Profile of metabolic bone disease in extremely low birth weight (ELBW) and very low birth weight (VLBW) neonates

Abstract

Background

Metabolic bone disease (MBD) is an important cause of morbidity in premature, very low birth weight (VLBW), and sick infants and, if left undiagnosed, may lead to structural deformities and spontaneous fractures. The objective of the present study was to study the profile of MBD and to determine the incidence of MBD in infants ≤ 32 weeks/≤ 1250 g at birth.

Method

A total of 57 infants ≤ 32 weeks/≤ 1250 g at birth admitted in our NICU from October 2020 to July 2021 were included in the study. These infants underwent screening for MBD at 4 weeks of age. They were stratified into three groups based on their gestation (≤ 28 weeks, 29–30 weeks, 31–32 weeks).

Results

MBD was observed in 100% of extreme preterm babies and 69% of very preterm babies. Overall, the incidence of MBD was 73%. Serum phosphorus level normalized by 42–44 weeks post menstrual age (PMA) across all gestations. Alkaline phosphatase (ALP) levels normalized by 42–44 weeks only in very preterm babies. Seventeen babies ≤ 30 weeks required inorganic phosphorus supplementation in addition to calcium phosphate supplementation in order to correct the MBD. Drugs like caffeine, steroids, and furosemide have significant impact on the development of MBD. The time to reach full feeds with fortification had no statistically significant effect on the incidence of MBD as detected by serum phosphorus level and serum ALP level.

Conclusion

The profile outlined in the present study matches the literature reports in many aspects, revealing the importance of characterizing this group for the prognosis and short- and long-term follow-up of newborns with bone metabolic disease.

Background

Metabolic bone disease (MBD) is defined as impaired bone mineralization in a neonate with lower-than-expected bone mineral levels in either a fetus or a neonate of comparable gestational age and/or weight, coupled with biochemical abnormalities with or without accompanying radiological manifestations. MBD has been reported to occur in 16% to 40% of extremely low birth weight neonates and presents by 6–16 weeks after birth [1]. Patients tend to be asymptomatic, and the diagnosis of the diseases is based on laboratory and radiographic findings of spontaneous, non-traumatic fractures, especially in the legs, arms, and ribs [2, 3].

However, MBD includes complications such as rickets and osteopenia; rickets is defined on the basis of radiographic evidence in the growing ends of long bones, while osteopenia implies low bone mineralization [4]. In addition, the prevalence of rickets in LBW infants and neonates with normal weight was reported to be 13.4% and 4.9%, respectively, between 12 and 14 weeks of corrected age [5].

The major causes of MBD among preterm infants are deficiency of calcium, phosphorus, and magnesium [6]. During pregnancy, the total amount of calcium increases from 5 g at the end of the second trimester to 30–35 g at term, and 80% of calcium and phosphorus transmission from the mother to fetus occurs during the third trimester of pregnancy; preterm birth disrupts this transfer [7]. On the other hand, breast milk and usual formula are not able to provide the infant with the essential calcium and phosphorus independently. Therefore, early provision of highly bioavailable minerals and the correction of vitamin D deficiency and phosphorus concentrations are recommended as helpful methods of MBD prevention [3]. Among other risk factors of MBD are delayed breastfeeding, soy formula, obstructive jaundice, prolonged use of total parenteral nutrition (TPN), and medications such as diuretics, methylxanthines, and corticosteroids [2].

In previous research, out of 32 ELBW infants, 18 cases showed radiologic signs of rickets, and 14 had osteopenia without rickets. Therefore, early screening for MBD is recommended in preterm infants, specialty those with birth weight of < 600 g or alkaline phosphatase (ALP) of > 800 IU/L [8]. Evidence suggests that ALP concentrations of > 750 IU/L may cause osteopenia of prematurity. As a result, weekly measurement of biochemical bone profile (i.e., Ca, P, and ALP) has been recommended for early detection of MBD in high-risk neonates [9]. Very few studies are available from India showing the exact prevalence of MBD. Normalization of biochemical parameters is expected by term-corrected age, as found in many studies done in well-developed countries. Whether this holds true in Indian babies is not known. This study aims to determine the incidence of MBD in extremely low and very low birth weight (VLBW) neonates. Secondarily, the time required for normalization of biochemical parameters in babies who were diagnosed with MBD during neonatal period was also be studied.

Methods

After obtaining written informed consent from the parent/s, this prospective observational cross-sectional study was conducted on 57 infants ≤ 32 weeks/≤ 1250 g at birth admitted in our NICU from October 2020 to July 2021. These neonates underwent screening for MBD at 4 weeks of age. Neonates with cholestatic disorder, skeletal anomalies, and genetic syndrome were excluded from the study.

As per existing unit policy, all neonates ≤ 32 weeks/≤ 1250 g at birth were started on feeds once hemodynamically stable with minimal enteral nutrition (MEN) of 10–20 ml/kg/day, increased at a rate of 15–30 ml/kg/day as per tolerance. The feed was fortified with human milk fortifier (HMF) when the baby tolerated feed of 100 ml/kg/day, and feeds were subsequently increased to 150–200 ml/kg/day, to provide a total calorie intake of 120–140 kCal/kg/day. One sachet of HMF of 1 g (Lactodex-HMF) provides 3.37 kCal energy, 0.27 g protein, 15.8 mg calcium, 7.9 mg phosphorus, and 132 IU vitamin D. They underwent laboratory evaluation with serum calcium, phosphorus, alkaline phosphatase (ALP), and urine analysis for tubular reabsorption of phosphorus (TRP) at 4 weeks of age as a screening test for detection of MBD. Serum calcium, phosphorus, ALP, and TRP levels were done as in-house tests in our NABL-accredited biochemistry laboratory. A urine calcium/creatinine ratio and calcium/phosphorus ratio were done fortnightly to calculate the TRP(percentage).

$$\textrm{TRP}=1\hbox{-} \left[\textrm{Up}/\textrm{Sp}\times \textrm{Scr}/\textrm{Ucr}\right]\times 100\ \left(\textrm{Up}-\textrm{urinary}\ \textrm{phosphorus},\textrm{Sp}-\textrm{serum}\ \textrm{phosphorus},\textrm{Scr}-\textrm{serum}\ \textrm{creatinine},\textrm{Ucr}-\textrm{urinary}\ \textrm{creatinine}\right)$$

MBD was diagnosed when serum phosphorus was < 5.5 mg/dl and/or ALP > 900 IU/L [10]. TRP > 95% indicates increased reabsorption of phosphorus and therefore phosphorus deficiency. After the initial measurement, subsequent measurements were done fortnightly till term-corrected age (40 weeks) by which time the risk of MBD becomes reasonably low. If serum phosphorus was < 5.5 mg/dl and ALP > 900 IU/L (as per existing MBD diagnostic protocol), the babies received calcium phosphate supplements and vitamin D along with fortified human milk. If TRP remained > 95% despite the above, only phosphorus supplementation was increased by adding sodium phosphate. Extra calcium was supplemented only if there is documented hypocalcemia (serum calcium < 7 mg/dl). Supplementation range used for calcium was 150–220 mg/kg/day; for phosphorus, it was 50–90 mg/kg/day and vitamin D 800–1600 IU/day (ESPHGAN recommendation). Subsequent investigations with the above biochemical tests were done every 2 weeks, for monitoring till normalization of the biochemical parameters.

Kidney, ureter, and urinary bladder ultrasonography (USG KUB) was done prior to discharge to look for nephrocalcinosis due to calcium supplementation. The growth of the baby during stay in the unit was monitored using Ehrenkranz and Intergrowth-21 growth chart for weight, length, and head circumference. Those babies who had abnormal biochemistry parameters at discharge were subsequently followed up in high-risk follow-up outpatient department (OPD). Serum calcium, phosphorus, and ALP were monitored in them monthly till normalization. The study was started after the approval of the institutional ethics committee (IEC/313/21).

Statistical analysis

Statistical analysis was carried out for qualitative and quantitative variables. Frequency and percentage were calculated for qualitative data. Mean and standard deviation were calculated for quantitative data. To test the association between three or more dependent variables, repeated measures “ANOVA” was used. P value (significance) of < 0.05 was deemed statistically significant. Microsoft Excel 2010 was used to code raw data, and SPSS 25.0 was used to analyze data.

Results

During the study period of 10 months from October 2020 to July 2021, 78 babies ≤ 32 weeks or with birth weight ≤ 1250 g were eligible for MBD screen at 4 weeks in our NICU. Of these, 57 babies developed MBD and were stratified into three groups based on their gestation (≤ 28 weeks, 29–30 weeks, 31–32 weeks). Their baseline demographic characteristics are given in Table 1, and feeding details are given in Table 2.

Table 1 Baseline demographic characteristics

Table 2 Baseline nutritional characteristics

Among the baseline nutritional characteristics, there was a significantly longer duration of parenteral nutrition in ≤ 28 weeks and 29–30 weeks as compared to 31–32 weeks (Table 2). The time to reach full feeds, initiation of fortification, and number of days of fortification prior to the first evaluation of MBD was significantly prolonged in the lower gestations. However, the maximum feed volume and calorie received, type of milk fed, and feed intolerance episodes were comparable among the three groups.

The overall incidence of MBD in the present study population was 73%, while 100% of ≤ 28 weeks babies had MBD at discharge (Table 3). Fifty-four percent of babies was available for at least 1 follow-up post discharge.

Table 3 Incidence of MBD and follow-up rates of babies with MBD

Seventeen babies of ≤ 30 weeks required inorganic phosphorus supplementation in addition to calcium phosphate supplementation in order to correct the MBD. All infants required almost the same average doses of calcium, phosphorus, or vitamin D supplementation with no significant difference in the average dosage in the three groups (Fig. 1).

Fig. 1

Average supplementation of calcium, phosphorus, and vitamin D

There was no significant difference found in the serum calcium and serum phosphorus of the three groups at the varied postnatal age (PNA) of estimation and also at follow-up when they received optimum supplementation. The serum calcium level was maintained throughout in all the three groups. USG KUB was done in 51 infants prior to discharge to look for nephrocalcinosis due to calcium supplementation. None of them had evidence of nephrocalcinosis. The serum phosphorus level improved to > 5.5 mg/dl by 14 weeks in ≤ 28 weeks babies, by 12 weeks in 29–30 weeks babies, and by 12 weeks in 31–32 weeks babies when optimally supplemented. Serum phosphorus level normalized in all the babies by 42–44 weeks PMA. However, no significant difference was found in the serum ALP level of the three groups at the varied PNA of estimation except at 12 weeks PNA as the serum ALP level normalized by this time, i.e., 42–44 weeks PMA. The serum ALP level did not normalize till the last follow-up, i.e., 1.5 month corrected age in ≤ 28 weeks, which indicates that a longer time is required for normalization of this biochemical parameter in this subgroup (Table 4).

Table 4 Metabolic profile of infants with MBD at different postnatal ages

Methylxanthines, diuretics, and corticosteroids are known to increase the chance of development of metabolic bone disease. Infants who received caffeine with or without steroids or diuretics had a significantly higher incidence of hypophosphatemia and higher alkaline phosphatase levels at 4 weeks of age and also at discharge (Table 5).

Table 5 Effect of drugs on metabolic bone disease

The time to reach full feeds with fortification had no statistically significant effect on the incidence of metabolic bone disease as detected by serum phosphorus level and serum alkaline phosphatase level (Table 6).

Table 6 Effect of feeding and fortification on metabolic bone disease

The mean duration of hospital stay was 10 weeks in ≤ 28 weeks, 8 weeks in 29–30 weeks, and 6 weeks in 31–32 weeks of babies. Thus, the duration of hospital stay was higher in ≤ 28 weeks as compared to the other two groups due to obvious reasons of prematurity-related complications and the longer time required by the extremely preterm babies to reach suck-swallow-breathing coordination.

Discussion

In the present study, MBD was observed in 100% of extreme preterm babies and 69% of very preterm babies. Overall, the incidence of MBD was 73% in study population of ≤ 32 weeks/≤ 1250 g babies. The incidence of MBD reported in prior studies is variable. Peruri GP et al. reported that the incidence of MBD in the entire cohort and among very low birth weight (VLBW) neonates was 10.9% (19 out of 174) and 18.5% (15 out of 81), respectively [11]. Abdallah et al. found evidence of osteopenia in x-ray in 13.3% of their prospective cohort of 120 newborns born at < 34 weeks gestational age and < 1500 g birth weight. All the osteopenic infants had < 1000 g birth weight [12]. In the Perrone M et al. study, the rate of MBD was 22% in VLBW neonates with gestational age < 32 weeks, in which they are at an increased risk of developing MBD during hospital stays [13]. In Avila-the Alvarez A et al. study, using early biochemical criteria to identify infants at greater risk of MBD, they detected a prevalence of MBD of 12.3% [14]. However, Chinoy A et al. retrospectively reviewed the data of 57 ELBW infants of which 19 were diagnosed with rickets (33%) [15].

In the present study, most of the babies diagnosed with MBD were either on MOM or on combination of MOM with DHM. Serum phosphorus level normalized (increased to > 5.5 mg/dl) by 12–14 weeks of PNA, i.e., 42–44 weeks PMA. The serum ALP level also normalized by 42–44 weeks PMA in the very preterm population. The group of ≤ 28 weeks extremely preterm gestation babies had high ALP levels even at 42–44 weeks PMA. A larger sample size with a longer duration of follow-up is required to opine about the time when biochemical parameter normalizes in the various subgroups of preterm babies. These findings are in accordance with the study conducted by Alejandro Avila-Alvarez et al. [14]. In a review done by Faienza et al., it was found that phosphate supplementation should be considered for values < 5.5 mg/dl (1.3 mmol/L), but one may wait till values fall below < 4 mg/dl (1 mmol/L), to promote bone mineralization and to prevent hypercalciuria [16]. In the present study too, we found that 17 babies required inorganic phosphorus supplementation in addition to calcium phosphate supplementation in order to correct the MBD. This was observed predominantly in babies less than 30 weeks age. Thus, there may be requirement of additional phosphorus in this group which may be a separate entity to MBD in 30 weeks and below.

All neonates below 30 weeks gestation required parenteral nutrition, and majority (88.8%) had developed MBD which is comparable to a retrospective case-control study of infants with gestation age < 30 weeks and birth weight < 1000 g wherein they found longer period of parenteral nutrition to be associated with MBD [17]. Average duration of parenteral nutrition was 21 days. Abdallah et al. found that birth weight and gestational age were significantly inversely related to serum ALP levels [12]. This was also observed in our study. They also considered that high levels of ALP can be considered a reliable biomarker to predict the status of bone mineralization and the need for radiological evaluation in premature infants particularly those with < 1000 g birth weight and < 32 weeks gestation. We found higher values of alkaline phosphatase in our cohort of ELBW babies too; however, we had only 10 extreme preterms in our cohort which is a limitation in our study.

Some drugs frequently used in the NICU have been previously associated with MBD in preterm infants. Examples include hypercalciuric drugs such as furosemide or methylxanthines, both of which increase calcium loss, and steroids, which decrease bone formation by supporting osteoclast differentiation and inhibiting osteoblast growth [14]. In the current study, we observed that use of drugs like caffeine, steroids, and furosemide can increase the risk for development of MBD. However, the time taken to reach full feeds with fortification had no significant effect on development of MBD. MBD is a significant problem in very preterm infants which can lead to both short term and long-term implications. These findings are correlated with the study done by Ukarapong et al. [17] and Ali et al. [18].

Limitations

The sample size was small. Only 54% of infants enrolled in the study completed the last follow-up because of the migrant nature of the population and the lockdowns imposed in the COVID-19 pandemic. Vitamin D and PTH levels were not done due to financial constraints.

Conclusion

The overall incidence of MBD was 73% in the study population of ≤ 1250 g babies. The profile outlined in the present study matches the literature reports in many aspects, revealing the importance of characterizing this group for the prognosis and short- and long-term follow-up of newborns with bone metabolic disease. However, drugs like caffeine, steroids, and furosemide have significant impact on the development of metabolic bone disease. A larger sample size with a longer duration of follow-up is required to opine about the time when the biochemical parameter normalizes in the various subgroups of pretem neonates.