Review of current and past maternity history

At this time you will ascertain the number of pregnancies a woman reports (gravida) and the number of births (parity). Note: The definitions of mortality differ between Australia and New Zealand. In both Australia and New Zealand, a live birth is recorded if the fetus is 20 weeks gestation or over, and weighs 400 g or more at the time of birth (AIHW 2004). The perinatal death rate in Australia includes stillbirth (or fetal death) and death up to 28 days after birth; in New Zealand, a perinatal death is a stillbirth (fetal death) or a death up to 7 days after birth. The denominator for perinatal death rate is per 1000 births: both live and stillborn over these time periods.

At this time it is important to ascertain whether a woman has had any spontaneous or elective abortions or whether she has experienced preterm labour and birth, or had a stillbirth or a neonatal death. This information helps to build a better picture about whether a woman might benefit from more-intensive screening than the usual methods offered to most women regardless of any risk markers.

Assessing the effectiveness of a test

There are four indices of the validity of a test: sensitivity, specificity, and predictive values of positive and negative (see Fig 35.1).

Figure 35.1 Assessing the effectiveness of a screening or diagnostic test

(based on Grimes & Schulz 2002)

Specificity

Specificity denotes the ability of a test to identify those without the condition (Grimes & Schulz 2002). In diagnostic testing, this refers to the chance of having a negative test result, given that you do not have the disease. A 100% specificity means that all those without the disease will test negative, but this is not the same the other way around. A woman could have a negative test result, yet still have the disease—this is called a ‘false-negative’. The specificity of a test is also related to its ‘positive predictive value’ (true-positives): a test with a specificity of 100% means that all those who get a positive test result definitely have the disease. To fully judge the accuracy of a test, its sensitivity must also be considered (NICE 2008, glossary).

For the purposes of this chapter, it is important to understand the concepts of sensitivity and specificity broadly, so that you understand how screening works in real life. For example, the trade-off between sensitivity and specificity (see Fig 35.2) is demonstrated by the following. For any continuous outcome measurement (e.g. blood glucose level), the sensitivity and specificity of a test will be inversely related. If we place the cut-off for abnormal blood glucose level at a very low point, thereby producing perfect sensitivity (i.e. we manage to identify all those women with diabetes), the low cut-off identifies all those with diabetes. However, the trade-off is poor specificity—that is, those women who also have blood glucose at the level specified but are part of the ‘healthy distribution’ (i.e. at one end of ‘normal’) are incorrectly identified as having abnormal values. Placing the cut-off higher yields the opposite result: all those who are healthy are correctly identified (perfect specificity), but the cost here is missing a proportion of women who do have diabetes. Selecting a cut-off point is a compromise, mislabelling some healthy people and some people with diabetes. Where the cut-off should be depends on the implications of the test.

Figure 35.2 Hypothetical trade-off between sensitivity and specificity

(based on Grimes & Schulz 2002)

As in other areas of maternity care, there is healthy debate about what constitutes evidence-based screening and what does not. This has been instrumental in driving a serious attempt to outline the most reasonable path to follow with screening in pregnancy. For the purposes of this chapter, the road map for screening will be that developed by the NICE. The NICE Guidelines were developed in the United Kingdom and drew on the first attempt in Australia to gather evidence-based guidelines to help practitioners and women navigate their way through the maze of screening technology available. The guidelines developed by this group in the United Kingdom represent:

the view of the Institute, which was arrived at after careful consideration of the evidence available. Health professionals are expected to take it fully into account when exercising their clinical judgment. The guidance does not, however, override the individual responsibility of health professionals to make decisions appropriate to the circumstances of the individual patient, in consultation with the patient and/or guardian or carer (NICE 2008).

Critical thinking exercise

Look closely at the references cited in the NICE Guidelines, the Three Centres Consensus Guidelines from Australia, and the New Zealand guidelines for antenatal screening, and determine how many interventions are based on Level 1 evidence.

Ultrasound scans

Early in pregnancy (usually around 10–13 weeks), the first ultrasound scan may be offered. It is used to estimate when the baby is due and to check whether there is more than one fetus. If you don’t see women this early, it is appropriate to offer the scan at the earliest time. Between 18 and 20 weeks appears to be the optimum time to offer a second scan to check for physical abnormalities in the fetus. Beyond these it is not recommended that women have any further routinely offered scans, as they have not been shown to be useful (NICE 2008).

Individual clinicians, maternity units and geographic regions have adopted widely varying strategies for combining LMP and early ultrasound (EUS) estimates of gestational age (GA). Some still use LMP, others rely exclusively on EUS, while many base GA on EUS only when the discrepancy with LMP exceeds a given limit, such as ±7, 10 or 14 days (Blondel et al 2002).

Table 35.1 shows the effect of estimating the GA in a fetus using the various methods of estimation in a study in 44,623 pregnancies in France. Note the discrepancies between post-term dates using the LMP method and the EUS method. Some authors recommend ultrasound estimation as the first choice for dating all pregnancies; others have voiced reservations about exclusive use of EUS estimates. Because of these different approaches to resolving discrepancies in LMP and EUS estimates, both among and within countries, geographical and temporal trends in preterm and post-term birth remain difficult to interpret (Blondel et al 2002).

Table 35.1 Rates of preterm and post-term birth, estimated from LMP or EUS, according to six algorithms for combining the two estimates (n = 44,623)

Estimates of gestational duration based on the timing of the last normal menstrual period (LNMP) are dependent on a woman’s ability to recall the dates accurately, the regularity or irregularity of her menstrual cycles and variations in the interval between bleeding and anovulation. Crowther and colleagues (1999) estimate that between 11% and 42% of GA estimates from LMP are reported as inaccurate. Ultrasound measurements are based on standard growth curves, but fetal growth may deviate from the standard, even early in gestation. Fetuses that grow faster than the standard curve will be given a biased, longer gestational length, and therefore an increased probability of being classified as post-term, even when they are not (Olesen et al 2004). The opposite will be seen for slow-growing fetuses. Lower early ultrasound-based GA estimates therefore seem attributable in some cases to slower fetal growth (Olesen et al 2004). This observation was also reported by Smith et al (1998).

In a study that examined the determinants and consequences of fetal growth restriction as a function of differences in GA estimates, Morin et al (2005) found that several maternal and fetal characteristics influenced the magnitude of the discrepancy between the GA estimates. These included socioeconomic differences and potential determinants of early fetal growth restriction (Morin et al 2005). Small differences in growth may also depend on fetal sex, maternal age, parity and smoking (Olesen et al 2004). The NICE guideline group conclude, however, that there is thought to be little variation in fetal growth rate up to mid-pregnancy and, therefore, estimates of fetal size by ultrasound scan provide estimates of GA that are not subject to the same human error as LNMP (NICE 2008).

Ultrasound assessment of GA at 10–13 weeks is usually calculated by measurement of the crown–rump length. For pregnant women who present in the second trimester, GA can be assessed with ultrasound measurement of biparietal diameter or head circumference (NICE 2008). Ultrasound measurement of biparietal diameter is reported to provide a better estimate of date of delivery for term births than first day of the LMP (NICE 2008, p 77). Routine ultrasound before 24 weeks is also associated with a reduction in rates of intervention for post-term pregnancies. A 1998 Cochrane Review (Neilson 1998) found that, compared with selective ultrasonography, routine prenatal ultrasonography before 24 weeks gestation provided better GA assessment and earlier detection of multiple pregnancies and fetal malformations. This review reported, in a long-term follow-up of Norwegian and Swedish children, no adverse influence on school performance or neurobehavioural function as a consequence of antenatal exposure to ultrasound; however, fewer of the ultrasound-exposed children are right-handed (Neilson 1998). The ‘Alesund’ trial, which was reported in 2000 (Eik-Nes et al 2000), claimed the possible benefits of the routine use of ultrasound screening in pregnancy as a lower incidence of induced labour due to apparent post-term pregnancies—approximately 70% lower in the ultrasound-screened group. Inductions from all causes were also less frequent among ultrasound-screened women. Among the controls, three pairs of twins remained undiagnosed until the mothers were admitted to the hospital in labour at between 36 and 38 weeks gestation. The authors concluded that for women who were screened with ultrasound, obstetricians were less likely to induce labour due to apparent post-term pregnancy than for women who were not screened (Eik-Nes et al 2000).

Accurately assessing the GA also permits optimal timing of antenatal screening for Down syndrome and fetal structural anomalies. Reliable dating is important when interpreting Down syndrome serum results, as it may reduce the number of false-positives for a given detection rate (NICE 2008).

Side-effects of ultrasound

The use of ultrasonography for evaluating the developing embryo/fetus continues to rise, although the potential risks from exposure are uncertain (Barnett et al 2000; Marinac-Dabic et al 2002; Miller et al 1998). An effect of ultrasound on intellectual performance cannot, at present, be ruled out. Researchers from Sweden who published their work in the journal Epidemiology in 2005 concluded that further action to evaluate possible neurotoxic effects of ultrasound should be taken, particularly because ultrasound exposure in the published studies looking at long-term effects of exposure was low compared with present levels of exposure (Kieler et al 2005).

Prenatal exposure to ionising irradiation in the second trimester has been found to affect the developing brain, resulting in intellectual impairment (Schull & Otake 1999). Whether other forms of radiation, particularly non-ionising radiation such as ultrasound, have similar neurotoxic effects is not yet known (Ziskin & Barnett 2000). Ultrasound has the potential to damage tissue by heating, cavitation or streaming, and the brain is most susceptible to environmental effects during its development (Barnett 1998; Miller et al 1998). Modern ultrasound machines report Mechanical Index (MI) and Thermal Index (TI) as real-time feedback to the operator of the potential for fetal effects. Tissue temperature elevations become progressively greater from b-mode to colour Doppler to spectral Doppler applications, and the precautionary principle of as little exposure as possible while obtaining needed clinical information is recommended. Although epidemiological research on possible adverse effects of ultrasound is sparse, there have been published reports of an association between

non-right-handedness/left-handedness in males and exposure to prenatal ultrasound. The significance of ‘handedness’ studies is the higher risk of an association of left-handedness in infants with lowered intellectual abilities (Kieler et al 2005). However, in a study examining the association between prenatal ultrasound exposure and intellectual performance, Kieler et al (2005) failed to demonstrate a clear association between ultrasound scanning and intellectual performance.

Screening in Australia and New Zealand

To date, there has been an ‘ad hoc’ approach to serum screening in Australia and New Zealand. In Australia and New Zealand, prenatal screening tests are available to identify pregnancies at increased risk of chromosome anomalies such as trisomy 21, trisomy 18 and some structural anomalies such as neural tube defects. Ultrasound and maternal serum screening tests identify fetuses with an increased likelihood of having one of these conditions. Sometimes these conditions are not compatible with live birth, some are associated with

long-term and serious morbidity, and some require neonatal investigation or treatment. There is usually no intrauterine fetal therapy (RANZCOG 2009).

Policies that detail the aims and operations of antenatal screening programs are available online for New South Wales, South Australia, Victoria and Western Australia. The Royal Australian and New Zealand College of Obstetricians and Gynaecologists (RANZCOG) also produces national policy statements on prenatal screening (RANZCOG 2009).

There does not appear to be a consistent view on whether screening should be offered to all women or targeted to women of a particular age group. Best estimates of the rates of participation in either first- or second-trimester screening tests varies between 39% (Queensland) and 80% (South Australia), with much lower uptake (17%) in the Northern Territory. Most states had between 50% and 70% participation rates, with the majority of screening tests performed in the first trimester. The states with higher rates of participation were also those with policies that provided information and guidance to both healthcare professionals and the community (O’Leary et al 2006). This is despite the introduction in Australia of a specific Medicare item number (MBS 66321) and rebate for the maternal serum screen. Only the New South Wales, South Australian and Western Australian governments have policies for antenatal screening for Down syndrome and other fetal anomalies (O’Leary et al 2006).

Recommendations for antenatal screening for Down syndrome in Australia are as follows:

In New Zealand, antenatal screening for Down syndrome has developed rapidly over the past 15 years. There is now a range of screening tests available, including:

Development of diagnostic testing methods

When the first screening test for Down syndrome was introduced in the United Kingdom, approximately 5% of pregnant women were 35 years of age or older, and 30% of all Down syndrome births were related to those women (Haddow 1998). The vast majority of babies do not have Down syndrome. However all women, whatever their age, have a small risk of delivering a baby with a physical and/or mental disadvantage. The majority of children with Down syndrome are born to women who are younger than 35 years old. The identification of younger women with an increased risk became possible after Merkatz et al (1984) reported an association between a low alpha-fetoprotein level and fetal aneuploidy, launching an era of maternal serum screening to assess the risk of fetal Down syndrome. Subsequent studies showed that the maternal serum levels of three markers—alpha-fetoprotein (AFP), beta human chorionic gonadotrophin (β-hCG), and unconjugated oestriol (uE₃)—occurring between 15 and 22 weeks of gestation could be used to make adjustments to the risk based on maternal age alone (Mennuti & Driscoll 2003).

Antenatal screening for Down syndrome became possible because two components of diagnostic testing had developed to the level of clinical applicability. The first was a sampling procedure where cells were taken from the amniotic fluid (amniocentesis). Cells in amniotic fluid are fetal in origin and can be cultured for chromosome analysis. Amniocentesis was originally offered during the early second trimester and carried about a 1% risk of spontaneous fetal loss (Tabor et al 1986). The second diagnostic component that had reached clinical significance was cell culture and chromosomal analysis via karyotype. Karyotyping is still the preferred test method for most pregnancies, ascertained either through a screening program for Down syndrome or for other referral reasons. Karyotyping detects a range of numerical and structural chromosome abnormalities in addition to the common autosomal trisomies—13 (Patau’s syndrome), 18 (Edwards’ syndrome) and 21 (Down syndrome)—and sex-chromosome abnormalities.

However, since amniotic fluid and chorionic villus cells are cultured before analysis, delays of up to 14 days or longer can occur before a result is issued in the case of rare anomalies (Caine et al 2005). The results of CVS for Down syndrome are usually available within in three days of the test in Australia (Fetal Medicine Foundation 2009). Notwithstanding that, this is currently recognised as among the most reliable diagnostic tests in clinical medicine (Haddow 1998). Culture failure is uncommon, and erroneous results from maternal cell contamination are rare (Haddow 1998). This degree of reliability is crucial, since it forms the basis for decisions that have far-reaching consequences. Chromosome analysis via karyotype, however, continues to be the cornerstone of Down syndrome diagnosis (Haddow 1998).

After the landmark discovery in 1984 that AFP concentrations in maternal serum were lower in the presence of Down syndrome and could be used as a screening test (Cuckle et al 1984), screening began to be offered to pregnant women younger than age 35. The efficacy of screening with AFP was comparable to that based on maternal age (a 20%–25% detection rate for a 5% false-positive rate), and since AFP was already being measured as a screening test for open spina bifida or neural tube defect, adding an interpretation for Down syndrome risk represented a negligible cost (Haddow 1998). Further progress in screening was achieved later in the 1980s with the discoveries of β-hCG and uE₃. Measures of these markers were combined with that for AFP to produce a ‘triple’ test, capable of detecting 60% of Down syndrome pregnancies, at a 5% false-positive rate (Wald et al 1988). This substantial step-up in detection led many women over the age of 35 to choose serum screening rather than amniocentesis. The addition of a fourth serum marker, inhibin, has been shown to boost detection of Down syndrome to about 75% (Haddow 1998).

With the advent of biochemical markers in combination with ultrasound, considerable attention was given in the 1990s to providing maternal serum screening earlier in gestation. Measurements of two markers—pregnancy-associated plasma protein A (PAPP-A) and free β-hCG detected between 10 and 13 weeks gestation—have become nearly as efficient as the second-trimester ‘triple’ test (Haddow 1998).

Studies in the 1990s showed an association between Down syndrome and increased nuchal translucency, a sonolucent space evident at the back of the fetus’s neck in the first trimester. In 1995, Wald et al demonstrated the feasibility of first-trimester serum screening for Down syndrome (Wald et al 1995).

Current practice

The most accurate way of estimating the risk of the fetus having Down syndrome is carried out at 11–13 weeks and depends on the:

Screening combinations

The NICE Guidelines recommend the following screening combinations for Down syndrome:

The NICE Guidelines (2008) also state that:

Some women undergo CVS during the first trimester rather than screening or amniocentesis during the second trimester. Some women, such as those carrying multiple fetuses, are not candidates for second-trimester serum screening. Others decline screening because they do not wish to undergo prenatal diagnosis or would not consider pregnancy termination on the basis of the results. The anxiety experienced by friends or relatives who had a positive screening test followed by normal findings on amniocentesis deters some women from undergoing screening. Women who are 35 years of age or older increasingly opt to undergo serum screening and ultrasonography before deciding about amniocentesis. A range of other ultrasound features such as nasal bone length and echogenic heart areas are currently being investigated as indicators (NICE 2008). With the emergence of alternative approaches earlier in pregnancy, decisions about screening are becoming more complex.

The RANZCOG also provides a comprehensive guide to screening that is updated as new evidence comes to hand (RANZCOG 2007).

Summary: screening for Down syndrome

Figure 35.3 illustrates the rates of use of various diagnostic tests for Down syndrome.

Figure 35.3 Diagnostic tests for Down syndrome (percentages)

(based on Botto et al 2000)

What to tell women

Rowe et al (2006) have found that Australian women participating in prenatal genetic screening are inadequately informed regarding aspects of testing, including the management of pregnancy in the event of an ‘increased risk’ result. Whether or not decisions were informed was not associated with differences in measures of depression and anxiety. They suggested that strategies to improve the quality of education for clinicians so that they are able to convey the necessary complex information effectively and sensitively, and the development of participant information that meets diverse learning capacities and takes into account the effects of pregnancy-related elevations in anxiety, are needed (Rowe et al 2006).

The following information is a brief outline of the sorts of issues midwives might discuss with women who are contemplating screening for Down syndrome.

Use of maternal age as a primary criterion for offering amniocentesis results in very high rates of use of this invasive test and is a suboptimum use of resources. Every woman’s choice to accept or reject chorionic villus sampling or amniocentesis should be based on counselling that uses the best possible estimate of her personal risk for fetal aneuploidy. Policy advisory groups in other countries should follow the UK initiative, and abandon obsolete guidelines that have advocated offering amniocentesis to all women aged 35 or more, without routinely incorporating serum and ultrasound screening into their risk assessment (Benn 2003; Rowe et al 2006).

Birth anomalies remain a significant public health problem in Australia and are a major reason for admission to hospital during infancy and childhood. They often result in disabilities and handicaps and, in some cases, death (AIHW 2002). An estimated 5% of all Australian births and terminations of pregnancy have a major birth anomaly (Stanley et al 1995), and birth anomalies account for about 13% of the disease burden for children aged 0–14 years (AIHW 1999). Birth anomalies are also a leading cause of infant mortality in Australia, with 25% of infant deaths in 2000 caused by them (Al-Yaman et al 2002).

Birth anomalies are caused by genetic (including chromosomal), environmental and unknown factors, or combinations of these factors. It is estimated that the cause of about 65%–75% of birth anomalies is unknown. About 15%–25% of birth anomalies have a genetic cause and about 10% have an environmental cause (AIHW 2004; Brent 2001).

Neural tube defects

Neural tube defects (NTDs) represent a common group of severe congenital malformations that result from a failure of closure of the embryonic neural tube during early development. Their aetiology is quite complex, involving environmental and genetic factors, and their underlying molecular and cellular pathogenic mechanisms remain poorly understood (Kibar et al 2007). The syndrome includes abnormality of the brain and spinal cord. It can

be an isolated abnormality or a part of a genetic syndrome. Genetic studies of NTDs have focused mainly on folate-related genes, based on the finding that perinatal folic acid supplementation reduces the risk of neural tube defects by 60%–70% (Kibar et al 2007) or by 50%–70% (Van der Linden et al 2006).

Neural tube defects are among the most common congenital malformations in humans, affecting 1–2 infants per 1000 births; their incidence varies among different populations (Copp et al 2003).

There are known inconsistencies in the Australian and New Zealand data on rates of NTD, with widely accepted best estimates of 1.32 per 1000 total births for 1999–2003 and 2.56 per 1000 births for 1996–2000 for Indigenous Australians (Bower et al 2006; Dalziel et al 2009). Australian and New Zealand guidelines are similar to those elsewhere (Department of Health [UK] 1992; Institute of Medicine, Food and Nutrition Board [US] 1998), and recommend that women planning or capable of becoming pregnant should consume folic acid as a supplement or in the form of fortified foods at a level of 400 mg/day for at least one month before and three months after conception, in addition to consuming food folate from a varied diet (Australian Government and New Zealand Ministry of Health 2006). The use of periconceptional supplements in Australia is estimated to be about 36%, with lower rates for Indigenous Australian women (Conlin et al 2006).

Stay updated, free articles. Join our Telegram channel

Join