Formula Milk Supplementation and Bone Acquisition in 4–6 Years Chinese Children: A 12-Month Cluster-Randomized Controlled Trial

Dairy foods are crucial for adequate calcium intake in young children, but scarce data are available on the effects of formula milk on bone acquisition. This cluster-randomized controlled trial investigated the effects of the supplementation of formula milk on bone health in rural children accustomed to a low-calcium diet between September 2021 and September 2022. We recruited 196 healthy children aged 4–6 years from two kindergartens in Huining County, Northwest China. A class-based randomization was used to assign them to receive 60 g of formula milk powder containing 720 mg calcium and 4.5 µg vitamin D or 20–30 g of bread per day for 12 months, respectively. Bone mineral density (BMD) and bone mineral content (BMC) at the left forearm and calcaneus, bone biomarkers, bone-related hormones/growth factors, and body measures were determined at baseline, 6, and 12 months. A total of 174 children completed the trial and were included in the analysis. Compared with the control group, formula milk intervention showed significant extra increments in BMD (3.77% and 6.66%) and BMC (4.55% and 5.76%) at the left forearm at 6th and 12th months post-intervention (all p < 0.001), respectively. Similar trends were observed in BMD (2.83%) and BMC (2.38%) in the left calcaneus at 6 months (p < 0.05). The milk intervention (vs. control) also showed significant changes in the serum concentrations of osteocalcin level (−7.59%, p = 0.012), 25-hydroxy-vitamin-D (+5.54%, p = 0.001), parathyroid hormone concentration (−15.22%, p = 0.003), and insulin-like growth factor 1 (+8.36%, p = 0.014). The percentage increases in height were 0.34%, 0.45%, and 0.42% higher in the milk group than in the control group after 3-, 6-, and 9-month intervention, respectively (p < 0.05). In summary, formula milk supplementation enhances bone acquisition at the left forearm in young Chinese children.


Introduction
Osteoporosis is a severe global public health concern, particularly in women [1][2][3]. Low bone mass acquisition in childhood is an important risk factor for osteoporosis at later ages [4][5][6]. Maximizing peak bone mass before skeletal maturity by improving modifiable factors, such as diet and physical activity [7], will delay the time for osteoporosis in the middle or elderly ages [8].
A previous meta-analysis comprising 19 randomized controlled trials (RCT) performed on 2859 children with an average age of 10 years determined that calcium supplementation (800 mg/d) for 1-2 years increased the bone mineral density (BMD) of the total body

Study Design and Intervention Programs
This study was a 12-month, cluster-randomized controlled trial. One hundred ninety-six children were randomly assigned to the formula milk and control groups by a lottery method according to their natural kindergarten class, matched for the year grade. In the milk group, each child was given 60 g/d of formula milk powder (blended into 300 mL of milk) targeted for 3-6 years children (QQ Star Milk Powder, Yili Industrial Group Co., Ltd., Hohhot, China). The daily dose contained 720 mg of calcium and 4.5 μg of vitamin D (Table S1). Briefly, the formula milk was given as a part of breakfast or an extra meal after lunch. The consumption was recorded daily by kindergarten teachers or their guardians. For the control group, 20-40 g of bread was given per day to each participant during the days in kindergarten.

Study Design and Intervention Programs
This study was a 12-month, cluster-randomized controlled trial. One hundred ninetysix children were randomly assigned to the formula milk and control groups by a lottery method according to their natural kindergarten class, matched for the year grade. In the milk group, each child was given 60 g/d of formula milk powder (blended into 300 mL of milk) targeted for 3-6 years children (QQ Star Milk Powder, Yili Industrial Group Co., Ltd., Hohhot, China). The daily dose contained 720 mg of calcium and 4.5 µg of vitamin D (Table S1). Briefly, the formula milk was given as a part of breakfast or an extra meal after lunch. The consumption was recorded daily by kindergarten teachers or their guardians. For the control group, 20-40 g of bread was given per day to each participant during the days in kindergarten.

Bone Mineral Status Measurements
The primary outcome indicators were BMD and BMC, measured at the left forearm and calcaneus by an EXA-3000 peripheral dual-energy X-ray bone densitometer obtained from OsteoSys (Seoul, Korea) at 0 (baseline), 6, and 12 months post-intervention. The specific test site for the forearm is the ulna and radius in the distal 1/3 of the left forearm, while that for calcaneus is the central area of the left calcaneus bone at approximately 2 cm 2 . For quality control, the final scans at all skeletal sites were analyzed based on a reference of the baseline scan to minimize localization errors in positioning the region of interest. Each site had two readings, and the mean value was used for further analysis. Two specialists conducted all the scanning. Among 58 children, the CVs of BMD and BMC were less than 2% at the measured sites for two consecutive measurements with repositioning on the same day.

Anthropometric Examinations
Body height and weight were assessed at baseline and at 3, 6, 9, and 12 months postintervention. Body height was determined to the nearest 0.1 cm and body weight to the nearest 0.5 kg, with the participants wearing light clothing and no shoes. In addition, the measurements of height were controlled to be taken within 2 h on the same day for quality control.

Assessments of Dietary Intake and Physical Activity
Demographic information of all participants (e.g., sex, age, gestational age at birth, birth weight, breastfeeding time after birth, dairy products intake, the history of disease and medication, physical activity, and sleeping) and their parents (e.g., age, height, weight, and education level) were obtained by using face-to-face questionnaire interviews at the enrollment. In addition, the participants were required to complete a 3-day (including one weekend day) food diary, noting adverse reactions and side effects, physical activity, and sleeping time with their guardians' assistance at each visit at 3, 6, 9, and 12 months [27]. Photographs of food portions were supplied to help to estimate food intake, and further analysis was performed using the average values of the 3-day food diaries. Moreover, the exercise (e.g., activity steps, walking distance, and running distance), heart rate, and sleeping time for seven consecutive days were monitored by health bracelets (Xiaomi NFC5, Guangdong, China).

Statistical Analysis
The sample size was calculated based on an 18-month intervention trial on the effects of calcium supplementation (300 mg) on forearm bone health in Chinese children aged seven years old [28], in which the percentage mean (SD) difference in the changes of BMC at the forearm between intervention and control group was 2.5 (4.5%) g/cm 2 . Seventy-four children per group were provided 85% power to detect a 2.5% difference in BMD change between the milk supplement group and control group at α = 0.05. Eighty-two participants were required if a 10% dropout rate was considered. Finally, 96 and 100 participants (103 girls and 93 boys) were recruited into the intervention and control groups.
Participants who completed all visits for BMD/BMC tests were included in the final analyses. Continuous variables were presented as means ± SDs or medians and interquartile (IQR), and categorical variables were expressed as numbers (percentages). Baseline characteristics were compared using Student's t-test for normally distributed data, Kruskal-Wallis H rank sum tests for skewed distributed data, and Chi-square tests for categorical data. For the effects of milk intervention, the primary analyses compared the mean difference of 12-month changes in BMD and BMC at the left forearm and calcaneus using Student's t-test. Additional analyses included comparisons of the differences between the formula milk and control groups in the study indices, including absolute changes and percentage changes in BMD, BMC, bone metabolism biomarkers, height, and weight at each visit using Student's t-test or Kruskal-Wallis H rank sum tests for the univariate analyses and using analysis of covariance (ANCOVA) after adjusting potential covariates. The ANCOVA was adjusted for factors related to demographics, economics, baseline values for each indicator, dietary intake, physical activity, and other outcome health indicators. Continuous variables with skewed distribution were transformed to normality before the ANCOVA. All statistical tests were two-sided. A p-value less than 0.05 was defined as a significant difference. SPSS version 25 (IBM Statistics, IBM Corporation, New York, NY, USA) and R 4.1.2 were used for the analyses.

Subjects' Characteristics and Intervention Adherence
Of the 196 children enrolled, 22 withdrew during the study period, and the final group sizes for the milk and control groups were 83 and 91, respectively. The reasons for their withdrawal are presented in Figure 1. There were no significant differences between the milk and control groups in dropout rates (13.54% vs. 9.0%, p = 0.435). The baseline characteristics were all similar in the two groups, as depicted in Table 1 (p > 0.05). The overall compliance in the milk group was 92.3%, equivalent to an intake of 665 mg of calcium from formula powder.

Dietary Intake and Physical Activity during the Study Period
Tables 2 and 3 display the mean daily intake of food categories and nutrients in milk and control groups during the intervention. The milk group showed a higher mean dairy consumption than the control group (median, 300 mL vs. 0 mL, p < 0.001) based on the estimations of four 3-day food diaries during the 12-month study period. However, there was no significant difference in the intake of other major food categories (e.g., cereals, beans, animal foods, vegetables, fruits, edible fungus, nuts, and beverages) between the milk and control groups ( Table 2). As shown in Table 3, the milk group had significantly increased the median daily intakes of protein (42.5 g vs. 34.0 g), vitamin D (VD) (5.4 µg vs. 1.0 µg), and calcium (839 mg vs. 196 mg) compared with the control group (all p < 0.001).
In addition, marked increases in other nutrients (such as vitamin A, vitamin B 2 , folic acid, potassium, magnesium, and zinc) intake were also observed in the milk group (all p < 0.05). On the other hand, there were no significant differences in the median values of walking steps (9628 vs. 9463), running distance (1341 m vs. 1239 m), and sleeping time (7.7 h vs. 7.5 h) per day between the control and supplementation groups according to the three 7-day health bracelets monitors (all p > 0.05) ( Table 4).      Table S2 shows no significant differences in baseline values of BMD and BMC at the left forearm and calcaneus between the milk and control groups (p > 0.05). However, after the 12-month intervention, bone mineral acquisition at the left forearm was significantly greater in the milk group than in the control group in both univariate and multivariate analyses ( Figure 2, Table S2). In the univariate model, the mean (±SD) increments in BMD and BMC between the milk and control groups were significantly higher in the 6th and 12th month (all p < 0.0001). The mean percentage differences in bone acquisition in the milk group compared to the control group were 3.77% and 6.66% for BMD and 4.55% and 5.76% for BMC at 6 and 12 months (all p < 0.0001). For BMD and BMC changes at the left calcaneus, a similar trend (but less significant) was observed in the milk (vs. control) group. The mean percentage differences in bone acquisition in the milk group compared to the control group were 2.83% for BMD (p = 0.013) and 2.38% for BMC (p = 0.024) in the 6th month. However, there was no significant difference between the groups in the left calcaneus BMD and BMC in the 12th month (all p > 0.05). Moreover, the statistical differences were not significantly altered after adjusting the potential confounders in the multivariate model.    Figure 3 shows that the percentage increases in height are 0.34%, 0.45%, and 0.42% higher in the milk group than in the control group after a 3-, 6-, and 9-month intervention, respectively (p < 0.05). However, the beneficial difference did not remain significant at the end of the trial (p > 0.05). There was no significant difference in the weight gain between the two study groups at each follow-up visit during the whole intervention period (p > 0.05).

Effects of Formula Milk Intervention on Bone Metabolism Biomarkers
As presented in Figure 4 and Table S3, the biomarkers of bone resorption and formation had no significant difference in their changes between the two groups throughout the trial except for osteocalcin in the 6th month and ALP in the 12th month.

Discussion
The results of this cluster randomized controlled trial indicated that 60 g/d formula milk supplementation over 6 and 12 months significantly increased bone mass acquisition.
In addition, milk supplementation for 3-9 months could marginally increase the height gain. The increases in 25(OH)D and IGF-1 and the decrease in PTH may support (at least in part) the beneficial changes in bone acquisition.

Formula Milk/Calcium Supplementation and Bone Mineral Acquisition
The magnitude of the 60 g/d of milk supplementation effect was considerable, with a significant increase in left forearm BMD of 6.66% and in BMC of 5.76% after 12 months in children aged 4-6 years. Our findings were consistent with those in previous studies, demonstrating that increased calcium [9,[29][30][31][32][33][34][35][36][37][38][39][40] or dairy products [11,21,41] intake increases bone mineral status. For example, a randomized, double-blind, placebo-controlled study in 160 Gambian 8-12 years children accustomed to a low-calcium diet (338 ± 142 mg/d) revealed that calcium supplementation (1000 mg/d) for 12 months significantly enhanced the BMC and BMD in the distal radius (8.4% and 7.0%) and midshaft (3.0% and 4.5%) (all p < 0.05) [18]. Another 18-month randomized trial on adolescent girls (mean age 12 y) showed that calcium supplementation (792 mg/d) had significantly greater BMD and BMC gains at the total body and upper limb (all p < 0.05) [42]. In a 2-year milk intervention trial in a school conducted on Beijing girls aged ten, a pronounced BMD increase in the total body (3.2%-5.3%) was observed compared to the control group [13]. Thus, these supported the beneficial effect of calcium supplementation on bone mineral gains in childhood populations. Our findings highlighted the importance of adequate milk consumption in rural young children with less calcium-enriched foods in China.
The milk-attributable gains in BMD and BMC were much more prominent in the non-weight-bearing forearm than in the weight-bearing forearm in this study. A metaanalysis involving 2859 children showed that an average dose of 800 mg/d of calcium supplementation also showed more benefit to BMD at the non-weight-bearing areas (total body and upper limb) than the weight-bearing sites (femoral neck and lumbar spine) in children with a mean age of 10 years [9]. A less significant beneficial effect of milk supplementation on BMD and BMC gains at the calcaneus (weight-bearing site) was found in this population. In accordance with our results, a randomized and placebo-controlled intervention study reported a significant effect of calcium supplementation on BMC gain at the radius-ulna but not the weight-bearing site (tibia-fibula) [7]. It might be because bone acquisition in the weight-bearing sites was more susceptible to gravity effects [18], which could easily mask the effects provided by environmental factors (e.g., diet). On the other hand, it has been reported that the combination of high doses of calcium and exercise showed more significant bone acquisition than that of either exercise or calcium alone at the weight-loaded sites of the hip and femoral [43]. Therefore, the effects of environmental factors (e.g., diet) on weight-bearing calcaneus bones has yet to be validated in studies with large samples, longer follow-up times, and a combination of calcium and exercise intervention in preschool children.

Formula Milk/Calcium Supplementation and Bone Metabolism Biomarkers
The benefit of milk intervention on bone gains was supported by bone-related biomarkers, such as increased serum 25(OH)D (mainly D 3 ) and IGF-1, as well as reduced PTH, as did in a previous study [44]. The results indicated that supplemental milk had no apparent changes in most bone resorption and formation biomarkers. Previous studies showed that VD supplementation and higher circulating VD were associated with better BMD in Children [45]. Sunlight irradiation could significantly affect the circulating 25(OH)D level [46]. We observed a decreased serum 25(OH)D concentration in both groups in the early Spring (March 2022) compared to the baseline values in September 2021 due to the reduced sunlight irradiation in a long winter. Therefore, it suggested that daily supplementation of 4.5 µg vitamin D was insufficient to prevent the decrease in circulating 25(OH)D in the winter/spring in young children in Northwest China. However, there was a less decreased 25(OH)D level in the formula milk group (vs. control). Furthermore, a similar 25(OH)D concentration was found at the end of the intervention in September 2022 in both groups and to those at the baseline, suggesting the intervention dosage (4.5 µg vitamin D) could not provide additional benefit to the circulating 25(OH)D in the summer/autumn in the study population.
Parathyroid hormone (PTH) is essential in bone metabolism, especially during pediatric growth. It has been reported that PTH concentrations increased with age during childhood. A peak PTH concentration was observed in the 10-14 years with the most rapid bone growth [47]. Our study also suggested an increasing trend in PTH with bone growth at 4-6 years of age. It has been reported that calcium supplementation could decrease bone remodeling by inhibiting PTH secretion. At the same time, an increase in BMD is the consequence of space-filling for remodeling in adults [48,49]. In the present study, relatively lower increases in plasma PTH concentration were observed in the formula milk group compared to the controls, suggesting that the above mechanism was similar in children [38]. A previous meta-analysis of randomized controlled trials also reported that increased dairy intake could elevate the IGF-1 level [50]. In addition, it has been demonstrated that periosteal bone apposition might be stimulated by IGF-1, contributing to a slightly larger skeletal envelope in the milk group [51]. Therefore, the changes in IGF-1 in the milk supplementation group may support the benefits of height and bone growth.

Nutritional Factors in Formula Milk and Bone Health
The effect of formula milk powder may be attributed to milk-contained nutrients. A daily dose of the milk contains 720 mg of calcium and 4.5 mg of VD, resulting in an extra intake of 643 mg of calcium and 4.4 µg VD in the milk group compared to the controls with deficient intakes of calcium (196 mg/d) and VD (1.0 µg/d). The remarkable improvement of BMD and BMC in the milk intervention group indicated that the habitual calcium and/or VD intake was much less than their requirements for optimal bone acquisition. It highlighted the importance of adequate milk consumption in young children. In addition, formula milk provides a relatively rich source of high-quality protein and other nutrients (e.g., vitamin K, folic acid, potassium, magnesium, and zinc) that could help improve bone health in children and adolescents [52,53].

Strengths and Limitations
This study has several strengths. First, the increased final study sizes (final/planned: 174/148) would improve the statistical power of this trial. The high adherence (92.3%) to the milk supplementation in the milk group reduced the likelihood of false negatives for the primary study outcome. Next, the consumption of dairy products did not significantly increase in the control group, showing no obvious contamination of the intervention. Finally, the comparability in baseline characteristics and physical activities, non-dairy foods, and sleeping time during the intervention period between the two groups would attenuate potential confounding bias.
Some limitations should be noted. Firstly, we could not exclude the possibility of introducing information bias due to no placebo control and double-blind measures. However, we reduced the potential by using objective outcomes, duplicated reading for BMD and BMC, and determining the laboratory indices without knowing the group status. Secondly, the cluster randomization method was less efficient in balancing the confounding factors than the individual-based randomized technique. However, the two groups had no significant differences in the main baseline characteristics. Lastly, the sample size of the trial was calculated based on the difference (SD) in changes of BMC at the upper limb, and we might not have sufficient statistical power to detect the differences in the other indices (e.g., bone biomarkers, BMD/BMC at the calcaneus).

Conclusions
In conclusion, the study demonstrates that supplementing 60g/d of formula milk powder with fortified calcium and VD over 6 and 12 months enhances bone mass acquisition at the left forearm in Chinese preschool children. The marginal increase of the height gain during the 3-9 months of milk intervention needed further replication in long-term intervention studies. The increase in 25(OH)D and IGF-1 and the decrease in PTH may support (at least in part) the beneficial changes in bone acquisition. Our findings highlight the importance of adequate milk consumption in young children in a rural region in China.
Supplementary Materials: The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/nu15082012/s1, Table S1: Composition of formula milk powder; Table S2: Comparison of means of bone mineral density (BMD) and bone mineral content (BMC) at baseline and their 12-month changes between the formula and control groups; Table S3: Comparison of means of bone metabolism markers at baseline and their 12-month changes between the formula and control groups.