Body composition assessment of premature infants in the neonatal intensive care unit

Thumbnail Image

Persistent link to this item

View Statistics

Journal Title

Journal ISSN

Volume Title


Body composition assessment of premature infants in the neonatal intensive care unit

Published Date




Thesis or Dissertation


Premature infants experience growth alterations that place them at risk for adverse metabolic and neurodevelopmental outcomes. Monitoring the quality of weight gain through body composition assessment in the neonatal intensive care unit (NICU) may help clinicians gauge the response to nutritional provision and guide future interventions that promote adequate growth and neurodevelopment while reducing the risk for obesity and metabolic disease. While length and weight are regularly tracked during an infant’s NICU stay, these measurements do not adequately represent total body adiposity shortly after birth. Thus, a method of body composition which is non-invasive, portable, and able to be frequently utilized in both critically ill and medically stable infants is desirable. Unfortunately, many current methods of body composition are invasive, expensive, involve ionizing radiation, or are unsuitable for repeated measurement in a medically fragile infant. Thus, this dissertation project explores methods to monitor body composition in premature infants in the NICU setting with a focus on ultrasound. The first study explored the ability of weight for length indices of the body to serve as proxies for adiposity in preterm infants. Indices examined include weight for length (W/L), body mass index (BMI), and ponderal index (PI). Each index was examined for its ability to predict fat mass (FM), fat-free mass (FFM), and percent body fat (%BF). None of the indices adequately reflected adiposity in preterm infants, indicating that assessing body composition in preterm infants requires more than weight and length measurements, and other methods of bedside assessment should be pursued. The second study examined the ability of ultrasound to assess body composition in premature infants in the NICU setting. Ultrasound images of the biceps, abdomen, and quadriceps were obtained for assessment of adipose and muscle thickness and were compared with body composition measurements (FFM, FM, %BF) taken using air displacement plethysmography (ADP). While ultrasound measurements of biceps and quadriceps muscle thickness correlated with total FFM, ultrasound measurements were not included in final models for predicting FFM. Biceps, abdomen, and quadriceps adipose thickness correlated with total FM and %BF, but only biceps adipose was selected in the final model predicting %BF. The sum of ultrasound adipose thickness measures was selected for the final model predicting %BF. However, all models had low predictive ability due to low proportion of variance explained (R2) and/or high prediction error (root mean square error, RMSE). While the study conducted here does not support the use of ultrasound measurements of adipose and muscle thickness of the biceps, abdomen, or quadriceps alone to predict body composition in preterm infants, exploration of additional sites or cross-sectional area may improve predictive ability. Additionally, ultrasound measurements may have some value as a prognostic tool for other clinical outcomes, such as neurodevelopment or readiness for NICU discharge. Regardless, this work highlights the need for clinical body composition methods appropriate for premature infants to help monitor for disease risk and assist in the refinement of current nutrition practices in the NICU.


University of Minnesota Ph.D. dissertation. August 2019. Major: Nutrition. Advisor: Carrie Earthman. 1 computer file (PDF); ix, 79 pages.

Related to




Series/Report Number

Funding information

Isbn identifier

Doi identifier

Previously Published Citation

Suggested citation

Nagel, Emily. (2019). Body composition assessment of premature infants in the neonatal intensive care unit. Retrieved from the University Digital Conservancy,

Content distributed via the University Digital Conservancy may be subject to additional license and use restrictions applied by the depositor. By using these files, users agree to the Terms of Use. Materials in the UDC may contain content that is disturbing and/or harmful. For more information, please see our statement on harmful content in digital repositories.