*2.7. Outcome Measures*

Primary endpoint: Change in BMI, BMI centile after 26 weeks of vitamin D supplementation. Secondary endpoints: change in body composition and bone mineral density and vitamin D level after 26 weeks of vitamin D supplementation. Additionally, we assessed changes in blood level of vitamin D.

#### *2.8. Sample Size*

Assuming probability of the event (at least 10% reduction in baseline BMI over the follow–up period) at 0.85 and 0.6 for the experimental and control group, respectively, minimum sample size providing 0.9 statistical power for alpha equal 0.05 or lower and beta equal 0.1 or lower was estimated at 130 (65 per group).

#### *2.9. Statistical Analysis*

Normal distribution of continuous variables was verified with the Shapiro–Wilk test. Descriptive statistics are presented as the mean or median and standard deviation from the mean. Between groups comparisons were carried out using the Mann–Whitney U test. Nonparametric tests were chosen because of the large number of significant Shapiro tests, which were used for normality assumption assessment. All statistical tests were two–tailed and performed at the 5% level of significance. All analyses were performed on the intention–to–treat basis, in which all of the participants in a trial are analyzed according to the intervention to which they were assigned, analyzing only participants who completed the whole weight management intervention program. Statistical analyses were performed with the Statistica 13 (TIBCO Software Inc., Tulsa, USA 2017).

#### **3. Results**

After obtaining and verifying the correctness of the recorded data (cross-checked with source data of 30% of patients), the data was decoded according to the protocol received from OCSR UCK in Gdansk.

After decoding, a full analysis of available anthropometric data and the results of densitometric examination was performed for patients divided into groups—receiving placebo and vitamin D.

A total of 170 patients were eligible for enrolment in the study during the enrolment period. A total of 152 patients were eventually included in the study (*n* = 8—no parental consent, *n* = 10—vitamin D intake during three months before the study). Subsequently the patients were randomly assigned to groups: 67 to the placebo group and 85 to the Vitamin D group. Out of the 152 qualified patients, 109 (72%) completed a full cycle of four visits scheduled in the program. In the placebo group, 64 patients completed the active phase of 26 weeks of supplementation, 53 completed the comprehensive treatment program (52 week duration).

The patient flow is shown in Figure 2.

Basic demographic and clinical data for both groups are presented in Table 1.


**Table 1.** Basic anthropometric and clinical data of studied groups (vitamin D and placebo) at visit 1.

*p* < 0.05 Mann–Whitney U test analysis.

**Figure 2.** Patient flowchart in the study

There were no statistically significant differences between the vitamin D and placebo groups at the start of the study.

The differences between the groups in terms of main endpoints were subsequently assessed. The results of basic anthropometric and 25(OH)D concentration tests are shown in Table 2.


**Table 2.** Results of basic anthropometric and 25(OH)D concentration tests in studied groups.

*p* < 0.05 Mann–Whitney U test, \* t–student test for independent samples.

The results of bioelectric impedance measurements are shown in Table 3.


**Table 3.** Results of bioelectric impedance measurements in studied groups.

*p* < 0.05 Mann–Whitney U test, \* t–student test for independent samples. BI—bioimpedance, FM—fat mass, Δ – delta–difference, MM—muscle mass.

The results of the dual X–ray absorptiometry (DXA) measurements are presented in Table 4.

**Table 4.** Results of dual X–ray absorptiometry (DXA) measurements in studied groups.


*p* < 0.05 Mann–Whitney U test, \* t–student test for independent samples. Sp—spine, BMD—bone mineral density, Δ—delta–difference, TBLH—total body less head, TFM—total fat mass, TLM—total lean mass.

Both groups had a reduction in BMI centiles. Although the reduction was greater in the vitamin D vs. placebo group (–4.28 ± 8.43 vs. –2.53 ±6.10) the difference was not statistically significant (*p* = 0.319). Similarly the reduction in fat mass—assessed both using bioimpedance and DEXa was achieved, yet the differences between the groups were not statistically significant, as shown in Tables 3 and 4.

The analysis showed statistically significant differences between the groups only in 25(OH) D3 concentration in the measurements taken after the supplementation period (24.99 vs. 16.25; *p* = 0.000) and in the difference between second and first measurement of vitamin D levels (6.06 vs. –4.24; *p* = 0.000), and in the difference between second and first measurement of bone mineral density in the spine (Sp BMD) (0.04 vs. 0.06; *p* < 0.0256). The difference was higher in the placebo group.

There was no difference between the placebo group and the vitamin D group in BMI reduction, BMI centile, fat tissue in kg, % of fat tissue (assessed both by performing BIA and densitometry analysis).

Additionally, we have performed an analysis of correlations (Spearman rank correlation coefficient) between vitamin D levels at Visit 1 and 4 and changes of vitamin D level between the visits dependent on several anthropometric variables. Results are presented in Table 5.


**Table 5.** Correlation between vitamin D levels and anthropometric variables.

Spearman rank correlation coefficient; bolded when *p* < 0.05, BI—bioimpedance, FM—fat mass, Δ—delta–difference, MM—muscle mass, BMD—bone mineral density, TBLH—total body less head, TFM—total fat mass, TLM—total lean mass.

The results show that there were no important correlation between the initial BMI centile and vitamin D level at the first visit. Additionally the correlation did not show any significant influence of BMI changes on changes in vitamin D level after the supplementation period. There were significant negative correlations (*p* < 0.05) between fat mass % and levels of vitamin D during both visits—which can confirm the relation between the fat mass and blood level of vitamin D. Yet this was a rather weak correlation (–0.25 to –0.21). Further, it was not confirmed in DXA measurements.

#### **4. Discussion**

#### *4.1. E*ff*ect of Vitamin D Supplementation on Body Mass Reduction*

Presented study is the first randomized trial to assess potential effects of vitamin D supplementation in body mass reduction in overweight and obese children. Results of present study show that supplementation of vitamin D did not have a statistically significant, put potentially clinically important, influence on body mass (BMI, BMI centile) body composition or bone mineral density comparing to placebo groups during an organized obesity management program in children.

Biological role of vitamin D in etiopathogenesis of metabolic syndrome represents an interesting issue. Previous studies conducted among children revealed inverse relationship between blood concentration of vitamin D and waist circumference, systolic blood pressure, insulin resistance, fasting glucose, total cholesterol, triglycerides and LDL cholesterol, as well as positive association between the concentration of vitamin D and HDL cholesterol [25,38,39]. It seems that vitamin D can interfere with secretion of insulin both directly—binding to its receptors [VDR] on pancreatic β cells, and indirectly by modulating concentration of calcium in extracellular space [40].

Importantly, a positive association between the concentration of vitamin D and sensitivity to insulin was observed in obese children, along with an inverse relationship between the level of this vitamin and concentration of glycated hemoglobin (HbA1c) [41]. Supplementation with vitamin D in obese adolescents resulted in decrease of insulin resistance, while levels of inflammatory markers [CRP, TNF–α, IL–6] remained unchanged [42]. Till now, to the best of our knowledge, no studies were carried out in children or adolescents regarding the effects of vitamin D supplementation on body mass reduction during an organized lifestyle modification program. As we presented, adding a 1200 IU/day dose of vitamin D did not lead to higher changes in BMI (BMI centile) or fat mass as well as fat free mass changes in children aged 6–14. It needs to be stated that the results presented show that although there was a reduction in BMI in vitamin D group compared to placebo group (–0.46 ± 1.80 vs. 0.11 ±1.84) and BMI centiles also showed higher reduction in children supplemented with vitamin D (–4.28 ± 8.43 vs. –2.53 ±6.10) none of those results was statistically significant (*p* = 0.203 and *p* = 0.319 respectively). Similar results can be found in bioimpedance measurements, but not DXA assessment. Presented results would be presented as clinically valid. Although this 26-week long supplementation had an influence on blood concentration level of 25(OH) D in the active treatment group, only 6 out of all patients reached a level above 30 ng/ml in the final assessment (2 in the placebo group) 52 weeks after the start of the intervention. This shows that vitamin D can potentially have an impact on weight loss level but possibly due to resignation ratio or sample size we were not able to show that effect.

As previous studies in adolescents and adults showed, low (~1000 IU/daily) or high doses (up to 300,000 IU/month) of vitamin D supplementation have a very mixed results in influencing changes in fat mass, free fat mass or muscle mass [43–45]. Data regarding effect of vitamin D supplementation on body mass/body fat changes are limited to studies and meta–analyses/ reviews regarding adults with wide range of interventions (medical weight loss, bariatric surgery, low–caloric diet) were used together with vitamin D supplementation. Additionally those studies were focused on finding the optimal dose of supplementation to reach the optimal (>30 ng/l) level of vitamin D [44] or assessing the association of vitamin level and body fat. Meta-analysis of studies in obese adults showed that although there was an impact of vitamin D supplementation on body fat, the results were not statistically significant [43]. This study also acknowledges that the impact of vitamin D supplementation on body fat reduction has a linear effect up to 2000 IU/d, with no benefits in increasing the dose in adults. The only review regarding impact of vitamin D supplementation on bone mineral density was focused on general population—as no studies was directly focused on overweight/obese children. In addition, no sub–analysis was performed to show the effect depending on body mass. Winzenberg states that the impact of vitamin D supplementation can be higher in vitamin D deficient children, but the effect is small—in most studies standardized mean difference between groups was small (<0,3) [45]. Yet all of these measures are only proxy measures to changes in body weight as a primary outcome of supplementation of vitamin D in obese children/adolescents. As for now, no studies have demonstrated effectiveness of such a strategy for supporting body fat reduction throughout a long–term interventional process.

#### *4.2. E*ff*ect of Vitamin D Supplementation on Bone Mineral Density During Weight Loss*

Metabolic effects of obesity on growth and maturation of bones are still not fully understood. Moreover, the results of previous studies analyzing bone mass and density in obese individuals are highly inconclusive. While some authors claimed a decrease in bone mass relative to body weight [46], others did not document significant differences in bone mineral density [47] or showed an increase in body mass and bone size in obese children, adolescents and adults. Increased bone mass and density observed in obese individuals is postulated to be a response to greater mechanical load, direct influence of leptin or enhanced enzymatic activity of aromatase [48,49]. Nevertheless, obesity markedly increases the risk of bone fractures in children [50]. Vitamin D plays important biological role in the process of bone maturation and mineralization. Previous studies documented an inverse relationship between blood concentration of vitamin D and bone mineral density [51,52]. One meta–analysis revealed that supplementation of vitamin D can improve both bone mineral density and bone mass in individuals with low blood levels of this vitamin [45]. The effects of supplementation are particularly favorable in premenarcheal girls with normal body weight, in whom administration of vitamin D resulted in increases of both bone mass and fat–free mass [53]. An analysis of 58 morbidly obese teenagers showed that individuals with physiological blood concentration of PTH (parathyroid hormone) have normal bone mineral density, irrespectively of their vitamin D levels [54]. In contrast, a recently published study involving a small group of adolescents with obesity (*n* = 24) and normal body weight (*n* = 25) showed that obese people present with higher bone mineral density, irrespectively their blood concentration of vitamin D and despite lower level of physical activity than their normal-weight peers. Moreover, the differences in bone mineral density turned out to be independent from fat-free mass content. Furthermore, bone mineral density was associated with blood concentrations of leptin and insulin [55]. Our study shows clearly that we can observe an increase of bone mineral density both spinal and subtotal BMD with body mass reduction. Yet there is no impact of vitamin D supplementation on the level of bone density assessed using DXA. This shows that the body mass reduction itself impacts the bone mineralization the most.

Apart from many unquestioned favorable health effects of losing excessive weight, this process may also be associated with enhanced bone turnover and decrease in bone mineral density. In recently published systemic review, the decrease of bone mass was reported following calorie-restricting diet but not in exercise–induced weight-loss [56]. However, this evidence originates mostly from studies conducted among adults [57–59], and to the best of our knowledge, the issue in question was a subject of only one study of adolescents after bariatric surgeries [60]. The results from previous intervention studies suggest that a low-calorie albeit high-protein (ca. 30%) diet, containing high amounts of dairy products, can prevent the loss of bone mass and a decrease in bone mineralization [61]. Yet supplementation with high doses of vitamin D can have a negative effect on bone mineralization in adults—as a recent study by Burt et al. showed [62].

Study limitations:

This study has some limitations that need to be taken into account when assessing the usefulness of the results:

– Children aged 6–14 years old were included in the study—this is not a homogenic group when it comes to maturation/puberty status—and this has an impact on both the ability to decrease body mass and bone turnover and mineralization;

– We were giving one dose of vitamin D (1200 IU) to all participants independent the body mass and age, which could have result in less effective increase of 25(OH)D level, which in turn could have impacted the changes in body mass or bone mineral density;

– 26 weeks may be too short to establish the effects of vitamin D supplementation on the rate of skeletal mineralization

– There was an almost 30% drop–out, seen especially in the second part of the program—after finishing the active supplementation with vitamin D/placebo period. This was unlikely to be due to the treatment itself (as there were no important side effects registered). The level of lost to follow-up patients in the interventional program is similar to other such programs seen in Poland and other European countries [63,64].

– Finally, the sample sizes were smaller than anticipated and there is a possibility that with the estimated reduction of 10% of BMI reduction the study sample is underestimated. It is possible that the study was underpowered to detect smaller changes in BMI/BMI centiles as well as other parameters.
