Sie sind auf Seite 1von 7

Research ajog.

org

OBSTETRICS
The association between fetal Doppler and
admission to neonatal unit at term
Asma A. Khalil, MD, MRCOG; José Morales-Rosello, MD; Malaz Elsaddig; Naila Khan;
Aris Papageorghiou, MD, MRCOG; Amar Bhide, MD, MRCOG; Basky Thilaganathan, PhD, MRCOG

OBJECTIVE: Fetal cerebroplacental ratio is emerging as a better proxy the babies requiring neonatal unit admission (P < .05). However,
than birthweight for placental insufficiency and as a marker of fetal the estimated fetal weight percentile was not significantly different
compromise at term. The extent to which these fetal Doppler changes between those who required neonatal unit admission and those who
are related to neonatal outcomes has not been systematically assessed. did not (P ¼ .087). According to multivariate logistic regression,
The main aim of this study was to evaluate the association between cerebroplacental ratio MoM (odds ratio, 0.39; 95% confidence inter-
estimated fetal weight percentile, cerebroplacental ratio recorded at val, 0.19e0.79; P ¼ .008) and gestational age at delivery (odds ra-
34þ0e35þ6 weeks’ gestation, and neonatal unit admission at term. tio, 0.70; 95% confidence interval, 0.61e0.80; P < .001) were
significantly associated with the risk of neonatal unit admission,
STUDY DESIGN: This was a retrospective cohort study in a tertiary
whereas maternal age and birthweight percentile were not (P ¼ .183
referral center over an 11 year period from 2002 to 2012. The um-
and P ¼ .460, respectively). Irrespective of birthweight or estimated
bilical artery pulsatility index (PI), middle cerebral artery PI, and cer-
fetal weight percentile, the fetal cerebroplacental ratio appears to be
ebroplacental ratio were recorded at 34þ0e35þ6 weeks. Weight
a better predictor of the need for neonatal unit admission (P < .001).
values were converted into percentiles and Doppler parameters into
multiples of the median (MoM), adjusting for gestational age. Logistic CONCLUSION: Lower cerebroplacental ratio and gestational age at
regression analysis was performed to identify, and adjust for, potential delivery, but not fetal size, were independently associated with the
confounders. need for admission to the neonatal unit at term in a high-risk patient
group. The extent to which fetal hemodynamic assessment could be
RESULTS: We identified 2518 pregnancies in which a scan was
used to predict perinatal morbidity and optimize the timing of delivery
performed at 34þ0e35þ6 weeks and delivery occurred at or beyond
merits further investigation.
37 weeks. In the 2485 pregnancies included in the analysis, the
umbilical artery PI MoM was significantly higher, and the middle ce- Key words: birthweight, cerebroplacental ratio, Doppler, neonatal unit
rebral artery PI and cerebroplacental ratio MoM significantly lower in admission, small for gestational age

Cite this article as: Khalil AA, Morales-Rosello J, Elsaddig M, et al. The association between fetal Doppler and admission to neonatal unit at term. Am J Obstet Gynecol
2015;213:57.e1-7.

S mall for gestational age (SGA),


defined as birthweight (BW) less
than the 10th percentile for gestation, is
and do not demonstrate signs of
placental insufficiency. The clinical
definition of SGA is further confused by
average risk group for fetal growth re-
striction (FGR).4,5 However, national
guidance in the United Kingdom and the
commonly used as a proxy to identify the finding that a proportion of average United States does not recommend the
fetuses at risk of adverse outcomes sec- for gestational age (AGA) infants also fail use of fetal Doppler as a screening tool
ondary to fetal growth restriction.1,2 to meet their growth potential or, more for placental insufficiency, except when
However, the majority of SGA babies correctly, suffer from occult placental the fetus is already known to be SGA.1,2
are not pathologically growth restricted insufficiency.3,4 The antenatal diagnosis of FGR using
We and others have recently reported fetal biometry alone has recently been
that fetal Doppler assessment immedi- challenged, and the use of fetal Doppler
From the Fetal Medicine Unit, St George’s ately prior to delivery at term might be of assessment has been proposed as a
Hospital, St George’s University of London, value in detecting AGA pregnancies that potentially better marker.4,6,7
London, England, UK. are at increased risk of adverse outcome It is now increasingly routine in many
Received June 15, 2014; revised Aug. 23, 2014; from fetal hypoxemia secondary to hospitals and individual practices to
accepted Oct. 7, 2014.
placental insufficiency and failure to undertake a 34-36 week ultrasound scan
The authors report no conflict of interest. reach growth potential.4,5 Fetal Doppler for fetal assessment. There is a paucity
Corresponding author: Asma Khalil, MD, indices assessed just prior to delivery of data to support such routine assess-
MRCOG. asmakhalil79@googlemail.com
appear to be a better marker than BW ment, but nevertheless its proponents
0002-9378/$36.00 or BW percentile for adverse pregnancy are convinced of its utility. The main
ª 2015 Published by Elsevier Inc.
http://dx.doi.org/10.1016/j.ajog.2014.10.013 outcomes, such as the need for operative aim of this study was to investigate the
delivery for fetal compromise in an association between fetal Doppler and

JULY 2015 American Journal of Obstetrics & Gynecology 57.e1


Research Obstetrics ajog.org

ultrasound parameters at the 34-36 week assess the difference between the SGA abnormalities, stillbirth, or missing out-
growth scan and the need for neonatal model, which relies on fetal biometry, come data, leaving 2485 pregnancies
unit admission at term. and the placental insufficiency model, included in the analysis. The maternal
which relies on fetal hemodynamic characteristics, ultrasound, and birth
M ATERIALS AND M ETHODS assessment. This was derived from our indices in the groups requiring and not
This was a retrospective cohort study previous work in which, to define the requiring admission to the neonatal unit
in a single tertiary referral center over threshold for the failure to achieve are shown in Table 1. The prevalence of
an 11 year period from 2002 to 2012. growth potential, we calculated the fifth SGA, defined as BW less than the 10th
Cases were identified by searching centile of the CPR in the group least percentile and the fifth percentile, was
the ViewPoint database (ViewPoint likely to present with failure to achieve 25.8% and 14.0%, respectively.
5.6.8.428; ViewPoint Bildverarbeitung growth potential (those with BW greater The overall neonatal admission rate
GmbH, Weßling, Germany) in the Fetal than the 90th percentile). This value in the study cohort was 6.7%. This fell
Medicine Unit, St George’s Hospital. corresponded to 0.6765.4 with advancing gestation from 16.8%
The inclusion criteria were singleton Data on admission to the neonatal to 8.7%, 4.2% and 4.2% at 37, 38, 39,
morphologically normal fetuses born at unit were ascertained from the maternity and longer than 40 weeks, respectively.
term that had previously had an ultra- and neonatal records. The indications The most common indications for
sound scan at 34þ0e35þ6 weeks’ gesta- for admission to the neonatal unit were admission to the neonatal unit were
tion for a variety of indications such as respiratory complications (defined as respiratory complications and post-
suspected poor/excessive fetal growth, respiratory distress syndrome and tran- resuscitation observation. There were 2
reduced fetal movements, history of sient tachypnea of the newborn, use of early neonatal deaths and 1 late neonatal
SGA or large for gestational age baby, continuous positive airway pressure, death.
high midtrimester uterine artery Dop- endotracheal intubation), hypoglycemia, The scatter plot of CPR MoM values
pler indices, and gestational diabetes. sepsis, need for phototherapy, post- against BW percentiles showed a signif-
Therefore, these pregnancies were at resuscitation observation, seizures, and icant linear relationship (Figure 1: CPR
risk of fetal growth disorders. hypothermia. MoM ¼ (0.0019  BW percentile) þ
Pregnancies complicated by fetal ab- 0.9705, R2 ¼ 0.049). At the ultrasound
normality, aneuploidy, or antepartum Statistical analysis assessment, UA PI MoM was signifi-
stillbirth were excluded from the anal- Categorical data were presented as cantly higher and MCA PI and CPR
ysis. Gestational age (GA) was calculated number (percentage) and were com- MoM were significantly lower in the
from the crown-rump length measure- pared using the Fisher exact test or the neonates requiring neonatal unit ad-
ment at 11-13 weeks, and only 1 exam- c2 test. Continuous data were presented mission (P < .05, Table 1, Figure 2), but
ination (the last in the 34þ0e35þ6 as median (interquartile range [IQR]). the EFW percentile was not significantly
week window) per fetus was included The D’Agostino and Pearson Omnibus different between the 2 study groups
in the analysis.8 Routine fetal biometry test was used to assess the normality of (P ¼ .087).
was performed according to a stand- the data. Nonparametric analysis using The results of the regression analysis
ard protocol and the estimated fetal the Mann-Whitney U test was then used are shown in Tables 2 and 3. According
weight (EFW) calculated using Had- to compare continuous data between the to a multivariate logistic regression, CPR
lock’s formula.9 Data on pregnancy study groups. A logistic regression anal- MoM, GA at delivery, maternal ethni-
outcomes were collected from hospital ysis was performed to identify, and city, and male gender were significantly
obstetric and neonatal records. adjust for, potential confounders. Both associated with the risk of neonatal unit
The umbilical artery (UA) and middle unadjusted and adjusted odds ratios admission (P < .05), whereas maternal
cerebral artery (MCA) Doppler were (ORs) were calculated. The analysis was age and BW percentile were not (P ¼
recorded using color Doppler, and the performed using the statistical software .183 and P ¼ .460, respectively). When
pulsatility index (PI) calculated accord- packages SPSS 18.0 (SPSS Inc, Chicago, divided into 4 groups according to a
ing to a standard protocol.10,11 The cer- IL), Stata 11 (release 11.2; College Sta- combination of a BW cutoff of the 10th
ebroplacental ratio (CPR) was calculated tion, TX), and GraphPad Prism 5.0 for percentile and an optimal CPR cutoff of
as the simple ratio of the MCA PI to the Windows (InStat; GraphPad Software 0.6765 MoM, the rates of neonatal unit
UA PI.12 All Doppler indices were con- Inc, San Diego, CA). admission were significantly different
verted into multiples of the median (P < .001, Figure 3).
(MoM) correcting for GA using refer- R ESULTS The incidence of admission to the
ence ranges, and BW values were con- We identified 2518 pregnancies with neonatal unit in the AGA neonates with
verted into percentiles.4,13,14 fetal Doppler assessment at 34þ0e35þ6 low CPR (9.8%) was almost double that
The study cohort was divided into 4 weeks, in which the delivery occurred at in the group of AGA with normal CPR
groups according to a combination of a or beyond 37 weeks’ gestation. We (5.5%). However, the difference in the
BW cutoff of the 10th percentile and an excluded 33 pregnancies (1.3%) because admission rates between the groups of
optimal CPR cutoff of 0.6765 MoM4 to they had aneuploidy, major structural AGA with low CPR and AGA with

57.e2 American Journal of Obstetrics & Gynecology JULY 2015


ajog.org Obstetrics Research

TABLE 1
Maternal and fetal characteristics in the study cohort according to admission to the neonatal unit
Neonatal unit admission No neonatal unit admission
Characteristic (n [ 167) (n [ 2318) P value
Maternal age, y, median (IQR) 31.0 (27.0e34.0) 31.0 (26.0e34.0) .133
Ethnicity .276
White, n (%) 101 (60.5) 1230 (53.1)
African, n (%) 30 (18.0) 416 (18.0)
Asian, n (%) 30 (18.0) 522 (22.5)
Mixed, n (%) 4 (2.4) 100 (4.3)
Other, n (%) 2 (1.2) 50 (2.2)
Gestational age at ultrasound (wks), median (IQR) 34.9 (34.4e35.4) 35.0 (34.4e35.6) .113
Estimated fetal weight (g), median (IQR) 2356 (2085e2651) 2426 (2183e2690) .025
Estimated fetal weight percentile, median (IQR) 28.92 (11.45e54.01) 32.89 (15.60e55.77) .087
UA PI 0.95 (0.85e1.08) 0.91 (0.80e1.03) .008
UA PI MoM 1.01 (0.89e1.14) 0.97 (0.85e1.09) .014
MCA PI 1.73 (1.50e1.98) 1.80 (1.60e2.02) .002
MCA PI MoM 1.17 (1.00e1.30) 1.21 (1.08e1.37) .001
Cerebroplacental ratio 1.87 (1.50e2.25) 2.00 (1.70e2.31) < .001
Cerebroplacental ratio MoM 0.95 (0.76e1.15) 1.02 (0.87e1.18) < .001
Gestational age at delivery (wks), median (IQR) 38.9 (37.7e40.0) 39.6 (38.6e40.6) < .001
Birthweight (g), median (IQR) 2955 (2500e3470) 3140 (2780e3520) < .001
Birthweight percentile, median (IQR) 19.01 (4.67e57.80) 28.32 (10.09e58.43) .005
Fetal sex male, n (%) 98 (59.4) 1140 (49.3) .012
IQR, interquartile range; MCA, middle cerebral artery; PI, pulsatility index; MoM, multiples of median; UA, umbilical artery.
Khalil. Doppler and neonatal unit admission. Am J Obstet Gynecol 2015.

normal CPR was not statistically signif- IQR, 1.06e1.35), and CPR MoM significantly higher in those with low
icant (P ¼ .203). This could be related (0.84, IQR, 0.67e0.99 vs 0.94, IQR, CPR MoM compared with those with
to the small number of cases in these 0.80e1.09). BW percentile, MCA PI normal CPR MoM (30.2% vs 12.2%, P ¼
2 groups because admission to the MoM, and CPR MoM were significantly .004).
neonatal unit in healthy newborns at associated with the risk of neonatal unit
term is a relatively uncommon event. admission (P < .001, Table 4), whereas
We also investigated the association the UA PI MoM and EFW percentile C OMMENT
between fetal Doppler parameters and were not (P ¼ .115 and P ¼ .213, The findings of this study suggest that
the risk of admission to the neonatal unit respectively). neonates that were admitted to the
in the pregnancies that were SGA (BW According to the multivariate logistic neonatal unit at term had significantly
less than 10th percentile, n ¼ 640) in regression in SGA neonates, both BW lower CPR at 34-36 weeks, whereas both
which the rate of admission to the percentile (OR, 0.87; 96% confidence ultrasound EFW and BW percentiles
neonatal unit was 10%. Compared with interval [CI], 0.78e0.96; P ¼ .008) and were not significantly different from
those not requiring admission, neonates CPR MoM (OR, 0.21; 96% CI, those not admitted. A multivariate lo-
requiring neonatal unit admission had 0.06e0.77; P ¼ .018) were significantly gistic regression demonstrated that CPR
significantly lower median BW percen- and independently associated with the MoM at 34-36 weeks was almost twice as
tile (2.70, IQR, 0.73e5.89 vs 4.69, risk of neonatal unit admission. The likely as GA at delivery to determine the
IQR, 2.34e7.38), GA at delivery in prevalence of SGA, defined as BW less need for neonatal unit admission. It was
weeks (38.35, IQR, 37.45e39.40 vs than the third percentile, was 8.5%. The also the case that, among the SGA neo-
39.10, IQR, 38.30e40.10), MCA PI neonatal unit admission rate in this nates at term, those with a lower CPR
MoM (1.11, IQR, 0.97e1.24 vs 1.19, cohort was 15.8%, which was MoM at 34-36 weeks had a higher risk

JULY 2015 American Journal of Obstetrics & Gynecology 57.e3


Research Obstetrics ajog.org

of a neonatal unit admission compared population, an abnormal CPR at term


FIGURE 1 with those with normal CPR MoMs. predicts neurobehavioral problems at
Scatter plot of CPR MoM values 18 months of age on internalizing and
against BW percentile somatic complaints scales.
Findings in term pregnancies
The novel observation in our cohort is
Findings in term SGA pregnancies
that CPR, but not BW, percentiles,
The role of fetal Doppler in term and
measured at 34-36 weeks’ gestation, are
near-term pregnancies may currently
significantly and independently associ-
be underestimated because of the rarity
ated with the need for neonatal unit
of abnormal UA Doppler indices at
admission at term. In a recent prospec-
this gestation and the fact that FGR-
tive study including 400 term pregnan-
related perinatal morbidity has been
cies (37-42 weeks), the investigators
noted to occur even when the UA blood
assessed fetal biometry and Doppler
flow is normal.16-18 In our cohort the
parameters immediately before estab-
rate of neonatal unit admission was
lished labor.5 Fetuses with CPR below
almost 3 times higher in SGA neonates
the 10th percentile were 6 times more
with low CPR MoM compared with
likely to be delivered by cesarean
those SGA neonates with normal CPR
delivery for presumed fetal compromise
Scatter plot of CPR MoM values against BW MoM (23% vs 8%).
than those with a CPR above the10th
percentile, showing a significant linear rela- The role of the MCA Doppler in SGA
percentile.5 Even after the exclusion of
tionship (CPR MoM ¼ [0.0019  BW cases with a BW below the 10th per-
pregnancies at term has been investi-
percentile] þ 0.9705, R2 ¼ 0.049). gated in previous studies.16,18-22 Chang
centile, the rate of cesarean delivery for
BW, birthweight; CPR, cerebroplacental ratio; MoM, multiple of et al18 reported that fetal Doppler PI
median. presumed fetal compromise remained
ratios recorded in the third trimester
Khalil. Doppler and neonatal unit admission. Am J Obstet significantly higher in fetuses with a
(aortic/middle cerebral and renal/mid-
Gynecol 2015. CPR less than the 10th percentile.5 In the
dle cerebral PI ratios) were superior to
Generation R Study (Rotterdam,
predelivery estimates of fetal size in
The Netherlands, 2003-2007), Roza
predicting suboptimal perinatal out-
et al15 reported that, even in the general
come in small fetuses delivering at term.
The suboptimal perinatal outcome in
the latter study was defined as acidemia
FIGURE 2
at birth, fetal distress requiring emer-
Cerebroplacental ratio MoM values and percentiles
gency cesarean delivery in labor, or
admission to the neonatal unit.18
Hershkovitz et al16 also demonstrated
an association between MCA blood
flow redistribution in structurally nor-
mal fetuses with an EFW below the fifth
percentile and an increased risk of ce-
sarean delivery and need for neonatal
unit admission. CPR has been shown to
correlate better with adverse outcome
than does UA and MCA Doppler.23,24

Findings in preterm SGA pregnancies


In pregnancies with SGA neonates, BW
percentiles and CPR MoMs at 34-36
weeks’ gestation, but not EFW percen-
tiles at 34-36 weeks, were independently
Box and whisker plots of cerebroplacental ratio MoM values and birthweight percentiles in newborns associated with neonatal unit admission
who were admitted to the neonatal unit and those who were not. The horizontal line in the box at term. These findings are consistent
represents the median, the box represents the IQR, and the whiskers indicate the minimum and with the recent Prospective Observa-
maximum values. tional Trial to Optimize Pediatric Health
IQR, interquartile range; MoM, multiple of median. in Intrauterine Growth Restriction
Khalil. Doppler and neonatal unit admission. Am J Obstet Gynecol 2015. study, in which in 1200 pregnancies
with EFW less than the 10th percentile

57.e4 American Journal of Obstetrics & Gynecology JULY 2015


ajog.org Obstetrics Research

TABLE 2 FIGURE 3
Results of the logistic regression analysis of factors associated with the Rates of neonatal unit admission
risk of admission to the neonatal unit in the 4 study groups
Risk factor Unadjusted OR 95% CI P value
Maternal characteristics
Maternal age, y 1.02 (1.00e1.05) .106
Maternal ethnicity 0.83 (0.70e0.98) .026
Ultrasound indices
EFW percentile 1.00 (0.99e1.00) .245
UA PI MoM 3.00 (1.23e7.28) .016
MCA PI MoM 0.26 (0.12e0.55) < .001 The rates of neonatal unit admission in the
4 study groups according to a combination of a
Cerebroplacental ratio MoM 0.83 (0.70e0.98) < .001
BW cutoff of the 10th percentile and a CPR
Fetal sex (female) 0.67 (0.48e0.92) .013 cutoff of 0.6765 MoM.
Indices at birth AGA, average for gestational age; BW, birthweight; CPR, cere-
broplacental ratio; SGA, small for gestational age.
GA at delivery, wk 0.67 (0.59e0.76) < .001 Khalil. Doppler and neonatal unit admission. Am J Obstet
Gynecol 2015.
BW percentile 0.99 (0.99e1.00) .079
BW, birthweight; CI, confidence interval; EFW, estimated fetal weight; GA, gestational age; MCA, middle cerebral artery;
MoM, multiples of median; OR, odds ratio; PI, pulsatility index; UA, umbilical artery.
difference may be due to variations in
Khalil. Doppler and neonatal unit admission. Am J Obstet Gynecol 2015.
gestation at assessment and delivery
as well as in neonatal unit admission
prospectively followed up, 95% of those parameter that was significantly associ- policies in the 2 studies. Despite this
with abdominal circumference below ated with adverse outcome was EFW discrepancy, the findings of both studies
the 10th percentile had a normal preg- below the third percentile.25 Despite the question the currently accepted defini-
nancy outcome.25 Importantly, the au- fact that the authors stated in the tion of FGR and emphasize the impor-
thors noted that the presence of an methodology that MCA Doppler had tance of fetal hemodynamic assessment
abnormal UA Doppler was significantly been performed, these data were not as a predictor of adverse pregnancy
associated with adverse outcome, irre- reported in the results.25 The neonatal outcome in all fetal weight categories.
spective of EFW or abdominal circum- admission rate in that SGA cohort was
ference measurement.25 Furthermore, 28%, compared with 10% in our cohort Strengths and limitations
the only sonographic weight-related of 640 pregnancies with SGA. This The strengths of our study include the
large number of pregnancies, adjusting
for GA, ascertainment of the outcome
TABLE 3 data, and the narrow GA range at ultra-
Results of the multivariate logistic regression analysis of factors sound assessment. However, the retro-
associated with the risk of admission to the neonatal unit spective design, lack of detailed robust
Risk factor Adjusted OR 95% CI P value data on the intrapartum events, and the
higher proportion of SGA (25.8%) in the
Maternal characteristics
study population limit the generaliz-
Maternal age, y 1.02 (0.99e1.05) .183 ability of our findings. The higher prev-
Maternal ethnicity 0.81 (0.68e0.97) .020 alence of SGA in our cohort could be
explained by the fact that pregnancies at
Ultrasound indices
risk of SGA were more likely to have an
Cerebroplacental ratio MoM 0.39 (0.19e0.78) .008 ultrasound assessment at 34-36 weeks’
Fetal sex (female) 0.67 (0.48e0.94) .020 gestation. However, we see this as a
Indices at birth
strength because despite the bias toward
lower BW and higher prevalence of SGA,
GA at delivery, wks 0.70 (0.61e0.80) < .001 CPR was more strongly associated with
BW percentile 1.00 (0.99e1.00) .460 neonatal unit admission. A high pro-
BW, birthweight; CI, confidence interval; GA, gestational age; MoM, multiples of median; OR, odds ratio. portion of neonatal unit admissions are
Khalil. Doppler and neonatal unit admission. Am J Obstet Gynecol 2015. short term, with full recovery and
discharge home. Prolonged neonatal

JULY 2015 American Journal of Obstetrics & Gynecology 57.e5


Research Obstetrics ajog.org

REFERENCES
TABLE 4 1. American College of Obstetricians and
Factors associated with neonatal unit admission in SGA newbornsa Gynecologists. Intrauterine growth restriction.
Risk factor Unadjusted OR 95% CI P value ACOG practice bulletin no. 12. Int J Gynecol
Obstet 2001;72:85-96.
Maternal characteristics 2. Royal College of Obstetricians and Gynae-
Maternal age, y 1.03 0.98e1.07 .255 cologists. The investigation and management of
the small-for-gestational-age fetus. Green-top
Maternal ethnicity 0.69 0.53e0.91 .008 guideline number 31. London (UK): Royal Col-
lege of Obstetricians and Gynaecologists; 2013.
Ultrasound indices
3. Maulik D. Fetal growth compromise: defini-
EFW percentile 0.98 0.96e1.01 .213 tions, standards, and classification. Clin Obstet
Gynecol 2006;49:214-8.
UA PI MoM 3.06 0.76e12.27 .115 4. Morales-Roselló J, Khalil A, Morlando M,
MCA PI MoM 0.88 0.24e0.32 < .001 Papageorghiou A, Bhide A, Thilaganathan B.
Changes in fetal Doppler as a marker of failure to
Cerebroplacental ratio MoM 0.12 0.03e0.42 .001 reach growth potential at term. Ultrasound
Fetal sex (female) 0.70 0.42e1.18 .184 Obstet Gynecol 2014;43:303-10.
5. Prior T, Mullins E, Bennett P, Kumar S. Pre-
Indices at birth diction of intrapartum fetal compromise using
GA delivery, wks 0.68 0.55e0.86 .001 the cerebroumbilical ratio: a prospective obser-
vational study. Am J Obstet Gynecol 2013;208:
BW percentile 0.84 0.76e0.92 < .001 124.e1-6.
BW less than fifth percentile 2.13 1.22e3.74 .008 6. Sebire NJ. Detection of fetal growth restriction
at autopsy in non-anomalous stillborn infants.
BW, birthweight; CI, confidence interval; EFW, estimated fetal weight; GA, gestational age; MCA, middle cerebral artery; Ultrasound Obstet Gynecol 2014;43:241-4.
MoM, multiples of median; OR, odds ratio; PI, pulsatility index; SGA, small for gestational age (defined as birthweight less than 7. Figueras F, Gratacós E. Update on the diag-
10th percentile); UA, umbilical artery.
a
nosis and classification of fetal growth restriction
Determined by logistic regression analysis. and proposal of a stage-based management
Khalil. Doppler and neonatal unit admission. Am J Obstet Gynecol 2015. protocol. Fetal Diagn Ther 2014;36:86-98.
8. Robinson HP, Fleming JE. A critical evalua-
tion of sonar “crown-rump length” measure-
ments. BJOG 1975;82:702-10.
9. Hadlock FP, Harrist RB, Sharman RS,
intensive care unit admission and These findings support the assertion Deter RL, Park SK. Estimation of fetal weight
neonatal morbidity are the subjects of a that fetal Doppler assessment could be with the use of head, body, and femur mea-
future study. of value in detecting fetal hypoxemia surements, a prospective study. Am J Obstet
Approximately 10% of term babies secondary to placental insufficiency Gynecol 1985;151:333-7.
10. Acharya G, Wilsgaard T, Berntsen GK,
may require admission to the neonatal and failure to reach their genetic growth Maltau JM, Kiserud T. Reference ranges for
unit.26 The commonest indications potential in apparently AGA pregnan- serial measurements of umbilical artery Doppler
include respiratory complications, post- cies. AGA pregnancies can demonstrate indices in the second half of pregnancy. Am J
resuscitation observation, and hypogly- fetal cerebral and placental blood flow Obstet Gynecol 2005;192:937-44.
11. Bahlmann F, Reinhard I, Krummenauer F,
cemia. These adverse events are more redistribution indicative of an increased
Neubert S, Macchiella D, Wellek S. Blood flow
likely to occur in FGR babies, or more risk of adverse pregnancy outcome as a velocity waveforms of the fetal middle cerebral
correctly, babies compromised by pla- consequence of fetal hypoxemia sec- artery in a normal population: reference values
cental insufficiency. Admission to the ondary to placental insufficiency.4 from 18 weeks to 42 weeks of gestation.
neonatal unit at term is usually an un- Despite this significant association J Perinat Med 2002;30:490-501.
12. Baschat AA, Gembruch U. The cere-
expected event for both parents and between fetal Doppler and the need for
broplacental Doppler ratio revisited. Ultrasound
health care professionals. It also repre- neonatal unit admission at term, its Obstet Gynecol 2003;21:124-7.
sents a burden on health care resources predictive accuracy as an isolated marker 13. Morales Roselló J, Hervás Marín D, Fillol
and reflects the quality of intrapartum is likely to be poor because of the con- Crespo M, Perales Marín A. Doppler changes
care. For these reasons, it is one of the founding effect of the process of labor in the vertebral, middle cerebral, and umbilical
arteries in fetuses delivered after 34 weeks:
national health care quality measures and birth.27 Whether fetal Doppler
relationship to severity of growth restriction.
in many Western countries, including assessment in AGA pregnancies is pre- Prenat Diagn 2012;32:960-7.
the United Kingdom and the United dictive of perinatal morbidity or neuro- 14. Yudkin PL, Aboualfa M, Eyre JA,
States. We have shown that lower CPR developmental delay is unknown. The Redman CW, Wilkinson AR. New birthweight
at 34-36 weeks and GA at delivery, but extent to which fetal hemodynamic and head circumference percentiles for gesta-
tional ages 24 to 42 weeks. Early Hum Dev
not BW percentiles at 34-36 weeks, were assessment could be used to optimize
1987;15:45-52.
independently associated with the need the timing of delivery and reduce neu- 15. Roza SJ, Steegers EA, Verburg BO, et al.
for admission to the neonatal unit at rodevelopmental impairment merits What is spared by fetal brain-sparing? Fetal cir-
term. further investigation. - culatory redistribution and behavioral problems

57.e6 American Journal of Obstetrics & Gynecology JULY 2015


ajog.org Obstetrics Research
in the general population. Am J Epidemiol 20. Odibo AO, Riddick C, Pare E, perinatal outcome in intrauterine growth restric-
2008;168:1145-52. Stamilio DM, Macones GA. Cerebroplacental tion. Am J Obstet Gynecol 1999;180:750-6.
16. Hershkovitz R, Kingdom JC, Geary M, Doppler ratio and adverse perinatal outcomes 24. Gramellini D, Folli MC, Raboni S, Vadora E,
Rodeck CH. Fetal cerebral blood flow redistri- in intrauterine growth restriction: evaluating Merialdi A. Cerebral-umbilical Doppler ratio as a
bution in late gestation: identification of com- the impact of using gestational age-specific predictor of adverse perinatal outcome. Obstet
promise in small fetuses with normal umbilical reference values. J Ultrasound Med 2005;24: Gynecol 1992;79:416-20.
artery Doppler. Ultrasound Obstet Gynecol 1223-8. 25. Unterscheider J, Daly S, Geary MP, et al.
2000;15:209-12. 21. Fong KW, Ohlsson A, Hannah ME, et al. Optimizing the definition of intrauterine growth
17. Beattie RB, Dornan JC. Antenatal screen- Prediction of perinatal outcomes in fetuses restriction: the multicenter prospective PORTO
ing for intrauterine growth retardation with um- suspected to have intrauterine growth restric- Study. Am J Obstet Gynecol 2013;208:290.
bilical artery Doppler ultrasonography. BJOG tion: Doppler US study of fetal cerebral, renal e1-6.
1989;298:631-5. and umbilical arteries. Radiology 1999;213: 26. Alkiaat A, Hutchinson M, Jacques A,
18. Chang TC, Robson SC, Spencer JAD, 681-9. Sharp MJ, Dickinson JE. Evaluation of the fre-
Gallivan S. Prediction of perinatal morbidity at 22. Harrington K, Carpenter RG, Nguyen M, quency and obstetric risk factors associated
term in small fetuses: comparison of fetal growth Campbell S. Changes observed in Doppler with term neonatal admissions to special care
and Doppler ultrasound. BJOG 1994;101:422-7. studies of the fetal circulation in pregnancies units. Aust N Z J Obstet Gynaecol 2013;53:
19. Cruz-Martínez R, Figueras F, Hernandez- complicated by pre-eclampsia or the delivery of 277-82.
Andrade E, Oros D, Gratacos E. Fetal brain a small-for-gestational-age baby. I. Cross- 27. Martinez-Biarge M, Diez-Sebastian J,
Doppler to predict cesarean delivery for non- sectional analysis. Ultrasound Obstet Gynecol Wusthoff CJ, Mercuri E, Cowan FM. Antepartum
reassuring fetal status in term small-for- 1995;6:19-28. and intrapartum factors preceding neonatal
gestational-age fetuses. Obstet Gynecol 2011; 23. Bahado-Singh RO, Kovanci E, Jeffres A, hypoxic-ischemic encephalopathy. Pediatrics
117:618-26. et al. The Doppler cerebroplacental ratio and 2013;132:e952-9.

JULY 2015 American Journal of Obstetrics & Gynecology 57.e7