Short cervix is a risk factor for preterm birth. Currently, both international and domestic studies about progesterone's effectiveness are limited to pregnant women at 18-24 weeks gestation. However, multiple studies indicated that cervical length was associated with preterm birth even before 32 weeks of gestation. Therefore, this study expanded the gestational week range to investigate whether progesterone can reduce the rate of preterm birth in singleton pregnant women with a short cervix at 18-32 weeks gestation. Pregnant women who underwent prenatal examination at Peking University First Hospital from January 2016 to August 2020 were prospectively followed. A total of 132 asymptomatic singleton pregnant women at 18-32 weeks gestation with a cervical length <25 mm were ultimately enrolled. According to the method of treatment, the participants were divided into progesterone group (80 patients) and control group (52 patients). The rate of preterm birth (PTB) at different stages was compared between two groups. (1) There was no significant difference in the total preterm birth rate (18.8% vs. 21.2%, RR 0.886[0.442-1.777], p = 0.734). (2) Stratified analysis found that, for pregnant women at <24 weeks gestation, there was a significant difference in the rate of PTB at <32 weeks (2.8% vs. 33.3%, p = 0.021). For women at 24-28weeks gestation, significant difference was not found in the rate of PTB at <37 weeks gestation (25% vs. 42.9%, RR = 0.583[0.186-1.831], p = 0.682), neither for women at after 28 weeks(12.5% vs. 11.1%,1.12[0.27-4.59], p = 1). (3) Vaginal progesterone was not associated with low birth weight (13.8% vs. 19.2%, p = 0.4), or preterm birth-related complications such as respiratory distress syndrome (3.8% vs. 7.7%, p = 0.555), aspiration pneumonia (22.5% vs. 19.2%, p = 0.653) and sepsis (2.5% vs. 7.7%, p = 0.331). For pregnant women with a short cervix at 18-24 weeks gestation, the rate of preterm birth before 32 weeks could be significantly reduced. For women with a short cervix at 24-28 weeks gestation, the rate of preterm birth could be reduced, while there was no significant effect for pregnant women. Further studies with a larger sample size and randomized controlled researches are needed.

目的 探讨维生素D结合蛋白（vitamin D binding protein，VDBP）对先兆早产孕妇发生早产的预测价值。 方法 研究对象为2015年9月1日至2016年1月31日在北京大学第一医院规律产前检查，孕20~35周因先兆早产（或先兆晚期流产）收入院的单胎妊娠孕妇92例。入院时留取宫颈阴道分泌物，用酶联免疫吸附试验（enzyme linked immunosorbent assay，ELISA）检测VDBP水平，半定量检测胎儿纤维连接蛋白（fetal fibronectin，fFN），并分析这些孕妇的妊娠结局。采用χ2检验（或Fisher精确概率法）对数据进行组间比较，绘制短期内（出现症状的3、7和14 d内）分娩的受试者工作特性（receiver operating characteristic，ROC）曲线，分析VDBP预测先兆早产孕妇短期内分娩的价值。 结果 （1）纳入研究的92例孕妇中，57例（62.0%）足月分娩，35例（38.0%）发生早产（包括1例孕25周流产）；其中17例为<孕34周的早产。3、7和14 d内分娩的孕妇分别为11例（11.9%）、13例（14.1%）和16例（17.4%）。早产组新生儿出生体重<2 500 g和<1 500 g的比例[68.6%（24/35）与3.5%（2/57），24.9%（8/35）与0.0%（0/57）；Fisher精确概率法，P值均<0.001]，以及新生儿转NICU的比例[85.7%（30/35）与29.8%（17/57），χ2=27.107，P<0.01]均高于足月组。早产组6例（17.1%）发生呼吸窘迫综合征，2例（5.7%）发生Ⅲ~Ⅳ度脑室内出血，2例（5.7%）发生新生儿败血症，而足月组均未发生。（2）VDBP预测先兆早产孕妇在3、7和14 d内分娩的灵敏度[OR（95%CI）]分别为0.909（0.587~0.998）、0.923（0.640~0.998）和0.938（0.698~0.998），特异度分别为0.889（0.800~0.948）、0.911（0.826~0.964）和0.934（0.853~0.978），阳性预测值分别为0.526（0.289~0.756）、0.632（0.384~0.837）和0.750（0.502~0.917），阴性预测值分别为0.986（0.926~1.000）、0.986（0.925~1.000）和0.986（0.925~1.000）（P值均<0.001）。（3）fFN预测先兆早产孕妇在3、7和14 d内分娩的灵敏度为0.429（0.099~0.816）、0.444（0.137~0.788）和0.455（0.167~0.766），特异度为0.873（0.755~0.947）、0.887（0.770~0.957）和0.902（0.786~0.967），阳性预测值分别为0.300（0.067~0.652）、0.400（0.111~0.755）和0.500（0.173~0.827），阴性预测值分别为0.923（0.815~0.979）、0.904（0.790~0.931）和0.885（0.766~0.956）。 结论 VDBP用于预测先兆早产孕妇在短期内分娩可能具有一定价值，但需要进一步研究加以证实。

Abstract Objective: The objective of this study is to understand the prevalence of short cervical length between 20 and 24 weeks gestation in China and to evaluate the efficacy of micronized progesterone for prolonging gestation in nulliparous patients with a short cervix. Methods: From May 2010 to May 2015, a total of 25 328 asymptomatic women with singleton pregnancies at Peking University First Hospital had their cervical length routinely measured between 20 and 24 weeks of gestation. A cervical length of 25 mm or less was defined as a shortened cervical length. The therapies prescribed include vaginal micronized progesterone capsules (200 mg each night) or bed rest from 20 to 34 weeks of gestation. The primary outcome was spontaneous delivery before 33 weeks. Results: (1) One hundred fourteen women had a cervical length of ≤25 mm (0.45%). (2) Twenty-nine of which with previous spontaneous preterm delivery or late pregnancy loss had cervical cerclage, the remaining 85 women by the use of vaginal progesterone or simply resting activity restriction to prevent preterm birth. (3) In 85 nulliparous women treated by progesterone or bed rest, progesterone use in cervical length between 10 and 20 mm was associated with a statistically significant reduction in the incidence of preterm birth at <33 weeks of gestation (9.5% versus 45.5%, p = 0.02) compared with bed rest. There were no significant differences in cervical length between 20 and 25 mm in their rates of spontaneous preterm delivery at <33 (5.3% versus 3.2%, p = 0.72), <37 (33.3% versus 54.5%, p = 0.25), or <35 weeks (14.3% versus 45.5, p = 0.06) of gestation between vaginal progesterone and bed rest. Conclusion: The rate of short cervical length was less than expected. Vaginal progesterone is efficacious for the prolonging of gestation in women with a cervical length of 10–20 mm in the mid-trimester for a singleton gestation and nulliparous women. For a cervical length of 20–25 mm in the mid trimester, vaginal progesterone compared with bed rest did not prolong pregnancy.