Risk assessment early in pregnancy

Before developing an antenatal care plan, women should be assessed for the factors listed in box 2. From meta-analysis and systematic review,8 the unadjusted relative risks of developing pre-eclampsia were: presence of antiphospholipid antibodies (9.72, 95% confidence interval 4.34 to 21.75), history of pre-eclampsia (7.19, 5.85 to 8.83), pre-existing diabetes (3.56, 2.54 to 4.99), multiple pregnancy (2.93, 2.04 to 4.21), nulliparity (2.91, 1.28 to 6.61), family history of pre-eclampsia (2.90, 1.70 to 4.93), women aged ≥ 40 (nulliparous women, 1.68, 1.23 to 2.29; multiparous women, 1.96, 1.34 to 2.87), and a raised body mass index at booking (1.55, 1.28 to 1.88). The risk of pre-eclampsia is also increased with pre-existing hypertension and renal disease, a pregnancy interval of ≥ 10 years, a raised diastolic blood pressure at booking, and confirmed proteinuria.9 The data did not show an increased risk for young women of ≥ 19, ≥ 17, or ≥ 16.

For the continuous variables, such as age and body mass index, we selected conservative thresholds for action available from the data. Below these cut-off points there is still an increased risk of pre-eclampsia. Data were insufficient to calculate absolute risk for each factor, to see how two factors interact, or to comment on migraine or change of partner. We did not consider donor egg and donor insemination.

Referral in early pregnancy for specialist input

Women should be offered specialist input before 20 weeks if they have one of the criteria listed in box 3. Input may concern further specialist investigation, clarification of risk, or advice on early intervention or pharmacological treatment. We do not prescribe subsequent obstetric care, which will be determined on an individual basis and may be led by specialists, general practitioners, or midwives, or by shared care.

Previous pre-eclampsia is associated with higher rates of moderate, severe, and early onset pre-eclampsia and adverse perinatal outcomes associated with preterm delivery.10 Recurrent pre-eclampsia occurs, on average, between zero and four weeks later than in the first pregnancy. We recommend that women who have asymptomatic proteinuria at booking, if persistent or confirmed by a 24 hour sample, be investigated for possible underlying renal disease (itself a risk factor for pre-eclampsia) or other conditions to accurately determine risk.

Data were lacking on the effect of two predisposing factors on the overall likelihood of developing pre-eclampsia. We recommend that women with two such factors be referred for early specialist input, individual assessment, and discussion of obstetric risk.

Community monitoring after 20 weeks' gestation

A Cochrane review comparing schedules of antenatal care does not provide evidence to recommend a particular schedule for women who do not qualify for early referral:11 no study was powered to identify differences in mortality, or serious outcomes associated with pre-eclampsia. We found absence of antenatal care to be strongly associated with eclampsia and fetal death.12 A UK study showed that reducing the frequency of antenatal care shifts costs to neonatal care, resulting in higher overall costs.13

Serious morbidity associated with pre-eclampsia can occur from 20 weeks' gestation to after delivery: placental abruption; haemolysis, elevated liver enzymes, and low platelet count syndrome; and renal failure are more common before 32 weeks, whereas eclampsia is most common at term.14 15 Onset before 32 weeks has the most serious outcome and the interval between diagnosis and delivery is on average 14 days (range 0-62 days), with a substantial number of women requiring delivery within 72 hours.16 We therefore recommend (see table 1) that before 32 weeks, women with one risk factor (and none from box 3) are seen at least once every three weeks, and then at least once every two weeks, until delivery.

Women with no risk factors for pre-eclampsia may still develop the condition. NICE recommends assessments for pre-eclampsia at weeks 16, 28, 34, 36, 38, 40, and 41 for healthy parous women with a single fetus. Given that pre-eclampsia can progress to a life threatening situation in, on average, two weeks from diagnosis, we recommend that these women are told that pre-eclampsia can develop between antenatal assessments, are made aware of symptoms, and know how to contact their healthcare professionals at all times.

Content of the assessment

After 20 weeks' gestation, women should be assessed for the signs and symptoms of pre-eclampsia (see box 4). Any one of these may be the first indication of pre-eclampsia. The method of measuring blood pressure is critical: errors have been implicated in maternal deaths.1 6 Our recommendations concur with NICE's guideline. In the community, fetal compromise is usually assessed by asking women about reduced fetal movements or by estimating a small for gestational age fetus. The guideline of the Royal College of Obstetricians and Gynaecologists provides evidence based recommendations. Thresholds for step-up assessment (see table 2) are based on the association with poor outcome and rates of progression. Oedema is not predictive, and weight change does not reliably precede other signs.

Women with new hypertension before 32 weeks have a 50% chance of developing pre-eclampsia:17 at 24-28 weeks, new hypertension is predictive of severe pre-eclampsia.18 On average a rise in diastolic blood pressure that does not reach 90 mm Hg at any time during pregnancy is associated with an uncomplicated pregnancy.19 Eclampsia is not always associated with severe hypertension; in a UK population study, 34% of eclamptic women had a maximum diastolic blood pressure of ≤ 100 mm Hg.15

New proteinuria with new hypertension is strongly associated with poor fetal and maternal outcome.20 21 Women may progress rapidly: 25-55% of women with hypertension of ≥ 160 mm Hg systolic or ≥ 110 mm Hg diastolic with new proteinuria (≥+) required delivery within 48 hours of admission.16

Quantified protein excretion is independently associated with undiagnosed underlying medical conditions and a poor obstetric outcome.9 The most reliable method for quantifying protein excretion is urine collection over 24 hours. Although NICE's guideline allows the use of protein creatinine ratios to quantify protein, more recent data22 suggest that although the test is useful for screening (≥ 30 mg/mmol on a random sample) local confirmation of performance is required for quantification, as the results may be modified by the method used to measure the proteinuria.

While + proteinuria with new hypertension is associated with poor outcome and should be considered as pre-eclampsia until otherwise confirmed, a + result on dipstick testing on its own is prone to false positives. Factors affecting the result include reader error (which can be minimised by training, or the use of automated readers) and concentration errors (avoided by the use of the protein creatinine ratio test). Accuracy is not increased by repeating the test on a new sample. A + result on dipstick testing is unlikely to be due to infection, unless the woman has symptoms.

In the presence of pre-eclampsia, headache is an independent risk factor for eclampsia, and epigastric pain and vomiting are independent risk factors for serious morbidity in women with severe pre-eclampsia.23 24 These symptoms should always be followed up immediately, by an assessment of blood pressure and proteinuria as a minimum.

Fetal compromise can be the first clinical indication of pre-eclampsia1 6 and should always be followed up by an assessment of blood pressure and proteinuria as well as following local management protocols.