Maternal serum alpha-fetoprotein in fetal neural tube and abdominal wall defects at 10 to 14 weeks of gestation

Authors

  • N. J. Sebire,

    Research Fellow
    1. Harris Birthright Research Centre for Fetal Medicine, Kings College Hospital Medical School, London
    Search for more papers by this author
  • K. Spencer,

    Consultant Biochemist
    1. Department of Clinical Biochemistry, Oldchurch Hospital, Romford
    Search for more papers by this author
  • P. L. Noble,

    Research Fellow
    1. Harris Birthright Research Centre for Fetal Medicine, Kings College Hospital Medical School, London
    Search for more papers by this author
  • K. Hughes,

    Research Fellow
    1. Harris Birthright Research Centre for Fetal Medicine, Kings College Hospital Medical School, London
    Search for more papers by this author
  • K. H. Nicolaides

    Professor (Fetal Medicine) , Corresponding author
    1. Harris Birthright Research Centre for Fetal Medicine, Kings College Hospital Medical School, London
      Correspondence: Professor K. H. Nicolaides, Harris Birthright Research Centre for Fetal Medicine, Kings College Hospital Medical School, Denmark Hill, London SE5 8RX, UK.
    Search for more papers by this author

Correspondence: Professor K. H. Nicolaides, Harris Birthright Research Centre for Fetal Medicine, Kings College Hospital Medical School, Denmark Hill, London SE5 8RX, UK.

Abstract

Maternal serum alpha-fetoprotein concentration was determined in nine pregnancies with fetal anencephaly, seven with exomphalos containing liver, two with spina bifida and 100 normal controls at 10 to 14 weeks of gestation. The median alpha-fetoprotein in the group with fetal anencephaly and exomphalos was significantly higher than in normal fetuses but the sensitivity of this test is likely to be only about 30% for a false positive rate of 5%.

At 16 to 18 weeks of gestation maternal serum alpha- fetoprotein concentration is geqslant R: gt-or-equal, slanted 2.5 multiples of the median (MoM) in about 3% of normal fetuses, 90% of those with anencephaly, 80% with open spina bifida and about 40% with anterior abdominal wall defects1,2. The aim of the present study is to examine whether maternal serum alpha-fetoprotein concentration is raised in pregnancies with fetal anencephaly or exomphalos at 10 to 14 weeks of gestation.

Methods

Maternal serum alpha-fetoprotein concentration was determined in nine pregnancies with fetal anencephaly, seven with exomphalos containg liver, two with spina bifida and 100 normal controls at 10 to 14 weeks of gestation. The fetal abnormalities were diagnosed by routine ultrasound examination, which was carried out at 10 to 14 weeks in self-referred patients attending for screening for chromosomal abnormalities by maternal age and fetal nuchal translucency thickness. All pregnancies with anencephaly and three with a large exomphalos had surgical termination at 12 to 14 weeks at the request of the parents. In a further two patients with exomphalos there was intrauterine death in a subsequent visit. In both cases of spina bifida the diagnosis was confirmed at 16 and 20 weeks of gestation, respectively, and the parents opted for termination of the pregnancy. All infants of normal control patients were liveborn and healthy.

Maternal blood was taken at the time of the ultrasound examination, separated and the serum stored at −20°C until the time of analysis. The patients with abnormalities were identified by searching for all such cases where a maternal serum sample was available. Maternal serum alpha-fetoprotein concentration was measured in two batches by a sensitive immunometric assay (IDS Ltd, Tyne and Wear, UK; detection limit 0.1 kU/L), without knowledge of the ultrasound findings. All samples were analysed in duplicate (within assay precision 10% at 10 kU/L, 4% at 20 kU/L, 2% at 50kU/L).

The normal medians of maternal serum alpha-fetoprotein concentration for each gestational week were calculated and individual values of alpha-fetoprotein in the abnormal and nonnl fetuses were expressed as multiples of the normal median (MoM). Significance of differences between groups was examined using Mann Whitney U test. Cuzicks test for trend was applied to examine the relationship between maternal serum alpha-fetoprotein concentration and gestational age.

Results

In the normal group the median maternal serum alpha- fetoprotein concentration increased with gestation from 3.5 kU/L at 10 weeks to 16.2 kU/L at 14 weeks (Fig. 1a; Z= 4.3, P < 0.0001). The median alpha-fetoprotein in the group with fetal anencephaly (2.2 MoM, range 0.4–12.7 MoM) and fetal exomphalos (2.9 MoM, range 1–10.8 MoM) was significantly higher than in normal fetuses (1 MoM, range 0.01–8.4 MoM; P= 0.048 and P= 0.002, respectively: Fig. 1b). In the two cases with fetal spina bifida the maternal serum alpha-fetoprotein concentration was not significantly different from normal (0.9 MoM and 0.24 MoM, respectively; P= 0.31). A minimum value of 3.5 MoM was observed in 5% of the normal group, 33% (3/9) of those with anencephaly, 28% (2/7) with exomphalos and neither of the two cases of spina bifida.

Figure 1.

(a) Maternal serum alpha-fetoprotein concentration (kU/L) by gestational age at 10 to 13 completed weeks in normal pregnancies (▪), and in pregnancies affected by fetal anencephaly (□), open spina bifida (D) and exomphalos (•). (b) Maternal serum alpha-fetoprotein concentration expressed as multiples of the normal median (MoM) in normal pregnancies (▪), and in pregnancies affected by fetal anencephaly (Anen; Δ), open spina bifida (SB; □) and exomphalos (Exom; •). The dashed line represents 3.5 MoM which is the 95th centile for the normal group.

Discussion

The data from this study demonstrate that at 10 to 14 weeks of gestation both fetal anenecephaly and exomphalos are associated with increased maternal serum alpha-fetoprotein values. However, the sensitivity of this test is likely to be only about 30% for a false positive rate of 5%. The findings with fetal neural tube defects are similar to previous reports. Thus in the UK Collaborative Study on alpha-fetoprotein the values were above the 95th centile in only four of the 20 anencephalic pregnancies and four of the 22 pregnancies with open spina bifida at 10 to 12 weeks of gestation3. In three other studies on a total of 12 cases of anencephaly and 25 cases of spina bifida at 10 to 14 weeks the median maternal serum alpha-fetoprotein was not significantly different from normal4–6.

At 16 to 18 weeks, compared with 10 to 14 weeks there is only a small improvement in the sensitivity of screening by maternal serum alpha-fetoprotein for exomphalos, whereas for anencephaly there is a threefold increase in sensitivity. A possible explanation for this differential improvement in sensitivity is the evolution of the neural tube defect with advancing gestation. Ultrasonographically, the fetal brain at 10 to 11 weeks looks normal apart from the absence of the echogenic cranium. Subsequently, there is a breakdown of the unprotected fetal brain and progressive change from acrania to anencephaly, with consequent increase in the leakage of alpha-fetoprotein from the fetal circulation through the expanding cerebrocranial defect. However, in anencephaly the median level of amniotic fluid alpha-fetoprotein at 16 to 18 weeks of gestation (20 MoM)2 is similar to the level at 10 to 14 weeks (17 MoM)7. Since in anencephaly the maternal serum alpha-fetoprotein level at 16 to 18 weeks is 7 MoM2, whereas at 10 to 14 weeks it is only about 2 MoM, with advancing gestation there is an increase in the transport of alpha-fetoprotein from the amniotic fluid into the maternal circulation. Consequently, the most likely explanation for the differential increase in maternal serum alpha-fetoprotein with advancing gestation in anencephaly compared with exomphalos is that in the latter there is a relative decrease in amniotic fluid alpha-fetoprotein presumably due to a decrease in the permeability of the exomphalos sac.

The findings of this study confirm those of previous reports that screening for neural tube defects by maternal serum alpha-fetoprotein at 10 to 14 weeks is unlikely to be beneficial. Furthermore the data suggest that the sensitivity of screening for exomphalos is likely to be as poor as in the second trimester. Nevertheless there is evidence that both anencephaly and exomphalos can be diagnosed reliably by appropriately trained ultrasonographers both in the first and second trimesters8,9.

Ancillary