|
摘要：
目的 探讨硬膜外分娩镇痛（labor epidural analgesia, LEA）对意愿阴道分娩产妇产后抑郁症（postpartum depression, PPD）发生的影响。 方法 纳入705例意愿阴道分娩的产妇，宫口开至2~3 cm时由产妇自主选择是否接受LEA。对试产中出现紧急剖宫产指征的产妇，由产科医师做出转剖决策。根据试产期间接受LEA情况和最终分娩方式将产妇分为4组：分娩镇痛阴道分娩组（V/LEA组，302例）、自然阴道分娩组（V/NLEA组，143例）、分娩镇痛试产转剖组（T/LEA组，85例）、自然试产转剖组（T/NLEA组，47例）。收集产妇的人口统计学数据、孕前病史、孕期状况、产前数据（包含心理学量表）和产妇配偶数据，记录分娩过程中最高疼痛数字分级评分法（Numerical Rating Scale, NRS）评分、出血量、家人陪产及新生儿情况，产后第1天记录母婴同室及母婴亲密接触情况、NRS评分和护理满意度，产后第3个月使用爱丁堡产后抑郁量表（the Edinburgh Postnatal Depression Scale, EPDS）评估PPD发生情况。分别探讨LEA对阴道分娩成功和转剖产妇PPD发生的影响，采用Logistic回归对PPD的相关危险因素进行分析，绘制森林图，采用受试者工作特征（receiver operating characteristic, ROC）曲线下面积（area under curve, AUC）评价模型拟合效果。 结果 4组产妇年龄、初产妇占比、产时出血量、最高NRS评分差异有统计学意义（P<0.05），V/LEA组和V/NLEA组产妇产程时长、会阴切开率差异有统计学意义（P<0.05），T/LEA组和T/NLEA组产妇转剖前宫口开度差异有统计学意义（P<0.05），其他指标差异无统计学意义（P>0.05）。产后第3个月随访结果显示产妇PPD总发生率为28.4%。在二次分析和调整组间混杂因素后，未发现接受LEA与PPD发生率降低相关［校正后优势比（odds ratio, OR） 0.774，95%CI 0.469~1.276，P=0.315］。Logistic回归分析显示，意愿阴道分娩产妇PPD的危险因素有：孕期慢性疼痛、产前EPDS得分较高、Zung氏焦虑自评量表（the Zung Self‑Rating Anxiety Scale, SAS）得分较高、社会支持评定量表（the Social Support Rating Scale, SSRS）得分较低、无家人陪产、产后第1天NRS评分较高。将上述因素拟合成逻辑回归模型，绘制ROC曲线，AUC=0.881，95%CI 0.851~0.911。 结论 接受LEA并不能降低意愿经阴道分娩产妇PPD的发生风险。孕期慢性疼痛，较高的EPDS、SAS得分，较低的SSRS得分，产时无家人陪产，产后第1天NRS评分较高等是PPD发生的独立危险因素。
|
|
Abstract: Objective To explore the effect of labor epidural analgesia (LEA) on the occurrence of postpartum depression (PPD) in women intending to deliver vaginally. Methods A total of 705 women who intended to deliver vaginally were enrolled. At 2‒3 cm cervical dilation, they were allowed to choose whether to accept LEA or not. For those with emergency cesarean section (EmCS) indications during trial delivery, the obstetrician made the decision whether to convert to cesarean section (CS). The parturient women were divided into four groups according to the use of LEA during childbirth and the final delivery method: a vaginal delivery with LEA group (group V/LEA, n=302), a natural vaginal delivery group (group V/NLEA, n=143), a cesarean delivery after failed attempt of trial delivery with LEA group (group T/LEA, n=85), and a cesarean delivery after failed attempt of natural vaginal delivery (group T/NLEA, n=47). Their demographic data, pre‑pregnancy history, pregnant conditions, prenatal data (including psychological scales) and maternal spouse data were collected. The highest Numerical Rating Scale (NRS) score during delivery, blood loss volume, accompaniment during childbirth and newborn conditions were recorded. On the first day after delivery, rooming‑in for new mother and infant, close contact between mother and infant, 1‑day NRS score and nursing satisfaction were recorded. The Edinburgh Postpartum Depression Scale (EPDS) was used to assess the occurrence of PPD at the third month postpartum. To explore the effect of LEA on the occurrence of PPD for the parturient women with successful vaginal delivery and cesarean delivery after failed attempt of trial delivery. The related risk factors of PPD were analyzed by logistic regression. The area under receiver operating characteristic (ROC) curve was used to fit the effects. Results There were statistical differences in age, the proportion of primiparas, blood loss volume, and the highest NRS score among the four groups (P<0.05); the labor time and episiotomy rate were statistically different between group V/LEA and group V/NLEA (P<0.05). There were statistical differences in the degree of cervical dilation before cesarean delivery between group T/LEA and group T/NLEA (P<0.05), while there were no statistical differences in other indicators (P>0.05). The three‑month follow‑up results showed that the total incidence of PPD at the third month postpartum was 28.4%. After secondary analysis and adjusting for confounding factors between groups, no association was found between the use of LEA and PPD ［adjusted odds ratio (OR) 0.774, 95%CI (0.469, 1.276), P=0.315］. Logistic regression analysis showed that the risk factors of PPD in women intending to vaginal delivery included chronic pain during pregnancy, high prenatal EPDS scores, high Zung's Self‑Rating Anxiety Scale (SAS) scores, low Social Support Rating Scale (SSRS) scored, no accompaniment during childbirth, and high 1‑day NRS score after delivery (P<0.05). These factors were adopted to fit a logistic regression model and a receiver operating characteristic (ROC) curve was plotted, with the area under curve (AUC) of 0.881, 95%CI (0.851, 0.911). Conclusions The use of LEA does not reduce the risk of PPD in women intending to deliver vaginally. Chronic pain during pregnancy, high EPDS and SAS scores, low SSRS scores, no accompaniment during childbirth, and high 1‑day NRS score are the independent risk factors of PPD.
|
