This open-label randomized controlled trial was conducted at Chakri Naruebodindra Medical Institute, Faculty of Medicine Ramathibodi Hospital, Mahidol University, Samut Prakan, Thailand from July 2019 to October 2020. The study protocol was approved by the Ramathibodi Hospital Institutional Review Board (ID 10-61-58). The study was pre-registered in the Thai Clinical Trials Registry (TCTR20190622001) on 22/06/2019.
Study population and randomization
The study participants were recruited from healthy full-term infants and their mothers who attended the well-baby clinic at Chakri Naruebodindra Medical Institute for a routine 2-month infant checkup and immunization. To be eligible, the infants were required to be 6 to 12 weeks old when they entered the study. Only mothers who performed exclusive breastfeeding and had an intention to continue exclusive breastfeeding until the infants were 6 months old were approached for their consent to participate in the trial. The exclusion criteria were premature infants with a gestational age of <37 weeks at birth, infants with a congenital anomaly, introduction of formula milk and/or complementary foods before 6 months of age, and participant withdrawal. After the mothers had provided informed consent, the infants were randomized into two groups: the control group and the intervention group. Randomization was performed using opaque, sealed, sequentially numbered envelopes opened after informed consent. Each envelope contained a computer-generated block of four intervention order randomization assignments. An enrollment log was kept to ensure all envelopes were accounted for and used in the correct order.
Intervention
Routine health supervision and immunization were provided to both groups of participants under individualized physician discretion. Infants in the intervention group were given 400 mg/day of vitamin D supplementation (400 mg/day) in the form of a daily 1-mL multivitamin drop (composition per mL: vitamin A, 2000 IU; vitamin B1, 2 mg; vitamin B2, 2 mg; vitamin B6, 1.8 mg; vitamin B12, 5 mcg; vitamin C, 40 mg; vitamin D3, 400 IU; nicotinamide, 15 mg; dexpanthenol) (Munti-Vim drops; B.L. Hua & Co., Ltd., Bangkok, Thailand). The multivitamin drop was used in the study because there was no commercially available infant vitamin D-only preparation in Thailand at the time of the study. No placebo was given to the infants in the control group.
At 4 months of age, follow-up appointments were made for all infants to ensure compliance with the study protocol and to refill the vitamin D supplementation in the intervention group. At 6 months of age, follow-up appointments were made for all infants and their mothers to conclude the study and collect blood samples. Both appointments at 4 and 6 months of age were performed in accordance with the infants' routine checkup and immunization schedule visits.
Data collection
The infants' demographic data (place of birth, season at birth, gestational age, and sex) and anthropometric data (weight, length, and head circumference) at birth and at the 2-month visit were collected at the time of study enrollment. At the 4- and 6-month visits, the infants' anthropometric measurements were repeated. At the 6-month visit, the following laboratory data were obtained for each infant: serum concentrations of vitamin D [25-hydroxyvitamin D2 (25OHD2), 25-hydroxyvitamin D3 (25OHD3), and total 25-hydroxyvitamin (25OHD)], intact parathyroid hormone (iPTH), calcium, phosphorous, and alkaline phosphatase (ALP). Maternal age and serum vitamin D concentrations were also collected at the 6-month visit.
Biochemical analyses
Serum vitamin D (25OHD2, 25OHD3, and 25OHD) was analyzed using liquid chromatography-tandem mass spectrometry assays (6490 Triple Quad LC/MS; Agilent Technologies, Santa Clara, CA, USA). The iPTH concentration was measured using an electrochemiluminescence immunoassay (cobas e601; Roche Diagnostics, Basel, Switzerland). The plasma concentrations of calcium, phosphorous, and ALP were measured using an automated analyzer (cobas c501; Roche Diagnostics). Vitamin D deficiency was defined as a 25OHD concentration of <12 ng/mL, insufficiency as 25OHD of 12 to 20 ng/mL, and sufficiency as 25OHD of >20 ng/mL (8).
Sample size estimation
A power calculation was used to calculate the sample size needed to evaluate the primary outcome (infants' 25OHD concentrations). A pilot study of 34 full-term, exclusively breastfed infants aged 1 to 6 months in our institution (Ramathibodi Hospital, Bangkok) showed that the infants' mean 25OHD concentration was 13.6 ± 7.7 ng/mL (unpublished data). We hypothesized that vitamin D supplementation would increase the 25OHD level by at least 50% in the intervention group or to a mean 25OHD concentration of up to 20.4 ng/mL (sufficiency level) with a standard deviation similar to that in the control group. According to this assumption, 40 infants in each group were required to detect a post-intervention difference in the 25OHD concentration with an alpha of 0.01, 90% power, two-sided. To allow for attrition, an additional 20% or 10 subjects were added to each group. Therefore, 50 was selected as the total number of infants in each group.
Statistical analysis
A univariate analysis was performed to identify significant differences between the groups. Student's t-test was used for continuous variables, and the results are presented as mean ± standard deviation. The Mann–Whitney U test was used if the distribution was not normal, and the results are presented as median (interquartile range). Pearson's chi-square test or Fisher's exact test was used for categorical variables, and the results are presented as total number (%). Multivariate linear regression analysis was used to determine the factors associated with infants' 25OHD concentration. To demonstrate the differences in serum bone markers between infants with and without vitamin D sufficiency, univariate analysis was applied using an infant serum 25OHD concentration of <20 ng/mL as the independent variable. A p-value of <0.05 was considered statistically significant. Stata Statistical Software version 15.1 (StataCorp LLC, College Station, TX, USA) was used for all statistical analyses.