4: Multicenter randomized trial of cerclage for preterm birth prevention in high-risk women with shortened mid-trimester cervical length


      To evaluate cerclage for preterm birth (PTB) prevention in women with a prior spontaneous PTB and shortened cervical length (CL).

      Study Design

      Randomized, intent to treat clinical trial at 16 U.S. centers. Women with prior spontaneous PTB <34 wks and a singleton underwent serial vaginal ultrasound at 16-22 6/7 wks. Those with CL <25 mm were assigned to cerclage or no-cerclage. Subsequent management was similar in both groups. A sample size of 300 was based on an 80% power for a 33% difference in PTB rates <35 wks and a 10% dropout rate.


      Of 1014 women who began ultrasound screening, 317 experienced CL shortening, and 302 were randomized. The cerclage and no-cerclage groups were similar with regard to: CL (18.7 v. 19.5 mm), and gestational age (GA) (19.4 v. 19.5 wks) at randomization, GA of earliest prior PTB (24.2 v. 24.5 wks) and race/ethnicity. Of 301 patients with outcomes, 148 assigned to cerclage, had a PTB <35 wk rate of 32% v. 42% in the 153 no-cerclage group (P=0.09). Survival analysis confirmed cerclage benefit (P=0.053). Planned interaction analysis between CL strata <15 mm v. 15-24 mm and cerclage was significant (P=0.03); cerclage benefit was highly concentrated in women with very short CL (Figure). The adjusted odds ratio (95% C.I.) for PTB <35 wk was 0.23 (0.08-0.66) in the <15 mm stratum v. 0.84 (0.49-1.4) in the 15-24 mm stratum.


      Women with prior spontaneous PTB and mid-trimester CL <25 mm may benefit from cerclage, but the benefit is most pronounced when the CL is <15 mm, suggesting the presence of a more significant, and treatable component of cervical insufficiency.