Cervical length measurement at mid gestation to predict spontaneous preterm birth in asymptomatic triplet pregnancies.

Objective To assess the predictive value of cervical length (CL) measured at mid gestation for spontaneous preterm birth (PTB) in triplet pregnancies. Methods Retrospective study on asymptomatic triplet pregnancies followed at five Italian tertiary referral centres, between 2002 and 2015. CL was measured transvaginally between 18–24 weeks of gestation. Pregnancies with a medically indicated PTB were excluded. Demographic and pregnancy characteristics of pregnancies complicated with PTB were analysed and the distributions of CL measurements in these patients were calculated. Logistic regression analysis was performed to assess the association between CL and PTB, adjusted for confounders. CL performance to predict PTB <28, <30 and <32 weeks of gestation was assessed. Results A total of 120 pregnancies were included in the final analysis. Median CL was 35 mm (interquartile range [IQR] 29–40) measured at a median gestational age of 20+2 (IQR 20+0-23+4) weeks. Overall, 23 (19.2%), 17 (14.2%) and 8 (6.7%) patients had a CL <25, <20 and <15 mm, respectively. Spontaneous PTB < 32 weeks occurred in 41 (34.2%) cases, <30 weeks in 23 (19.2%) and <28 weeks in 12 (10%) cases. A CL <15 mm was significantly more frequent in the group of patients delivered <28 and <30 weeks (p = 0.03 and p = 0.01), while a cervical length <20 mm was more common in triplet pregnancies delivered <32 weeks (p = 0.03). Logistic regression analysis was possible for PTB <32 weeks: after adjustment for confounders CL was not significantly associated with PTB (OR 0.97, CI 0.94 to 1.01). CL showed an area under the receiver operating characteristic curve of 0.41 (95% confidence interval [CI] 0.20 to 0.62), 0.41 (95% CI 0.26 to 0.56) and 0.42 (95% CI 0.31 to 0.54) for the prediction of PTB < 28, < 30 and < 32 weeks, respectively. Conclusions CL assessed at mid gestation is a poor predictor of PTB in asymptomatic triplet pregnancies.