The Journal of Maternal-Fetal & Neonatal Medicine

ISSN: 1476-7058 (Print) 1476-4954 (Online) Journal homepage: https://www.tandfonline.com/loi/ijmf20

Cervical cerclage, pessary, or vaginal progesterone

in high-risk pregnant women with short cervix: a
randomized feasibility study

A. Care, R. Jackson, E. O’Brien, S. Leigh, C. Cornforth, A. Haycox, M.

Whitworth, T. Lavender & Z. Alfirevic

To cite this article: A. Care, R. Jackson, E. O’Brien, S. Leigh, C. Cornforth, A. Haycox, M.

Whitworth, T. Lavender & Z. Alfirevic (2019): Cervical cerclage, pessary, or vaginal progesterone in
high-risk pregnant women with short cervix: a randomized feasibility study, The Journal of Maternal-
Fetal & Neonatal Medicine, DOI: 10.1080/14767058.2019.1588245

To link to this article: https://doi.org/10.1080/14767058.2019.1588245

View supplementary material

Published online: 21 Mar 2019.

Submit your article to this journal

Article views: 10

View Crossmark data

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=ijmf20
THE JOURNAL OF MATERNAL-FETAL & NEONATAL MEDICINE
https://doi.org/10.1080/14767058.2019.1588245

ORIGINAL ARTICLE

Cervical cerclage, pessary, or vaginal progesterone in high-risk pregnant

women with short cervix: a randomized feasibility study
A. Carea , R. Jacksonb, E. O’Brienb, S. Leighc, C. Cornforthb, A. Haycoxc, M. Whitworthd, T. Lavendere
and Z. Alfirevica
a
Centre for Women and Children’s Health Research, Harris-Wellbeing Preterm Birth Research Group, University of Liverpool, Liverpool
Women’s Hospital, Liverpool, UK; bLiverpool Clinical Trials Unit, University of Liverpool, Liverpool, UK; cManagement School,
University of Liverpool, Liverpool, UK; dSaint Mary’s Hospital, Manchester, UK; eDivision of Nursing, Manchester, UK

ABSTRACT ARTICLE HISTORY

Objective: To assess feasibility for a definitive randomized controlled trial (RCT) comparing three Received 26 June 2018
treatments for short cervix in a population at high risk for spontaneous preterm birth (sPTB) Accepted 24 February 2019
over a 1-year period.
KEYWORDS
Design: Three arm, open label feasibility randomized clinical study.
Cerclage; pessary;
Methods: Women with singleton pregnancy with risk factors for sPTB (history of sPTB or prela- pregnancy; preterm birth;
bor premature rupture of membranes (PPROM) <34 weeks or significant cervical surgery), and progesterone; short cervix
short cervix on transvaginal ultrasound scan detected between 16þ0 and 24þ6 weeks gestation
were randomized to receive either cervical cerclage, vaginal pessary, or vaginal progesterone
200 mg nocte. Pregnancy outcomes and treatment costs were collected from hospital records,
NHS Reference costs, and British National Formulary costs.
Main outcome measures: Feasibility targets were defined as (i) at least 55% of eligible women
randomized; (ii) maximum 5% failure to adhere to the protocol per arm; (iii) maximum 5% loss
to short-term follow-up.
Results: Of 417 women screened between October 2015 and 2016, 25 (6%) were eligible for
trial inclusion, of whom 18 (72%) agreed to participate at the rate 0.75 participants/site/month.
Adherence to protocol was 100% in pessary and cerclage arms and 80% in vaginal progesterone
arm (95% CI 24–100%). No participants were lost to follow up. Cost of interventions accounted
for 6% (95% CI 2–10%) of overall health care expenditure.
Conclusions: A definitive clinical trial comparing treatments for prevention of sPTB in high-risk
women with short cervix is feasible but will be challenging due to small numbers of eligible
participants.

Introduction 35% of delay in trial completion is due to slow recruit-

A history of midtrimester loss [1], spontaneous pre-
term birth (sPTB) [1,2], prelabor premature rupture of
membranes (PPROM) [2], or excisional cervical treat-
ments [3,4] are currently considered triggers for cer-
vical length screening during pregnancy. The
detection of a short cervix by transvaginal ultrasound
scan in these high-risk women further increases the
risk of sPTB [5] and cerclage [6], vaginal pessary [7],
and vaginal progesterone [8] have all performed bet-
ter than placebo or expectant management in these
circumstances. It remains unclear which of these is the
most effective as only indirect comparisons have been
published so far [9].
Conducting a large multicenter randomized trial
poses significant organizational challenges. In the USA,
ment; nearly one-fifth of investigators do not enroll
any women and about one-third enroll only 5% of eli-
gible women [10]. Women may refuse randomization
based on personal preferences or reluctance to accept
one of the proposed interventions. Clinicians may
exhibit strong beliefs based on anecdotal evidence,
interpretation of current data, or a preference for a
less invasive treatment.
There has been considerable confusion in current
literature regarding the use of the terms “pilot” and
“feasibility” study relating to randomized clinical trials
(RCTs). The NIHR states that feasibility studies are
pieces of research done before a main study in order
to answer the question “Can this study be done?”
They are used to estimate important parameters that

CONTACT Angharad Care angharad.care@liv.ac.uk University Department, Liverpool Women’s Hospital, Crown Street, Liverpool L8 7SS, UK
Supplemental data for this article can be accessed here.
ß 2019 Informa UK Limited, trading as Taylor & Francis Group
2 A. CARE ET AL.
are needed to design the main study. If a feasibility
study is a small randomized controlled trial, it does
not need to have a power calculation. Instead, the
sample size is often used to estimate the critical
parameters (e.g. recruitment rate) to the necessary
degree of precision [11].
We describe the findings from a 12-month random-
ized feasibility study of cerclage, vaginal pessary, or
vaginal progesterone in high-risk women with an
asymptomatic singleton pregnancy and short cervix.
Materials and methods
This was a three-arm, open-label randomized feasibil-
ity clinical trial (EudraCT no. 2014-003112-36) of cerc-
lage, vaginal pessary or vaginal progesterone. Ethical
approval was awarded by NRES Committee North
West – Preston (REC Reference 14/NW/1392).
Recruitment lasted from 6 October 2015 to 19
October 2016 at two specialist sPTB prevention clinics
in large teaching hospitals in the United Kingdom.
Both clinics see approximately 150 women per annum
for serial transvaginal ultrasound measurement of cer-
vical length from 16 weeks of gestation.
Trial participants
Women were eligible for the study if they met the
following criteria:
i. Singleton pregnancy between 16 þ0
and 24 þ6
weeks of gestation
ii. history of previous spontaneous preterm birth
(sPTB) or preterm prelabor rupture of membranes
(PPROM) between 16þ0 and 33þ6 weeks of gesta-
tion or previous significant cervical surgery,
defined as two large loop excision of the trans-
formation zone (LLETZ) procedures, or a single
knife cone biopsy (KCB),
iii. cervical length below the third centile for gesta-
tional age [12],
iv. treating clinician in equipoise as to
best treatment,
v. 18 years or older.
Exclusions included: (i) known or suspected
chromosomal fetal abnormality, (ii) inability to give
informed consent, (iii) treatment with history indicated
cerclage, (iv) treatment with vaginal progesterone
within 2 weeks of randomization, (v) peanut allergy in
patient or partner (contraindication to use of proges-
terone preparation), (vi) known, past, or suspected
mammary or genital tract carcinoma, (vii) severe hep-
atic dysfunction, (viii) porphyria, (ix) thrombophlebitis,
(x) thromboembolic disorders, (xi) thrombophlebitis,
and (xii) cerebral hemorrhage.
Recruitment
Women attending clinic were screened for a short
cervix and recruited at the point of requiring treat-
ment. Standardization of cervical length scanning
was achieved using the technique described by the
Fetal Medicine Foundation for cervical length assess-
ment [13]. An information leaflet was given and writ-
ten informed consent was obtained from all
participants before enrollment. A high vaginal swab
was taken and treatment with clindamycin 2% cream
was given if positive for bacterial vaginosis. A cen-
tralized web-based service at the Liverpool Trials Unit
(LCTU), University of Liverpool was used to randomly
allocate women to their treatment arms on a 1:1:1
basis, using a pre-prepared randomization list with
randomly permuted blocks of three and six. The
intention was to initiate treatment within 72 h of
randomization.
Trial interventions
For women assigned to cerclage, the method of cerc-
lage placement (e.g. McDonald or Shirodkar), or suture
material (braided/monofilament) was not specified. An
experienced clinician performed the procedure with
whichever method of cerclage placement they were
most familiar.
Women assigned to pessary were fitted with a CE-
certified Arabin pessary (CE 0482/EN ISO 13485: 2003
annex III of the council Directive 93/42 EEC) immedi-
ately on the day of randomization without anesthetic.
The fitting was performed as described by its design-
ers [14]. Training was performed at site initiation visits
on how to insert and remove a pessary.
Pessary and cerclage were removed in the event of
confirmed PPROM. Otherwise, both cervical cerclage
and Arabin pessary were removed as an outpatient
procedure at 37þ0 weeks of gestation.
Women assigned to vaginal progesterone were pre-
scribed Uterogestan 200 mg pv which was collected
from the local site pharmacy after randomization. The
study medication was prescribed until 37þ0 weeks
gestation when the participant was advised to discon-
tinue the treatment.

Figure 1. RECAP study visits.
Participant assessment and follow up
A semi-structured interview was conducted 1-week
post randomization to evaluate participant experience
of the trial. A study visit to the clinic occurred 2 weeks
post randomization to review cervical length following
treatment. Any further visits to the PTB clinic were at
the discretion of the treating clinician (Figure 1). A
postnatal visit 6 weeks after birth was arranged as a
part of the qualitative study to evaluate women’s
experience. Once all the participants had completed
the study; focus groups were held to identify emerg-
ing themes. The results of this qualitative study will be
published separately.
To assess the feasibility of collecting health eco-
nomic data was assessed using EQ-5D, the most
widely used measure of patient-reported outcome in
THE JOURNAL OF MATERNAL-FETAL & NEONATAL MEDICINE 3
clinical trials, which a form was completed by research
staff at baseline and 6 weeks postnatally.
Treatment was discontinued before 37þ0 weeks if
pessary repeatedly displaced or fell out, in cases of
allergic reaction to vaginal progesterone, prelabor pre-
term rupture of the membranes (PPROM), participant
request or a clinician’s decision to stop treatment in
the best interest of the woman. Short-term neonatal
data were recorded after hospital discharge.
An international consortium of the PROspective
Meta-Analysis for Pessary Trials (PROMPT
Collaborative) agreed the minimal data collection
points used in the RECAP study to allow for an indi-
vidual patient data meta-analysis of data from
ongoing and planned clinical trials of pessary
(PROSPERO CRD42018067740) [15].
4 A. CARE ET AL.
Outcomes
In order to demonstrate the feasibility of a definitive
trial within the intended study population outcomes
were defined as follows:
i. recruitment rate, measured as the number of
women recruited in a 1-year period
ii. willingness of the women to be randomized—
measured as the rate of screening failures in
each hospital
iii. adherence to protocol by doctor and the
woman—measured as number of protocol devi-
ations as defined by the trial monitoring plan
iv. preliminary piloting of appropriate measures of
outcome and procedures for definitive trial
v. participant and clinician acceptability of treat-
ment—measured by the number of patient
withdrawals or changes of treatment
vi. safety outcomes—incidence of AEs, PTB,
and PPROM
vii. participant satisfaction with proposed current
trial model—qualitative methods
viii. expected relative resource impact of using cer-
vical cerclage, Arabin pessary, or vaginal proges-
terone for the prevention of PTB—nested health
economic study
ix. clinical outcomes—we collected data in accord-
ance with the international clinical consensus on
minimal data collection for pessary trials as out-
lined by international consortium PROspective
meta-analysis for pessary trials (PROMPT
Collaborative) and to reflect a primary outcome
of birth less than 34 weeks and markers of neo-
natal morbidity and mortality
Our a priori criteria for feasibility were:
 A minimum of 85% of eligible women to be
approached by healthcare team
 A minimum of 55% of eligible women to choose to
be randomized to a treatment
 No more than 5% failure of adherence to protocol
per arm
 No more than 5% loss to short term follow up.
Statistical analysis
A sample size for RECAP was not set as recruitment
rate was a main outcome measure. Therefore, data
from number of participants obtained over 12 months
was described per treatment arm and presented as
mean monthly rates with 95% confidence internal (CI).
For demographics, continuous outcomes are pre-
sented as median (IQR), categorical outcomes are pre-
sented as frequencies of counts with associated
percentages.
Health economics
A standard cost-effectiveness approach was applied,
whereby both the total costs per completed preg-
nancy episode, and the clinical efficacy of the treat-
ments, measured in terms of the percentage success
in avoiding PTB for each treatment, were jointly con-
sidered. The primary economic outcome was the
incremental cost–effectiveness ratio (ICER), character-
ized as “the cost per PTB avoided.” Unit costs for all
services utilized by patients, and subsequently
accounted for in the health economic analysis, were
obtained from both NHS reference costs 2016, and the
British National Formulary (BNF), and are provided in
Table S1.
Results
Recruitment rate
Over a 12-month recruitment period, 417 pregnant
women were screened of whom 25 (6%) were eligible
and 18 were randomized (Figure 2). The main reason
for the reduction in number from screening to enroll-
ment was that most women did not develop a short
cervix by 24 weeks. One woman was recruited for
every 23 women screened with average recruitment
rate of 0.75 participants/site/month.
Willingness to be randomized
Seven eligible women declined participation; one
woman expressed a definite preference for cerclage,
two for vaginal pessary, two women refused to accept
a vaginal pessary as treatment option, one did not
accept vaginal progesterone, and one did not give a
reason for declining. Women who declined randomiza-
tion were treated with first-line treatments at their
respective units.
One woman withdrew consent immediately the fol-
lowing randomization to cerclage as she did not want
surgical intervention and had hoped to be allocated
to one of the other treatments.
Adherence to the protocol
Two major protocol deviations occurred; recruitment
to the study at 15þ6 weeks and one cervical cerclage
 THE JOURNAL OF MATERNAL-FETAL & NEONATAL MEDICINE 5

Assessed for eligibility (n=417)

Enrollment

Excluded (n=399)
♦ Not meeting inclusion criteria (n=392)
♦ Declined to participate (n=7)
Reasons for declining:
Single treatment arm preference (n=3)
- Cerclage (n=1)
- Pessary (n=2)
Single treatment arm refusal (n=3)
- Pessary (n=2)
- Progesterone (n=1)
Randomized (n=18)

Allocaon
Allocated to Pessary (n=6)
♦ Received allocated intervention (n=6)
♦ Did not receive allocated intervention
(give reasons) (n=0)
Allocated to Cerclage (n=7)
♦ Received allocated intervention (n=6)
♦ Did not receive allocated intervention
(withdrew consent once randomised
– did not want cerclage) (n=1)
Allocated to Vaginal Progesterone (n=5)
♦ Received allocated intervention (n=5)
♦ Did not receive allocated intervention
(give reasons) (n=0)

Follow-Up
Lost to follow-up (n=0)
Discontinued intervention (n=1)
- Active preterm labour 20+5 (n=1)
Lost to follow-up (n= 0)
Discontinued intervention (n=2)
- Induction of labour for IUGR
36+4 (n=1)
- Threatened preterm labour 35+4
(n=1)
Lost to follow-up (n= 0)
Discontinued intervention (give reasons)
(n=2)
- Vaginal discomfort 34+1 (n=1)
- Clinician decision 18+4 –
cerclage inserted for presumed
failure of progesterone (n=1)

Analysis
Analysed (n=6) Analysed (n=6) Analysed (n=5)
♦ Excluded from analysis (n=0) ♦ Excluded from analysis (n=1; withdrew ♦ Excluded from analysis (n=0)
consent)

Figure 2. RECAP CONSORT Flow diagram.

was sited outside the initial 72-h window. Twelve par-
ticipants were associated with minor protocol devia-
tions including vaginal swabs not taken (n ¼ 3) and
postnatal EQ5D not completed/received (n ¼ 9).
Piloting of data collection
There were no feasibility issues related to data collec-
tion. The baseline demographics, maternal complica-
tions, and labor information of participants are shown
by allocated treatment arm (Table 1). EQ5D as a health
economics assessment tool was considered not feas-
ible for postnatal data collection as >50% were
not completed.
Acceptability of treatment
Five women did not end treatment as per protocol.
One woman voluntarily discontinued progesterone at
34þ1 weeks due to vaginal discomfort and four were
physician decisions to discontinue treatment. A cerc-
lage was removed at 20þ5 weeks due to active pre-
term labor (PTL), a pessary was removed at 36þ4
weeks for an induction of labor for IUGR, and one pes-
sary was removed between 35þ4 weeks for threatened
PTL. Only one woman had her treatment changed
from progesterone to cerclage after cervical shorten-
ing and presumed treatment failure of progesterone
at 18þ4 weeks.
6 A. CARE ET AL.

Table 1. Baseline demographics, maternal complications, and outcomes reported per randomized arm.
Variable Arabin pessary Cervical cerclage Vaginal progesterone
No. participants 6 7 5
Withdrawal 0 1 0
Baseline demographics
Previous preterm birth/PPROM 4 6a 4
Previous cervical surgery 2 2a 1
Gestation at recruitment, weeksa 20.6 (20.3, 23.1) 21.7 (19.6, 22.3) 20.6 (18.1, 22.0)
Cervical length at recruitmenta 17 (13, 24) 21 (19.5, 21.5) 21 (20, 21)
Agea 31.5 (28.8, 35.8) 30 (28.5, 30.5) 29 (27, 30)
BMIa 27.35 (22.65, 30.925) 26.5 (22, 34.1) 26.3 (25.8, 30.1)
Current smoker 2 (25%) 3 (38%) 3 (38%)
Maternal complications
Genital tract infection 0 (0%) 1 (17%) 0 (0%)
Presumed chorioamnionitis 0 (0%) 0 (0%) 1 (20%)
Urinary tract infection 0 (0%) 1 (17%) 1 (20%)
Attendance to emergency room 6 (100%) 5 (83%) 4 (80%)
Admission to hospital 3 (50%) 1 (17%) 2 (40%)
Tocolytics given 1 (17%) 1 (17%) 1 (20%)
Magnesium sulfate for neuroprotection 0 (0%) 0 (0%) 2 (40%)
Antenatal steroid given 3 (50%) 1 (17%) 3 (60%)
PPROM 3 (50%) 1 (17%) 3 (60%)
Antibiotic required 0 (0%) 2 (33%) 1 (20%)
Gestation at treatment discontinuation
34 weeks 1 1 1
34–37þ0 4 1 3
37þ0 1 4 1
Delivery outcomes
Time between randomization and birth, weeks 16.2 (15.9, 16.4) 15.5 (15.2, 16.7) 14.4 (13.2, 15.6)
Onset of labor, spontaneous 3 (50%) 3 (50%) 2 (40%)
IOL, PPROM 0 0 1 (20%)
Gestation at birth, weeksa 38.5 (36.8, 39.9) 37.9 (37.2, 38.8) 35.8 (35, 36.5)
<24 weeks 0 1 (17%) 0
24þ0–28þ0 weeks 0 0 1 (20%)
34þ0–36þ6 weeks 2 (33%) 0 3 (60%)
37 weeks 4 (67%) 5 (50%) 1 (20%)
Livebirth 6 (100%) 6 (100%) 4 (80%)
Birthweight, gb 3155 (2350, 3833) 2665 (2413, 3120) 2800 (2660, 2995)
Neonatal complications
Neonatal death 0 1 (17%) 0
Fetal anomaly 0 0 0
Early sepsis 0 0 0
a
Median (IQR).
b
One participant had both previous preterm birth and cervical surgery.

Safety outcomes Health economic impact

Table 1 lists maternal complications by treatment arm.
One neonatal death attributed to extreme prematurity
occurred to a participant allocated to the cervical cerc-
lage arm. One term intrauterine death (IUD) occurred
at 37þ4 weeks due to placental abruption in the vagi-
nal progesterone arm and was unrelated to PTB pre-
vention treatment.
Participant satisfaction
Overall women were happy to recommend their treat-
ment to others if they achieved a healthy delivery.
Commonly reported problems associated with both
pessary and cerclage were increased discharge and
vaginal pain or discomfort. Interestingly, participant
confidence in pessary and cerclage was higher than
with progesterone and described as a “physical” treat-
ment “keeping the baby in place.”
The mean total postrandomization cost for mothers
requiring treatment for the prevention of PTB was
£8963 (95% CI £663–£17,262). Following the removal
of a single outlier observation in which 88 days of
neonatal specialist care were required (49 critical care
39 high dependency unit), this fell to £5146 (95% CI
£3164–£7128). Most of the healthcare expenditure was
associated with inpatient admission following delivery
at £2933 (95% CI £1794–£4071), with women spend-
ing a mean (median) 4 days (2 days) in hospital post-
partum, accounting for 58.8% (95% CI 46.1–71.5%) of
overall healthcare expenditure. Neonatal costs, includ-
ing time in the high dependency and critical care
units, accounted for approximately 13.9% (95% CI
3.6–24.2%) of healthcare expenditure, at an average
£4004 (95% CI £0–£11,397) per patient. Unscheduled
hospital admissions accounted for 9.4% of total
expenditure (95% CI 5.9–12.9%) at an average cost of
 THE JOURNAL OF MATERNAL-FETAL & NEONATAL MEDICINE 7

Table 2. Estimated number of participants based on recruitment rates from RECAP feasibility study.
Overall Expected reduction Sample size No. centers No. centers
estimated rate with superior treatment (alpha 5% þ power recruiting recruiting for
Core outcome with treatment (25% risk reduction) 90% – NIHR standard) for 2 years eacha 4 years eacha
Gestational age
PTB < 37 w 29% 22% 2000 112 56
PTB < 34 w 14% 10.5% 4500 250 125
Child
Neonatal mortality 1.2% 0.9% 59000 3278 1639
Composite of mortality 8% 6% 8300 462 231
and morbidity
Estimated rates produced from Cochrane database figures from published clinical trials of cerclage, vaginal progesterone, and Arabin pessary.
a
Recruitment period does not include time needed for follow-up, analysis, and write-up.

£1198 (95% CI £0–£2457) per patient. Additionally,
participants experienced a mean (median) 2.8 (2)
unscheduled antenatal appointments, at an average
cost of £514 (95% CI £249–£779) per patient.
Cervical cerclage participants experienced signifi-
cantly higher “upfront” costs, at £549 (cerclage), com-
pared to £58.25 (progesterone), and £40.00 (Arabin
pessary), p ¼ .0023. However, combined treatment
costs associated with the treatments for PTB under
consideration in this study (cervical cerclage, Arabin
pessary, and progesterone), on average, accounted for
just 6.2% (95% CI 2.3–10.3%) of overall healthcare
expenditure.
feasibility data are used to quantify the event rates to
determine if proceeding to a full RCT is possible and
practical [16]. In particular, actual recruitment figures
lack precision due to wide confidence intervals. The
best example is the evaluation of a priori criterion; “No
more than 5% failure of adherence to protocol per
arm”. Despite full adherence in two arms, one partici-
pant discontinued treatment at 34 weeks due to vagi-
nal soreness in the progesterone arm. With only five
women recruited to this arm, the confidence intervals
around this percentage are too wide which led the
trial oversight committee to conclude that this could
not be a reason to determine a full RCT unfeasible.

Clinical outcomes Interpretation

The study was not powered to detect clinically import-
ant differences in observed core outcomes; therefore,
we have not performed any further statistical analysis;
data of interest are shown in Table 1.
Discussion
Main findings
This feasibility study demonstrates that a definitive
clinical trial, although feasible, would be challenging
in terms of recruitment rate. Our two centers have
dedicated regional sPTB prevention clinics with more
than 150 new high-risk referrals and yet the number
of eligible participants was low. The majority of eli-
gible women agreed to be randomized and once
randomized, protocol violations were rare. Actual
treatment costs account for no more than 10% of
overall expenditure and are, therefore, not a major
consideration in future studies.
Strengths and limitations
Our study was planned and organized as if this was a
definitive clinical trial with a priori targets to assess
feasibility. However, limitations are unavoidable when
For any future “definitive” clinical trial, the selection of
primary outcome(s) will be a key consideration in cal-
culating appropriate sample size and the number of
study sites required to recruit. Recently published core
set of outcomes for PTB prevention includes nine out-
comes related to offspring and four to mother. In
order to illustrate the size of the challenge to provide
robust estimates of treatment effects in a randomized
trial, we have provided the power calculation for sev-
eral core outcomes and calculated how many centers
that would need to participate to recruit these num-
bers within 2 and 4 years periods (Table 2). We have
assumed that all participating centers are open for
recruitment at the beginning of 2-year period, which
is rarely, if ever, the case. Our recruitment estimates
are based on our two sites chosen because both have
a dedicated spontaneous PTB clinic and approximately
8500 births per annum. Every year, both preterm labor
clinics screen minimum of 150 high-risk women, there-
fore, our recruitment estimates would need to be
adjusted downward for smaller units.
One way to improve the feasibility of a study of
this nature would be to identify a higher risk popula-
tion, which we do not think is possible at present, or
one with more women with a short cervix. To do this
8 A. CARE ET AL.
would require universal screening policies which have
been proposed [17] but not yet recommended by
high-quality guidelines [18]. Even if ultrasound screen-
ing for short cervix becomes universally accepted, the
number needed to screen will have to be estimated
conservatively because recent data show that the
prevalence of short cervix in low-risk populations is
likely to be much lower than anticipated [19].
Our economic analysis has shown that examining
the cost-effectiveness of treatments for PTB on a
larger scale is feasible but probably unnecessary. The
cost of treatment is proportionally very small (<10%)
compared to the cost of the rest of the care for moth-
ers and babies. Therefore, it is very unlikely that cost-
effectiveness will be an important consideration. The
effectiveness in reducing the risk of core outcomes
and patient preferences will remain the key drivers for
clinical decision-making.
Conclusion
This study demonstrates a definitive multicenter clin-
ical trial comparing all three current treatments for
prevention of sPTB in women with risk factors for pre-
term birth and a short cervix may be possible but will
require significant investment of time and resources.
As the incidence of eligible women in any future trial
is likely to be low, a large number of specialist centers
will be needed recruiting for several years with an
expected recruitment rate of approximately 1 recruit
for every 23 women screened and approximately two-
thirds of eligible women wiling to participate.
Acknowledgments
We thank all the women who participated in this study at
such a vulnerable time in their pregnancy. We thank the
members of the Trial Steering Committee (Professor Anna
David – Chair, Mrs. Jenny Ross, Dr. Rachel McFarland, Dr.
Richard Jackson, Dr. Sarah Seaton, and Dr. Sam Oddie) and
Independent Safety and Data Monitoring Committee (Dr.
Nigel Simpson – Chair, Dr. Adrian Hughes and Dr. Mark
Pilling). We are very grateful to Besins Healthcare for provid-
ing the study supply of vaginal progesterone, and
HOLOGICV for providing quantitative fetal fibronectin. We
R
acknowledge the persons acting on behalf of the study
sponsors (Louise Hardman, Alex Astor, and Karen Wilding).
We would also like to thank the members of the patient and
public involvement group at the Harris–Wellbeing Preterm
Birth Centre in Liverpool who contributed to final trial
design and patient information leaflet design and content.
Finally, we would like to acknowledge the hard work of the
research midwives who assisted in recruitment and data col-
lection throughout the study.
Disclosure statement
No potential conflict of interest was reported by the authors.
Funding
This report is independent research funded by the National
Institute for Health Research (Research for Patient Benefit
Programme, three-arm randomized trial of Arabin pessary,
cervical cerclage, and progesterone to prevent spontaneous
preterm birth in an asymptomatic high-risk cohort: a feasibil-
ity study, PB-PG-0213-30106). The views expressed in this
publication are those of the authors and not necessarily
those of the NHS, the National Institute for Health Research
or the Department of Health. This work was produced by Dr.
Angharad Care/University of Liverpool under the terms of a
commissioning contract issued by the Secretary of State
for Health.
ORCID
A. Care http://orcid.org/0000-0003-2131-0406
T. Lavender http://orcid.org/0000-0003-1473-4956
References
[1] Esplin MS, O’Brien E, Fraser A, et al. Estimating recur-
rence of spontaneous preterm delivery. Obstet
Gynecol. 2008;112(3):516–523.
[2] Goldenberg RL, Culhane JF, Iams JD, et al.
Epidemiology and causes of preterm birth. Lancet.
2008;371(9606):75–84.
[3] Ortoft G, Henriksen T, Hansen E, et al. After conisation
of the cervix, the perinatal mortality as a result of
preterm delivery increases in subsequent pregnancy.
BJOG. 2010;117(3):258–267.
[4] Poon LC, Savvas M, Zamblera D, et al. Large loop
excision of transformation zone and cervical length in
the prediction of spontaneous preterm delivery.
BJOG. 2012;119(6):692–698.
[5] To MS, Skentou CA, Royston P, et al. Prediction of
patient-specific risk of early preterm delivery using
maternal history and sonographic measurement of
cervical length: a population-based prospective study.
Ultrasound Obstet Gynecol. 2006;27(4):362–367.
[6] Berghella V, Rafael TJ, Szychowski JM, et al. Cerclage
for short cervix on ultrasonography in women with
singleton gestations and previous preterm birth: a
meta-analysis. Obstet Gynecol. 2011;117(3):663–671.
[7] Goya M, Pratcorona L, Merced C, et al. Cervical pes-
sary in pregnant women with a short cervix (PECEP):
an open-label randomised controlled trial. Lancet.
2012;379(9828):1800–1806.
[8] Romero R, Nicolaides KH, Conde-Agudelo A, et al.
Vaginal progesterone decreases preterm birth
34 weeks of gestation in women with a singleton
pregnancy and a short cervix: an updated meta-ana-
lysis including data from the OPPTIMUM study.
Ultrasound Obstet Gynecol. 2016;48(3):308–317.

[9] Alfirevic Z, Owen J, Carreras Moratonas E, et al.
Vaginal progesterone, cerclage or cervical pessary for
preventing preterm birth in asymptomatic singleton
pregnant women with a history of preterm birth and
a sonographic short cervix. Ultrasound Obstet
Gynecol. 2013;41(2):146–151.
[10] Getz K. Swift predominance of ex-U.S. sites. Appl Clin
Trials. 2005;14(12):21–22.
[11] National Institute of Health Research. Available
from: https://www.nihr.ac.uk/funding-and-support/
documents/funding-for-research-studies/research-pro-
grammes/PGfAR/CCF-PGfAR-Feasibility-and-Pilot-stud-
ies.pdf [accessed 2018 Mar 21].
[12] Salomon LJ, Diaz-Garcia C, Bernard JP, et al.
Reference range for cervical length throughout preg-
nancy: non-parametric LMS-based model applied to
a large sample. Ultrasound Obstet Gynecol. 2009;
33(4):459–464.
[13] Fetal Medicine Foundation. Cervical assessment.
Available from: https://www.fetalmedicine.org/educa-
tion/cervical-assessment [accessed 2018 Mar 21].
[14] Arabin B, Halbesma JR, Vork F, et al. Is treatment with
vaginal pessaries an option in patients with a
THE JOURNAL OF MATERNAL-FETAL & NEONATAL MEDICINE 9
sonographically detected short cervix? J Perinat Med.
2003;31(2):122–133.
[15] National Institute of Health Research. PROSPERO inter-
national prospective register of systematic reviews
available from: https://www.crd.york.ac.uk/prospero/
display_record.php?RecordID¼67740 [accessed 2018
Mar 21].
[16] Collinson M, Anwar S, Graham L, et al. Designing and
analysing feasibility studies of complex interventions:
challenges related to assessing stop/go criteria. Trials.
2013;14(Suppl 1):O20.
[17] Souka AP, Papastefanou I, Pilalis A, et al.
Implementation of universal screening for preterm
delivery by mid-trimester cervical length measure-
ment. Ultrasound Obstet Gynecol. 2019;53:396–401.
[18] Medley N, Poljak B, Mammarella S, et al. Clinical guide-
lines for prevention and management of preterm birth:
a systematic review. BJOG. 2018;125(11):1361–1369.
[19] Peixoto AB, da Cunha Caldas TMR, Tahan LA, et al.
Second trimester cervical length measurement for
prediction spontaneous preterm birth in an unse-
lected risk population. Obstet Gynecol Sci. 2017;60(4):
329–335.