通过单因素、多因素Logistic回归分析筛选出影响治疗有效率的指标，根据回归方程结果构建预测模型，并分析其预测效能。结果：共纳入277例AA患者临床资料进行分析，其中补肾生血组136例，益气养血组141例，治疗6个月后疗效评价治疗有效共219例，无效共58例。将18个相关指标纳入单因素分析后，P<0.1的自变量则进入多因素Logistic分析，治疗分组中设置益气养血组为哑变量，补肾生血组（P<0.001,OR=2.88,95%CI[1.50,5.74]）,PLT（P=0.03,OR=1.01,95%CI[1.00,1.02]）,CD₃⁺CD₈⁺（P=0.04,OR=1.02,95%CI[1.00,1.05]），提示治疗分组、PLT、CD₃⁺CD₈⁺是预测AA中西医结合治疗效果的保护性因素，根据结果使用R 4.1.3软件生成疗效预测列线图，列线图模型ROC的AUC为0.71,95%CI[0.63,0.78]，模型内部验证校准曲线和标准曲线较为贴近，重合度较高，拟合度良好，Hosmer-Lemeshow拟合优度检验为模型一致性良好（χ²=4.02,P=0.86）。结论：T淋巴细胞亚群CD₃⁺CD₈⁺中是AA疗效的重要影响因素，结合中医治疗分组、PLT建立的列线图模型可以用于中西医结合治疗AA患者的疗效预测。更多还原

【Abstract】 Objective:To investigate the characteristics of T lymphocyte subpopulation changes in patients with aplastic anemia （AA） treated mainly by the method of nourishing kidney and engendering blood and the method of boosting qi and nourishing blood and to construct a multifactorial assessment model for efficacy prediction.Methods:Clinical data of patients in a prospective,multicenter,randomized controlled trial were retrospectively collected,including treatment subgroups,gender,comorbidities,age,disease age,HGB,PLT,WBC,ANC,and various T lymphocyte subsets.According to the Diagnostic and Efficacy Criteria for Hematological Diseases,the efficacy of AA patients after 6 months of treatment was evaluated,‘effective’and‘ineffective’were used as the dependent varivables,and the indexes affecting treatment efficiency were screened out by univariate and multivariate Logistic regression analysis,and prediction models were constructed according to the results of regression equations and their predictive efficacy was analyzed.Results:A total of 277 AA patients’clinical data were included for analysis,including 136 cases in the group of nourishing kidney and engendering blood,141 cases in the group of boosting qi and nourishing blood,and a total of 219 cases were evaluated as effective and 58 cases were ineffective in treatment after 6months of treatment.After including 18 relevant indexes in the univariate analysis,the independent variables with P<0.1 were then entered into the multivariate Logistic analysis,and the treatment grouping was set as dummy variables for the group of boosting qi and nourishing blood,the group of nourishing kidney and engendering blood (P<0.001,OR=2.88,95%CI[1.50,5.74],PLT (P=0.03,OR=1.01,95%CI[1.00-1.02],CD₃⁺CD₈⁺（P=0.04,OR=1.02,95%CI[1.00,1.05]） were protective factors for predicting efficacy.Based on the results,the nomogram model of efficacy prediction was generated using R 4.1.3 sofeware.The AUC of the ROC of the nomogram model was 0.71,95%CI[0.63,0.78],the calibration curve and the standard curve of the internal validation of the model were relatively close,the degree of coincidence was high,and the degree of fit was good.The Hosmer-Lemeshow goodness of fit test was good for model consiseency （χ²=4.02,P=0.86）.Conclusion:CD₃⁺CD₈⁺in T lymphocyte subsets is an important factor affecting the efficacy of AA,and the nomogram model established by combining traditional Chinese medicine treatment grouping and PLT can be used to predict the efficacy of integrated traditional Chinese and Western medicine in the treatment of AA patients.更多还原