Advertisement

2 Pessary plus progesterone to prevent preterm birth in women with a short cervix (P5 trial)

      Objective

      While there it is known that progestogens reduce spontaneous preterm birth in women with a mid-trimester short cervix, studies on cervical pessary show conflicting results. We compared the additional effectiveness of cervical pessary for the prevention of preterm birth in women with a mid-trimester short cervix already using progesterone.

      Study Design

      We performed a multicenter randomized controlled trial in 18 perinatal centers in Brazil (RBR-3t8prz UTN:U1111-1164-2636). Asymptomatic women with a singleton or twin pregnancy and a CL ≤ 30mm, measured at 16+0 – 22+6 weeks gestation, were randomized to cervical pessary + vaginal progesterone or vaginal progesterone (200mg daily) alone. Treatments were used from randomization to 36 weeks of gestation or delivery. The primary outcome was a composite of neonatal mortality and morbidity. Secondary outcomes were among others delivery <37 weeks and delivery <34 weeks. We planned subgroup analysis according to cervical length (≤ 25 mm), parity and the number of fetuses (singletons or twins).

      Results

      Between July 15, 2015 and March 29, 2019, we measured cervical length in 8,490 women, of which 475 women were randomized to receive pessary + progesterone and 461 women to progesterone alone. The primary composite perinatal outcome occurred in 89/463 19.2% versus (91/436) 20.9%, respectively (RR 0·85, 95% CI 0·55 to 1·1). Delivery rates < 37 weeks were 29·1% versus 31·4% (RR 0·86 CI 0·72 - 1·04) while < 34 weeks, delivery rates were 9·9% versus 13·9% (RR 0·66 CI 0·47-0·93). In women with a nulliparous singleton pregnancy and cervical length ≤ 25 mm, the composite perinatal outcomes occurred in 20% versus 33% of the women (RR 0.59, 95% CI 0.37 – 0.94, p-value 0.02).

      Conclusion

      In asymptomatic women with a short cervix, the combination of cervical pessary + vaginal progesterone did not statistically significantly improve perinatal outcome over vaginal progesterone alone. However, delivery <34 weeks occurred significantly less, specifically in nulliparous women with a CL <25 mm
      Figure thumbnail fx1