Eating and Weight Disorders - Studies on Anorexia, Bulimia and Obesity https://doi.org/10.1007/s40519-018-0511-4

ORIGINAL ARTICLE

Vitamin D deficiency in relation to general and abdominal obesity among high educated adults

Masoume Mansouri1 · Ali Miri2 · Mehdi Varmaghani3 · Rowshanak Abbasi4 · Parisa Taha1 · Shadi Ramezani1 · Elnaz Rahmani1 · Rohangyz Armaghan1 · Omid Sadeghi5

Received: 30 January 2018 / Accepted: 30 April 2018 © Springer International Publishing AG, part of Springer Nature 2018

Abstract Purpose To assess the association of vitamin D deficiency with general and abdominal obesity among high educated Iranian adults. Methods Current cross-sectional study was done on 500 Iranian professors aged 35 years or more. Complete data on general and abdominal obesity as well as serum 25(OH)D concentrations were available for 352 persons. Obesity was considered as body mass index ≥ 30, and abdominal obesity as waist circumference ≥ 80 cm for women and ≥ 94 cm for men. Furthermore, vitamin D deficiency was defined as serum 25(OH)D < 30 ng/ml. Results Mean age of study population was 53.03 ± 7.15 years. Compared with those in the first quartile of serum 25(OH) D, participants in the fourth quartile were less likely to be generally obese (OR 0.46, 65% CI 0.22–0.99). Such finding was also seen even after taking potential confounders into account. Furthermore, we found an inverse association between serum 25(OH)D and abdominal obesity in fully adjusted model (OR 0.44, 95% CI 0.22–0.86). In addition, a significant positive association was found between vitamin D deficiency and obesity; such that after controlling for potential confounders, par- ticipants with vitamin D deficiency had 2.16 and 2.04 times greater odds for having general (OR 2.16, 95% CI 1.05–4.45) and abdominal obesity (OR 2.04, 95% CI 1.16–3.60), respectively, than those with normal levels of vitamin D. Conclusion Serum 25(OH)D concentrations were inversely associated with general and abdominal obesity. In addition, vitamin D deficiency was positively associated with both general and abdominal obesity. Level of evidence Level V, cross-sectional descriptive study.

Keywords Anthropometry · Obesity · Vitamin D · Abdominal obesity

Introduction

Obesity has been a global epidemic public health issue during the past decades that imposes a large burden to the * Omid Sadeghi healthcare system [1]. Epidemiological studies have shown [email protected] that 55% of adult population in US are overweight or obese 1 Student Health Services, Health Center of Tarbiat Modares [2]. In Iran, it has been estimated that about half of adult University, Tehran, Iran population are affected [3]. Obesity is a known risk fac- 2 Department of Nutrition, School of Health, Zabol University tor for diabetes, hypertension, osteoarthritis, coronary heart of Medical Sciences, Zabol, Iran disease (CHD) and some cancers [1, 4, 5]. In addition to 3 Social Determinants of Health Research Center, Mashhad general obesity, distribution of fat in body is also important. University of Medical Sciences, Mashhad, Iran Abdominal obesity has been associated with a greater risk 4 Research Center for Prevention of Cardiovascular Disease, of metabolic diseases and mortality [6, 7]. Therefore, the Institute of Endocrinology and Metabolism, Iran University assessment of contributing factors for general and abdominal of Medical Sciences, Tehran, Iran obesity is of great importance. 5 Department of Community Nutrition, School of Nutritional It seems that genetic and environmental factors contribute Sciences and Dietetics, Tehran University of Medical to the obesity epidemic [8]. Several studies have focused on Sciences, P.O. Box 14155‑6117, Tehran, Iran

Vol.:(0123456789)1 3 Eating and Weight Disorders - Studies on Anorexia, Bulimia and Obesity micronutrients deficiencies [9–11]. Among micronutrients, participant between December 2016 and March 2017. In vitamin D has received a great attention [12–14]. Serum addition, fasting blood sample was taken to assess vitamin 25(OH)D concentrations have been reported to be inversely D status. Participants were excluded if had chronic diseases associated with impaired glucose regulation, insulin resist- affecting their diet, followed an arbitrary special dietary ance, β-cell dysfunction, and risk of metabolic syndrome regimen such as weight loss diet or were pregnant and lac- [15, 16]. Vitamin D may provide mentioned favourable tating during the previous year. In addition, we excluded effects by affecting on adipose tissue. In an experimental participants who had a history of vitamin D supplementa- study, it had been shown that vitamin D intake led to the tion or presented incomplete data. Finally, 352 participants activation of calcium-mediated apoptotic pathway in adi- (including 295 men and 57 women) remained for the final pose tissue and counteraction of body and fat weight gain analysis. All participants were willing to participate in the [17]. In another similar study, vitamin D up-regulated genes current study and provided a written informed consent. The involved in fatty acid oxidation and mitochondrial metabo- study was approved by Board of Directors and the ethics lism, followed by a significant rise in energy expenditure committee of Tarbiat Modares University, Tehran, Iran. [18]. Therefore, vitamin D deficiency may have a role in the obesity epidemic. On the other hand, in parallel in the increasing prevalence of vitamin D deficiency, the preva- Dietary assessment lence of obesity also is rising [19, 20]. Therefore, to shed light facts on the association between vitamin D deficiency We used some questions to assess consumption frequency and obesity, further studies are required. of some food items affecting the association of vitamin D Despite of several studies on the association between status with general and abdominal obesity. These questions vitamin D status and obesity, findings are conflicting. Some were about frequency of fruits intake (rarely, < 2 serv/day, studies have shown an inverse association between serum 2–4 serv/day, > 4 serv/day), vegetables (rarely, < 3 serv/ 25(OH)D concentrations and obesity [21, 22], while others day, 3–5 serv/day, > 5 serv/day), dairy products (rarely, < 2 failed to find any significant relationship [23]. Moreover, serv/day, 2–3 serv/day, > 3 serv/day), meat (rarely, < 2 serv/ prior studies are mostly from western countries and few data day, 2–3 serv/day, > 3 serv/day), red meat (daily, < 1 time/ are available from Asian countries, in particular from the day), whole grains (rarely, 1 time/week, 2–3 times/week, ≥ 1 understudies region of the Middle Eastern. In mentioned time/day) and refined sugar (rarely, 1 time/week, 2–3 times/ area, the prevalence of vitamin D deficiency has been esti- week, ≥ 1 time/day). Because of infrequently consumption mated to be high [24, 25]. Furthermore, assessing obesity is of whole grains in Iranian food culture [27], we changed particularly relevant for Middle Eastern population, due to the four response categories to two categories as “weekly” the prevalence of a Middle Eastern pattern of obesity, char- and “<1 time/week”. In the current study, whole-grain acterized by abdominal fat accumulation and enlarged waist foods were defined as wheat, diet bread and dark breads circumference (WC) [26]. Furthermore, mentioned associa- including Iranian breads of Sangak and Barbari. In addi- tion might be important among high educated individuals tion, red meat was defined as beef, veal, mutton and lamb, who are socioeconomically different than others. Therefore, dairy products as milk, cheese, yogurt and dough (yogurt current study aimed to assess the association of vitamin D drink) and refined sugar as sweets, cakes and biscuits. To status with general and abdominal obesity among high edu- include dietary intakes with the four response categories cated Iranian adults. into the multivariate analysis, we considered two of the first categories as “low intake” and two of the last categories as “adequate intake”. Subjects and methods

Participants Biochemical assessment

Current study was cross-sectional which was performed on To quantify serum concentrations of 25(OH)D, we collected all professors (assistant, associate and full faculty mem- blood samples after a 12-h overnight fast. Blood sampling bers) affiliated to Tarbiat Modares University, Tehran, Iran. was done in winter. Venous blood samples were drawn into First, 500 faculty members, aged 35 years or more, were 10-mL syringes and then centrifuged for 10 min at 500 g at invited to Health Center of Tarbiat Modares University by 4C within 30 to 45 min of collection. Serum 25(OH)D levels sending invitation letter via Email, SMS and post. Among were assessed by a commercial ELISA kit (IDS). Partici- them, 360 persons agreed to take part in the study. Required pants were classified into three categories based on serum data on demographic variables along with anthropometric levels of 25(OH)D: vitamin D sufficient (≥ 30 ng/ml), insuf- measurements and dietary habits were collected from each ficient (20–29 ng/ml) and deficient (< 20 ng/ml) [28].

1 3 Eating and Weight Disorders - Studies on Anorexia, Bulimia and Obesity

Anthropometric assessment the second model. In the final model, further controlling was done for dietary intakes of fruits (< 2/> 2 serv/day), Weight was measured with minimal clothing and without vegetables (< 3/> 3 serv/day), dairy (< 2/> 2 serv/day), red shoes by analogue scale with a precision of 100 g and height meat (daily/< 1 time/day), whole grains (weekly/< 1 time/ was determined in a standing position without shoes by a week) and refined sugar (< 2/> 2 serv/day). In all analyses, tape measure with the nearest 0.5 cm. Body mass index participants in the first quartile of serum 25(OH)D or with (BMI) was calculated as weight in kilograms divided by the normal levels of 25(OH)D (≥ 30 ng/ml) were considered height in square meters. Obesity was considered as body as the reference group. To determine the overall trend of mass index ≥ 30. To measure waist circumference (WC), we odds ratios across increasing categories of serum 25(OH) measured the middle of bottom ribs and pelvic bones after D, we considered these categories as an ordinal variable in a normal exhale using an inelastic tape. Abdominal obesity the logistic regression models. All statistical analyses were was defined based on criteria reported by Lean et al. and the performed using SPSS software (version 19.0; SPSS Inc, National Cholesterol Education Program (NCEP), respec- Chicago IL). P values were considered significant at < 0.05. tively [29, 30]. Participants were categorized into 2 groups: normal (< 80 cm for women, < 94 cm for men) and abdomi- naly obese (≥ 80 cm for women and ≥ 94 cm for men). Results

Assessment of other variables Mean age of study population was 53.03 ± 7.15 years. Vita- min D deficiency was prevalent among 60.5% of partici- Data on age, marital status (single/married), smoking (non- pants. In addition, 20.3% of participants were affected by smoker/former smoker/current smoker) and sleep pattern general obesity and 54.5% were abdominally obese. was obtained using a pre-tested questionnaire. In terms of General characteristics of faculty members across catego- sleep pattern, participants were asked two questions as fol- ries of serum 25(OH)D are indicated in Table 1. Compared low: “how is your pattern of sleeping and awaking?” and with the lowest quartile, participants in the highest quartile “how many hours do you sleep in a day?” The response cat- of serum 25(OH)D were older, more likely to be females and egories for the first question were “regular”, “irregular” and had lower weight and BMI. In addition, the prevalence of for the second question were as follow: “<6 h/day”, “6–8 h/ smoking was different across quartiles of serum 25(OH)D. day”, “8–10 h/day”, >10 h/day. To assess physical activity, In terms of standard ranges of vitamin D, participants with we asked a question as follow: “how often do you exercise vitamin D deficiency had higher weight, BMI and WC com- lasting 30 min?” The response categories of this question pared with those who were vitamin D sufficient. No other was “rarely”, “1–2 times/week”, “3–4 times/week”, “>5 significant difference was found for general characteristics times/week”. We collected no information about economic across categories of serum 25(OH)D concentrations. status of participants due to favourable economic status of Multivariable adjusted odds ratios and 95% confidence faculty members. intervals for general and abdominal obesity across categories of serum 25(OH)D concentrations are presented in Table 2. Statistical analysis Compared with those in the first quartile of serum 25(OH) D, participants in the fourth quartile were less likely to be First, we categorized participants according to quartiles of generally obese (OR 0.46, 65% CI 0.22–0.99); such that after serum 25(OH)D concentrations and also standard ranges taking potential confounders into account, individuals in the of 25(OH)D levels [sufficiency (≥ 30 ng/ml), insufficiency highest quartile of serum 25(OH)D had 60% less odds for (20–29 ng/ml), deficiency (< 20 ng/ml)]. One-way analy- having general obesity compared with those in the lowest sis of variance (ANOVA) was applied to assess significant quartile (OR 0.40, 95% CI 0.17–0.91). In terms of abdomi- differences in continuous variables across categories of nal obesity, when the potential confounders were controlled, serum 25(OH)D concentrations. To assess the distribu- a significant inverse association was found between serum tion of participants in terms of categorical variables across 25(OH)D and abdominal obesity; such that participants in categories of serum 25(OH)D, we used Chi-square test. the top quartile of serum 25(OH)D had 56% less likely to Binary logistic regression in different models was applied be abdominally obese compared with those in the bottom to assess the association of serum 25(OH)D with general quartile (OR 0.44, 95% CI 0.22–0.86). In addition, a sig- and abdominal obesity. In the first model, age (continuous) nificant positive association was found between vitamin D and gender were controlled. Additionally, adjustment was deficiency and both kinds of obesity; such that after control- made for marital status (single/married), physical activity ling for potential confounders, participants with vitamin D (< 3/> 3 times/week), smoking (non-smoker/former smoker/ deficiency had 2.16 and 2.04 times greater odds for having current smoker) and sleeping pattern (regular/irregular) in general (OR 2.16, 95% CI 1.05–4.45) and abdominal obesity

1 3 Eating and Weight Disorders - Studies on Anorexia, Bulimia and Obesity

Table 1 General characteristics and dietary intakes of participants across categories of serum 25(OH)D concentrations Variables Quartiles of serum 25(OH)D Based on standard ranges of vitamin D levels Q1 Q2 Q3 Q4 P Sufficiency Insufficiency Deficiency P (≥ 30 ng/ (20–29 ng/ (< 20 ng/ ml) ml) ml)

N 89 90 85 88 93 46 213 Age (year) 51.4 ± 7.3 52.6 ± 7.3 53.6 ± 6.9 54.3 ± 6.7 0.03 54.2 ± 6.6 52.9 ± 7.2 52.4 ± 7.2 0.12 Gender (female) (%) 5.6 21.1 15.3 22.7 0.008 22.6 19.6 12.7 0.07 Weight (Kg) 81.7 ± 12.8 80.8 ± 14.04 79.3 ± 10.6 75.3 ± 12.6 0.005 75.7 ± 12.5 77.8 ± 10.9 81.2 ± 12.9 0.002 BMI 27.8 ± 3.8 27.7 ± 3.9 27.3 ± 3.2 26.4 ± 3.7 0.05 26.3 ± 3.6 26.5 ± 4.7 27.7 ± 3.8 0.008 WC (cm) 94.4 ± 10.3 92.8 ± 9.9 92.4 ± 13.08 90.4 ± 9.6 0.10 90.5 ± 9.5 89.7 ± 13.4 94.0 ± 10.5 0.007 Marital status (married) (%) 98.9 97.8 95.3 95.4 0.43 95.7 95.7 97.7 0.57 Physical activity (> 3 times/ 32.6 25.6 30.1 27.3 0.74 27.2 31.1 29.1 0.88 week) (%) Sleep pattern (regular) (%) 85.4 83.3 82.4 79.3 0.75 78.3 82.6 84.5 0.41 Sleep duration (> 6 h/day) (%) 80.7 83.3 78.8 77.3 0.76 75.3 78.3 82.5 0.32 Current smoker (%) 29.2 8.9 36.5 29.5 0.001 30.1 34.8 22.1 0.11 Fruits (≥ 2 serv/day) (%) 78.7 71.1 70.6 77.3 0.49 76.3 73.9 73.7 0.88 Vegetable (≥ 3 serv/day) (%) 36.0 29.2 28.6 33.3 0.31 32.6 35.6 30.7 0.80 Dairy (≥ 2 serv/day) (%) 48.9 42.2 37.6 50.0 0.31 50.5 39.1 43.4 0.36 Red meat (daily) (%) 87.5 68.5 75.9 77.0 0.27 76.1 82.2 76.7 0.68 Refined sugar (≥ 2 serv/day) 55.1 59.6 56.5 48.9 0.54 49.5 54.3 57.5 0.42 (%) Whole grains (weekly) (%) 79.3 88.9 81.0 81.6 0.33 81.5 82.6 83.3 0.92

Data are presented as mean ± SD or percent BMI body mass index, WC waist circumference

Table 2 Multivariable adjusted odds ratios and 95% confidence intervals for general and abdominal obesity based on categories of serum 25(OH)D concentrations Quartiles of serum 25(OH)D Based on standard ranges of vitamin D levels Q1 Q2 Q3 Q4 P-trend Sufficiency Insufficiency Deficiency P-trend (≥ 30 ng/ (20–29 ng/ml) (< 20 ng/ml) ml)

General obesity N 89 90 85 88 93 46 213 Crude 1 0.77 (0.39–1.53) 0.58 (0.28–1.20) 0.46 (0.22–0.99) 0.03 1 1.10 (0.40–2.98) 1.98 (1.02–3.86) 0.03 Model 1 1 0.69 (0.34–1.39) 0.52 (0.25–1.09) 0.39 (0.18–0.87) 0.01 1 1.15 (0.42–3.14) 2.16 (1.10–4.26) 0.01 Model 2 1 0.63 (0.30–1.30) 0.49 (0.23–1.05) 0.36 (0.16–0.81) 0.01 1 1.18 (0.42–3.30) 2.30 (1.13–4.67) 0.01 Model 3 1 0.64 (0.30–1.37) 0.54 (0.25–1.19) 0.40 (0.17–0.91) 0.02 1 1.16 (0.41–3.28) 2.16 (1.05–4.45) 0.02 Abdominal obesity N 89 90 85 88 93 46 213 Crude 1 1.11 (0.61–2.02) 0.83 (0.46–1.52) 0.71 (0.39–1.28) 0.17 1 0.78 (0.38–1.60) 1.50 (0.92–2.46) 0.06 Model 1 1 0.91 (0.49–1.69) 0.66 (0.36–1.24) 0.50 (0.26–0.94) 0.02 1 0.85 (0.41–1.78) 1.88 (1.12–3.16) 0.009 Model 2 1 0.85 (0.45–1.61) 0.64 (0.33–1.22) 0.48 (0.25–0.91) 0.01 1 0.91 (0.42–1.94) 1.85 (1.08–3.17) 0.01 Model 3 1 0.81 (0.41–1.57) 0.67 (0.34–1.30) 0.44 (0.22–0.86) 0.01 1 0.99 (0.45–2.16) 2.04 (1.16–3.60) 0.008

Model 1: Adjusted for age and gender Model 2: Further adjustment for marital status, physical activity, smoking and sleeping pattern Model 3: Additionally control for consumption of fruits, vegetables, dairy, red meat, whole grains and refined sugar

1 3 Eating and Weight Disorders - Studies on Anorexia, Bulimia and Obesity

(OR 2.04, 95% CI 1.16–3.60), respectively, than those with both normal-weight and obese participants [38]. Mentioned normal levels of vitamin D. findings were confirmed by some clinical trials showing vitamin D supplementation increases insulin sensitivity [39], improves glucose homeostasis [40], decreases the leptin and Discussion increases the adiponectin levels [41, 42], and followed by a rise in energy expenditure [18]. Furthermore, in a meta- In this study, we found a significant inverse association analysis, it had been shown that vitamin D supplementation between serum 25(OH)D concentrations, general and decreased the incidence of Type 1 diabetes by 25% [43]. In abdominal obesity. Such finding was significant even after the current study, the prevalence of vitamin D deficiency was adjusting for confounding variables. In addition, participants high (60.5%). However, individuals participated in this study with vitamin D deficiency were more likely to be generally was high educated. Therefore, it seems that high education and abdominally obese compared with those with normal level is not a reason for the lack of nutritional deficiencies. levels of vitamin D when potential confounders were taken Different theories can be proposed to explain that vita- into account. To the best of our knowledge, this study was min D deficiency might favour greater adiposity. Vitamin the first in Middle East to examine the association between D deficiency increases parathyroid hormone levels and vitamin D deficiency and obesity among high educated greater inflow of calcium into adipocytes. Both conditions adults. induce lipogenesis [44]. Accumulated evidence showed that The World Health Organization has projected that there 1.25(OH)D inhibits adipogenesis by actions modulated by are approximately 1.5 billion overweight adults around the vitamin D-dependent receptors [45]. In an experimental world [31]. Taking into account the association between study, murine 3T3-L1 pre-adipocytes exposed to vitamin vitamin D deficiency and obesity, these two conditions might D and/or alendronate (ALN), showed an increment in vita- constitute important health issues in future years. Based on min D receptors (VDR) mRNA expression and a signifi- our findings, serum 25(OH)D concentrations were inversely cant reduction in levels of peroxisome proliferator-activated associated with general and abdominal obesity. In line with receptor gamma (PPARγ), the master gene of adipogenesis our findings, Zhang et al. reported a significant inverse asso- [46]. Murine 3T3-L1 pre-adipocytes are one of the best ciation between serum 25(OH)D levels and obesity, particu- characterized and widely used cellular models to study larly abdominal obesity [21]. In a cross-sectional study, an adipocyte differentiation [47]. Vitamin D deficiency might inverse relationship was reported between serum 25(OH)D lead to excessive differentiation of pre-adipocytes to adipo- levels and risk of metabolic obesity among male subjects cytes [48]. In addition, a study on a diet-induced obesity in [13]. In a systematic review and meta-analysis, it had been mouse models revealed that intake of high doses of vitamin shown that vitamin D intake improved the cardiovascular D, especially if associated with high intake of calcium, leads risk factors in adults with obesity [32]. Another study, done to the activation of calcium-mediated apoptotic pathway in by Afzal et al. showed that the low plasma 25(OH)D con- adipose tissue, reduction of body and fat weight gain, and centrations might be a modest mediator between obesity and improvement in adiposity markers (plasma concentrations increased risk of diabetes [22]. In contrast, Khan et al. failed of glucose, insulin, and adiponectin) [17]. to find any significant association between vitamin D status Vitamin D is a fat-soluble vitamin that is stored in adi- and obesity [23]. Finding no significant association in that pose tissue and released in small amounts from this tissue, study might be due to low quality of this study and lack of depending on the stored quantity [41]. As a consequence, adjustment for confounding variables. a greater adipose mass decreases the available vitamin D The prevalence of obesity has been increasing at an in blood [41]. Therefore, it must be kept in mind that obe- alarming rate [20]. Recently, the effects of low concentra- sity may be a reason for vitamin D deficiency. Prior studies tions of serum 25(OH)D on imprinting during childhood have also shown that obese individuals are more likely to be and adolescence and subsequent occurrence of bone diseases vitamin D deficient compared with normal-weight people have been assessed, along with the growing risk of develop- [41]. This finding was also confirmed by studies assessed ment of metabolic syndrome and cardiovascular, respiratory the effect of sleeve gastrectomy on serum level of 25(OH) and psychological disorders [33, 34]. On the other hand, D [49–52]. In a clinical trial, after 3 month weight loss by vitamin D deficiency predisposes the adults to increased sleeve gastrectomy in morbidly obese patients, 25(OH)D risk of chronic diseases such as hypertension, diabetes, car- concentrations were increased significantly 50[ ]. In another diovascular diseases, different types of cancer and excess similar study, weight loss along with increment in 25(OH) weight [35–37]. In Third National Health and Nutrition D levels occurred after sleeve gastrectomy improved bone Examination Survey (NHANES III), after a median follow- mineral density in obese patients [51]. Søvik et al. reported up of 19.1 years, vitamin D deficiency was associated with that weight loss due to gastric bypass not only increased the increased risk of cardiometabolic diseases and mortality in 25(OH)D concentrations, it also improved cardiovascular

1 3 Eating and Weight Disorders - Studies on Anorexia, Bulimia and Obesity risk factors [52]. Beneficial effects on risk factors related to References cardiovascular diseases in mentioned study might be due to increased levels of vitamin D. Therefore, it is still unclear 1. World Health Organization. WHO Media center. Obesity and that vitamin D deficiency develops obesity or vice versa. overweight. World Health Organization, Geneva 2. Brooks B, Rajeshwari R, Nicklas TA, Yang S-J, Berenson GS Further studies, particularly of a prospective nature, are (2006) Association of calcium intake, dairy product consumption needed to shed light facts in this regard. with overweight status in young adults (1995–1996): the Bogalusa The prevalence of vitamin D deficiency in individuals Heart Study. J Am Coll Nutr 25:523–532 participated in the current study was high (60.5%). On rea- 3. Janghorbani M, Amini M, Willett WC, Gouya MM, Delavari A, Alikhani S et al (2007) First nationwide survey of prevalence of son for this high prevalence might be seasonal variation in overweight, underweight, and abdominal obesity in Iranian adults. serum 25(OH)D. Blood sampling in this study was in win- Obesity 15:2797–2808. https​://doi.org/10.1038/oby.2007.332 ter and in this season, sunlight exposure is low. However, 4. Silverwood V, Blagojevic-Bucknall M, Jinks C, Jordan J, Pro- sunshine exposure in Middle East is potentially high and theroe J, Jordan K (2015) Current evidence on risk factors for knee osteoarthritis in older adults: a systematic review and meta- the effects of seasonal variation on serum 25(OH)D might analysis. Osteoarthr Cartil 23:507–515. https://doi.org/10.1016/j.​ be low [19]. In future studies, seasonal variation in serum joca.2014.11.019 25(OH)D should be considered on the association between 5. Deng T, Lyon CJ, Bergin S, Caligiuri MA, Hsueh WA (2016) vitamin D levels and obesity. Obesity, inflammation, and cancer. Annu Rev Pathol 11:421–449. https​://doi.org/10.1016/j.yexmp​.2017.01.003 This study has several limitations that should be con- 6. Ekelund U, Ward HA, Norat T, Luan JA, May AM, Weiderpass sidered when interpreting our findings. The design of our E et al (2015) Physical activity and all-cause mortality across study was cross-sectional that prohibits inferring a causal levels of overall and abdominal adiposity in European men and link between vitamin D status and obesity. Therefore, women: the European Prospective Investigation into Cancer and Nutrition Study (EPIC)–. Am J Clin Nutr 101:613–621. https​:// prospective studies are needed to confirm our findings. In doi.org/10.3945/ajcn.114.10006​5 addition, participants with obesity may have reduced their 7. Azadbakht L, Esmaillzadeh A (2008) Dietary and non-dietary dietary intakes to lose weight that affect intake of foods rich determinants of central adiposity among Tehrani women. Public in vitamin D. Therefore, actual results may be even stronger Health Nutr 11:528–534. https​://doi.org/10.1017/S1368​98000​ 70008​82 than those obtained. Despite several adjustments, further 8. Bank SS, Ghanjali N, Ghalaeh RS, Azadbakht L (2013) Associa- control for other residual confounding variables such as tion between dairy and calcium intake and general and central dietary intake of energy, fish and mushrooms (as sources of obesity among female students. J Educ Health Promot 31:16. https​ vitamin D) intake, skin type, sun exposure and psychologi- ://doi.org/10.4103/2277-9531.11268​5 9. Krzizek EC, Brix JM, Herz CT, Kopp HP, Schernthaner GH, cal factors might be needed to reach an independent associa- Schernthaner G et al (2018) Prevalence of micronutrient defi- tion between vitamin D status with general and abdominal ciency in patients with morbid obesity before bariatric sur- obesity. gery. Obes Surg 28:550–551. https​://doi.org/10.1007/s1169​ In conclusion, serum 25(OH)D concentrations were 5-017-3027-5 10. Poli VFS, Sanches RB, Moraes ADS, Fidalgo JPN, Nascimento inversely associated with general and abdominal obesity MA, Bresciani P et al (2017) The excessive caloric intake and either before or after controlling for potential confounders. micronutrient deficiencies related to obesity after a long-term Furthermore, in fully adjusted model, vitamin D deficiency interdisciplinary therapy. Nutrition 38:113–119. https​://doi. was positively associated with both kinds of obesity. org/10.1016/j.nut.2017.01.012 11. Ribaldone DG, Saracco GM, Pellicano R (2018) Prevalence of Acknowledgements micronutrient deficiency in patients with morbid obesity before Authors appreciate the valuable assistance of all bariatric surgery: what about celiac disease? Obes Surg 28:550– subjects. We also would like to thank the authorities of Health Center 551. https​://doi.org/10.1007/s1169​5-017-3027-5 of Tarbiat Modares University, Tehran, Iran, for their cooperation. 12. Censani M, Hammad HT, Christos PJ, Schumaker T (2018) Vitamin D deficiency associated with markers of cardiovas- Funding This study was supported by Health Center of Tarbiat cular disease in children with obesity. Glob Pediatr Health Modares University, Tehran, Iran. 5:2333794X17751773. https://doi.org/10.1177/23337​ 94x17​ 75177​ ​ 3 Compliance with ethical standards 13. Ha J, Jo K, Lim DJ, Lee JM, Chang SA, Kang MI et al (2017) Parathyroid hormone and vitamin D are associated with the risk of Conflict of interest metabolic obesity in a middle-aged and older Korean population Authors declared no personal or financial conflicts with preserved renal function: a cross-sectional study. PLoS One of interest. 12:e0175132. https​://doi.org/10.1371/journ​al.pone.01751​32 Ethical approval 14. Mansouri M, Abasi R, Nasiri M, Sharifi F, Vesaly S, Sadeghi All procedures performed in this study were in O et al (2018) Association of vitamin D status with metabolic accordance with the ethical standards of the institutional and/or syndrome and its components: a cross-sectional study in a pop- national research committee and with the 1964 Helsinki declaration ulation of high educated Iranian adults. Diabetes Metab Syndr and its later amendments or comparable ethical standards. 17:30480–30480. https​://doi.org/10.1016/j.dsx.2018.01.007 Informed consent 15. Forouhi NG, Luan JA, Cooper A, Boucher BJ, Wareham NJ Informed consent was obtained from all subjects (2008) Baseline serum 25-hydroxy vitamin D is predictive included in the study.

1 3 Eating and Weight Disorders - Studies on Anorexia, Bulimia and Obesity

of future glycemic status and insulin resistance: the Medical obesity: a systematic review and meta-analysis. Eur J Clin Invest Research Council Ely Prospective Study 1990–2000. Diabetes 45:1113–1126. https​://doi.org/10.1111/eci.12510​ 57:2619–2625. https​://doi.org/10.2337/db08-0593 33. Holick MF (2006) Resurrection of vitamin D deficiency and rick- 16. Kayaniyil S, Vieth R, Retnakaran R, Knight JA, Qi Y, Ger- ets. J Clin Invest 116:2062–2072. https​://doi.org/10.1172/JCI29​ stein HC et al (2010) Association of vitamin D with insulin 449 resistance and β-cell dysfunction in subjects at risk for type 2 34. McGrath J (2001) Does ‘imprinting’with low prenatal vitamin D diabetes. Diabetes Care 33:1379–1381. https://doi.org/10.2337/​ contribute to the risk of various adult disorders? Med Hypotheses dc09-2321 56:367–371. https​://doi.org/10.1054/mehy.2000.1226 17. Sergeev IN, Song Q (2014) High vitamin D and calcium intakes 35. Palomer X, González-Clemente J, Blanco-Vaca F, Mauricio D reduce diet-induced obesity in mice by increasing adipose tis- (2008) Role of vitamin D in the pathogenesis of type 2 diabetes sue apoptosis. Mol Nutr Food Res 58:1342–1348. https​://doi. mellitus. Diabetes Obes Metab 10:185–197. https​://doi.org/10.1 org/10.1002/mnfr.20130​0503 111/j.1463-1326.2007.00710​.x 18. Marcotorchino J, Tourniaire F, Astier J, Karkeni E, Canault M, 36. Cheng S, Massaro JM, Fox CS, Larson MG, Keyes MJ, McCabe Amiot M-J et al (2014) Vitamin D protects against diet-induced EL et al (2010) Adiposity, cardiometabolic risk, and vitamin D obesity by enhancing fatty acid oxidation. J Nutr Biochem status: the Framingham Heart Study. Diabetes 59:242–248. https​ 25:1077–1083. https​://doi.org/10.1016/j.jnutb​io.2014.05.010 ://doi.org/10.2337/db09-1011 19. Lips P (2007) Vitamin D status and nutrition in Europe and Asia. J 37. Parker J, Hashmi O, Dutton D, Mavrodaris A, Stranges S, Kandala Steroid Biochem Mol Biol 103:620–625. https://doi.org/10.1016/j.​ N-B et al (2010) Levels of vitamin D and cardiometabolic disor- jsbmb​.2006.12.076 ders: systematic review and meta-analysis. Maturitas 65:225–236. 20. Sadeghi O, Saneei P, Nasiri M, Larijani B, Esmaillzadeh A (2017) https​://doi.org/10.1016/j.matur​itas.2009.12.013 Abdominal obesity and risk of hip fracture: a systematic review 38. Al-khalidi B, Kimball SM, Kuk JL, Ardern CI (2018) Meta- and meta-analysis of prospective studies. Adv Nutr 8:728–738. bolically healthy obesity, vitamin D, and all-cause and cardio- https​://doi.org/10.3945/an.117.01554​5 metabolic mortality risk in NHANES III. Clin Nutr. https​://doi. 21. Zhang Y, Zhang X, Wang F, Zhang W, Wang C, Yu C et al (2016) org/10.1016/j.clnu.2018.02.025 The relationship between obesity indices and serum vitamin D 39. Cefalo CM, Conte C, Sorice GP, Moffa S, Sun VA, Cinti F et al levels in Chinese adults from urban settings. Asia Pac J Clin Nutr (2018) Effect of Vitamin D supplementation on obesity-induced 25:333–339. https​://doi.org/10.6133/apjcn​.2016.25.2.15 insulin resistance: a double-blind, randomized, placebo-controlled 22. Afzal S, Brondum-Jacobsen P, Bojesen SE, Nordestgaard BG trial. Obesity (Silver Spring) 26:651–657. https://doi.org/10.1002/​ (2014) Vitamin D concentration, obesity, and risk of diabetes: oby.22132​ a mendelian randomisation study. Lancet Diabetes Endocrinol 40. Wimalawansa SJ (2016) Associations of vitamin D with insulin 2:298–306. https​://doi.org/10.1016/s2213​-8587(13)70200​-6 resistance, obesity, type 2 diabetes, and metabolic syndrome. J 23. Khan AH, Iqbal R, Naureen G, Dar FJ, Ahmed FN (2012) Preva- Steroid Biochem Mol Biol 175:177–189. https://doi.org/10.1016/j.​ lence of vitamin D deficiency and its correlates: results of a com- jsbmb​.2016.09.017 munity-based study conducted in Karachi, Pakistan. Arch Osteo- 41. Caprio M, Infante M, Calanchini M, Mammi C, Fabbri A (2017) poros 7:275–282. https​://doi.org/10.1007/s1165​7-012-0108-x Vitamin D: not just the bone. Evidence for beneficial pleiotropic 24. Goswami R, Mishra SK, Kochupillai N (2008) Prevalence and extraskeletal effects. Eat Weight Disord 22:27–41. https​://doi. potential significance of vitamin D deficiency in Asian Indians. org/10.1007/s4051​9-016-0312-6 Indian J Med Res 127:229–238 42. Stokić E, Kupusinac A, Tomic-Naglic D, Smiljenic D, Kovacev- 25. van der Meer IM, Middelkoop BJ, Boeke AJ, Lips P (2011) Preva- Zavisic B, Srdic-Galic B et al (2015) Vitamin D and dysfunc- lence of vitamin D deficiency among Turkish, Moroccan, Indian tional adipose tissue in obesity. Angiology 66:613–618. https​:// and sub-Sahara African populations in Europe and their countries doi.org/10.1177/00033​19714​54351​2 of origin: an overview. Osteoporos Int 22:1009–1021. https://doi.​ 43. Zipitis CS, Akobeng AK (2008) Vitamin D supplementation in org/10.1007/s0019​8-010-1279-1 early childhood and risk of type 1 diabetes: a systematic review 26. Esmaillzadeh A, Azadbakht L (2008) Major dietary patterns in and meta-analysis. Arch Dis Child 93:512–517. https​://doi. relation to general obesity and central adiposity among Iranian org/10.1136/adc.2007.12857​9 women. J Nutr 138:358–363. https://doi.org/10.1093/jn/138.2.358​ 44. Wood RJ (2008) Vitamin D and adipogenesis: new 27. Kimiagar M (1995) National food consumption survey. National molecular insights. Nutr Rev 66. https​://doi.org/10.111 Nutrition and Food Technology Research Institute, Tehran 1/j.1753-4887.2007.00004​.x 28. Holick MF, Binkley NC, Bischoff-Ferrari HA, Gordon CM, Han- 45. Martini LA, Wood RJ (2006) Vitamin D status and the metabolic ley DA, Heaney RP et al (2011) Evaluation, treatment, and pre- syndrome. Nutr Rev 64:479–486 vention of vitamin D deficiency: an Endocrine Society clinical 46. Mammi C, Calanchini M, Antelmi A, Feraco A, Gnessi L, Fal- practice guideline. J Clin Endocrinol Metab 96:1911–1930. https​ cone S et al (2013) Bisphosphonates and adipogenesis: evidence ://doi.org/10.1210/jc.2011-0385 for alendronate inhibition of adipocyte differentiation in 3T3-L1 29. Lean ME, Han TS, Morrison CE (1995) Waist circumference preadipocytes through a vitamin D receptor mediated effect. Nat as a measure for indicating need for weight management. BMJ Sci 5:955 311:158–161 47. Armani A, Mammi C, Marzolla V, Calanchini M, Antelmi A, 30. Third report of the National Cholesterol Education Program Rosano G et al (2010) Cellular models for understanding adi- (NCEP) Expert panel on detection, evaluation, and treatment of pogenesis, adipose dysfunction, and obesity. J Cell Biochem high blood cholesterol in adults (adult treatment panel III) final 110:564–572. https​://doi.org/10.1002/jcb.22598​ report. Circulation 2002;106:3143–3421 48. Pereira-Santos M, Costa PR, Assis AM, Santos CA, Santos DB 31. Wahl S, Drong A, Lehne B, Loh M, Scott WR, Kunze S et al (2015) Obesity and vitamin D deficiency: a systematic review (2017) Epigenome-wide association study of body mass index, and meta-analysis. Obes Rev 16:341–349. https://doi.org/10.1111/​ and the adverse outcomes of adiposity. Nature 541:81–86. https​ obr.12239​ ://doi.org/10.1038/natur​e2078​4 49. Zubiaga Toro L, Ruiz-Tovar Polo J, Díez-Tabernilla M, Giner 32. Manousopoulou A, Al-Daghri NM, Garbis SD, Chrousos GP Bernal L, Arroyo Sebastián A, Calpena Rico R (2014) Fórmula (2015) Vitamin D and cardiovascular risk among adults with CUN-BAE y factores bioquímicos como marcadores predictivos

1 3 Eating and Weight Disorders - Studies on Anorexia, Bulimia and Obesity

de obesidad y enfermedad cardiovascular en pacientes pre y laparoscopic sleeve gastrectomy. Obes Surg 23:861–866. https​:// post gastrectomía vertical. Nutr Hosp 30:281–286. https​://doi. doi.org/10.1007/s1169​5-013-0866-6 org/10.3305/nh.2014.30.2.7581 52. Søvik TT, Aasheim ET, Taha O, Fagerland MW, Björkman S, 50. Ruiz-Tovar J, Oller I, Tomas A, Llavero C, Arroyo A, Calero A Kristinsson J et al (2011) Weight loss, cardiovascular risk fac- et al (2012) Mid-term effects of sleeve gastrectomy on calcium tors, and quality of life after gastric bypass and duodenal switch: metabolism parameters, vitamin D and parathormone (PTH) a randomized trial. Ann Intern Med 155:281–291. https​://doi. in morbid obese women. Obes Surg 22:797–801. https​://doi. org/10.7326/0003-4819-155-5-20110​9060-00005​ org/10.1007/s1169​5-011-0582-z 51. Ruiz-Tovar J, Oller I, Priego P, Arroyo A, Calero A, Diez M et al (2013) Short-and mid-term changes in bone mineral density after

1 3