zyxwvutsrqp
zyxwvutsrqp
British Journal of Obstetrics and Cynaecology
April 1988, Vol. 95, pp. 334-341
Maternal serum unconjugated oestriol as an antenatal
screening test for Down’s syndrome
N. J. WALD, H. S. CUCKLE, J . W. DENSEM, K. NANCHAHAL,
J. A. CANICK, J. E. HADDOW, G. J . KNIGHT, G. E. PALOMAKI
zyxwvutsr
Summary. The median maternal serum unconjugated oestriol level
between 13 and 27 weeks gestation in 77 pregnancies associated with
Down’s syndrome was lower than the median level in 385 unaffected
control pregnancies matched for maternal age, gestational age, and
duration of serum sample storage (P<O*OOl). The median level for the
affected pregnancies was 73% of that in the controls. Low unconjugated
oestriol levels can be used to detect fetal Down’s syndrome; at cut-off
levels selected to detect at least 3S% of affected pregnancies, unconjug-
ated serum oestriol was a better screening test than either maternal age
or serum alpha-fetoprotein (AFP). The use of all three variables in
combination to select women with a 1:250 or greater risk of a Down’s
syndrome term pregnancy would yield a 45% detection rate with a false-
positive rate of 5.2%. The same detection rate using maternal age alone
or using age and serum AFP in combination would yield higher false-
positive rates, 15% and 9.8% respectively. The addition of unconjug-
ated oestriol to a Down’s syndrome screening programme would there-
fore be more efficient than the use of age and AFP alone; for a given
detection rate fewer women would need an amniocentesis or, for a given
percentage of women having an amniocentesis, more pregnancies with
Down’s syndrome would be detected.
Department of Environmental and Preventive Our observation that maternal serum unconjug-
Medicine, St Bartholomew’s Hospital Medical ated oestriol levels were significantly lower in
College, London EClM 6BQ, UK pregnancies associated with Down’s syndrome
N . J . WALD Professor and Head of Department than in unaffected pregnancies of the same ges-
H. S. CUCKLE CRC Senior Lecturer tational age (Canick et al. 1988) prompted us to
J. W. DENSEM Computer Manager investigate the association further. Using a bank
K. NANCHAHAL Compuler Programmer of mid-trimester antenatal serum samples we
Department of Pathology and Laboratory Medicine, designed a study to investigate whether serum
Brown University Women and Infant’s Hospital, unconjugated oestriol measurement would be a
zyxwvutsrqpon
Providence, Rhode Island, USA useful antenatal screening test for Down’s syn-
J . A CANICK Associate Professor
drome and, if so, how it might be used in con-
Foundation for Blood Research, Scarborough, junction with the determination of maternal age
Maine, USA
and maternal serum alpha-fetoprotein (AFP).
J . E. HADDOW Associate Medical Director
G . .I.KNIGHT Direclor, Radioimrnunoassuy
Laboratory
G. E. PALOMAKI Biostatisficiun
Methods
Correspondence: N. J. Wald, Department of Environ- General
mental and Preventive Medicine, St Bartholomew’s
Hospital Medical College, Charterhouse Square, A -40°C antenatal serum bank was used relat-
London EClM 6BQ ing to pregnancies from women attending for
334
, antenatal care at the John Radcliffe Maternity
Hospital, Oxford, between 1973 and 1983. Sev-
enty-seven antenatal serum samples were identi-
zyxw
zyxwv
zyxw
Serum oestriol in screening Down’ssyndrome
expressed in MOM but a log linear regression
was used.
335
fied from 77 singleton pregnancies associated Method of estimating detection rates and false-
with Down’s syndrome (cases). (The extra nine positive rates from our resultCs
cases compared to our previously reported series
of 68 (Cuckle et al. 1987) represent pregnancies For each screening variable (maternal age,
from which serum had been collected and stored unconjugated oestriol and AFP) considered
alone and in combination, wc estimated the
but which had not undergone AFP screening.)
false-positive rate (proportion of unaffected
Each sample was matched with five stored con-
pregnancies regarded as positive) corresponding
trol samples from unaffected pregnancies
to a specified detection rate (proportion of
matched for maternal age (within the same
zyxwvutsrqpo
affected pregnancies regarded as positive).
year), gestational age (usually within 1 week) (i) For maternal age alone these rates were
and duration of storage (within 1 year) of the derived from the age distributions of affected
sample. The samples were retrieved from and unaffected pregnancies. The age distri-
storage and assayed for unconjugated oestriol bution for Down’s syndrome was estimated by
and AFP. None of the cases or controls were applying the risk of Down’s syndrome for
from pregnancies associated with a neural-tube specific years and days of age to the age distri-
defect, and all the samples were collected bution in years of maternities in England and
between 13 and 27 completed weeks gestation. Wales between 1981 and 1985 after dividing the
The affected pregnancies were identified from number each year by 365. The risk for specific
the same sources as described previously years and days of age was estimated as described
(Cuckle et al. 1984) namely (i) the records of the previously (Cuckle et al. 1987). The age distri-
Department of Medical Genetics, Churchill bution for unaffected pregnancies was obtained
Hospital, Oxford, (ii) the delivery records at the by subtraction.
John Radcliffe Maternity Hospital, Oxford, (iii) (ii) For AFP alone the detection and false-
notification of congenital malformations made positive rates were derived from areas under the
at the Office of Population Censuses and Sur- Gaussian frequency distributions of log AFP
veys, (iv) the Oxford obstetric data system for previously defined (Cuckle et al. 1987).
infants born at the John Radcliffe, (v) records (iii) For unconjugated oestriol alone we first
held by the local hospital for handicapped chil- confirmed that the data fitted Gaussian fre-
dren (Park Hospital) and (vi) the handicapped quency distributions satisfactorily (see Results).
zyxwvutsr
q j s t e r for Oxfordshire. Since the assays were performed under research
Unconjugated oestriol was assayed using a conditions within 1 month, an estimate of long-
lirect, non-extraction radioimmunoassay term between-batch assay variance was added to
Amcrlex oestriol RIA kit, Amersham). As the observed variances of unconjugated oestriol
iefore (Canick et al. 1988),the assay was modi- in order to obtain estimates of the standard
ied by doubling the sample volume to 4 0 ~ for1 deviations that would be found in routine prac-
dl specimens analysed. Maternal serum AFP tice. The detection and false-positive rates were
vas assayed using an immunoradiometric assay then estimated from Gau n distributions with
Boots-Celltech Diagnostics Ltd). All the assays the adjusted parameters. The long-term
vere performed within 1month. Cases and con- between-batch assay variance was estimated to
rols were always assayed in the same analytical be 0.0034. This was obtained by the repeated
iatch, and the assays were done without know- assay of two serum samples in many batches: (a)
edge of whether the samples were from affected a serum sample with mean unconjugated
ir unaffected pregnancies. oestriol concentration of 1.39 MOM at 17 weeks
Unconjugated oestriol levels were expressed gestation assayed 58 times over an &month
n units of mass Concentration and in multiples of period and (b) a serum sample with mean con-
he median for unaffected pregnancies of the centration of 0.50 MOM at 17 weeks gestation
ame gestational age (MOM) using a linear assayed 47 times over a 7-month period.
egression of unaffected medians on gestational (iv) For combinations of variables detection
ige weighted for the number of women tested at and false-positive rates were estimated using the
:ach week of gestation. AFP levels were also method described previously (Cuckle et al. 1987)
