Does late second-trimester cervical length predict unscheduled preterm birth in asymptomatic triplet pregnancies?

Multiple gestation increases the risk of unscheduled preterm birth (PTB), both spontaneous and indicated, leading to increased neonatal morbidity and additional healthcare costs. The purpose of this study was to determine whether cervical length (CL) assessment by 28 weeks could predict unscheduled PTB <34 weeks in triplet pregnancies. Secondary outcomes included prediction of PTB <30 weeks, prediction of PTB based on degree of cervical change and effect of ART-use on PTB.

This was a retrospective cohort of women with triplet pregnancies. The exposure variable of interest was short cervix < 25 and <20 millimeters (mm) by 28 weeks. Maternal characteristics were described. The distribution of CLs was analyzed by the primary outcome of unscheduled PTB < 34 weeks, and by PTB <30 weeks (secondary outcome). Gestational age at delivery was compared between women with and without a short cervix. Changes in CL were compared between the groups with unscheduled PTB and those delivering ≥34 and ≥30 weeks. Statistical analyses were performed using appropriate tests.

Of 92 triplet pregnancies, 51 met the criteria, with 1233 total (411 shortest) CL measurements from 16 to 34 weeks' gestation. The overall rate of PTB <34 weeks was 31.4% and <30 weeks was 9.8%. The median gestational age at delivery was 32.7 (IQR 2.3) weeks. There were no statistically significant differences in rates of unscheduled PTB in women who had a short cervix and those that did not: PTB <34 weeks with CL <25 mm (p = .53) and CL <20 mm (p = .70); PTB <30 weeks with CL <25 mm (p = .38) and CL <20 mm (p = .26). The degree of cervical change from 18 to 28 weeks was not statistically significant for predicting unscheduled PTB <34 and <30 weeks. Of 70.6% of triplet pregnancies conceived by ARTs, 13.9% had unscheduled PTB <30 weeks, whereas no spontaneously-conceived pregnancies delivered <30 weeks (p = .14).

Short cervix did not predict unscheduled spontaneous PTB <34 weeks nor <30 weeks in our triplet cohort, nor did the degree of cervical change by 28 weeks predict PTB. Triplets conceived by ARTs may have an increased risk of unscheduled PTB.