# Customized Fetal Body Mass Index as a Better Predictive Marker for Neonatal Nutritional Status

## TL;DR

A new fetal body mass index model is shown to better predict neonatal nutritional status than existing methods.

## Contribution

A customized fetal BMI model is introduced with higher diagnostic accuracy for neonatal nutrition.

## Key findings

- The new fetal BMI model achieved an AUC of 0.95 for predicting malnutrition.
- Existing methods (GROW and IG21st) had AUCs of 0.79 and 0.80, respectively.
- The model showed high sensitivity (0.83) and specificity (0.90) for malnutrition detection.

## Abstract

Background/Objectives: The diagnosis of fetal nutritional status is of great importance for the accurate evaluation and monitoring of these pregnancies. The objective of the present study is to develop a model that allows for the prenatal assessment of fetal body mass index and to evaluate its diagnostic efficacy in predicting neonatal nutritional status. Methods: A retrospective cohort study was conducted to develop and evaluate a new model in the diagnosis of alterations in fetal nutritional status based on the customized fetal body mass index. By establishing the relationship between weight and length, we can calculate the fetal body mass index, which could correlate more effectively with nutritional status. Results: A total of 12,633 subjects were recruited, and 9499 were included in our study. Capacities to predict both neonatal malnourishment and overnutrition were calculated for each of the three methods analyzed (BMI, GROW, and IG21st). The receiver operating characteristic curve for each method was developed. The sensitivity and specificity for the assessment of malnutrition were 0.83 and 0.90, respectively. The area under the ROC curve of our method was 0.95 for malnutrition, while for IG21st and GROW, it was 0.80 and 0.79, respectively. Conclusions: This study demonstrates a superior diagnostic capacity for alterations in fetal and neonatal nutritional status of this new fetal BMI curve compared to the previously used fetal weight percentile curves.