Key Points
Question
Does placement of a cervical pessary reduce the rate of preterm birth or fetal death before 37 weeks in nonlaboring singleton pregnancies with a cervical length of 20 mm or less?
Findings
In this randomized trial of 544 singleton pregnancies, the rate of preterm birth or fetal death before 37 weeks did not differ between those randomized to pessary (45.5%) and those randomized to usual care (45.6%) (relative risk, 1.00; 95% CI, 0.83-1.20). Fetal or neonatal/infant death occurred more frequently in those randomized to receive a pessary (13.3%) than those randomized to receive usual care (6.8%) (relative risk, 1.94; 95% CI, 1.13-3.32).
Meaning
Cervical pessary does not benefit singleton pregnancies with a cervical length of 20 mm or less.
Abstract
Importance
A short cervix as assessed by transvaginal ultrasound is an established risk factor for preterm birth. Study findings for a cervical pessary to prevent preterm delivery in singleton pregnancies with transvaginal ultrasound evidence of a short cervix have been conflicting.
Objective
To determine if cervical pessary placement decreases the risk of preterm birth or fetal death prior to 37 weeks among individuals with a short cervix.
Design, Setting, and Participants
We performed a multicenter, randomized, unmasked trial comparing a cervical pessary vs usual care from February 2017 through November 5, 2021, at 12 centers in the US. Study participants were nonlaboring individuals with a singleton pregnancy and a transvaginal ultrasound cervical length of 20 mm or less at gestations of 16 weeks 0 days through 23 weeks 6 days. Individuals with a prior spontaneous preterm birth were excluded.
Interventions
Participants were randomized 1:1 to receive either a cervical pessary placed by a trained clinician (n = 280) or usual care (n = 264). Use of vaginal progesterone was at the discretion of treating clinicians.
Main Outcome and Measures
The primary outcome was delivery or fetal death prior to 37 weeks.
Results
A total of 544 participants (64%) of a planned sample size of 850 were enrolled in the study (mean age, 29.5 years [SD, 6 years]). Following the third interim analysis, study recruitment was stopped due to concern for fetal or neonatal/infant death as well as for futility. Baseline characteristics were balanced between participants randomized to pessary and those randomized to usual care; 98.9% received vaginal progesterone. In an as-randomized analysis, the primary outcome occurred in 127 participants (45.5%) randomized to pessary and 127 (45.6%) randomized to usual care (relative risk, 1.00; 95% CI, 0.83-1.20). Fetal or neonatal/infant death occurred in 13.3% of those randomized to receive a pessary and in 6.8% of those randomized to receive usual care (relative risk, 1.94; 95% CI, 1.13-3.32).
Conclusions and Relevance
Cervical pessary in nonlaboring individuals with a singleton gestation and with a cervical length of 20 mm or less did not decrease the risk of preterm birth and was associated with a higher rate of fetal or neonatal/infant mortality.
Trial Registration
ClinicalTrials.gov Identifier: NCT02901626
This randomized clinical trial assesses the effect of cervical pessary placement on preterm birth and fetal mortality among pregnant individuals with a cervical length of 20 mm or less.
Introduction
Preterm birth remains the most frequent cause of perinatal mortality,1,2 and children born prematurely are at greater risk of chronic medical conditions3,4 and developmental delays.5 These risks are inversely proportional to gestational age at birth.6 Improvements in survival have largely been attributed to the use of antenatal corticosteroids7 and improved neonatal care.8 Shortened cervical length (<25 mm) as measured by transvaginal sonography in the second trimester has been shown to correlate with the risk of preterm delivery9 and occurs in approximately 1.0% to 2.5% of the population.10,11 Randomized clinical trials have demonstrated that intravaginal progesterone decreases the risk of preterm birth among pregnancies with a shorter cervical length of 20 mm or less12,13 and that cerclage uniquely benefits individuals with a short cervix who also have a history of preterm birth.12 Nevertheless, these treatments have been shown to result in a decrease in preterm birth before 34 weeks of only approximately 45%13,14 and no differences in preterm birth before 37 weeks. Therefore, additional treatments are needed to further reduce the rates of preterm birth in this at-risk population.
Cervical pessary placement is one treatment that has been evaluated to prevent preterm birth in those with a singleton pregnancy and a short cervical length. Randomized clinical trials have differed on the effect of pessary placement, with some studies15,16 showing lower rates of preterm birth before 34 and 37 weeks and other studies not demonstrating benefit.17,18,19 Reasons for these disparate outcomes have been hypothesized to be the result of varying use of progesterone,15 inclusion of individuals with a prior preterm birth,15,17 and lack of adequate training in pessary placement.
The randomized Trial of Pessary in Singleton Pregnancies With a Short Cervix (TOPS) was performed to test the hypothesis that nonlaboring individuals carrying a singleton gestation, with no prior preterm birth history and a cervical length of 20 mm or less (measured by transvaginal ultrasound) prior to 24 weeks of gestation, would have a lower rate of preterm birth or fetal death before 37 weeks of gestation with placement of a cervical pessary.
Methods
Study Design and Participants
We conducted a multicenter, open-label, randomized clinical trial comparing a pessary (Dr Arabin GmBH & Co) vs usual care at 12 clinical centers participating in the Eunice Kennedy Shriver National Institute of Child Health and Human Development’s Maternal-Fetal Medicine Units Network from February 2017 through November 5, 2021. The protocol (Supplement 1) was approved by each individual center’s institutional review board prior to participant enrollment. An independent data and safety monitoring board was convened prior to study initiation, approved the protocol, and determined a schedule of ongoing data oversight. The trial was performed under an FDA Investigational Device Exemption (IDE No. G160176). The statistical analysis plan is available in Supplement 2.
Individuals with nonlaboring, singleton pregnancies who had a transvaginal cervical length of 20 mm or less at gestations of 16 weeks 0 days through 23 weeks 6 days were identified at each individual hospital per the local protocol for assessment of cervical length.
Individuals were not eligible for the study if they had a history of spontaneous preterm birth; a cervix that was dilated 3 cm or greater on digital examination; prolapse of the membranes beyond the cervical os; preterm rupture of membranes or evidence of preterm labor (>6 contractions per hour); a major fetal or müllerian anomaly; a planned or placed cerclage; or another medical condition that increased the risk of medically indicated preterm birth (eTable 1 in Supplement 3). Race and ethnicity were self-reported by participants.
Randomization and Masking
Individuals meeting eligibility requirements and accepting trial participation were randomized in a 1:1 fashion to either pessary or usual care using a secure web-based system developed by the data coordinating center (Figure 1). Randomization sequences were generated by the data coordinating center using the simple-urn method and stratified by site.20 Research staff had no role in determining the randomization assignment.
Figure 1. Participant Flow in a Study of Cervical Pessary to Reduce Preterm Birth or Fetal Death in Singleton Pregnancy.
aRandomization was stratified by site.
Interventions
All research staff, sonographers, and pessary placers were trained and certified for this study. The certification process for sonographers included both demonstration of knowledge of cervical length assessment and demonstration of appropriate technique in accordance with Fetal Medicine Foundation or Perinatal Quality Foundation standards. Pessary placement was required to be performed by physicians who were certified following standardized education and demonstration of competence by passing a standardized test.
Following individual written informed consent, eligibility confirmation via speculum pelvic examination was performed, and vaginal swabs for bacterial analysis and cervicovaginal fluid collection were obtained.
Participants randomized to receive a pessary were positioned in lithotomy position. The pessary was placed on the day of randomization around the cervix and then flexed posteriorly with digital pressure to alter the cervicouterine angle.15 Pessary placers were allowed to choose from 3 different sizes of pessary (nulliparous: internal diameter, 32 mm, external diameter, 65 mm; multiparous: internal diameter, 32 mm, external diameter, 70 mm; and alternative: internal diameter, 35 mm, external diameter 70 mm) based on participant obstetric history, anatomy, and tolerance and the ability of the pessary, once placed, to maintain the correct position.
All participants were to be contacted within 1 week of randomization to assess adherence and adverse effects of the pessary. Participants were asked to return every 4 weeks for a study visit, which initially was planned to occur in person but, after the onset of the COVID-19 pandemic, could be conducted virtually. An additional set of vaginal samples and a pessary check were scheduled between 24 weeks and 32 weeks of gestation. In the event the pessary was dislodged or removed, it was to be replaced by a trained placer. Pessary position at the time of planned removal at 36 weeks and 6 days of gestation was noted. Medically indicated removal of the pessary prior to study completion was required in the setting of active labor, preterm premature rupture of membranes, clinical chorioamnionitis, active vaginal bleeding, or cervical edema or laceration. The pessary could also be removed for other reasons at the discretion of treating obstetric clinicians or due to participant request. The remainder of obstetrical care, including the use of progesterone and examination-indicated cerclage (ie, cerclage placed based on physical examination changes following randomization), was at the discretion of treating clinicians. Following delivery, maternal, fetal, and neonatal outcomes were abstracted from medical records by certified research staff members.
Outcomes
The primary outcome was delivery or fetal death before 37 weeks 0 days of gestation. Prespecified secondary maternal outcomes were preterm birth or fetal death before 35 weeks, 32 weeks, and 28 weeks of gestation; interval from randomization to delivery or fetal death; gestational age at delivery; preterm birth (before 37 weeks 0 days of gestation), premature rupture of membranes; vaginal infection as assessed by clinical diagnosis; maternal antibiotic or antifungal prescription for vaginal infection; adverse effects; and discontinuation of the pessary and reasons for its removal. Prespecified secondary neonatal/fetal outcomes included fetal or neonatal (birth to 28 days of life)/infant (>28 days of life to 1 year) death, duration of ventilator support, duration of supplemental oxygen, seizures requiring treatment, and small-for-gestational-age birth (less than the fifth percentile for gestational age per the methods of Alexander et al; written communication; Greg R. Alexander, RS, MPH, ScD; 2000).21,22 The composite neonatal outcome included the occurrence of any of the following: fetal death, neonatal/infant death, respiratory distress syndrome, grade III/IV intraventricular hemorrhage, periventricular leukomalacia, stage 2 or 3 necrotizing enterocolitis, bronchopulmonary dysplasia, stage 3 or higher retinopathy of prematurity, or proven early sepsis. A complete list of prespecified outcomes is shown in eTable 2 in Supplement 3.
Sample Size Calculation
Based on available literature at the time of protocol development, we assumed the rate of preterm birth before 37 weeks or fetal death in those randomized to usual care was 28% and that a sample size of 850 would have at least 90% power to detect a 33% reduction in preterm birth with a 2-sided type I error of .05. We also determined that this sample size would yield more than 80% power to detect a 30% reduction in preterm birth or fetal death prior to 37 weeks.
Statistical Analysis
The analyses were performed using a randomized approach. Participants randomized to the pessary group who had a pessary placed comprised the as-treated population. We compared categorical baseline variables using χ2 and Fisher exact tests. We compared continuous variables using Wilcoxon rank sum tests. Unadjusted log-binomial models were used to assess treatment effect for the primary outcome of delivery or fetal death prior to 37 weeks. If the treatment groups were found to differ on a pretreatment factor known to be a risk factor for the outcome, the statistical analysis adjusted for these differences. Per protocol, prespecified subgroup analyses were conducted only if there was a statistical difference in the primary outcome between the 2 treatment groups.
Categorical outcomes were evaluated using relative risks and 95% confidence intervals. Quantile regression was used to evaluate differences in medians for continuous outcomes. We did not perform sensitivity analyses or imputation for missing data. A 2-tailed P < .05 was considered to indicate statistical significance. Statistical analyses were performed using SAS version 9.4 (SAS Institute Inc).
The timing of interim analyses was at the discretion of the data and safety monitoring board and not prespecified in the protocol. Three interim analyses were performed for the primary outcome. For the interim analyses, we used a group sequential method to control the type I error, with the Lan-Demets characterization of the O’Brien-Fleming boundary providing the stopping rule for benefit.23 The data and safety monitoring board also used conditional power analysis as a stopping guideline for futility. If the conditional power was low given the observed data and assuming the alternative hypothesis for the remainder of the trial, termination for futility could be considered. Although the protocol did not specify stopping rules for safety, the data and safety monitoring board reviewed maternal and neonatal safety outcomes and considered them in their deliberations and recommendations for trial continuation or termination.
After adjusting the planned type I error rate of .05 for interim analyses, the 2-tailed α value indicating statistical significance for the primary outcome was .04756. We report 95% confidence intervals throughout because the type I error adjustment for interim analyses was minimal. This trial is part of the PROMPT collaborative, an international prospective individual patient data meta-analysis of pessary trials (PROSPERO 2018 CRD42018067740).
Results
Trial Population
From February 2017 through November 5, 2021, 2105 individuals with a cervical length of 20 mm or less were identified by referral or institutional protocol for cervical length screening and assessed for eligibility. A total of 1019 were considered eligible and 544 provided consent and were randomized (Figure 1; eTable 6 in Supplement 3). At the third interim analysis, which included complete data for 511 participants (60.1% of the planned sample size), the data and safety monitoring board recommended termination of the trial due to a safety concern of a higher frequency of fetal and neonatal/infant death in the pessary group as well as futility. Among the 280 participants randomized to receive a pessary, 4 were not able to tolerate the device at the time of the initial pessary placement. A total of 264 participants were randomized to usual care. One participant in each study group was lost to follow-up (Figure 1).
The characteristics of the participants were balanced between the 2 groups (Table 1). Vaginal progesterone use was prescribed for 98.9% in each group. The proportions of individuals treated for vaginal infection prior to randomization did not differ between groups.
Table 1. Baseline Participant Characteristics.
| Characteristics | Pessary (n = 280) | Usual care (n = 264) |
|---|---|---|
| Maternal age, mean (SD), y | 29.9 (6.0) | 29.1 (5.9) |
| Race and ethnicity, No. (%)a | ||
| American Indian or Alaska Native | 1 (0.4) | 0 |
| Asian | 12 (4.3) | 16 (6.1) |
| Non-Hispanic Black | 132 (47.1) | 127 (48.1) |
| Hispanic | 62 (22.1) | 56 (21.2) |
| Native Hawaiian or Pacific Islander | 0 | 1 (0.4) |
| Non-Hispanic White | 61 (21.8) | 53 (20.1) |
| More than 1 race | 2 (1.0) | 2 (1.0) |
| Not reported or unknown | 10 (3.6) | 9 (3.4) |
| Body mass index at first clinic visit, median (IQR)b | 28.8 (24.3-34.2) (n = 276) | 28.6 (23.2-33.4) (n = 263) |
| <30 | 153 (55.4) | 162 (61.6) |
| 30.0-34.9 | 61 (22.1) | 49 (18.6) |
| 35.0-39.9 | 29 (10.5) | 30 (11.4) |
| ≥40.0 | 33 (12.0) | 22 (8.4) |
| Cigarette use during pregnancy, No. (%) | 30 (10.7) | 21 (8.0) |
| Married or living with partner, No. (%) | 167 (59.6) | 168 (63.6) |
| Employed full- or part-time, No. (%) | 191 (68.2) | 172 (65.2) |
| More than 12 y of education, No./total (%) | 176/280 (62.9) | 155/263 (58.9) |
| Insurance, No. (%) | ||
| Private | 145 (51.8) | 137 (51.9) |
| Government | 127 (45.4) | 121 (45.8) |
| Self-pay | 8 (2.9) | 6 (2.3) |
| Prior cervical surgery, No. (%)c | 33 (11.8) | 31 (11.7) |
| Previous pregnancy regardless of outcome, No. (%) | 103 (36.8) | 87 (33.0) |
| Total prior term deliveries, median (IQR) | 0 (0-1) | 0 (0-1) |
| Type of pregnancy, No. (%) | ||
| Spontaneous | 257 (91.8) | 245 (92.8) |
| In vitro fertilization | 18 (6.4) | 15 (5.7) |
| Ovulation induction/artificial insemination | 5 (1.8) | 4 (1.5) |
| Vaginal infection prior to enrollment, No. (%)d | 57 (20.4) | 55 (20.8) |
| Cervical length at screening, median (IQR), mm | 13.3 (9.1-17.2) | 13.7 (8.3-17.5) |
| <10 mm, No. (%) | 80 (28.6) | 82 (31.1) |
| Amniotic cavity debris on screening ultrasound, No./total (%) | 50/276 (18.1) | 44 (16.7) |
| Gestation at enrollment, median (IQR), wk | 21.7 (20.6-22.9) | 21.9 (20.7-23.2) |
| Time from screening to randomization, median (IQR), d | 0 (0-3) | 0 (0-3) |
| Current or planned progesterone use, No. (%) | 277 (98.9) | 261 (98.9) |
| Nugent score, No./total (%)e | ||
| 0-3 (normal flora) | 168/251 (66.9) | 161/234 (68.8) |
| 4-6 (intermediate) | 39/251 (15.5) | 34/234 (14.5) |
| 7-10 (bacterial vaginosis) | 44/251 (17.5) | 39/234 (16.7) |
Race and ethnicity are based on participant self-report.
Calculated as weight in kilograms divided by height in meters squared.
Any cervical surgery, including cervical cone, or loop electrosurgical excision procedure.
Includes chlamydia, gonorrhea, syphilis, trichomonas, bacterial vaginosis, genital herpes, HIV, hepatitis B, hepatitis C, hepatitis D, pyelonephritis, cystitis, or candidiasis diagnosed during the current pregnancy.
The Nugent score is a summary evaluation from the Gram stain of vaginal swabs. It ranges from 0 to 10, with scores of 7 or higher indicating bacterial vaginosis.
Outcomes
Primary Outcome
The primary outcome of preterm birth before 37 weeks or fetal death occurred in 127 participants (45.5%) randomized to pessary and in 120 (45.6%) randomized to usual care (relative risk in the pessary group, 1.00; 95% CI, 0.83-1.20) (Table 2).
Table 2. Preterm Delivery and Fetal Death and Maternal Secondary Outcomes.
| Outcomes | Pessary (n = 279) | Usual care (n = 263) | Absolute median difference (95% CI) | Unadjusted relative risk (95% CI) |
|---|---|---|---|---|
| Primary outcome | ||||
| Preterm delivery or fetal death <37 wk, No. (%) | 127 (45.5) | 120 (45.6) | 1.00 (0.83-1.20) | |
| Preterm delivery <37 wk, No. (%) | 109 (39.1) | 111 (42.2) | ||
| Fetal death <37 wk, No. (%)a | 18 (6.5) | 9 (3.4) | ||
| Secondary outcomes | ||||
| Time from enrollment to delivery or fetal death, median (IQR), d | 105 (42-20) | 103 (56-118) | 2 (−4 to 8) | |
| Gestational age at delivery, median (IQR), wk | 37.3 (27.4-39.1) | 37.4 (29.6-39.1) | 0.1 (−0.8 to1.1) | |
| Preterm delivery or fetal death, No. (%) | ||||
| <28 wk | 72 (25.8) | 54 (20.5) | 1.26 (0.92-1.71) | |
| <32 wk | 93 (33.3) | 73 (27.8) | 1.20 (0.93-1.55) | |
| <35 wk | 111 (39.8) | 90 (34.2) | 1.16 (0.93-1.45) | |
| Preterm premature rupture of membranes, No./total (%) | 64/277 (23.1) | 46/207 (17.8) | 1.30 (0.92-1.82) | |
| Spontaneous preterm delivery, No. (%) | 109 (39.1) | 106 (40.3) | 0.97 (0.79-1.19) | |
| Indicated preterm delivery, No. (%) | 18 (6.5) | 14 (5.3) | 1.21 (0.61-2.39) | |
| Cesarean delivery, No. (%) | 86 (30.8) | 80 (30.4) | 1.01 (0.79-1.31) | |
| Chorioamnionitis, No. (%) | 18 (6.5) | 16 (6.1) | 1.06 (0.55-2.04) | |
| Antepartum length of stay, median (IQR), d | 1 (0-1) | 1 (0-1) | 0 (0 to 0) | |
| Vaginal infection, No./total (%)b | 57/226 (25.2) | 41/220 (18.6) | 1.36 (0.95-1.94) | |
| Antibiotic or antifungal treatment, No./total (%)b | 55/226 (24.3) | 37/220 (16.8) | 1.45 (0.99-2.06) | |
| Treatment for preterm labor, No./total (%)c | 150/258 (58.1) | 143/250 (57.2) | 1.02 (0.88-1.18) | |
| Cerclage placement, No. (%) | 13 (4.7) | 24 (9.1) | 0.51(0.27-0.98) | |
| Pessary removal reasons, No. (%) | n = 276 | |||
| Medically indicated | 230 (83.3) | |||
| Final study visit | 104 (37.7) | |||
| Preterm premature rupture of membranes | 53 (19.2) | |||
| Preterm labor | 42 (15.2) | |||
| Delivery | 9 (3.3) | |||
| Vaginal bleeding | 7 (2.5) | |||
| Uterine contractions | 7 (2.5) | |||
| Cervical dilation ≥3 cm | 7 (2.5) | |||
| Confirmed clinical chorioamnionitis | 1 (0.4) | |||
| Not medically indicated | 46 (16.7) | |||
| Patient request (any reason) | 23 (8.3) | |||
| Cerclage | 7 (2.5) | |||
| Clinician request | 5 (1.8) | |||
| Self-removed | 4 (1.5) | |||
| Spontaneous dislodgement | 3 (1.1) | |||
| Recruitment closure | 2 (0.7) | |||
| Cervical erosion | 1 (0.4) | |||
| Incontinence | 1 (0.4) | |||
Includes only delivery prior to 20 weeks or fetal deaths that occurred prior to delivery, and not infant deaths following delivery at or after 20 weeks.
Participants with at least 1 study visit. Vaginal infection includes chlamydia, gonorrhea, syphilis, trichomonas, bacterial vaginosis, genital herpes, HIV, hepatitis B, hepatitis C, hepatitis D, pyelonephritis, cystitis, or candidiasis diagnosed during the current pregnancy. Antibiotics include those administered after randomization and before delivery via intravenous, oral, and vaginal routes.
Treatment for preterm labor includes cerclage, bed rest, tocolysis, antenatal corticosteroids for fetal lung maturity, and magnesium sulfate for neuroprotection.
Prespecified Secondary Outcomes
Rates of preterm birth or fetal death before 35 weeks, 32 weeks, and 28 weeks did not differ between participants randomized to pessary and those randomized to usual care. The proportion of individuals who remained pregnant across gestational ages did not differ between the 2 groups (hazard ratio in the pessary group, 1.9; 95% CI, 0.8-4.7) (Figure 2). Other obstetrical outcomes were similar between the 2 groups except cerclage placement, which occurred in 13 participants (4.7%) randomized to pessary and 24 (9.1%) randomized to usual care (relative risk, 0.51; 95% CI, 0.51-0.98). Reviewers who were masked to study group evaluated cerclage indications and determined cerclage placement to be examination indicated in 32 participants and non–examination indicated in 5 participants.
Figure 2. Kaplan-Meier Plot of Continued Pregnancy Without Delivery or Fetal Death.
The median time from randomization to delivery or fetal death was 105 days (IQR, 42-120 days) in the pessary group and 103 days (IQR, 56-118 days) in the usual care group.
Medically indicated removal of the pessary occurred in 83.3% of participants, with the most common nonmedical reason being participant request (n = 23 [8.3%]; 16 for discomfort, 1 for vaginal discharge, and 6 had no reason listed).
Suspected neonatal sepsis was less likely in the pessary group compared with the usual care group (14.7% vs 20.1%; relative risk, 0.67; 95% CI, 0.47-0.97); however, confirmed early and late sepsis did not differ between groups. All other neonatal outcomes and care were similar between the 2 groups (Table 3).
Table 3. Neonatal Outcomes Among Live Births.
| Outcomes | Pessary (n = 279) | Usual care (n = 263) | Absolute median difference (95% CI) | Unadjusted relative risk (95% CI) |
|---|---|---|---|---|
| Fetal or neonatal/infant death, No. (%) | 37 (13.3) | 18 (6.8) | 1.94 (1.13-3.32) | |
| Live births | n = 261 | n = 254 | ||
| Ventilator or continuous positive airway pressure use, No./total (%) | 75/259 (29.0) | 82/254 (32.3) | 0.90 (0.69-1.17) | |
| Ventilator or continuous positive airway pressure, median (IQR), d | 6 (0-43) [n = 97] | 5 (0-41) [n = 104] | 0 (−12 to 12) | |
| Supplemental oxygen, median (IQR), d | 7 (0-41) [n = 95] | 6 (0-51) [n = 105] | 1 (−12 to 14) | |
| Seizures requiring treatment, No./total (%) | 1/259 (0.4) | 1/254 (0.4) | 0.98 (0.06-15.7) | |
| Small for gestational age (less than fifth percentile), No./total (%) | 5/256 (1.9) | 6/257 (2.4) | 0.82 (0.25-2.65) | |
| Intraventricular hemorrhage grade 3 or 4, No./total (%) | 4/258 (1.6) | 7/254 (2.8) | 0.56 (0.17-1.90) | |
| Retinopathy of prematurity, No./total (%) | 27/259 (10.4) | 23/254 (9.1) | 1.15 (0.68-1.96) | |
| Respiratory distress syndrome, No./total (%) | 67/259 (25.9) | 69/254 (27.2) | 0.95 (0.71-1.27) | |
| Bronchopulmonary dysplasia, No./total (%) | 14/260 (5.4) | 19/254 (7.5) | 0.72 (0.37-1.41) | |
| Necrotizing enterocolitis stage 2 or 3, No./total (%) | 7/259 (2.7) | 6/254 (2.4) | 1.14 (0.39-3.37) | |
| Hyperbilirubinemia, No./total (%)a | 84/259 (32.4) | 82/253 (32.4) | 1.00 (0.78-1.29) | |
| Neonatal infectious morbidity, No./total (%) | 53/259 (20.5) | 65/254 (25.6) | 0.80 (0.58-1.10) | |
| Suspected sepsis | 38 (14.7) | 51 (20.1) | 0.67 (0.47-0.97) | |
| Early-onset sepsisb | 5 (1.9) | 2 (0.8) | 2.45 (0.48-12.57) | |
| Late-onset sepsisb | 8 (3.1) | 12 (4.7) | 0.65 (0.27-1.58) | |
| Pneumonia | 2 (0.8) | 4 (1.6) | 0.49 (0.09-2.66) | |
| Composite neonatal outcome, No./total (%)c | 81/260 (31.2) | 74/254 (29.1) | 1.05 (0.81-1.37) | |
| Hospital length of stay, median (IQR), d | 3 (2-20) [n = 258] | 3 (2-22) [n = 253] | 0 (0 to 0) | |
| Special care admission, No./total (%) | 98/260 (37.7) | 105/254 (41.3) | 0.91 (0.74-1.13) | |
| Special care length of stay, median (IQR), d | 50 (10-99) [n = 97] | 37 (11-88) [n = 104] | 10 (−16 to 36) | |
| Birth weight, median (IQR), g | 2847 (1410-3300) [n = 258] | 2810 (1690-3290) [n = 253] | 35 (−194 to 264) | |
| Congenital malformation, No./total (%) | 4/259 (1.5) | 2/253 (0.8) | 1.95 (0.36-10.61) |
Peak total bilirubin of 15 mg/dL or higher or phototherapy use.
Early-onset sepsis defined as diagnosis at or prior to 72 hours postbirth. Late-onset sepsis defined as diagnosis after 72 hours postbirth.
Includes any of death, respiratory distress syndrome, intraventricular hemorrhage grade 3 or 4, periventricular leukomalacia, necrotizing enterocolitis stage 2 or 3, bronchopulmonary dysplasia, retinopathy of prematurity stage 2 or higher, or early-onset sepsis.
In analyses of the as-treated population (5 patients in the pessary group were not analyzed [4 did not receive the pessary and 1 was lost to follow-up]; in the usual care group, 1 was not analyzed due to being lost to follow-up), fetal or neonatal/infant death occurred in 36 of 275 participants (13.1%) in the pessary group compared with 18 of 263 (6.8%) in the usual care group (relative risk, 1.91; 95% CI, 1.11-3.29; P = .02). More specifically, fetal death occurred in 17 (6.2%) of 275 pregnancies in the pessary group and in 9 (3.4%) of 263 in the usual care group (relative risk, 1.81; 95% CI, 0.82-3.99; P = .14), and there were 19 neonatal/infant deaths (7.4%) among 258 pregnancies in the pessary group and 9 neonatal/infant deaths (3.5%) among 254 pregnancies in the usual care group (relative risk, 2.34; 95% CI, 1.04-5.25; P = .04). The median gestation at fetal death was 20.5 weeks (IQR, 18.7-22.4 weeks) in the pessary group and 20.1 weeks (IQR, 19.1-21.3 weeks) in the usual care group. A summary of fetal deaths is provided in eTable 3 in Supplement 3.
In a post hoc analysis comparing those who experienced a fetal death (death prior to delivery or delivery prior to 20 weeks) or a neonatal/infant death (death following delivery at or after 20 weeks) with those who did not, the average cervical length at screening was shorter among those with a death and was more likely to be less than 10 mm (eTable 4 in Supplement 3). The qualifying cervical length was detected at an earlier mean gestational age than among those who did not have a fetal or neonatal/infant death. Those who experienced a fetal or neonatal/infant death were less likely to have had a prior term delivery. The mean gestational age at delivery in those who experienced a fetal or neonatal/infant death was 22.1 weeks (SD, 2.4 weeks) and there was a shorter time from randomization to delivery. Similarly, preterm premature rupture of membranes occurred more frequently among those receiving a pessary (50.0%) than among those receiving usual care (17.4%) (P < .001).
Adverse Events
Participants in the pessary group who had a pessary placed were more likely to report adverse effects (eTable 5 in Supplement 3). Specifically, they were more likely to report vaginal discharge, soreness, odor, itching, and leakage of fluid.
Discussion
The results of this trial demonstrate that in nonlaboring individuals with singleton pregnancies and a cervical length of 20 mm or less, cervical pessary placement did not result in a lower frequency of preterm birth or fetal death prior to 37 weeks of gestation. Pessary placement was associated with a higher rate of fetal or neonatal/infant death. Other neonatal outcomes, including adverse events and care utilization, were otherwise similar between the 2 groups. Per-protocol adherence to the pessary was high, although adverse effects of vaginal discharge, soreness, odor, and leakage of fluid were more commonly reported by individuals in the pessary group.
Our results differ from a 2012 clinical trial that found a significant reduction in preterm birth prior to 34 weeks and prior to 37 weeks in individuals with short cervical length of less than 25 mm randomized to receive a pessary.13 A key difference in our trial was the routine use of progesterone for short cervix and a requirement of a shorter cervical length (<20 mm). Although 2 prior trials had larger sample sizes,15,17 the sample size in this trial is larger than in other published randomized trials of pessary for preterm birth prevention in singleton gestations.13,14,16,24 Most randomized trials of pessary published since 2012 have included routine provision of progesterone and, similar to our trial, did not demonstrate a reduction in preterm birth.17,18,19 However, in a trial by Saccone et al16 that included individuals with a cervical length of 25 mm or less in which progesterone was provided to those with a cervical length of 20 mm or less, lower rates of preterm birth before 34 weeks and 37 weeks were found in the pessary group. Our study differs from other trials in that we excluded individuals with a history of preterm birth, required a shorter qualifying cervical length (<20 mm), and randomized participants at an earlier gestational age.
In our trial, cerclage was more common in participants randomized to receive usual care. For most participants, cerclage was examination indicated; there is controversy regarding this indication. This may have potentially advantaged the usual care group. This study adds to the literature suggestions that there is no incremental benefit of use of a cervical pessary when progesterone is initially prescribed.
This trial raises a new concern, given the higher rate of fetal or neonatal/infant death among individuals who received a pessary; this finding has not been reported in other randomized trials. Our ability to detect this may be due to recruitment of a particularly high-risk group of individuals with severe cervical length shortening found early in gestation, which was more common in those who experienced a fetal or neonatal/infant death. The overwhelming majority of fetal deaths (death prior to delivery or delivery prior to 20 weeks) and neonatal/infant deaths (death following delivery at or after 20 weeks) reflected periviable births, with both groups having a median gestational age of 22.1 weeks.
Limitations
This trial has several limitations. First, the open-label design has the potential to introduce bias. We attempted to minimize bias to the greatest extent possible through the rigor of the protocol and by selecting a primary outcome that was not prone to ascertainment bias. However, the difference in cerclage placement may result from an unmeasured bias. Second, our planned effect size differed from the trial results. The final cohort had a shorter cervical length than anticipated, and this likely contributed to the higher incidence of the primary outcome. Third, we are unable to precisely report the number of individuals who had a short cervical length in the population that may have potentially been eligible, as centers varied in their approach to clinical screening (ranging from universal screening to chance finding). In addition, referrals from outside centers and screening occurring at private clinicians’ offices complicated quantification of screening logs in a uniform fashion.
Conclusions
This trial demonstrates that cervical pessary placement in nonlaboring individuals with singleton gestations and a cervical length of 20 mm or less did not decrease the rate of preterm birth before 37 weeks. Furthermore, cervical pessary placement was associated with a higher risk of fetal and neonatal/infant mortality.
Trial Protocol
Statistical Analysis Plan
Trial Sites and Study Personnel
eTable 1. Trial Exclusion Criteria
eTable 2. Outcomes
eTable 3. Perinatal Death Summaries
eTable 4. Post-Hoc Analysis of Baseline Characteristics and Maternal Outcomes by Fetal or Neonatal/Infant Death (As Treated)
eTable 5. Side Effects (As Treated and Received)
eTable 6. Reasons for Exclusion
Nonauthor Collaborators. Eunice Kennedy Shriver National Institute of Child Health and Human Development Maternal-Fetal Medicine Units (MFMU) Network
Data Sharing Statement
References
- 1.Liu L, Oza S, Hogan D, et al. Global, regional, and national causes of under-5 mortality in 2000-15: an updated systematic analysis with implications for the Sustainable Development Goals. Lancet. 2016;388(10063):3027-3035. doi: 10.1016/S0140-6736(16)31593-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Martin JA, Hamilton BE, Osterman MJK, Driscoll AK. Births: final data for 2019. Natl Vital Stat Rep. 2021;70(2):1-51. [PubMed] [Google Scholar]
- 3.Moster D, Lie RT, Markestad T. Long-term medical and social consequences of preterm birth. N Engl J Med. 2008;359(3):262-273. doi: 10.1056/NEJMoa0706475 [DOI] [PubMed] [Google Scholar]
- 4.Crump C, Sundquist J, Winkleby MA, Sundquist K. Gestational age at birth and mortality from infancy into mid-adulthood: a national cohort study. Lancet Child Adolesc Health. 2019;3(6):408-417. doi: 10.1016/S2352-4642(19)30108-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Blencowe H, Lee AC, Cousens S, et al. Preterm birth-associated neurodevelopmental impairment estimates at regional and global levels for 2010. Pediatr Res. 2013;74(Suppl 1)(suppl 1):17-34. doi: 10.1038/pr.2013.204 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Tyson JE, Parikh NA, Langer J, Green C, Higgins RD; National Institute of Child Health and Human Development Neonatal Research Network . Intensive care for extreme prematurity—moving beyond gestational age. N Engl J Med. 2008;358(16):1672-1681. doi: 10.1056/NEJMoa073059 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Crowley PA. Antenatal corticosteroid therapy: a meta-analysis of the randomized trials, 1972 to 1994. Am J Obstet Gynecol. 1995;173(1):322-335. doi: 10.1016/0002-9378(95)90222-8 [DOI] [PubMed] [Google Scholar]
- 8.Richardson DK, Gray JE, Gortmaker SL, Goldmann DA, Pursley DM, McCormick MC. Declining severity adjusted mortality: evidence of improving neonatal intensive care. Pediatrics. 1998;102(4 pt 1):893-899. doi: 10.1542/peds.102.4.893 [DOI] [PubMed] [Google Scholar]
- 9.Iams JD, Goldenberg RL, Meis PJ, et al. ; National Institute of Child Health and Human Development Maternal Fetal Medicine Unit Network . The length of the cervix and the risk of spontaneous premature delivery. N Engl J Med. 1996;334(9):567-572. doi: 10.1056/NEJM199602293340904 [DOI] [PubMed] [Google Scholar]
- 10.Orzechowski KM, Boelig RC, Baxter JK, Berghella V. A universal transvaginal cervical length screening program for preterm birth prevention. Obstet Gynecol. 2014;124(3):520-525. doi: 10.1097/AOG.0000000000000428 [DOI] [PubMed] [Google Scholar]
- 11.Esplin MS, Elovitz MA, Iams JD, et al. Predictive accuracy of serial transvaginal cervical lengths and quantitative vaginal fetal fibronectin levels for spontaneous preterm birth among nulliparous women. JAMA. 2017;317(10):1047-1056. doi: 10.1001/jama.2017.1373 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Berghella V, Rafael TJ, Szychowski JM, Rust OA, Owen J. 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. doi: 10.1097/AOG.0b013e31820ca847 [DOI] [PubMed] [Google Scholar]
- 13.Hassan SS, Romero R, Vidyadhari D, et al. ; PREGNANT Trial . Vaginal progesterone reduces the rate of preterm birth in women with a sonographic short cervix: a multicenter, randomized, double-blind, placebo-controlled trial. Ultrasound Obstet Gynecol. 2011;38(1):18-31. doi: 10.1002/uog.9017 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Fonseca EB, Celik E, Parra M, Singh M, Nicolaides KH; Fetal Medicine Foundation Second Trimester Screening Group . Progesterone and the risk of preterm birth among women with a short cervix. N Engl J Med. 2007;357(5):462-469. doi: 10.1056/NEJMoa067815 [DOI] [PubMed] [Google Scholar]
- 15.Goya M, Pratcorona L, Merced C, et al. ; Pesario Cervical Para Evitar Prematuridad (PECEP) Trial Group . Cervical pessary in pregnant women with a short cervix (PECEP): an open-label randomised controlled trial. Lancet. 2012;379(9828):1800-1806. doi: 10.1016/S0140-6736(12)60030-0 [DOI] [PubMed] [Google Scholar]
- 16.Saccone G, Maruotti GM, Giudicepietro A, Martinelli P; Italian Preterm Birth Prevention (IPP) Working Group . Effect of cervical pessary on spontaneous preterm birth in women with singleton pregnancies and short cervical length: a randomized clinical trial. JAMA. 2017;318(23):2317-2324. doi: 10.1001/jama.2017.18956 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Nicolaides KH, Syngelaki A, Poon LC, et al. A randomized trial of a cervical pessary to prevent preterm singleton birth. N Engl J Med. 2016;374(11):1044-1052. doi: 10.1056/NEJMoa1511014 [DOI] [PubMed] [Google Scholar]
- 18.Dugoff L, Berghella V, Sehdev H, Mackeen AD, Goetzl L, Ludmir J. Prevention of Preterm Birth With Pessary in Singletons (PoPPS): randomized controlled trial. Ultrasound Obstet Gynecol. 2018;51(5):573-579. doi: 10.1002/uog.18908 [DOI] [PubMed] [Google Scholar]
- 19.Pacagnella RC, Silva TV, Cecatti JG, et al. ; P5 Working Group . Pessary plus progesterone to prevent preterm birth in women with short cervixes: a randomized controlled trial. Obstet Gynecol. 2022;139(1):41-51. doi: 10.1097/AOG.0000000000004634 [DOI] [PubMed] [Google Scholar]
- 20.Rosenberger WF, Lachin J. Randomization in Clinical Trials: Theory and Practice. John Wiley & Sons; 2002. doi: 10.1002/0471722103 [DOI] [Google Scholar]
- 21.Baschat AA. Pathophysiology of fetal growth restriction: implications for diagnosis and surveillance. Obstet Gynecol Surv. 2004;59(8):617-627. doi: 10.1097/01.OGX.0000133943.54530.76 [DOI] [PubMed] [Google Scholar]
- 22.Alexander GR, Kogan MD, Himes JH. 1994-1996 U.S. singleton birth weight percentiles for gestational age by race, Hispanic origin, and gender. Matern Child Health J. 1999;3(4):225-231. doi: 10.1023/a:1022381506823 [DOI] [PubMed] [Google Scholar]
- 23.Lan KKG, DeMets DL. Discrete sequential boundaries for clinical trials. Biometrika. 1983;70(3):659. doi: 10.2307/2336502 [DOI] [Google Scholar]
- 24.Hui SYA, Chor CM, Lau TK, Lao TT, Leung TY. Cerclage pessary for preventing preterm birth in women with a singleton pregnancy and a short cervix at 20 to 24 weeks: a randomized controlled trial. Am J Perinatol. 2013;30(4):283-288. doi: 10.1055/s-0032-1322550 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Trial Protocol
Statistical Analysis Plan
Trial Sites and Study Personnel
eTable 1. Trial Exclusion Criteria
eTable 2. Outcomes
eTable 3. Perinatal Death Summaries
eTable 4. Post-Hoc Analysis of Baseline Characteristics and Maternal Outcomes by Fetal or Neonatal/Infant Death (As Treated)
eTable 5. Side Effects (As Treated and Received)
eTable 6. Reasons for Exclusion
Nonauthor Collaborators. Eunice Kennedy Shriver National Institute of Child Health and Human Development Maternal-Fetal Medicine Units (MFMU) Network
Data Sharing Statement