MATERNAL-FETAL MEDICINE

Progestogens for preterm birth prevention: a systematic reviewand meta-analysis by drug route

Digna R. Velez Edwards • Frances E. Likis • Jeffrey C. Andrews •

Alison L. Woodworth • Rebecca N. Jerome • Christopher J. Fonnesbeck •

J. Nikki McKoy • Katherine E. Hartmann

Received: 14 December 2012 / Accepted: 5 March 2013 / Published online: 27 March 2013

� Springer-Verlag Berlin Heidelberg 2013

Abstract

Purpose Progestogen has been investigated as a pre-

ventive intervention among women with increased preterm

birth risk. Our objective was to systematically review the

effectiveness of intramuscular (IM), vaginal, and oral

progestogens for preterm birth and neonatal death

prevention.

Methods We included articles published from January

1966 to January 2013 and found 27 randomized trials with

data for Bayesian meta-analysis.

Results Across all studies, only vaginal and oral routes

were effective at reducing preterm births (IM risk ratio

[RR] 0.95, 95 % Bayesian credible interval [BCI]:

0.88–1.03; vaginal RR 0.87, 95 % BCI: 0.80–0.94; oral RR

0.64, 95 % BCI: 0.49–0.85). However, when analyses were

limited to only single births all routes were effective at

reducing preterm birth (IM RR 0.77, 95 % BCI: 0.69–0.87;

vaginal RR 0.80, 95 % BCI: 0.69–0.91; oral RR 0.66, 95 %

BCI: 0.47–0.84). Only IM progestogen was effective at

reducing neonatal deaths (IM RR 0.78, 95 % BCI:

0.56–0.99; vaginal RR 0.75, 95 % BCI: 0.45–1.09; oral RR

0.72, 95 % BCI: 0.09–1.74). Vaginal progestogen was

effective in reducing neonatal deaths when limited to sin-

gletons births.

Conclusions All progestogen routes reduce preterm births

but not neonatal deaths. Future studies are needed that

directly compare progestogen delivery routes.

Keywords Preterm birth � Meta-analysis � Review �Pregnancy � Progestogen

Electronic supplementary material The online version of thisarticle (doi:10.1007/s00404-013-2789-9) contains supplementarymaterial, which is available to authorized users.

D. R. Velez Edwards (&) � J. C. Andrews � K. E. Hartmann

Vanderbilt Epidemiology Center, Institute for Medicine and

Public Health, 2525 West End Ave., Suite 600 6th Floor,

Nashville, TN 37203, USA

e-mail: digna.r.velez.edwards@vanderbilt.edu

D. R. Velez Edwards � J. C. Andrews � K. E. Hartmann

Department of Obstetrics and Gynecology, Vanderbilt

University Medical Center, Nashville, TN, USA

F. E. Likis � K. E. Hartmann

Vanderbilt Evidence-based Practice Center, Institute for

Medicine and Public Health, Nashville, TN, USA

F. E. Likis � J. Nikki McKoy

Department of Medicine, Vanderbilt University Medical Center,

Nashville, TN, USA

A. L. Woodworth

Department of Pathology, Microbiology and Immunology,

Nashville, TN, USA

R. N. Jerome

Department of Biomedical Informatics, Nashville, TN, USA

R. N. Jerome

Eskind Biomedical Library, Vanderbilt University Medical

Center, Nashville, TN, USA

C. J. Fonnesbeck

Department of Biostatistics, Vanderbilt University Medical

Center, Nashville, TN, USA

123

Arch Gynecol Obstet (2013) 287:1059–1066

DOI 10.1007/s00404-013-2789-9

Purpose

Preterm birth is the predominant cause of perinatal mor-

tality in the United States [1]. Efforts to predict impending

preterm birth and delay its occurrence have experienced

limited success [2–4], thus a focus on primary intervention

has become a priority [5]. Attention has increasingly

focused on methods to prevent preterm birth using earlier

interventions to treat women based on risks rather than

symptoms.

Progestogen administration has been investigated as a

preventive intervention across several groups of women

with increased risk of preterm birth, such as those with

prior preterm birth, multiple gestation, a short cervix, or

symptoms of preterm labor. Progestogens, the encom-

passing term for natural progesterone, pharmacologic

agents that are chemically identical to endogenous pro-

gesterone, and synthetic progestins that are similar but not

identical in chemical structure [6], can be administered

orally, vaginally, or via injection. In February 2011, the US

Food and Drug Administration approved the Makena

(hydroxyprogesterone caproate) injection for the preven-

tion of preterm birth for women with a history of at least

one spontaneous preterm birth [7]. However, in January of

2012, the FDA did not approve progesterone gels for the

prevention of preterm birth among at-risk women with a

short cervix.

There is currently limited understanding about the

pharmaceutical mechanisms of action of progestogens and

there are no studies directly comparing progestogen drug

delivery routes. This is of concern to clinicians prescribing

progestogens for the prevention of preterm birth due to the

substantial expense of administering progestogens, the

poor understanding of the effects of sex hormones on

pregnant women and their fetuses, and the multiple for-

mulations and delivery routes that are available. In order to

build upon prior systematic reviews that have examined the

effectiveness of individual progestogen delivery routes

within subsets of women with specific indications for use,

we examined the effectiveness across the different pro-

gestogen delivery routes and formulations for preventing

preterm birth [8–12]. Our objective was to systematically

review the effectiveness of intramuscular, vaginal, and oral

administration of progestogens for prevention of preterm

birth.

Methods

Sources

We searched MEDLINE� and EMBASE for English lan-

guage articles published from January 1966 to January

2013. Controlled vocabulary terms served as the founda-

tion of our search, complemented by additional keyword

phrases to represent the myriad ways in which progesto-

gens and preterm labor are referred to in the clinical lit-

erature [13]. We also hand-searched references of included

articles to identify studies. This was a systematic review of

existing literature. No IRB approval is required for syn-

thesis of findings from existing literature.

Study selection

We included randomized controlled trials (RCTs) with 20

or more women in order to have adequate power for sta-

tistical analyses. We included all formulations and drug

delivery routes. Our searches were executed between

August 2009 and January 2013.

Two reviewers separately evaluated the abstracts for

inclusion or exclusion. If one reviewer concluded the

article could be eligible based on the abstract, we retained

it. Full publications were then independently reviewed for

final inclusion. All team members shared the task of

entering information into evidence tables [13]. After data

extraction, another member checked table entries for

accuracy, completeness, and consistency; inconsistencies

were adjudicated by team members. The team evaluating

the literature included two physicians, a certified nurse-

midwife and nurse practitioner, three biomedical

researchers, and two library scientists.

The primary outcomes extracted from articles were

preterm birth [less than 33 (singleton with short cervix), 34

(multiples and singleton with short cervix), 35 (multiples),

Non duplicate articles identified in search

n = 524Literature search: n = 433Hand-search: n = 91

Full text articles reviewedn = 204

Articles excluded:n = 320

Articles excluded:n = 177

• Not related to the use of progestogens to prevent PTB (n= 83)

• Did not address study question(n=19)

• Not original research (n=33)• Ineligible study size (n=53)• Ineligible study type/not RCT

(n=5)• Ineligible secondary analysis of

primary data already included (n=1)

Unique full text articles included

in reviewn = 27 RCTs

Fig. 1 Flow chart of study selection. The number of manuscripts

excluded based on the different criteria in the fourth box do not add

up to the articles excluded (n = 177) because manuscripts may have

met more than one exclusion criteria

1060 Arch Gynecol Obstet (2013) 287:1059–1066

123

and 37 (singleton) weeks’ gestation] and neonatal death.

We also limited analyses only to major indications for

progestogen treatment that include prior preterm births,

preterm labor, short cervix, and multiple gestations. From

among 524 abstracts, and the review yielded a total of 27

full publications including relevant data for meta-analysis

(Fig. 1).

We conducted a Bayesian meta-analysis [14, 15] to

provide aggregate estimates of the effectiveness of pro-

gestogen treatment for preventing preterm birth and

reducing neonatal death. We constructed mixed-effects

models grouping RCTs by fixed treatment effects for the

progestogen drug delivery route for which the progestogens

were administered in the study (prior preterm births, pre-

term labor, short cervix, and multiple gestations). Study-

level baseline variation was modeled using random effects.

A total of 27 studies were included in the meta-analyses: 13

studies used injected 17-hydroxyprogesterone (17OHP) as

the drug delivery route, 11 used vaginal, and three used

oral. Due to an insufficient number of studies (1) for oral

delivery, this delivery route was excluded from the neo-

natal death outcome analysis. Statistical models or preterm

birth outcome also included a covariate adjustment for the

gestational age cut-off used to define preterm birth status,

as well as indications for treatment. Finally, two additional

sub-analyses were performed limiting studies to singleton

births and multiple gestations in order to assess whether the

effectiveness of the drug delivery route varied by singleton

compared to multiple gestation.

Meta-analyses were implemented in PyMC version 2.1

[16], to fit Bayesian hierarchical models using Markov

chain Monte Carlo (MCMC) algorithms [14]. To check the

fit of the model, we conducted posterior predictive checks,

which generate simulated datasets based on the fitted

model. The distribution of simulated datasets was then

compared to the observed data from the studies in the meta-

analysis. The observed data for each study fell within the

2.5th and 97.5th quantiles of the distributions of datasets

simulated from our model, suggesting acceptable fit of the

model to the data, with the exception of two studies of

preterm birth that had poor fit [17, 18]. Posterior estimates

of risk ratios (RRs) with 95 % Bayesian credible intervals

(BCIs) are provided.

To assess quality, two reviewers independently used a

worksheet to capture key elements of study design and

conduct [19]. Inconsistencies were resolved through review

of the publications, discussion, and consensus with the

team. Assessment of internal validity included: randomized

allocation, method of randomization, use of masking,

description of inclusion criteria, loss to follow-up, dropout

rates, power calculation, and recognition and description of

statistical issues. Assessment of external validity included:

description of baseline population, clear specification of

intervention, description of the outcomes, length of follow-

up for infant, outcome measurement, and reliability of

outcome measurement. A composite quality score of good,

fair, or poor was calculated for both internal and external

validity, and the study as a whole [13]. Individual study

quality scores are provided in Supplemental Material

(Table 1S).

Extracted data were used to assess the strength of evi-

dence of the aggregate literature for assessing effectiveness

of the intervention in relation to specific outcomes.

Strength was assessed in four domains: risk of bias, con-

sistency, directness and precision [20]. The possible grades

were: (1) High confidence that the evidence reflects the true

effect. Further research is unlikely to change estimates;

(2) Moderate confidence that the evidence reflects the true

effect. Further research may change our confidence in the

estimate of effect and may change the estimate; (3) Low

confidence that the evidence reflects the true effect. Further

research is likely to change confidence in the estimate

of effect and is also likely to change the estimate; or

(4) Insufficient meaning evidence is either unavailable or

does not permit a conclusion [13].

Results

We identified 27 publications meeting the criteria for

inclusion in meta-analysis [3, 4, 17, 18, 21–43]. Studies

that were secondary analyses of the same datasets were

counted only once. The most common progestogen in these

studies was the synthetic progestin 17OHP. Other inject-

able forms of progesterone used include crystalline pro-

gesterone and natural progesterone. Vaginal progestogens

were administered via suppositories, gels, and capsules.

Oral progestogens included medroxyprogesterone acetate

(trade names Provera and Perlutex), allylestrenol, and oral

chlormadinone acetate. Of these oral formulations, only

medroxyprogesterone acetate is currently available in the

United States. Across all studies, the most common pro-

gestogens and doses, by route, were: 250 mg of 17OHP

injected intramuscularly weekly, 200 mg progesterone

vaginal suppository or gel daily, and 100 to 1,000 mg oral

micronized progestogen daily. RCTs making direct com-

parisons of one form of progestogen to another have not

been conducted with currently available progestogens. A

single 1979 study comparing intramuscular 17OHP injec-

tions and oral chlormadinone acetate was the only head-

to-head comparison; however, it did not meet criteria for

inclusion in meta-analysis [44]. Among the studies that

reported on intramuscular 17OHP, vaginal, and oral pro-

gestogen the majority of studies (n = 9) were conducted in

the US [22–25, 28–30, 34, 38], seven in Europe [3, 4, 18,

26, 32, 35, 39], six in Asia [21, 33, 37, 40–42], three from

Arch Gynecol Obstet (2013) 287:1059–1066 1061

123

Table 1 Summary of RCTs by progestogen drug delivery route

Study country Formulation Dosea Outcomeb Effectc Target EGA, start; end

(weeks’)

Injected 17OHP

Grobman et al. [30] US IM 250 mg q 7d \35

\37

ND

RR = 0.84 (0.58,

1.21)

RR = 1.03 (0.79,

1.35)

RR = 0.76 (0.27,

2.16)

16–20; 36 ? 6

Rozenberg et al. [40]

France

IM 500 mg b.i.w. \34

\37

ND

NR p = 0.69

NR p = 0.76

NR p [ 0.99

24-31 ? 6; 36

Tan et al. [43] Malaysia IM 250 mg (single dose) \34

\37

RR = 0.61 (0.30,

1.27)

RR = 0.85 (0.63,

1.45)

22-35; NR

Combs et al. [26] US IM 250 mg q 7d \34 RR = 1.4 (0.7, 2.6) 16–23; 34

Lim et al. [33]

Netherlands

IM 250 mg q 7d ND Only RR = 0.60 (0.25,

1.43)

16-20; 36

Ibrahim et al. [32] Egypt IM 250 mg q 7d \37 RR = 0.079 (0.021,

0.302)

[14; 36

Combs et al. [25] US IM 250 mg q 7d \ 35 NR p = 0.15 \21 (80.4 %); 36

Caritis et al. [24] US IM 250 mg q 7d \ 35 RR = 1.0 (0.9, 1.1) 16–21; 35

Briery et al. [23] US IM 250 mg q 7d \35 NR p = 0.117 20–30; NR

Facchinetti et al. [27] Italy IM 341 mg q 4d \ 37 NR p = 0.049 25–34; 36

Rouse et al. [39] US IM 250 mg q 7d \35 RR = 1.1 (0.09, 1.3) 16–20; 36

Facchinetti et al. [3] Italy IM 341 mg q 4d \37 NR p = 0.004 25–34; 36

Meis et al. [35] US IM 250 mg q 7d \37 RR = 0.66 (0.54,

0.81)

16–21; 36

Vaginal

Saleh Gargari et al. [41]

Iran

Vaginal suppository 400 mg qd ND only NR p = 0.08 24–34; 37

Wood et al. [44] Canada Vaginal gel 90 mg qd \35

\37

RR = 0.87 (0.47,

1.59)

RR = 0.93 (0.66,

1.30)

16-20; 35 ? 6

Serra et al. [18] Spain VaginalPessaries 200 mg qd \34 NR p = NS 20–34; 34

400 mg qd \37 ND NR p = NS

NR p = NS

Rode et al. [17]

Multinational

VaginalPessaries 200 mg qd \34

\37

ND

OR = 0.8 (0.5–1.2)

OR = 0.8 (0.6–1.1)

NR

20–23 ? 6; 33 ? 6

Arikan et al. [22] Turkey Vaginal (NR) 200 mg qd \37 NR p = NS 24–34; 37

Fonseca et al. [28]

Multinational

Vaginal capsule 200 mg qd \34 RR = 0.60 (0.35,

0.94)

24; 34

Sharami et al. [42] Iran VaginalSuppository 200 mg qd \37 NR p = 0.098 28–36; 36

Hassan et al. [31] US Vaginal gel 90 mg qd \33 RR = 0.55 (0.33,

0.92)

C24 (98.5 %); 34

Majhi et al. [34] India Vaginal capsule 100 mg qd \37 RR = 0.315 (0.137,

0.724)

20–24; 36

O’Brien et al. [37]

Multinational

Vaginal gel 90 mg qd \37 OR = 0.93 (0.66,

1.32)

16–23; 37

1062 Arch Gynecol Obstet (2013) 287:1059–1066

123

multiple countries [17, 27, 36], one from Canada [43], and

one form Africa [42].

Intramuscular 17OHP Injection

We identified thirteen RCTs in which intramuscular

17OHP injections were administered for prevention of

preterm birth [3, 22–26, 29, 31, 32, 34, 38, 39, 42]. Exact

combinations of dose, interval, and target gestational age

windows are included in Table 1. Three different indica-

tions were used for intramuscular 17OHP treatment

including prior preterm birth in one trial [34], preterm

labor in five [3, 26, 31, 39, 42], multiple gestation in six

[22–25, 32, 38], and short cervix in one [29]. Four of

thirteen demonstrated effectiveness of 17OHP [3, 26, 31,

34], nine had nonsignificant findings [22–25, 29, 32, 38,

39, 42], and none found significant advantage for the

placebo group. One study only had data for neonatal death

analyses [32]. Aggregate estimates indicated that intra-

muscular17OHP injections were effective at reducing risk

for neonatal death (meta-estimate RR 0.78, 95 % BCI 0.56

to 0.99) but not effective at reducing risk of preterm birth

(meta-estimate RR 0.95, 95 % BCI 0.88 to 1.03) (Fig. 2).

However, when limiting to only singleton births meta-

estimates for both neonatal death (meta-estimate RR 0.53,

95 % BCI 0.22 to 0.85) and preterm birth showed evi-

dence of protection (meta-estimate RR 0.77, 95 % BCI

0.69 to 0.87). Among multiple gestations intramuscular

17OHP was not effective at reducing risk or neonatal

deaths (meta-estimate RR 0.93, 95 % BCI 0.61 to 1.28) or

preterm births (meta-estimate RR 1.05, 95 % BCI 0.95 to

1.15).

Vaginal

We identified eleven publications that reported on a vaginal

gel, capsule, or suppository for administering progestogen

treatment (Table 1) [17, 18, 21, 27, 30, 33, 35, 36, 40, 41, 43].

These studies included two different doses. The indication

for treatment was history of prior preterm birth in two

studies [33, 36], preterm labor in three [21, 40, 41], multiple

gestation in three [17, 18, 35, 43], and short cervix in two

[27, 30]. Overall, three of eleven demonstrated effective-

ness of vaginal progestogens [27, 30, 33], eight had non-

significant findings [17, 18, 21, 35, 36, 40, 41, 43], and none

found significant advantage for the placebo group. One

study only had data for neonatal death analyses [40].

Aggregate estimates indicated that vaginal progestogens

was not effective at reducing risk for neonatal death (meta-

estimate RR 0.75, 95 % BCI 0.45 to 1.09), but was effective

at reducing risk of preterm birth (meta-estimate RR 0.87,

95 % BCI 0.80 to 0.94) (Fig. 2). The meta-estimates again

showed evidence for protection from neonatal death (meta-

estimate RR 0.47, 95 % BCI 0.20 to 0.0.75) and preterm

birth (meta-estimate RR 0.80, 95 % BCI 0.69 to 0.91)

outcomes after limiting analyses to singleton births. The

meta-estimates did not show evidence for protection from

neonatal deaths (meta-estimate RR 1.20, 95 % BCI 0.53

to 1.92) or preterm births among multiple gestations

(meta-estimate RR 0.93, 95 % BCI 0.83 to 1.03).

Table 1 continued

Study country Formulation Dosea Outcomeb Effectc Target EGA, start; end

(weeks’)

Norman et al. [36] UK Vaginal gel 90 mg qd \34 OR = 0.74 (0.48,

1.12)

24; 34

Oral

Glover et al. [29] US Progesterone

(Micronized)

400 mg qd \37 RR = 0.55 (0.26,

1.16)

16–19; 33.9

Rai et al. [38] India Progesterone

(Micronized)

100 mg b.i.d \37 NR p = 0.002 18–24; 36

Noblot et al. [4] France Progesterone

(Micronized)

4 9 100 mg q6 h for

24 h;

4 9 100 mg q8 h for

24 h; then

3 9 100 mg q8 h

\37 NR p = NS NR

EGA estimated gestational agea Dose reported in milligrams, q 7d and q 4d indicates every seven and every four days, b.i.d. twice per day, b.i.w. twice per week, q6 h once

every 6 h, q8 h once every 8 hb ND neonatal deathc NS not statistically significant, NR not reported, RR relative risk

Arch Gynecol Obstet (2013) 287:1059–1066 1063

123

Oral

Three RCTs used oral progestogens alone or in combination

with ritodrine in the treatment group [4, 28, 37]. The three

RCTs reported prematurity outcomes at less than 37 week’s

(Table 1). Each of the three RCTs used different doses. Two

had prior preterm birth as an indication for treatment [28,

37] and one had preterm labor [4]. One of three demon-

strated effectiveness of oral progestogens [37], two had

non-significant findings [4, 28], and none found significant

advantage for the placebo group. Aggregate estimates

indicated that oral progestogens were the most effective at

reducing risk of preterm birth relative to the meta-estimates

for 17OHP and vaginal progestogens (meta-estimate RR

0.64, 95 % BCI 0.49 to 0.85) (Fig. 2). There was not suf-

ficient data to analyze neonatal death outcome among oral

progestogen studies. None of the studies included multiple

gestations as an indication for treatment and therefore no

stratified analyses were performed.

Quality of evidence

Overall the strength of evidence was insufficient or low.

Deficiencies in the strength of evidence most often related to

a preponderance of study designs with high-risk of bias;

inconsistent findings across studies and inconsistencies

among outcomes that would be expected to show corre-

sponding benefit; use of intermediate outcomes; and small

studies with poor precision. The quality ratings of individual

studies are provided in Supplemental Material (Table 1S).

Conclusions

In this systematic review of progestogens for prevention of

preterm birth, we sought evidence that drug delivery route

might modify treatment response. No direct comparisons of

one progestogen drug delivery route to another were

identified that would allow us to directly compare the

effectiveness of the formulations and routes of the pro-

gestogen intervention. However, meta-analyses stratifying

studies by drug delivery routes indicated that all three drug

delivery routes (intramuscular 17OHP injection, oral, and

vaginal) were effective at reducing the number of preterm

births across studies. The meta-estimates for risk of neo-

natal death indicated that intramuscular 17OHP and vagi-

nal progestogen routes were effective at reducing neonatal

death among singleton births.

Fig. 2 Estimates of risk ratio

for each study (with 95 %

Bayesian credible interval),

along with meta-estimates for

the effectiveness of each

progestogen drug delivery route

for the prevention of preterm

birth. Results for meta-analyses

of RCTs by drug delivery route

are presented for: intramuscular

17OHP injection (IM); vaginal

progestogen; oral progestogen.

The solid line on the X axis

indicates where RR = 1 and the

bolded line indicates the overall

meta-estimate RR. Black circlesindicate the RR for each

individual RCT and the overall

meta-estimate RR, while

horizontal bounded black linesindicate the 95 % BCI. Effect

sizes for individual studies were

calculated from analyses of data

abstracted from the published

manuscripts

1064 Arch Gynecol Obstet (2013) 287:1059–1066

123

In order to make direct comparisons across drug for-

mulations and delivery route other factors of study design

would need to be fully comparable to allow isolation of the

factor of interest such as drug delivery route. Unfortu-

nately, no head-to-head trials of currently available pro-

gestogens have been conducted (one 1979 trial of poor

quality is the only publication) [44]. Although we were

able to assess the consistency of effect size estimates by

drug delivery routes through meta-analysis, we were not

able to directly compare and determine the effectiveness of

one route compared to another without a head-to-head

comparison. As an extension of the lack of direct com-

parisons, it is not possible to determine with confidence

whether acceptability, adherence, adverse effects, or safety

of progestogens vary by progestogen route. Without head-

to-head comparisons it is possible that differences in

effectiveness of progestogen treatments arise solely from

underlying differences in populations.

We found no direct comparator studies that allowed us to

assess how the gestational age timing of progestogen drug

delivery may be influenced by drug delivery route. Among

the studies examined, only two clinical care cohorts had

varied timing of initiation of treatment (for reasons other

than randomization) and showed no significant difference in

preterm birth rates based upon whether the progestogen was

initiated before or after 21 weeks gestation [45, 46]. No

RCTs directly addressed modification of effectiveness by

timing of initiation by drug delivery route. Given variation

in pharmaceutical agents being studied, it is not possible to

extrapolate from trends in study findings to determine an

optimal time for initiation of treatment.

Without direct comparisons of different progestogen

drug delivery routes we were unable to assess whether

benefit or risk of harm varies for different progestogen drug

delivery routes. However, a detailed description of known

harms and side effects of progestogen treatments has been

previously published [13]. Reported side effects varied

with route of administration, the most common being

injection site reactions and vaginal discharge. However,

more severe effects have been reported, but are much less

common [13]. Given lack of detailed reporting about harms

and lack of consistent definitions, it is not meaningful to

extrapolate among routes from what data are available.

Harms data were not uniformly collected so comparisons

across studies cannot provide meaningful data to inform

clinical decisions. Overall, there is no evidence to help

inform selection of the progestogen with the fewest side

effects and/or lowest risk of harms.

The overall strength of evidence was insufficient or low,

supporting the need for further research into progestogen

drug delivery route. This review is built upon prior reviews

of progestogen delivery route by comprehensively exam-

ining progestogen indications for use and delivery routes,

as well by providing a more updated assessment of the

literature. The findings of this systematic review indicate

that all routes of progestogen are effective at reducing

preterm birth risk and worthy of further investigation, as

route is likely to influence acceptability, adherence, and

costs. Head-to-head comparisons of benefits and harms are

needed to determine the ideal route.

Acknowledgments This project was supported by the Agency for

Healthcare Research and Quality (contract number: 290-2007-10065-I).

We are indebted to a tireless and exceptional group of colleagues who

made this report possible. Each step of a systematic review draws on

the skills and attention of an entire team. Additional members of our

team include Dr. Steven Fox, Dr. Melissa McPheeters, Ms. Rachel

Weaver, Mr. Jeffrey Seroogy, Ms. Allison Glasser, Ms. Cheena

Clermont, Dr. Chris Slaughter, and Ms. Tracy Shields.

Conflict of interest We declare that we have no conflict of

interest.

References

1. Behrman RE, Butler AS, Institute of Medicine (US) (2007)

Committee on understanding premature birth and assuring heal-

thy outcomes. Preterm birth: causes, consequences, and preven-

tion. National Academies Press, Washington, xvi, 772

2. Borna S, Sahabi N (2008) Progesterone for maintenance tocolytic

therapy after threatened preterm labour: a randomised controlled

trial. Aust N Z J Obstet Gynaecol 48:58–63

3. Facchinetti F, Paganelli S, Comitini G, Dante G, Volpe A (2007)

Cervical length changes during preterm cervical ripening: effects

of 17-alpha-hydroxyprogesterone caproate. Am J Obstet Gynecol

196:453

4. Noblot G, Audra P, Dargent D, Faguer B, Mellier G (1991) The

use of micronized progesterone in the treatment of menace of

preterm delivery. Eur J Obstet Gynecol Reprod Biol 40:203–209

5. Kauppila A, Hartikainen-Sorri AL, Janne O, Tuimala R, Jarvinen

PA (1980) Suppression of threatened premature labor by

administration of cortisol and 17 alpha-hydroxyprogesterone

caproate: a comparison with ritodrine. Am J Obstet Gynecol

138:404–408

6. Brucker MC, Likis FE (2011) Steroid hormones. Pharmacology

for Women’s Health

7. FDA approves drug to reduce risk of preterm birth in at-risk preg-

nant women (2011) United States Food and Drug Administration

8. Dodd JM, Flenady V, Cincotta R, Crowther CA (2006) Prenatal

administration of progesterone for preventing preterm birth.

Cochrane Database Syst Rev CD004947

9. Gallo MF, Grimes DA, Lopez LM, Schulz KF, d’Arcangues C

(2008) Combination injectable contraceptives for contraception.

Cochrane Database Syst Rev CD004568

10. Gallo MF, Nanda K, Grimes DA, Lopez LM, Schulz KF (2008)

20 lg versus [20 lg estrogen combined oral contraceptives for

contraception. Cochrane Database Syst Rev CD003989

11. Romero R, Nicolaides K, Conde-Agudelo A, Tabor A, O’Brien

JM, Cetingoz E et al (2012) Vaginal progesterone in women with

an asymptomatic sonographic short cervix in the midtrimester

decreases preterm delivery and neonatal morbidity: a systematic

review and metaanalysis of individual patient data. Am J Obstet

Gynecol 206:124

Arch Gynecol Obstet (2013) 287:1059–1066 1065

123

12. Smith V, Devane D, Begley CM, Clarke M, Higgins S (2009) A

systematic review and quality assessment of systematic reviews

of randomised trials of interventions for preventing and treating

preterm birth. Eur J Obstet Gynecol Reprod Biol 142:3–11

13. Likis FE, Andrews JC, Woodworth AL, Velez Edwards DR,

Jerome RN, Fonnesbeck CJ et al (2012) Progestogens for pre-

vention of preterm birth. Agency for Healthcare Research and

Quality, Rockville

14. Gelman A, Carlin J, Stern H, Rubin DB (2004) Bayesian Data

Analysis

15. Sutton A, Abrams K (2001) Bayesian methods in meta-analysis

and evidence synthesis

16. Patil A, Huard D, Fonnesbeck C (2010) Bayesian stochastic

modelling in python. J Stat Softw 35:1–80

17. Rode L, Klein K, Nicolaides KH, Krampl-Bettelheim E, Tabor A

(2011) Prevention of preterm delivery in twin gestations (PRE-

DICT): a multicenter, randomized, placebo-controlled trial on the

effect of vaginal micronized progesterone. Ultrasound Obstet

Gynecol 38:272–280

18. Serra V, Perales A, Meseguer J, Parrilla JJ, Lara C, Bellver J et al

(2013) Increased doses of vaginal progesterone for the prevention

of preterm birth in twin pregnancies: a randomised controlled

double-blind multicentre trial. BJOG Int J Obstet Gynaecol

120:50–57

19. Schulz KF, Chalmers I, Hayes RJ, Altman DG (1995) Empirical

evidence of bias. Dimensions of methodological quality associ-

ated with estimates of treatment effects in controlled trials.

JAMA 273:408–412

20. Hadjigeorgiou E, Malamitsi-Puchner A, Lolis D, Lazarides P,

Nicolopoulos D, Kaskarelis D (1982) Cardiovascular birth

defects and antenatal exposure to female sex hormones. Dev

Pharmacol Ther 5:61–67

21. Arikan I, Barut A, Harma M, Harma IM (2011) Effect of pro-

gesterone as a tocolytic and in maintenance therapy during pre-

term labor. Gynecol Obstet Invest 72:269–273

22. Briery CM, Veillon EW, Klauser CK, Martin RW, Chauhan SP,

Magann EF et al (2009) Progesterone does not prevent preterm

births in women with twins. South Med J 102:900–904

23. Caritis SN, Rouse DJ, Peaceman AM, Sciscione A, Momirova V,

Spong CY et al (2009) Prevention of preterm birth in triplets

using 17 alpha-hydroxyprogesterone caproate: a randomized

controlled trial. Obstet Gynecol 113:285–292

24. Combs CA, Garite T, Maurel K, Das A, Porto M (2010) Failure

of 17-hydroxyprogesterone to reduce neonatal morbidity or pro-

long triplet pregnancy: a double-blind, randomized clinical trial.

Am J Obstet Gynecol 203:248

25. Combs CA, Garite T, Maurel K, Das A, Porto M (2011)

17-hydroxyprogesterone caproate for twin pregnancy: a double-

blind, randomized clinical trial. Am J Obstet Gynecol 204:221

26. Facchinetti F, Dante G, Venturini P, Paganelli S, Volpe A (2008)

17alpha-hydroxy-progesterone effects on cervical proinflamma-

tory agents in women at risk for preterm delivery. Am J Perinatol

25:503–506

27. Fonseca EB, Celik E, Parra M, Singh M, Nicolaides KH (2007)

Progesterone and the risk of preterm birth among women with a

short cervix. N Engl J Med 357:462–469

28. Glover MM, McKenna DS, Downing CM, Smith DB, Croom CS,

Sonek JD (2011) A randomized trial of micronized progesterone

for the prevention of recurrent preterm birth. Am J Perinatol

28:377–381

29. Grobman WA, Thom EA, Spong CY, Iams JD, Saade GR,

Mercer BM et al (2012) 17 alpha-hydroxyprogesterone caproate

to prevent prematurity in nulliparas with cervical length less than

30 mm. Am J Obstet Gynecol 207:390

30. Hassan SS, Romero R, Vidyadhari D, Fusey S, Baxter JK,

Khandelwal M et al (2011) Vaginal progesterone reduces the rate

of preterm birth in women with a sonographic short cervix: a

multicenter, randomized, double-blind, placebo-controlled trial.

Ultrasound Obstet Gynecol 38:18–31

31. Ibrahim M, Mohamed Ramy AR, Younis MAF (2010) Proges-

terone supplementation for prevention of preterm labor: A ran-

domized controlled trial. 15 Middle East. Fertility Soc J 1:39–41

32. Lim AC, Schuit E, Bloemenkamp K, Bernardus RE, Duvekot JJ,

Erwich JJ et al (2011) 17alpha-hydroxyprogesterone caproate for

the prevention of adverse neonatal outcome in multiple pregnan-

cies: a randomized controlled trial. Obstet Gynecol 118:513–520

33. Majhi P, Bagga R, Kalra J, Sharma M (2009) Intravaginal use of

natural micronised progesterone to prevent pre-term birth: a

randomised trial in India. J Obstet Gynaecol 29:493–498

34. Meis PJ, Klebanoff M, Thom E, Dombrowski MP, Sibai B,

Moawad AH et al (2003) Prevention of recurrent preterm delivery

by 17 alpha-hydroxyprogesterone caproate. N Engl J Med 348:

2379–2385

35. Norman JE, Mackenzie F, Owen P, Mactier H, Hanretty K,

Cooper S et al (2009) Progesterone for the prevention of preterm

birth in twin pregnancy (STOPPIT): a randomised, double-blind,

placebo-controlled study and meta-analysis. Lancet 373:2034–

2040

36. O’Brien JM, Adair CD, Lewis DF, Hall DR, Defranco EA, Fusey

S et al (2007) Progesterone vaginal gel for the reduction of

recurrent preterm birth: primary results from a randomized,

double-blind, placebo-controlled trial. Ultrasound Obstet Gyne-

col 30:687–696

37. Rai P, Rajaram S, Goel N, Ayalur Gopalakrishnan R, Agarwal R,

Mehta S (2009) Oral micronized progesterone for prevention of

preterm birth. Int J Gynaecol Obstet 104:40–43

38. Rouse DJ, Caritis SN, Peaceman AM, Sciscione A, Thom EA,

Spong CY et al (2007) A trial of 17 alpha-hydroxyprogesterone

caproate to prevent prematurity in twins. N Engl J Med 357:454–461

39. Rozenberg P, Chauveaud A, Deruelle P, Capelle M, Winer N,

Desbriere R et al (2012) Prevention of preterm delivery after

successful tocolysis in preterm labor by 17 alpha-hydro-

xyprogesterone caproate: a randomized controlled trial. Am J

Obstet Gynecol 206:e1–9

40. Saleh Gargari S, Habibolahi M, Zonobi Z, Khani Z, Sarfjoo FS,

Kazemi Robati A et al (2012) Outcome of vaginal progesterone

as a tocolytic agent: Randomized clinical trial. ISRN Obstet

Gynecol 2012:607906

41. Sharami SH, Zahiri Z, Shakiba M, Milani F (2010) Maintenance

therapy by vaginal progesterone after threatened idiopathic pre-

term labor: a randomized placebo-controlled double-blind trial.

Int J Fert Steril 4:45–50

42. Tan PC, King AS, Vallikkannu N, Omar SZ (2012) Single dose

17 alpha-hydroxyprogesterone caproate in preterm labor: a ran-

domized trial. Arch Gynecol Obstet 285:585–590

43. Wood S, Ross S, Tang S, Miller L, Sauve R, Brant R (2012)

Vaginal progesterone to prevent preterm birth in multiple preg-

nancy: a randomized controlled trial. J Perinat Med 40:593–599

44. Breart G, Lanfranchi M, Chavigny C, Rumeau-Rouquette C,

Sureau C (1979) A comparative study of the efficiency of

hydroxyprogesterone caproate and of chlormadinone acetate in

the prevention of premature labor. Int J Gynaecol Obstet 16:

381–384

45. Gonzalez-Quintero VH, Istwan NB, Rhea DJ, Smarkusky L,

Hoffman MC, Stanziano GJ (2007) Gestational age at initiation

of 17-hydroxyprogesterone caproate (17P) and recurrent preterm

delivery. J Matern Fetal Neonatal Med 20:249–252

46. How HY, Barton JR, Istwan NB, Rhea DJ, Stanziano GJ (2007)

Prophylaxis with 17 alpha-hydroxyprogesterone caproate for

prevention of recurrent preterm delivery: does gestational age at

initiation of treatment matter? Am J Obstet Gynecol 197:260

1066 Arch Gynecol Obstet (2013) 287:1059–1066

123