A Silicone Ring, a Short Cervix, and Twenty Years of Trying to Answer One Question

What three studies published in 2025 and 2026 say about whether a cervical pessary can delay an early birth

A short cervix found midway through pregnancy is one of the clearest signs that a baby may arrive too early. For twenty years, doctors have been testing whether a small silicone ring called a cervical pessary can hold a pregnancy longer. Three studies published in 2025 and 2026 give three different answers — and the reasons they disagree are as important as the answers themselves.

Why a short cervix matters so much

The cervix is the lower part of the womb: a firm, closed neck of tissue that keeps a pregnancy where it belongs until it is time for labour. In most pregnancies it stays long and closed until the very end. In some it begins to shorten and soften far too early, often with no symptoms at all, and that quiet shortening is one of the strongest warnings doctors have that a baby will be born prematurely. Around 13.4 million babies worldwide were born before 37 weeks in 2020 [1], and being born early remains the biggest single reason babies need intensive care.

Measuring the cervix is easy — a routine ultrasound scan does it in minutes. Doing something useful about a short measurement has proved far harder. Three approaches exist: progesterone, a hormone given as a nightly vaginal capsule; a cerclage, a stitch placed surgically around the cervix to hold it closed; and the cervical pessary, a soft silicone ring fitted in an ordinary clinic appointment that sits in the vagina and cradles the cervix, changing the angle at which the weight of the pregnancy presses down on it. The pessary is by far the gentlest option — no operation, no anaesthetic, removable at any time. That is exactly why so many people have wanted it to work.

What families faced before any of this was tested

Before the mid-2000s, a mother told her cervix was shortening was offered very little. No treatment had convincing evidence behind it, so the usual advice was rest, watchfulness and repeat scans — another way of saying that everyone waited to see what would happen. Many of those pregnancies ended in a very small baby, and families lived through weeks of anxiety knowing nothing was being done because nothing was known to help.

That began to change when a study compared progesterone against a dummy treatment in women whose cervix measured 15 millimetres or less. Among those given progesterone, 19.2% gave birth before 34 weeks; among those given the dummy, 34.4% did [2]. For the first time, a short cervix was something to act on rather than simply worry about. Progesterone became standard care, and the search began for something to add to it.

The pessary's first big test looked spectacular. The PECEP trial (a study of the cervical pessary in women with a short cervix), involving 385 women at maternity hospitals across Spain, reported a large reduction in early births [3]. Enthusiasm spread quickly — and then the replication studies arrived. A trial across 16 maternity hospitals in England, Slovenia, Portugal, Chile, Australia, Italy, Albania, Germany and Belgium found no benefit in women expecting one baby [4]. The ProTWIN trial (a Dutch study of pessaries in twin and other multiple pregnancies) found none either [5], and the STOPPIT-2 trial (the Arabin pessary in twin pregnancies with a short cervix), run across 57 antenatal clinics in the United Kingdom and Europe, also found nothing [6].

Then came the result that made many hospitals stop using the device altogether. The P5 trial, run across 17 health centres in Brazil between 2015 and 2019, had found fewer babies dying before birth in the group given a pessary [7]. But the TOPS trial, run at 12 centres in the United States, found the opposite — more babies dying before birth in the pessary group [8]. Faced with a possible safety problem, a great many units simply put the pessaries away. That is where matters stood when the three new studies were published.

The three new studies, and what each one asked

The first study, published in June 2025, is what researchers call a meta-analysis: instead of running a new trial, the authors gathered every existing trial that met strict quality standards and combined their results into one larger, more reliable picture [9]. They searched the medical databases up to June 2024, registered their plan in advance so they could not change the rules partway through, and ended up with nine randomised trials covering 3,813 women. A randomised trial is one in which a coin-toss decides who receives the treatment — the only reliable way to be sure a difference between groups is caused by the treatment rather than by who was chosen for it.

The second study, published in January 2026, went back to the Brazilian P5 trial data and asked a narrower question [10]. What about the women at the very highest risk — those who had already had one premature baby and now had a short cervix? For that group, 155 women in total, does adding a pessary on top of progesterone help?

The third study, published in February 2026, looked at 120 twin pregnancies cared for at a university hospital in Alexandroupolis, Greece, between 2019 and 2024 [11]. Every woman received progesterone. Some also received a pessary, some a stitch, some neither. Crucially, no coin was tossed: doctors chose case by case, based on what each woman's cervix looked like and what her history was.

What the studies found

The meta-analysis found no clear effect either way [9]. Combining all nine trials, the pessary neither reduced nor increased the risk of a baby dying around the time of birth, and neither reduced nor increased the risk of birth before 28 weeks. The results were what statisticians call imprecise — the range the data leave open still includes real benefit and real harm.

Two things in that paper are worth holding onto. First, when the authors removed the American TOPS trial from the pool and re-ran the numbers, the alarming safety signal did not reappear; the evidence as a whole does not show that pessaries cause babies to die. Second, the authors estimated how much data would be needed before a firm answer becomes possible, and showed that the world's trials have never accumulated enough. The honest summary is not "pessaries don't work" but "after twenty years, we still cannot say."

The re-analysis of the Brazilian trial was clearer, and negative [10]. For women who had both a previous early birth and a short cervix now, adding a pessary to progesterone made no difference at any stage — not before 37 weeks, not before 34, 32 or 28. Both groups' pregnancies lasted essentially the same length of time. The same paper produced something useful: a list of features that best predict a repeat early birth, including carrying twins, previous surgery on the cervix, a funnel-shaped opening at the top of the cervix on the scan, and previous miscarriages.

The Greek twin study found something apparently remarkable [11]. Twenty-six of the 120 pregnancies (21.6%) ended before 33 weeks, and women who had received a pessary carried their pregnancies dramatically longer than those who had received a stitch — a calculated risk reduction of around 91%.

Why the numbers disagree, and which to believe

A 91% reduction is an extraordinary claim for a silicone ring, and the Greek study contains its own clue that something is amiss. Funnelling — that funnel-shaped opening at the top of the cervix — is a known danger sign, and the Brazilian analysis confirmed it as one [10]. Yet in the Greek study's calculations, funnelling came out looking protective [11]; the authors say so themselves and call the finding paradoxical.

When a study makes a known warning sign look like a good sign, it is telling you the comparison is not fair. And there is a plain reason why it would not be. In Greece, no coin was tossed. Doctors decided who got a stitch and who got a ring, and they naturally reserved the stitch — the more serious surgical option — for women whose cervices were already in the most trouble. Comparing those women with the pessary group is not comparing two treatments; it is comparing two groups who were different before treatment started. This is one of the commonest traps in medical research, and exactly why coin-toss trials exist.

So the meta-analysis [9] and the re-analysis of a randomised trial [10] carry the weight; the Greek cohort [11] is interesting but cannot settle anything. A separate 2025 review of 17 studies covering 5,704 women did find a benefit from pessaries specifically in twin pregnancies with a short cervix, and none in single pregnancies [12] — so twins remain the one situation where the question is genuinely open.

What this means for you and your baby

If you have been told your cervix is short, the treatment with the strongest evidence behind it is vaginal progesterone [2]. If you have had a previous early birth as well, the best current evidence says adding a pessary to progesterone will not help you [10] — and knowing that spares you a device, extra appointments and false hope. If you are expecting twins the picture is less settled, and both a pessary and a stitch may reasonably be discussed; there is good evidence that a stitch helps twins whose cervix is already short or beginning to open [13].

If you read about the TOPS result and were frightened by it, the useful thing to know is that when all the trials were pooled, that safety signal did not hold up [9],[8]. And if you already have a pessary in place, nothing here suggests you should be alarmed: in the Brazilian protocol the ring's position was checked by ultrasound straight after fitting and every four weeks after that, and it was removed at 36 weeks or sooner if needed [10].

The message is honest uncertainty rather than bad news. Doctors offering or declining a pessary are not withholding a proven cure or pushing a proven failure. They are working at the edge of what is known.

What researchers are working on next

Three things need to happen. Trials need to stop lumping very different women together — a woman with twins, a woman with a previous early birth, and a woman with an unexpectedly short cervix and no other risk factor are not the same patient, and pooling them has hidden whatever effect exists [10],[12]. Timing needs its own study: if the process that shortens the cervix is well under way by 18 weeks, every trial so far may have acted too late [10]. And head-to-head comparisons are under way — a trial in two hospitals in Ho Chi Minh City, Vietnam, has compared the stitch against the ring, with and without progesterone, in twin pregnancies with a short cervix [14].

Until those answers arrive, the fair thing to say is that a pessary is unproven, not disproven — and families deserve to hear it described that way.

References

  1. Ohuma EO, Moller AB, Bradley E, et al. National, regional, and global estimates of preterm birth in 2020, with trends from 2010: a systematic analysis. Lancet. 2023;402(10409):1261–1271. doi:10.1016/S0140-6736(23)00878-400878-4) ↩
  2. Fonseca EB, Celik E, Parra M, Singh M, Nicolaides KH. 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 ↩
  3. Goya M, Pratcorona L, Merced C, et al. 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)60030-060030-0) ↩
  4. Nicolaides KH, 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 ↩
  5. Liem S, et al. Cervical pessaries for prevention of preterm birth in women with a multiple pregnancy (ProTWIN): a multicentre, open-label randomised controlled trial. Lancet. 2013;382(9901):1341–1349. doi:10.1016/S0140-6736(13)61408-761408-7) ↩
  6. Norman JE, et al. Evaluation of the Arabin cervical pessary for prevention of preterm birth in women with a twin pregnancy and short cervix (STOPPIT-2): an open-label randomised trial and updated meta-analysis. PLoS Med. 2021;18(3):e1003506. doi:10.1371/journal.pmed.1003506 ↩
  7. Pacagnella RC, Silva TV, Cecatti JG, et al. 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 ↩
  8. Hoffman MK, Clifton RG, Biggio JR, et al. Cervical pessary for prevention of preterm birth in individuals with a short cervix: the TOPS randomized clinical trial. JAMA. 2023;330(4):340–348. doi:10.1001/jama.2023.10812 ↩
  9. Provinciatto HG, Araujo Júnior E, Callado GY, et al. Pessary for prevention of preterm birth and perinatal mortality in pregnancies with a short cervix: systematic review and meta-analysis of randomized controlled trials. Diagnostics. 2025;15(12):1466. doi:10.3390/diagnostics15121466 ↩
  10. França MS, França GUS, Hatanaka AR, et al. Efficacy of combined cervical pessary and progesterone in women at high-risk of preterm birth. Diagnostics. 2026;16(3):402. doi:10.3390/diagnostics16030402 ↩
  11. Pagkaki C, Kritsotaki N, Bothou A, et al. Comparative effectiveness of pessary placement, cervical cerclage, or expectant management in preventing preterm delivery in twin pregnancies. J Pers Med. 2026;16(2):104. doi:10.3390/jpm16020104 ↩
  12. Santos ACFF, Pimenta NS, Pereira AGA, et al. Short cervix and use of cervical pessary for preventing preterm birth in singleton and twin pregnancies: a systematic review and meta-analysis. Rev Bras Ginecol Obstet. 2025;47:e-rbgo10. doi:10.61622/rbgo/2025rbgo10 ↩
  13. D'Antonio F, Eltaweel N, Prasad S, et al. Cervical cerclage for prevention of preterm birth and adverse perinatal outcome in twin pregnancies with short cervical length or cervical dilatation: a systematic review and meta-analysis. PLoS Med. 2023;20(8):e1004266. doi:10.1371/journal.pmed.1004266 ↩
  14. He YTN, Pham HNH, Nguyen TC, et al. Cervical cerclage versus cervical pessary with or without vaginal progesterone for preterm birth prevention in twin pregnancies and a short cervix: a two-by-two factorial randomised clinical trial. PLoS Med. 2025;22(2):e1004526. doi:10.1371/journal.pmed.1004526 ↩