Assessing skeletal mineralization in preterm infants: Insights from dual energy X-ray absorptiometry and their limitations

Metabolic bone disease of prematurity remains a significant challenge in neonatal medicine. Infants born prematurely are deprived of the rapid mineral accretion that occurs during the third trimester, when approximately 80% of fetal calcium and phosphorus deposition takes place. Consequently, preterm infants frequently exhibit reduced bone mineral content (BMC), impaired skeletal mineralization, and an increased risk of fractures.^[1] In this context, the study by Singh et al. provides an important insight into the clinical and nutritional determinants of early skeletal mineralization, while highlighting both the utility and limitations of dual-energy X-ray absorptiometry (DXA) in neonatal research.^[2]

In their prospective observational study, Singh et al. evaluated 44 preterm neonates born at ≤33 weeks’ gestation and measured total body less head BMC at 6 weeks of postnatal age using a GE Lunar DPX DXA system.^[2] BMC correlated strongly with indices of growth and maturity, including gestational age, birth weight, and birth length, the latter emerging as the strongest independent predictor. In addition, markers of illness severity and nutritional exposure were independently associated with skeletal outcomes. Prolonged total parenteral nutrition, extended mechanical ventilation, and delayed fortification of human milk were associated with lower BMC, whereas earlier fortification was associated with improved mineralization. These findings underscore both the biological dependence of skeletal mineral accretion on growth and the modifiable impact of neonatal nutritional practices.

These observations align with the current understanding that bone mineral accretion in preterm infants is determined by gestational maturity, mineral supply, systemic illness, medication exposure, and reduced mechanical loading during neonatal intensive care.^[1,3,4] Contemporary reviews emphasize the interplay between prenatal mineral accretion and postnatal influences, including nutrition, immobilization, and clinical interventions.^[1] From a clinical perspective, the study reinforces the central importance of early nutrition and growth in shaping skeletal outcomes in this vulnerable population.

Interpretation of these findings, however, requires careful consideration of the methodological constraints of DXA in very small infants. DXA is widely used because it provides rapid measurements with low radiation exposure and acceptable reproducibility.^[5,6] It also enables simultaneous assessment of body composition, facilitating exploration of relationships between lean mass, fat mass, and skeletal mineralization.

A key limitation of DXA is its two-dimensional nature. Bone mineral content is expressed relative to projected bone area to derive areal bone mineral density, which does not account for bone depth.^[5] Consequently, measurements are inherently influenced by skeletal size, and smaller bones may yield lower values despite similar intrinsic mineralization. The strong association between BMC and birth length reported by Singh et al.^[2] therefore warrants cautious interpretation, as part of this relationship likely reflects size dependence rather than differences in mineralization alone. This effect may also contribute to differences observed between small-for-gestational-age and appropriately grown infants. Although size-adjusted indices such as bone mineral apparent density have been proposed, these rely on simplifying geometric assumptions that may not fully capture the complexity of skeletal growth in early life.^[5]

Important insights into skeletal development in preterm infants have also been obtained using complementary techniques. Mercy et al. demonstrated, using quantitative ultrasound, that tibial speed of sound declines postnatally while linear growth continues in very low birthweight infants, and that nutrient intake influences lower-limb growth.^[4] These findings provide additional perspective on cortical bone status and longitudinal growth and highlight the value of integrating alternative approaches alongside DXA.

Further methodological considerations relate to the DXA technology used. Singh et al. employed a GE Lunar DPX densitometer, a pencil-beam system originally developed for adult applications.^[2] Although this design minimizes magnification artifact, it is associated with longer acquisition times and lower spatial resolution compared with newer fan-beam DXA scanners. In unsedated infants, prolonged scan duration increases susceptibility to motion artifact, which may influence projected bone area and calculated BMC. These technical factors should be considered when interpreting measurements in neonatal studies.

Interpretation is further constrained by the availability of appropriate analysis software and reference standards. Singh et al. do not specify the software used for analysis.^[2] The GE Lunar DPX platform was developed primarily for adult densitometry, and infant-specific analysis tools and neonatal reference data for this older system are not well characterized in the literature.

Despite these limitations, DXA has substantially advanced the understanding of skeletal development in preterm infants. Early studies established the feasibility of measuring bone mineralization in infancy, while subsequent longitudinal cohorts have provided insight into longer-term skeletal outcomes. In particular, follow-up of infants enrolled in neonatal nutrition trials suggests that early nutritional exposures may influence bone mass into adolescence and early adulthood.^[7]

The study by Singh et al. emphasizes that skeletal health in preterm infants is not solely a consequence of prematurity but is shaped by clinical care, particularly nutritional management.^[2] At the same time, DXA-derived measurements in this population require careful interpretation in light of their inherent methodological limitations.