General Obstetrics and Gynecology: Obstetrics
Customized versus population-based birth weight standards for identifying growth restricted infants: A French multicenter study

https://doi.org/10.1016/j.ajog.2005.10.816Get rights and content

Objective

This study was undertaken to describe the characteristics of pregnancies according to a customized definition of fetal growth restriction and to determine the association between customized standards and adverse pregnancy outcomes.

Study design

Two definitions of growth restriction, a population and a customized standard, were applied to 56,606 births in 5 tertiary maternity hospitals in France from 1997 to 2002. The customized definition was adjusted for maternal height and weight, parity, fetal gender, and gestational age. Odds ratios and 95% CIs for neonatal morbidity and mortality were calculated to compare small for gestational age and non-small for gestational age births.

Results

By using customized standards, 2.7% of births were reclassified as small for gestational age. These births were to taller, heavier, multiparous women. Compared with non-small for gestational age births, these newly detected small-for-gestational-age newborn infants showed an increased risk of stillbirth (odds ratio = 4.52, 95% CI 2.47-8.14) and perinatal death (odds ratio = 2.60, 95% CI 1.62-4.15). These infants were also more likely to be born to women with hypertensive disease in pregnancy (7.0%) versus those reclassified as non-small for gestational age (2.3%) and those non-small for gestational age by both standards (5.5%).

Conclusion

These findings highlight the interest of using customized birth weight standard adjusted for maternal and neonatal characteristics to identify fetuses at risk, particularly among apparently normal fetuses. Individual growth norms should be used to define small for gestational age.

Section snippets

Description of sample

Five maternity hospitals participated; 4 were regional teaching hospitals with an on-site neonatal care unit. The last hospital is a large level II center. Two are located in northern France, in Lille and Roubaix, 2 others in Paris, and the last in Nancy in the east of France. We used routinely prospectively recorded data, including demographic data and information on complications during pregnancy, delivery, and neonatal period. We included all singleton pregnancies with delivery occurring at

Results

The final sample included 56,606 births that occurred between 1997 and 2002 in the 5 participating hospitals. Table I displays maternal and neonatal characteristics in each maternity hospital. The maternity hospitals in the north and the east of France are located in areas with a more socially disadvantaged population than in the Parisian region. The prevalence of young primiparous or grand multiparous women with less schooling is higher. Average maternal weight was higher in Lille and Roubaix

Comment

This study used 2 methods for identifying SGA newborn infants: a population-based standard, adjusted only for gestational age and fetal gender, and a customized standard, based on the prediction of optimal birth weight at term given maternal characteristics. The latter identified a group of infants at risk for adverse perinatal outcome that would otherwise be considered appropriate for gestational age using population standards. This group represented almost 3% of births in our sample.

Acknowledgments

We thank the obstetric team of the Roubaix maternity hospital, and are especially grateful for the assistance provided by Mrs S. Carlier.

References (27)

  • P. Owen et al.

    Fetal growth velocity in the prediction of intrauterine growth retardation in a low population

    BJOG

    (1998)
  • I.K. Rossavik et al.

    Mathematical modeling of fetal growth: I, basic principles

    L Clin Ultrasound

    (1984)
  • R.L. Deter et al.

    Mathematic modeling of fetal growth: development of individual growth curve standards

    Obstet Gynecol

    (1986)
  • Cited by (175)

    View all citing articles on Scopus

    Reprints not available from the authors.

    View full text