UOG Journal Club: Optimal risk assessment of small-for-gestational-age fetuses and changes in fetal Doppler indices as markers of failure to reach growth potential

UOG Journal Club: March 2014

Optimal risk assessment of small-for-gestational-age fetuses using 31–34-week biometry in a low-risk population

J. J. Stirnemann, G. Benoist, L. J. Salomon, J.-P. Bernard and Y. VilleVolume 43, Issue 3, Date: March 2014, pages 311-316

Changes in fetal Doppler indices as a marker of failure to reach growth potential at term

J. Morales-Roselló, A. Khalil, M. Morlando, A. Papageorghiou,A.Bhide and B. Thilaganathan

Volume 43, Issue 3, Date: March 2014, pages 303-310

Journal Club slides prepared by Dr Aly Youssef(UOG Editor for Trainees)

UOG Journal Club: March 2014

Optimal risk assessment of small-for-gestational-age fetuses

using 31–34-week biometry in a low-risk population

J. J. Stirnemann, G. Benoist, L. J. Salomon,J.-P. Bernard and Y. Ville


Introduction

• Late-pregnancy intrauterine growth restriction (IUGR) remains a leading cause of unanticipated perinatal death and morbidity after 34 weeks’ gestation

• The detection and follow-up of fetuses at risk are necessary for optimal management and planning of delivery

• Estimated fetal weight (EFW) using a cross-sectional age-specific percentile as a selection criterion of altered growth remains the most widely used method to prenatally assess the likelihood of IUGR


Stirnemann et al., UOG 2014

To compare the performance of traditional growth charts for

EFW and a validated pragmatic probabilistic approach using

biometry at 31–34weeks’ gestation to screen for late

pregnancy small-for-gestational age (SGA) fetuses in

a low-risk population.



Objective

Training dataset

•Records of 7755 women presenting at 31–34 weeks following normal aneuploidy screening in the 1st or 2nd trimester, and normal 20-24 wks scan.

•Only cases with known birth weight and gestational age at delivery, with gestational age at delivery ≥37 weeks were included.

•Prenatal and postnatal malformations and cases with absent or reversed diastolic flow in the umbilical artery were excluded.

Methods

Validation dataset

•1725 women recruited at 11-14 weeks onwards.

• At 31-34 weeks biparietal diameter, head circumference, abdominal circumference and femur length were measured.

Methods

• A logistic regression model was built for each type of ultrasound measurement, computed using locally-derived growth charts.

• The results were displayed in terms of false-positive rate and detection rate of SGA, bypassing the intermediate step of EFW calculation.

• The potential additional value of maternal characteristics such as smoking status, parity, body mass index (BMI) and age were also investigated.

• External validation was performed by comparing observed prevalence and predictions given by the model in the second independent dataset.

This model was compared with the routinely used multistep approach involving estimation of fetal weight and consecutive screening by a percentile cut-off.

The probabilistic model

Defining SGA

•Analysis of data was conducted for different definitions of SGA including birth weight <3rd, 5th, and 10th centiles, and birth weight <2500 grams.

Probabilistic approach

Model using EFW P value

Area under the ROC curve

0.832 0.828 0.007

ResultsAccuracy of predicting birth weight <10th centile

• For a 10% false-positive rate, the probabilistic approach yielded a 51% detection rate for birth weight<10th centile, compared to the 32% and 48% detection by the 10th centile cut-off of two EFW reference charts.

• Adding maternal characteristics significantly improved detection of SGA by 2%, from 51% to 53%.



• As the main goal of ultrasound biometry is to detect abnormal growth, the suggested probabilistic model using biometric measurements seems a reasonable and pragmatic approach.

• This may unify screening procedures and simplify counselling at 31-34 weeks.

• In addition, it allows the direct incorporation of maternal-specific characteristics, thus having the potential to replace customized growth charts.

• This screening strategy is however intended for low-risk population and is not validated in high risk pregnancies, or outside the 31-34 weeks window.

Conclusion



Weaknesses• Some important maternal covariates, such as ethnicity,

were not recorded.

• The sample size may still limit the accuracy of parameter estimation for very low birth weights.

• This approach is not validated for high risk pregnancies.



Discussion points• Should low-risk women be offered an ultrasound scan with fetal

weight estimation in the third trimester?

• Would the probabalistic method be superior to the use of customized growth charts to detect SGA at term?

• Which cut-off of estimated fetal weight/fetal biometric measurements should be used to define IUGR?

• How should pregnancies with SGA fetuses at term with normal Doppler indices be managed?

• How should pregnancies with AGA fetuses at term with abnormal Doppler indices be managed?

UOG Journal Club: March 2014Changes in fetal Doppler indices as a marker of failure to

reach growth potential at term

J. Morales-Roselló, A. Khalil, M. Morlando, A. Papageorghiou, A. Bhide and B. Thilaganathan


SGA = fetus with EFW <10th centile (i.e. small-for-gestational age).FGR = fetus unable to achieve its genetically determined size as a consequence of placental insufficiency (typically defined as SGA with evidence of placental insufficiency.

• The standard definitions of both SGA and FGR exclude apparently appropriate for gestational age (AGA) infants that are growth restricted (e.g. birth-weight (BW) on the 40th centile with genetic potential to be born on the 80th centile).

• To date, the identification of AGA fetuses affected by occult chronic placental insufficiency, fetal hypoxemia and failure to reach growth potential (FRGP) remains challenging.

Background

Fetal doppler changes as a marker of failure to reach growth potential at termMorales-Roselló et al., UOG 2014

To determine whether AGA fetuses at term exhibit changes in

middle cerebral artery (MCA) and umbilical artery (UA) Doppler

indices that may be of value in identifying those that are

affected by placental insufficiency and subsequent FRGP.


Objective

Retrospective cohort in a tertiary centre from 2002-2012.Inclusion criteria

•Singleton pregnancies.•Morphologically normal and term fetuses.•Ultrasound performed within 14 days before date of delivery.

Methods


Exclusion criteria•Fetal abnormality.•Aneuploidy.•Antepartum stillbirth.

• UA and MCA were examined using color Doppler and the pulsatility index (PI) was measured.

• Cerebroplacental ratio (CPR) was calculated = MCA PI / UA PI.• All Doppler indices were converted into multiples of the median (MoM) for

gestational age.• Doppler PI MoM values were represented in scatter graphs according to BW

centile, and linear regression analysis was calculated evaluating the presence of statistical significance.

• Doppler measurements were then grouped according to BW quartiles and compared.

Methods


• As the group of largest fetuses (>90th centile) was assumed to include those least likely to be FRGP, the CPR 5th centile from this group (optimal CPR) was preliminarily established as the normality threshold to indicate placental insufficiency.

• The proportion of fetuses with FRGP was estimated in each group by subtracting the proportion of fetuses with a CPR below this optimal CPR limit.

Results

Linear regression analysis showed

that AGA fetuses with lower

BW centiles had significantly:

1.higher UA PI MoM

2.lower MCA PI MoM

3.lower CPR MoM


Cerebroplacental ratio MoM

BW centiles

MC

A P

I/UA

PI M

oM

Results

Compared to fetuses >75th BW

centile, AGA fetuses in the

lower quartiles had lower CPR

MoM, suggesting that some of

these pregnancies were

affected by placental

insufficiency and FRGP.


Cerebroplacental ratio MoM intervals

BW centiles

CP

R M

oM

BW centile groups

Results: proportion of fetuses with failure to reach growth potential (FRGP)(% of fetuses with FRGP were calculated after subtracting those cases with CPR MoM <5th

centile observed in the group with BW >90th centile)


Pro

port

ion

of fe

tuse

s w

ith F

RG

P

BW centile% of fetuses with

FRGP

75-90th centile 1%

50-75th centile 1.7%



• The data presented in this study demonstrate that in term AGA pregnancies, Doppler indices suggestive of fetal hypoxemia are more prevalent in the lower AGA BW centiles.

• The study findings imply that Doppler indices have the potential to identify AGA fetuses that are affected by placental insufficiency and failing to reach their genetic growth potential, as evidenced by blood flow redistribution.

• This finding challenges the conventional paradigm that only SGA fetuses are at risk of placental insufficiency, fetal hypoxemia and FRGP.

• It is still unknown whether the degree of placental insufficiency leading to FRGP in these AGA fetuses is predictive for perinatal complications and childhood developmental problems as seen in SGA and IUGR neonates.

Discussion


Discussion points• How should women at term with appropriate for gestational age fetuses

and evidence of circulatory redistribution be managed?

• Which cut-off of cerebroplacental ratio should be used to define fetal blood flow redistribution?

• Should middle cerebral artery and umbilical artery be performed in women with appropriate for gestational age fetuses?

Future studies are needed to evaluate the performance of the CPR in AGA fetuses in the prediction of neonatal neurodevelopmental impairment, with the aim of optimizing the timing of delivery and reducing long-term neonatal handicap.

Future perspectives


UOG Journal Club: Optimal risk assessment of small-for-gestational-age fetuses and changes in fetal Doppler indices as markers of failure to reach growth potential

Health & Medicine