Cervical pessary versus progesterone for preterm birth prevention in twin pregnancy with short cervix

Photo by Thought Catalog / Unsplash

The Australian & New Zealand journal of obstetrics & gynaecology Published April 3, 2026 Medically reviewed May 26, 2026 Study authors: McGannon Claire, Au Ling Shan, Neil Peter, Roberts Calum, Mol Ben W J PubMed ↗ DOI ↗ By Dr. Sofia Müller, MD · Lifespan & Whole-Person Care

Key Takeaway

Interpret small RCT on pessary vs progesterone for twin pregnancy with short cervix as preliminary due to early termination.

This randomized clinical trial, conducted at a single center in Australia, enrolled 34 individuals with a twin pregnancy and a mid-trimester cervical length at or below the 25th percentile (≤38 mm). Participants were randomized to receive either placement of a cervical pessary (Arabin pessary, n=16) or self-administered daily vaginal progesterone (200 mg, n=18). One participant in the progesterone group withdrew consent after randomization, leaving 33 individuals for analysis.

The primary outcome was a composite adverse neonatal outcome. This occurred in 41% of neonates (13/32) in the pessary group compared to 29% of neonates (10/34) in the progesterone group (risk ratio [RR] 1.38, 95% CI 0.71-2.69). For the secondary outcome of delivery before 28 weeks, the rate was 19% (3/16 individuals) in the pessary group versus 5.9% (1/17) in the progesterone group (RR 3.19, 95% CI 0.37-27.58). Neither comparison was statistically significant, as all confidence intervals included 1.0.

Safety and tolerability data were not reported. The study has important limitations: it was halted early after other studies indicated no benefit from pessary, resulting in a very small sample size. This early termination, combined with the wide confidence intervals, means the evidence has low certainty. The findings do not demonstrate harm from the pessary or superiority of progesterone, but they do not support a benefit of cervical pessary over progesterone for preventing preterm birth in this specific population.

Study Details

Study typeRct

EvidenceLevel 2

Follow-up5.3 mo

PublishedApr 2026

PMID41879343

View Original Abstract ↓

BACKGROUND: Both twin pregnancy and short cervix are risk indicators for preterm birth. Cervical pessary and progesterone have been suggested to prevent preterm birth. A randomised trial was conducted in individuals carrying twins with cervix below the 25th centile (less than 38 mm). METHODS: We conducted a single centre open-label randomised clinical trial at Monash Medical Centre in Victoria, Australia (Trial registration number: ACTRN12616000875404). Individuals with a twin pregnancy and mid-trimester (16 to 23 weeks) cervical length < 25th percentile were invited to participate. After informed consent, participants were randomly assigned to placement of cervical pessary or self-administered daily vaginal progesterone (200 mg), until 36 weeks of gestation or until onset of labour, whichever occurred first. Primary outcome was a composite adverse neonatal outcome, with preterm birth before 28 weeks as an important secondary outcome. RESULTS: Between 2 December 2019 and 27 July 2023, 159 individuals were screened, of whom 66 had a cervical length ≤ 38 mm, and 34 were randomised to Arabin pessary (N = 16) or progesterone (N = 18). One woman in the progesterone group withdrew consent after randomisation. The study was halted as studies elsewhere had indicated no benefit from pessary. Composite adverse neonatal outcome occurred in 41% (13/32) of the neonates in the Arabin pessary group compared with 29% (10/34) of the neonates in the progesterone group (risk ratio [RR] 1.38, 95% CI 0.71-2.69). Delivery before 28 weeks occurred in 19% (3/16) and 5.9% (1/17) (RR: 3.19, 95% CI 0.37-27.58). CONCLUSION: In individuals with a twin pregnancy and a short cervix, we could not report benefit from cervical pessary over progesterone.