基于倾向性评分匹配法探讨血小板与淋巴细胞比值与维持性血液透析患者抑郁的相关性

Relationship of platelet-to-lymphocyte ratio with depression in patients with maintenance hemodialysis based on propensity score matching

  • 摘要:
    目的 基于倾向性评分匹配法探讨血小板与淋巴细胞比值(PLR)与维持性血液透析(MHD)患者抑郁的相关性。
    方法 收集2023年6—8月贵州省内20家三甲医院及三级医院血液透析中心患者的年龄、性别、基础疾病等一般资料,以及生化指标和贝克抑郁量表评分(BDI)。根据是否发生抑郁分为抑郁组和非抑郁组,比较各组临床特征的差异。采用1∶1倾向性评分匹配(PSM)获得基线资料(性别、年龄、教育程度)相似的抑郁及无抑郁人群,比较2组临床特征的差异; 根据PLR四分位数将研究对象分为Q1、Q2、Q3、Q4组,比较各组临床特征的差异; 采用Logistic回归分析评估PLR与MHD患者抑郁发病的关系; 采用受试者工作特征(ROC)曲线分析PLR对MHD患者抑郁的评估价值。
    结果 本研究共纳入2 081例MHD患者,男1 286例(61.8%), 年龄(54.3±14.7)岁,抑郁共368例(患病率为17.7%), 其中PSM后成功配对359对; 抑郁组的糖尿病患者比率、PLR等较高,而血肌酐值、血红蛋白较低,差异均有统计学意义(P < 0.05)。PLR Q4组抑郁风险为Q1组的22.654倍(95%CI: 12.713~40.369, P < 0.001); 初步校正年龄、糖尿病、高血压后, PLR Q4组抑郁风险为Q1组的28.272倍(95%CI: 15.390~51.937, P < 0.001); 进一步校正血肌酐、血红蛋白、白蛋白、白细胞计数、总胆固醇、单核细胞与淋巴细胞比值、全身免疫炎症指数、中性粒细胞与淋巴细胞比值后, PLR Q4组抑郁风险为Q1组的10.557倍(95%CI: 3.461~32.201, P < 0.001)。ROC曲线分析结果提示, PLR预测MHD患者抑郁的灵敏度为78.0%, 特异度为73.5%, 最佳阈值为114.14, 曲线下面积为0.798(95%CI: 0.765~0.832, P < 0.001)。
    结论 PLR为MHD患者抑郁的独立危险因素,对MHD患者抑郁具有较高的预测价值。

     

    Abstract:
    Objective To investigate the correlation of platelet-to-lymphocyte ratio (PLR) with depression in patients with maintenance hemodialysis (MHD) based on propensity score matching (PSM).
    Methods General information such as age, gender and underlying diseases, biochemical indicators, and Beck Depression Inventory (BDI) scores of patients in hemodialysis centers of 20 Grade Ⅲ and Level A hospitals in Guizhou Province from June to August 2023 were collected. Patients were divided into depression group and non-depression group based on the occurrence of depression, and differences in clinical characteristics were compared between the groups. A 1 to 1 propensity score matching was used to obtain patients with similar baseline data (gender, age, education level) in the depression and non-depression groups, and differences in clinical characteristics between the two groups were compared. Participants were divided into Q1, Q2, Q3 and Q4 groups according to the quartiles of PLR, and differences in clinical characteristics among the groups were compared. Logistic regression analysis was used to evaluate the relationship between PLR and depression in MHD patients. Receiver operating characteristic (ROC) curve was used to assess the value of PLR in predicting depression in MHD patients.
    Results A total of 2, 081 MHD patients were included in this study, including 1, 286 males (61.8%), with a mean age of (54.3±14.7) years. A total of 368 patients (17.7%) had depression, and 359 pairs were successfully matched after propensity score matching. The depression group had higher rates of diabetes and PLR, but lower serum creatinine and hemoglobin levels (P < 0.05). The risk of depression in the PLR Q4 group was 22.654 times than that of the Q1 group (95%CI, 12.713 to 40.369, P < 0.001). After initially adjusting for age, diabetes, and hypertension, the risk of depression in the PLR Q4 group was 28.272 times than that of the Q1 group (95%CI, 15.390 to 51.937, P < 0.001). After further adjusting for serum creatinine, hemoglobin, albumin, white blood cell count, total cholesterol, monocyte-to-lymphocyte ratio, systemic immune-inflammation index, and neutrophil-to-lymphocyte ratio, the risk of depression in the PLR Q4 group was 10.557 times than that of the Q1 group (95%CI, 3.461 to 32.201, P < 0.001). The ROC curve analysis results indicated that the sensitivity of PLR in predicting depression in MHD patients was 78.0%, the specificity was 73.5%, the optimal threshold was 114.14, and the area under the curve was 0.798 (95%CI, 0.765 to 0.832, P < 0.001).
    Conclusion PLR is an independent risk factor for depression and has a high predictive value for depression in MHD patients.

     

/

返回文章
返回