文献速递流行病学与生物统计临床试验与 GCP

孕激素预防早产的国际合作个体参与者数据荟萃分析

Evaluating Progestogens for Preventing Preterm birth International Collaborative (EPPPIC): meta-analysis of individual participant data from randomised controlled trials.
PubMed 文献 · Lancet · 2021-01-01

高危单胎孕妇中阴道孕酮和17-OHPC均降低34周前早产,口服孕酮证据不足;双胎中阴道孕酮及双胎/三胎中17-OHPC未见降低34周前早产。

要点

  • 高危单胎孕妇阴道孕酮可降低34周前早产(RR 0·78)。
  • 双胎使用阴道孕酮未见降低34周前早产(RR 1·01)。
  • 多胎使用17-OHPC增加34周前胎膜早破风险(RR 1·59)。

核心结论

  • 在高危单胎妊娠中,阴道孕酮和17-OHPC均降低34周前早产,绝对风险降低在宫颈短者更大。
  • 口服孕酮证据不足,不支持其使用。
  • 无证据支持对未选择的雙胎妊娠使用孕激素治疗。

研究设计与样本

  • 系统综述与个体参与者数据荟萃分析,检索MEDLINE、Embase、CINAHL等数据库及试验注册库。
  • 纳入比较阴道孕酮、肌注17-OHPC或口服孕酮与对照或相互比较的随机试验,数据收集完成于2016年7月30日前。
  • 采用一步法广义线性混合模型进行主要荟萃分析,并纳入随机效应以考虑试验间异质性。
  • 研究注册于PROSPERO,CRD42017068299。

主要发现

  • 单胎中阴道孕酮降低34周前早产(9项试验,3769名孕妇;RR 0·78,95% CI 0·68-0·90)。
  • 单胎中17-OHPC降低34周前早产(5项试验,3053名孕妇;RR 0·83,95% CI 0·68-1·01)。
  • 单胎中口服孕酮降低34周前早产(2项试验,181名孕妇;RR 0·60,95% CI 0·40-0·90)。
  • 双胎中阴道孕酮未降低34周前早产(8项试验,2046名孕妇;RR 1·01,95% CI 0·84-1·20)。
  • 双胎或三胎中17-OHPC未降低34周前早产(8项试验,2253名孕妇;RR 1·04,95% CI 0·92-1·18)。

关键数值与推荐

  • 单胎34周前早产:阴道孕酮RR 0·78(95% CI 0·68-0·90),17-OHPC RR 0·83(95% CI 0·68-1·01)。
  • 双胎34周前早产:阴道孕酮RR 1·01(95% CI 0·84-1·20)。
  • 双胎或三胎34周前早产:17-OHPC RR 1·04(95% CI 0·92-1·18)。
  • 多胎34周前胎膜早破:17-OHPC RR 1·59(95% CI 1·15-2·22)。

临床怎么用

  • 高危单胎妊娠可与孕妇共同决策,讨论个体风险、潜在获益与危害及干预可行性。
  • 宫颈短的高危单胎孕妇因基础风险高,绝对风险降低更大,治疗可能最有用。
  • 不应对未选择的雙胎妊娠常规使用孕激素治疗。

局限性

  • 部分结局(如其他出生及新生儿结局)结果一致有利但不确定性较大。
  • 母体并发症可能增加,但证据不确定。
  • 亚组分析对无宫颈短者疗效提出质疑,但未发现一致的交互作用证据。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

原文

BACKGROUND: Preterm birth is a global health priority. Using a progestogen during high-risk pregnancy could reduce preterm birth and adverse neonatal outcomes. METHODS: We did a systematic review of randomised trials comparing vaginal progesterone, intramuscular 17-hydroxyprogesterone caproate (17-OHPC), or oral progesterone with control, or with each other, in asymptomatic women at risk of preterm birth. We identified published and unpublished trials that completed primary data collection before July 30, 2016, (12 months before data collection began), by searching MEDLINE, Embase, CINAHL, the Maternity and Infant Care Database, and relevant trial registers between inception and July 30, 2019. Trials of progestogen to prevent early miscarriage or immediately-threatened preterm birth were excluded. Individual participant data were requested from investigators of eligible trials. Outcomes included preterm birth, early preterm birth, and mid-trimester birth. Adverse neonatal sequelae associated with early births were assessed using a composite of serious neonatal complications, and individually. Adverse maternal outcomes were investigated as a composite and individually. Individual participant data were checked and risk of bias assessed independently by two researchers. Primary meta-analyses used one-stage generalised linear mixed models that incorporated random effects to allow for heterogeneity across trials. This meta-analysis is registered with PROSPERO, CRD42017068299. FINDINGS: Initial searches identified 47 eligible trials. Individual participant data were available for 30 of these trials. An additional trial was later included in a targeted update. Data were therefore available from a total of 31 trials (11 644 women and 16185 offspring). Trials in singleton pregnancies included mostly women with previous spontaneous preterm birth or short cervix. Preterm birth before 34 weeks was reduced in such women who received vaginal progesterone (nine trials, 3769 women; relative risk [RR] 0·78, 95% CI 0·68-0·90), 17-OHPC (five trials, 3053 women; 0·83, 0·68-1·01), and oral progesterone (two trials, 181 women; 0·60, 0·40-0·90). Results for other birth and neonatal outcomes were consistently favourable, but less certain. A possible increase in maternal complications was suggested, but this was uncertain. We identified no consistent evidence of treatment interaction with any participant characteristics examined, although analyses within subpopulations questioned efficacy in women who did not have a short cervix. Trials in multifetal pregnancies mostly included women without additional risk factors. For twins, vaginal progesterone did not reduce preterm birth before 34 weeks (eight trials, 2046 women: RR 1·01, 95% CI 0·84-1·20) nor did 17-OHPC for twins or triplets (eight trials, 2253 women: 1·04, 0·92-1·18). Preterm premature rupture of membranes was increased with 17-OHPC exposure in multifetal gestations (rupture <34 weeks RR 1·59, 95% CI 1·15-2·22), but we found no consistent evidence of benefit or harm for other outcomes with either vaginal progesterone or 17-OHPC. INTERPRETATION: Vaginal progesterone and 17-OHPC both reduced birth before 34 weeks' gestation in high-risk singleton pregnancies. Given increased underlying risk, absolute risk reduction is greater for women with a short cervix, hence treatment might be most useful for these women. Evidence for oral progesterone is insufficient to support its use. Shared decision making with woman with high-risk singleton pregnancies should discuss an individual's risk, potential benefits, harms and practicalities of intervention. Treatment of unselected multifetal pregnancies with a progestogen is not supported by the evidence. FUNDING: Patient-Centered Outcomes Research Institute.

来源:Lancet,查看原文。版权归原作者所有。

引用

  1. Evaluating Progestogens for Preventing Preterm birth International Collaborative (EPPPIC): meta-analysis of individual participant data from randomised controlled trials. Lancet. 2021. PMID: 33773630. doi:10.1016/S0140-6736(21)00217-8

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。