Determinants of bone mineral content in preterm neonates

INTRODUCTION

Compromised bone health in preterm neonates has significant short- and long-term implications. Metabolic bone disease (MBD) of prematurity is associated with low bone mass, rickets, and fractures in the neonatal period and osteoporosis later in life.^[1,2] The prevalence of MBD increases with decreasing gestational age and is estimated to be around 40% in preterm neonates.^[3] This is related to the fact around 80% of fetal calcium accretion occurs in the third trimester.^[4]

Understanding the determinants of bone health in preterm neonates is expected to help devise strategies to improve it. Maternal factors such as smoking, pre-eclampsia, gestational weight gain, placental size, and chorioamnionitis have been linked to bone mass of their offsprings.^[5,6] Postnatal factors such as nutrition, level of physical activity, use of medications (steroids, diuretics, caffeine), and comorbidities like sepsis and necrotizing enterocolitis (NEC) also influence bone health. Delayed breastmilk fortification and prolonged use of total parenteral nutrition (TPN) are associated with higher prevalence of rickets and elevated serum alkaline phosphatase (ALP) due to the insufficient calcium and phosphate content in non-fortified breastmilk.^[7,8]

Given the paucity of Indian data, we conducted this study to evaluate the bone mineral content (BMC) of Indian preterm neonates and identify clinical and biochemical predictors of impaired bone health.

MATERIAL AND METHODS

This prospective observational study was conducted at our hospital from November 2020 to July 2022. All preterm neonates (period of gestation ≤33 weeks) admitted to the neonatal intensive care unit (NICU) during the study period were screened for eligibility. Infants who were stable and able to undergo a dual-energy x-ray absorptiometry (DXA) scan at 6 weeks of postnatal age were included. Ethical clearance was obtained from the Institutional Ethics Committee (IEC number RHL-IEC-16052, November 19, 2020). Written informed consent was obtained from the parents or legal guardians at the time of NICU admission.

Baseline data (mode of conception and delivery, weeks of gestation, birth weight and length, gender, duration of ventilatory support, duration of steroid and diuretic use, duration of TPN, and day of initiation of fortified feeds) were recorded. All neonates underwent anthropometric evaluation (weight, length, and head circumference), biochemical analysis (calcium, phosphorus, ALP, and 25-hydroxyvitamin D), and DXA scanning at 6 weeks of postnatal age in view of clinical stability, feasibility, and in line with international DXA literature. Delayed fortification was defined as initiation after day 7 of life. Weight was measured using a calibrated electronic weighing scale (accuracy ±10 g), length was measured to the nearest 0.1 cm using an infantometer, and head circumference was recorded with a non-stretchable measuring tape. The DXA scans were performed by a single technician using a Lunar GE DPX pencil beam machine (GE Healthcare), with all scans conducted in a uniform, nonsedated, swaddled state. Only scans without motion artifacts were included for analysis. Total body less head (TBLH) BMC was recorded.

Poor weight gain was defined as gain below 15 g/kg/day. MBD was diagnosed in the presence of radiological features or serum ALP above 400 IU/L. Neonates who remained critically ill or required life support at 6 weeks or whose parents did not consent were excluded.

Data were entered in Microsoft Excel and analyzed using IBM SPSS Statistics for Windows, Version 23.0 (IBM Corp., Armonk, NY, USA). Continuous variables are presented as mean ± standard deviation (SD), while categorical variables are expressed as frequencies and percentages. Independent t-tests or Mann–Whitney U tests were used for group comparisons as appropriate. Pearson correlation analysis was performed to assess associations with BMC. Univariate and multivariate linear regression models were used to identify independent predictors of BMC. Only variables with significant univariate associations were included.

RESULTS

Of the 51 preterm neonates approached for the study, four were unfit for DXA, two declined consent, and one scan was excluded due to poor image quality. The final cohort comprised 44 neonates (25 boys), with a mean gestational age of 29.8 ± 1.7 weeks, birth weight of 1158.2 ± 293.2 g, and birth length of 37.8 ± 2.6 cm. Among these, 30 (68.2%) were appropriate for gestational age, 13 (29.5%) small for gestational age (SGA), and one (2.3%) large for gestational age.

Thirty-eight neonates (86.4%) had respiratory distress syndrome while NEC and culture-positive sepsis were observed in 9 each (20.5%), respectively. A total of 14 infants (31.8%) required invasive mechanical ventilation, with a mean duration of 6.9 ± 6.5 days. TPN was administered to 41 neonates (93.2%) for an average of 7.0 ± 6.9 days. Delay in breastmilk fortification beyond the 1^st week was observed in 16 neonates (36.4%).

At the 6-week follow-up, mean serum calcium, phosphorus, ALP, and 25-hydroxyvitamin D levels were 8.4 ± 0.6 mg/dL, 5.7 ± 1.4 mg/dL, 434 ± 132 U/L, and 45.9 ± 8.7 ng/mL, respectively. The mean TBLH BMC was 15.0 ± 6.4 g. None of the subjects had clinical evidence of fractures.

BMC correlated positively with gestational age (r = 0.65, P < 0.001), birth weight (r = 0.77, P < 0.001), and birth length (r = 0.82, P < 0.01, [Table 1 and Figure 1]). Biochemical markers such as serum calcium (r = 0.37, P = 0.01) and phosphorus (r = 0.64, P < 0.001) were also positively correlated with BMC, whereas ALP levels were negatively correlated (r = –0.65, P < 0.001). Duration of TPN (r = –0.72, P ≤ 0.001) and mechanical ventilation (r = –0.50, P = 0.002) were inversely correlated with BMC. Data on antenatal factors were collected but not included in the model due to lack of significance on univariate model. Steroid and diuretic exposure showed no significant associations.

Close

Figure 1:

Correlation matrix for bone mineral content (all P-values are significant).

Export to PPT

Anthropometric measurements at 6 weeks showed a mean length of 42.7 ± 2.9 cm, weight of 1671 ± 458 g, and head circumference of 30.9 ± 2.1 cm. BMC was positively correlated with both weight and length at 6 weeks (r = 0.83 for both, P < 0.001) and with weight gain from birth (r = 0.46, P = 0.002). Poor weight gain was noted in 12 (27.3%) neonates.

Neonates with NEC had lower BMC values (8.2 ± 5.7 as against 16.5 ± 5.3 g, P = 0.002), as did those with culture-positive sepsis (9.1 ± 5.2 as against 16.3 ± 5.7 g, P = 0.003), MBD (9.5 ± 5.9 as against 17.3 ± 5.3 g, P < 0.001), need for mechanical ventilation (10.0 ± 5.0 as against 17.4 ± 5.7 g, P < 0.001), and poor weight gain (11.3 ± 6.7 as against 16.1 ± 5.7 g, P = 0.04 [Table 2 and Figure 2]). None of the neonates had radiological or clinical evidence of fractures. No gender difference in BMC was observed.

Multivariate linear regression identified birth length (β = 0.52, P = 0.01), duration of TPN (β = –0.70, P = 0.001), mechanical ventilation (β = –0.57, P = 0.002), and delay in breastmilk fortification (β = –0.48, P = 0.002) as significant predictors of TBLH BMC [Table 3]. The model explained 77% of the variance in BMC (R² = 0.768). The final model maintained more than 10 events per variable and no multicollinearity (variance inflation factor <2). A 1-week delay in breastmilk fortification or prolongation of TPN was estimated to reduce BMC by 3.7 g (24.7%) and 3.9 g (35.7%), respectively, while an additional week of ventilation reduced BMC by 2.0 g (13.3%). SGA infants had lower BMC on univariate analysis, but not in the multivariate model once length and gestation were included.

Close

Figure 2:

Effects of risk factors on bone mineral content. t-test; P < 0.05 significant.

Export to PPT

DISCUSSION

Our study highlights the impact of perinatal variables on early bone accretion in preterm neonates. Gestational age, size at birth, and neonatal complications emerged as key determinants of BMC, aligning with earlier studies on the influence of fetal growth on postnatal skeletal development.^[8,9] This study quantifies clinical determinants of BMC, evaluates impact of modifiable NICU practices, and identifies birth length as a key predictor.

Prolonged exposure to TPN and mechanical ventilation was associated with significantly reduced BMC, consistent with observations on iatrogenic contributors to osteopenia of prematurity.^[8] We were able to quantify the estimate of this impact with 25.7% and 13.3% decline with each additional week of TPN and ventilation, respectively. These observations underscore the need for judicious and time-bound use of these modalities.

The impact of TPN duration and delayed initiation of fortified feed on BMC in multivariate models reinforces the role of early enteral nutrition in promoting bone health. The association of phosphorus, but not vitamin D or ALP, confirms the key role of phosphate in neonatal bone mineralization.^[9] The lack of association with vitamin D and ALP may be due to the interplay of growth, mineral accretion, and hormonal regulators, and the limitations of using single time-point measurements. Moreover, single-time vitamin D levels may not reflect cumulative exposure. This is consistent with the observation of lack of impact of daily 800 IU of vitamin D supplementation on bone mineralization.^[10]

Birth length emerged as a stronger predictor of BMC than weight or head circumference, suggesting the major impact of linear growth on early bone mineral accretion. The association between poor postnatal weight gain and reduced BMC supports the importance of optimizing early growth to prevent osteopenia. Lower BMC in neonates with NEC, sepsis, and MBD highlights the inhibitory effect of systemic inflammation and catabolic stress on bone accretion.^[11] This study quantifies clinical determinants of BMC, evaluates impact of modifiable NICU practices, and identifies birth length as a key predictor. It supports the need for earlier fortification and minimization of TPN duration.

This study is limited by its single-center design and cross-sectional assessment of BMC at 6 weeks. Sample size was relatively small but adequate as the post-hoc power exceeded 95% for the strongest correlation (birth length vs. BMC, r = 0.82). The exclusion of sickest infants unable to undergo DXA could have resulted in bias. Longitudinal and multicentric studies are needed to evaluate the long-term skeletal outcomes of preterm infants and validate these findings across diverse settings with different nutrition and care practices. Serial assessments of mineral metabolism may further elucidate the dynamic nature of neonatal bone health.

CONCLUSION

Prolonged TPN, ventilation, and delayed fortification are modifiable determinants of BMC in preterm neonates. Early enteral optimization and timely fortification may improve skeletal outcomes.