Double‐blind randomized trial of progesterone to prevent preterm birth in second‐trimester bleeding

Accepted Article DR RAED SALIM (Orcid ID : 0000-0002-3603-8452) Article type : Original Research Article Double-blind randomized trial of progesterone to prevent preterm birth in second trimester bleeding Raed Salim1,2 , Marwan Hakim3,4 , Noah Zafran1,2 , Zohar Nachum1,2 , Shabtai Romano1,2 , Gali Garmi1,2 1 Department of Obstetrics and Gynecology, Emek Medical Center, Afula, Israel 2 Rappaport Faculty of Medicine, Technion, Haifa, Israel 3 Nazareth Hospital EMMS, Nazareth, Israel 4 Faculty of Medicine in the Galilee, Bar Ilan University, Ramat Gan, Israel Corresponding author Raed Salim Department of Obstetrics and Gynecology, Emek Medical Center, Afula 18101, Israel E. Mail: salim_ra@clalit.org.il Conflict of Interest None Funding The described project was supported by an award from the Legacy Heritage Fund Limited (HLMH, Inc.). Project no. 4 – 2010. ABSTRACT Introduction: Second-trimester vaginal bleeding increases the risk for spontaneous preterm birth. We aimed to examine the efficacy of vaginal progesterone to reduce preterm birth rate in women with second-trimester vaginal bleeding. Material and methods: Two-center, doubleThis article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. Please cite this article as doi: 10.1111/aogs.13641 This article is protected by copyright. All rights reserved. Accepted Article blind, placebo-controlled trial involving pregnant women with second-trimester vaginal bleeding. Women with documented uterine bleeding were randomly assigned in a 1:1 ratio to receive 200 mg of micronized vaginal progesterone or placebo once daily at 16 to 26 weeks until 36 weeks of gestation. Women who had prior preterm birth or short cervix diagnosed before recruitment were not eligible. The primary outcome was spontaneous delivery <37 weeks. Clinical Trial Registration: clinicaltrials.gov Identifier: NCT01269450. Results: Between March 2011 and January 2017, 128 women gave consent and were randomized; 16 withdrew consent and 3 had a second-trimester termination of pregnancy. The final analysis included 109 women: 60 in the progesterone group and 49 in the placebo group. Demographic and obstetric characteristics did not differ between the groups. Primary outcome occurred in 19 (31.7%) and 12 (24.5%) in the progesterone and placebo groups, respectively (OR 1.32; 95% CI 0.55 to 3.16; p=0.53). The proportion of births <34 weeks was similar between the groups (OR 1.19; 95% CI 0.47 to 3.02; p=0.72), as were the survival curves from randomization to delivery (hazard ratio, 1.24; 95% CI, 0.60 to 2.56; p=0.57). There were no significant differences in neonatal morbidities between the groups. The study was ended prematurely because of slow recruitment. Conclusions: Antepartum vaginal progesterone does not seem to reduce the incidence of preterm birth in women with second-trimester bleeding. Key words Preterm birth; Second trimester bleeding; Vaginal progesterone Key Message The findings do not provide support for progesterone administration to women with second trimester bleeding in order to reduce spontaneous preterm birth rate. INTRODUCTION Preterm birth is a major contributor to perinatal mortality and morbidity and affects up to 12% of births in developed countries.1 Women with singleton gestations at great risk for preterm birth are those with prior preterm birth, short cervical length, and antepartum bleeding, particularly when this occurred in the second trimester.1 Data suggest that progesterone is an effective intervention as a primary preventive treatment in reducing the risk of preterm birth when given to women who had a second-trimester short cervix or a previous preterm birth.2,3 Additionally, progesterone supplementation has been reported to decrease the incidence of recurrent miscarriages and seems to be safe when given in the first trimester.4 This article is protected by copyright. All rights reserved. Accepted Article Nevertheless, the efficacy of progesterone was challenged after publication of the OPPTIMUM study, which reported no significant association between vaginal progesterone use and reduction of preterm birth risk (odds ratio (OR) 0.86; 95% confidence interval (CI) 0.61 to 1.22).5 Antepartum bleeding complicates about 25% of all pregnancies and occurs in the second trimester in up to 2.5%.6 The risk of preterm birth is 30% or more when the bleeding occurs in the second trimester.7-9 The mechanism that leads to preterm birth is not clear. It has been assumed that bleeding and thrombin formation may lead to the formation of proteolytic enzymes and cytokines that lead to rupture of membranes and preterm birth.10-12 Progesterone has been reported to reduce in vitro the production of cytokines induced by thrombin.12 It is unknown whether progesterone in vivo affects this risk of preterm birth among women with second trimester bleeding. We aimed in the current study to examine the efficacy of vaginal progesterone to reduce the risk of spontaneous preterm birth in women with secondtrimester vaginal bleeding. MATERIAL AND METHODS Study Design This randomized, double blind placebo controlled trial was conducted from March 2011 through January 2017. During the study period four maternity units in four teaching medical institutions in Israel joined the study but only two (Emek Medical Center, Afula and The Nazareth Hospital, E.M.M.S, Nazareth) were able to recruit women for the trial. Pregnant women admitted with bleeding between 13 to 26 weeks were given the option to participate in the trial. Eligible women were randomly assigned in a 1:1 ratio to receive once daily 200 mg of micronized vaginal progesterone or placebo at 16 weeks or once bleeding was diagnosed, but before 26 weeks. The trial drug was continued until 36 weeks of gestation or until delivery, whichever came first. Inclusion criteria included vaginal bleeding that originated from the uterus and confirmed by vaginal examination, singleton pregnancy, normal clotting tests, hemodynamic stability, and gestational age between 13 and 26 weeks at first episode of bleeding. Women were excluded in case of vaginal bleeding of cervical origin, rupture of membranes at time of presentation or evidence of preterm labor, fetal malformations incompatible with life, uterine malformations This article is protected by copyright. All rights reserved. Accepted Article and maternal conditions that include liver disease, breast cancer, active thromboembolic state, uncontrolled hypertension, or uncontrolled diabetes with or without involvement of target organs, heart failure, renal failure, and psychiatric disorders. One woman with epilepsy had a worsening in epilepsy attacks during the study period and for that reason epilepsy was subsequently considered an exclusion criterion. Women who had an indication for progesterone treatment, i.e., prior spontaneous preterm birth or short cervix diagnosed before recruitment, were not eligible. Cervical length measurement was not part of the study protocol. The decision for cervical length measurement was determined by the woman's provider. Women with confirmed vaginal bleeding of uterine origin verified by speculum examination at admission were invited to participate in this study. Bleeding was categorized as light or heavy (requires pad usage), single or repeated episode (>one episode). Fetal viability, confirmation of pregnancy dating, placental location, and sonographic signs of placental abruption were examined with ultrasound at recruitment. Randomization and follow-up Each study participant was given a blister pack that contained either 200 mg capsules of micronized progesterone (Utrogestan; Besins Healthcare International, Belgium) or identicalappearing capsules of placebo containing safflower oil (Medicaps). Each blister pack contained 35 capsules. The drug and placebo were purchased from the company, which provided no financial support and had no involvement in study design, data collection and analysis, study interpretation, or the decision to publish. The placebo tablets were identical to the Utrogestan tablets with respect to size, thickness, and appearance. The blisters were packaged by DHP Clinical Trial Supplies and labeled as trial drug one or trial drug two and then shipped to Emek Medical Center, the responsible center. The blisters were received and stored by the pharmacist in charge according to the manufacturer’s storage instructions. The site pharmacist dispensed the trial drug to the local coordinator who taught the study participants how to self-administer the study drug and explained the trial procedures. This article is protected by copyright. All rights reserved. Accepted Article Each woman was instructed to introduce one capsule into her vagina every night before going to sleep from 16 to 36 weeks of gestation or until delivery, whichever came first. Women’s providers received written document regarding the women’s participation in the study. Follow-up visits for maternal and fetal assessment were carried out monthly until 36 weeks of gestation or until delivery, whichever came first. At each visit, women were asked regarding bleeding episodes, side effects of the drug, and adherence that was verified by counting the remaining capsules of the prior pack. At each visit a new blister pack was provided. The investigators, participants, clinicians, women’s health care providers, and data analysis individuals were unaware of the treatment assignments. Information on the characteristics of the women, including demographic data, and obstetrical and medical histories, was obtained from the patients at recruitment and entered into a computer database. Data on pregnancy outcomes were obtained from the hospital labor and neonatal records. The obstetrical records of all women delivering before 37 weeks were examined to determine whether the delivery was medically indicated or spontaneous. Spontaneous deliveries included those with spontaneous onset of labor and those with rupture of membranes before labor. Randomization was performed using a computer randomization sequence generation program stratified according to participating center. Eligible women who signed an informed consent were randomly assigned in a 1:1 ratio to receive trial drug one or two. The randomization sequence was kept in a closed study box and the sequence was concealed until intervention was assigned. Women were allocated to randomization code numbers in chronological order. The allocation list was stored at Emek Medical Center and communication with each center was performed by telephone. Outcome measures The primary outcome measure was spontaneous delivery before 37 weeks of gestation. The secondary outcome measures were neonatal mortality and morbidity that includes any of the following: respiratory distress syndrome, intraventricular hemorrhage, necrotizing enterocolitis, offspring infection, ventilation with positive airway pressure or intubation, phototherapy, and admission to a neonatal intensive care unit. This article is protected by copyright. All rights reserved. Accepted Article All outcomes were analyzed before the randomization code of the trial was broken. The study was approved by the local ethics committees of the two participating hospitals, and by the Israel Ministry of Health. Each eligible woman signed an informed consent. Quality control of screening, handling of data, supervision of the stored trial drug, and verification of compliance to protocol was performed by a local data monitoring committee. The funding organization and the company that supplied and distributed the trial drug had no role in the trial design, the collection, analysis, or interpretation of the data, the writing of the manuscript, or the decision to submit the manuscript for publication. Sample size We expected that progesterone treatment will decrease the rate of spontaneous preterm birth (<37 weeks) among women with second trimester bleeding. To detect a reduction in the incidence of spontaneous preterm birth from 30%,8 to 15% in the placebo and progesterone groups, respectively, 134 women in each group were needed. The analysis was performed according to the intention-to-treat principle. Statistical analyses Women’s characteristics at the recruiting hospitals were compared by the chi-squared test or Fisher’s exact test where appropriate for categorical variables and by t-test for continuous variables. Demographic and obstetric characteristics of women of the two treatment groups were compared by logistic regression for categorical variables and by ANCOVA for continuous variables adjusting for recruitment center (hospital). Test of group differences in the outcomes were made by logistic regression for categorical outcomes and by linear regression for the continuous variables adjusting for recruitment center. Survival times in the two groups were computed via Cox regression analysis adjusting for hospital for the primary outcome, i.e., spontaneous delivery before 37 weeks. The study was conducted in accordance with the CONSORT (Consoli-dated Standards of Reporting Trials) guidelines.13 Ethical Approval The study was approved by the local institutional board - IRB no: 0023-08-EMC; Date of approval: 29.11.2010, Emek Medical Center, Afula, Israel. Clinical Trial Registration: clinicaltrials.gov Identifier: NCT01269450. This article is protected by copyright. All rights reserved. Accepted Article RESULTS The trial started at Emek Medical Center, Afula, Israel, in March 2011 followed by the Nazareth Hospital, E.M.M.S, Nazareth. Another two centers were not able to recruit women for the trial due to rejection of eligible women to participate. Recruitment was stopped in January 2017. The duration of this trial was remarkably extended compared with the original plan. For that reason recruitment was ended prematurely because of slow recruitment in six years despite not achieving the planned sample size. The most obvious obstacle to recruiting women was related to the use of a placebo arm. During the study period, 128 women gave consent and were randomized; 16 withdrew consent and three women had a second-trimester termination of pregnancy (trisomy 21, antenatal Cytomegalovirus infection, and the third due to second trimester severe oligohydramnios). Overall, 109 women were included in the final analysis, 60 in the progesterone group and 49 in the placebo group (Figure 1). There were no statistically significant differences in the demographic and obstetric characteristics of women between the Utrogestan and the placebo groups (fixed effects of center, Table 1). Overall, 31 (28.4%) women who were included in this trial had spontaneous preterm birth. The primary outcome, i.e., spontaneous delivery before 37 weeks, occurred in 19 (31.7%) and 12 (24.5%) in the progesterone and placebo groups, respectively (OR, 1.32; 95% CI, 0.55 to 3.16; p = 0.53) (Table 2). The risk of spontaneous preterm birth in the two groups was assessed using Cox regression analysis (Figure 2). The cumulative percentage of women who did not deliver spontaneously before 37 weeks, did not differ significantly between the progesterone and the placebo groups (hazard ratio, 1.24; 95% CI, 0.60 to 2.56; p = 0.57). In addition, survival analysis revealed no difference between the two groups for the primary outcome (mean days placebo: 140.8 vs. study: 134.3; log rank chi-squared=0.49, p = 0.49) or birth less than 24 weeks [median days placebo (7 women): 144.9 vs. study (9 women): 127.4; log rank chi-squared=2.17, p = 0.14]. The proportion of births < 34 weeks was also similar between the groups (p = 0.72). Of all 109 women, six stopped the trial drug during pregnancy and vaginal Utrogestan was initiated due to short cervix diagnosed by the women’s providers, five in the placebo group and one in the active drug group. This article is protected by copyright. All rights reserved. Accepted Article Table 3 presents the neonatal outcomes in the two trial groups. The analysis included only women who delivered at 24 weeks or more. There were no significant differences in neonatal morbidities between the groups. One neonate who was born at 27 weeks in the Utrogestan group, died at age seven days due to complications related to necrotizing enterocolitis. Nine (15%) and seven (14.2%) women in the Utrogestan and placebo groups respectively delivered spontaneously before 24 weeks. Mean gestational age was 18.6 (±2.0) and 19.4 (±3.0) weeks respectively. None survived after delivery. DISCUSSION The results of this randomized trial demonstrate that in women with second trimester bleeding, daily vaginal administration of progesterone, 200 mg, starting at 16 to 26 weeks, up until 36 weeks of gestation did not affect the rate of spontaneous preterm birth. Additionally, there was no significant reduction in perinatal mortality or neonatal morbidity. The overall incidence of spontaneous preterm birth found in the current trial among women with second trimester bleeding was comparable to a risk of 30% that had been reported previously in the literature.7,8 The main limitation of our study was related to terminating the recruitment phase without achieving the planned sample size. Recruiting women in an obstetrics context has been reported as being very difficult.14 In fact, in this context, the idea of being randomized to a group, so that chance would decide instead of a professional’s decision based on evidence and standard practice, is hard to accept.14 The difficulty is even more stressed in double blinded trials with a placebo arm.15 Strengths of the current trial are related to its design. Neither the patients nor the health care provider monitoring the outcome knew which women were receiving which treatment until the trial was stopped and the data was analyzed. Furthermore, including more than one center increases the generalizability of the trial. In addition, we searched the PubMed English language literature from inception through February 2018 using the MeSH headings that included combinations of the terms progesterone ("progesterone", "progestins", "progestogen", "progestational agent"), preterm birth ("preterm", "premature"), bleeding ("bleeding", "hemorrhage"), and randomized ("randomized", "randomized", "randomize"), and found that there were no randomized trials that examined the effect of progesterone among women with second trimester bleeding. This article is protected by copyright. All rights reserved. Accepted Article Progesterone supplementation has been proven in reducing the risk of spontaneous preterm birth among a precise group of women at risk for preterm birth. The American College of Obstetricians and Gynecologists and the Society for Maternal-Fetal Medicine recommend the use of intramuscular 17-alpha hydroxyprogesterone caproate to women with a singleton gestation and a prior spontaneous preterm singleton birth to reduce the risk of recurrent spontaneous preterm birth. Vaginal progesterone is recommended as a management option to reduce the risk of spontaneous preterm birth in asymptomatic women with a singleton gestation without a prior preterm birth with an incidentally identified short cervix before or at 24 weeks of gestation.16,17 Second trimester bleeding increases the risk of spontaneous preterm birth two to four times and may lead to other adverse pregnancy outcomes including higher rates of low birth weight, low Apgar score, operative delivery, and peripartum mortality.7-9,18 Although antepartum bleeding and subsequent preterm birth are typically related to placenta previa and placental abruption, the majority of reasons have been related to unknown origin. Bleeding of unknown origin has also been associated with high preterm birth and perinatal mortality rates.19-21 To improve pregnancy outcome, increased antenatal monitoring had been advocated with or without hospitalization.21 Additionally, avoidance of activities that may stimulate uterine contractions and/or cervical irritation, such as vigorous exercise and intercourse, has also been prescribed.22 These suggestions are based mostly on expert opinion, and evidence of their benefit on pregnancy outcome is uncertain. It has been assumed that bleeding and thrombin formation may lead to the formation of proteolytic enzymes and cytokines that lead to rupture of membranes and preterm birth.10-12 Additionally, thrombin may act as a direct uterotonic agent, initiate functional progesterone withdrawal, and subsequently lead to uterine contractions and preterm birth.11,23-27 Additionally, progesterone has been reported to reduce in vitro the production of cytokines induced by thrombin.12 Notwithstanding, vaginal progesterone use in the current trial did not lead to reduction in spontaneous preterm birth rate or to improvement in perinatal outcome. Failure to improve outcome may be related to the futility of vaginal progesterone in reducing the risk of spontaneous preterm birth in women with second trimester bleeding. Alternatively, premature termination of the trial leaves the possibility that progesterone may be effective had the trial continued till the planned sample size. Different dose or route of administration may be other reasonable alterations that may affect the outcome. This article is protected by copyright. All rights reserved. Accepted Article CONCLUSION Approximately 30% of women experiencing second trimester bleeding may have spontaneous preterm births. Currently no effective treatment has been reported to affect pregnancy outcome in this situation. The findings of the current trial do not provide support for prophylactic administration of 200 mg vaginal progesterone to women with second trimester bleeding in order to reduce the incidence of spontaneous preterm birth. Acknowledgments The authors thank Professor Eliezer Shalev, MD, Rappaport Faculty of Medicine, Technion, Haifa, Israel, for his advice and instructions; Mgr. Samira Younis, BSc, Pharm, MHA, Deputy Pharmacy Director, Manager of Clinical Trials in Pharmacy, Emek Medical Center, for the professional practice and ensuring that the drug trial supervision meets institutional medication guidelines, and adheres to all regulations and standards; Mrs Marlene Khammar Hakim, BSc, Pharm, Pharmacy Director, Manager of Clinical Trials in Pharmacy, Nazareth Hospital EMMS and Paula S. Herer, biostatistician, MSc., MPH, for statistical guidance and assistance. This article is protected by copyright. All rights reserved. Accepted Article References 1. Society for Maternal Fetal Medicine Publications Committee. ACOG Committee Opinion number 419 October 2008 (replaces no. 291, November 2003). Use of progesterone to reduce preterm birth. Obstet Gynecol. 2008;112:963-965. 2 Meis PJ, Klebanoff M, Thom E, et al. Prevention of recurrent preterm delivery by 17 alphahydroxyprogesterone caproate. N Engl J Med. 2003;348:2379-2385. 3. Romero R, Nicolaides KH, Conde-Agudelo A, et al. Vaginal progesterone decreases preterm birth ≤ 34 weeks of gestation in women with a singleton pregnancy and a short cervix: an updated meta-analysis including data from the OPPTIMUM study. Ultrasound Obstet Gynecol. 2016;48:308-317. 4. Saccone G, Schoen C, Franasiak JM, Scott RT Jr, Berghella V. Supplementation with progestogens in the first trimester of pregnancy to prevent miscarriage in women with unexplained recurrent miscarriage: a systematic review and meta-analysis of randomized, controlled trials. Fertil Steril. 2017;107:430-438. 5. Norman JE, Marlow N, Messow CM, et al.; OPPTIMUM Study Group. Vaginal progesterone prophylaxis for preterm birth (the OPPTIMUM study): a multicentre, randomised, double-blind trial. Lancet. 2016;387:2106-2116. 6. Axelsen SM, Henriksen TB, Hedegaard M, Secher NJ. Characteristics of vaginal bleeding during pregnancy. Eur J Obstet Gynecol Reprod Biol. 1995;63:131-1314. 7. Sipila P, Hartikainen-Sorri A. Perinatal outcome of pregnancies complicated by vaginal bleeding. Br J Obstet Gynecol. 1992;99:959-963. 8. Parant O, Clouet-Delannoy M, Connan L, Duclusaud A, Chale J, Fournié A. Metrorrhagia during the second trimester of pregnancy: obstetrical and perinatal outcome. A retrospective study including 85 cases. J Gynecol Obstet Biol Reprod. (Paris) 2000;29:66-72. 9. Koifman A, Levy A, Zaulan Y, Harlev A, Mazor M, Wiznitzer A, Sheiner E. The clinical significance of bleeding during the second trimester of pregnancy. Arch Gynecol Obstet. 2008;278:47-51. 10. Elovitz MA, Baron J, Phillippe M. The role of thrombin in preterm parturition. Am J Obstet Gynecol. 2001;185:1059-1063. 11. Rosen T, Kuczynski E, O’Neill LM, Funai EF, Lockwood CJ. Plasma levels of thrombinantithrombin complexes predict preterm premature rupture of the fetal membranes. J Matern Fetal Med. 2001;10:297-300. 12. Cakmak H, Schatz F, Huang ST, et al. Progestin suppresses thrombin- and interleukin1beta-induced interleukin-11 production in term decidual cells: implications for preterm This article is protected by copyright. All rights reserved. Accepted Article delivery. J Clin Endocrinol Metab. 2005;90:5279-5286. 13. Moher D, Hopewell S, Schulz KF, et al. CONSORT 2010 explanation and elaboration: updated guidelines for reporting parallel group randomised trials. BMJ. 2010;340:c869. 14. Tooher RL, Middleton PF, Crowther CA. A thematic analysis of factors influencing recruitment to maternal and perinatal trials. BMC Pregnancy Childbirth. 2008;8:36. 15. Watson J, Torgerson D. Increasing recruitment to randomised trials: a review of randomised controlled trials. BMC Med Res Methodol. 2006;6:34. 16. Committee on Practice Bulletins—Obstetrics, The American College of Obstetricians and Gynecologists. Practice bulletin no. 130: prediction and prevention of preterm birth. Obstet Gynecol. 2012;120:964-973. 17. Society for Maternal-Fetal Medicine (SMFM) Publications Committee. The choice of progestogen for the prevention of preterm birth in women with singleton pregnancy and prior preterm birth. Am J Obstet Gynecol. 2017;216:B11-13. 18. Signore CC, Sood AK, Richards DS. Second-trimester vaginal bleeding: correlation of ultrasonographic findings with perinatal outcome. Am J Obstet Gynecol. 1998;178:336-340. 19. Hammouda A. Bleeding in the first two trimesters of pregnancy. Review of 1000 cases. Int Surg. 1966;45:447-449. 20. Scott J. Vaginal bleeding in the midtrimester of pregnancy. Am J Obstet Gynecol. 1972;113:329-334. 21. Chan CC, To WW. Antepartum hemorrhage of unknown origin—what is its clinical significance? Acta Obstet Gynecol Scand. 1999;78:186-190. 22. Gabbe SG, Niebyl JR, Simpson JL, eds. Antepartum and Postpartum Hemorrhage. 7th ed. St; Louis: CV Mosby, 1983:403. 23. Lockwood C, Kuczynski E. Markers of risk for preterm delivery. J Perinat Med. 1999;27:5-20. 24. Elovitz MA, Baron J, Phillippe M. The role of thrombin in preterm parturition. Transactions of the Twenty-First Annual Meeting of the Society for Maternal-Fetal Medicine. Am J Obstet Gynecol. 2001;185:1059-1063. 25. Buhimschi CS, Schatz F, Krikun G, Buhimschi IA, Lockwood CJ. Novel insights into molecular mechanisms of abruption-induced preterm birth. Expert Rev Mol Med. 2010;12:e35. 26. Thachil J, Toh CH. Disseminated intravascular coagulation in obstetric disorders and its acute haematological management. Blood Rev. 2009;23:167. 27. Lockwood CJ, Kayisli UA, Stocco C, et al. Abruption-induced preterm delivery is associated with thrombin-mediated functional progesterone withdrawal in decidual cells. Am J This article is protected by copyright. All rights reserved. Accepted Article Pathol. 2012;181:2138. Legends of tables and figures Table 1: Demographic and obstetric characteristics of women according to trial group. Table 2: Pregnancy outcomes according to trial group. Table 3: Neonatal outcomes delivered at 24 weeks or more, according to trial group. Figure 1. Trial profile. Figure 2. Cox regression analysis of the cumulative percentage of continued. pregnancy without delivery in the progesterone and the placebo groups. This article is protected by copyright. All rights reserved. Table 1: Demographic and obstetric characteristics of women according to trial group Vaginal progesterone Placebo n=60 n=49 28.7±4.8 29.8±6.0 [28, 24.3-32.0] [29, 26-36] Arabic origin 33 (55.0) 26 (53.1) Jewish origin 27 (45.0) 23 (46.9) Pre-gestational body mass index, kg/m2 23.8±4.8 24.2±4.2 [22.1,20.5-26.8] [23.5, 20.7-26.2] Smoking 5 (8.3) 1 (2.0) Gravidity 2.9±1.5 2.8±1.4 [3, 2-4] [3, 2-4] 2.2±1.1 2.3±1.1 [2, 1-3] [2, 2-3] Primiparous (first delivery) 17 (28.3) 11 (22.4) First trimester bleeding 33 (55.0) 29 (59.2) Bleeding days in second trimester before randomization 2.8±4.2 2.4±1.8 [1, 1-3] [2, 1-3] 34 (56.7) 22 (44.9) 4 (6.7) 3 (6.1) 22 (36.7) 24 (49.0) Placental abruption 15 (25.0) 16 (32.7)a Low lying or placenta previa 15 (25.0) 16 (32.7) Gestational age at recruitment 18.9±4.0 18.0±3.5 [18, 15.4-22.9] [17.3,15.0-20.8] 19.3±3.6 18.5±2.9 [18, 16.0-22.9] [17.3, 16.0-20.8] 8 (13.3) 4 (8.2) Accepted Article Variable Maternal age, years Ethnicity Parity Severity of bleeding in second trimester before randomization Mild Moderate Heavy Gestational age at trial drug initiation Pregnancy complications b a Data are mean ± standard deviation, [median, IQR], or n (%) unless otherwise specified. a One women had both low lying and placental abruption. b Pregnancy complications were defined as presence of any of the following: thrombophilia, hypothyroidism, ulcerative colitis, chronic hypertension, diabetes. This article is protected by copyright. All rights reserved. Table 2: Pregnancy outcomes according to trial group Placebo p OR (95% CI) n=60 n=49 Delivery at < 37weeks 19 (31.7) 12 (24.5) 0.53 1.32 (0.55-3.16) Gestational age of pregnancies delivered > 24 weeks 37.4±3.0 38.1±2.5 0.22 ---- (38.1; 37.2-39.0) (38.6; 37.5-39.4) Delivery between 34 to 36+6 weeks 4 (6.7) 2 (4.1) 0.57 1.66 (0.29-9.53) Delivery < 34 weeks 15 (25.0) 10 (20.4) 0.72 1.19 (0.47-3.02) Delivery > 24 weeks 51 (85.0) 42 (85.7) 0.91 0.94 (0.31-2.84) Delivery < 24 weeks 9 (15.0) 7 (14.3) 0.92 1.06 (0.35-3.19) Days from initiation of treatment until delivery 107.2±51.6 118.5±52.8 0.33 --- (112.4; 75.2-154.0) (140.7; 96.6-154.7) 47 (78.3) 40 (81.6) -- 1.00 (reference) 11 (18.3) 8 (16.3) 0.80 1.14 (0.42-3.12) 2 (3.3) 1 (2.0) 0.98 1.03 (0.07-17.97) Days of bleeding after 0.45±1.11 0.94±2.83 0.20 --- initiation of treatment (0.0, 0-0) (0.0, 0-0) Mode of delivery of women delivered > 24 weeks N=51 N=42 37 (72.5) 34 (81.0) 1.00 (reference) 14 (27.5) 8 (19.0) 1.58 (0.58-4.25) Side effects of trial drug 2 (3.3) 1 (2.0) 0.64 1.79 (0.16-20.46) Adherence rate (%) 80.9±30.1 79.4±35.1 0.90 --- (100; 74.5-100.0) (100.0; 77.0-100.0) Accepted Article Vaginal progesterone Bleeding after initiation of treatment None Mild Moderate Vaginal Cesarean 0.37 Data are mean ± standard deviation, [median, IQR] or n (%) unless otherwise specified. Side effects of treatment: epilepsy outbreaks and genital itching in the progesterone group and genital itching in the placebo group. This article is protected by copyright. All rights reserved. Table 3: Neonatal outcomes delivered at 24 weeks or more, according to trial group Placebo p OR (95% CI) n=51 n=42 2926±696 3082±579 0.29 ---- (3036, 2615-3430) (3201, 2925-3457) Apgar score <7 at 5 minutes 1 (2.0) 2 (4.8) 0.59 0.40 (0.04-4.57) Cord artery pH<7.1 0 (0.0) 0 (0.0) --- --- Celestone treatment 14 (27.5) 9 (21.4) 0.50 1.39 (0.53-3.62) Respiratory distress syndrome 5 (9.8) 0 (0.0) 0.06 Infinity (0.78-infinity) Intraventricular hemorrhage 0 (0.0) 0 (0.0) --- --- Necrotizing enterocolitis 1 (2.0) 0 (0.0) 0.99 Infinity (0.02-infinity) Offspring infection 3 (5.9) 1 (2.4) 0.62 2.56 (0.26-25.59) Admission to intensive care unit 7 (13.7) 5 (11.9) 0.79 1.18 (0.34-4.02) Ventilation with positive airway pressure or intubation 2 (3.9) 0 (0.0) 0.50 Infinity (0.15-infinity) Phototherapy 8 (15.7) 6 (14.3) 0.85 1.11 (0.35-3.52) Any neonatal morbiditya 14 (27.5) 9 (21.4) 0.50 1.39 (0.53-3.62) Days to discharge 7.2±12.7 6.0±10.8 0.37 --- (3; 2-5) (3; 2-5) 1 (2.0) 0 (0.0) >0.99 Infinity (0.02-infinity) Accepted Article Vaginal progesterone Birth weight, grams Offspring mortality Data are mean ± standard deviation, [median, IQR], or n (%) unless otherwise specified. a Any neonatal morbidity was defined as presence of any of the following: Respiratory distress syndrome, Intraventricular hemorrhage, Necrotizing enterocolitis, Offspring infection, Admission to intensive care unit, Ventilation or Phototherapy. This article is protected by copyright. All rights reserved. Accepted Article This article is protected by copyright. All rights reserved. Accepted Article This article is protected by copyright. All rights reserved.