Planned early delivery versus expectant management for hypertensive disorders from 34 weeks gestation to term

Background Hypertensive disorders in pregnancy are significant contributors to maternal and perinatal morbidity and mortality. These disorders include well‐controlled chronic hypertension, gestational hypertension (pregnancy‐induced hypertension) and mild pre‐eclampsia. The definitive treatment for these disorders is planned early delivery and the alternative is to manage the pregnancy expectantly if severe uncontrolled hypertension is not present, with close maternal and fetal monitoring. There are benefits and risks associated with both, so it is important to establish the safest option. Objectives To assess the benefits and risks of a policy of planned early delivery versus a policy of expectant management in pregnant women with hypertensive disorders, at or near term (from 34 weeks onwards). Search methods We searched Cochrane Pregnancy and Childbirth Trials Register (12 January 2016) and reference lists of retrieved studies. Selection criteria Randomised trials of a policy of planned early delivery (by induction of labour or by caesarean section) compared with a policy of delayed delivery ("expectant management") for women with hypertensive disorders from 34 weeks' gestation. Cluster‐randomised trials would have been eligible for inclusion in this review, but we found none. Studies using a quasi‐randomised design are not eligible for inclusion in this review. Similarly, studies using a cross‐over design are not eligible for inclusion, because they are not a suitable study design for investigating hypertensive disorders in pregnancy. Data collection and analysis Two review authors independently assessed eligibility and risks of bias. Two review authors independently extracted data. Data were checked for accuracy. Main results We included five studies (involving 1819 women) in this review. There was a lower risk of composite maternal mortality and severe morbidity for women randomised to receive planned early delivery (risk ratio (RR) 0.69, 95% confidence interval (CI) 0.57 to 0.83, two studies, 1459 women (evidence graded high)). There were no clear differences between subgroups based on our subgroup analysis by gestational age, gestational week or condition. Planned early delivery was associated with lower risk of HELLP syndrome (RR 0.40, 95% CI 0.17 to 0.93, 1628 women; three studies) and severe renal impairment (RR 0.36, 95% CI 0.14 to 0.92, 100 women, one study). There was not enough information to draw any conclusions about the effects on composite infant mortality and severe morbidity. We observed a high level of heterogeneity between the two studies in this analysis (two studies, 1459 infants, I2 = 87%, Tau2 = 0.98), so we did not pool data in meta‐analysis. There were no clear differences between subgroups based on our subgroup analysis by gestational age, gestational week or condition. Planned early delivery was associated with higher levels of respiratory distress syndrome (RR 2.24, 95% CI 1.20 to 4.18, three studies, 1511 infants), and NICU admission (RR 1.65, 95% CI 1.13 to 2.40, four studies, 1585 infants). There was no clear difference between groups for caesarean section (RR 0.91, 95% CI 0.78 to 1.07, 1728 women, four studies, evidence graded moderate), or in the duration of hospital stay for the mother after delivery of the baby (mean difference (MD) ‐0.16 days, 95% CI ‐0.46 to 0.15, two studies, 925 women, evidence graded moderate) or for the baby (MD ‐0.20 days, 95% CI ‐0.57 to 0.17, one study, 756 infants, evidence graded moderate). Two fairly large, well‐designed trials with overall low risk of bias contributed the majority of the evidence. Other studies were at low or unclear risk of bias. No studies attempted to blind participants or clinicians to group allocation, potentially introducing bias as women and staff would have been aware of the intervention and this may have affected aspects of care and decision‐making. The level of evidence was graded high (composite maternal mortality and morbidity), moderate (caesarean section, duration of hospital stay after delivery for mother, and duration of hospital stay after delivery for baby) or low (composite infant mortality and morbidity). Where the evidence was downgraded, it was mostly because the confidence intervals were wide, crossing both the line of no effect and appreciable benefit or harm. Authors' conclusions For women suffering from hypertensive disorders of pregnancy after 34 weeks, planned early delivery is associated with less composite maternal morbidity and mortality. There is no clear difference in the composite outcome of infant mortality and severe morbidity; however, this is based on limited data (from two trials) assessing all hypertensive disorders as one group. Further studies are needed to look at the different types of hypertensive diseases and the optimal timing of delivery for these conditions. These studies should also include infant and maternal morbidity and mortality outcomes, caesarean section, duration of hospital stay after delivery for mother and duration of hospital stay after delivery for baby. An individual patient meta‐analysis on the data currently available would provide further information on the outcomes of the different types of hypertensive disease encountered in pregnancy