European Journal of Clinical Nutrition (2015) 69, 367–372 © 2015 Macmillan Publishers Limited All rights reserved 0954-3007/15 www.nature.com/ejcn

ORIGINAL ARTICLE

The effect of monthly 50 000 IU or 100 000 IU vitamin D supplements on vitamin D status in premenopausal Middle Eastern women living in Auckland H Mazahery1, W Stonehouse2 and PR von Hurst1 BACKGROUND/OBJECTIVES: Middle Eastern female immigrants are at an increased risk of vitamin D deficiency and their response to prescribed vitamin D dosages may not be adequate and affected by other factors. The objectives were to determine vitamin D deficiency and its determinants in Middle Eastern women living in Auckland, New Zealand (Part-I), and to determine serum 25hydroxyvitamin D (serum-25(OH)D) response to two prescribed vitamin D dosages (Part-II) in this population. PARTICIPANTS/METHODS: Women aged ⩾ 20 (n = 43) participated in a cross-sectional pilot study during winter (Part-I). In Part-II, women aged 20–50 years (n = 62) participated in a randomised, double-blind placebo-controlled trial consuming monthly either 50 000, 100 000 IU vitamin D3 or placebo for 6 months (winter to summer). RESULTS: All women in Part-I and 60% women in Part-II had serum-25(OH)D o50 nmol/l. Serum-25(OH)D was higher in prescribed vitamin D users than nonusers (P = 0.001) and in Iranians than Arab women (P = 0.001; Part-I). Mean (s.d.) serum-25(OH)D increased in all groups (time effect, P o0.001) and differed between groups (time × dosage interaction, P o 0.001; 50 000 IU: from 44.0 ± 16.0 to 70.0 ± 15.0 nmol/l; 100 000 IU: 48.0 ± 11.0 to 82.0 ± 17.0 nmol/l; placebo: 45.0 ± 18.0 to 54.0 ± 18.0 nmol/l). Only 32% and 67% achieved serum-25(OH)D ⩾ 75 nmol/l with 50 000 and 100 000 IU/month, respectively. Predictors of 6-month change in serum-25 (OH)D were dose (B-coefficient ± s.e.; 14.1 ± 2.4, P o0.001), baseline serum-25(OH)D (−0.6 ± 0.1, P o 0.001) and body fat percentage (−0.7 ± 0.3, P = 0.01). CONCLUSIONS: Vitamin D deficiency/insufficiency is highly prevalent in this population. Monthly 100 000 IU vitamin D for 6 months is more effective than 50 000 IU in achieving serum-25(OH)D ⩾ 75 nmol/l; however, a third of women still did not achieve these levels. European Journal of Clinical Nutrition (2015) 69, 367–372; doi:10.1038/ejcn.2014.264; published online 10 December 2014

INTRODUCTION Vitamin D has a well-established role in bone metabolism, including calcium and phosphorus homeostasis, and its deficiency is associated with several adverse health consequences. 25-hydroxyvitamin D (serum-25(OH)D) concentration has been associated with musculoskeletal1 and non-musculoskeletal health outcomes such as cancer, cardiovascular diseases and autoimmune diseases.2 Concurrently, vitamin D deficiency/insufficiency is re-emerging as a worldwide public health issue.3 Evidence shows that Middle Eastern women living in their home country4–6 or in other countries as immigrants7–9 have low serum-25(OH)D concentrations. Comparable to Turkish women, 45.0% of Iranian women living in Oslo, Norway, had serum levels o20 nmol/l.7 Although no data is available about this population in New Zealand, anecdotal evidence from health service providers suggests that vitamin D deficiency is highly prevalent.10 These women are at a greater risk of vitamin D deficiency owing to many lifestyle risk factors such as conservative clothing style. Approximately 90–100% of vitamin D requirement comes from exposure of the skin to UVB radiation,11 and if skin synthesis of vitamin D is either insufficient or absent, vitamin D can be obtained from the increased ingestion of vitamin D rich foods.

Because dietary sources of vitamin D are very limited,12 and the fortification of foods with vitamin D is neither mandatory nor common, in New Zealand13 supplements are considered to be the best alternative. Daily doses of 400–800 IU are recommended age dependently, but in some populations with vitamin D deficiency/ insufficiency up to 2000 IU vitamin D/day is required,14–16 and dosages up to 10 000I U/day have been considered to be safe.15 In New Zealand, pharmaceutical preparations containing a monthly dosage of 50 000 IU vitamin D3 might be considered by family doctors for those populations that are at risk a of vitamin D deficiency.12 However, the adequacy of this dose in some populations, especially in Middle Eastern women, remains doubtful.17 Talwar et al.18 predicted that patients with serum-25 (OH)D concentrations 445 and o45 nmol/l needed 2800 and 4000 IU/d vitamin D, respectively, to achieve concentrations 475 nmol/l, indicating that dosages 450 000 IU/month might be needed to raise serum-25(OH)D concentrations of Middle Eastern women to at least 75 nmol/l. Individual response to vitamin D supplementation is predicted by a variety of factors such as body composition,19 basal 25(OH)D concentration,20 dose,21 type of vitamin D,22 season,21 oestrogen use23 and calcium intake.19 Zittermann et al.,24 in their systematic review, showed that ~ 34% of the variation in circulating 25(OH)D

1 Institute of Food, Nutrition and Human Health, Massey University, Auckland, New Zealand and 2Commonwealth Scientific Industrial Research Organisation, Food and Nutrition Flagship, Adelaide, South Australia, Australia. Correspondence: Dr PR von Hurst, Institute of Food, Nutrition and Human Health, College of Health, Massey University, Albany Campus, Private Bag 102904, North Shore City, Auckland 0745, New Zealand. E-mail: [email protected] Received 12 September 2014; revised 24 October 2014; accepted 12 November 2014; published online 10 December 2014

Vitamin D dose response trial in Middle Eastern women H Mazahery et al

368

MATERIALS AND METHODS Study protocol This study was divided into two parts; Part-I was designed as an epidemiological cross-sectional pilot study commencing in July–August 2012 (winter), and Part-II as a 6-month randomised, double-blind, placebocontrolled dose–response trial commencing in late June 2013 (winter) and concluding in late January 2014 (summer).

Participants A total of 43 (Part-I) and 62 (Part-II) healthy adult Middle Eastern women (either the participant, or both parents were born in Middle Eastern countries) were recruited from the community in Auckland, New Zealand. Women were included if they were ⩾ 20 years, and were excluded if they had: (1) major systemic illnesses; (2) diseases such as digestive disorders, kidney or liver diseases or were taking medications affecting vitamin D metabolism or absorption such as glucocorticoids; (3) bleeding disorders or were taking blood thinning medication. In Part-II, women were also excluded if they were in postmenopausal stage or had treatment with vitamin D within the last 6 months (other than multivitamins).

Part-I n=43 Winter 2012

Middle Eastern Women’s Health Study (MEWH Study)

was explained by body weight, followed by type of supplement (9.8%), age (3.7%), calcium intake (2.4%) and basal 25(OH)D concentrations (1.9%). However, the significance of these factors in response to vitamin D supplements in Middle Eastern women has not been determined. Accordingly, the aims of the present study were: (1) Part-I: to identify the likelihood of vitamin D deficiency and its determinants in Middle Eastern women living in Auckland, New Zealand; and (2) Part-II: to investigate the adequacy of different available and recommended dosages of vitamin D and to identify predictors of response to vitamin D supplements in this population.

Part-II Visit One – Baseline Measures: n=62 • Consent form Winter 2013 to • Questionnaires: (1) participant details and demographic questionnaire, (2) medical summer 2014 history questionnaire, (3) Fitzpatrick skin colour scale questionnaire and (4) physical activity level questionnaire • Bioelectrical Impedance Analysis (weight, body fat percentage and body mass index) • Anthropometry (height) • Blood test [serum-25(OH)D and serum calcium] • Instructions on completion of food diary Randomisation to Two Intervention and One Placebo Arm Randomisation Performed by Third Party Not Associated With the Study Visit Two – 3-months Measures: • Blood test [serum-25(OH)D and serum calcium] • Change of lifestyle questionnaire Visit Three – 6-month (Endpoint) Measures: • Bioelectrical Impedance Analysis(weight, body fat percentage and body mass index) • Blood test [serum-25(OH)D and serum calcium] • Change of lifestyle questionnaire

Data collection All participants’ data and non-fasting venous blood samples were collected during participants’ visit to the Massey University Human Nutrition Research Unit. For a presentation of the study procedure refer to Figure 1. The use of any vitamin D-containing supplements and other medications was recorded in both the studies. Sun-exposure questionnaire was adopted from von Hurst et al.25 Skin colour and physical activity level were determined using Fitzpatrick skin colour questionnaire26 and New Zealand Physical Activity Questionnaire Short Form (NZPAQ-SF).27 Body fat percentage was measured using bioelectrical impedance analysis (BIA, Inbody 230, Seoul, Korea). To assess the calcium intake, participants were given recorded, verbal and written instructions on how to collect diet records of three weekdays and one weekend day (4-day food diary) within the next 14 days. Average dietary calcium intake was assessed over four reported days using FoodWorks Professional Edition 7 (Xryis Software, Brisbane, QLD, Australia, 2012). To record any changes in lifestyle including prescribed medicine and supplement use between the two visits, participants completed the change of lifestyle questionnaire. The study was registered at http://www.anzctr.org.au as ACTRN12613000 383763. The ethical approval of Part-I was granted by the Massey University Human Ethics Committee (Southern A), Reference No. 12/16, and of Part-II by the Health and Disability Ethics Committee: Reference No. 13/STH/40. All volunteers were provided with an information sheet explaining the study protocol in detail and signed an informed consent form.

Treatment At the baseline visit, participants were randomly assigned to one of two monthly vitamin doses (50 000 IU+placebo, and 100 000 IU (2 × 50 000 IU)) or placebo (2 × placebos) for 6 months. Study tablets, Cal.D.Forte, contained 50 000 IU vitamin D3 and were provided by PSM HealthCare Ltd (trading as API Consumer Brands, Auckland, New Zealand). The placebo was identical to vitamin D tablets in appearance and contained no active ingredients. Participants were reminded to take their monthly ‘tablets’ by monthly text message. Randomisation was generated using the Website Randomisation.com (http://www.randomization.com). The researchers and participants were blinded and became unblinded at the end of study following the data analysis. Compliance was calculated using European Journal of Clinical Nutrition (2015) 367 – 372

Visit One: • Consent form • Questionnaires: (1) participant details, demographic and lifestyle questionnaire, (2) medical history questionnaire and (3) sunexposure questionnaire • Anthropometry (height, weight and body mass index) • Blood test [serum-25(OH)D]

Figure 1.

A presentation of study procedure.

cumulative pill counts at the end of the study, and adherence was measured as a percentage; (number of pills supplied minus number of pills returned)/number of pills supplied × 100.

Biochemical analyses Blood samples were drawn 10–14 days after taking the final supplement to allow participants to return to a steady state. Serum was used for the analysis of 25(OH)D and calcium. The blood was protected from light and allowed to clot for 30 min and centrifuged for 10 min at 2000 r.p.m. at 4 °C within 2 h of sampling. Aliquots of serum were collected in Eppendorf tubes and stored at − 80 °C until the analysis of serum-25(OH)D at the end of the study in one batch for each study. Serum-25(OH)D was measured using ADVIA Centaur Vitamin D Total assay, Siemens Healthcare Diagnostics Inc (IL, USA) with a coefficient of variation of 4.2–11.9%. Calcium and albumin were measured by a Flex reagent cartridge system by Siemens Health care Diagnostics (Australia and New Zealand) with a coefficient of variation of 2.2–3.0%. We used both cutoffs of ⩾ 5028 and ⩾ 75 nmol/l29,30 to define vitamin D sufficiency. Hypercalaemia was defined as a serum calcium level of 42.63 mmol/l.31

Statistical analysis Sample size was calculated using the uniform withdrawal rate of 10% and the following formula: N = 2α2K/(μ2-μ1)2 (N, sample size/group; α, s.d. serum-25(OH)D (25 nmol/l);32–34 K, constant (7.9); μ2-μ1, clinically meaningful difference in means (22 nmol/l)17,35). Descriptive statistics were used for population characteristics. Normality was explored using Kolmogorov–Smirnov or Shapiro–Wilk tests. Nonnormal variables were log transformed and checked for normality again. Normal and non-normal variables were reported as mean ± s.d. and © 2015 Macmillan Publishers Limited

Vitamin D dose response trial in Middle Eastern women H Mazahery et al

369 median (25th–75th percentiles), respectively. Transformed variables were back transformed into mean (95% confidence interval) from summary statistics. Between-groups differences were examined using χ2-test, Student’s t-test, Kruskal–Wallis tests and one-way independent analysis of variance. A Bonferroni correction was applied for multiple comparisons (0.05/number of comparisons). Effect size (r) was calculated using Z/√n and a value of 0.1, 0.3 and 0.5 represents small, medium and large size effect, respectively.36 The primary outcome variable was serum-25(OH)D concentration. In Part-II, repeated measures mixed models were used with maximum likelihood estimation in an intention-to-treat analysis of each outcome measure using all available data. The data were checked for outliers. One outlier with baseline serum-25(OH)D concentration of 189 nmol/l was excluded from the end point analysis, but was included in the analyses determining the baseline characteristics. Adherence to protocol was defined as ⩾ 80% and the data were tested per protocol. There was no difference in outcomes between intention-to-treat and per-protocol analyses as such the results of intention-to-treat analyses were reported here. Mixed model design (variance components structure) was used to explore the dose–response curve for serum-25(OH)D. Dose as continuous and time as categorical (baseline, 3 and 6 months) were included as fixed effects, and participant was included as a random effect to account for the repeated measures within individuals. Multiple linear regression was used: (1) to determine predictors of log transformed serum-25(OH)D from independent variables in Part-I. Variables included were ethnicity (Persian/Arab), prescribed vitamin D-use (yes/no), time spent in private outdoor area ( o2/ ⩾ 2 h/week), and years of education; and (2) to investigate factors influencing the 6-month serum-25 (OH)D change in Part-II. Variables included were dose, baseline values, body fat percentage and calcium intake. All assumptions for regression analysis were met (multicolinearity, independence of variables, normally distributed residuals, and the same variance of the dependent variables). Statistical analysis was performed using SPSS software (version 18 and 21, SPSS Inc, Chicago IL, USA in Part-I and Part-II, accordingly). A P-value o0.05 was considered significant.

RESULTS Part-I Data were collected from 43 participants. The characteristics of the study population are presented in Table 1. Determinants of serum-25(OH)D concentrations were ethnicity (B-coefficient ± s.e.: − 0.6 ± 0.2, P = 0.003) and prescribed vitamin D-use (0.6 ± 0.2, P = 0.005). The median serum-25(OH)D concentration was higher in prescribed vitamin D users (31.0 (24.0–41.0) nmol/l) than nonusers (16.0 (10.0–21.0) nmol/l), U = 37.0, n = 43, Po 0.001, r = − 0.5, and in Persians (20.0 (14.0–31.0) nmol/l) than Arab women (9.0 (9.0–11.0) nmol/l), U = 20.0, n = 43, P = 0.001, r = –0.5). Part-II Sixty-one participants completed the study out of the 62 initial recruits (one participant moved to overseas). The median followup was 163 days (range, 157–197 days). The baseline characteristics of the study population are presented in Table 2. Half of the participants had tertiary education, and the majority of participants were students or employed (74.1%), 61.3% reported o 2 h of moderate physical activity per week and 85.0% Fitzpatrick skin type V and VI. There were four women taking multivitamins containing both supplemental calcium and vitamin D, and 11 taking multivitamins containing vitamin D (100–800 IU vitamin D/ tablet). The mean adherence was 97.8%. Among all the study groups, there were no significant changes in the levels of serum calcium, and there were no reports of serum calcium (corrected for albumin) ⩾ 2.7 mmol/l (hypercalcaemia) and serum-25(OH)D 4225 nmol/l (hypervitaminosis D). In the mixed model analyses, we found a significant effect of time of the follow-up, F (2, 74.0) = 11.3, Po0.001, dose, F (1, 169.5) = 46.9, © 2015 Macmillan Publishers Limited

Table 1.

Baseline characteristics of study population in Part-I

Parameters

Study population (n = 43)

Serum-25(OH), nmol/la Serum-25(OH)D, range, nmol/l

18.2 (15.5–21.3) o9.0–46.0

Vitamin D status, n (%) o12.5 12–25 25–50 ⩾ 50

12 (28.0) 20 (46.5) 11 (25.6) 0 43.5 ± 11.8 26.5 (23.2–29.2)

Age, years** BMI, kg/m2 b Ethnicity, n (%) Persian Arab

37 (86.0) 6 (14.0)

Smokers, n (%) Clothing style, Islamic, n (%) Sun avoidance behaviour, n (%)c Using vitamin D-containing multivitamins, n (%) Using vitamin D prescribed by family doctor

1 (2.3) 10 (23.3) 41 (95.4) 16 (37.0) 9 (21.0)

Abbreviations: BMI, body mass index; 25(OH)D, 25-hydroxyvitamin D. **Values are reported as mean ± s.d. aLog-transformed values are reported as means (95% confidence interval) calculated by back transformation from summary statistics. bValues are reported as median (25th–75th percentile). c Women protected their skin from sun-exposure sometimes or always.

P o0.001, and the interaction between time of follow-up and dose, F (2, 73.6) = 11.7, P o 0.001 (Figure 2). The mean serum-25(OH)D concentrations increased from a baseline of 44.0 ± 16.0 and 48.0 ± 11.0 nmol/l to 70.0 ± 15.0 and 82.0 ± 17.0 nmol/l at 6 months in the 50 000 and 100 000IU groups, respectively (P o 0.001 for both the treatment groups). The mean serum-25(OH)D concentration in the placebo group increased from 45.0 ± 18.0 nmol/l at baseline to 54.0 ± 18.0nmol/l at 6 months (P o 0.01). The proportion of participants reaching serum-25(OH)D above cutoff points of ⩾ 50 and ⩾ 75 nmol/l is shown in Table 3. At a 6month visit, all participants in the 50 000 and 1 00 000 IU groups, except for one in the 50 000 IU group, had serum-25(OH)D concentrations ⩾ 50 nmol/l. Serum-25(OH)D concentration increased to ⩾ 75 in 31.6% and 66.7% of participants in the 50 000 and 1 00 000 IU groups, respectively, at the 6-month visit. Determinants of the 6-month change in serum-25(OH)D included baseline serum-25(OH)D concentration, body fat percentage, dose and dressing code (Table 4). These variables accounted for 57.6% of the variance in the change in serum-25(OH)D concentrations. Larger dose, lower baseline serum-25(OH)D concentration and lower body fat percentage were significantly associated with larger 6-month change in the serum-25(OH)D concentration. DISCUSSION Vitamin D deficiency/insufficiency in both studies was highly prevalent in Middle Eastern women living in Auckland, although mean serum-25(OH)D concentration was much higher in Part-II. The Middle Eastern community in New Zealand is small and the information spread rapidly by word-of-mouth. Part-I of the study had increased the awareness and promoted actions against vitamin D deficiency in this community, (likely) resulting in the higher mean serum-25(OH)D concentration in Part-II. Furthermore, European Journal of Clinical Nutrition (2015) 367 – 372

Vitamin D dose response trial in Middle Eastern women H Mazahery et al

370 Table 2.

Baseline characteristics of study population in Part-II

Parameters

Study group

Serum-25(OH)D, nmol/lb Vitamin D status, n (%) o 25 25–50 50–75 ⩾ 75

All participants (n = 62)

Placebo (n = 21)

50 000 IU (n = 20)

100 000 IU (n = 21)

P-valuea

43.8 (39.6–48.9)

45.2 (35.5–56.8)

40.9 (33.8–49.9)

46.1 (40.4–52.5)

0.65

2 (9.5) 12 (57.1) 5 (23.8) 2 (9.5)

3 (15.0) 9 (45.0) 8 (40.0) 0

1 (4.8) 10 (47.6) 10 (47.6) 0

— — — —

2.25 (2.19–2.28) 36.7 ± 9.0 24.7 (22.0–28.1) 36.9 ± 7.6

2.23 (2.19–2.27) 37.1 ± 8.5 24.3 (22.0–27.6) 37.0 ± 6.5

2.27 (2.22–2.31) 35.1 ± 10.1 24.5 (22.1–27.4) 36.8 ± 8.1

2.23 (2.19–2.29) 36.1 ± 8.6 24.9 (21.7–28.5) 37.0 ± 8.3

— 0.78 0.92 1.00

48 (77.4) 14 (22.6)

16 (76.2) 5 (23.8)

15 (75.0) 5 (25.0)

17 (81.0) 4 (19.0)

0.89 0.89

19 (30.6) 11 (17.7) 7 (11.3) 529 (469–596)

6 (28.6) 3 (14.3) 1 (4.8) 488 (388–614)

7 (35.0) 4 (20.0) 3 (15.0) 539 (433–679)

6 (28.6) 4 (19.0) 3 (14.3) 561 (446–706)

0.88 0.88 — 0.65

c

6 31 23 2

Serum calcium, mmol/ld Age, yearse BMI, kg/m2 d Body fat (%)e Ethnicity, n (%) Persian Arab Islamic clothing style, n (%) Smokers, n (%) Oestrogen users, n (%) Calcium intake, mg/df

(9.7) (50.0) (37.1) (3.2)

Abbreviations: BMI, body mass index; 25(OH)D, 25-hydroxyvitamin D. aP-values o 0.05 are considered significant; for normally distributed data and nonnormally distributed data, one-way analysis of variance and Kruskal–Wallis tests, respectively, were performed; for categorical variables, χ2-test was performed. P-value for some variables has not been reported because the χ2-test was not performed owing to the violation of χ2-test assumptions. bAll participants were included and log values are reported as means (95% confidence interval) calculated by back transforming summary statistics. cOne participant had serum-25 (OH)D concentration of 189 nmol/l, and there were no significant differences among the groups in the prevalence of serum-25(OH)D status based on the two cutoffs of ⩾ 50 nmol/l and ⩾ 75 nmol/l. dValues are reported as medians (25th–75th percentile). eValues are reported as means ± s.d. fDerived from 4-day food diary (diet+supplements); 54 volunteers (18 per treatment group) returned their food diaries. Values are reported as means (95% confidence interval) calculated by back transforming summary statistics.

Table 3. Percentage of participants with serum-25(OH)D concentrations ⩾ 75 nmol/l and ⩾ 50nmol/l at baseline, after 3 and 6 months in the three groups

Mean serum-25(OH)D concentration (nmol/L)

100 100000 IU*

80

Variablea

Study group

50000 IU*

60

Placebo*

40 20 0

Baseline

3 months Follow up**

6 months

Error Bars: 95% CI

Figure 2. The dose–response curve to vitamin D supplementation. The pattern of change in serum-25(OH)D concentrations over the study period (baseline, 3 months and 6 months) within each of the treatment groups have been presented. Reference lines at 50 (dotted line) and 75 nmol/l (hyphened line) were added for clarification. *Significant effect of dose of treatment F (1, 169.5) = 46.9, P o0.001. **Significant effect of time of follow-up F (2, 74.0) = 11.3, Po 0.001. We also found a significant interaction between time of follow-up and the dose of treatment, F (2, 73.6) = 11.7, Po 0.001.

in Part-I, despite the use of prescribed vitamin D supplements by 30% of the study population, all women had serum-25(OH)D concentrations o 50 nmol/l. These findings raised two main questions; (1) Why is vitamin D deficiency/insufficiency highly prevalent in this population? and (2) Why did serum-25(OH)D European Journal of Clinical Nutrition (2015) 367 – 372

Placebo (n = 21)

50 000 IU (n = 20)b

100 000 IU (n = 21)

P-valuec

Serum-25(OH)D Baseline 3 months 6 months

⩾ 75 nmol/l 9.5 (2) 4.8 (1) 14.3 (3)

0.0 (0) 30.0 (6) 31.6 (6)

0.0 (0) 47.6 (10) 66.7 (14)

— 0.007 0.002

Serum-25(OH)D Baseline 3 months 6 months

⩾ 50 nmol/l 33.3 (7) 33.3 (7) 61.9 (13)

40.0 (8) 95.0 (19) 94.7 (18)

47.6 (10) 100.0 (21) 100.0 (21)

0.64 o0.001 —

Abbreviation: 25(OH)D, 25-hydroxyvitamin D. aValues are reported as percentage (number). bOne participant was lost to the follow-up at the 6-month visit. cχ2- test was performed because the variables were categorical. The P-value for some variables has not been reported because of the violation of χ2- test assumptions.

concentrations of prescribed vitamin D users not reach the sufficient levels and what is the adequate dose? Ethnicity is a recognised determinant of vitamin D status,28 and Middle Eastern populations compare unfavourably with Caucasians.28 A study from New Zealand37 and several other studies from European countries7–9 reported that vitamin D deficiency/insufficiency is widespread in this ethnic group. Furthermore, the baseline blood samples (both studies) were collected in late June-August (winter). Season is a well-established determinant of vitamin D status.21,38 In latitudes above 35° (Auckland is at a latitude of 36.5°), synthesis of 25(OH)D decreases © 2015 Macmillan Publishers Limited

Vitamin D dose response trial in Middle Eastern women H Mazahery et al

371 Table 4.

a

Predictors of change in serum-25(OH)D concentrations over study period (6 months)

Change in serum-25(OH)D model Model Intercept Dosed Baseline serum-25(OH)Db Body fat (%)e Dressing codef

Coefficient (B)

s.e.(B)

95% CI B

35.8 14.1 − 0.6 − 0.7 7.3

13.2 2.4 0.1 0.3 4.3

9.3, 62.3 9.3, 19.0 − 0.8, − 0.3 − 1.2, − 0.2 − 1.4, 15.9

Standardised B

0.6 − 0.4 − 0.3 0.2

Rb

P-value

0.576c

o0.001 o0.001 o0.001 o0.01 0.1

Abbreviations: CI, confidence interval; serum-25(OH)D, 25-hydroxyvitamin D. Regression equation: change in serum-25(OH)D after 6 months (nmol/l) = 35.8 +14.1 × dose–0.6 × baseline serum-25(OH)D (nmol/l)–0.7 × body fat (%) +7.3 × dressing code. aBackward stepwise technique. The following variables were included in the model: dose, baseline serum-25(OH)D concentration (nmol/l), body fat (%), dressing code and dietary calcium intake (diet+supplements) (mg). Calcium intake was removed from this model during stepwise analysis (P = 0.98). Valid number of participants: 53. bFor each decrease of one unit in baseline serum-25(OH)D concentration, the change in serum-25(OH)D is expected to increase by 0.6 nmol/l. cF (4, 52) = 16.3, Po0.001; ** F (5, 52) = 12.8, Po 0.001. dFor each increase of one unit in vitamin D of 50000IU, serum-25(OH)D is expected to increase by 14.1 nmol/l. eFor each decrease of one unit in body fat (%), the change in serum-25(OH)D is expected to increase by 0.7 nmol/l. fDressing code was coded as 1 = Islamic dressing code and 2 = non-Islamic dressing code; having non-Islamic dressing code was associated with an increase of 7.3nmol/l in change in serum-25(OH)D concentration after 6 months.

in winter.1 The seasonality of vitamin D status was shown by the increase in mean serum-25(OH)D concentration in the placebo group at the final visit (approaching summer). In their studies conducted in New Zealand, Logan et al.21 and Rockell et al.38 reported a drop of 44 (from summer to winter) and 30 nmol/l (from summer to spring) in the geometric mean of placebo, respectively. The lower seasonal effect in our study population could be attributed to the more conservative clothing style of Middle Eastern women, having more sun avoidance behaviour (95.4%) or darker skin colour (Fitzpatrick skin type V and VI) than the study populations in the aforementioned studies.21,38 In Part-I, prescribed vitamin D use was associated with higher serum-25(OH)D concentrations, but it did not raise serum-25(OH)D concentrations to ⩾ 50 nmol/l. Because crude information on the consumption of this supplement was available, we could not exclude the potential measurement errors. Part-II was then designed to elucidate these findings. In Part-II, serum-25(OH)D concentration reached a plateau at 3 months, a finding confirmed by others.18,39 Participants assigned to the 50 000 and 1 00 000 IU groups exhibited significantly higher serum-25(OH)D concentrations than the placebo group. Both the doses were adequate in raising summer serum-25(OH)D concentrations ⩾ 50 nmol/l in all women except for one in the 50 000 IU group, a finding apparently contradicting our Part-I observations. The mean serum-25(OH)D concentration in the 50 000 IU group was lower relative to the 1 00 000 IU group, and the mean increased significantly (+10 nmol/l) in our placebo group. Hence, we cannot rule out the effect of season on serum-25(OH)D concentrations in our supplemented groups. By inference, 10 nmol/l of the 6-month change in the supplemented groups could be attributed to the season, meaning that the mean of our 50 000 and 1 00 000 IU groups may drop to 60 and 72 nmol/l, respectively, in winter. As such serum-25(OH)D concentration of a large proportion of women in the 50 000IU, but not 1 00 000 IU group, may fall again below 50 nmol/l in winter. The proportion of participants with serum-25(OH)D o50 and o75 nmol/l decreased more in women receiving a monthly dose of 1 00 000 IU. Although more effective than monthly 50 000 IU, even with this dose, ~ 1/3 of the participants still failed to reach optimum levels (⩾75 nmol/l). Other studies also found that accumulative doses of 1 00 000–1 20 000 IU/month were insufficient to raise serum-25(OH)D concentrations to ⩾ 75 nmol/l.6,40 The study by Zabihiyeganeh et al.,6 showed that among Iranians (serum-25(OH)D o 75 nmol/l), 65.6% of the patients who received accumulative dose of 300 000 IU over 3 months reached serum-25(OH)D concentration ⩾ 75 nmol/l after 6 months,6 a finding comparable to ours (66.7%). So, larger doses than monthly © 2015 Macmillan Publishers Limited

1 00 000 IU are required to shift vitamin D status to ⩾ 75 nmol/l in the populations with high prevalence of vitamin D insufficiency. Baseline serum-25(OH)D concentration was the second most significant predictor of the 6-month change and explained 20.2% of the variance of serum-25(OH)D response. Other studies have also reported that larger increases are seen in those with lower baseline levels.18,20,23,35 Lower body fat percentage appeared to be one of the factors influencing a better response to oral vitamin D supplementation. The difference in serum-25(OH)D response to supplementation based on body fat percentage is consistent with the findings of Blum et al.41 and Nelson et al.23 The postulated mechanism is that, because vitamin D is a fat soluble vitamin and is stored in body fat,42,43 the larger the mass of adipose tissue the more likely vitamin D is sequestered.44 Experimental support for sequestration comes from animal43 and human45 studies; a study in Wistar rats showed that under supplementation with high-dose vitamin D, plasma-25(OH)D concentration increased rapidly until it reached a plateau.43 The plasma-25(OH)D and adipose tissue vitamin D3 accumulation occurred linearly and rapidly, and the accumulated vitamin D3 was released slowly into circulation in the condition of energy balance. Recent evidence of sequestration of vitamin D in human adipose tissue lends credence to these observations.45 Vitamin D supplementation was not associated with any adverse events in this study, and others have demonstrated safety with larger doses than those used here.46 The main strength of this study lies in its design and study population. Part-I gave insight into this relatively new migrant population and helped us to set the objectives and inclusion/ exclusion criteria for further studies. Part-II was sufficiently powered (sample size) to detect a difference in the primary outcome, and was of sufficient duration to investigate the effect of supplementation on serum-25(OH)D concentration over time. A limitation of this study is that the results may not apply to other ethnic groups, those living in other regions or countries or those with disease. In conclusion, the prevalence of vitamin D deficiency/insufficiency is high in Middle Eastern women. Although monthly intake of 1 00 000 IU vitamin D for 6 months is more effective than 50 000 IU in achieving serum-25(OH)D concentrations ⩾ 75 nmol/l, a third still did not achieve adequate levels. A better response to supplementation was observed in those, taking the larger dose, having lower baseline levels or lower body fat percentage. Further research is warranted to investigate who are the most at risk of, and the reasons for, not achieving the adequate 25(OH)D concentration with the higher dose. European Journal of Clinical Nutrition (2015) 367 – 372

Vitamin D dose response trial in Middle Eastern women H Mazahery et al

372

CONFLICT OF INTEREST The authors declare no conflict of interest.

ACKNOWLEDGEMENTS We wish to thank research project manager, Owen Mugridge, and laboratory manager, PC Tong, for their ongoing support and help The study was registered under the Trial registration no: ACTRN12613000383763.

REFERENCES 1 Holick MF. Vitamin D: A D-lightful health perspective. Nutr Rev 2008; 66: S182–S194. 2 Kulie T, Groff A, Redmer J, Hounshell J, Schrager S. Vitamin D: an evidencebased review. J Am Board Fam Med 2009; 22: 698–706. 3 Lips P. Worldwide status of vitamin D nutrition. J Steroid Biochem Mol Biol 2010; 121: 297–300. 4 Arabi A, Baddoura R, El-Rassi R, El-Hajj Fuleihan G. PTH level but not 25 (OH) vitamin D level predicts bone loss rates in the elderly. Osteoporos Int 2012; 23: 971–980. 5 Muhairi SJ, Mehairi AE, Khouri AA, Naqbi MM, Maskari FA, Kaabi J et al. Vitamin D deficiency among healthy adolescents in Al Ain, United Arab Emirates. BMC Public Health 2013; 13: 33. 6 Zabihiyeganeh M, Jahed A, Nojomi M. Treatment of hypovitaminosis D with pharmacologic doses of cholecalciferol, oral vs intramuscular; an open labeled RCT. Clin Endocrinol (Oxf) 2013; 78: 210–216. 7 Holvik K, Meyer HE, Haug E, Brunvand L. Prevalence and predictors of vitamin D deficiency in five immigrant groups living in Oslo: The Oslo immigrant health study, Norway. Eur J Clin Nutr 2005; 59: 57–63. 8 Van Der Meer IM, Karamali NS, Boeke AJP, Lips P, Middelkoop BJC, Verhoeven I et al. High prevalence of vitamin D deficiency in pregnant non-Western women in the Hague, Netherlands. Am J Clin Nutr 2006; 84: 350–353. 9 Hobbs R, Habib Z, Alromaihi D, Idi L, Parikh N, Blocki F et al. Severe Vitamin D Deficiency in Arab-American women living in Dearborn, Michigan. Endocr Pract 2009; 15: 35–40. 10 Perumal LHealth needs assessment of Middle Eastern, Latin American and African people living in the Auckland region. Auckland District Health Board: Auckland, New Zealand, 2010. 11 Holick MF. Sunlight "D"ilemma: risk of skin cancer or bone disease and muscle weakness. Lancet 2001; 357: 4–6. 12 Ministry of Health and Cancer Society of New Zealand. Consensus Statement on Vitamin D and Sun Exposure in New Zealand. In: Health Mo. eds, Ministry of Health: Wellington, New Zealand, 2012. 13 Ministry of Health, NHMRC. Nutrient Reference Values for Australia and New Zealand Including Recommended Dietary Intakes. Ministry of Health and Nutritional Health and Medical Research Council, 2006. 14 Ross AC, Manson JE, Abrams SA, Aloia JF, Brannon PM, Clinton SK et al. The 2011 report on dietary reference intakes for calcium and vitamin D from the Institute of Medicine: what clinicians need to know. J Clin Endocrinol Metab 2011; 96: 53–58. 15 Rizzoli R, Boonen S, Brandi M-L, Bruyère O, Cooper C, Kanis JA et al. Vitamin D supplementation in elderly or postmenopausal women: a 2013 update of the 2008 recommendations from the European Society for Clinical and Economic Aspects of Osteoporosis and Osteoarthritis (ESCEO). Curr Med Res Opin 2013; 29: 305–313. 16 Bischoff-Ferrari HA, Willett WC, Orav EJ, Lips P, Meunier PJ, Lyons RA et al. A pooled analysis of vitamin D dose requirements for fracture prevention. N Engl J Med 2012; 367: 40–49. 17 Al-Daghri NM, Alkharfy KM, Al-Othman A, El-Kholie E, Moharram O, Alokail MS et al. Vitamin D supplementation as an adjuvant therapy for patients with T2DM: an 18-month prospective interventional study. Cardiovasc Diabetol 2012; 11: 85. 18 Talwar SA, Aloia JF, Pollack S, Yeh JK. Dose response to vitamin D supplementation among postmenopausal African American women. Am J Clin Nutr 2007; 86: 1657–1662. 19 Gallagher JC, Peacock M, Yalamanchili V, Smith LM. Effects of vitamin D supplementation in older African American women. J Clin Endocrinol Metab 2013; 98: 1137–1146. 20 Aloia JF, Patel M, DiMaano R, Li-Ng M, Talwar SA, Mikhail M et al. Vitamin D intake to attain a desired serum 25-hydroxyvitamin D concentration. Am J Clin Nutr 2008; 87: 1952–1958.

European Journal of Clinical Nutrition (2015) 367 – 372

21 Logan VF, Gray AR, Peddie MC, Harper MJ, Houghton LA. Long-term vitamin D3 supplementation is more effective than vitamin D2 in maintaining serum 25-hydroxyvitamin D status over the winter months. Br J Nutr 2013; 109: 1082–1088. 22 Biancuzzo RM, Clarke N, Reitz RE, Travison TG, Holick MF. Serum concentrations of 1,25-Dihydroxyvitamin D2 and 1,25-Dihydroxyvitamin D3 in response to vitamin D2 and vitamin D3 supplementation. J Clin Endocrinol Metab 2013; 98: 973–979. 23 Nelson ML, Blum JM, Hollis BW, Rosen C, Sullivan SS. Supplements of 20 μg/d cholecalciferol optimized serum 25-Hydroxyvitamin D concentrations in 80% of premenopausal women in winter. J Nutr 2009; 139: 540–546. 24 Zittermann A, Prokop S, Gummert F, Borgermann J, Safety J. Issues of Vitamin D Supplementation. Anticancer Agents Med Chem 2013; 13: 4–10. 25 von Hurst PR, Stonehouse W, Coad J. Vitamin D status and attitudes towards sun exposure in South Asian women living in Auckland, New Zealand. Public Health Nutr 2010; 13: 531–536. 26 Fitzpatrick TB. The validity and practicality of sun-reactive skin types i through vi. Arch Dermatol 1988; 124: 869–871. 27 SPARC, McLean G, Tobias M. The New Zealand Physical Activity Questionnaires: Report on the Validation and Use of the NZPAQ-LF and NZPAQ-SF Self-report Physical Activity Survey Instruments. 2004. http://www.sparc.org.nz/researchpolicy/research/nzspas-97-01/nzpaq. 28 Ministry of Health Vitamin D Status of New Zealand Adults. Findings from the 2008/09 New Zealand Adult Nutrition Survey. Ministry of Health, 2012. 29 Barger-Lux MJ, Heaney RP. Effects of above average summer sun exposure on serum 25-Hydroxyvitamin D and calcium absorption. J Clin Endocrinol Metab 2002; 87: 4952–4956. 30 Giovannucci E, Liu Y, Hollis BW, Rimm EB. 25-Hydroxyvitamin D and risk of myocardial infarction in men: a prospective study. Arch Intern Med 2008; 168: 1174–1180. 31 Carroll MF, Schade DS. A practical approach to hypercalcemia. Am Fam Physician 2003; 67: 1959–1966. 32 Maghbooli Z, Hossein-Nezhad A, Shafaei AR, Karimi F, Madani FS, Larijani B. Vitamin D status in mothers and their newborns in Iran. BMC Pregnancy Childbirth 2007; 7: 1. 33 Rassouli A, Milanian I, Moslemi-Zadeh M. Determination of serum 25-hydroxyvitamin D3 levels in early postmenopausal Iranian women: relationship with bone mineral density. Bone 2001; 29: 428–430. 34 Hosseinpanah F, Pour SH, Heibatollahi M, Moghbel N, Asefzade S, Azizi F. The effects of air pollution on vitamin D status in healthy women: a cross sectional study. BMC Public Health 2010; 10: 519. 35 Saadi HF, Dawodu A, Afandi BO, Zayed R, Benedict S, Nagelkerke N. Efficacy of daily and monthly high-dose calciferol in vitamin D–deficient nulliparous and lactating women. Am J Clin Nutr 2007; 85: 1565–1571. 36 Field APDiscovering statistics using SPSS. SAGE: Los Angeles, 2009. 37 Judkins A, Eagleton C. Vitamin D deficiency in pregnant New Zealand women. N Z Med J 2006; 119: 1241. 38 Rockell JEP, Skeaff CM, Venn BJ, Williams SM, Green TJ. Vitamin D insufficiency in New Zealanders during the winter is associated with higher parathyroid hormone concentrations: implications for bone health? N Z Med J 2008; 121: 75–84. 39 Von Hurst PR, Stonehouse W, Coad J. Vitamin D supplementation reduces insulin resistance in South Asian women living in New Zealand who are insulin resistant and vitamin D deficient-a randomised, placebo-controlled trial. Br J Nutr 2010; 103: 549–555. 40 Vieth R, Chan P-CR, MacFarlane GD. Efficacy and safety of vitamin D3 intake exceeding the lowest observed adverse effect level. Am J Clin Nutr 2001; 73: 288–294. 41 Blum M, Dolnikowski G, Syoum E, Harris SS, Booth SL, Peterson J et al. Vitamin D3 in fat tissue. Endocr Pract 2008; 33: 90–94. 42 Holick MF. Vitamin D Deficiency. N Engl J Med 2007; 357: 266–281. 43 Brouwer DA, van Beek J, Ferwerda H, Brugman AM, van der Klis FR, van der Heiden HJ et al. Rat adipose tissue rapidly accumulates and slowly releases an orally-administered high vitamin D dose. Br J Nutr 1998; 79: 527–532. 44 Wortsman J, Matsuoka LY, Chen TC, Lu Z, Holick MF. Decreased bioavailability of vitamin D in obesity. Am J Clin Nutr 2000; 72: 690–693. 45 Piccolo BD, Dolnikowski G, Seyoum E, Thomas AP, Gertz ER, Souza EC et al. Association between subcutaneous white adipose tissue and serum 25-hydroxyvitamin D in overweight and obese adults. Nutrients 2013; 5: 3352–3366. 46 Heaney RP, Davies KM, Chen TC, Holick MF, Barger-Lux MJ. Human serum 25hydroxycholecalciferol response to extended oral dosing with cholecalciferol. Am J Clin Nutr 2003; 77: 204–210.

© 2015 Macmillan Publishers Limited

Copyright of European Journal of Clinical Nutrition is the property of Nature Publishing Group and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.

The effect of monthly 50,000 IU or 100,000 IU vitamin D supplements on vitamin D status in premenopausal Middle Eastern women living in Auckland.

Middle Eastern female immigrants are at an increased risk of vitamin D deficiency and their response to prescribed vitamin D dosages may not be adequa...
348KB Sizes 2 Downloads 6 Views