 , .

17:4: 305–309 (1997)

THE ASSOCIATION OF EARLY-ONSET FETAL

GROWTH RESTRICTION, ELEVATED
MATERNAL SERUM ALPHA-FETOPROTEIN,
AND THE DEVELOPMENT OF SEVERE
PRE-ECLAMPSIA
 . *   -†
*Vincent Memorial Obstetrics and Gynecology Service, Massachusetts General Hospital and †Department of
Obstetrics and Gynecology, Brigham and Women’s Hospital, Boston, Massachusetts, U.S.A.

Received 25 July 1996

Revised 28 October 1996
Accepted 5 November 1996

SUMMARY
From Antenatal Diagnostic Center referrals over 22 months, consultations for early-onset fetal growth restriction
versus skeletal dysplasia were retrospectively identified. Those with elevated maternal serum alpha-fetoprotein
(MSAFP) levels are the focus of this report. All had an early ultrasound confirming menstrual dates and subsequent
sonography at <28 weeks with at least two fetal biometric measures delayed by ¢2 standard deviations from mean
values. Of the five patients identified, the mean gestational age at the time of diagnosis of fetal growth restriction was
23·3&2·9 weeks. All had normal karyotypes and normal amniotic fluid AFP. None of the patients had evidence of
hypertension or pre-eclampsia at diagnosis of fetal growth restriction. All five gravidas subsequently developed
severe pre-eclampsia from 5·5 to 12·5 weeks after documentation of fetal growth delay. Three developed HELLP
syndrome. Pregnancies were continued a mean duration of 10·2 weeks, with all five delivering at preterm gestations
(mean=33·5&1·7 weeks) for maternal indications of severe pre-eclampsia. Unexplained early-onset fetal growth
restriction in conjunction with unexplained elevations of MSAFP together consistently heralded the subsequent
development of severe pre-eclampsia. ? 1997 by John Wiley & Sons, Ltd.
 : IUGR; MSAFP; pre-eclampsia

INTRODUCTION women with unexplained elevations of MSAFP

For fetuses without sonographically identified
structural defects, elevated levels of maternal
serum alpha-fetoprotein (MSAFP) have been
associated with a variety of obstetrical complica-
tions. These fetuses are at a greater risk of low
birth weight (Brock et al., 1977; Wald et al., 1977),
perinatal death, and hypertension or pre-eclampsia
(Katz et al., 1990; Brazerol et al., 1994; Milunsky
et al., 1989; Gordon et al., 1978), among other
measures of poor obstetrical outcome. Most
Correspondence to: Thomas D. Shipp, MD, Vincent
Memorial Obstetrics and Gynecology Service, Founders
4—Massachusetts General Hospital, Fruit Street, Boston, MA
02114, U.S.A.
will have good pregnancy outcomes (Katz et al.,
1990). By further defining the natural history of
such pregnancies and identifying covariate risk
factors, obstetric management and possible thera-
peutic interventions for these high-risk pregnancies
with unexplained elevations of MSAFP will
become clearer.
The increasing use and sophistication of ultra-
sound in obstetrics have resulted in improved
detection of structural defects and disorders of
fetal growth which can now be confidently ident-
ified prior to the third trimester. We have retro-
spectively identified pregnancies referred for
intrauterine growth restriction (IUGR) prior to 28
weeks’ gestation. A subpopulation of this data set,

CCC 0197–3851/97/040305–05 $17.50

? 1997 by John Wiley & Sons, Ltd.
306 . .   . -
those growth-restricted fetuses with elevated
MSAFP, consistently developed subsequent severe
pre-eclampsia, necessitating preterm delivery.
MATERIALS AND METHODS
From the Antenatal Diagnostic Center referrals
at Brigham and Women’s Hospital over a
22-month period from May 1993 through Febru-
ary 1995, 13 consultations for early-onset IUGR
versus skeletal dysplasia were identified. A subset
of the study population, those continuing IUGR
pregnancies with unexplained elevations of second-
trimester MSAFP, constitutes the focus of this
report.
Study inclusion criteria were (1) first or early
second-trimester ultrasound dating confirming
menstrual dates; (2) unexplained elevations of
MSAFP ¢2·0 multiples of the median for the
gestational age; and (3) subsequent sonography at
less than 28 weeks with at least two fetal biometric
measures delayed by ¢2 standard deviations from
the mean for the gestational age (Merz et al., 1987;
Tamura and Sabbagha, 1980). Pregnancies were
excluded if they were complicated by (1) pre-
existing maternal hypertension or diabetes; (2)
maternal tobacco or drug use; (3) skeletal abnor-
malities suggesting severe dysplasia; (4) major fetal
malformations; or (5) multiple gestations.
The identified pregnancies were reviewed retro-
spectively for antenatal course, ultrasound
findings, pregnancy outcome, maternal com-
plications, and placental pathology. The classifi-
cation of severe pre-eclampsia was made if one of
the following obstetrical criteria was present: (1)
blood pressure ¢160/110 on two occasions at
least 6 h apart; (2) urine protein of 3+ or more
and/or ¢5 g of urinary protein over 24 hours; (3)
oliguria with ¡500 cm3 urine output over 24 h;
(4) the presence of cerebral or visual disturbances
(ACOG, 1986). HELLP syndrome was diagnosed
when there was laboratory evidence of haemoly-
sis, elevated liver enzymes, and thrombocyto-
penia. Fetal growth restriction was not included
as a criterion for the diagnosis of severe
pre-eclampsia for this study.
All ultrasounds were performed with an Acuson
128 with a 3·5 MHz or 5 MHz variable focus
transducer (Acuson, Mountain View, CA, U.S.A.).
Biometric measurements evaluated for fetal
growth included biparietal diameter, abdominal
circumference, and femur length. Umbilical artery
Doppler flow determinations were obtained.
Amniotic fluid was subjectively assessed. A thor-
ough fetal structural survey was performed on all
ultrasound examinations. All MSAFP levels were
assessed within the range of 15–22 weeks. Values
were adjusted for maternal weight and race, and
were reported as multiples of the median for the
specified gestational age at sample procurement.
Gestational age means are presented &1
standard deviation.
RESULTS
Thirteen patients were referred to the Antenatal
Diagnostic Center over the 22-month period for
early-onset IUGR versus skeletal dysplasia. Five
of these 13 patients were identified for the study.
All five had unexplained elevations of MSAFP,
early-onset IUGR, and subsequently developed
severe pre-eclampsia (Table I). Of the eight other
patients, none of the continuing pregnancies had
unexplained elevations of MSAFP, and none
developed pre-eclampsia.
The mean gestational age of diagnosis of IUGR
was 23·2&2·9 weeks (median=24·1 weeks). The
specific fetal biometric parameters used are repre-
sented in Table I. All cases had a normal amniotic
fluid volume at the time of diagnosis of IUGR, and
one case (3) developed mild oligohydramnios by
the time of delivery. Three patients had umbilical
artery Doppler wave forms that appeared normal;
one case (3) showed absent diastolic flow; and
another (4) showed decreased diastolic flow. These
patterns persisted throughout the pregnancy.
All fetuses were karyotyped via amniocentesis,
and all were found to have a normal karyotype
and normal amniotic fluid AFP levels. TORCH
titres were unremarkable for four patients (1, 3, 4
and 5). Anticardiolipin antibody and lupus antico-
agulant levels were obtained on three patients (2, 3
and 4), and were negative. Two patients were
begun on 81 mg of aspirin per day at the first
ultrasound evidence of impaired fetal growth (23
and 24·5 weeks), and it was continued through-
out the pregnancies. All five women received
betamethasone prior to delivery.
All women had normal blood pressures early in
gestation, and none had evidence of pre-eclampsia
or hypertension at the time of diagnosis of IUGR.
Two gravidas were nulliparas (3 and 5). The
remaining three patients (1, 2, and 4) previously
delivered four appropriate-for-gestational-age
Table I—Clinical characteristics and outcomes of study pregnancies

Biometric parameters Gestational

Gestational (SD<mean) age at Gestational
age at Amniotic Umbilical diagnosis age at
MSAFP diagnosis fluid artery pre-eclampsia delivery
Patient (MOM) IUGR BPD AC FL volume Doppler Aspirin (weeks) (weeks) Outcome Placental pathology

1 2·5 27·0 0 2·0 2.0 Normal Normal No 35·7 35·9 Severe 360 g, multiple small placental
pre-eclampsia, infarcts, multiple intervillous
induction, thrombi, extensive intravillous
SVD, male, 1710 g fibrin, excessive umbilical cord
spiraling
2 2·4 24·1 2·8 2·0 5·6 Normal Normal Yes 32·3 32·4 HELLP syndrome, 310 g, multiple microscopic
induction, SVD, infarcts, intravillous fibrin,
female, 1260 g, focally extensive
grade I IVH
3 2·1 24·5 2·0 1·8 5·3 Normal (mild Absent Yes 30·0 31·5 HELLP syndrome, 190 g, multiple small marginal
  , 

oligohydramnios diastolic C/S, malpresenta- infarcts

at delivery) flow tion, female, 1050 g
4 8·3 20·3 3·2 — 4·0 Normal Decreased No 30·3 34·3 Severe pre- 300 g, multiple infarcts, largest
diastolic eclampsia, C/S, 9 cm
MSAFP

flow malpresentation,
female, 1365 g
5 3·1 20·5 2·0 — 2·0 Normal Normal No 33·0 33·3 HELLP syndrome, 180 g multiple small infarcts
C/S, non-reassuring
fetal testing,
female, 1415 g

MSAFP=maternal serum alpha-fetoprotein; MOM=multiples of median; IUGR=intrauterine growth restriction; BPD=biparietal diameter; AC=abdominal
circumference; FL=femur length; SD=standard deviation; SVD=spontaneous vaginal delivery; IVH=intraventricular haemorrhage; C/S=Caesarean section.
 -
307
308 . .   . -
infants at term. None of these prior pregnancies
had been complicated by pre-eclampsia.
The mean gestational age at the time of diagno-
sis of pre-eclampsia was 32·3&2·1 weeks. Follow-
ing the ultrasound diagnosis of fetal growth delay,
pregnancies continued a mean duration of
9·0&2·3 weeks (range 5·5–12·5 weeks) until pre-
eclampsia was diagnosed. The mean gestational
age at delivery was 33·5 weeks&1·7 weeks
(median=33·3 weeks). Following the ultrasound
diagnosis of fetal growth restriction, pregnancies
continued a mean duration of 10·2 weeks&3·0
weeks (range 7–14 weeks, median=8·9 weeks). All
pregnancies were delivered for severe pre-
eclampsia. Three pregnancies developed HELLP
syndrome (2, 3 and 5). The 5-min Apgar score was
more than 7 in all cases. All neonates were small
for gestational age at delivery, and on newborn
examinations none was found to be dysmorphic,
nor were structural anomalies present.
Common to all placentae was the finding of
multiple placental infarcts. These ranged from
microscopic to extensive and were found in each
case.
DISCUSSION
Pregnancies at risk for low birth weight ident-
ified by unexplained elevations in MSAFP levels
contribute only a small proportion to the popula-
tion of low birth weight infants. Furthermore,
whether low birth weight pregnancies with
elevated MSAFP levels are due to preterm delivery
and/or IUGR has been a matter of debate (Wald et
al., 1977; Davis et al., 1992; Waller et al., 1993;
Brazerol et al., 1994; Robinson et al., 1989). The
association of elevated MSAFP, pre-eclampsia,
and subsequent low birth weight has been pre-
viously suggested (Walters et al., 1983, 1985). In
this study, however, the low birth weight was
secondary to preterm delivery necessitated by pre-
eclampsia rather than by IUGR. Our study differs
in that we have chosen pregnancies with elevated
MSAFP levels and early-onset growth restriction
in which severe pre-eclampsia subsequently devel-
oped remote from the IUGR diagnosis.
A few important points concerning our paper
deserve comment. First, we use a discriminating
value of 2·0 MOM for our MSAFP screening;
however, a number of centres use 2·5 MOM. At
this level, two of our five patients would not have
been identified as their MSAFP was between 2 and
2·5 MOM. Second, the use of a rigorous definition
of fetal growth restriction, i.e. two biometric par-
ameters with measurements ¢2 standard devia-
tions from the mean for the gestation, was chosen
so as to identify clearly those fetuses with a dis-
order of fetal growth. Third, it must be
re-emphasized that our study patients represent a
very select group of pregnancies complicated by
unexplained elevated MSAFP levels. Fourth, sev-
eral weeks elapsed after the diagnosis of IUGR,
prior to the development of the clinical signs of
pre-eclampsia. This provides less support for pre-
eclampsia as aetiologic in the early-onset IUGR
that we observed, unlike the IUGR which is more
commonly identified late in gestation associated
with pre-eclampsia.
As the amniotic fluid AFP levels were normal
for the entire group, the most likely reason for the
elevated MSAFP levels involves an abnormal pla-
cental process. A placental pathology would be
consistent with a disease process which affected
these fetuses with an early-onset growth restric-
tion. All placentae showed multiple infarcts. Pla-
cental vascular lesions have been previously
associated with elevated MSAFP levels (Salafia et
al., 1988) and are commonly seen in pre-eclampsia
and growth restriction (Gersell et al., 1987).
We have reviewed a subset of patients who were
predicted to be at higher risk for an abnormal
pregnancy outcome based on the elevated MSAFP
result. The identification of early-onset IUGR in
these pregnancies appeared to further increase
their risk for the development of severe pre-
eclampsia. The fact that the growth restriction
antedated the development of severe pre-eclampsia
by a mean duration of 9·6 weeks is noteworthy and
may have implications for management. Two cases
(2 and 3) were begun on aspirin in the hope of
improving outcome, though this was not initiated
until the growth restriction was well established,
and no obvious improvement in outcome was
realized. As there seems to be a great risk for the
development of severe pre-eclampsia, these preg-
nancies complicated by both unexplained eleva-
tions in MSAFP and early-onset IUGR warrant
close surveillance for maternal complications and
interventions in anticipation of a preterm delivery.
It has already been firmly established that preg-
nancies with unexplained elevations of MSAFP
levels are at risk for pre-eclampsia and low birth
weight. We have identified a subset of women with
increased MSAFP in whom a covariate risk of
early-onset fetal growth restriction consistently
   , 
heralds the subsequent development of severe pre-
eclampsia. Therapies to improve the outcome in
such cases remain to be investigated.
REFERENCES
ACOG (1986). Management of Preeclampsia, American
College of Obstetricians and Gynecologists Technical
Bulletin No. 91, Washington, DC: American College
of Obstetricians and Gynecologists.
Brazerol, W.F., Grover, S., Donnenfeld, A.E. (1994).
Unexplained elevated maternal serum á-fetoprotein
levels and perinatal outcome in an urban clinic popu-
lation, Am. J. Obstet. Gynecol., 171, 1030–1035.
Brock, D.J.H., Barron, L., Jelen, P., Watt, M.,
Scrimgeour, J.B. (1977). Maternal serum alpha-
fetoprotein measurements as an early indicator of low
birth-weight, Lancet, 2, 267–268.
Davis, R.O., Goldenberg, R.L., Boots, L., Hoffman,
H.J., Cooper, R., Cutter, G.R., Dubard, M.B., Cliver,
S.P., Smith, R.K. (1992). Elevated levels of midtrimes-
ter maternal serum á-fetoprotein are associated with
preterm delivery but not with fetal growth retarda-
tion, Am. J. Obstet. Gynecol., 167, 596–601.
Gersell, D.J., Kraus, F.T., Riffle, M.B. (1987). Diseases
of the placenta. In: Kurman, R.J. (Ed.). Blaustein’s
Pathology of the Female Genital Tract, 3rd edn, New
York: Springer-Verlag, 769–834.
Gordon, Y.B., Grudzinskas, J.G., Kitau, M.J.,
Usherwood, M.M., Letchworth, A.T., Chard, T.
(1978). Fetal wastage as a result of an alpha-
fetoprotein screening programme, Lancet, 1, 677–678.
Katz, V.L., Chescheir, N.C., Cefalo, R.C. (1990). Un-
explained elevations of maternal serum alpha-
fetoprotein, Obstet. Gynecol. Surv., 45, 719–726.
MSAFP  - 309
Merz, E., Kim-Kern, M.-S., Pehl, S. (1987). Ultrasonic
mensuration of fetal limb bones in the second and
third trimesters, J. Clin. Ultrasound., 15, 175–183.
Milunsky, A., Jick, S.S., Bruell, C.L., MacLaughlin,
D.S., Tsung, Y.K., Jick, H., Rothman, K.J., Willett,
W. (1989). Predictive values, relative risks, and overall
benefits of high and low maternal serum á-fetoprotein
screening in singleton pregnancies: new epidemiologic
data, Am. J. Obstet. Gynecol., 161, 291–297.
Robinson, L., Grau, P., Crandall, B.F. (1989). Preg-
nancy outcomes after increasing maternal serum
alpha-fetoprotein levels, Obstet. Gynecol., 74, 17–20.
Salafia, C.M., Silberman, L., Herrera, N.E., Mahoney,
M.J. (1988). Placental pathology at term associated
with elevated midtrimester maternal serum
á-fetoprotein concentration, Am. J. Obstet. Gynecol.,
158, 1064–1066.
Tamura, R.K., Sabbagha, R.E. (1980). Percentile ranks
of sonar fetal abdominal circumference measure-
ments, Am. J. Obstet. Gynecol., 138, 475–479.
Wald, N., Cuckle, H., Stirrat, G.M., Bennett, M.J.,
Turnbull, A.C. (1977). Maternal serum alpha-
fetoprotein and low birth-weight, Lancet, 2, 268–270.
Waller, D.K., Lustig, L.S., Smith, A.H., Hook, E.B.
(1993). Alpha-fetoprotein: a biomarker for pregnancy
outcome, Epidemiology, 4, 471–476.
Walters, B.N.J., Lao, T.T., Smith, V.A., Blau, K., de
Swiet, M. (1983). Severe pre-eclampsia of early onset
(Letter), Br. Med. J. (Clin. Res. Ed.), 287, 1302.
Walters, B.N.J., Lao, T., Smith, V., de Swiet, M. (1985).
á-Fetoprotein elevation and proteinuric pre-
eclampsia, Br. J. Obstet. Gynaecol., 92, 341–344.