Evidence

Surveillance report 2017 – Hypertension in pregnancy: diagnosis and management (2010) NICE guideline CG107

Commentary on selected new evidence

Commentary on selected new evidence

With advice from topic experts we selected 1 study for further commentary.

Advice and follow-up care at transfer to community care – risk of recurrence

We selected the individual patient data meta-analysis by van Oostwaard et al. (2015) for a full commentary because it is a large, good quality, individual patient data meta-analysis which has a potential impact on guideline recommendations and provides useful information for women and clinicians.

What the guideline recommends

NICE guideline CG107 recommends telling women who had gestational hypertension that their risk of developing:

  • gestational hypertension in a future pregnancy ranges from about 1 in 6 (16%) pregnancies to about 1 in 2 (47%) pregnancies

  • pre-eclampsia in a future pregnancy ranges from 1 in 50 (2%) to about 1 in 14 (7%) pregnancies.

NICE guideline CG107 also recommends telling women who had pre-eclampsia that their risk of developing:

  • gestational hypertension in a future pregnancy ranges from about 1 in 8 (13%) pregnancies to about 1 in 2 (53%) pregnancies

  • pre-eclampsia in a future pregnancy is up to about 1 in 6 (16%) pregnancies

  • pre-eclampsia in a future pregnancy is about 1 in 4 (25%) pregnancies if their pre-eclampsia was complicated by severe pre-eclampsia, HELLP syndrome or eclampsia and led to birth before 34 weeks, and about 1 in 2 (55%) pregnancies if it led to birth before 28 weeks.

Methods

van Oostwaard et al. (2015) conducted an IPD meta-analysis aiming to calculate the risk of recurrence of hypertensive disorders of pregnancy (HDP), defined in the paper as:

  • Pre-eclampsia: hypertension (diastolic blood pressure at least 90 mm Hg or systolic blood pressure at least 140 mm Hg on 2 occasions that were 4 to 5 hours apart) in combination with proteinuria (a positive [0.3g/L] proteinuria dipstick test, a protein/creatinine ratio of at least 30 mg/mmol in a random sample or a urine protein excretion of at least 300 mg for 24 hours) after 20 weeks' gestation.

  • Gestational hypertension: hypertension at later than 20 weeks' gestation without proteinuria or a significant rise in blood pressure (if a woman had known chronic hypertension).

  • Superimposed pre-eclampsia: women with chronic hypertension and proteinuria or a sudden increase in proteinuria if already present.

  • Hemolysis, elevated liver enzymes and low platelets (HELLP) syndrome: (elevated lactate dehydrogenase levels [at least 600 U/L], elevated liver enzymes by levels of aspartate transaminase or alanine transferase at least 70 U/L, and low platelets less than 100,000/mm).

  • Small for gestational age (SGA) was defined as birthweight below 10th percentile, and was adjusted for gestational age based on a local reference population.

Inclusion criteria

There were no reported study design inclusion criteria, except that case control studies were required to report recurrence. All studies needed to include women with a history of a hypertensive disorder, as defined above, that resulted in a delivery at any gestational age, and a subsequent pregnancy. Studies indicating that recurrence of pre-eclampsia was reported in the original data were eligible, even if this was not available in the publication.

Data collection

Data were collected for eligible studies from consenting authors from either the original database or a separate file. The authors stated that data were included for a subsequent pregnancy following the hypertensive pregnancy. Data were collected on various demographic characteristics at the index pregnancy, including:

  • age

  • body mass index

  • cardiovascular risk factors

  • the clinical syndrome for the index pregnancy.

Outcomes

The primary outcome was the recurrence of any HDP in the next subsequent pregnancy, and the secondary outcomes were recurrences of individual hypertensive disorders.

Results

In total, 22 studies were included covering a total of 99,415 women. The majority of women were included in 3 large retrospective cohort studies (n=88,067), none of which were included in NICE guideline CG107.

Overall and individual recurrence rates

  • The recurrence rate of any HDP following any HDP at the index pregnancy was 20.7% (95% confidence interval [CI], 20.4 to 20.9%) and:

    • 8.6% for gestational hypertension (95% CI 8.4 to 8.8)

    • 13.8% for pre-eclampsia (95% CI 13.6 to 14.1)

    • 0.2% for HELLP (95% CI 0.16 to 0.25)

    • 3.4% for SGA (95% CI 3.2 to 3.6).

  • For gestational hypertension at the index pregnancy, there was a:

    • 21.5% risk of any HDP

    • 14.5% recurrence rate of gestational hypertension

    • 7.1% risk of pre-eclampsia

    • 0.1% risk of HELLP

    • 3.6% risk of SGA.

  • For pre-eclampsia at the index pregnancy, there was a:

    • 20.4% risk of any HDP

    • 6% risk of gestational hypertension

    • 16% recurrence of pre-eclampsia

    • 0.2% risk of HELLP

    • 3.3% risk of SGA.

  • For HELLP syndrome at the index pregnancy, there was a:

    • 36.3% risk of any HDP

    • 18.4% risk of gestational hypertension

    • 17.8% risk of pre-eclampsia

    • 7.2% recurrence of HELLP

    • 5.9% risk of SGA.

  • For SGA accompanying HDP at the index pregnancy, there was a:

    • 22.2% risk of any HDP

    • 12.9% risk of gestational hypertension

    • 14.3% risk of pre-eclampsia

    • 0.6% risk of HELLP

    • 6.6% recurrence of SGA.

Premature delivery

Premature delivery in addition to recurrent hypertensive disorder occurred at:

  • less than 37 weeks' gestation in 3,316 women (3.3%; 95% CI 3.2 to 3.5)

  • less than 34 weeks' gestation in 1,224 women (1.2%; 95% CI 1.2 to 1.3)

  • less than 28 weeks gestation in 179 women (0.18%; 95% CI 0.16 to 0.22).

Sensitivity analyses

The authors conducted sensitivity analyses to address risk of bias, sample size and study design. From the analyses, the authors reported similar associations between the index and subsequent pregnancy, for both overall recurrence of HDP and for individual disorders. However, no quantitative data were reported for any of the 3 analyses.

Reasons for refraining from subsequent pregnancy

The authors stated that from three included databases reporting information about the reason for refraining from a subsequent pregnancy, 140 (29%) of 471 women refrained because of high perceived risk of recurrence of HDP.

Strengths and limitations

Strengths

  • Although mild and severe hypertension were not distinguished in the majority of included studies, the definitions used for gestational hypertension and pre-eclampsia in the meta-analysis were consistent with those in NICE guideline CG107.

  • The results were based on the largest published database of recurrence of HDP to date. The majority of included studies, including those with the largest sample sizes, were not covered by NICE guideline CG107.

  • The use of individual patient data enabled the analysis of a very large sample size and consequently reduced the risk of bias inherent in single studies.

  • To address significant heterogeneity, a random effects model was used and three post hoc sensitivity analyses were performed for:

    • smaller or larger studies (less than or more than 200 participants)

    • exclusion of retrospective studies

    • exclusion of studies with a high or unclear risk of bias.

  • In addition to risk of recurrence, the aggregated data allowed the calculation of separate risk factors on an individual level, and indicated reasons for refraining from a next pregnancy. This may further assist in the provision of advice and follow up care.

Limitations

  • The primary outcome of the meta-analysis was stated as recurrence at the next subsequent pregnancy, as distinct from any subsequent pregnancy, which was used in NICE guideline CG107. This may limit the applicability to the guideline. Furthermore, the terminology describing subsequent pregnancy was inconsistent, and it is unclear whether the original patient data covered next or any subsequent pregnancies.

  • Confidence intervals were not reported for the recurrence rates of individual disorders.

  • The search strategies for study identification were not reported in full, and the reported terms were inadequate. For example, pre-eclampsia was the only population term listed.

  • Not all required data was available or accessible, resulting in the inclusion of only 22 of 88 (25%) eligible studies.

  • It is unclear whether more than one reviewer was involved in the data checking and quality assessment of included studies.

  • There were no quantitative results presented for the sensitivity analyses. The authors also acknowledged that the sensitivity analyses based on study design and size showed some conflicting results, due to loss of power.

  • Data were combined from heterogeneous study designs, settings, and populations, which led to a very wide range of recurrence rates of 6% to 83%. The authors did not state whether definitions of individual HDP differed between included studies.

  • Prophylactic trials were included in the analysis, which may have influenced the overall results where the interventions were effective in reducing recurrence. However these studies had only small sample sizes, which may lower the risk of confounding the data.

  • Reporting bias was acknowledged as a limitation by the authors, such as in the reporting of eclampsia.

  • Variables that were seldom registered limited the analyses of the data. For example, HELLP syndrome, SGA accompanying HDP, and multiple pregnancies were rarely registered.

Impact on guideline

The meta-analysis indicates that for gestational hypertension at the index pregnancy, which relates to recommendation 1.10.4.1 in NICE guideline CG107, there was a 14.5% recurrence rate of gestational hypertension and a 7.1% recurrence of pre-eclampsia. Recommendation 1.10.4.1 advises telling women with gestational hypertension that recurrence rate of gestational hypertension is 16%-47% and recurrence rate of pre-eclampsia is 2%-7%. There is a potential impact on the guideline to review the recommended recurrence rates for both gestational hypertension and pre-eclampsia.

For pre-eclampsia at the index pregnancy, which relates to recommendation 1.10.4.2 in NICE guideline CG107, there was a 6% recurrence rate of gestational hypertension, a 16% recurrence of pre-eclampsia and 0.2% HELLP. Recommendation 1.10.4.2 advises telling women who had pre-eclampsia that recurrence rate of gestational hypertension is 13%–53% and recurrence rate of pre-eclampsia is 16%. There is a potential impact on the guideline to review the recommended recurrence rate for gestational hypertension (6% vs 13-53%) but not for pre-eclampsia (16% vs 16%).

The recurrence rates established in the evidence review for NICE guideline CG107 were based on a qualitative synthesis of 5 studies. The 3 largest (n=1943) of these 5 studies were also included in the meta-analysis by van Oostwaard et al. (2015), along with 19 other studies. However, the vast majority of patient data used in the meta-analysis was derived from 3 large registry based studies (n=88,067) that were not included in NICE guideline CG107. It should be noted, though, that the primary outcome of the IPD meta-analysis was recurrence at the next subsequent pregnancy, as distinct from the outcome of recurrence at any subsequent pregnancy, which was used in the guideline.


This page was last updated: