prematurity: review progestogens as maintenance treatment in … · 2016. 11. 8. · prematurity:...

Prematurity: Review

Progestogens as Maintenance Treatment inArrested Preterm LaborA Systematic Review and Meta-analysis

Montse Palacio, MD, PhD, Stefania Ronzoni, MD, PhD, Luis Sánchez-Ramos, MD,and Kellie E. Murphy, MD, MSc

OBJECTIVE: To evaluate the efficacy of maintenance

tocolysis with progestogens compared with placebo or

no treatment in women with singleton pregnancies and

arrested preterm labor.

DATA SOURCES: Studies without language restrictions

were identified from MEDLINE, EMBASE, PubMed, Sco-

pus, the Cochrane Pregnancy and Childbirth Group’s Tri-

als Register, the Cochrane Central Register of Controlled

Trials, and ClinicalTrials.gov from inception to June 2015.

MeSH headings for progestogens were combined with

terms regarding labor, tocolysis, or preterm birth. Refer-

ence lists of included studies and GoogleSearch were

also reviewed.

METHODS OF STUDY SELECTION: Randomized con-

trolled trials that compared progestogens as a mainte-

nance treatment after arrested preterm labor in singleton

pregnancies with placebo or no treatment were identi-

fied. Selected studies evaluated delivery before 37 or 34

weeks of gestation or the latency period from random-

ization to delivery. Excluded studies used progestogens

as prevention in asymptomatic women at risk. Risk of

bias assessment, subgroup analysis on type of proges-

togens used, and sensitivity analysis by high-quality

studies were performed.

TABULATION, INTEGRATION, AND RESULTS: Sixteen

randomized controlled trials consisting of 1,917 partic-

ipants were included. Study characteristics and quality

were recorded. Preterm delivery at less than 37 weeks of

gestation was decreased (38.2% compared with 44.3%;

relative risk 0.79, 95% confidence interval [CI] 0.65–0.97)

and pregnancy was prolonged (mean difference 8.1 days;

95% CI 3.8–12.4) when women treated with progesto-

gens were compared with placebo or no treatment.

There were no differences in the outcome of delivery

at less than 34 weeks of gestation (15.6% compared with

18.3%; relative risk 0.77, 95% CI 0.53–1.12). However,

sensitivity analysis including five high-quality studies

showed no significant differences for preterm delivery

at less than 37 weeks of gestation (37.2% compared with

36.9%; relative risk 0.91, 95% CI 0.67–1.25) or latency

period (mean difference 0.6 days; 95% CI 23.7 to 4.9).

CONCLUSION: There is insufficient high-quality data to

inform clinicians and patients about the use of proges-

togens as maintenance treatment after arrested preterm

labor to reduce the incidence of preterm birth or

pregnancy prolongation.

(Obstet Gynecol 2016;128:989–1000)

DOI: 10.1097/AOG.0000000000001676

More than 1 in 10 of the world’s children born in2010 were born preterm, defined as before 37weeks of gestation.1 Current tocolytic agents appear tobe superior to placebo at delaying delivery at both 48hours and 7 days but do not reduce the incidence ofpreterm birth or improve perinatal outcome.2 Main-tenance tocolytic therapy has not been proven toimprove the outcomes in women treated because of

From the BCNatal–Barcelona Center for Maternal-Fetal and Neonatal Medicine(Hospital Clínic and Hospital Sant Joan de Deu), Fetal i+D Fetal MedicineResearch Center, IDIBAPS, University of Barcelona, Barcelona, and the Centrefor Biomedical Research on Rare Diseases (CIBER-ER), Spain; the Departmentof Obstetrics and Gynaecology, Mount Sinai Hospital, University of Toronto,Toronto, Ontario, Canada; and the Department of Obstetrics and Gynecology,Division of Maternal-Fetal Medicine, University of Florida Health ScienceCenter, Jacksonville, Florida.

Montse Palacio was supported by Instituto de Salud Carlos III (BA15/00028and EC07/90023) and Ministerio de Sanidad y Política Social (TRA-096).

The authors thank Daphne Horn, Research Librarian at Mount Sinai Hospital,for help with the literature search, and Josie Chundamala, Scientific GrantEditor in the Department of Obstetrics and Gynaecology at Mount Sinai Hos-pital, for editorial assistance.

Corresponding author: Montse Palacio, MD, PhD, BCNatal, Barcelona Centre forMaternal-Fetal and Neonatal Medicine, Hospital Clínic, Sabino de Arana 1, 08028Barcelona, Spain; e-mail: [email protected].

Financial DisclosureThe authors did not report any potential conflicts of interest.

© 2016 by The American College of Obstetricians and Gynecologists. Publishedby Wolters Kluwer Health, Inc. All rights reserved.ISSN: 0029-7844/16

VOL. 128, NO. 5, NOVEMBER 2016 OBSTETRICS & GYNECOLOGY 989

Copyright ª by The American College of Obstetriciansand Gynecologists. Published by Wolters Kluwer Health, Inc.

Unauthorized reproduction of this article is prohibited.

http://ClinicalTrials.govRichardResaltado

RichardResaltado

RichardResaltado

preterm labor and the risk of preterm birth in thesepatients remains high.3,4

Progestogens have been shown to reduce thespontaneous preterm birth rate, especially in womenat risk for preterm birth,5 with a history of prior pre-term birth,6,7 or with an asymptomatic short cervix.8–10

It is reasonable to hypothesize that progestogens, withtheir inhibitory effect on uterine contractility11 and rolein maintaining pregnancy until term,12 would be a goodintervention in arrested preterm labor. As such, both17-hydroxyprogesterone13,14 and vaginal naturalprogesterone15–20 have been studied as maintenancetreatment for women with arrested preterm laborwith conflicting results. The authors of two recentmeta-analyses21,22 on the topic were cautious withtheir conclusions because of the limited quality ofthe studies included. Given that recent data fromadditional large studies were not included20,23 andother small studies were missed, a review of the topicat this time is warranted.

The objective of this systematic review and meta-analysis was to examine whether progestogens are aneffective maintenance treatment for women afterarrested preterm labor.

SOURCES

A systematic review was conducted according to theprotocol as outlined subsequently and reported fol-lowing Preferred Reporting Items for Systematic Re-views and Meta-Analyses guidelines.24 The followingsources were searched from inception to June 2015:MEDLINE, EMBASE, the Cochrane Central Regis-ter of Systematic Reviews, the Cochrane Central Reg-ister of Controlled Trials, Web of Science, andClinicalTrials.gov database. The full search strategycan be found in Appendix 1, available online athttp://links.lww.com/AOG/A869. The reference listsof all included primary and review articles and Goo-gleSearch were also examined to identify articles notcaptured by the electronic searches. No language re-strictions were applied.

STUDY SELECTION

All articles were screened by two authors (M.P., S.R.)and articles likely to meet selection criteria werereviewed. Three reviewers (M.P., S.R., K.E.M.) madethe final inclusion and exclusion decisions accordingto adherence to the following: 1) population: pregnantwomen with singleton pregnancies who receivedprogestational agents for maintenance of pregnancyafter arrested preterm labor; 2) progestogen: use ofany kind of progestogen and by any route (ie,intramuscular, vaginal, or oral) had to be tested; 3)

main outcome: delivery at less than 37 weeks ofgestation, delivery at less than 34 weeks of gestation,and latency period to delivery; and 4) study design:randomized clinical trials in which progestationalagents were compared with a nontreatment arm orto placebo. Trials that included women with rupturedmembranes, trials without a placebo or a “no treat-ment” arm, and other studies were excluded.

All articles were assessed independently and dataabstracted by two reviewers (M.P., S.R.). A thirdreviewer (K.E.M. or L.S.-R.) resolved by consensus orthrough arbitration in case of disagreement. In case ofduplicate publications or those with overlapping datasets, only data from the most complete study wereincluded.

All included papers were assessed for methodo-logic quality and risk of bias by two investigators(M.P., S.R.) using Jadad’s criteria25 and the CochraneRisk of Bias tool.26 Disagreements were resolved byconsensus with a third reviewer (K.E.M.). Trials thatmet all of Jadad’s criteria were considered highquality.

Data were extracted for the main outcomes:preterm birth at less than 37 weeks of gestation, lessthan 34 weeks of gestation, and latency period fromrandomization to delivery. Secondary outcomes suchas gestational age at delivery, recurrence of pretermlabor, neonatal outcomes as admissions to the neo-natal intensive care unit, and neonatal respiratorydistress syndrome were also explored. Studies report-ing continuous variables as medians and interquartileranges were converted to means and standard devia-tions as per Hozo’s method.27

Data analysis was completed independently bytwo authors (M.P., L.S.-R.) using Review Manager5.3. Any differences between the analyses wereresolved by independent analysis by the coauthors.

A random-effects meta-analysis model ofDerSimonian and Laird was chosen for each test toobtain the pooled relative risks (RRs) estimate whenthere was evidence of statistical heterogeneity betweenstudies (P value of the Cochrane Q statistic ,.1). Oth-erwise, a fixed-effects model was selected. Statisticalheterogeneity between studies was assessed using theCochrane Q statistic and I2 statistics of Higgins et al.26

The summary measures were reported as RRs with95% confidence interval (CI) for binary outcomesand weighted mean difference for continuous variables.A P value ,.05 was considered statistically significant.To explore potential sources of heterogeneity, we per-formed planned a priori subgroup analysis to deter-mine whether the type of progestogen used wouldaffect the results. An a priori sensitivity analysis was

990 Palacio et al Progestogens as Maintenance Treatment OBSTETRICS & GYNECOLOGY



http://ClinicalTrials.govhttp://links.lww.com/AOG/A869

also performed consisting of high-quality trials basedon Jadad’s criteria.25

To test for the presence of publication bias,a visual inspection of the funnel plots displayingindividual study log RR with the standard error RRfor binary outcomes and mean difference with stan-dard error of mean difference for continuous out-comes was performed. Asymmetry in such funnelplots is usually caused by small trials that reportgreater effects, on average, than large trials, whichsuggests publication or other biases.28 To assess pub-

lication bias statistically, Beggs’ test for small studyeffects was used (P,.05).

All analyses were conducted and summary resultsgenerated as forest plots using Review Manager(RevMan) 5.3.

RESULTS

Sixteen randomized controlled trials met the inclu-sion criteria and were included in this meta-analysis(Fig. 1).13–18,20,29–36 One trial32 had three arms: one usingprogesterone, one using 17-hydroxyporgesterone, and

Fig. 1. Flowchart of includedstudies. PROM, prelabor rupture ofmembranes; RCT, randomizedcontrolled trial.

Palacio. Progestogens as MaintenanceTreatment. Obstet Gynecol 2016.

VOL. 128, NO. 5, NOVEMBER 2016 Palacio et al Progestogens as Maintenance Treatment 991



Table 1. Descriptive Data for Each Trial and Primary Outcomes

Author, Year,Country Years

S orM

Total Patients (Progestogenvs Control)

Dose (mg),Route Control Primary Tocolytic Agent

PretermLabor When

StartingMedication

GA Range atRandomization

(wk)

Borna and Sahabi,17

2008, Iran2004–2005 S 70 (37 vs 33) 400 OD, VAG No T MgSO4 Arrested, tocolysis discontinued 24–34 6/7

El-Abidin et al,35

2009, Egypt2007–2008 S 40 (20 vs 20) 200 BID, VAG No T MgSO4 Arrested, tocolysis discontinued 24–34

Sharami et al,33 2010,Iran

2007–2009 S 163 (80 vs 83) 200 OD, VAG Placebo MgSO4 Arrested, tocolysis not specified 28–34 6/7

Arikan et al,18 2011,Turkey

NA S 83 (43 vs 40) 200 OD, VAG No T Ritodrine Acute preterm labor 24–34 6/7

Saleh Gargari et al,30

2012, Iran2007–2010 S 144 (72 vs 72) 400 OD, VAG No T MgSO4 Arrested, tocolysis not specified 24–34 6/7

Areia et al,15 2013,Portugal

2008–2010 S 52 (26 vs 26) 200 OD, VAG No T Atosiban Arrested, tocolysis discontinued 24–34 6/7

Lotfakizadeh et al,32

2013, Iran2010 S 73 (37 vs 36) 400 OD, VAG No T MgSO4 or nifedipine Arrested, on tocolysis 26–36 6/7

Mishra and Inamdar,36

2014, India2012–2014 S 100 (50 vs 50) 400 OD, VAG No T Isoxsuprine hydrochloride Arrested, tocolysis discontinued 28–36

Martinez de Tejadaet al,20 2015,Switzerland andArgentina

2006–2012 M 379 (193 vs 186) 200 OD, VAG Placebo B-mimetic, atosiban, or Ca+channel blockers

Arrested, tocolysis not specified 24–33 6/7

Palacio et al,23 2016,Spain

2008–2012 M 258 (126 vs 132) 200 OD, VAG Placebo B-mimetic, atosiban, nifedipine Arrested, tocolysis discontinued 24–33 6/7

Noblot et al,29 1991,France

1987 S 44 (22 vs 22) 300 TID, PO Placebo Ritrodrine Acute preterm labor ,35

Choudhary et al,16

2014, India2010–2012 S 90 (45 vs 45) 200 OD, PO Placebo Nifedipine Arrested, on tocolysis 24–34 6/7

Facchinetti et al,31

2007, Italy2004–2006 S 60 (30 vs 30) 341 bwk, IM No T Atosiban Not specified if arrested, on tocolysis 25–33 6/7

Rozenberg et al,13

2012, France2006–2008 M 188 (94 vs 94) 500 bwk, IM No T Nifedipine, nicardipine

salbutamolArrested, tocolysis not specified 24–31 6/7

Regmi et al,34 2012,Nepal

2009–2010 S 60 (29 vs 31) 250 wk, IM No T Nifedipine Arrested, tocolysis discontinued 28–34 6/7

Lotfakizadeh et al,32

2013, Iran2010 S 73 (37 vs 36) 250 wk, IM No T MgSO4 or nifedipine Arrested, still on tocolysis 26–36 6/7

Briery et al,14 2014,United States

39 mo S 45 (22 vs 23) 250 wk, IM Placebo MgSO4 or Ca+ channel blockersor anti-PGE

Arrested, tocolysis discontinued 20–30 6/7

S, single center; M, multicenter; GA, gestational age; PTL, preterm labor; PTB, preterm birth; OD, once a day; VAG, vaginal; No T, no treatment; MgSO4, magnesium sulphate; ctxs,contractions; NA, not available; BID, twice a day; TID, every 8 hours; bwk, biweekly; PO, per os; IM, intramuscular; anti-PGE, antiprostaglandins.

992Palacio

etal

Progesto

gensas

Main

tenan

ceTreatm

ent

OBSTET

RIC

S&

GYNEC

OLO

GY



Mean GA atRandomization

(wk)Definitionof PTL

Prior PTB Progestogens vsControl (%) Study Primary Outcomes

PTB at Less Than 37 Wk ofGestation, n (%)

PTB at Less Than 34 Wkof Gestation, n (%)

Latency (d;mean)

31/32 .6 ctxs/309+cervical changes(manual)

5/37 (13.5) vs 4/33 (12.1) Latency period, recurrent PTL NA NA 36.1 vs 24.5


9/20 (45) vs 7/20 (35) Latency period and recurrentPTL

NA 2/20 (10), 11/20 (55) 8.3 vs 5.0

33/34 $6 ctxs/309 for 30sec+cervicalchanges (manual)

1/80 (1.3) vs 3/83 (3.6) Latency period and PTB lessthan 37 and 34 wk ofgestation

33/80 (41), 45/83 (54) 8/80 (10), 9/83 (11) 23.9 vs 16.7


4/43 (9.3) vs 3/40 (7.5) Latency period, GA at deliveryand PTB less than 37 wk ofgestation

20/43 (47), 28/40 (70) 20/43 (47), 28/40 (70) 32.1 vs 21.2

32/32 $4 ctxs/209 or $8 ctxs/609+cervical changes(manual)

NA Latency period, GA at delivery NA NA 28.0 vs 9.8

28/29 4 ctxs/209 or 8 ctxs/609+cervical changes (TVUS orfFN+)

9/26 (34.6) vs 9/26 (34.6) Latency period 9/26 (35), 13/26 (50) 3/26 (12), 6/26 (23) 55 vs 38

34/33 4 ctxs/209+2 cm cervicaldilatation or progressivecervical changes (manual)

NA Rate of recurrent PTL NA NA 31 vs 26

Stratified forrange of GA

4 ctxs/209 or 8 ctxs/609+cervical dilatation .1 cmand eff .80%

Latency, birth weight, NICUadmission

17/50 (34), 34/50 (68) 4/50 (8), 13/50 (26) 23.4 vs 11.7

29/29 $2 ctxs/109 for 309+cervicalchanges (TVUS or manual,or fFN+)

46/193 (23.8) vs 39/186 (21.0) PTB less than 37 wk ofgestation

82/193 (43), 66/186 (36) 38/193 (20), 24/186 (13) 45 vs 52

32/32 $2 ctxs/109 for 309+cervicalchanges (TVUS and manual)

11.9 vs 12.1 PTB less than 37 and 34 wk ofgestation

36/126 (29), 29/132 (22) 9/126 (7), 10/132 (8) NA

32/31 Regular ctxs /109 for 1h+cervical change (manual)

NA NA 8/22 (36), 6/22 (27) 44 vs 42

32/32 4 ctxs/209 or 8 ctxs/609+cervical dilatation .1 cmand eff .80%

6/45 (13.3) vs 2/45 (4.4) Latency period 15/45 (33), 26/45 (58) 8/45 (18), 9/45 (20) 33.3 vs 23.1

30/30 .6 ctxs/309 for 30sec+cervicalchanges (manual)

1/30 (3.3) vs 2/30 (6.6) Cervical shortening 5/30 (16), 17/30 (57) 3/30 (10), 7/30 (23) 35.3 vs 25.5

28/28 $2 ctxs/109 for 609+TVUS 11/94 (11.7) vs 22/92 (23.9) Latency period 37/94 (35), 36/94 (38) 15/94 (16), 19/94 (20) 61 vs 6333/33 .6 ctxs/309 for 30sec+cervical

changes (manual)11/29 (37.9) vs 20/31 (64.5) Latency period and rate of

recurrent PTL within 48 hNA NA 25 vs 16

34/33 4 ctxs/min+2 cm cervicaldilatation or progressivecervical changes (manual)

NA Rate of recurrent PTL NA NA 36 vs 26

29/27 Painful ctx every 5 min andcervical dilatation (manual)

8/22 (36.4) vs 7/23 (30.4) PTB less than 37 wk ofgestation

19/22 (86), 22/23 (96) 14/22 (63), 21/23 (91) 23 vs 16

VOL.

128,NO.5,NOVEM

BER

2016

Palacio

etal

Progesto

gensas

Main

tenan

ceTreatm

ent

993



the other received no treatment. Another29 includeda small number of multiple pregnancies (four in total),but all authors agreed to include these data because theeffect on the overall data was considered to be negligible.

Overall, 1,917 women were included in this systematicreview. Descriptive characteristics of the studies includedand primary outcomes are shown in Table 1. The trialswere relatively small with only two studies recruiting 200or more participants20,23 and six trials used placebo in thecontrol arm.14,16,20,23,29,33 Five trials fulfilled all of Jadad’scriteria and thus were considered to be high qual-ity.16,20,23,29,33 The risk of bias for the included studies isdetailed in Figure 2. Of the 16 included trials, 7 failed inat least half of the Cochrane Risk of Bias tool criteria.

The rate of preterm birth at less than 37 weeks ofgestation in the placebo or no treatment arms in theincluded studies ranged from 22%23 to 96%,14

although the rate was reduced to 22%23 to 58%16 whenonly high-quality studies were considered. All butthree studies20,23,29 reported a rate of preterm birthat less than 37 weeks of gestation of 50% or moreand five13,14,16,35,36 out of eight that reported the resultsof preterm birth at less than 34 weeks of gestationreported a rate from 20% to 91.3% for this outcomein the placebo or no treatment arm.

Five trials reported the use of magnesium sul-phate as a tocolytic agent.17,30,32,33,35 Other trials wereusing betamimetics, calcium channel blockers, atosi-ban, or antiprostaglandin drugs (Table 1).

Twelve trials tested natural progesterone eitherby the vaginal15,17,18,20,23,30,32,33,35,36 or oral16,29

route and five trials used intramuscular 17-hydroxyprogesterone.13,14,31,32,34 Nine trials reported theuse of a standard dose of progesterone (ie, 200 mg perday)15,16,18,20,23,33 or 17-hydroxyprogesterone (ie, 250 mgintramuscularly weekly).14,32,34 Eight other trialsused higher doses of progesterone17,29,30,32,35,36 or17-hydroprogesterone.13,31

In one study,23 women were randomized once thedecision to discharge was taken and medication wasstarted the first day at home. In all the other trials,women were randomized and started on study treat-ment while in the hospital. However, the rate of pre-term delivery before hospital discharge was not statedin any study. In two trials,18,29 study medication wasstarted together with the acute tocolysis or this was notspecified31 and was prolonged as a maintenance treat-ment. In six trials, study medication was started aftercessation of uterine contractions but concomitant totocolysis16,32 or the latter was not specified.13,20,30,33 Inseven trials, study medication was administered aftera contraction-free period after discontinuation of acutetocolysis.14,15,17,23,34–36

Secondary outcomes of the included studies areshown in Appendix 2, available online at http://links.lww.com/AOG/A869. Preterm delivery at less than37 weeks of gestation, less than 34 weeks of gestation,

Fig. 2. Risk of bias summary: review authors’ judgmentsabout each risk of bias item for each included study.

Palacio. Progestogens as Maintenance Treatment. Obstet Gynecol2016.




http://links.lww.com/AOG/A869http://links.lww.com/AOG/A869

and latency to delivery was reported in 11, 8, and 16out of 16 studies, respectively.

The forest plots in Figures 3–5 provide study-specific details regarding evaluation of pooled pre-term delivery at less than 37 weeks of gestation, lessthan 34 weeks of gestation, and latency periodbetween preterm labor and delivery. Preterm deliveryat less than 37 weeks of gestation was decreased(38.2% compared with 44.3%; RR 0.79, 95% CI0.65–0.97) and pregnancy was prolonged (mean dif-ference 8.1 days; 95% CI 3.8–12.4) when women trea-ted with progestogens were compared with womenwho received placebo or no treatment. There were

no differences in the outcome of delivery at less than34 weeks of gestation (15.6% compared with 18.3%;RR 0.77, 95% CI 0.53–1.12). Significant heterogeneityamong studies was noted for preterm delivery at lessthan 37 weeks of gestation (I2 65%, P5.001), less than34 weeks of gestation (I2 56%, P5.03), and latencyperiod (I2 96%, P,.001).

Only five studies fulfilled all of Jadad’s criteria.Sensitivity analysis of these studies showed no signif-icant differences in preterm delivery at less than 37weeks of gestation (36.9% compared with 37.2%; RR0.91, 95% CI 0.67–1.25), preterm delivery at less than34 weeks of gestation (14.2% compared with 11.7%;

Fig. 3. A. Forest plot comparison (progestational agents compared with nontreatment or placebo) for outcome: preterm birth atless than 37 weeks of gestation. B. Forest plot comparison (progestational agents compared with nontreatment or placebo) foroutcome: preterm birth at less than 37 weeks of gestation (high-quality studies). IV, inverse-variance; CI, confidence interval.

Palacio. Progestogens as Maintenance Treatment. Obstet Gynecol 2016.




RR 1.20, 95% CI 0.86–1.69), or for latency period(mean difference 0.6 days; 95% CI 23.7 to 4.9). Sen-sitivity analysis did not explain the source of hetero-geneity for preterm delivery at less than 37 weeks ofgestation (I2 65%, P5.02) or the latency period (I2

86%, P,.001), but did explain the heterogeneity forpreterm delivery at less than 34 weeks of gestation (I2

0%, P5.55).There were limited data reporting on the second-

ary outcomes of gestational age at delivery (13 of 16studies), recurrence of preterm labor (8 of 16 studies),neonatal admission to the neonatal intensive care unit(12 of 16 studies), or neonatal respiratory distresssyndrome (10 of 16 studies). Other outcomes relatedto neonatal morbidity (intraventricular hemorrhage,

necrotizing enterocholitis, or sepsis) were inconsis-tently reported (Appendix 2, http://links.lww.com/AOG/A869).

Table 2 summarizes the overall results and theresults of the sensitivity analysis on the five high-quality studies (Jadad score of 5) for all outcomes.Use of progestogens did not improve the primary orsecondary outcomes. Appendix 3, available online athttp://links.lww.com/AOG/A869, summarizes the re-sults for the secondary outcomes of the subgroup anal-ysis when the type of progestogen used wasconsidered. Subgroup analysis did not show addi-tional benefit for any of the progestogens used.

Analysis of publication bias based on Egger’s testshowed publication bias for the outcome of latency to

Fig. 4. A. Forest plot comparison (progestational agents compared with nontreatment or placebo) for outcome: preterm birth atless than 34 weeks of gestation. B. Forest plot comparison (progestational agents compared with nontreatment or placebo) foroutcome: preterm birth at less than 34 weeks of gestation (high-quality studies). IV, inverse-variance; CI, confidence interval.





http://links.lww.com/AOG/A869http://links.lww.com/AOG/A869http://links.lww.com/AOG/A869

delivery (17 studies, P5.010) (Fig. 6) and suggests biasfor preterm birth at less than 37 weeks of gestation (11studies, P5.095) (Appendix 4, available online athttp://links.lww.com/AOG/A869).

DISCUSSION

There is insufficient high-quality evidence to eval-uate the use of progestogens as maintenance treat-ment after arrested preterm labor to either reducepreterm birth or increase latency to delivery.Studies included in this meta-analysis were variablein both quality and design because they varied insize, type of progestogen, route of administration,and whether they included a placebo arm.Although differences in the primary outcomes werefound when all trials were included, sensitivityanalysis performed on high-quality trials showed

no differences in preterm birth at less than 37 weeksof gestation, less than 34 weeks of gestation, or inthe latency period to delivery between the pro-gestogen and placebo arms.

Two recent meta-analyses21,22 have shown thatprogestogens, when used after arrested preterm labor,might be of some benefit. However, the authors werecautious with their conclusions because of the limitedquality of the studies published (ie, small sample sizeor lack of blinding). Since that publication, a subse-quent meta-analysis showed conflicting results,37

recent data have been published,23 and additionalstudies in which progestogen was introduced shortlyafter preterm labor was arrested were not included inthe earlier meta-analyses20; a review of the topic at thistime is timely. Furthermore, previous systematic re-views21,22,37 did not include small trials not cited in

Fig. 5. A. Forest plot comparison (progestational agents compared with nontreatment or placebo) for outcome: latency todelivery. B. Forest plot comparison (progestational agents compared with nontreatment or placebo) for outcome: latency todelivery (high-quality studies). SD, standard deviation; IV, inverse-variance; CI, confidence interval.





http://links.lww.com/AOG/A869RichardResaltado

PubMed.35,36 This systematic review also addressesthis deficiency.

The strengths of this systematic review includedits comprehensive search strategy, methodologicdesign, and statistical analysis. In particular, it synthe-sized the results of existing studies in which pro-gestational agents were used as a maintenancetreatment after arrested preterm labor whether ornot the treatment was initiated together with acutetocolysis. The inclusion of a range of outcomes suchas preterm birth, latency period from randomizationto delivery, recurrence of preterm labor, gestationalage at delivery, admissions to the neonatal intensive

care unit, and neonatal distress syndrome ensureda complete evaluation of the intervention. In addition,a subgroup analysis was performed and allowed thecomparison of the estimated effect of different proges-togens, which is relevant for different clinical practicesand sensitivity analysis showed results derived frominclusion of high-quality studies.

The studies’ heterogeneity was this systematicreview’s most obvious limitation and limits the val-idity of the combined results, especially becausefindings from larger and low risk of bias trials con-flicted with those from smaller trials. The potentialfor selection bias exists in randomized studies whenthere is no blinding or inadequate concealment,which happened in 10 of 16 studies (Fig. 2), limitingthe quality of the data. Indeed, heterogeneity mightalso be explained by the fact that inclusion criteriaamong studies were very different. Evidence for thatis the high rate of preterm birth in some studiescompared with others. On the other hand, some out-comes (ie, preterm delivery at less than 34 weeks ofgestation) were reported in some studies but not inothers. The inclusion of core outcome sets for pre-term delivery in future research may avoid hetero-geneity resulting from this fact.

In addition, the risk of publication bias wasassessed by visual inspection of the funnel andasymmetric plots and suggested publication bias. Thissuggests underreporting of negative trials. Thus, thiscomprehensive systematic review provides data toevaluate these limitations and forms the basis forfuture studies.

Table 2. Summary of the Overall Results and the Sensitivity Analysis (High-Quality Studies Jadad’s Score 5)

Outcomes

Overall Analysis Sensitivity Analysis (High-Quality Studies)

No. ofStudies

No. ofParticipants I2 (%) P

Effect Estimate(Progestogens vsNontreatment or

Placebo)No. ofStudies

No. ofParticipants I2 (%) P

Effect Estimate(Progestogens vsNontreatment or

Placebo)

PTB at less than 37wk of gestation

11 1,462 65 .001 0.79 (0.65–0.97)* 5 934 65 .02 0.91 (0.67–1.25)*

PTB at less than 34wk of gestation

8 1,263 56 .03 0.77 (0.53–1.12)† 4 890 0 .55 1.20 (0.86–1.69)†

Latency to delivery (d) 16 1,917 96 ,.001 8.1 (3.8–12.4)* 5 929 86 ,.001 0.6 (23.7–4.9)*GA at delivery 13 1,672 87 ,.001 1.31 (0.80–1.81)‡ 4 890 69 .02 0.14 (20.51–0.78)‡

Recurrent pretermlabor

8 1,047 0 .56 0.63 (0.52–0.78)† 3 725 20 .29 0.74 (0.53–1.02)†

Admission to NICU 12 1,672 33 .13 0.91 (0.75–1.10)† 4 882 0 .98 1.15 (0.87–1.52)†

Respiratory distresssyndrome

10 1,057 7 .38 0.70 (0.50–0.97)† 3 506 0 .97 0.75 (0.43–1.31)†

PTB, preterm birth; GA, gestational age; NICU, neonatal intensive care unit; CI, confidence interval.* Risk ratio (IV, random, 95% CI).† Risk ratio (M-H, fixed, 95% CI).‡ Mean difference (IV, random, 95% CI).

Fig. 6. Funnel plot of studies (comparing progestationalagents and nontreatment or placebo) for outcome: latencyto delivery.

Palacio. Progestogens as Maintenance Treatment. Obstet Gynecol2016.




The findings from this meta-analysis suggest thatthere is insufficient high-quality data to inform clini-cians and patients about the use of progestogens asmaintenance treatment after arrested preterm labor toreduce preterm birth or prolong pregnancy. There-fore, the use of progestogens for this indication shouldbe limited to research protocols until results fromadditional large and well-designed randomized trialsbecome available.

REFERENCES1. Blencowe H, Cousens S, Oestergaard MZ, Chou D, Moller AB,

Narwal R, et al. National, regional, and worldwide estimates ofpreterm birth rates in the year 2010 with time trends since 1990for selected countries: a systematic analysis and implications.Lancet 2012;379:2162–72.

2. Haas DM, Imperiale TF, Kirkpatrick PR, Klein RW,Zollinger TW, Golichowski AM. Tocolytic therapy: a meta-analysis and decision analysis. Obstet Gynecol 2009;113:585–94.

3. Roos C, Spaanderman ME, Schuit E, Bloemenkamp KW,Bolte AC, Cornette J, et al. Effect of maintenance tocolysis withnifedipine in threatened preterm labor on perinatal outcomes:a randomized controlled trial. JAMA 2013;309:41–7.

4. Sanchez-Ramos L, Kaunitz AM, Gaudier FL, Delke I. Efficacyof maintenance therapy after acute tocolysis: a meta-analysis.Am J Obstet Gynecol 1999;181:484–90.

5. Sanchez-Ramos L, Kaunitz AM, Delke I. Progestational agentsto prevent preterm birth: a meta-analysis of randomized con-trolled trials. Obstet Gynecol 2005;105:273–9.

6. da Fonseca EB, Bittar RE, Carvalho MH, Zugaib M. Prophy-lactic administration of progesterone by vaginal suppository toreduce the incidence of spontaneous preterm birth in women atincreased risk: a randomized placebo-controlled double-blindstudy. Am J Obstet Gynecol 2003;188:419–24.

7. Meis PJ, Klebanoff M, Thom E, Dombrowski MP, Sibai B,Moawad AH, et al. Prevention of recurrent preterm deliveryby 17 alpha-hydroxyprogesterone caproate. N Engl J Med2003;348:2379–85.

8. Fonseca EB, Celik E, Parra M, Singh M, Nicolaides KH; FetalMedicine Foundation Second Trimester Screening Group. Pro-gesterone and the risk of preterm birth among women witha short cervix. N Engl J Med 2007;357:462–9.

9. Hassan SS, Romero R, Vidyadhari D, Fusey S, Baxter JK,Khandelwal M, et al. Vaginal progesterone reduces the rateof preterm birth in women with a sonographic short cervix:a multicenter, randomized, double-blind, placebo-controlledtrial. Ultrasound Obstet Gynecol 2011;38:18–31.

10. Romero R, Nicolaides K, Conde-Agudelo A, Tabor A,O’Brien JM, Cetingoz E, et al. Vaginal progesterone in womenwith an asymptomatic sonographic short cervix in the mid-trimester decreases preterm delivery and neonatal morbidity:a systematic review and metaanalysis of individual patient data.Am J Obstet Gynecol 2012;206:124.e1–19.

11. Lye SJ, Porter DG. Demonstration that progesterone “blocks”uterine activity in the ewe in vivo by a direct action on themyometrium. J Reprod Fertil 1978;52:87–94.

12. Henderson D, Wilson T. Reduced binding of progesteronereceptor to its nuclear response element after human laboronset. Am J Obstet Gynecol 2001;185:579–85.

13. Rozenberg P, Chauveaud A, Deruelle P, Capelle M, Winer N,Desbrière R, et al. Prevention of preterm delivery after success-

ful tocolysis in preterm labor by 17 alpha-hydroxyprogesteronecaproate: a randomized controlled trial. Am J Obstet Gynecol2012;206:206.e1–9.

14. Briery CM, Klauser CK, Martin RW, Magann EF, Chauhan SP,Morrison JC. The use of 17-hydroxy progesterone in womenwith arrested preterm labor: a randomized clinical trial.J Matern Fetal Neonatal Med 2014;27:1892–6.

15. Areia A, Fonseca E, Moura P. Progesterone use after successfultreatment of threatened pre-term delivery. J Obstet Gynaecol2013;33:678–81.

16. Choudhary M, Suneja A, Vaid NB, Guleria K, Faridi MM.Maintenance tocolysis with oral micronized progesterone forprevention of preterm birth after arrested preterm labor. Int JGynaecol Obstet 2014;126:60–3.

17. Borna S, Sahabi N. Progesterone for maintenance tocolytictherapy after threatened preterm labour: a randomised con-trolled trial. Aust N Z J Obstet Gynaecol 2008;48:58–63.

18. Arikan I, Barut A, Harma M, Harma IM. Effect of progesteroneas a tocolytic and in maintenance therapy during preterm labor.Gynecol Obstet Invest 2011;72:269–73.

19. Bomba-Opon DA, Kosinska-Kaczynska K, Kosinski P,Wegrzyn P, Kaczynski B, Wielgos M. Vaginal progesteroneafter tocolytic therapy in threatened preterm labor. J MaternFetal Neonatal Med 2012;25:1156–9.

20. Martinez de Tejada B, Karolinski A, Ocampo M, Laterra C,Hösli I, Fernández D, et al. Prevention of preterm delivery withvaginal progesterone in women with preterm labour (4P): rand-omised double-blind placebo-controlled trial. BJOG 2015;122:80–91.

21. Suhag A, Saccone G, Berghella V. Vaginal progesterone formaintenance tocolysis: a systematic review and metaanaly-sis of randomized trials. Am J Obstet Gynecol 2015;213:479–87.

22. Saccone G, Suhag A, Berghella V. 17-alpha-hydroxyprogester-one caproate for maintenance tocolysis: a systematic review andmetaanalysis of randomized trials. Am J Obstet Gynecol 2015;213:16–22.

23. Palacio M, Cobo T, Antolín E, Ramirez M, Cabrera F, Mozo deRosales F, et al. Vaginal progesterone as maintenance treatmentafter an episode of preterm labour (PROMISE) study: a multi-center, double-blind, randomized, placebo-controlled trial.BJOG 2016 Mar 30 [Epub ahead of print].

24. Moher D, Liberati A, Tetzlaff J, Altman DG; PRISMA Group.Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. BMJ 2009;339:b2535.

25. Jadad AR, Moore RA, Carroll D, Jenkinson C, Reynolds DJ,Gavaghan DJ, et al. Assessing the quality of reports of random-ized clinical trials: is blinding necessary? Control Clin Trials1996;17:1–12.

26. Higgins JP, Altman DG, Gøtzsche PC, Jüni P, Moher D,Oxman AD, et al. The Cochrane Collaboration’s tool for as-sessing risk of bias in randomised trials. BMJ 2011;343:d5928.

27. Hozo SP, Djulbegovic B, Hozo I. Estimating the mean andvariance from the median, range, and the size of a sample.BMC Med Res Methodol 2005;5:13.

28. Dickersin K. The existence of publication bias and risk factorsfor its occurrence. JAMA 1990;263:1385–9.

29. Noblot G, Audra P, Dargent D, Faguer B, Mellier G. The use ofmicronized progesterone in the treatment of menace of pretermdelivery. Eur J Obstet Gynecol Reprod Biol 1991;40:203–9.

30. Saleh Gargari S, Habibolahi M, Zonobi Z, Khani Z, Sarfjoo F,Kazemi Robati A, et al. Outcome of vaginal progesterone as




RichardResaltado

a tocolytic agent: randomized clinical trial. ISRN Obstet Gyne-col 2012;2012:607906.

31. Facchinetti F, Paganelli S, Comitini G, Dante G, Volpe A. Cer-vical length changes during preterm cervical ripening: effects of17-alpha-hydroxyprogesterone caproate. Am J Obstet Gynecol2007;196:453.e1–4.

32. Lotfalizadeh M, Ghomian N, Reyhani A. The effects of pro-gesterone therapy on the gestation length and reduction of neo-natal complications in patients who had received tocolytictherapy for acute phase of preterm labor. Iran Red CrescentMed J 2013;15:e7947.

33. Sharami SH, Zahiri Z, Shakiba M, Milani F. Maintenance ther-apy by vaginal progesterone after threatened idiopathic pretermlabor: a randomized placebo-controlled double-blind trial. Int JFertil Steril 2010;4:45–50.

34. Regmi M, Rijal P, Agrawal A, Uprety D. Progesterone forprevention of recurrent preterm labor after arrested pretermlabor—a randomized controlled trial. Gynecol Obstet 2012:2:125.

35. El-Abidin E, Mahmood M, Rashwan H, Hanafy A. Efficacy ofvaginal progesterone for maintenance tocolysis: a recent trendin the management of arrested preterm labour. Med J CairoUniv 2009;77:99–104.

36. Mishra G, Inamdar S. Role of micronised progesterone in main-tenance therapy following arrested preterm labor: a randomisedcontrolled trial. Int J Reprod Contracept Obstet Gynecol 2014;3:1050–55.

37. Eke AC, Chalaan T, Shukr G, Eleje GU, Okafor CI. A system-atic review and meta-analysis of progestogen use for mainte-nance tocolysis after preterm labor in women with intactmembranes. Int J Gynaecol Obstet 2016;132:11–6.




prematurity: review progestogens as maintenance treatment in … · 2016. 11. 8. · prematurity:...

Documents