Hypertension in pregnancy: diagnosis and management

Why the committee made the recommendations

The committee were aware that there is often over-reliance on a proteinuria result in the diagnosis of pre-eclampsia, and agreed that healthcare professionals should use the results of a full clinical review, including severity of hypertension and other signs and symptoms, before making a diagnosis of pre-eclampsia.

The committee amended the recommendation on automated dipstick tests from the 2010 guideline to emphasise that this should be used as a screening tool for proteinuria. The committee highlighted the importance of using automated dipstick analysis in secondary care rather than visual analysis, which they were aware from their experience has a higher error rate.

Protein:creatinine ratio (PCR) and albumin:creatinine ratio (ACR) were both shown to have high specificity and high sensitivity at the chosen thresholds (30 mg/mmol and 8 mg/mmol respectively), and therefore either could be used depending on local availability. The committee agreed that using both tests together did not have any additional diagnostic benefit.

There was some evidence that using the first morning urine void in assessment of proteinuria can lead to lower diagnostic accuracy, and so the committee recommended against using this.

As PCR and ACR show very high diagnostic accuracy, they should be used in place of 24‑hour urine collection, which is awkward for women and could delay identification of proteinuria. However, there are rare occasions when it might be more appropriate to use 24‑hour collection (for example, women with renal complications), and so the committee agreed it should not be ruled out entirely.

There was good evidence that a PCR of 30 mg/mmol had good diagnostic accuracy, showing high sensitivity and specificity and should be used as the threshold for significant proteinuria. However, the committee recommended retesting for results above 30 mg/mmol if there is still diagnostic uncertainty (for example, the woman has no other clinical signs or symptoms of pre-eclampsia) because there is large variation in protein excretion during the day and from day to day. The committee agreed that this would prevent women being diagnosed with pre-eclampsia on the basis of a single raised PCR result.

Evidence from a single study showed high sensitivity and specificity for an ACR result of 8 mg/mmol to diagnose proteinuria. However, the committee were also aware of further results from a large, UK‑based study, which provided further evidence for the efficacy of a threshold of 8 mg/mmol in the diagnosis of severe pre-eclampsia. The committee were aware that this threshold is different to that used for detection of microalbuminuria in the non-pregnant population. However, they agreed that, on the basis of the evidence reviewed, it was appropriate to use a threshold of 8 mg/mmol for pregnant women.

As with PCR, the committee were aware that women are sometimes diagnosed with pre-eclampsia on the basis of a single raised ACR, and that this may lead to over-diagnosis. Therefore, they made a recommendation to consider repeating the ACR measurement if there was ongoing clinical uncertainty about the diagnosis.

No evidence was reviewed that examined the timing of repeat testing for either ACR or PCR, and so no recommendations could be made regarding this.

How the recommendations might affect practice

The recommendation to take account of other clinical features when assessing women for suspected pre-eclampsia might lead to an increased need for follow‑up and surveillance. However, this will also reduce the chance that a diagnosis of pre-eclampsia is missed by raising awareness of the multi-system nature of the disease, and so could reduce the number of women who go on to develop complications from undiagnosed pre-eclampsia.

Not all secondary care units currently use automated dipstick analysis to screen for proteinuria, so the recommendations might increase the need for automated reagent-strip reading devices. However, the accuracy and reliability of screening will be improved, reducing the need for further investigations for some women and correctly identifying more women who need further testing or investigations.

Moving from 24‑hour urine collection to spot urine ACR or PCR will save time, with potential for faster diagnosis, and a reduction in inaccuracies because of incomplete samples. It is also likely to improve quality of life, as the process of completing a 24‑hour urine collection is time-consuming and awkward.

A PCR of 30 mg/mmol is already used routinely as a diagnostic threshold and therefore should not change practice. Currently units may use different ACR levels for diagnosis and so the recommendation to use 8 mg/mmol will standardise practice.

Recommendations have been made for the use of either ACR or PCR allowing local decisions to use whichever test is available, so this should not affect practice.

Repeating the PCR or ACR test may incur a small additional cost. However, this should reduce the false positive rate, and mean some women will avoid unnecessary follow‑up or intensive monitoring (such as hospital admission) if their proteinuria resolves and is shown to be transient.

Return to recommendations

Why the committee made the recommendations

The committee agreed that pregnant woman with chronic hypertension should be offered lifestyle advice similar to other adults with hypertension, and in line with the NICE guideline on hypertension in adults.

There was very little evidence available on treatment initiation thresholds for chronic hypertension in pregnancy, so the committee based their recommendations on the values specified in the recent Control of Hypertension in Pregnancy Study (CHIPS) and the NICE guideline on hypertension in adults. There was evidence for target blood pressure levels from the large CHIPS trial, so the committee made recommendations based on this.

There was some very limited evidence of both benefits and harms for different antihypertensive medicines. However, there was not enough evidence to recommend one treatment over another. As labetalol, nifedipine and methyldopa had been recommended in the previous guideline (for gestational hypertension and pre-eclampsia), and these medicines had been used for many years in pregnancy, the committee agreed they should be preferred treatment options for chronic hypertension in pregnancy. Labetalol is specifically licensed for use in pregnancy and so is suggested as the first-line option, with nifedipine as the next alternative, and methyldopa as the third option (as it may lead to more side effects and be the least effective option of the 3).

There was some very limited evidence for the benefits of aspirin in reducing preterm births and neonatal unit admissions, so the committee retained the recommendation on aspirin from the previous guideline, but incorporated it into the section on the treatment of chronic hypertension in pregnancy. The committee noted that the studies used different doses of aspirin (ranging from 50 mg to 150 mg daily), and that common practice in the UK was to offer 75 mg to 150 mg, therefore this dose range was recommended.

The committee were aware of the link between chronic hypertension and both pre-existing and gestational diabetes, therefore they made an overarching recommendation at the beginning of the guideline to cross-refer to the existing NICE guideline on diabetes in pregnancy.

The committee made a new recommendation referring to the NICE diagnostics guidance on placental growth factor (PLGF) testing as this may be applicable to women with chronic hypertension.

As there is currently a lack of evidence on the difference in outcomes between different antihypertensive medications, and concerns about possible adverse neonatal events from beta blockers, the committee made research recommendations on these topics.

How the recommendations might affect practice

Based on these recommendations, a clear blood pressure target should now be set for women with chronic hypertension in pregnancy who need antihypertensive treatment to improve consistency of treatment targets.

Starting treatment for hypertension and offering aspirin to women with chronic hypertension who are pregnant are standard care, so these recommendations are not expected to change practice significantly.

Return to recommendations

Why the committee made the recommendations

The committee updated the table from the previous guideline on the management of pregnancy with gestational hypertension. There was very little evidence available on treatment initiation thresholds for gestational hypertension in pregnancy, so the committee made recommendations using the values specified in the recent Control of Hypertension in Pregnancy Study (CHIPS). There was evidence for target blood pressure levels from the large CHIPS trial, so the committee made recommendations based on this. The committee made a new recommendation referring to the NICE diagnostics guidance on placental growth factor (PLGF) testing as this is applicable to women with gestational hypertension.

The committee were aware of the link between gestational hypertension and both pre-existing and gestational diabetes, therefore they made an overarching recommendation at the beginning of the guideline to cross-refer to the existing NICE guideline on diabetes in pregnancy.

There was no evidence on fetal monitoring in gestational hypertension, so the committee made a research recommendation on the optimal fetal monitoring strategy to detect small for gestational age infants.

How the recommendations might affect practice

The recommendations reflect current clinical practice in many units, but may help standardise practice across the NHS for units that currently use other blood pressure targets.

Return to recommendations

Why the committee made the recommendations

The committee agreed, based on their clinical expertise, that women with pre-eclampsia should have a full clinical assessment at every antenatal appointment and should be admitted to hospital if there are any concerns about the wellbeing of the woman or her baby. The committee agreed that this would include (but was not limited to) reasons such as severe hypertension or other features of pre-eclampsia indicating increased risk of adverse outcomes.

There was some evidence that the fullPIERS and PREP‑S models can help identify women at different risks of adverse outcomes because of pre-eclampsia. There was more extensive validation of the fullPIERS model, and the validation studies were conducted in populations from a range of healthcare settings. The PREP‑S model had been developed using a UK population, and validated using data from similar multinational settings. It was noted that further validation of PREP‑S was unlikely to be conducted, because of the cost of conducting these studies. The committee therefore agreed that both models could be considered as options, in addition to a full clinical assessment, to help guide decisions relating to interventions and place of care.

The tools predict adverse outcomes in women, but are not designed to predict outcomes for babies. The committee agreed it was important to highlight this.

How the recommendations might affect practice

The use of a full clinical assessment and validated models to predict risk may improve consistency in current practice with regard to admission to hospital for women with pre-eclampsia. Some centres offer admission to all women with pre-eclampsia, whereas others only offer it to a small proportion of women. The guidance might increase the number of women who are admitted to hospital in some centres if admission is not currently routine, but might decrease admission in other centres, thus standardising practice.

Return to recommendations

Why the committee made the recommendations

The committee updated the table from the previous guideline on the management of pregnancy with pre-eclampsia. There was limited evidence on the best place of treatment for women with pre-eclampsia. Because of this, the committee made recommendations based on other evidence they reviewed (see evidence review C), which showed that women should be admitted if there were concerns for the wellbeing of the woman or her baby, or a high risk of complications of pre-eclampsia predicted using the fullPIERS or PREP‑S model. (See the section of the guideline on assessing pre-eclampsia and evidence review C for more details on the use of the fullPIERS and PREP‑S models.)

There was no evidence on treatment initiation thresholds or target blood pressure levels for pre-eclampsia, so the committee based their recommendations on the NICE guideline on hypertension in adults and the values specified in the Control of Hypertension In Pregnancy Study (CHIPS; see evidence review A), which included women with chronic or gestational hypertension.

There was some very limited evidence of both benefits and harms for different pharmacological interventions. However, as there was not enough evidence to recommend one treatment over another, the committee adopted the choices from the previous guideline and recommended choosing a treatment based on previous treatments, side-effect profiles and the woman's preferences. Labetalol is specifically licensed for use in pregnancy and so is suggested as the first-line option, with nifedipine as the next alternative, and methyldopa as the third option (as it may lead to more side effects and be the least effective option of the 3).

There was limited evidence on the benefits and harms of planned early birth compared with expectant management of pregnancy in women with pre-eclampsia, so the committee recommended that decisions about timing of birth should be based on whether the woman and baby are at risk of adverse outcomes if pregnancy is prolonged. These recommendations were based on those from the previous guideline, and expanded based on international guidelines, which were used by the committee in their clinical practice. Based on the data from HYPITAT‑II study, the committee also agreed that pregnancies in women with pre-eclampsia could be managed with continued surveillance to 37 weeks, unless there were specific concerns or indications to offer a planned early birth before then.

There was limited evidence to guide the best place of care for women with pre-eclampsia and their babies, so the committee made a research recommendation.

How the recommendations might affect practice

The recommendations are in line with current best clinical practice, so are unlikely to cause a significant change in practice.

Currently, some units admit all women with pre-eclampsia routinely, some only admit women who they believe to be at a high risk of complications, and some admit very few. Standardising practice could therefore increase or reduce the number of women who will be admitted, depending on a unit's current practice, but is likely to reduce unwanted variance between units.

Return to recommendations

Why the committee made the recommendations

There was very little evidence on the efficacy and safety of antihypertensive agents in postnatal women, so the committee made recommendations based on the NICE guideline on hypertension in adults, with adaptations based on the potential effects of medicines on the baby. The committee therefore recommended the use of an angiotensin converting enzyme (ACE) inhibitor as first-line treatment, except in women of African or Caribbean family origin, in whom a calcium-channel blocker would be used first line. The choice of second-line medicine was modified from the NICE guideline on hypertension in adults as angiotensin receptor blockers, thiazide and thiazide-like diuretics are not recommended during breastfeeding. Therefore, the committee agreed that beta-blockers should be used as the second-line antihypertensive agent. The committee also agreed that the medicines with once-daily administration should be used wherever possible and for this reason the committee recommended enalapril in preference to captopril (which is taken 3 times daily) and atenolol as an alternative to labetalol (which is taken 2 to 4 times daily).

Based on their experience, the committee made recommendations on advice for women who wish to breastfeed while taking antihypertensives, and on the monitoring of babies whose mothers are taking antihypertensives.

As there was very little evidence on the effectiveness and safety of antihypertensives for postnatal use, the committee revised the research recommendation made in the 2010 guideline.

How the recommendations might affect practice

There is currently wide variation in practice over use of antihypertensive treatment in the postnatal period, and these recommendations may reduce variation in practice. The recommendations could lead to an increase in the use of atenolol instead of labetalol in the postnatal period.

Return to recommendations

Why the committee made the recommendations

Long-term follow‑up studies of women who have experienced hypertensive disorders during pregnancy showed an increased risk of long-term cardiovascular disease and a higher prevalence of hypertensive disorders in subsequent pregnancies compared with women unaffected by hypertensive disorders.

There was no evidence on which interventions could reduce the risk of recurrence of hypertensive disorders of pregnancy or future cardiovascular disease, so the committee made a research recommendation.

How the recommendations might affect practice

Providing guidance and advice to women on future risks and signposting appropriate care and lifestyle advice may be an additional activity for some healthcare professionals, compared with current practice.

Return to recommendations