Participants and study design
The present study is part of a double-blind, randomized clinical trial, which aimed to investigate the effects of 2000 and 4000 IU/d of vitamin D3 supplementation on serum 25(OH) D concentration, body composition, and anthropometric measures in nursing mothers with overweight or obesity, as well as to assess the growth and risk of infection in their infants. Postpartum nursing mothers were recruited at the private hospital’s maternity ward in Qazvin province from November 2018 to March 2019 (autumn to winter). Inclusion criteria were participants aged 20–49 with a BMI 25–39.9 kg/m2 who delivered at term (gestational age of 37–42 weeks),and a birth weight appropriate for gestational age (2500–3900 kg). In addition, the mothers had to undertake to continue breastfeeding for the 3 months of the study. The exclusion criteria were having diagnosed gastrointestinal disorders interfering with bowel function, having severe hepatic, renal, inflammatory, cancer, diabetes, hypertension, epilepsy, and thyroid diseases, or taking any medication, a history of smoking or alcohol consumption in the past month and adhering to a specific diet during the past 12 weeks.
The sample size was calculated according to Roosta et al.’s study [26] by two mean comparison formulas, type I error (α) =0.05, type II error (β) =0.2, and mean (Standard Deviation) of waist circumference (WC) changes. Mean changes of WC after supplementation of vitamin D at the end of the study of Roosta et al. [26] were equal to 1.91 ± 1.7 cm and 0.55 ± 1.04 cm in the intervention and control groups, respectively. For power = 0.8 and (α) value = 0.05, the sample size was calculated to equal 18. Since there were 3 groups in this study, the calculated sample size was multiplied by √ (the number of groups) [27]. Finally, the required sample size for this study was 90 participants.
The protocol for the present study was published previously [28] and was registered in the Iranian Registry of Clinical Trials (http://www.irct.ir: IRCT20140413017254N6) on 11-04-2018. The Ethics Committee of Tehran University of Medical Sciences approved the study (IR.TUMS REC 1397429).
Outcomes
The outcomes of the study were WC, weight, BMI, body fat percentage (BFP), FM, fat-free mass (FFM), skeletal muscle mass, relative fat mass index (RFMI), fat mass index (FMI), serum concentrations of 25(OH) D, calcium, iPTH and phosphorus.
Randomization and intervention
Ninety nursing women with overweight or obesity were randomly allocated to three groups: two groups of vitamin D3 supplementation 2000 IU/d (VD1, n = 32), 4000 IU/d (VD2, n = 29) and placebo group (PL, n = 29). The participants were randomly assigned to groups with a 1:1:1 randomization ratio. Randomization was performed by an assistant using permuted block randomization method, and stratified randomization was employed to match the women based on age (20–34 and 35–49 years) and BMI (25–29.9 and 30–39.9 kg/m2). The intervention allocation was blinded for both investigators and participants. Vitamin D3 (cholecalciferol) and placebo (lactose) supplements were in the form of nano microcapsules, which have provided by the Nano Hayat Darou Industrial Co. (Tehran, Iran). Mothers are instructed to consume two nano micro capsules daily, one capsule with lunch and dinner. The capsules were identical in size, color, and shape. The vitamin D3 supplement and placebo were packed by the Pharmacy. The double-dummy method was used for the double-blind study. The intervention started approximately 3 days after delivery and continued up to 12 weeks. Mothers were called up every week. The number of returned capsules were recorded at the final visit to calculate compliance and adherence to the intervention.
Socio-demographic, sunlight exposure, physical activity measurement and dietary analysis
Questionnaire inquiries on socio-demographic data were completed by participants at the beginning of the study. The researcher asked sunlight exposures and physical activity at the beginning and the end of the study. Duration of outdoor activity and sunlight avoidance histories such as usage of umbrella and sunscreen were asked. Participants were also inquired about wearing long or short sleeves clothes, short or long pants and the traditional Islamic veil. This questionnaire was obtained from a previous study, which investigated the validity and reliability of the questionnaire [29]. The reliability of the questionnaire was calculated as a pilot and through (Test re Test) for 20 students. The coefficient of stability was obtained as 0.85 [29]. The validity of the questionnaire was determined by the opinion of ten faculty members of the Kurdistan University of Medical Sciences.
The short form of the International Physical Activity Questionnaire (IPAQ) was used to assess physical activity patterns during the previous week (http://www.ipaq.ki.se 2017). The questionnaire consists of seven questions to evaluate the intensity of activities. Scores less than 600 metabolic equivalents (MET) minutes per week show low activity, scores expressed in 600–3000 MET minutes per week show moderate physical activity, and scores more than 3000 MET minutes per week represent intense physical activity.
For the assessment of dietary intakes, two 24-hour food recalls were completed at the baseline and end of the study. One of the 24 -hour recalls was filled for one of the working days of a week and another one on holidays. The Nutritionist IV software version 4.1 (First Databank Division, the Hearst Corporation, and San Bruno, CA) was used to estimate dietary intake of nutrients.
Anthropometry and body composition assessment
Height was measured using a fixed stature meter (Model No.26 SM). The BMI was calculated as weight in kilograms divided by the square of the height in meters. Bioelectrical Impedance Analysis (BIA) was performed using In Body model 270 as a body composition analyzer (In Body Co., Ltd. Seoul, Korea), to assess weight, FM (kg), FFM (kg), skeletal muscle mass (kg), BFP (%), FMI as FM (kilogram) divided by the square of height (meter) [30], and RFMI for women calculated as 76 − (20 height × WC in meters) [31]. All measurements were done at 8–10 a.m.
Biochemical assay
After an 8-h overnight fasting, the registered staff nurses took the venous blood samples between 9 and 10 a.m. Then blood samples were centrifuged at 3000 rpm for 10 min at 4 °C to obtain the serum and were stored at − 21 °C until biochemical analyses. Serum was used for analyzing calcium, phosphorus, iPTH, and 25(OH) D concentrations at the baseline and end of the study. Assay performance was measured using the kit, and standard laboratory procedures, and performance were within acceptable limits. Calcium and phosphorus concentrations were measured by colorimetric enzymatic test (Pars Azmun Co., Tehran, Iran) by photometric UV test BILT1500. 25(OH) D and iPTH were measured by Enzyme-linked Immune Sorbent Assay (ELISA) (Monobind, Inc. Lake Forest, CA (92630), the USA) and (Biomerica, Inc. Irvine, CA (92614) USA), respectively. The inter- and intra-assay coefficients of variation (CV) for calcium, phosphorus, 25(OH) D, and iPTH were 1.04 and 2.01%, 1.61 and 2.22%, 3.87 and 4.55%, 4.5, and 3.9%, respectively.
Statistical analyses
All statistical analyses were performed by SPSS version 24 (SPSS Inc., Chicago, IL, USA). Analyses were done based on ITT analyses. The ITT population consisted of all the enrolled and randomized participants. In the ITT method, information on the baseline was considered a covariate. Multiple imputation methods were used to impute missing values. The missing data were imputed using a linear regression imputation method. The BMI and serum 25(OH) D concentrations in each study group were used for multiple imputations. The normal distribution of variables was tested and confirmed by Kolmogorov–Smirnov test. The baseline measurements and dietary intakes of participants in the three groups were compared using one-way analysis of variance (ANOVA) test for quantitative variables with normal distribution, the Kruskal Willis test for non-parametric, and the Chi-square test for qualitative variables. The differences between before and after values of body composition measures were computed; then, the independent t-student test was used to determine the relationship between the mean changes of body composition measures and serum 25(OH) D status.
Two-way repeated measure ANOVA (mixed ANOVA) with Bonferroni correction was applied to assess the time effect and time-by-treatment (two doses of vitamin D or placebo) interaction effect on all outcome measures. The models were adjusted for saturated fatty acids (SFA) intake that was different between treatment groups at the baseline with a p-value < 0.05. Since weight loss during lactation might complicate the findings relating to vitamin D supplementation, the body composition variables (except for weight and body mass index) were also adjusted for mean change of weight. P < 0.05 was considered statistically significant.