1. UOG Journal Club: August 2017
Customized vs population-based growth charts to identify neonates at risk of adverse outcome: systematic review and Bayesian meta-analysis of observational studies
G. Chiossi, C. Pedroza, M. M. Costantine, V. T. T. Truong, G. Gargano and
G. R. Saade
Volume 50, Issue 2, Date: August (pages 156–166)
Journal Club slides prepared by Dr Yael Raz
(UOG Editor for Trainees)

2. Customized vs population-based growth charts to identify neonates at risk of adverse outcome
G Chiossi et al. UOG 2017
Introduction
Birth weight is determined by a combination of genetic, intrauterine and environmental influences.
Infants born SGA (birth weight <10th centile) or LGA (birth weight >90th centile) are at increased risk of adverse health outcomes.
Population-based norms do not differentiate between abnormal growth and constitutionally large or small, but otherwise healthy, fetuses.
Relying on these norms may lead to misclassification of birth weight and over- or underdiagnosis of fetal growth abnormalities.

3. Customized vs population-based growth charts to identify neonates at risk of adverse outcome
G Chiossi et al. UOG 2017
Introduction
Customized norms model optimal fetal growth by accounting for individual variables that are known to affect growth
Customized norms allow assessment of deviations from a presumed ideal fetal growth potential, rather than deviations from a poppulation-based norm.
Early studies suggested that customized norms can better identify neonates at risk BUT more recent reports do not support this conclusion.

4. Customized vs population-based growth charts to identify neonates at risk of adverse outcome
G Chiossi et al. UOG 2017
Aim of the study
To determine which classification (customized or population-based growth charts) has the strongest association with adverse outcomes when birth weight is outside the norm.

5. Methods Studies were included if they
Customized vs population-based growth charts to identify neonates at risk of adverse outcome
G Chiossi et al. UOG 2017
Methods
Studies were included if they
compared adverse neonatal outcomes
compared adverse maternal outcomes
Assessed both SGA or LGA neonates
Used both customized and population-based norms
A random-effects Bayesian meta-analysis was performed to compute pooled odds ratios (ORs) of adverse outcomes among SGA and LGA pregnancies, classified according to population-based or customized norms, when compared with non-SGA and non-LGA neonates

6. Customized vs population-based growth charts to identify neonates at risk of adverse outcome
G Chiossi et al. UOG 2017
Results

7. Customized vs population-based growth charts to identify neonates at risk of adverse outcome
G Chiossi et al. UOG 2017
Results
Customized centiles were obtained according to the approach described by Gardosi et al. which takes into account maternal height, weight, parity, ethnic origin and gestational age at delivery.
One study referred to the more complex model by Bukowski et al. that also takes into consideration prior pregnancies, marital status, maternal education and first- and second-trimester screening results.
Population-based centiles were derived from sex-adjusted standards specific for the population enrolled in each study.

8. Intrauterine Fetal Death in SGA neonates
Customized vs population-based growth charts to identify neonates at risk of adverse outcome
G Chiossi et al. UOG 2017
Results
Intrauterine Fetal Death in SGA neonates
Population-based
Customized

9. Neonatal Death in SGA neonates
Customized vs population-based growth charts to identify neonates at risk of adverse outcome
G Chiossi et al. UOG 2017
Results
Neonatal Death in SGA neonates
Population-based
Customized

10. Perinatal Death in SGA neonates
Customized vs population-based growth charts to identify neonates at risk of adverse outcome
G Chiossi et al. UOG 2017
Results
Perinatal Death in SGA neonates
Population-based
Customized

11. NICU admission in SGA neonates
Customized vs population-based growth charts to identify neonates at risk of adverse outcome
G Chiossi et al. UOG 2017
Results
NICU admission in SGA neonates
Population-based
Customized

12. Shoulder dystocia in LGA neonates
Customized vs population-based growth charts to identify neonates at risk of adverse outcome
G Chiossi et al. UOG 2017
Results
Shoulder dystocia in LGA neonates
Population-based
Customized
LGA pregnancies had a similar risk of perinatal death, NICU admission, maternal third- and fourth-degree perineal lacerations and neonatal hypoglycemia when compared with non-LGA pregnancies – by population or customized norms

13. Discussion SGA neonates:
Customized vs population-based growth charts to identify neonates at risk of adverse outcome
G Chiossi et al. UOG 2017
Discussion
SGA neonates:
Appear to be at greater risk for IUFD, neonatal death, perinatal death and NICU admission than do non-SGA neonates, using either customized or population-based growth charts.
A customized approach led to higher ORs for the studied outcomes, suggesting that it may better identify patients at risk.
However, the strength of any conclusions regarding the superiority of this method is hindered by the overlapping CIs.

14. Discussion LGA neonates:
Customized vs population-based growth charts to identify neonates at risk of adverse outcome
G Chiossi et al. UOG 2017
Discussion
LGA neonates:
Did not appear to be at increased risk for perinatal death, NICU admission, maternal 3rd/4th-degree perineal lacerations or neonatal hypoglycemia, using either customized or population-based growth charts.
Both approaches indicated an increased risk for shoulder dystocia
These findings may be because of the limited number of studies addressing these outcomes.
It may be that absolute weight is more important in determining outcome in these large fetuses compared with an individualized approach.

15. Customized vs population-based growth charts to identify neonates at risk of adverse outcome
G Chiossi et al. UOG 2017
Discussion
IUGR constitutes the largest single at-risk category among normally formed fetuses experiencing in-utero demise and is also associated with increased risks of perinatal mortality, morbidity and long-term adverse health outcomes.
Most cases of IUGR are due to late-onset (term) placental insufficiency.
The majority of deaths associated with fetal-growth problems are potentially avoidable through better assessment and surveillance.
This meta-analysis does not conclude that customized norms can better define a population of pathologically small fetuses that are at higher risk of adverse perinatal outcomes, although a recent prospective observational study suggested such a conclusion.

16. Customized vs population-based growth charts to identify neonates at risk of adverse outcome
G Chiossi et al. UOG 2017
Limitations
No randomized controlled studies addressed the research question and therefor only observational studies are included.
Studies differed in their definitions of various outcomes:
IUFD – after 20 weeks’ gestation / after 22 weeks’ gestation
NICU admission – stay of >48h / stay of >7 days.
Patient-level data not available - impossible to determine if the ORs for each outcome were significantly different when fetuses were classified by customized vs population-based norms.

17. Customized vs population-based growth charts to identify neonates at risk of adverse outcome
G Chiossi et al. UOG 2017
Conclusions
Both customized and population-based growth charts can detect intrauterine growth disturbances that are associated with adverse perinatal outcomes, particularly for SGA neonates.
Although point estimates of pooled ORs may differ for some of these outcomes, their overlapping CIs prevents any firm conclusions about the superiority of one method or the other.

18. Customized vs population-based growth charts to identify neonates at risk of adverse outcome
G Chiossi et al. UOG 2017
Points for discussion
Should customized growth charts be used assessing SGA, LGA or for all cases?
Why does customization not improve detection of adverse outcome?
Is birth weight customization for a factor related to adverse outcome (such as obesity and racial origin) biologically justified?
Are prescriptive (prospective) population-based charts such as those from INTERGROWTH-21st any more likely to predict adverse outcome?
Is lack of nutrition (poor growth) or lack of oxygen (lack of respiration) more likely to result in fetal demise and adverse neonatal outcome?
Is deviation of fetal growth from the ideal/optimum always an accurate proxy for failure of placental oxygen transfer?