Differential classification of infants in United States neonatal intensive care units for weight, length, and head circumference by United States and international growth curves

• Published in: Annals of human biology 2020-08, Vol.47 (6), p.564-571
• Main Authors: Ferguson, A. Nicole, Olsen, Irene E., Clark, Reese H., Yockey, Bryan D., Boardman, Jonathan, Biron, Kyle, Jannuzzo, Cooper, Waskiewicz, Daniel, Mendoza, Amanda, Lawson, M. Louise
• Format: Article
• Language: English
• Subjects: large for gestational age; Preterm growth charts; preterm growth curves; small for gestational age
• ISSN: 0301-4460; 1464-5033
• DOI: 10.1080/03014460.2020.1817555

Clinicians and researchers use a variety of intrauterine growth curves to classify NICU infants as small (SGA), appropriate (AGA), or large for gestational age (LGA). Since curve creation methods and samples vary, SGA/AGA/LGA cut-offs and resulting subgroups of infants vary among curves and impact outcome study findings - limiting generalisability. Determine how two international and two US-specific curves classified US NICU infants. Classified 192,888 infants from US NICUs (2013-2016) as SGA or LGA for birthweight, length, and head circumference, using the international Fenton and INTERGROWTH-21st curves and US-specific Olsen and Lubchenco (historical) curves. Modern curves classified approximately 10% of infants as SGA up to 32 weeks, but older infants had increased variability. The INTERGROWTH-21st curves consistently had rates above 10% for LGA after 32 weeks. While Olsen and Fenton both fit, the Olsen curves had overall best-fit for our sample of US NICU infants. The INTERGROWTH-21st curves fit the definitions for SGA and LGA for younger ages, but inferences outside of these definitions are unwarranted due to limited sample size. The INTERGROWTH-21st sample used for 33 weeks and older infants was physically smaller at the upper percentiles than our sample of US infants.