Which chart and which cut-point: deciding on the INTERGROWTH, World Health Organization, or Hadlock fetal growth chart

• Published in: BMC pregnancy and childbirth Vol. 22; no. 1; p. 25
• Main Authors: Liauw, Jessica, Mayer, Chantal, Albert, Arianne, Fernandez, Ariadna, Hutcheon, Jennifer A
• Format: Journal Article
• Language: English
• Published: England BioMed Central 10-01-2022; BMC
• Subjects: Adult; Biometrics; Birth weight; British Columbia - epidemiology; Cohort Studies; Confidence intervals; Endpoint Determination; Ethnicity; Female; Fetal Development - physiology; Fetal growth; Fetal weight; Fetal Weight - physiology; Fetuses; Gestational Age; Growth Charts; Hospitals; Humans; Incidence; Infant, Newborn; Morbidity; Mortality; Multiple births; Perinatal Mortality; Population; Predictive Value of Tests; Pregnancy; Risk; Sensitivity and Specificity; Standard deviation; Stillbirth; Ultrasonic imaging; Ultrasonography, Prenatal; Womens health; British Columbia; British Columbia Canada; Canada; Fetal growth; Growth charts; Fetal weight
• ISSN: 1471-2393; 1471-2393
• DOI: 10.1186/s12884-021-04324-0

To determine how various centile cut points on the INTERGROWTH-21st (INTERGROWTH), World Health Organization (WHO), and Hadlock fetal growth charts predict perinatal morbidity/mortality, and how this relates to choosing a fetal growth chart for clinical use.
We linked antenatal ultrasound measurements for fetuses > 28 weeks' gestation from the British Columbia Women's hospital ultrasound unit with the provincial perinatal database. We estimated the risk of perinatal morbidity/mortality (decreased cord pH, neonatal seizures, hypoglycemia, and perinatal death) associated with select centiles on each fetal growth chart (the 3rd, 10th, the centile identifying 10% of the population, and the optimal cut-point by Youden's Index), and determined how well each centile predicted perinatal morbidity/mortality.
Among 10,366 pregnancies, the 10th centile cut-point had a sensitivity of 11% (95% CI 8, 14), 13% (95% CI 10, 16), and 12% (95% CI 10, 16), to detect fetuses with perinatal morbidity/mortality on the INTERGROWTH, WHO, and Hadlock charts, respectively. All charts performed similarly in predicting perinatal morbidity/mortality (area under the curve [AUC] =0.54 for all three charts). The statistically optimal cut-points were the 39th, 31st, and 32nd centiles on the INTERGROWTH, WHO, and Hadlock charts respectively.
The INTERGROWTH, WHO, and Hadlock fetal growth charts performed similarly in predicting perinatal morbidity/mortality, even when evaluating multiple cut points. Deciding which cut-point and chart to use may be guided by other considerations such as impact on workflow and how the chart was derived.