Comparative performances of five neonatal anthropometric charts in classifying small- and large-for-gestational age Nigerian term neonates and predicting insulin resistance: An exploratory cross-sectional analysis
Year: 2026
Type: preprint
Abstract: Background: Abnormal intrauterine foetal growth, manifesting as small-for-gestational-age (SGA) or large-for-gestational-age (LGA) birth, increases the risk of insulin resistance (IR), a precursor of later adverse cardiometabolic outcomes. Neonatal anthropometric charts are used for identifying SGA or LGA babies, but their identification of these and association with neonatal outcomes vary across populations but sparingly reported from sub-Saharan Africa with largest burden of SGA worldwide. Notably, their association with neonatal IR is not known. We thus compared the performances of five charts—INTERGROWTH-21ST, Fenton-2013, Mokuolu, Lubchenco and WHO Multicentre Growth Reference Study (MGRS)—in identifying SGA/LGA and predicting neonatal IR associated with SGA/LGA. Methods: This exploratory secondary cross-sectional analysis included 121 term neonates from a prospective maternal-infant cohort at Lagos State University Teaching Hospital, Ikeja, Lagos, Nigeria. Birthweights were classified with each chart. Neonatal IR was defined as Homeostatic Model Assessment of Insulin Resistance >1.195. We calculated frequencies of SGA/LGA, diagnostic accuracy (sensitivity, specificity) of detecting IR associated with SGA or LGA and inter-chart agreement (Cohen’s κ). Results: One-quarter of neonates (24.8%) had IR. The frequency of SGA and LGA varied widely across charts (SGA: 4.2–10.1%; LGA: 4.1–14.2%). INTERGROWTH-21ST and Fenton-2013 identified more SGA (10.1% and 10.0%, respectively), while INTERGROWTH-21ST and Mokuolu identified more LGA (12.6% and 12.5%, respectively). Across charts, sensitivities for predicting IR were low (13.3–40.0%), while specificities were higher (73.3–76.6%). Lubchenco and WHO charts showed relatively better balance of sensitivity and specificity, albeit with very wide confidence intervals. Concordance was highest between INTERGROWTH-21ST and Fenton-2013 for SGA (κ=1.0), and between Mokuolu, INTERGROWTH-21ST, and Lubchenco for LGA (κ=0.93). Conclusion: This largely exploratory findings suggests that neonatal anthropometric charts vary widely in their classification of neonatal birthweight. While none of the charts reliably predicted neonatal IR, Lubchenco and WHO-MGRS charts performed relatively better but unreliably due to wide confidence intervals. There is need for larger, multicentre studies in African populations. Clinical Trial Number: Not applicable
Source: Research Square
Language: English
Field: Medicine
Domain: Health Sciences
Open Access status: green
OpenAlex ID: W7225389651