vaginal progesterone vs cervical cerclage for the ... · birth.67-74,116-122 it is unlikely that...
TRANSCRIPT
t
aELSa
R
S
Reports of Major Impact www.AJOG.org
Vaginal progesterone vs cervical cerclage for the preventionof preterm birth in women with a sonographic short cervix,previous preterm birth, and singleton gestation: a systematicreview and indirect comparison metaanalysisAgustin Conde-Agudelo, MD, MPH; Roberto Romero, MD, DMedSci; Kypros Nicolaides, MD; Tinnakorn Chaiworapongsa, MD;John M. O’Brien, MD; Elcin Cetingoz, MD; Eduardo da Fonseca, MD; George Creasy, MD; Priya Soma-Pillay, MD;Shalini Fusey, MD; Cetin Cam, MD; Zarko Alfirevic, MD; Sonia S. Hassan, MD
OBJECTIVE: No randomized controlled trial has compared vaginal pro-gesterone and cervical cerclage directly for the prevention of pretermbirth in women with a sonographic short cervix in the mid trimester, sin-gleton gestation, and previous spontaneous preterm birth. We per-formed an indirect comparison of vaginal progesterone vs cerclage us-ing placebo/no cerclage as the common comparator.
STUDY DESIGN: Adjusted indirect metaanalysis of randomized con-rolled trials.
RESULTS: Four studies that evaluated vaginal progesterone vs placebo(158 patients) and 5 studies that evaluated cerclage vs no cerclage(504 patients) were included. Both interventions were associated with astatistically significant reduction in the risk of preterm birth at �32
weeks of gestation and composite perinatal morbidity and mortalityshort cervix, previous preterm birth, and singleton gestation: a systematic review and indirec
See related editorial, page 1
42.e1 American Journal of Obstetrics & Gynecology JANUARY 2013
compared with placebo/no cerclage. Adjusted indirect metaanalysesdid not show statistically significant differences between vaginal pro-gesterone and cerclage in the reduction of preterm birth or adverseperinatal outcomes.
CONCLUSION: Based on state-of-the-art methods for indirect compar-isons, either vaginal progesterone or cerclage are equally efficacious inthe prevention of preterm birth in women with a sonographic short cer-vix in the mid trimester, singleton gestation, and previous preterm birth.Selection of the optimal treatment needs to consider adverse events,cost and patient/clinician preferences.
Key words: birthweight, cervix, neonatal intensive care unit, perinatalmortality, perinatal morbidity, premature, prematurity, progestin, 17�-
hydroxyprogesterone caproate, 17PCite this article as: Conde-Agudelo A, Romero R, Nicolaides K, et al. Vaginal progesterone vs cervical cerclage for the prevention of preterm birth in women with a sonographic
t comparison metaanalysis. Am J Obstet Gynecol 2013;208:42.e1-18.From the Perinatology Research Branch, NICHD/NIH/DHHS, Bethesda, MD, and Detroit, MI (Drs Conde-Agudelo, Romero, Chaiworapongsa, andHassan); the Department of Obstetrics and Gynecology, King’s College Hospital, London, UK (Dr Nicolaides), Wayne State University/Hutzel Hospital,Detroit, MI (Drs Chaiworapongsa and Hassan); the Perinatal Diagnostic Center, Central Baptist Hospital, and the Department of Obstetrics and Gynecology,University of Kentucky, Lexington, KY (Dr O’Brien); the Department of Obstetrics and Gynecology, Zeynep Kamil Women and Children Diseases Educationnd Research Hospital, Uskudar, Istanbul, Turkey (Drs Cetingoz and Cam); the Departamento de Obstetrícia e Ginecologia, Hospital do Servidor Publicostadual “Francisco Morato de Oliveira” and School of Medicine, University of São Paulo, São Paulo, Brazil (Dr da Fonseca); Columbia Laboratories, Inc,ivingston, NJ (Dr Creasy); the Department of Obstetrics and Gynaecology, Steve Biko Academic Hospital, and the University of Pretoria, Pretoria,outh Africa (Dr Soma-Pillay); the Department of Obstetrics and Gynecology, Government Medical College and Hospital, Maharashtra, India (Dr Fusey);nd the University of Liverpool, Liverpool, UK (Dr Alfirevic).
eceived July 10, 2012; revised Oct. 12, 2012; accepted Oct. 17, 2012.
upported in part by the Intramural Research Program of the Eunice Kennedy Shriver National Institute of Child Health and Human Development,National Institutes of Health, Department of Health and Human Services, Bethesda, MD.
Most of the authors report no conflict of interest, except as stated in this paragraph. J.M.O. was involved in studies of progesterone geltreatment for preterm birth prevention sponsored by Columbia Laboratories, Inc (which is the manufacturer of the preparation used in thePREGNANT trial) and a previous trial of vaginal progesterone in women at risk for preterm delivery, serves on advisory boards, and is aconsultant for Watson Pharmaceuticals (which is a company with a financial interest in marketing vaginal progesterone gel for the prevention ofpreterm birth). He and others are listed in the patent on the use of all progesterone compounds to prevent preterm birth (US Patent Number7,884,093: Progesterone for the Treatment and Prevention of Spontaneous Preterm Birth). G.C. is an employee of Columbia Laboratories, Inc.
Reprints not available from the authors.
0002-9378/free • © 2013 Published by Mosby, Inc. • http://dx.doi.org/10.1016/j.ajog.2012.10.877
For Editors’ Commentary, see Contents
ti
uitp
bttscs
s
f
vpchg
ic
hcwsc
brtcpapw
i
tmaca
i
I
www.AJOG.org Reports of Major Impact
Most of the efforts to prevent pre-term birth have been focused on
the treatment of symptoms or signs ofactivation of the common pathway ofparturition1-5 (ie, increased uterine con-tractility,6-28 preterm cervical ripen-ing,29-41 and/or membrane decidual ac-ivation42-56). Although the detection ofncreased uterine contractility57-66 has
been the focus of clinicians and repro-ductive biologists for decades, emergingclinical67-76 and laboratory-based evi-dence77-96 suggests that focusing on the
terine cervix may yield approaches todentify the patient who is at risk for pre-erm delivery as well as interventions torevent it.97-115
A sonographic short cervix has emergedas a powerful predictor of pretermbirth.67-74,116-122 It is unlikely that thiscondition is due to a single cause; a mul-tiple causation model of a sonographicshort cervix has been proposed (eg, ashort cervix is syndromic in na-ture).75,123,124 Such model would have
iologic, diagnostic, prognostic, andherapeutic implications.3,75 Indeed, pa-ients may have a short cervix after diethyl-tilbestrol exposure in utero,125-127 aervical conization,128-145 a loop electro-urgical excision procedure,146-150 intra-
uterine infection/inflammation,151-161
a decline in progesterone action,162-165
and the challenging condition clini-cally referred to as idiopathic cervicalinsufficiency.166-173
Three interventions have been pro-posed to treat patients with a sono-graphic short cervix: (1) vaginal proges-terone administration,101-105 (2) cervicalcerclage for patients with a history ofpreterm birth,174-177 and (3) vaginal pes-ary.97-99 Recently, a combination of
vaginal progesterone and a pessary hasbeen reported to be a successful methodto reduce the rate of preterm delivery intwin gestations with a cervix of �25mm.178
Two independent randomized clinicaltrials101,104 and an individual patientdata (IPD) metaanalysis showed thatvaginal progesterone decreases the rateof preterm delivery and neonatal mor-bidity/mortality in women with a sono-graphic short cervix.105 This is the case
or patients with or without a history of ppreterm birth.105 The placement of a cer-ical cerclage appears to be indicated inatients with acute cervical insufficien-y,179-186 and perhaps, in some with aistory of preterm birth and a sono-raphic short cervix of �25 mm.174-177
Thus, there appear to be 2 interventionsthat may reduce the rate of preterm de-livery in patients with a history of pre-term birth and a cervix of �25 mm: vag-nal progesterone administration or aervical cerclage.
Recently, 2 professional organizationsave recommended that cerclage may beonsidered for the treatment of womenith a singleton gestation, previous
pontaneous preterm birth, and a cervi-al length �25 mm at �24 weeks of ges-
tation.187,188 This recommendation wasased mainly on an IPD metaanalysis ofandomized controlled trials that showhat cerclage is associated with a statisti-ally significant reduction in the risk ofreterm birth at �37, �35, �32, �28,nd �24 weeks of gestation, and com-osite perinatal morbidity and mortalityhen compared with no cerclage.176
However, another IPD metaanalysisdemonstrated that vaginal progesteroneadministration to women with a sono-graphic short cervix (�25 mm) in themid trimester significantly decreased therisk of preterm birth at �35, �34, �33,�30, and �28 weeks of gestation andcomposite neonatal morbidity and mor-tality when compared with placebo.179
In addition, a subgroup analysis showedthat vaginal progesterone was associatedwith a significant reduction in the risk ofpreterm birth at �33 weeks of gestationand composite neonatal morbidity andmortality in women with a short cervix(�25 mm), singleton gestation, and pre-vious spontaneous preterm birth.105
The availability of vaginal progester-one and cerclage for the prevention ofpreterm birth in women with a short cer-vix, singleton gestation, and previousspontaneous preterm birth could createa dilemma for physicians and patientsabout the optimal choice of treat-ment.189 Thus far, there are no random-zed controlled trials comparing vaginal
rogesterone and cerclage directly. InJANUARY 2013 Ameri
he absence of this evidence, indirectetaanalysis has emerged as an accepted
nd valid method for the comparison ofompeting interventions with the use ofcommon comparator.190-193
We performed an adjusted indirectmetaanalysis to compare the treatmenteffects of vaginal progesterone vs cer-clage in asymptomatic women with acervical length �25 mm in the mid tri-mester, singleton gestation and previousspontaneous preterm birth for the pre-vention of preterm birth. Previously, wehad conducted an IPD metaanalysis toevaluate the efficacy of vaginal progester-one vs placebo in patients with suchcharacteristics. Then, the summary esti-mates and measures of uncertainty wereused together with those reported in theIPD metaanalysis that evaluated cerclagevs no cerclage177 to perform the adjustedndirect comparison metaanalysis.
MATERIALS AND METHODSThe study was conducted based on a pro-spectively prepared protocol and is re-ported with the use of the Preferred Re-porting Items for Systematic reviews andMetaanalyses (PRISMA) guidelines formetaanalyses of randomized controlledtrials194 and suggested guidelines forPD195 and indirect metaanalyses.196
Literature searchWe searched MEDLINE, EMBASE, CI-NAHL, and LILACS (all from inceptionto October 31, 2012), the Cochrane CentralRegister of Controlled Trials (1960 to Octo-ber 31, 2012; http://www.mrw.interscience.wiley.com/cochrane/cochrane_clcentral_articles_fs.html), ISIWebofScience(1960toOctober 31, 2012; http://www.isiknowledge.com), research registers of ongoing trials(www.clinicaltrials.gov, www.controlled-trials.com, www.centerwatch.com, www.anzctr.org.au, http://www.nihr.ac.uk, andwww.umin.ac.jp/ctr), and Google scholarusing a combination of keywords and textwords related to progesterone (progesterone,progestins, progestogen, progestagen, pro-gestational agent), cervical cerclage (cerclage,cervical stitch, cervical suture, cervical liga-tion, Shirodkar suture, Shirodkar opera-tion, Shirodkar stitch, Shirodkar proce-dure, McDonald suture, McDonald
procedure, McDonald method, McDon-can Journal of Obstetrics & Gynecology 42.e2
stgcttwswpoaimpc(w(wtpetvdip(ltt1
sottwhribtosracIcwnwmppbc
dtrdntt�t
titartcooamborMfipttver
Reports of Major Impact www.AJOG.org
ald technique), short cervix (short cervicallength, short cervix, cervical shortening),and preterm birth (preterm, premature).Congress proceedings of international so-ciety meetings of maternal-fetal and repro-ductive medicine and international meet-ings on preterm birth, reference lists ofidentified studies, textbooks, previouslypublished systematic reviews, and reviewarticles were also searched. Experts in thefield were contacted to identify furtherstudies. No language restrictions wereapplied.
Study selectionWe included randomized controlled tri-als in which asymptomatic women witha sonographic short cervix (cervicallength, �25 mm) in the mid trimester,ingleton gestation, and previous spon-aneous preterm birth at �37 weeks ofestation were allocated randomly to re-eive vaginal progesterone vs placebo/noreatment or cerclage vs no cerclage forhe prevention of preterm birth. Trialsere included if the primary aim of the
tudy was to (1) prevent preterm birth inomen with such characteristics; or (2)revent preterm birth in women withther characteristics, but outcomes werevailable for patients with a prerandom-zation cervical length �25 mm in the
id trimester, singleton gestation, andrevious preterm birth. Trials were ex-luded if they (1) were quasirandomized,2) evaluated the interventions inomen with only multiple gestations,
3) evaluated vaginal progesterone inomen with actual or threatened pre-
erm labor, second trimester bleeding, orremature rupture of membranes, (4)valuated the administration of proges-erone in the first trimester only to pre-ent miscarriage, (5) assessed history-in-icated cerclage (placed for the sole
ndication of poor obstetric history),hysical examination–indicated cerclageplaced for second trimester cervical di-ation), or compared different cerclageechniques or outpatient cerclage vs inpa-ient cerclage, (6) compared cerclage with7�-hydroxyprogesterone caproate, or (7)
did not provide data for women with a cer-vical length �25 mm in the mid trimester,singleton gestation, and previous preterm
birth.42.e3 American Journal of Obstetrics & Gynecolog
All published studies that were deemedsuitable were retrieved and reviewed inde-pendently by 2 authors (A.C-A. and R.R.)to determine inclusion. Disagreementswere resolved through consensus.
Data collectionFor the IPD metaanalysis that evaluatedvaginal progesterone vs placebo, we con-tacted the corresponding authors to re-quest access to the data. Authors wereasked to supply anonymized data (with-out identifiers) about patient baselinecharacteristics, experimental interven-tion, control intervention, cointerven-tions, and prespecified outcome mea-sures for every randomly assignedubject and were invited to become partf the collaborative group with joint au-horship of the final publication. Datahat were provided by the investigatorsere merged into a master database thatad been constructed specifically for theeview. Data were checked for missingnformation, errors, and inconsistenciesy cross-referencing the publications ofhe original trials. Quality and integrityf the randomization processes were as-essed by a review of the chronologicandomization sequence and pattern ofssignment and the balance of baselineharacteristics across treatment groups.nconsistencies or missing data were dis-ussed with the authors and correctionsere made when deemed necessary. Fi-ally, data were extracted for womenith a cervical length �25 mm in theid trimester, singleton gestation, and
revious preterm births. A similar ap-roach was used in the IPD metaanalysisy Berghella et al177 that evaluated cer-lage vs no cerclage.
Outcome measuresThe prespecified primary outcome mea-sures were preterm birth �32 weeks ofgestation and composite perinatal mor-bidity and mortality (defined as the oc-currence of any of the following events:respiratory distress syndrome, gradeIII/IV intraventricular hemorrhage, ne-crotizing enterocolitis, neonatal sepsis,bronchopulmonary dysplasia, or perina-tal mortality). Secondary outcome mea-sures included preterm birth at �37, �35,
and �28 weeks of gestation, respiratoryy JANUARY 2013
istress syndrome, necrotizing enterocoli-is, grade III/IV intraventricular hemor-hage, neonatal sepsis, bronchopulmonaryysplasia, perinatal mortality, a compositeeonatal morbidity outcome (defined as
he occurrence of any of the above men-ioned neonatal morbidities), birthweight
1500 g and �2500 g, and admission tohe neonatal intensive care unit (NICU).
Assessment of risk of biasThe risk of bias in each included studywas assessed by the use of the criteria re-cently outlined in the Cochrane Hand-book for Systematic Reviews of Interven-tions.197 Seven domains that are relatedo the risk of bias were assessed in eachncluded trial because there is evidencehat these issues are associated with bi-sed estimates of treatment effect: (1)andom sequence generation, (2) alloca-ion concealment, (3) blinding of parti-ipants and personnel, (4) blinding ofutcome assessment, (5) incompleteutcome data, (6) selective reporting,nd (7) other bias. Review authors’ judg-ents were categorized as “low risk” of
ias, “high risk” of bias, or “unclear risk”f bias. The assessments considered theisk of material bias rather than any bias.
aterial bias was defined as a bias of suf-cient magnitude to have a notable im-act on the results or conclusions of therial. The risk of bias in each includedrial was assessed individually by 2 re-iewers (A.C-A. and R.R.). Any differ-nces of opinion regarding assessment ofisk of bias were resolved by discussion.
Data extractionTwo authors (A.C-A. and R.R.) ex-tracted data from each study on partici-pants (inclusion and exclusion criteria,number of women and fetuses/infants inrandomized groups, baseline character-istics, and country and date of recruit-ment), study characteristics (randomiza-tion procedure, concealment allocationmethod, blinding of clinicians, womenand outcome assessors, completeness ofoutcome data for each outcome, which in-cluded attrition and exclusions from theanalysis, and intention-to-treat analysis),details of interventions (aim, gestationalage at trial entry, daily dose of vaginal pro-
gesterone and duration of treatment, cer-
gttb
r
ut
Ha
s
mcmsng
www.AJOG.org Reports of Major Impact
clage type and suture used, and cointer-ventions), and outcomes (number ofoutcome events/total number in womenwith a cervical length �25 mm, singletonestation, and previous spontaneous pre-erm birth). Women with multiple gesta-ions, no previous spontaneous pretermirth, or cervical length �25 mm were ex-
cluded. For studies that assessed cerclage,data on proportions and relative risks
FIGUREStudy selection process
Records identified through database searching (n = 11,587)
Records
Full-text articles a ●Vaginal proge ●Cerclage (n =
Iden
tific
atio
n Sc
reen
ing
Elig
ibili
ty
Incl
uded Studies included in qualitativ
●Vaginal progesteron ●Cerclage vs no cerc
Conde-Agudelo. Vaginal progesterone vs cervical cerclage. Am
(RRs) with 95% confidence intervals (CIs)
for each outcome measure were extractedfrom the IPD metaanalysis by Berghella etal.177 Disagreements in extracted data wereesolved by discussion among reviewers.
Statistical analysisStatistical analyses were based on an in-tent-to-treat basis and included all ran-domly assigned women and their fetus-es/infants. For studies that assessed
1
Additional records identified through other sources (n = 4)
Duplicate records excluded (n = 5985)
eened (n = 5606)
Records excluded on basis of title or abstract (n = 5574)
ssed for eligibility (n = 32) rone (n = 11)
ynthesis and metaanalyses (n = 9) placebo (n = 4)101,104,227,228 (n = 5)113,115,174,229,230
Full-text articles excluded (n = 23) Vaginal progesterone (n = 7) ●Cervical length was not measured or collected before randomization (n = 2)204,205 ●Vaginal progesterone evaluated only in women with twin gestations (n = 2)206,207 ●Vaginal progesterone evaluated in women with actual or threatened preterm labor (n = 3)208-210 Cerclage (n = 16) ●History-indicated cerclage vs no cerclage (n = 5)211-215 ●History- vs ultrasound-indicated cerclage (n = 3)216-218 ●Comparison of different cerclage techniques (n = 3)219-221 ●Cerclage evaluated only in women with multiple gestations (n = 2)222,223 ●Quasirandomized and compared cerclage vs pessary (n = 1)224 ●Outpatient vs inpatient cerclage (n = 1)225 ●Cerclage vs 17α-hydroxyprogesterone
caproate (n = 1)226
bstet Gynecol 2013.
vaginal progesterone, IPD were com-
JANUARY 2013 Ameri
bined in a 2-stage approach in whichoutcomes were analyzed in their originaltrial, then summary statistics were com-bined with the use of standard summarydata metaanalysis techniques to give anoverall measure of effect (summary RRwith 95% CI).198 A similar approach was
sed in the IPD metaanalysis of trialshat evaluated cerclage vs no cerclage.177
Heterogeneity of the results among stud-ies was tested with the quantity I2 inthe IPD metaanalysis of vaginal proges-terone vs placebo199 and the Mantel-
aenszel Q statistics in the IPD meta-nalysis of cerclage vs no cerclage. I2
values of �50% or a probability valueof � .10 for Mantel-Haenszel Q statisticsindicated a substantial level of heteroge-neity. Fixed-effects models were used ifsubstantial statistical heterogeneity wasnot present. Otherwise, random-effectsmodels were used.
The number needed to treat for benefitor harm (with their 95% CIs) were cal-culated for the primary outcomes forwhich there was a statistically significantreduction or increase in risk differencebased on control event rates in the in-cluded trials.200 Publication and relatedbiases were assessed visually by an exam-ination of the symmetry of funnel plotsand statistically by the use of the Eggertest.201 A probability value of � .1 was con-idered to indicate significant asymmetry.
The adjusted indirect comparisonetaanalysis of vaginal progesterone vs
erclage was performed according to theost widely applied indirect compari-
on method by Bucher et al.202 The Ca-adian Agency for Drugs and Technolo-ies in Health192 and others190,193,203
have identified this method as the mostsuitable approach for performing indi-rect treatment comparisons of random-ized controlled trials. In this method, therandomization of each trial is main-tained, and the direct comparisons A vsB and C vs B with the common compar-ator link B are used to yield an indirectcomparison of A vs C. Because vaginalprogesterone and cerclage have beencompared with placebo and no cerclage,respectively, indirect comparison wasenabled by the “common” placebo/nocerclage arms. An extension of the Bu-
scr
sseste 21)
e se vslage
J O
cher approach was used to convert the
can Journal of Obstetrics & Gynecology 42.e4
wsunbR
lgactattasoofiigwgpafcr
sacfIwDtr
pon
tnfesHSp
mtt
s(
ctgtwp
ttvmsmw
tf
au
gai
pei
wai
ap
Reports of Major Impact www.AJOG.org
summary estimates (lnRRs) and mea-sures of uncertainty (variances) from the2 metaanalyses into a RR (95% CI) thatrepresented the difference between vagi-nal progesterone (p) and cerclage (c) asin the following equations:192
ln(RRpc Indirect) � �ln(RRpc)
95% CI of ln�RRpc Indirect� � �ln�RRpc�
� Z�⁄2��Var(ln�RRpc�)
here Var indicates the square of thetandard error (variance) and Z�/2 is thepper 95% percentile of the standardormal distribution. All values wereack transformed to give the estimate ofRpc with a 95% CI.To examine the assumption of simi-
arity of treatment effects, we investi-ated the effect of patient and trial char-cteristics on both direct and indirectomparison results with the use of sensi-ivity analyses. A predefined sensitivitynalysis was conducted by excluding pa-ients who received progesterone in trialshat evaluated cerclage vs no cerclagend patients who received a cerclage intudies that compared vaginal progester-ne with placebo to explore the impactf these cointerventions on the effect sizeor preterm birth and perinatal mortal-ty. This analysis was performed becauset is unclear whether the effects of pro-esterone and cerclage are additive inomen with a short cervix, singletonestation, and previous spontaneousreterm birth. An additional sensitivitynalysis was planned to evaluate the ef-ect of study quality on the main out-omes by the exclusion of trials with highisk of bias.
One author (A.C-A.) conducted alltatistical analyses using Review Man-ger software (version 5.1.6; Nordic Co-hrane Centre, Copenhagen, Denmark)or performing direct metaanalyses andndirect Treatment Comparison soft-are (version 1.0; Canadian Agency forrugs and Technologies in Health, Ot-
awa, Canada) to perform adjusted indi-ect comparison metaanalyses.
Informed consent was provided by theatients on enrollment in the each of theriginal trials. In this study, the data were
ot used for any other purpose other42.e5 American Journal of Obstetrics & Gynecolog
han those of the original trial, and noew data were collected. Therefore, in-
ormed consent specifically for this proj-ct was not considered necessary. Thistudy was exempted for review by theumanInvestigationsCommitteeofWayne
tate University. No patient identifiers wererovided by any investigator.
RESULTSOf the 5606 relevant citations that wereidentified, the abstracts were reviewed,and 32 studies were retrieved becausethey were considered potentially rele-vant to this indirect metaanalysis. Twen-ty-three studies were excluded204-226
(Figure). The remaining 9 trials met theinclusion criteria and provided data for662 women with a cervical length of �25
m at mid trimester, singleton gesta-ion, and previous spontaneous pre-erm birth at �37 weeks of gesta-
tion.101,104,113,115,174,227-230 Four studiesevaluated vaginal progesterone vs pla-cebo (158 women),101,104,227,228 and 5tudies evaluated cerclage vs no cerclage504 women).113,115,174,229,230
The main characteristics of studiesthat were included in this indirect com-parison metaanalysis are presented inTable 1. All 4 studies that evaluated vag-inal progesterone were double-blind,placebo-controlled trials.101,104,227,228
None of the studies that assessed cer-clage were double-blind. Seven trials(2 that evaluated vaginal progester-one101,104 and all 5 that evaluated cer-lage113,115,174,229,230) examined the in-erventions in women with a sono-raphic short cervix, 1 study evaluatedhe use of vaginal progesterone inomen with a history of spontaneousreterm birth,227 and the remaining
study evaluated the use of vaginal pro-gesterone in women with a previousspontaneous preterm birth, uterine mal-formations, or twin gestation.228 Only 1rial was designed specifically to evaluatehe use of cerclage in women with a cer-ical length of �25 mm in the mid tri-ester, singleton gestation, and previous
pontaneous preterm birth.174 The pri-ary outcome was preterm birth at �37eeksofgestationfor1trial,228�35weeksof
gestation for 2 trials,113,174 �34 weeks of ges-
y JANUARY 2013
tation for 2 trials,101,115 �33 weeks of gesta-ion for 2 trials,104,230 �32 weeks of gestationor 1 trial,227 and gestational age at delivery
for the remaining study.229
Gestational age at cervical lengthscreening varied between 14 and 25weeks of gestation, although most stud-ies performed screening at �25 weeks ofgestation.104,113,115,174,227-230 Of the 4 tri-ls that evaluated vaginal progesterone, 2sed gel (90 mg/d),104,227 1 used capsules
(200 mg/d),101 and the other used sup-positories (100 mg/d).228 The treatmentwas initiated at 24 weeks of gestation in 2trials,101,228 between 20 and 23 weeks ofestation in 1 trial,104 and between 18nd 22 weeks of gestation in the remain-ng study.227 Two studies reported that
participating women received studymedication from enrollment until 34weeks of gestation, 101,228 and 2 studies re-
orted that medication was given fromnrollment until 36 6/7 weeks of gestat-on. 104,227 In the study by Hassan et al,104
5 women received an emergencycerclage. Among the 5 trials that evaluatedcerclage, 4 used the McDonald proce-dure,113,115,174,229 and 1 used the Shirodkartechnique.230Rescuecerclageinwomenwho
ere allocated to the no cerclage group wasllowed in 3 studies based on physical exam-nation174 or based on ultrasonographic cer-vical changes.115,229 Prophylactic antibioticsnd tocolytics were administered to mostarticipating women in 3 studies,115,229,230
whereas bed rest was recommended toall women who were recruited in 2 tri-als.113,115 In the trial by Owen et al,174
99 women received 17�-hydroxypro-gesterone caproate, and 1 woman re-ceived vaginal progesterone.231
All 9 studies that were included in themetaanalysis had adequate random se-quence generation and allocation con-cealment, were free of selective outcomereporting, and had no obvious risk ofother biases. In the 4 trials that evaluatedvaginal progesterone, there was blindingof participants, health care providers,and outcome assessors. In the 5 trialsthat evaluated cerclage, study partici-pants and health care providers were notblinded, and it was unclear whether out-come assessors were masked to interven-tion allocations after inclusion of pa-
tients into the study. However, we
TABLE 1Characteristics of studies included in this systematic review
StudyParticipatingcountries
Primary targetpopulation Inclusion/exclusion criteria
Women with cervicallength <25 mm,singleton gestation,and previous pretermbirth, n Gestational
age atscreening,wk Intervention Cointerventions Primary outcome
Interventiongroup
Controlgroup
Vaginal progesteronecompared withplacebo............................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................
Fonseca et al,2007101
United Kingdom,Chile, Brazil,Greece
Women with ashort cervix
Inclusion: women with a singleton or twin pregnancy anda sonographic cervical length �15 mm
15 23 20-25 Vaginal progesteronecapsule (200 mg/d)or placebo from 24-33 6/7 weeks ofgestation
No Spontaneouspreterm birth �34wk
Exclusion: major fetal abnormalities, painful regularuterine contractions, a history of ruptured membranes, orcervical cerclage
............................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................
O’Brien et al,2007227
United States,South Africa,India, CzechRepublic, Chile,El Salvador
Women with ahistory ofspontaneouspreterm birth
Inclusion: women with a singleton pregnancy, 18-45years old, and a history of spontaneous singletonpreterm birth at 20-35 wk of gestation in theimmediately preceding pregnancy
9 13 16-22 Vaginal progesteronegel (90 mg/d) orplacebo from 18-22to 37 0/7 weeks ofgestation, rupture ofmembranes orpreterm delivery,whichever occurredfirst
No Preterm birth �32wk
Exclusion: planned cervical cerclage, history of adversereaction to progesterone, treatment with progesteronewithin 4 wk before enrollment, treatment for a seizuredisorder, a psychiatric illness or chronic hypertension atthe time of enrolment, history of acute or chroniccongestive heart failure, renal failure, uncontrolleddiabetes mellitus, active liver disorder, HIV infection witha CD4 count of �350 cells/mm3 that require multipleantiviral agents, placenta previa, history or suspicion ofbreast or genital tract malignancy, history or suspicion ofthromboembolic disease, Müllerian duct anomaly, majorfetal anomaly or chromosomal disorder, or multifetalgestation
............................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................
Cetingoz et al,2011228
Turkey Women at highrisk of pretermbirth
Inclusion: women with a least 1 previous spontaneouspreterm birth, uterine malformation or twin pregnancy
3 3 20-24 Vaginal progesteronesuppository (100mg/d) or placebofrom 24-34 wk ofgestation
No Preterm birth �37wk
Exclusion: in-place or planned cervical cerclage orserious fetal anomalies
............................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................
Conde-Agudelo. Vaginal progesterone vs cervical cerclage. Am J Obstet Gynecol 2013. (continued )
ww
w.A
JOG
.orgR
epor
tso
fM
ajo
rIm
pac
t
JANUARY2013
American
JournalofObstetrics&
Gynecology42.e6
TABLE 1Characteristics of studies included in this systematic review (continued)
StudyParticipatingcountries
Primary targetpopulation Inclusion/exclusion criteria
Women with cervicallength <25 mm,singleton gestation,and previous pretermbirth, n Gestational
age atscreening,wk Intervention Cointerventions Primary outcome
Interventiongroup
Controlgroup
Hassan et al,2011104
United States,Republic ofBelarus, Chile,Czech Republic,India, Israel,Italy, Russia,South Africa,Ukraine
Women with ashort cervix
Inclusion: women with a singleton pregnancy,transvaginal sonographic cervical length of 10-20 mm,and no signs or symptoms of preterm labor
48 44 19-23 Vaginal progesteronegel (90 mg/d) orplacebo from 20-23to 36 6/7 weeks ofgestation, rupture ofmembranes orpreterm delivery,whichever occurredfirst
Emergency cervical cerclage(4 in vaginal progesteronegroup [8.3%] and 1 [2.3%] inplacebo group)
Preterm birth �33wk
Exclusion: planned cerclage, acute cervical dilation,allergic reaction to progesterone, current or recentprogestogen treatment within the previous 4 wk, chronicmedical conditions that would interfere with studyparticipation or evaluation of the treatment, major fetalanomaly or known chromosomal abnormality, uterineanatomic malformation, vaginal bleeding, or known orsuspected clinical chorioamnionitis
............................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................
Rust et al, 2001229 United States Women with ashort cervix
Inclusion: women with a singleton or multiple gestationand transvaginal sonographic dilation of the internal oswith either membrane prolapse into the endocervicalcanal at least 25% of the total cervical length but notbeyond the external os or a cervical length �25 mm
53 49 16-24 McDonald procedurewith a single stitchof permanentmonofilament or nocerclage
Clindamycin and indomethacinthat were discontinued atapproximately 24 hr afterrandom assignment (bothgroups); rescue cerclage (bothgroups)
Gestational age atdelivery andneonatal morbidity
Exclusion: membrane prolapse beyond the external os,any fetal lethal congenital or chromosomal anomaly,abruption placenta, unexplained vaginal bleeding,chorioamnionitis, persistent uterine activity with cervicalchange, or any other contraindication to a cerclageprocedure
............................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................
Althuisius et al,2001115
The Netherlands Women with ashort cervix
Inclusion: women with a singleton gestation, risk factorsand/or symptoms of cervical incompetence, and acervical length �25 mm
14 12 14-23 McDonald procedurewith braidedpolyester thread orno cerclage
Amoxicillin/clavulanic acid for7 days and bed rest (bothgroups); two 100-mgsuppositories of indomethacin(cerclage group); rescuecerclage (no cerclage group)
Preterm birth �34wk and neonatalmorbidity andmortality
Exclusion: fetal congenital/chromosomal anomalies,preterm rupture of membranes, membranes bulging intothe vagina, or intrauterine infection
............................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................
To et al, 2004230 United Kingdom,Brazil, SouthAfrica, Slovenia,Greece, Chile
Women with ashort cervix
Inclusion: women with a singleton gestation and cervicallength �15 mm
21 23 22-24 Shirodkar suturewith Mersilene tapeor no cerclage
Prophylactic corticosteroids forfetal lung maturation (bothgroups); single dose oferythromycin (cerclage group)
Preterm birth �33wk
............................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................
Conde-Agudelo. Vaginal progesterone vs cervical cerclage. Am J Obstet Gynecol 2013. (continued )
Repo
rts
of
Ma
jor
Impa
ct
ww
w.A
JOG
.org
42.e7Am
ericanJournalofObstetrics
&Gynecology
JANUARY2013
TABLE 1Characteristics of studies included in this systematic review (continued)
StudyParticipatingcountries
Primary targetpopulation Inclusion/exclusion criteria
Women with cervicallength <25 mm,singleton gestation,and previous pretermbirth, n Gestational
age atscreening,wk Intervention Cointerventions Primary outcome
Interventiongroup
Controlgroup
Exclusion: major fetal abnormalities, painful regularuterine contractions, history of ruptured membranes andcervical cerclage in situ, and dilated cervix duringtransvaginal sonography
............................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................
Berghella et al,2004113
United States Women with ashort cervix
Inclusion: women with a singleton or twin gestation andcervical length �25 mm or significant funneling (�25%)
14 17 14-23 McDonald procedurewith Mersilene tapeor no cerclage
Bed rest (both groups) Preterm birth �35wk
Exclusion: history indicated prophylactic cerclage, lastpregnancy delivered at term, major fetal anomaly, tripletsor higher order pregnancy, previous inclusion in anothertrial, current drug abuse, or regular contractions leadingto preterm labor after identification of abnormal cervix byultrasonography
............................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................
Owen et al, 2009116 United States Women with ashort cervix,singletongestation, andpreviousspontaneouspreterm birth
Inclusion: women with a singleton gestation, at least 1previous spontaneous preterm birth between 17-33weeks of gestation, and mid trimester cervical length�25 mm
148 153 16-22 McDonald procedurewith nonabsorbablesuture (braided tape)or no cerclage
17�-hydroxyprogesteronecaproate (47 [31.8%] incerclage group and 52[34.0%] in no cerclage group);vaginal progesterone (1 [0.7%]in no cerclage group); rescuecerclage (no cerclage group)
Preterm birth �35wk
Exclusion: fetal anomaly, planned history indicatedcerclage for a clinical diagnosis of cervical insufficiency,and clinically significant maternal/fetal complications (eg,fetal red cell isoimmunization, treated chronichypertension, insulin-dependent diabetes mellitus) andcervical insufficiency that indicated cerclage in aprevious pregnancy.
............................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................
Conde-Agudelo. Vaginal progesterone vs cervical cerclage. Am J Obstet Gynecol 2013.
ww
w.A
JOG
.orgR
epor
tso
fM
ajo
rIm
pac
t
JANUARY2013
American
JournalofObstetrics&
Gynecology42.e8
ABLE
2ire
ctan
din
dire
ctco
mpa
rison
s
utco
me
Dire
ctco
mpa
rison
sIn
dire
ctco
mpa
rison
:va
gina
lpro
gest
eron
evs
cerc
lage
Vagi
nalp
roge
ster
one
vspl
aceb
oCe
rcla
gevs
noce
rcla
ge
Tria
ls,
nIn
terv
entio
n,n/
N(%
)Co
ntro
l,n/
N(%
)Re
lativ
eris
k(9
5%CI
)He
tero
gene
ityI2 ,%
Tria
ls,
nIn
terv
entio
n,n/
N(%
)Co
ntro
l,n/
N(%
)Re
lativ
eris
k(9
5%CI
)He
tero
gene
ity,
Pva
luea
Rela
tive
risk
(95%
CI)
Pva
lueb
rimar
you
tcom
es...
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
Pret
erm
birth
�32
wk
49/
75(1
2.0)
24/8
3(2
8.9)
0.47
(0.2
4–0.
91)
05
48/2
50(1
9.2)
75/2
54(2
9.5)
0.66
(0.4
8–0.
91)
�.1
00.
71(0
.34–
1.49
).8
8...
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
Com
posi
tepe
rinat
alm
orbi
dity
/mor
talit
yc4
7/75
(9.3
)20
/83
(24.
1)0.
43(0
.20–
0.94
)0
539
/250
(15.
6)63
/254
(24.
8)0.
64(0
.45–
0.91
)�
.10
0.67
(0.2
9–1.
57)
.86
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
.....
econ
dary
outc
omes
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
...
Pret
erm
birth
�37
wk
434
/75
(45.
3)46
/83
(55.
4)0.
84(0
.61–
1.14
)0
510
5/25
0(4
2.0)
154/
254
(60.
6)0.
70(0
.58–
0.83
)�
.10
1.20
(0.8
4–1.
72)
.94
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
...
Pret
erm
birth
�35
wk
420
/75
(26.
7)35
/83
(42.
2)0.
66(0
.42–
1.04
)0
571
/250
(28.
4)10
5/25
4(4
1.3)
0.70
(0.5
5–0.
89)
�.1
00.
94(0
.56–
1.58
).9
8...
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
Pret
erm
birth
�28
wk
46/
75(8
.0)
14/8
3(1
6.9)
0.51
(0.2
2–1.
18)
05
32/2
50(1
2.8)
51/2
54(2
0.1)
0.64
(0.4
3–0.
96)
�.1
00.
80(0
.31–
2.02
).9
2...
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
Resp
irato
rydi
stre
sssy
ndro
me
43/
75(4
.0)
12/8
3(1
4.5)
0.38
(0.1
3–1.
07)
74
13/2
07(6
.3)
21/1
96(1
0.7)
0.61
(0.3
2–1.
19)
�.1
00.
62(0
.18–
2.16
).8
4...
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
Grad
eIII
/IVin
trave
ntric
ular
hem
orrh
age
41/
75(1
.3)
3/83
(3.6
)0.
50(0
.08–
2.96
)8
40/
207
4/19
6(2
.0)
0.28
(0.0
5–1.
64)
�.1
01.
79(0
.15–
22.0
0).8
0...
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
Necr
otiz
ing
ente
roco
litis
40/
751/
83(1
.2)
0.47
(0.0
2–10
.32)
04
1/20
7(0
.5)
2/19
6(1
.0)
0.62
(0.0
8–4.
67)
�.1
00.
76(0
.02–
31.4
9).9
0...
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
Neon
atal
seps
is4
0/75
6/83
(7.2
)0.
25(0
.05–
1.37
)0
48/
207
(3.9
)17
/196
(8.7
)0.
47(0
.21–
1.05
)�
.10
0.53
(0.0
8–3.
35)
.78
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
...
Bron
chop
ulm
onar
ydy
spla
sia
20/
511/
47(2
.1)
0.31
(0.0
1–7.
32)
01
7/13
5(5
.2)
6/12
7(4
.7)
1.10
(0.3
8–3.
18)
�.1
00.
28(0
.01–
9.01
).5
8...
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
Com
posi
tene
onat
alm
orbi
dity
d4
3/75
(4.0
)16
/83
(19.
3)0.
29(0
.11–
0.81
)0
417
/207
(8.2
)28
/196
(14.
3)0.
60(0
.34–
1.06
)�
.10
0.48
(0.1
5–1.
53)
.75
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
...
Perin
atal
mor
talit
y4
5/75
(6.7
)7/
83(8
.4)
0.73
(0.2
5–2.
10)
05
22/2
50(8
.8)
35/2
54(1
3.8)
0.65
(0.4
0–1.
07)
�.1
01.
12(0
.35–
3.63
).9
6...
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
Adm
issi
onto
neon
atal
inte
nsiv
eca
re4
11/7
5(1
4.7)
33/8
3(3
9.8)
0.39
(0.2
1–0.
71)
04
57/2
07(2
7.5)
67/1
96(3
4.2)
0.63
(0.3
4–1.
18)
�.1
00.
62(0
.26–
1.48
).8
4...
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
Birth
wei
ght�
2500
g4
30/7
5(4
0.0)
42/8
3(5
0.6)
0.79
(0.5
5–1.
13)
05
86/2
50(3
4.4)
117/
249
(47.
0)0.
65(0
.42–
1.00
)�
.10
1.22
(0.6
9–2.
14)
.93
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
...
Birth
wei
ght�
1500
g4
8/75
(10.
7)18
/83
(21.
7)0.
53(0
.26–
1.11
)0
542
/250
(16.
8)66
/249
(26.
5)0.
64(0
.45–
0.90
)�
.10
0.83
(0.3
7–1.
85)
.93
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
.....
Fort
heM
ante
l-Hae
nsze
lQst
atis
tics(
test
ofhe
tero
gene
ity);
bFo
rthe
test
ofas
soci
atio
n;c
Occu
rrenc
eof
anyo
fthe
follo
win
gev
ents
:res
pira
tory
dist
ress
synd
rom
e,gr
ade
III/IV
intra
vent
ricul
arhe
mor
rhag
e,ne
crot
izing
ente
roco
litis
,neo
nata
lsep
sis,
bron
chop
ulm
onar
ydy
spla
sia,
orpe
rinat
alm
orta
lity;
dOc
curre
nce
ofan
yof
the
follo
win
gev
ents
:res
pira
tory
dist
ress
synd
rom
e,gr
ade
III/IV
intra
vent
ricul
arhe
mor
rhag
e,ne
crot
izing
ente
roco
litis
,neo
nata
lsep
sis,
orbr
onch
opul
mon
ary
dysp
lasi
a.
onde
-Agu
delo
.Vag
inal
prog
este
rone
vsce
rvic
alce
rcla
ge.A
mJO
bste
tGyn
ecol
2013
.
Reports of Major Impact www.AJOG.org
judged that assessment and measure-ment of most outcomes that were in-cluded in our review are considered ob-jective in nature and were not likely to beinfluenced by a lack of blinding in thestudies that evaluated cerclage. All butone study104 had adequate handling ofincomplete outcome data. Overall, all 9trials were considered to be at low risk ofbias.
Direct comparisonsThe use of either vaginal progesterone orcerclage in patients with a cervical lengthof �25 mm in the mid trimester, single-ton gestation, and previous spontaneouspreterm birth was associated with a sig-nificant reduction in the risk of pretermbirth at �32 weeks of gestation (RR,0.47; 95% CI, 0.24�0.91 for vaginal pro-gesterone and RR, 0.66; 95% CI,0.48�0.91 for cerclage) and compositeperinatal morbidity and mortality (RR,0.43; 95% CI, 0.20�0.94 for vaginal pro-gesterone and RR, 0.64; 95% CI,0.45�0.91 for cerclage) when comparedwith placebo and no cerclage, respec-tively (Table 2). The number of patientswho needed to be treated with vaginalprogesterone rather than with placebo toprevent either 1 case of preterm birth at�32 weeks of gestation or 1 case of com-posite perinatal morbidity/mortality was7 (95% CI, 5�38 for preterm birth at�32 weeks of gestation and 5�69 forcomposite perinatal morbidity/mortal-ity). The corresponding numbersneeded to treat for cerclage were 10 (95%CI, 7�38) and 11 (95% CI, 7�45),respectively.
Infants whose mothers received vagi-nal progesterone had a significantlylower risk of composite neonatal mor-bidity and admission to NICU than in-fants whose mothers had received pla-cebo. Patients who were allocated tocerclage had a statistically significant re-duction in the risk of preterm birth at�37, �35, and �28 weeks of gestationand a birthweight of �1500 g when com-pared with those who were allocated tono cerclage.
Indirect comparisonAdjusted indirect comparison meta-
analyses showed that, compared with42.e9 American Journal of Obstetrics & Gynecolog
T D O P . S . a C
y JANUARY 2013
pwtcbTsc
dfrwdimttc
. Am
www.AJOG.org Reports of Major Impact
cerclage, treatment with vaginal proges-terone was associated with a nonsignifi-cant 29% reduction in the risk of pre-term birth at �32 weeks of gestation(RR, 0.71; 95% CI, 0.34�1.49) and a33% nonsignificant decrease in the riskof composite perinatal morbidity andmortality (RR, 0.67; 95% CI, 0.29�1.57). Adjusted indirect comparison be-tween vaginal progesterone and cerclageindicated that there was no significantdifference for any of the secondary out-come measures. Estimated RRs rangedfrom 0.28 for bronchopulmonary dys-plasia to 1.79 for grade III/IV intraven-tricular hemorrhage, but all 95% CIs in-cluded 1. These results indicate thatvaginal progesterone and cerclage arenot significantly different in terms ofefficacy for the reduction of the risk ofpreterm birth and adverse perinataloutcomes.
Sensitivity analysisVaginal progesterone and cerclage sig-nificantly decreased the risk of pretermbirth at �32 and �35 weeks of gestationin a sensitivity analysis that excludedboth patients who received progestogensin trials that evaluated cerclage and thosein whom a cerclage was placed in trialsthat evaluated vaginal progesterone (Ta-ble 3). Cervical cerclage, compared withno intervention, was associated with areduction in the rate of preterm birth at�37 weeks of gestation and perinatal
TABLE 3Sensitivity analysis of direct and in
Outcome
Direct comparisons
Vaginal progesterone vs
Intervention,n/N (%)
Control,n/N (%)
Preterm birth..........................................................................................................
�32 wk 8/71 (11.3) 24/82 (29...........................................................................................................
�28 wk 5/71 (7.0) 14/82 (17...........................................................................................................
�35 wk 17/71 (23.9) 35/82 (42...........................................................................................................
�37 wk 31/71 (43.7) 46/82 (56....................................................................................................................
Perinatal mortality 4/71 (5.6) 7/82 (8.5...................................................................................................................a For the test of association.
Conde-Agudelo. Vaginal progesterone vs cervical cerclage
mortality; however, indirect compari- a
sons between vaginal progesterone andcerclage indicate that there were no sig-nificant differences between the 2 inter-ventions. Sensitivity analyses based ontrial quality were not performed becauseall trials were considered at low risk forbiases.
There was low statistical heterogeneityin all but 2 metaanalyses (admission toNICU and birthweight �2500 g in com-parison of cerclage vs no cerclage). Fun-nel plots showed no asymmetry, either visu-ally or in terms of statistical significance.
COMMENTPrincipal findings of the studyIn women with a sonographic short cer-vix in the mid trimester, singleton gesta-tion, and previous spontaneous pretermbirth, (1) vaginal progesterone adminis-tration was associated with a significant53% reduction in the risk of pretermbirth at �32 weeks of gestation, a 57%decrease in the risk of composite perina-tal morbidity and mortality, and a signif-icantly lower rate of composite neonatalmorbidity and admission to NICU whencompared with placebo; (2) the place-ment of a cervical cerclage showed a sig-nificant 34% reduction in the risk of pre-term birth at �32 weeks of gestation, a36% decrease in the risk of compositeperinatal morbidity and mortality, and asignificantly lower rate of preterm birthat �37, �35, and �28 weeks of gestation
ect comparison metaanalyses
cebo Cerclage vs no cerclage
Relative risk(95% CI)
Intervention,n/N (%)
Control, n/N(%)
.........................................................................................................................
0.44 (0.22–0.87) 40/203 (19.7) 64/201 (31.8.........................................................................................................................
0.46 (0.19–1.11) 28/203 (13.8) 43/201 (21.4.........................................................................................................................
0.59 (0.37–0.95) 57/203 (28.1) 85/201 (42.3.........................................................................................................................
0.80 (0.58–1.10) 82/203 (40.4) 123/201 (61.2.........................................................................................................................
0.61 (0.20–1.91) 19/203 (9.4) 33/201 (16.4.........................................................................................................................
J Obstet Gynecol 2013.
nd a birthweight of �2500 g when com- o
JANUARY 2013 Americ
ared with no cerclage; and (3) thereere no significant differences between
he efficacy of vaginal progesterone anderclage in the prevention of pretermirth or adverse perinatal outcomes.hese findings were consistent with sen-
itivity analyses in which patients who re-eived progestogens (eg, 17�-hydroxypro-
gesterone caproate or vaginal progesterone),and cerclage were excluded.
Strengths and limitations of the studyStrengths of the study include (1) use ofthe most rigorous methodology for per-forming an indirect comparison meta-analysis of randomized controlled trials.Specifically, we applied the best avail-able method to undertake the indirectcomparisons, assessed the assumption ofsimilarity of treatment effects using sen-sitivity analyses, evaluated statistical ho-mogeneity, and reported the results fol-lowing the recommended guidelines forthis type of study; (2) indirect compari-sons were performed by using data ob-tained from IPD metaanalyses105,177 of 2
irect comparisons; (3) the access to datarom individual patients enabled a moreigorous analysis from what is possibleith published data; (4) a broad andeep literature search was performed to
dentify relevant studies; (5) the highethodologic quality of the majority of
rials included in the review; (6) all pa-ients included in the review had a cervi-al length �25 mm at or before 25 weeks
Indirect comparison:vaginal progesterone vscerclage
elative risk5% CI)
Relative risk(95% CI) P valuea
..................................................................................................................
.63 (0.45–0.88) 0.70 (0.33–1.50) .88..................................................................................................................
.65 (0.43–1.00) 0.71 (0.27–1.88) .88..................................................................................................................
.67 (0.51–0.88) 0.88 (0.51–1.52) .96..................................................................................................................
.67 (0.55–0.82) 1.19 (0.82–1.74) .94..................................................................................................................
.58 (0.35–0.98) 1.05 (0.30–3.64) .98..................................................................................................................
dir
pla
R(9
......... .........
3) ) 0......... .........
1) ) 0......... .........
7) ) 0......... .........
1) ) 0......... .........
) ) 0......... .........
f gestation, singleton gestation, and
an Journal of Obstetrics & Gynecology 42.e10
crttln
v
dwehptomwaaidtahcw
rtvdvT
wcggt0bstc
Om
i1
hc
g
Reports of Major Impact www.AJOG.org
previous spontaneous preterm birth atless than 37 weeks of gestation; (7) theremarkable similar rates of preterm birthand adverse perinatal outcomes found incontrol groups of trials that evaluatedvaginal progesterone and cerclage (Table2) making more homogeneous the com-mon comparator placebo/no cerclage inindirect metaanalyses; (8) the evidenceof clinical and statistical homogeneityfor most of the outcomes evaluated; (9)the sensitivity analysis, by excluding pa-tients who received both progesteroneand cerclage, was consistent with (andthus supportive of) our overall findings;and (10) the symmetric funnel plots sug-gesting absence of publication and re-lated biases in our metaanalyses.
There are some potential limitationsof our study. In recent years, the adjustedindirect comparison method has beenused in health care decision-making tocompare competing treatments in theabsence of direct evidence about theirrelative effectiveness. The largest evalua-tion of the consistency between directand indirect comparisons of trials foundthat there was a statistically significantinconsistency in 16 of 112 comparisons(14%), which may be more commonwith subjectively assessed outcomes,comparisons that include a lower num-ber of trials in the analyses, and with sta-tistically significant results from eitherdirect or indirect comparisons.232 A re-ent simulation study reported that indi-ect comparisons may be underpoweredo determine treatment differences, par-icularly when there is a moderate-to-arge between-study degree of heteroge-eity.233 In addition, the risk of
overestimation could be high when theindirect comparison of interest relies onjust 1 trial for 1 of the 2 direct compari-sons. However, virtually all outcomemeasures that were included in our studywere assessed objectively, and most ofthe direct metaanalyses had a low degreeof statistical heterogeneity. Moreover,the comparisons of cerclage vs no cer-clage and vaginal progesterone vs pla-cebo relied mainly on 2 trials each (Owenet al174 and Rust et al229 for cerclage andHassan et al104 and Fonseca et al101 for
aginal progesterone).42.e11 American Journal of Obstetrics & Gynecolo
Another potential limitation of this in-irect metaanalysis was that 20% ofomen in the control group of trials that
valuated cervical cerclage received 17�-ydroxyprogesterone caproate com-ared with none in the control group ofrials that evaluated vaginal progester-ne. This difference could potentiallyean that the control groups, whichere used as the common comparator,
re not similar. Nevertheless, sensitivitynalyses that were performed by exclud-ng these patients showed no significantifferences in the results that were ob-ained with overall metaanalyses. Inddition, there is no evidence that 17�-ydroxyprogesterone caproate can de-rease the risk of preterm birth in womenith a short cervix.234
Given the apparent equivalence in ef-ficacy between vaginal progesterone andcerclage, differences in adverse effects arekey variables that clinicians and patientswith a singleton pregnancy and a previ-ous spontaneous preterm birth shouldconsider when selecting an optimaltreatment for a sonographic short cervixin the mid trimester. The IPD meta-analysis by Romero et al105 showed thatates of maternal adverse effects, discon-inuation of treatment because of ad-erse effects, and congenital anomaliesid not differ significantly between theaginal progesterone and placebo groups.he IPD metaanalysis by Berghella et al,177
which evaluated cerclage vs no cerclage,did not provide data on adverse events, butthe trial by Owen et al,174 which contrib-uted 60% of patients to that metaanalysis,reported that surgical and anesthetic com-plications that were associated with cer-clage placement were uncommon. None-theless, a recently updated Cochranereview that assessed the use of cerclage inwomen with a singleton gestation whowere at high risk of pregnancy loss foundthat, compared with no treatment, cer-clage was associated with a statistically sig-nificant increased risk of maternal fever(RR, 2.39; 95% CI, 1.35�4.23) and cesar-ean delivery (RR, 1.19; 95% CI,1.01�1.40).235 The authors of that re-view speculated that the higher rates ofcesarean delivery that were associatedwith cerclage could be due to biased
diagnosis of failed induction or failure wgy JANUARY 2013
to progress in labor when cliniciansknew that a woman had a cerclage ear-lier in pregnancy.
Implications for practiceThe optimal method to compare the ef-ficacy and safety of vaginal progesteroneand cerclage in women with a sono-graphic short cervix in the mid trimester,singleton gestation, and previous spon-taneous preterm is by a direct compari-son with a randomized controlled clini-cal trial. It is unknown whether such atrial will be forthcoming in the near fu-ture. We have performed a sample sizecalculation to estimate the number ofpatients who would be required to con-duct such a trial. Assuming a reductionin the frequency of preterm birth at �32
eeks of gestation from 19.2% in the cer-lage group to 12.0% in the vaginal pro-esterone group, 800 patients (400 perroup) would be required for this studyo have an 80% power with an alpha of.05. In the absence of such a trial, weelieve that the findings of the currenttudy provide the best available evidenceo counsel patients and inform physi-ians at this time.
Currently, the American College ofbstetricians and Gynecologists recom-ends the administration of 17�-
hydroxyprogesterone caproate for theprevention of preterm birth in womenwith a history of a spontaneous singletonpreterm birth at �37 weeks of gesta-tion.236 Thus far, there are no random-zed controlled trials that have compared7�-hydroxyprogesterone caproate
vs placebo, 17�-hydroxyprogesteronecaproate vs vaginal progesterone, or 17�-
ydroxyprogesterone caproate plus cer-lage vs cerclage alone or 17�-hydroxy-
progesterone caproate alone in womenwith a short cervix, singleton gestation,and previous spontaneous preterm birthfor the prevention of preterm birth. A re-cently published secondary analysis of thetrial by Owen et al174 evaluated the efficacyof cerclage vs no cerclage in patients with asingleton gestation and previous sponta-neous preterm birth who developed ashort cervix (�25 mm) in the second tri-mester while receiving 17�-hydroxypro-esterone caproate.237 Of the 99 women
ho received 17�-hydroxyprogesterone
awg1cwc2Twtaoogwot
hp�sbvags1
ctoccgbcamtpnsg
www.AJOG.org Reports of Major Impact
caproate, 47 were allocated to have a cer-clage, and 52 were allocated to the groupwho were treated without cerclage. Therates of preterm birth at �32, �28, �35,nd �37 weeks of gestation amongomen who received 17�-hydroxypro-esterone caproate and a cerclage were7%, 9%, 30%, and 49%, respectively. Theorresponding rates among womenho received 17�-hydroxyprogesterone
aproate in the no cerclage group were1%, 15%, 38%, and 60%, respectively.heseoutcomemeasures forpretermbirthere not significantly different between
he cerclage and no cerclage groups. Theuthors concluded that cerclage does notffer additional benefit for the preventionf preterm birth in women with a singletonestation and a cervical length of �25 mmho are receiving 17�-hydroxyprogester-ne caproate because of a previous pre-erm birth.
In our sensitivity analysis reportederein, we found that the frequency ofreterm birth at �32, �28, �35, and37 weeks of gestation in women with a
ingleton gestation, previous pretermirth, and a short cervix who receivedaginal progesterone was 11%, 7%, 24%,nd 44%, respectively. These data sug-est that vaginal progesterone is at leastimilar in efficacy to the combination of7�-hydroxyprogesterone caproate and
cerclage in the prevention of pretermbirth in women with a singleton gesta-tion, previous spontaneous pretermbirth, and a cervical length of �25 mm.Recently, a randomized clinical trial re-ported that 17�-hydroxyprogesteroneaproate did not reduce the risk of pre-erm birth at �32, �35, and �37 weeksf gestation when compared with pla-ebo in nulliparous women with a shortervix (�30 mm).234 In addition, sub-roup analyses did not demonstrate aenefit from 17�-hydroxyprogesteroneaproate administration to women withcervical length of �15 mm or 10-20m. Therefore, based on the totality of
he current available evidence, we pro-ose that women with a singleton preg-ancy who have a history of singletonpontaneous preterm birth and have be-un treatment with 17�-hydroxyproges-
terone caproate between 16 and 20 weeks
of gestation be followed with serial cer-vical length measurements using trans-vaginal sonography beginning at ap-proximately 18 weeks and continuingevery 2 weeks until 23 6/7 weeks. If cer-vical length is �25 mm, vaginal proges-terone should be offered to the patientbecause this intervention has beenproven to be effective in women with ashort cervix and a history of pretermbirth. If cervical length is 25-30 mm,these patients could be followed with ad-ditional ultrasound examinations be-cause they may still benefit from vaginalprogesterone. There is no evidence tosupport the continued administration of17�-hydroxyprogesterone caproate inpatients with a short cervix if vaginalprogesterone is used. This approachwould address the safety concerns thathave been outlined by the Food andDrug Administration in the package in-sert of the commercially available formof 17�-hydroxyprogesterone caproate(http://www.accessdata.fda.gov/drugsatfda_docs/label/2011/021945s000lbl.pdf) andthe lack of evidence that this syntheticprogestin would be effective if the pa-tient is already receiving vaginal proges-terone because of a history of pretermbirth and a cervix of �25 mm.
The key finding of this study is thatvaginal progesterone and cervical cer-clage have similar efficacy for the pre-vention of preterm birth and adverseperinatal outcomes in patients with ashort cervix and a history of pretermbirth. Given similar efficacy, therapeu-tic decision-making can be informedby reports about adverse events andcost-effectiveness of the interventions,as well as the patient and physician’spreferences.
The current recommendation that pa-tients with a short cervix and a history ofpreterm birth should be treated with cer-vical cerclage must be revisited in light ofthe results of the present study. Medicaltreatment with vaginal progesterone candecrease the risks that are associated withanesthesia and a surgical procedure;therefore, it is important to disclose theavailability of a non-surgical therapeuticchoice to patients with a history of pre-
term birth and a short cervix. fJANUARY 2013 Americ
REFERENCES1. Romero R, Mazor M, Munoz H, Gomez R,Galasso M, Sherer DM. The preterm labor syn-drome. Ann N Y Acad Sci 1994;734:414-29.2. Romero R, Gomez R, Mazor M, Ghezzi F,Yoon BH. The preterm labor syndrome. In: El-der M, Romero R, Lamont R, eds. Preterm la-bor. New York: Churchill Livingstone; 1997:29-49.3. Gotsch F, Romero R, Erez O, et al. The pre-term parturition syndrome and its implicationsfor understanding the biology, risk assessment,diagnosis, treatment and prevention of pretermbirth. J Matern Fetal Neonatal Med 2009;22(suppl 2):5-23.4. Di Renzo GC, Roura LC, Facchinetti F, et al.Guidelines for the management of spontaneouspreterm labor: identification of spontaneouspreterm labor, diagnosis of preterm prematurerupture of membranes, and preventive tools forpreterm birth. J Matern Fetal Neonatal Med2011;24:659-67.5. Romero R, Espinoza J, Kusanovic JP, et al.The preterm parturition syndrome. BJOG 2006;113(suppl 3):17-42.6. Simhan HN, Caritis SN. Prevention of pre-term delivery. N Engl J Med 2007;357:477-87.7. Fuchs F, Fuchs AR, Poblete VF Jr, Risk A.Effect of alcohol on threatened premature labor.Am J Obstet Gynecol 1967;99:627-37.8. Fuchs F. Use of ethanol in prevention of pre-mature delivery. Am J Obstet Gynecol 1971;110:1148-9.9. Lauersen NH, Merkatz IR, Tejani N, et al. In-hibition of premature labor: a multicenter com-parison of ritodrine and ethanol. Am J ObstetGynecol 1977;127:837-45.10. Briscoe CC. Failure of oral isoxsuprine toprevent prematurity. Am J Obstet Gynecol1966;95:885-6.11. Csapo AI, Herczeg J. Arrest of prematurelabor by isoxsuprine. Am J Obstet Gynecol1977;129:482-91.12. Ingemarsson I. Effect of terbutaline on pre-mature labor: a double-blind placebo-con-trolled study. Am J Obstet Gynecol 1976;125:520-4.13. Rodier P, Miller RK, Brent RL. Does treat-ment of premature labor with terbutaline in-crease the risk of autism spectrum disorders?Am J Obstet Gynecol 2011;204:91-4.14. Richter R. Evaluation of success in treat-ment of threatening premature labor by betami-metic drugs. Am J Obstet Gynecol 1977;127:482-6.15. Niebyl JR, Blake DA, White RD, et al. Theinhibition of premature labor with indomethacin.Am J Obstet Gynecol 1980;136:1014-9.16. Stubblefield PG. Indomethacin to stop pre-mature labor. Am J Obstet Gynecol 1976;125:571.17. Csapo AI, Puri CP, Tarro S, Henzl MR. De-activation of the uterus during normal and pre-mature labor by the calcium antagonist nicardi-
pine. Am J Obstet Gynecol 1982;142:483-91.an Journal of Obstetrics & Gynecology 42.e12
Reports of Major Impact www.AJOG.org
18. Carr DB, Clark AL, Kernek K, Spinnato JA.Maintenance oral nifedipine for preterm labor: arandomized clinical trial. Am J Obstet Gynecol1999;181:822-7.19. Papatsonis DN, Van Geijn HP, Dekker GA.Nifedipine as a safe and effective tocolytic agentin the treatment of preterm labor. Am J ObstetGynecol 2000;183:513-4.20. Conde-Agudelo A, Romero R, KusanovicJP. Nifedipine in the management of pretermlabor: a systematic review and metaanalysis.Am J Obstet Gynecol 2011;204:134.e1-20.21. Phaneuf S, Asboth G, MacKenzie IZ, MelinP, Lopez Bernal A. Effect of oxytocin antago-nists on the activation of human myometrium invitro: atosiban prevents oxytocin-induced de-sensitization. Am J Obstet Gynecol 1994;171:1627-34.22. Romero R, Sibai BM, Sanchez-Ramos L, etal. An oxytocin receptor antagonist (atosiban) inthe treatment of preterm labor: a randomized,double-blind, placebo-controlled trial with toco-lytic rescue. Am J Obstet Gynecol 2000;182:1173-83.23. Valenzuela GJ, Sanchez-Ramos L, RomeroR, et al. Maintenance treatment of preterm laborwith the oxytocin antagonist atosiban: the Ato-siban PTL-098 Study Group. Am J Obstet Gy-necol 2000;182:1184-90.24. Husslein P, Roura LC, Dudenhausen J, etal. Clinical practice evaluation of atosiban in pre-term labour management in six European coun-tries. Bjog 2006;113(suppl 3):105-10.25. Thornton S, Goodwin TM, Greisen G, He-degaard M, Arce JC. The effect of barusiban, aselective oxytocin antagonist, in threatenedpreterm labor at late gestational age: a random-ized, double-blind, placebo-controlled trial.Am J Obstet Gynecol 2009;200:627.e1-10.26. Baumbach J, Shi SQ, Shi L, Balducci J,Coonrod DV, Garfield RE. Inhibition of uterinecontractility with various tocolytics with andwithout progesterone: in vitro studies. Am J Ob-stet Gynecol 2012;206:254.e1-5.27. Morrison JC, Martin JN Jr, Martin RW,Gookin KS, Wiser WL. Prevention of pretermbirth by ambulatory assessment of uterine ac-tivity: a randomized study. Am J Obstet Gynecol1987;156:536-43.28. Han S, Crowther CA, Moore V. Magnesiummaintenance therapy for preventing pretermbirth after threatened preterm labour. CochraneDatabase Syst Rev 2010:CD000940.29. Buckingham JC, Buethe RA Jr, DanforthDN. Collagen-muscle ratio in clinically normaland clinically incompetent cervices. Am J Ob-stet Gynecol 1965;91:232-7.30. Buckingham JC, Selden R, Danforth DN.Connective tissue changes in the cervix duringpregnancy and labor. Ann N Y Acad Sci1962;97:733-42.31. Danforth DN. The fibrous nature of the hu-man cervix, and its relation to the isthmic seg-ment in gravid and nongravid uteri. Am J ObstetGynecol 1947;53:541-60.32. Danforth DN, Veis A, Breen M, Weinstein
HG, Buckingham JC, Manalo P. The effect of42.e13 American Journal of Obstetrics & Gynecolo
pregnancy and labor on the human cervix:changes in collagen, glycoproteins, andglycosaminoglycans. Am J Obstet Gynecol1974;120:641-51.33. Chwalisz K, Garfield RE. Role of nitric oxidein the uterus and cervix: implications for themanagement of labor. J Perinat Med 1998;26:448-57.34. Sennstrom MK, Brauner A, Lu Y, Gran-strom LM, Malmstrom AL, Ekman GE. Interleu-kin-8 is a mediator of the final cervical ripening inhumans. Eur J Obstet Gynecol Reprod Biol1997;74:89-92.35. Leppert PC, Woessner JF, Leppert PC,Woessner JF. The extracellular matrix of theuterus, cervix and fetal membranes: synthesis,degradation and hormonal regulation. In: Lip-pert PC, Woessner JF, editors. The extracellularmatrix of the uterus cervix and fetal membranes.Ithaca (NY): Perinatology Press; 1991:68-70.36. Leppert PC, Yu SY, Keller S, Cerreta J,Mandl I. Decreased elastic fibers and des-mosine content in incompetent cervix. Am JObstet Gynecol 1987;157:1134-9.37. Naftolin F, Stubblefield P. Dilatation of theuterine cervix: connective tissue biology andclinical management. New York: Raven PressBooks; 1980.38. Osmers R, Rath W, Adelmann-Grill BC, etal. Collagenase activity in the human cervix dur-ing parturition: the role of polymorphonuclearleukocytes. In: Lippert PC, ed. The extracellularmatrix of the uterus, cervix and fetal mem-branes: synthesis, degradation and hormonalregulation. Ithaca,NY:PerinatologyPress;1991:113-8.39. Romero R, Mazor M, Gomez R, GonzalezR, Galasso M, Cotton D. Cervix, incompetenceand premature labor. The Fetus 1993;3:1-10.40. Shirodkar VN. A new method of operativetreatment for habitual abortions in the secondtrimester of pregnancy. Antiseptic 1955;52:299-300.41. McDonald IA. Suture of the cervix for inevi-table miscarriage. J Obstet Gynaecol Br Emp1957;64:346-50.42. Lockwood CJ, Senyei AE, Dische MR, et al.Fetal fibronectin in cervical and vaginal secre-tions as a predictor of preterm delivery. N EnglJ Med 1991;325:669-74.43. Iams JD, Casal D, McGregor JA, et al. Fetalfibronectin improves the accuracy of diagnosisof preterm labor. Am J Obstet Gynecol 1995;173:141-5.44. Gomez R, Romero R, Medina L, et al. Cer-vicovaginal fibronectin improves the predictionof preterm delivery based on sonographic cer-vical length in patients with preterm uterine con-tractions and intact membranes. Am J ObstetGynecol 2005;192:350-9.45. Rozenberg P, Goffinet F, Malagrida L, et al.Evaluating the risk of preterm delivery: a com-parison of fetal fibronectin and transvaginalultrasonographic measurement of cervical
length. Am J Obstet Gynecol 1997;176:196-9.gy JANUARY 2013
46. Yoon BH, Romero R, Moon JB, et al. Thefrequency and clinical significance of intra-am-niotic inflammation in patients with a positivecervical fetal fibronectin. Am J Obstet Gynecol2001;185:1137-42.47. Conde-Agudelo A, Romero R. Cervicovagi-nal fetal fibronectin for the prediction of sponta-neous preterm birth in multiple pregnancies: asystematic review and meta-analysis. J MaternFetal Neonatal Med 2010;23:1365-76.48. Bartnicki J, Casal D, Kreaden US, Saling E,Vetter K. Fetal fibronectin in vaginal specimenspredicts preterm delivery and very-low-birth-weight infants. Am J Obstet Gynecol 1996;174:971-4.49. Peaceman AM, Andrews WW, Thorp JM,Cliver SP, Lukes A, Iams JD, Coultrip L, EriksenN, Holbrook RH, Elliott J, Ingardia C, PietrantoniM. Fetal fibronectin as a predictor of pretermbirth in patients with symptoms: a multicentertrial. Am J Obstet Gynecol 1997;177:13-8.50. Greenhagen JB, Van Wagoner J, Dudley D,et al. Value of fetal fibronectin as a predictor ofpreterm delivery for a low-risk population. Am JObstet Gynecol 1996;175:1054-6.51. Andrews WW, Sibai BM, Thom EA, et al.Randomized clinical trial of metronidazole pluserythromycin to prevent spontaneous pretermdelivery in fetal fibronectin-positive women. Ob-stet Gynecol 2003;101:847-55.52. Diaz-Cueto L, Dominguez-Lopez P, Tena-Alavez G, Cuica-Flores A, Rosales-Ortiz S, Are-chavaleta-Velasco F. Effect of clindamycintreatment on vaginal inflammatory markers inpregnant women with bacterial vaginosis and apositive fetal fibronectin test. Int J Gynaecol Ob-stet 2009;107:143-6.53. Shennan A, Crawshaw S, Briley A, et al. Arandomised controlled trial of metronidazole forthe prevention of preterm birth in women posi-tive for cervicovaginal fetal fibronectin: the PRE-MET Study. Bjog 2006;113:65-74.54. Chandiramani M, Di Renzo GC, GottschalkE, et al. Fetal fibronectin as a predictor of spon-taneous preterm birth: a European perspective.J Matern Fetal Neonatal Med 2011;24:330-6.55. Bolt LA, Chandiramani M, De Greeff A,Seed PT, Kurtzman J, Shennan AH. The valueof combined cervical length measurement andfetal fibronectin testing to predict spontaneouspreterm birth in asymptomatic high-riskwomen. J Matern Fetal Neonatal Med 2011;24:928-32.56. Lee SM, Romero R, Park JW, et al. Theclinical significance of a positive Amnisure test inwomen with preterm labor and intact mem-branes. J Matern Fetal Neonat 2012;25:1690-8.57. Wapner RJ, Cotton DB, Artal R, Librizzi RJ,Ross MG. A randomized multicenter trial as-sessing a home uterine activity monitoringdevice used in the absence of daily nursingcontact. Am J Obstet Gynecol 1995;172:1026-34.58. Iams JD, Newman RB, Thom EA, et al. Fre-
quency of uterine contractions and the risk of
www.AJOG.org Reports of Major Impact
spontaneous preterm delivery. N Engl J Med2002;346:250-5.59. Lawson Y, Dombrowski MP, Carter S, Hag-glund KH. Does external tocodynamometry in-crease maternal perception of uterine contrac-tions? Am J Obstet Gynecol 2003;189:1396-7.60. Millar LK, DeBuque L, Wing DA. Uterinecontraction frequency during treatment of pye-lonephritis in pregnancy and subsequent risk ofpreterm birth. J Perinat Med 2003;31:41-6.61. Berghella V, Iams JD, Newman RB, et al.Frequency of uterine contractions in asymp-tomatic pregnant women with or without a shortcervix on transvaginal ultrasound scan. Am JObstet Gynecol 2004;191:1253-6.62. Newman RB. Uterine contraction assess-ment. Obstet Gynecol Clin North Am 2005;32:341-67.63. Smith R, Van Helden D, Hirst J, et al. Path-ological interactions with the timing of birth anduterine activation. Aust N Z J Obstet Gynaecol2007;47:430-7.64. Most O, Langer O, Kerner R, David GB, Cal-deron I. Can myometrial electrical activity iden-tify patients in preterm labor? Am J Obstet Gy-necol 2008;199:378.e1-6.65. Lucovnik M, Maner WL, Chambliss LR, etal. Noninvasive uterine electromyography forprediction of preterm delivery. Am J Obstet Gy-necol 2011;204:228.e1-10.66. Rabotti C, Oei SG, van ’t Hooft J, Mischi M.Electrohysterographic propagation velocity forpreterm delivery prediction. Am J Obstet Gyne-col 2011;205:e9-10.67. Andersen HF, Nugent CE, Wanty SD,Hayashi RH. Prediction of risk for preterm deliv-ery by ultrasonographic measurement of cervi-cal length. Am J Obstet Gynecol 1990;163:859-67.68. Heath VC, Southall TR, Souka AP, ElisseouA, Nicolaides KH. Cervical length at 23 weeks ofgestation: prediction of spontaneous pretermdelivery. Ultrasound Obstet Gynecol 1998;12:312-7.69. Hassan SS, Romero R, Berry SM, et al. Pa-tients with an ultrasonographic cervical length� or � 15 mm have nearly a 50% risk of earlyspontaneous preterm delivery. Am J Obstet Gy-necol 2000;182:1458-67.70. Iams JD, Goldenberg RL, Meis PJ, et al. Thelength of the cervix and the risk of spontaneouspremature delivery: National Institute of ChildHealth and Human Development Maternal FetalMedicine Unit Network. N Engl J Med 1996;334:567-72.71. Berghella V, Tolosa JE, Kuhlman K, WeinerS, Bolognese RJ, Wapner RJ. Cervical ultra-sonography compared with manual examina-tion as a predictor of preterm delivery. Am JObstet Gynecol 1997;177:723-30.72. Owen J, Yost N, Berghella V, et al. Canshortened midtrimester cervical length predictvery early spontaneous preterm birth? Am JObstet Gynecol 2004;191:298-303.73. Rust OA, Atlas RO, Kimmel S, Roberts WE,Hess LW. Does the presence of a funnel in-
crease the risk of adverse perinatal outcome ina patient with a short cervix? Am J ObstetGynecol 2005;192:1060-6.74. Iams JD, Cebrik D, Lynch C, Behrendt N,Das A. The rate of cervical change and the phe-notype of spontaneous preterm birth. Am J Ob-stet Gynecol 2011;205:130.e1-6.75. Romero R, Espinoza J, Erez O, Hassan S.The role of cervical cerclage in obstetric prac-tice: can the patient who could benefit from thisprocedure be identified? Am J Obstet Gynecol2006;194:1-9.76. Romero R. Prevention of spontaneous pre-term birth: the role of sonographic cervicallength in identifying patients who may benefitfrom progesterone treatment. Ultrasound Ob-stet Gynecol 2007;30:675-86.77. Mackler AM, Iezza G, Akin MR, McMillan P,Yellon SM. Macrophage trafficking in the uterusand cervix precedes parturition in the mouse.Biol Reprod 1999;61:879-83.78. Uldbjerg N, Forman A, Reece EA, HobbinsJC. Biomechanical and biochemical changes ofthe uterus and cervix during pregnancy. In: Re-ece EA, Hobbins JC, eds. Medicine of the fetusand mother. Philadelphia: Lippincott-RavenPublishers; 1999:921-33.79. Winkler M, Rath W. Changes in the cervicalextracellular matrix during pregnancy and par-turition. J Perinat Med 1999;27:45-60.80. Word RA, Landrum CP, Timmons BC,Young SG, Mahendroo MS. Transgene inser-tion on mouse chromosome 6 impairs functionof the uterine cervix and causes failure of partu-rition. Biol Reprod 2005;73:1046-56.81. Hassan SS, Romero R, Haddad R, et al.The transcriptome of the uterine cervix beforeand after spontaneous term parturition. Am JObstet Gynecol 2006;195:778-86.82. Mahendroo MS, Cala KM, Russell DW. 5alpha-reduced androgens play a key role in mu-rine parturition. Mol Endocrinol 1996;10:380-92.83. Mahendroo MS, Porter A, Russell DW,Word RA. The parturition defect in steroid 5al-pha-reductase type 1 knockout mice is due toimpaired cervical ripening. Mol Endocrinol1999;13:981-92.84. Hassan SS, Romero R, Tarca AL, et al. Themolecular basis for sonographic cervical short-ening at term: identification of differentially ex-pressed genes and the epithelial-mesenchymaltransition as a function of cervical length. Am JObstet Gynecol 2010;203:472.e1-14.85. Akgul Y, Holt R, Mummert M, Word A, Ma-hendroo M. Dynamic changes in cervical gly-cosaminoglycan composition during normalpregnancy and preterm birth. Endocrinology2012;153:3493-503.86. Read CP, Word RA, Ruscheinsky MA, Tim-mons BC, Mahendroo MS. Cervical remodelingduring pregnancy and parturition: molecularcharacterization of the softening phase in mice.Reproduction 2007;134:327-40.87. Mahendroo M. Cervical remodeling in termand preterm birth: insights from an animal
model. Reproduction 2012;143:429-38.JANUARY 2013 Americ
88. Word RA, Li XH, Hnat M, Carrick K. Dynam-ics of cervical remodeling during pregnancy andparturition: mechanisms and current concepts.Semin Reprod Med 2007;25:69-79.89. Andersson S, Minjarez D, Yost NP, WordRA. Estrogen and progesterone metabolism inthe cervix during pregnancy and parturition.J Clin Endocrinol Metab 2008;93:2366-74.90. Akins ML, Luby-Phelps K, Bank RA, Ma-hendroo M. Cervical softening during pregnan-cy: regulated changes in collagen cross-linkingand composition of matricellular proteins in themouse. Biol Reprod 2011;84:1053-62.91. Holt R, Timmons BC, Akgul Y, Akins ML,Mahendroo M. The molecular mechanisms ofcervical ripening differ between term and pre-term birth. Endocrinology 2011;152:1036-46.92. Timmons B, Akins M, Mahendroo M. Cervi-cal remodeling during pregnancy and parturi-tion. Trends Endocrinol Metab 2010;21:353-61.93. Gonzalez JM, Dong Z, Romero R, Girardi G.Cervical remodeling/ripening at term and pre-term delivery: the same mechanism initiated bydifferent mediators and different effector cells.PLoS One 2011;6:e26877.94. Gonzalez JM, Franzke CW, Yang F, Ro-mero R, Girardi G. Complement activation trig-gers metalloproteinases release inducing cervi-cal remodeling and preterm birth in mice. Am JPathol 2011;179:838-49.95. Hassan SS, Romero R, Tarca AL, et al. Thetranscriptome of cervical ripening in humanpregnancy before the onset of labor at term:identification of novel molecular functions in-volved in this process. J Matern Fetal NeonatalMed 2009;22:1183-93.96. Kuon RJ, Shi SQ, Maul H, et al. Pharmaco-logic actions of progestins to inhibit cervical rip-ening and prevent delivery depend on theirproperties, the route of administration, and thevehicle. Am J Obstet Gynecol 2010;202:455.e1-9.97. Arabin B, Halbesma JR, Vork F, Hubener M,van Eyck J. Is treatment with vaginal pessariesan option in patients with a sonographically de-tected short cervix? J Perinat Med 2003;31:122-33.98. Kimber-Trojnar Z, Patro-Malysza J, Leszc-zynska-Gorzelak B, Marciniak B, Oleszczuk J.Pessary use for the treatment of cervical incom-petence and prevention of preterm labour. JMatern Fetal Neonatal Med 2010;23:1493-9.99. Goya M, Pratcorona L, Merced C, et al. Cer-vical pessary in pregnant women with a shortcervix (PECEP): an open-label randomised con-trolled trial. Lancet 2012;379:1800-6.100. Campbell S. Universal cervical-lengthscreening and vaginal progesterone preventsearly preterm births, reduces neonatal morbid-ity and is cost saving: doing nothing is no longeran option. Ultrasound Obstet Gynecol 2011;38:1-9.101. Fonseca EB, Celik E, Parra M, Singh M,Nicolaides KH. Progesterone and the risk ofpreterm birth among women with a short cervix.
N Engl J Med 2007;357:462-9.an Journal of Obstetrics & Gynecology 42.e14
Reports of Major Impact www.AJOG.org
102. DeFranco EA, O’Brien JM, Adair CD, et al.Vaginal progesterone is associated with a de-crease in risk for early preterm birth and im-proved neonatal outcome in women with ashort cervix: a secondary analysis from a ran-domized, double-blind, placebo-controlled trial.Ultrasound Obstet Gynecol 2007;30:697-705.103. O’Brien JM, Defranco EA, Adair CD, et al.Effect of progesterone on cervical shortening inwomen at risk for preterm birth: secondary anal-ysis from a multinational, randomized, double-blind, placebo-controlled trial. Ultrasound Ob-stet Gynecol 2009;34:653-9.104. Hassan SS, Romero R, Vidyadhari D, et al.Vaginal progesterone reduces the rate of pre-term birth in women with a sonographic shortcervix: a multicenter, randomized, double-blind,placebo-controlled trial. Ultrasound Obstet Gy-necol 2011;38:18-31.105. Romero R, Nicolaides K, Conde-AgudeloA, et al. Vaginal progesterone in women with anasymptomatic sonographic short cervix in themidtrimester decreases preterm delivery andneonatal morbidity: a systematic review andmetaanalysis of individual patient data. Am JObstet Gynecol 2012;206:124.e1-19.106. Guzman ER, Forster JK, Vintzileos AM,Ananth CV, Walters C, Gipson K. Pregnancyoutcomes in women treated with elective ver-sus ultrasound-indicated cervical cerclage. Ul-trasound Obstet Gynecol 1998;12:323-7.107. Althuisius S, Dekker G, Hummel P, Beke-dam D, Kuik D, van Geijn H. Cervical Incompe-tence Prevention Randomized Cerclage Trial(CIPRACT): effect of therapeutic cerclage withbed rest vs. bed rest only on cervical length.Ultrasound Obstet Gynecol 2002;20:163-7.108. Hibbard JU, Snow J, Moawad AH. Shortcervical length by ultrasound and cerclage. JPerinatol 2000;20:161-5.109. Groom KM, Shennan AH, Bennett PR. Ul-trasound-indicated cervical cerclage: outcomedepends on preoperative cervical length andpresence of visible membranes at time of cer-clage. Am J Obstet Gynecol 2002;187:445-9.110. Hassan SS, Romero R, Maymon E, BerrySM, Blackwell SC, Treadwell MC, Tomlinson M.Does cervical cerclage prevent preterm deliveryin patients with a short cervix? Am J ObstetGynecol 2001;184:1325-9.111. Baxter JK, Airoldi J, Berghella V. Shortcervical length after history-indicated cerclage:is a reinforcing cerclage beneficial? Am J ObstetGynecol 2005;193:1204-7.112. Belej-Rak T, Okun N, Windrim R, Ross S,Hannah ME. Effectiveness of cervical cerclagefor a sonographically shortened cervix: a sys-tematic review and meta-analysis. Am J ObstetGynecol 2003;189:1679-87.113. Berghella V, Odibo AO, Tolosa JE. Cer-clage for prevention of preterm birth in womenwith a short cervix found on transvaginal ultra-sound examination: a randomized trial. Am JObstet Gynecol 2004;191:1311-7.114. Kurup M, Goldkrand JW. Cervical incom-petence: elective, emergent, or urgent cer-
clage. Am J Obstet Gynecol 1999;181:240-6.42.e15 American Journal of Obstetrics & Gynecolo
115. Althuisius SM, Dekker GA, Hummel P,Bekedam DJ, van Geijn HP. Final results of theCervical Incompetence Prevention Random-ized Cerclage Trial (CIPRACT): therapeutic cer-clage with bed rest vs bed rest alone. Am JObstet Gynecol 2001;185:1106-12.116. Owen J, Yost N, Berghella V, et al. Mid-trimester endovaginal sonography in women athigh risk for spontaneous preterm birth. JAMA2001;286:1340-8.117. Taipale P, Hiilesmaa V. Sonographic mea-surement of uterine cervix at 18-22 weeks’ ges-tation and the risk of preterm delivery. ObstetGynecol 1998;92:902-7.118. To MS, Skentou C, Liao AW, Cacho A,Nicolaides KH. Cervical length and funneling at23 weeks of gestation in the prediction of spon-taneous early preterm delivery. Ultrasound Ob-stet Gynecol 2001;18:200-3.119. Guzman ER, Pisatowski DM, VintzileosAM, Benito CW, Hanley ML, Ananth CV. A com-parison of ultrasonographically detected cervi-cal changes in response to transfundal pres-sure, coughing, and standing in predictingcervical incompetence. Am J Obstet Gynecol1997;177:660-5.120. Guzman ER, Vintzileos AM, McLean DA,Martins ME, Benito CW, Hanley ML. The naturalhistory of a positive response to transfundalpressure in women at risk for cervical incompe-tence. Am J Obstet Gynecol 1997;176:634-8.121. Kushnir O, Vigil DA, Izquierdo L, Schiff M,Curet LB. Vaginal ultrasonographic assessmentof cervical length changes during normal preg-nancy. Am J Obstet Gynecol 1990;162:991-3.122. Okitsu O, Mimura T, Nakayama T, Aono T.Early prediction of preterm delivery by transvag-inal ultrasonography. Ultrasound Obstet Gyne-col 1992;2:402-9.123. Romero R. Prenatal medicine: the child isthe father of the man. Prenat Neonatal Med1996;1:8-11.124. Di Renzo GC. The great obstetrical syn-dromes. J Matern Fetal Neonatal Med 2009;22:633-5.125. Levine RU, Berkowitz KM. Conservativemanagement and pregnancy outcome in dieth-ylstilbestrol-exposed women with and withoutgross genital tract abnormalities. Am J ObstetGynecol 1993;169:1125-9.126. Ludmir J, Landon MB, Gabbe SG, Samu-els P, Mennuti MT. Management of the diethyl-stilbestrol-exposed pregnant patient: a pro-spective study. Am J Obstet Gynecol 1987;157:665-9.127. Mangan CE, Borow L, Burtnett-RubinMM, Egan V, Giuntoli RL, Mikuta JJ. Pregnancyoutcome in 98 women exposed to diethylstil-bestrol in utero, their mothers, and unexposedsiblings. Obstet Gynecol 1982;59:315-9.128. Raio L, Ghezzi F, Di Naro E, Gomez R,Luscher KP. Duration of pregnancy after carbondioxide laser conization of the cervix: influenceof cone height. Obstet Gynecol 1997;90:978-82.129. Bruinsma FJ, Quinn MA. The risk of pre-
term birth following treatment for precancerousgy JANUARY 2013
changes in the cervix: a systematic review andmeta-analysis. Bjog 2011;118:1031-41.130. Bevis KS, Biggio JR. Cervical conizationand the risk of preterm delivery. Am J ObstetGynecol 2011;205:19-27.131. Armarnik S, Sheiner E, Piura B, MeirovitzM, Zlotnik A, Levy A. Obstetric outcome follow-ing cervical conization. Arch Gynecol Obstet2011;283:765-9.132. Nam KH, Kwon JY, Kim YH, Park YW.Pregnancy outcome after cervical conization:risk factors for preterm delivery and the efficacyof prophylactic cerclage. J Gynecol Oncol2010;21:225-9.133. Andia D, Mozo de Rosales F, Villasante A,Rivero B, Diez J, Perez C. Pregnancy outcomein patients treated with cervical conization forcervical intraepithelial neoplasia. Int J GynaecolObstet 2011;112:225-8.134. van de Vijver A, Poppe W, Verguts J, Ar-byn M. Pregnancy outcome after cervical coni-sation: a retrospective cohort study in the Leu-ven University Hospital. Bjog 2010;117:268-73.135. Ortoft G, Henriksen T, Hansen E, PetersenL. After conisation of the cervix, the perinatalmortality as a result of preterm delivery in-creases in subsequent pregnancy. Bjog2010;117:258-67.136. Masamoto H, Nagai Y, Inamine M, et al.Outcome of pregnancy after laser conization:implications for infection as a causal link withpreterm birth. J Obstet Gynaecol Res 2008;34:838-42.137. Albrechtsen S, Rasmussen S, ThoresenS, Irgens LM, Iversen OE. Pregnancy outcomein women before and after cervical conisation:population based cohort study. Bmj 2008;337:a1343.138. Patrelli TS, Anfuso S, Vandi F, et al. Pre-term delivery and premature rupture of mem-branes after conization in 80 women: prelimi-nary data. Minerva Ginecol 2008;60:295-8.139. Sjoborg KD, Vistad I, Myhr SS, et al. Preg-nancy outcome after cervical cone excision: acase-control study. Acta Obstet GynecolScand 2007;86:423-8.140. Klaritsch P, Reich O, Giuliani A, TamussinoK, Haas J, Winter R. Delivery outcome aftercold-knife conization of the uterine cervix. Gy-necol Oncol 2006;103:604-7.141. Kristensen J, Langhoff-Roos J, Wittrup M,Bock JE. Cervical conization and preterm deliv-ery/low birth weight; a systematic review of theliterature. Acta Obstet Gynecol Scand 1993;72:640-4.142. Hagen B, Skjeldestad FE. The outcome ofpregnancy after CO2 laser conisation of the cer-vix. BJOG 1993;100:717-20.143. Kristensen GB. The outcome of preg-nancy and preterm delivery after conization ofthe cervix. Arch Gynecol 1985;236:127-30.144. Leiman G, Harrison NA, Rubin A. Preg-nancy following conization of the cervix: com-plications related to cone size. Am J Obstet Gy-necol 1980;136:14-8.145. Moinian M, Andersch B. Does cervix
conization increase the risk of complications in
www.AJOG.org Reports of Major Impact
subsequent pregnancies? Acta Obstet GynecolScand 1982;61:101-3.146. Arbyn M, Kyrgiou M, Simoens C, et al.Perinatal mortality and other severe adversepregnancy outcomes associated with treat-ment of cervical intraepithelial neoplasia: meta-analysis. Bmj 2008;337:a1284.147. Sadler L, Saftlas A. Cervical surgery andpreterm birth. J Perinat Med 2007;35:5-9.148. Jakobsson M, Gissler M, Paavonen J,Tapper AM. Loop electrosurgical excision pro-cedure and the risk for preterm birth. ObstetGynecol 2009;114:504-10.149. Fischer RL, Sveinbjornsson G, Hansen C.Cervical sonography in pregnant women with aprior cone biopsy or loop electrosurgical exci-sion procedure. Ultrasound Obstet Gynecol2010;36:613-7.150. Shin MY, Seo ES, Choi SJ, et al. The role ofprophylactic cerclage in preventing preterm de-livery after electrosurgical conization. J GynecolOncol 2010;21:230-6.151. Romero R, Gonzalez R, Sepulveda W, etal. Infection and labor: VIII, microbial invasion ofthe amniotic cavity in patients with suspectedcervical incompetence: prevalence and clinicalsignificance. Am J Obstet Gynecol 1992;167:1086-91.152. Mays JK, Figueroa R, Shah J, Khakoo H,Kaminsky S, Tejani N. Amniocentesis for selec-tion before rescue cerclage. Obstet Gynecol2000;95:652-5.153. Gomez R, Romero R, Nien JK, et al. Ashort cervix in women with preterm labor andintact membranes: a risk factor for microbial in-vasion of the amniotic cavity. Am J Obstet Gy-necol 2005;192:678-89.154. Hassan S, Romero R, Hendler I, et ak. Asonographic short cervix as the only clinicalmanifestation of intra-amniotic infection. J Peri-nat Med 2006;34:13-9.155. Kiefer DG, Keeler SM, Rust OA, WayockCP, Vintzileos AM, Hanna N. Is midtrimestershort cervix a sign of intraamniotic inflamma-tion? Am J Obstet Gynecol 2009;200:374.e1-5.156. Vaisbuch E, Hassan SS, Mazaki-Tovi S, etal. Patients with an asymptomatic short cervix(�15 mm) have a high rate of subclinical intra-amniotic inflammation: implications for patientcounseling. Am J Obstet Gynecol 2010;202:433.e1-8.157. Holst RM, Jacobsson B, Hagberg H,Wennerholm UB. Cervical length in women inpreterm labor with intact membranes: relation-ship to intra-amniotic inflammation/microbial in-vasion, cervical inflammation and preterm deliv-ery. Ultrasound Obstet Gynecol 2006;28:768-74.158. Lee SE, Romero R, Park CW, Jun JK,Yoon BH. The frequency and significance of in-traamniotic inflammation in patients with cervi-cal insufficiency. Am J Obstet Gynecol 2008;198:633.e1-8.159. Vaisbuch E, Romero R, Erez O, et al. Clin-ical significance of early (�20 weeks) vs late
(20-24 weeks) detection of sonographic shortcervix in asymptomatic women in themid-trimester. Ultrasound Obstet Gynecol2010;36:471-81.160. Rizzo G, Capponi A, Vlachopoulou A,Angelini E, Grassi C, Romanini C. Ultrasono-graphic assessment of the uterine cervix andinterleukin-8 concentrations in cervical secre-tions predict intrauterine infection in patientswith preterm labor and intact membranes. Ul-trasound Obstet Gynecol 1998;12:86-92.161. Vaisbuch E, Romero R, Mazaki-Tovi S, etal. The risk of impending preterm delivery inasymptomatic patients with a nonmeasurablecervical length in the second trimester. Am JObstet Gynecol 2010;203:446.e1-9.162. Straach KJ, Shelton JM, Richardson JA,Hascall VC, Mahendroo MS. Regulation of hya-luronan expression during cervical ripening.Glycobiology 2005;15:55-65.163. Roberson AE, Hyatt K, Kenkel C, HansonK, Myers DA. Interleukin 1beta regulates pro-gesterone metabolism in human cervical fibro-blasts. Reprod Sci 2012;19:271-81.164. Yellon SM, Ebner CA, Sugimoto Y. Partu-rition and recruitment of macrophages in cervixof mice lacking the prostaglandin F receptor.Biol Reprod 2008;78:438-44.165. Ledger WL, Webster MA, Anderson AB,Turnbull AC. Effect of inhibition of prostaglandinsynthesis on cervical softening and uterine ac-tivity during ovine parturition resulting from pro-gesterone withdrawal induced by epostane. JEndocrinol 1985;105:227-33.166. Danforth DN. The distribution and func-tional activity of the cervical musculature. Am JObstet Gynecol 1954;68:1261-71.167. Danforth DN, Buckingham JC. Cervical in-competence: a re-evaluation. Postgrad Med1962;32:345-51.168. Danforth DN, Buckingham JC, RoddickJW Jr. Connective tissue changes incident tocervical effacement. Am J Obstet Gynecol1960;80:939-45.169. Danforth DN, Veis A, Breen M, WeinsteinHG, Buckingham JC, Manalo P. The effect ofpregnancy and labor on the human cervix:changes in collagen, glycoproteins, and glycos-aminoglycans. Am J Obstet Gynecol 1974;120:641-51.170. Uldbjerg N. Cervical connective tissue inrelation to pregnancy, labour, and treatmentwith prostaglandin E2. Acta Obstet GynecolScandSuppl 1989;148:1-40.171. Uldbjerg N. Preterm delivery. Acta ObstetGynecol Scand 2005;84:515.172. Uldbjerg N, Ekman G, Malmstrom A, Ol-sson K, Ulmsten U. Ripening of the human uter-ine cervix related to changes in collagen, gly-cosaminoglycans, and collagenolytic activity.Am J Obstet Gynecol 1983;147:662-6.173. Sundtoft I, Sommer S, Uldbjerg N. Cervi-cal collagen concentration within 15 months af-ter delivery. Am J Obstet Gynecol 2011;205:59.e1-3.174. Owen J, Hankins G, Iams JD, et al. Multi-center randomized trial of cerclage for preterm
birth prevention in high-risk women with short-JANUARY 2013 Americ
ened midtrimester cervical length. Am J ObstetGynecol 2009;201:375.e1-8.175. Berghella V, Haas S, Chervoneva I, HyslopT. Patients with prior second-trimester loss:prophylactic cerclage or serial transvaginalsonograms? Am J Obstet Gynecol 2002;187:747-51.176. Berghella V, Odibo AO, To MS, Rust OA,Althuisius SM. Cerclage for short cervix on ul-trasonography: meta-analysis of trials using in-dividual patient-level data. Obstet Gynecol2005;106:181-9.177. Berghella V, Rafael TJ, Szychowski JM,Rust OA, Owen J. Cerclage for short cervix onultrasonography in women with singleton ges-tations and previous preterm birth: a meta-anal-ysis. Obstet Gynecol 2011;117:663-71.178. Zacharakis D, Daskalakis G, PapantoniouN, et al. Is treatment with cervical pessaries anoption in pregnant women with a mid-trimestershort cervix? J Matern Fetal Neonatal Med2012;25:52.179. Althuisius SM, Dekker GA, Hummel P, vanGeijn HP. Cervical incompetence preventionrandomized cerclage trial: emergency cerclagewith bed rest vs bed rest alone. Am J ObstetGynecol 2003;189:907-10.180. Daskalakis G, Papantoniou N, MesogitisS, Antsaklis A. Management of cervical insuffi-ciency and bulging fetal membranes. ObstetGynecol 2006;107:221-6.181. Stupin JH, David M, Siedentopf JP,Dudenhausen JW. Emergency cerclage versusbed rest for amniotic sac prolapse before 27gestational weeks: a retrospective, compara-tive study of 161 women. Eur J Obstet GynecolReprod Biol 2008;139:32-7.182. Ventolini G, Genrich TJ, Roth J, Neiger R.Pregnancy outcome after placement of ’rescue’Shirodkar cerclage. J Perinatol 2009;29:276-9.183. Cockwell HA, Smith GN. Cervical incom-petence and the role of emergency cerclage. JObstet Gynaecol Can 2005;27:123-9.184. Matijevic R, Olujic B, Tumbri J, Kurjak A.Cervical incompetence: the use of selective andemergency cerclage. J Perinat Med 2001;29:31-5.185. Novy MJ, Gupta A, Wothe DD, Gupta S,Kennedy KA, Gravett MG. Cervical cerclage inthe second trimester of pregnancy: a historicalcohort study. Am J Obstet Gynecol 2001;184:1447-54.186. Novy MJ, Haymond J, Nichols M. Shirod-kar cerclage in a multifactorial approach to thepatient with advanced cervical changes. Am JObstet Gynecol 1990;162:1412-9.187. American College of Obstetricians andGynecologists. Committee opinion no. 522: in-cidentally detected short cervical length. ObstetGynecol 2012;119:879.188. Progesterone and preterm birth preven-tion: translating clinical trials data into clinicalpractice. Am J Obstet Gynecol 2012;206:376-86.189. Parry S, Simhan H, Elovitz M, Iams J. Uni-versal maternal cervical length screening during
the second trimester: pros and cons of a strat-an Journal of Obstetrics & Gynecology 42.e16
Reports of Major Impact www.AJOG.org
egy to identify women at risk of spontaneouspreterm delivery. Am J Obstet Gynecol 2012;207:101-6.190. Song F, Altman DG, Glenny AM, Deeks JJ.Validity of indirect comparison for estimating ef-ficacy of competing interventions: empirical ev-idence from published meta-analyses. BMJ2003;326:472.191. Sutton A, Ades AE, Cooper N, Abrams K.Use of indirect and mixed treatment compari-sons for technology assessment. Pharmaco-economics 2008;26:753-67.192. Wells GA, Sultan A, Chen L, Khan M,Coyle D. Indirect evidence: indirect treatmentcomparisons in meta-analysis. Ottawa, Cana-da: Canadian Agency for Drugs and Technolo-gies in Health; 2009.193. Edwards SJ, Clarke MJ, Wordsworth S,Borrill J. Indirect comparisons of treatmentsbased on systematic reviews of randomisedcontrolled trials. Int J Clin Pract 2009;63:841-54.194. Liberati A, Altman DG, Tetzlaff J, et al. ThePRISMA statement for reporting systematic re-views and meta-analyses of studies that evalu-ate healthcare interventions: explanation andelaboration. BMJ 2009;339:b2700.195. Riley RD, Lambert PC, Abo-Zaid G. Meta-analysis of individual participant data: rationale,conduct, and reporting. BMJ 2010;340:c221.196. Donegan S, Williamson P, Gamble C, Tu-dur-Smith C. Indirect comparisons: a review ofreporting and methodological quality. PLoSOne 2010;5:e11054.197. Higgins JPT, Altman DG, Sterne JAC.Chapter 8: Assessing risk of bias in includedstudies. In: Higgins JPT, Green S, eds. Co-chrane handbook for systematic reviews of in-terventions, version 5.1.0 (updated March2011). The Cochrane Collaboration, 2011.Available at: www.cochrane-handbook.org.Accessed Nov. 2, 2012.198. Simmonds MC, Higgins JP, Stewart LA,Tierney JF, Clarke MJ, Thompson SG. Meta-analysis of individual patient data from random-ized trials: a review of methods used in practice.Clin Trials 2005;2:209-17.199. Higgins JP, Thompson SG, Deeks JJ, Alt-man DG. Measuring inconsistency in meta-analyses. BMJ 2003;327:557-60.200. Altman DG. Confidence intervals for thenumber needed to treat. BMJ 1998;317:1309-12.201. Egger M, Davey Smith G, Schneider M,Minder C. Bias in meta-analysis detected by asimple, graphical test. BMJ 1997;315:629-34.202. Bucher HC, Guyatt GH, Griffith LE, WalterSD. The results of direct and indirect treatmentcomparisons in meta-analysis of randomizedcontrolled trials. J Clin Epidemiol 1997;50:683-91.203. Glenny AM, Altman DG, Song F, et al. In-direct comparisons of competing interventions.Health Technol Assess 2005;9:1-134, iii-iv.204. da Fonseca EB, Bittar RE, Carvalho MH,Zugaib M. Prophylactic administration of pro-
gesterone by vaginal suppository to reduce the42.e17 American Journal of Obstetrics & Gynecolo
incidence of spontaneous preterm birth inwomen at increased risk: a randomizedplacebo-controlled double-blind study. Am JObstet Gynecol 2003;188:419-24.205. Majhi P, Bagga R, Kalra J, Sharma M. In-travaginal use of natural micronised progester-one to prevent pre-term birth: a randomised trialin India. J Obstet Gynaecol 2009;29:493-8.206. Norman JE, Mackenzie F, Owen P, et al.Progesterone for the prevention of preterm birthin twin pregnancy (STOPPIT): a randomised,double-blind, placebo-controlled study andmeta-analysis. Lancet 2009;373:2034-40.207. Rode L, Klein K, Nicolaides K, Krampl-Bettelheim E, Tabor A. Prevention of pretermdelivery in twin gestations (PREDICT): a multi-centre randomised placebo-controlled trial onthe effect of vaginal micronised progesterone.Ultrasound Obstet Gynecol 2011;38:272-80.208. Borna S, Sahabi N. Progesterone formaintenance tocolytic therapy after threatenedpreterm labour: a randomised controlled trial.Aust N Z J Obstet Gynaecol 2008;48:58-63.209. Sharami SH, Zahiri Z, Shakiba M, Milani F.Maintenance therapy by vaginal progesteroneafter threatened idiopathic preterm labor: a ran-domized placebo-controlled double-blind trial.Int J Fertil Steril 2010;4:45-50.210. Arikan I, Barut A, Harma M, Harma IM.Effect of progesterone as a tocolytic and inmaintenance therapy during preterm labor. Gy-necol Obstet Invest 2011;72:269-73.211. Rush RW, Isaacs S, McPherson K, JonesL, Chalmers I, Grant A. A randomized controlledtrial of cervical cerclage in women at high risk ofspontaneous preterm delivery. BJOG 1984;91:724-30.212. Lazar P, Gueguen S, Dreyfus J, Renaud R,Pontonnier G, Papiernik E. Multicentred con-trolled trial of cervical cerclage in women atmoderate risk of preterm delivery. BJOG1984;91:731-5.213. Szeverenyi M, Chalmels J, Grant A, et al.[Surgical cerclage in the treatment of cervicalincompetence during pregnancy (determiningthe legitimacy of the procedure)]. Orv Hetil1992;133:1823-6.214. Final report of the Medical ResearchCouncil/Royal College of Obstetricians andGynaecologists multicentre randomised trial ofcervical cerclage. MRC/RCOG Working Partyon Cervical Cerclage. BJOG 1993;100:516-23.215. Ezechi OC, Kalu BK, Nwokoro CA. Pro-phylactic cerclage for the prevention of pretermdelivery. Int J Gynaecol Obstet 2004;85:283-4.216. Kassanos D, Salamalekis E, Vitoratos N,Panayotopoulos N, Loghis C, Creatsas C. Thevalue of transvaginal ultrasonography in diagno-sis and management of cervical incompetence.Clin Exp Obstet Gynecol 2001;28:266-8.217. Beigi A ZF. Elective versus ultrasound-in-dicated cervical cerclage in women at risk forcervical incompetence. MJIRI 2005;19:5.218. Simcox R, Seed PT, Bennett P, Teoh TG,Poston L, Shennan AH. A randomized con-trolled trial of cervical scanning vs history to de-
termine cerclage in women at high risk of pre-gy JANUARY 2013
term birth (CIRCLE trial). Am J Obstet Gynecol2009;200:623.e1-6.219. Caspi E, Schneider D, Sadovsky G, Wein-raub Z, Bukovsky I. Diameter of cervical internalos after induction of early abortion by laminariaor rigid dilatation. Am J Obstet Gynecol1983;146:106-8.220. Tsai YL, Lin YH, Chong KM, Huang LW,Hwang JL, Seow KM. Effectiveness of doublecervical cerclage in women with at least oneprevious pregnancy loss in the second trimes-ter: a randomized controlled trial. J ObstetGynaecol Res 2009;35:666-71.221. Broumand F, Bahadori F, Behrouzilak T,Yekta Z, Ashrafi F. Viable extreme preterm birthand some neonatal outcomes in double cerclageversus traditional cerclage: a randomized clinicaltrial. ScientificWorldJournal 2011;11:1660-6.222. Dor J, Shalev J, Mashiach S, Blankstein J,Serr DM. Elective cervical suture of twin preg-nancies diagnosed ultrasonically in the first tri-mester following induced ovulation. GynecolObstet Invest 1982;13:55-60.223. Rust O, Atlas R, Wells M, Rawlinson K.Cerclage in multiple gestation with midtrimesterdilation of the internal os. Am J Obstet Gynecol2001;185(suppl):S111.224. Forster F DR, Schwarzlos G. [Therapy ofcervix insufficiency: cerclage or support pes-sary?] Zentralbl Gynakol 1986;108:7.225. Blair O, Fletcher H, Kulkarni S. A ran-domised controlled trial of outpatient versus in-patient cervical cerclage. J Obstet Gynaecol2002;22:493-7.226. Keeler SM, Kiefer D, Rochon M, QuinonesJN, Novetsky AP, Rust O. A randomized trial ofcerclage vs 17 alpha-hydroxyprogesteronecaproate for treatment of short cervix. J PerinatMed 2009;37:473-9.227. O’Brien JM, Adair CD, Lewis DF, et al. Pro-gesterone vaginal gel for the reduction of recur-rent preterm birth: primary results from a ran-domized, double-blind, placebo-controlled trial.Ultrasound Obstet Gynecol 2007;30:687-96.228. Cetingoz E, Cam C, Sakalli M, Karateke A,Celik C, Sancak A. Progesterone effects on pre-term birth in high-risk pregnancies: a random-ized placebo-controlled trial. Arch Gynecol Ob-stet 2011;283:423-9.229. Rust OA, Atlas RO, Reed J, van Gaalen J,Balducci J. Revisiting the short cervix detectedby transvaginal ultrasound in the second trimes-ter: why cerclage therapy may not help. Am JObstet Gynecol 2001;185:1098-105.230. To MS, Alfirevic Z, Heath VC, et al. Cervicalcerclage for prevention of preterm delivery inwomen with short cervix: randomised con-trolled trial. Lancet 2004;363:1849-53.231. Berghella V, Figueroa D, Szychowski JM, etal. 17-Alpha-hydroxyprogesterone caproate forthe prevention of preterm birth in women withprior preterm birth and a short cervical length.Am J Obstet Gynecol 2010;202:351.e1-6.232. Song F, Xiong T, Parekh-Bhurke S, et al. In-consistency between direct and indirect compari-sons of competing interventions: meta-epide-
miological study. BMJ 2011;343:d4909.
www.AJOG.org Reports of Major Impact
233. Mills EJ, Ghement I, O’Regan C, Thorlund K.Estimatingthepowerof indirectcomparisons:asim-ulation study. PLoS One 2011;6:e16237.234. Grobman WA, Thom EA, Spong CY, et al.17 alpha-hydroxyprogesterone caproate toprevent prematurity in nulliparas with cervicallength less than 30 mm. Am J Obstet Gynecol
2012;207:390.e1-8.235. Alfirevic Z, Stampalija T, Roberts D, Jor-gensen AL. Cervical stitch (cerclage) for pre-venting preterm birth in singleton pregnancy.Cochrane Database Syst Rev 2012;4:CD008991.236. American College of Obstetricians andGynecologists and Society for Maternal Fetal
Medicine. Use of progesterone to reduce pre-JANUARY 2013 Americ
term birth: ACOG committee opinion, no.419. Obstet Gynecol 2008;112:963-5.237. Szychowski JM, Berghella V, Owen J, etal. Cerclage for the prevention of preterm birthin high risk women receiving intramuscular17-alpha-hydroxyprogesterone caproate. JMatern Fetal Neonatal Med 2012 [Epub
ahead of print].an Journal of Obstetrics & Gynecology 42.e18