Should cervical favourability play a role in the decision for labour induction in gestational hypertension or mild pre-eclampsia at term? an exploratory analysis of the HYPITAT trial: Editorial comment

Gestational hypertension/preeclampsia is the most common obstetric complication of pregnancy. Management in the near term and early term periods remains controversial; the 2 options are induction of labor versus expectant management. Induction at an earlier gestational age presents risks of cesarean delivery (CD) and increased rates of neonatal morbidities. Maternal risks of expectant management include progression to severe gestational hypertension, eclampsia, or placental abruption. The Hypertension and Pre-eclampsia Intervention Trial at Term (HYPITAT) compared labor induction with expectant management and showed that induction reduced the occurrence of high-risk situations and was not associated with an increase in CDs or adverse neonatal outcomes. This post hoc analysis of the HYPITAT trial data was performed to evaluate the association between ripeness of the cervix and outcomes of labor induction and expectant management.

The HYPITAT trial included 756 women with a singleton pregnancy at 36 to 41 weeks’ gestation and complicated by gestational hypertension or mild preeclampsia; they were allocated to either labor induction or expectant management. In the induction group, labor was induced within 48 hours of randomization. Women in the expectant management group were monitored until the onset of spontaneous labor. The primary outcome was a composite of high-risk situations (ie, maternal mortality, maternal morbidity, progression to severe disease, and major postpartum hemorrhage). Secondary outcomes were CD and a composite of adverse neonatal outcomes. Before labor, the cervix was assessed by cervical length measurement and Bishop score calculation. The predictors in each model were treatment, cervical length, and an interaction between these factors.

The HYPITAT trial found that labor induction reduced the occurrence of high-risk situations without increasing the risk of CD or adverse neonatal outcomes. The median cervical length was 30 mm (range, 0–64 mm), and median Bishop score was 3 (range, 0–9). For the expectant management group, each centimeter of increase in cervical length was associated with a 32% increase in the likelihood of high-risk situations. For the induced women, each centimeter increase in baseline cervical length carried a 3% risk reduction for maternal high-risk situations (P = 0.03 for interaction between the treatment options). Fifty-one (13.5%) of 377 women who were induced developed an indication for CD compared with 68 (17.9%) of 379 women monitored expectantly. In both groups, the probability of CD was lower when the cervix was more favorable. For each centimeter increase in the length of cervix, the risk of CD was 31% and 14% higher for expectant management and labor induction, respectively (P = 0.48 for interaction). No association was found between cervical length and risk of adverse neonatal outcomes. Analyses using the Bishop score found similar, but not statistically significant, associations between cervical favorability and high-risk maternal situations, CD, and adverse neonatal outcomes. The women were categorized into those with a favorable or unfavorable cervix at baseline. In the induction group, the average times to delivery with a favorable or unfavorable cervix were 1.9 and 2.6 days, respectively. In women managed expectantly, the respective times to delivery were 7.7 and 9.1 days. In the induction group, 33% and 32% of women with a favorable or unfavorable cervix, respectively, had high-risk situations. In the expectant management group, 39% and 49%, respectively, experienced high-risk situations. The risk of CD was comparable after induction for women with (14.6%) and without (14.8%) a favorable cervix. In women managed expectantly, the risks of CD were 18.2% and 21.1%, respectively, for women with a favorable or unfavorable cervix. No increase was found in risks of adverse neonatal outcomes when women with and without a favorable cervix in each treatment group were compared.

This post hoc analysis showed that the likelihood of developing high-risk situations depends on the level of cervical ripeness only when women are managed expectantly. If labor is induced, the development of high-risk situations is not associated with cervical ripeness. A comparable risk of CD after labor induction occurs in women with a favorable or unfavorable cervix. Women with an unripe cervix are at increased risk for CD with expectant management. Despite more monitoring and earlier detection of deterioration in the induction of labor group, more women developed high-risk situations in the expectant management group.