目的:探讨抑郁症躯体症状与皮质醇(cortisol，COR)节律改变及其他免疫代谢相关指标之间的关系，从内分泌及免疫调节角度了解其发生机制。方法:采用病例对照研究，以符合DSM-5诊断标准的抑郁症住院患者为研究对象，根据患者健康问卷(patient health questionnaire，PHQ-15)，将PHQ-15≥10分定为伴躯体症状的抑郁症组(S-MDD组)，共纳入73例；将PHQ-15<5分定为不伴躯体症状的抑郁症组(NS-MDD组)，共纳入70例。检测患者同日8∶00、16∶00及24∶00血浆皮质醇(COR8、COR16、COR24)及同日8∶00时的C反应蛋白(C-reactive protein，CRP)、白介素6(interleukin-6，IL-6)水平，血清尿酸(UA)、血糖(GLU)、血脂(TC、TG、HDL、LDL)水平；采用独立样本n t检验、非参数检验、卡方检验、重复测量方差分析、协方差分析及多因素Logistic回归分析进行数据处理。n 结果:①采用重复测量方差分析，发现COR水平的时间效应、组间效应及时间与组间的交互效应均有统计学意义(n P<0.05)。以年龄为协变量，采用协方差分析发现，S-MDD组的COR16、皮质醇分泌总量(total cortisol output/area under the curve，AUC)及COR8-16斜率[(90.50±40.57)μg/L、(1 425.12±564.78)、(-6.43±5.76)]大于NS-MDD组[(68.74±31.51)μg/L、(1 251.57±456.61)、(-8.77±5.48)]，差异有统计学意义(n F＝8.971，4.320，8.731，n P<0.05)。②S-MDD组CRP水平[(1.41±1.06)mg/L]高于NS-MDD组[(0.61±0.53)mg/L]，差异有统计学意义(n F＝25.436，n P<0.05)。S-MDD组高CRP(CRP≥1 mg/L)患者人数比例(58%)高于NS-MDD组(23%)，差异有统计学意义(χn 2＝17.824，n P<0.01)。③多因素Logistic回归分析发现，CRP(n OR＝4.953，95%n CI：2.407～10.193)、COR8-16(n OR＝3.451，95%n CI：1.380～8.633)是抑郁症躯体症状的主要危险因素。n 结论:日间皮质醇节律紊乱及高CRP水平可能是抑郁症患者躯体症状产生的神经生物学基础。“,”Objective:To explore the relationship between somatic symptoms of major depressive disorder(MDD)and cortisol(COR) rhythm, C-reactive protein(CRP) and other immune-metabolism-related indicators, and understand its mechanism from the perspective of endocrine and immune regulation.Methods:A case-control study was conducted in hospitalized patients with MDD who met DSM-5 diagnostic criteria.According to the Patient Health Questionnaire (PHQ-15), PHQ-15 ≥10 were classified as the somatic major depressive disorder group(S-MDD group) and 73 patients were enrolled.PHQ-15 <5 was classified as the non-somatic depressive disorder group (NS-MDD group) and 70 patients were enrolled.Plasma cortisol (COR8, COR16 and COR24) levels were measured at 8∶00, 16∶00 and 24∶00 on the same day, plasma CRP and interleukin-6 (IL-6) level, serum uric acid (UA), blood glucose (GLU), blood lipid (TC, TG, HDL, LDL) level were detected at 8∶00.Independent sample n t test, non-parametric test, chi-square test, repeated ANOVA, covariance analysis, and multivariate Logistic regression were used for statistical analysis.n Results:①Time effect, grouping effect and the interaction effect of the time and grouping in the level of COR were statistically significant (n P<0.05). Covariance analysis excluded age as an influential factor, COR16, AUC(total cortisol output/area under the curve, AUC) and COR8-16 in S-MDD group ((90.50±40.57)μg/L, (1 425.12±564.78), (-6.43±5.76))were higher than those in NS-MDD group((68.74±31.51)μg/L, (1 251.57±456.61), (-8.77±5.48)), and the difference was statistically significant (n F=8.971, 4.320, 8.731, n P<0.05). ②CRP in S-MDD group ((1.41±1.06)mg/L) were higher than that in NS-MDD group((0.61±0.53)mg/L), and the difference was statistically significant (n F=25.436, n P<0.05). The proportion of patients with higher CRP level(CRP≥1 mg/L) in S-MDD group(58%) was higher than that in NS-MDD group(23%), and the difference was statistically significant(χn 2=17.824, n P<0.01). ③Multivariate logistic regression analysis found that CRP (n OR=4.953, 95%n CI: 2.407-10.193), COR8-16 (n OR=3.451, 95%n CI: 1.380-8.633) were main risk factors of somatic symptoms of MDD (n P<0.05).n Conclusion:Cortisol rhythm disturbance and high CRP level may be the biological basis of somatic symptoms in patients with MDD.