The value of cervical length changes for the prediction of preterm birth with normal Mid-trimester cervical length; a prospective longitudinal study

Abstract

Spontaneous preterm birth (sPTB) remains a leading cause of neonatal morbidity and mortality. Mid-trimester transvaginal cervical length (CL) measurement is an established screening tool, but it is unclear whether longitudinal change in CL between the 1st and 2nd trimesters (ΔCx) adds predictive value in women with a normal mid-trimester cervix. Methods This prospective longitudinal cohort study included singleton pregnancies with 1st and 2nd trimester transvaginal CL measurements. Women with a normal mid-trimester cervix and no antenatal progesterone exposure comprised the primary low-risk cohort, while women with a short cervix were analyzed separately as an exploratory comparison cohort. CL was measured at 11–14 weeks (C1) and 18–24 weeks (C2). ΔCx and cervical slope were calculated. The primary outcome was sPTB <37 weeks. Additional analyses examined preterm prelabor rupture of membranes (PPROM) and PROM at term. Results A total of 2,354 women had cervical length data in both trimesters; 2,292 constituted the primary low-risk cohort and 62 the exploratory short-cervix cohort. The overall rate of sPTB was 8.00 and 30.6% in the low-risk and short cervix cohorts, respectively (p < 0.001). In the low-risk cohort, C1 (35.74 ± 4.61 mm vs. 35.50 ± 4.77 mm, p = 0.497) and C2 (37.44 ± 5.50 mm vs. 36.83 ± 5.55 mm, p = 0.148), ΔCx (1.69 ± 7.19 mm vs. 1.33 ± 6.81 mm, p = 0.515), and cervical slope were similar between term and preterm pregnancies (p = 0.558). In multivariable analysis, cervical slope was not associated with sPTB and did not improve discrimination beyond C2 and maternal characteristics. By contrast, the short-cervix cohort had markedly shorter C1 (30.31 ± 5.51 mm vs. 35.72 ± 4.62 mm, p < 0.001), C2 (21.61 ± 6.55 mm vs. 37.39 ± 5.50 mm, p < 0.001) a more negative cervical slope (–0.99 ± 1.00 mm vs. 0.19 ± 0.81 mm, p < 0.001). PPROM was more frequent (12.7% vs. 43.1%, p < 0.001) among preterm births in the low-risk cohort, and cervical slope was associated with PPROM (p = 0.030). Conclusion In women with a normal mid-trimester cervix, longitudinal CL change does not improve prediction of sPTB beyond a single 2nd trimester measurement. The clinical value of cervical screening appears to lie primarily in identifying the short-cervix subgroup.