Estimation of gestational age

Accurate estimation of gestational age is not only important in the assessment of fetal size and growth, but also guides decisions regarding other obstetric interventions, such as prenatal testing, whether administration of prophylactic corticosteroids for fetal lung maturity and transfer to another healthcare setting is appropriate in cases of preterm labour, or when labour induction in prolonged pregnancy should occur [4]. It is also important in interpretation of results of first‐trimester screening for chromosomal abnormalities using a combination of nuchal translucency, pregnancy‐associated plasma protein‐A and free β‐human chorionic gonadotrophin (hCG) [5].

The typical length of gestation after conception is 266 days or 38 weeks (i.e. ‘conceptual age’). However, gestational age is traditionally estimated from the last menstrual period (LMP), adding 2 weeks to ‘postmenstrual age’, giving 280 days or 40 weeks. This assumes that ovulation and conception occur 14 days after LMP. This is not always the case: irregular menses, unknown or uncertain dates, oral contraceptive use or recent pregnancy or breastfeeding may all influence the accuracy of this method, and this inaccuracy is significant in a large proportion of women [6,7]. Bleeding during the first trimester can also add to difficulty in confirming gestational age clinically based on the period of amenorrhoea.

Because of this, guidelines in most developed countries support the estimation of gestational age by first‐trimester ultrasound using crown–rump length (CRL). Although this is more accurate at estimating gestational age at population level, it is important to recognize that this method too has limitations when interpreting individual results. For instance, there is an underlying assumption that all fetuses of the same size are of the same gestation, ignoring physiological differences and biological variability in size. In addition, aberrations in normal growth at very early stages of pregnancy exist and are associated with adverse outcome.

It is generally the case that assessment of gestational age in late pregnancy is less accurate than late pregnancy dating. This is because fetal ultrasound measurements are associated with a larger absolute error with advancing gestation, and because fetal growth disturbances become more prevalent, meaning that an abnormally small fetus could be misjudged to have lower gestational age (while a macrosomic fetus may be ascribed a more advanced gestational age). This limitation is of particular relevance in women who attend for their first antenatal care visit late in pregnancy and where no other reliable estimation of gestational age is available. It is known that unreliable reporting of LMP and late antenatal care are both associated with adverse pregnancy outcome; because of this, a clinically cautious approach is important when gestational age is assigned late, and particularly in the third trimester. Thus, the potential for error should be taken into account in order to ensure safe obstetric practice: for example, in preterm labour where late estimation of gestational age suggests a value above 34 weeks, prophylactic steroids or neonatal transfer should still be carried out as the gestational age may be lower by 2 weeks; in contrast, post‐dates labour induction may be appropriate at 39 weeks after late assessment of gestational age, as this could be as late as 41 weeks [8].

Although a CRL measurement may be the most accurate measure of gestational age in most pregnancies, it has been argued that clinical judgement is required in practice to determine the best approximation of the true gestational age. First, the earliest reliable ultrasound scan should be used to ascribe an estimated due date and this should not be changed subsequently as this can lead to potential dating errors. Second, all information collected at the time of that first visit (including the reported LMP and assessment of its reliability) should be taken into account. When a reliable LMP and ultrasound estimate concur, small discrepancies with actual gestational age may still exist due to inherent CRL measurement variability. Conversely, an apparently reliable and accurate LMP with a substantial difference in estimated gestational age based on CRL should be considered as an indicator of possible growth disturbance or underlying pathology that may merit further assessment [9].

Measurement of the fetus

The most common methods for estimating fetal size at any one time are by measuring fetal biometry using ultrasound; or clinically, but also less accurately, by measurement of the maternal fundal height. It has been shown that universal third‐trimester ultrasound (compared with selective ultrasound, which is only carried out based on risk factors or abnormal symphysis fundal height) is associated with greater diagnostic effectiveness as a screening test for SGA: those fetuses with reduced growth velocity were at increased risk of neonatal morbidity [10]. Nevertheless, meta‐analysis of randomized trials has failed to demonstrate benefit of routine late pregnancy ultrasound in low‐risk or unselected populations, in terms of perinatal mortality, preterm birth less than 37 weeks, caesarean section rates, and induction of labour rates [11]. It is possible that these two seemingly contradictory findings are the result of previous randomized trials lacking the use of an effective intervention after screening, or other flaws such as lack of statistical power [10].

Ultrasound

Estimation of the fetal head circumference (HC), abdominal circumference (AC), and femur length (FL) is undertaken using standard ultrasonographic planes (Fig. 3.1). Based on these parameters it is possible to calculate an estimated fetal weight (EFW). Although there are some advantages in using this estimation (for example, it is helpful in counselling parents and enabling paediatricians to make management decisions), there are disadvantages of using only a single summary measure of size. This is because individual measurement errors are compounded, resulting in 95% confidence intervals for random error in the region of 14% of birthweight. Importantly, this error is highest in exactly those pregnancies where accurate estimation is more important, namely babies with low and high birthweight [12]. Additional ultrasound measurements, including assessment of amniotic fluid and Doppler studies of uteroplacental and fetal blood flow, may aid in the clinical management of fetuses with (or at risk of) abnormal growth.