作者：黄小妹，胡填，叶玉焊，徐吟文，古剑雄

单位：

1.广东医科大学附属医院康复医学科2.广东医科大学医学技术学院3.中山大学附属第三医院儿童发育行为中心

单位（英文）：

1.Department of Rehabilitation Medicine, Affiliated Hospital of Guangdong Medical University, Zhanjiang 524000, China 2.School of Medical Technology, Guangdong Medical University, Zhanjiang 524000, China 3.Child Developmental and Behavior Center, the Third Affiliated Hospital of Sun Yat-sen University, Guangzhou 510000, China

关键词：

卒中; 卒中后抑郁; 系统性炎症反应指数; 倾向评分匹配; 逆概率处理加权法;

关键词（英文）：

Stroke; Post-stroke depression; Systemic inflammatory response index; Propensity score matching; Inverse probability of treatment weighting

中图分类号：

DOI：

10.12114/j.issn.1008-5971.2023.00.161

基金项目：

国家自然科学基金重点项目（61431007）；湛江市非资助科技攻关计划项目（2022B01004）；广东医科大学青年培育基金项目（GDMUQ2021034）

摘要：

目的 基于倾向评分匹配(PSM)和逆概率处理加权法(IPTW)探讨系统性炎症反应指数(SIRI)与卒中后抑郁(PSD)的关系。方法 选取2007—2018年美国国家健康和营养检查调查(NHANES)数据库中自我报告的年龄>20岁的脑卒中患者1 016例为研究对象，收集患者的临床资料、计算SIRI。采用患者健康问卷(PHQ-9)评估脑卒中患者的抑郁症情况。采用PSM和IPTW控制混杂因素，采用标准化均数差(SMD)评估低SIRI组与高SIRI组各协变量的均衡性，以SMD<0.1为组间均衡性较好；采用多因素Logistic回归分析探讨SIRI与PSD的关系，采用亚组分析探讨不同特征患者中SIRI与PSD的关系。结果 1 016例脑卒中患者中，189例(18.60%)合并抑郁症(PSD组)，827例(81.40%)未合并抑郁症(非PSD组)。PSD组与非PSD组性别、年龄、BMI、婚姻状况、吸烟史、心血管疾病史比较，差异有统计学意义(P<0.05)。患者的SIRI为1.2×109(0.8×109,1.9×109)。根据SIRI诊断PSD的最佳截断值将患者分为低SIRI组(SIRI<1.5×109,n=627)和高SIRI组(SIRI≥1.5×109,n=389)，采用PSM处理后低SIRI组和高SIRI组各389例，采用IPTW处理后低SIRI组和高SIRI组各1 015例。匹配前低SIRI组与高SIRI组性别、年龄、受教育程度、吸烟史、高血压史、糖尿病史、心血管疾病史比较，差异有统计学意义(P<0.05)。PSM、IPTW处理后低SIRI组与高SIRI组性别、年龄、BMI、受教育程度、婚姻状况、吸烟史、饮酒史、高血压史、糖尿病史、心血管疾病史比较，差异无统计学意义(P>0.05)。匹配前低SIRI组和高SIRI组性别、年龄、BMI、受教育程度、婚姻状况、吸烟史、饮酒史、高血压史、糖尿病史、心血管疾病史SMD>0.1;PSM处理后低SIRI组和高SIRI组受教育程度、心血管疾病史SMD=0.1，性别、年龄、BMI、婚姻状况、吸烟史、饮酒史、高血压史、糖尿病史SMD<0.1;IPTW处理后低SIRI组和高SIRI组性别、年龄、BMI、受教育程度、婚姻状况、吸烟史、饮酒史、高血压史、糖尿病史、心血管疾病史SMD<0.1。多因素Logistic回归分析结果显示，匹配前SIRI与PSD不相关[OR=1.298,95%CI(0.941,1.787),P=0.111];PSM处理后，SIRI≥1.5×109是PSD的独立危险因素[OR=1.556,95%CI(1.075,2.265),P=0.020];IPTW处理后，SIRI≥1.5×109是PSD的独立危险因素[OR=1.444,95%CI(1.158,1.805),P<0.001]。亚组分析结果显示，在男性、年龄<60岁、BMI≥30 kg/m2、有高血压史、有心血管疾病史患者中，SIRI≥1.5×109是PSD的独立危险因素(P<0.05)。交互作用检验结果显示，年龄可能是影响SIRI与PSD关系的因素(P<0.05)。结论 SIRI≥1.5×109是PSD的独立危险因素，尤其在男性、年龄<60岁、BMI≥30 kg/m2、有高血压史、有心血管疾病史的脑卒中患者中。SIRI可作为PSD的辅助诊断标志物，以帮助临床医生早期对PSD做出诊断。

英文摘要：

Objective To explore the relationship between system inflammation response index (SIRI) and poststroke depression (PSD) using propensity score matching (PSM) and inverse probability of treatment weighting (IPTW) . Methods A total of 1 016 self-reported stroke patients aged > 20 years in the National Health and Nutrition Examination Survey (NHANES) database from 2007 to 2018 were selected as subjects, the clinical data of patients were collected, and SIRI was calculated. Patient Health Questionnaire-9 (PHQ-9) was used to assess depression in stroke patients. The PSM and IPTW were used to control for confounders, standardized mean difference (SMD) was used to evaluate the balance between the low SIRI group and the high SIRI group, with SMD < 0.1 indicating good inter group balance; multivariate Logistic regression analysis was used to analyze the relationship between SIRI and PSD, and subgroup analysis was used to analyze the relationship between SIRI and PSD in patients with different characteristics. Results Among 1 016 stroke patients, 189 (18.60%) had depression (PSD group) and 827 (81.40%) did not have depression (non-PSD group) . There were significant differences in gender, age, BMI, marital status, history of smoking, and history of cardiovascular disease between PSD group and non-PSD group (P < 0.05) . The SIRI of patients was 1.2×109 (0.8×109 , 1.9×109 ) . Patients were divided into the low SIRI group (SIRI < 1.5×109 , n=627) and the high SIRI group (SIRI ≥ 1.5×109 , n=389) according to the best cut-off value for the diagnosis of PSD by SIRI, 389 cases in the low SIRI group and 389 cases in the high SIRI group after PSM, and 1 015 cases in the low SIRI group and 1 015 cases in the high SIRI group after IPTW. There were significant differences in gender, age, education level, history of smoking, history of hypertension, history of diabetes, and history of cardiovascular disease between the low SIRI group and the high SIRI group before matching (P < 0.05) . There was no statistically significant difference in gender, age, BMI, education level, marital status, history of smoking, history of drinking, history of hypertension, history of diabetes, and history of cardiovascular disease between the low SIRI group and the high SIRI group after PSM and IPTW (P > 0.05) . The gender, age, BMI, education level, marital status, history of smoking, history of drinking, history of hypertension, history of diabetes, and history of cardiovascular disease SMD > 0.1 between the low SIRI group and the high SIRI group before matching; the education level, and history of cardiovascular disease SMD = 0.1, and the gender, age, BMI, marital status, history of smoking, history of drinking, history of hypertension, history of diabetes SMD < 0.1 between the low SIRI group and the high SIRI group after PSM; the gender, age, BMI, education level, marital status, history of smoking, history of drinking, history of hypertension, history of diabetes, and history of cardiovascular disease SMD < 0.1 between the low SIRI group and the high SIRI group after IPTW. Multivariate Logistic regression analysis showed that SIRI was not associated with PSD before matching [OR=1.298, 95%CI (0.941, 1.787) , P=0.111] ; SIRI ≥ 1.5×109 was an independent risk factor for PSD after PSM [OR=1.556, 95%CI (1.075, 2.265) , P=0.020] ; SIRI ≥ 1.5×109 was an independent risk factor for PSD after IPTW [OR=1.444, 95%CI (1.158, 1.805) , P < 0.001] . The results of subgroup analysis showed that SIRI ≥ 1.5×109 was an independent risk factor for PSD in the males and patients with age < 60 years, BMI ≥ 30 kg/m2 , history of hypertension, and history of cardiovascular disease. The results of interaction tests showed that age may be a factor influencing the relationship between SIRI and PSD (P < 0.05) . Conclusion SIRI ≥ 1.5×109 is an independent risk factor for PSD, especially in male stroke patients and patients with age < 60 years, BMI ≥ 30 kg/m2 , history of hypertension, and history of cardiovascular disease. SIRI can be used as an auxiliary diagnostic marker of PSD to help clinicians to make a early diagnosis of PSD.

参考文献：