Do different settings and techniques for measuring blood pressure during pregnancy help to improve outcomes for women and babies?

What is the issue?

Regular blood pressure (BP) measurements are crucial during pregnancy for the diagnosis and management of high BP. BP can be measured in various settings (e.g. self-measurement at home versus in clinic) and using different techniques (e.g. measurement based on different blood flow sounds). They may have different effects on diagnosing and monitoring high BP, and reducing the risk of serious illness or death in both woman and baby.

Why is this important?

If high BP in pregnancy is not detected and managed in a timely fashion, serious complications can develop. This review is needed to establish the benefits and risks of these settings and techniques for women and their babies.

What evidence did we find?

We searched for evidence from randomised controlled trials in April 2020, and identified three studies (involving 536,607 women). Overall, the studies were conducted in such a way that we are not certain of the findings, mainly due to the small size of two of the studies and the design of the other.

One study (154 women) compared BP settings in the UK: self-monitoring at home versus usual measurement in clinic. The other two studies compared BP techniques: one (220 women) compared two different blood flow sounds to determine diastolic BP (the bottom number) in Australia, and the other (536,233 deliveries) investigated the introduction of a semi-automated BP monitor and an education package (CRADLE intervention) compared with usual care across Africa, India and Haiti.

None of the studies measured high BP, the number of women admitted to hospital before birth, how long babies stayed in the neonatal unit, or what extra help babies received for their breathing.

Self-measurement of BP at home compared with usual BP measurement in clinic

Self-monitoring BP may lead to more women being diagnosed with pre-eclampsia compared with usual care but the evidence is uncertain.

We are uncertain if self-monitoring BP increases the likelihood of stillbirth, baby deaths (after birth), women giving birth early, or women admitted to the intensive care unit.

Self-monitoring BP may have little to no effect on the likelihood of women having their labour induced compared with usual care.

Self-monitoring BP may lead to slightly more newborns being admitted to a neonatal unit compared with usual care.

This trial had no maternal deaths, and did not report the number of baby deaths, before or shortly after birth.

Measuring BP using different blood flow sounds - Korotkoff phase IV (K4, softer, muffled sound) compared with Korotkoff phase V (K5, when the sound disappears) to measure diastolic BP

There may be little to no difference between using K4 or K5 to diagnose pre-eclampsia; the evidence is uncertain.

We are uncertain if there is an effect on baby deaths, before or shortly after birth.

This trial had no maternal deaths, and did not report the number of women admitted to intensive care, women who needed their labour induced, women giving birth early, stillbirths, baby deaths (after birth), or babies admitted to the neonatal unit.

CRADLE intervention (semi-automated BP monitor and an education package) compared with usual care

The CRADLE BP monitor may make little or no difference to the risk of maternal death.

The trial did not report the number of women with pre-eclampsia, women who needed their labour induced, women giving birth early, baby deaths (before and after birth), or the number of babies admitted to the neonatal unit. The number of women admitted to intensive care, stillbirths, and baby deaths (after birth) were only reported for a subgroup of women in this trial, so we did not include these results.

What does this mean?

More evidence is needed on whether self-monitoring BP in pregnant women with high BP is beneficial, because the study exploring this was small.

Current practice of using K5 (no blood flow sound) to measure diastolic BP is supported in pregnant women with high BP.

The trial using the CRADLE device to monitor BP in pregnancy had limitations in its design, and we are uncertain about its benefit.

Authors' conclusions: 

The benefit, if any, of self-monitoring BP in hypertensive pregnancies remains uncertain, as the evidence is limited to one feasibility study. Current practice of using K5 to measure diastolic BP is supported for women with pregnancy hypertension. The benefit, if any, of using the CRADLE device to measure BP in pregnancy remains uncertain, due to the limitations and instability of the trial study design.

Read the full abstract...
Background: 

Regular blood pressure (BP) measurement is crucial for the diagnosis and management of hypertensive disorders in pregnancy, such as pre-eclampsia. BP can be measured in various settings, such as conventional clinics or self-measurement at home, and with different techniques, such as using auscultatory or automated BP devices. It is important to understand the impact of different settings and techniques of BP measurement on important outcomes for pregnant women.

Objectives: 

To assess the effects of setting and technique of BP measurement for diagnosing hypertensive disorders in pregnancy on subsequent maternal and perinatal outcomes, women's quality of life, or use of health service resources.

Search strategy: 

We searched Cochrane Pregnancy and Childbirth’s Trials Register, ClinicalTrials.gov, the WHO International Clinical Trials Registry Platform (ICTRP) on 22 April 2020, and reference lists of retrieved studies.

Selection criteria: 

Randomised controlled trials (RCTs) involving pregnant women, using validated BP devices in different settings or using different techniques.

Data collection and analysis: 

Two authors independently extracted data, assessed risk of bias, and used the GRADE approach to assess the certainty of the evidence.

Main results: 

Of the 21 identified studies, we included three, and excluded 11; seven were ongoing. Of the three included RCTs (536,607 women), one was a cluster-RCT, with a substantially higher number of participants (536,233 deliveries) than the other two trials, but did not provide data for most of our outcomes. We generally judged the included studies at low risk of bias, however, the certainty of the evidence was low, due to indirectness and imprecision. Meta-analysis was not possible because each study investigated a different comparison.

None of the included studies reported our primary outcome of systolic BP greater than or equal to 150 mmHg. None of the studies reported any of these important secondary outcomes: antenatal hospital admissions, neonatal unit length of stay, or neonatal endotracheal intubation and use of mechanical ventilation.

Setting of BP measurement: self-measurement versus conventional clinic measurement (one study, 154 women)

There were no maternal deaths in either the self-monitoring group or the usual care group. The study did not report perinatal mortality.

Self-monitoring may lead to slightly more diagnoses of pre-eclampsia compared with usual care (risk ratio (RR) 1.49, 95% confidence interval (CI) 0.87 to 2.54; 154 women; 1 study; low-certainty evidence) but the wide 95% CI is consistent with possible benefit and possible harm.

Self-monitoring may have little to no effect on the likelihood of induction of labour compared with usual care (RR 1.09, 95% CI 0.82 to 1.45; 154 women; 1 study; low-certainty evidence).

We are uncertain if self-monitoring BP has any effect on maternal admission to intensive care (RR 1.54, 95% CI 0.06 to 37.25; 154 women; 1 study; low-certainty evidence), stillbirth (RR 2.57, 95% CI 0.13 to 52.63; 154 women; 1 study; low-certainty evidence), neonatal death (RR 1.54, 95% CI 0.06 to 37.25; 154 women; 1 study; low-certainty evidence) or preterm birth (RR 1.15, 95% CI 0.37 to 3.55; 154 women; 1 study; low-certainty evidence), compared with usual care because the certainty of evidence is low and the 95% CI is consistent with appreciable harms and appreciable benefits.

Self-monitoring may lead to slightly more neonatal unit admissions compared with usual care (RR 1.53, 95% CI 0.65 to 3.62; 154 women; 1 study; low-certainty evidence) but the wide 95% CI includes the possibility of slightly fewer admissions with self-monitoring.

Technique of BP measurement: Korotkoff phase IV (K4, muffling sound) versus Korotkoff phase V (K5, disappearance of sound) to represent diastolic BP (one study, 220 women)

There were no maternal deaths in either the K4 or K5 group.

There may be little to no difference in the diagnosis of pre-eclampsia between using K4 or K5 for diastolic BP (RR 1.16; 95% CI 0.89 to 1.49; 1 study; 220 women; low-certainty evidence), since the wide 95% CI includes the possibility of more diagnoses with K4.

We are uncertain if there is a difference in perinatal mortality between the groups because the quality of evidence is low and the 95% CI is consistent with appreciable harm and appreciable benefit (RR 1.14, 95% CI 0.16 to 7.92; 1 study, 220 women; low-certainty evidence).

The trial did not report data on maternal admission to intensive care, induction of labour, stillbirth, neonatal death, preterm birth, or neonatal unit admissions.

Technique of BP measurement: CRADLE intervention (CRADLE device, a semi-automated BP monitor with additional features, and an education package) versus usual care (one study, 536,233 deliveries)

There may be little to no difference between the use of the CRADLE device and usual care in the number of maternal deaths (adjusted RR 0.80, 95% CI 0.30 to 2.11; 536,233 women; 1 study; low-certainty evidence), but the 95% CI is consistent with appreciable harm and appreciable benefit.

The trial did not report pre-eclampsia, induction of labour, perinatal mortality, preterm birth, or neonatal unit admissions. Maternal admission to intensive care and perinatal outcomes (stillbirths and neonatal deaths) were only collected for a small proportion of the women, identified by an outcome not by baseline characteristics, thereby breaking the random allocation. Therefore, any differences between the groups could not be attributed to the intervention, and we did not extract data for these outcomes.

Share/Save