Abstract: Background　The etiology and pathogenesis of schizophrenia have yet not been very clear.There is a theory suggests that schizophrenia roots in the perinatal period，and the subsequent emergence of psychiatric symptoms is the manifestation instead of the onset of the disease.However，the causal relationship between maternal stress during pregnancy and the onset of schizophrenia in offspring has not been determined.Objective　To explore the influence of prenatal maternal stress on the incidence of schizophrenia in offspring.Methods　Databases of Medline，EMBase，PubMed，CNKI，CSTJ，CBM，VIP and Wanfang Data Knowledge Service Platform were systematically searched for cohort studies on prenatal maternal exposure to adverse life events and offspring schizophrenia from inception to May 2019 by two researchers independently according to search strategy.Data were extracted，including the title，first author，stressful event，publication time，number of cases，diagnostic criteria，and pregnancy time，and literature quality was evaluated.RevMan 5.3 was used for meta-analysis.Pooled relative risk（RR）and 95% confidence interval（CI）were used to assess the pooled effect size.Stata 15.0 was used to carry out Egger's test for publication bias.Results　Finally，13 articles were incorporated containing 3 356 707 participants，including 536 142 cases（stress group）and 2 806 687 controls（control group）.Result of meta-analysis suggested that the incidence of offspring schizophrenia in stress group was significantly higher than that of control group〔RR=1.43，95%CI（1.17，1.73），P=0.000 01〕.According to the subgroup analysis of stress duration，the incidence of schizophrenia in offspring of prenatal maternal exposure to short-term stress subgroup was higher than that of the control group〔RR=1.10，95%CI（1.01，1.20），P=0.03〕，and the incidence of schizophrenia in offspring of the prenatal maternal exposure to long-term stress subgroup was also higher than that of the control group〔RR=1.69，95%CI（1.58，1.80），P<0.000 01〕.Egger's test showed there was no significant publication bias（t=0.79，P=0.79）.Conclusion　 Prenatal maternal exposure to stress，regardless of short-term or long-term，may increase the incidence of offspring schizophrenia.

Key words: Schizophrenia, Fertile period, Stress, Prenatal exposure delayed effects, Life experiences, Disaster, Meta-analysis

摘要： 背景　精神分裂症的病因及发病机制一直以来不是十分清楚，发育起源学说认为精神分裂症起始于围生期，后期出现精神症状不是疾病的开始而是疾病中后期的表现，母孕期经历应激与子代精神分裂症发病的因果关系尚未定论。目的　探索母孕期经历应激对子代精神分裂症发病的影响。方法　计算机检索Medline、EMBase、PubMed、中国知网（CNKI）、中国科技期刊数据库、中国生物医学数据库（CBM）、维普网、万方数据知识服务平台等数据库，筛选关于母孕期经历应激事件的队列研究文献。检索时间为数据库建库至2019年5月。双人独立根据检索策略进行文献检索，提取题目、第一作者、应激事件、发表时间、例数、诊断标准、孕期时间等信息并评价文献质量。采用RevMan 5.3软件进行统计学分析，采用RR及其95%CI评估合并效应量。采用State 15.0软件进行发表偏倚Egger's检验。结果　本研究共纳入13篇文献，共3 342 829例研究对象，其中应激组536 142例，对照组2 806 687例。Meta分析结果显示，应激组子代精神分裂症发病率高于对照组〔RR=1.43，95%CI（1.17，1.73），P=0.000 01〕。按照应激持续时间进行亚组分析，结果显示，母孕期经历短期应激时应激组子代精神分裂症发病率高于对照组〔RR=1.10，95%CI（1.01，1.20），P=0.03〕，母孕期经历长期应激时应激组子代精神分裂症发病率高于对照组〔RR=1.69，95%CI（1.58，1.80），P<0.000 01〕。Egger's检验结果显示，无发表偏倚（t=0.79，P=0.79）。结论　母孕期经历应激可能使子代精神分裂症发病率增加，且长期应激与短期应激均可能使子代精神分裂症发病率增加。

关键词: 精神分裂症, 受孕期, 应激, 产前暴露迟发效应, 生活事件, 灾难, Meta 分析