ORIGINAL ARTICLE

Relationship between 25-hydroxyvitamin D concentrations,serum calcium, and parathyroid hormone in apparently healthySyrian people

Rima Sayed-Hassan & Nizar Abazid & Zaynab Alourfi

Received: 27 November 2013 /Accepted: 24 March 2014# International Osteoporosis Foundation and National Osteoporosis Foundation 2014

AbstractSummary Vitamin D deficiency (25-hydroxyvitamin D(25OHD) <25 nmol/L) was common in a convenience sampleof apparently healthy Syrian adults. Female gender, season,and concealing clothing were independent predictors of vita-min D deficiency. Community-based research is needed toidentify vulnerable subgroups and inform public healthactions.Purpose Optimal vitamin D status for bone health has beeninferred from the determination of serum 25OHD levels be-low which there is an increase in serum parathyroid hormone(PTH). Studies worldwide showed high prevalence ofhypovitaminosis D even in sunny countries. There is littleevidence about its prevalence among Syrian adult population.We aimed to assess the serum levels of 25OHD and factorsrelated to vitamin D inadequacy and its relation to serum PTHand calcium among apparently healthy adults.Methods Serum 25OHD and PTH measurements were ob-tained from 372 subjects aged 18–62 years living in Damascusand its surroundings, between April 2011 and March 2013.Binary logistic regression was used to assess risk factors forhypovitaminosis D.Results The mean (standard deviation (SD)) 25OHD level was24.7 (16.9) nmol/L [9.8 (6.7) ng/mL] and was higher in menthan women (p<0.001). Levels <25, <50, and <75 nmol/L

were detected in 61, 90.1, and 99.2 % of the participants,respectively. Season influenced vitamin D status in men butnot in women (p<0.001). Female gender and wearing the veil(hijab) were independent predictors of vitamin D deficiency(25OHD<25 nmol/L). PTHwas significantly higher below thisthreshold (p<0.001). Serum 25OHD <25 nmol/L, sex, and age≥35 years were statistically significant factors for PTHelevation.Conclusions Vitamin D deficiency was highly prevalent inour sample. Further research is needed to identify populationgroups vulnerable for hypovitaminosis D and specify its pre-dictors and inform the necessary public health measures.

Keywords Syria . VitaminD deficiency . Secondaryhyperparathyroidism . Seasons . Concealing clothing

Introduction

Adequate vitamin D status is crucial for optimal global healthacross the lifespan [1–3]. In addition to the key role of vitaminD in bone health, recent research has demonstrated the impor-tance of maintaining adequate vitamin D status to preventvarious extraskeletal clinical disorders [1–4]. Serum 25-hydroxyvitamin D (25OHD) concentration is the best bio-marker to determine vitamin D status [2–6].

Severe chronic vitamin D deficiency causes skeleton min-eralization defects resulting in rickets in children and osteo-malacia in adults [1–5]. Less severe vitamin D inadequacy,termed “vitamin D insufficiency,” contributes to secondaryhyperparathyroidism and osteoporosis with increased boneturnover, progressive bone loss, and increased risk of fragilityfracture as well as increased risk for falls caused by muscleweakness [2–7].

Based on a comprehensive review of vitamin D literature inrelation to diverse health outcomes, the Institute of Medicine

R. Sayed-Hassan (*) : Z. AlourfiDepartment of Internal Medicine, Faculty of Medicine,Damascus University, P.O. Box 9241, Damascus, Syriae-mail: rimasayedhassan@gmail.com

Z. Alourfie-mail: zaynabarfi@hotmail.com

N. AbazidDepartment of Family & Community Medicine,Faculty of Medicine, Damascus University, Damascus, Syriae-mail: sdcm@pisem.net

Arch Osteoporos (2014) 9:176DOI 10.1007/s11657-014-0176-1

and the European Society for Clinical and Economic Aspectsof Osteoporosis and Osteoarthritis estimated that a level of50 nmol/L (20 ng/mL) was adequate for bone health in 97 %of the population [6, 8]. On the other hand, the US EndocrineSociety andmany experts agree that the recommended level ofvitamin D for optimal health benefit should be greater than75 nmol/L (30 ng/mL) [9–11], especially in elderly subjectswho are at elevated risk for falls and fractures [6, 12, 13].

The inverse relationship between serum 25OHD and para-thyroid hormone (PTH) levels was mainly used to determinenormal serum 25OHD level; therefore, this optimal level hasbeen proposed as the range where the PTH concentrationsbegan to plateau at their nadir, with maximal efficiency ofintestinal calcium absorption and adequate bone mineraldensity [3, 10]. Different cutoff points of serum 25OHDsuggested for maximal suppression of PTH and preven-tion of secondary hyperparathyroidism range from 25 to122 nmol/L (10 to 48.8 ng/mL) [7, 11, 14–20]. However, notall studies from different age groups have found such a plateau[13, 15, 16, 21].

Circulating 25OHD levels for skeletal health depend onenvironmental, cultural, and biological factors, including sunexposure, latitude, season, concealing clothing, age, gender,race/ethnicity, skin color, body mass index, socioeconomicstatus, medication use, and vitamin D dietary intake (fromfortified food or supplements) [2, 3, 5, 6, 14, 22–25].

Regardless of the differences in assay methodology, vita-min D inadequacy is highly prevalent across age groups andcountries, whatsoever their latitude, as suggested by a plethoraof research worldwide [1, 3, 7, 14, 22–26], with higher prev-alence rates in the Middle East [6, 7, 25]. It has been estimatedthat 20–80% of apparently healthy individuals inMiddle Eastcountries have hypovitaminosis D, with mean 25OHD levelreported near or below 25 nmol/L (10 ng/mL) across age

groups [22–24]. Hypovitaminosis D, low calcium intake,and inadequate exposure to sunlight were recognized as themain risk factors for low peak bonemineral density among theArab population and, consequently, for osteoporosis, com-pared to European and North American subjects [27, 28].

Syria is an Eastern Mediterranean country character-ized by a temperate climate that is mostly sunny aroundthe year. There is little evidence concerning vitamin Dstatus in Syrian adults living in Damascus. We hypoth-esized that Syrian population is not an exception, andvitamin D inadequacy is relatively prevalent amongdifferent age groups. Therefore, this study aimed todetermine the prevalence of hypovitaminosis D and itsdeterminants and its relation to serum PTH and calciumconcentrations in apparently healthy Syrian adults toprovide an estimation of the magnitude of the problemin our country.

Subjects and methods

Study subjects

This cross-sectional study was conducted in Damascus whichis located at latitude 33.25° north and longitude 36.31° east,between April 2011 and March 2013. Participants were re-cruited at Al-AssadUniversity Hospital. A total of 372 healthySyrian volunteers aged 18–62 years were included in theanalysis (Fig. 1). The study protocol was approved by theDamascus University institutional review board. A writteninformed consent was obtained from each participant.

All study subjects underwent a full clinical assessment.Exclusion criteria included any current acute or chronic illness;impaired renal or liver function; intestinal malabsorption; and

412Baseline screening: apparently healthy subjects

Clinical approach and biochemical screening:

372 recruited into the study

15

21

4 creatinine clearance < 60

taking vitamin D and calcium, bisphosphonates

chronic diseases (diabetes, familial Mediterranean fever, hypertension)

Fig. 1 Patient recruitment

176, Page 2 of 10 Arch Osteoporos (2014) 9:176

medication influencing bone metabolism such as calcium orvitamin D supplements, anticonvulsants, or corticosteroids; andfamily history of hypocalcemia or vitamin D disorders. Womenwho were pregnant or lactating within the last 3 years or takingoral contraceptives were also excluded. All the subjects whohad 25OHD <50 nmol/L (20 ng/mL) received appropriatetreatment, whether or not they were ultimately included in thestudy.

Clinical measurements

Anthropometric measurements were performed using thesame standardized techniques and calibrated digital scale(Seca, Germany), throughout the study. Body mass index(BMI) was calculated, for all subjects wearing light clothingand no shoes, as weight in kilograms (kg) divided by squaredheight in meters (m).

Participants were classified into two age groups of 18–34and ≥35 years. This cutoff point was chosen for two reasons.First, a report from Lebanon, a neighboring country, hassuggested that peak bone mass was achieved by the age of35 years [28]; second, after mid-30s, bone mineral densitystarts to decline slowly [29]. Additionally, women were di-vided into two groups: non-veiled and veiled, i.e., wearing thetraditional Islamic clothes (hijab: head, arms, and legs coveredbut not face or hands). Previous number of pregnancies andtotal duration of lactation estimated by months were noted.

Information on calcium intake and physical activity wereindicated by dichotomous variables (yes/no), using the ques-tion “Do you consume three servings or more of dairy prod-ucts per week?,” based on the average consumption (132 g perperson/day) of milk and dairy products as reported by theWorld Health Organization for the Middle Eastern region[30]. Physical activity was categorized using the question“Do you exercise three or more times per week?” into “inac-tive or less active” (irregular exercise or no physical activity)vs. “active” (regular exercise ≥3 times/week).

Sunlight exposure was evaluated based on sunscreen useand average of self-reported hours per week (≤3 or >3 h/week)spent outdoors between 10 a.m. and 3 p.m., for daily activities.

Educational levels attained were classified as secondaryschool (12 grades) or less, vs. higher than secondaryeducation.

We stratified the sample into two “seasonal” groups basedon month of blood collection. We divided blood samples intosummertime, May to October, and wintertime, November toApril.

Laboratory measurements

A morning fasting blood sample was collected from eachsubject. All tests were performed in Al-Assad UniversityHospital clinical laboratory and analyzed immediately after

collection by the same team of laboratory technicians, and thesamemethod was used throughout the study period, using kitsprovided by the same manufacturer.

Serum calcium, phosphorus, albumin, creatinine, total al-kaline phosphatase (TAP), alanine transaminase (ALT), aspar-tate transaminase (AST), gamma-glutamyl transferase (GGT),and magnesium were measured by standard colorimetricmethods using the Roche Hitachi 912 autoanalyzer (RocheDiagnostics, Mannheim, Germany).

Serum calcium levels were adjusted for serum albuminusing the formula: Corrected calcium (mmol/L)=measuredtotal Ca (mmol/L)+[40−serum albumin (g/L)]×0.02 [31].

Creatinine clearance was calculated from plasma creatinineby the Cockcroft-Gault equation as follows: Creatinine clear-ance=[(140−ageyears)×weightkg×1.2/serum creatinineμmol/L](multiplied by 0.85 if female), and was standardized for abody surface area of 1.73 m2 [32, 33].

Both intact parathyroid hormone (iPTH) and total25OHD levels were measured using automatedelectrochemiluminescence immunoassay (Elecsys 2010 ana-lyzers, Roche Diagnostics GmbH,Mannheim, Germany). TheElecsys vitamin D total assay is a competitive protein bindingassay which uses vitamin D binding protein for the detectionof 25-hydroxyvitamin D. For 25OHD, the manufacturer’smeasuring range as reported in the kit was 7.5–175 nmol/L(3–70 ng/mL). The intra- and interassay coefficients of varia-tion were <8 and <11 %, respectively. The lower limit ofdetection was 7.5 nmol/L (3 ng/mL). For iPTH, the manufac-turer’s normal range as reported in the kit was 1.6 to 6.9 pmol/L (15–65 pg/mL). The intra- and interassay coefficients ofvariation were <5 and <7 %, respectively. The lower limit ofdetection was <0.127 pmol/L (<1.20 pg/mL).

Secondary hyperparathyroidism was defined as serumiPTH concentration >6.9 pmol/L (65 pg/mL) and correctedserum calcium <2.55 mmol/L (10.2 mg/dL) (upper limit of themanufacturer’s normal reference range).

We grouped vitamin D values into normal [75 nmol/L(30 ng/mL) or above], vitamin D insufficiency [50–<75 nmol/L (20–<30 ng/mL)], and vitamin D deficiency[<50 nmol/L (<20 ng/mL)], which in turn was classified intomild [25–50 nmol/L (10–<20 ng/mL)], moderate [12.5–<25 nmol/L (5–<10 ng/mL)], and severe [<12.5 nmol/L(<5 ng/mL)] [3, 34].

Data analysis

Statistical analysis was carried out using Predictive AnalyticsSoftware Statistics (PASW Statistics) version 18. Chi-squaretest was used to analyze group differences for categoricalvariables. Normally distributed numeric variables were ana-lyzed using Student’s t test and presented as mean and stan-dard deviation (SD). When the distribution was not normal,medians and quartiles (interquartile range, IQR) were reported

Arch Osteoporos (2014) 9:176 Page 3 of 10, 176

and the differences assessed using Mann–Whitney U test.Nevertheless, we also report means and SD for non-normally distributed variables to make comparisons with oth-er studies possible. A p value less than 0.05 was consideredsignificant.

In the absence of a consensus value defining mild vitaminD deficiency, 25OHD levels were divided into 2.5 nmol/L(1 ng/mL) increments, in order to find the threshold at whichthe median serum iPTH concentration started to increase.

Univariate logistic regression models were used to assessthe odds ratio for vitamin D inadequacy categories for eachpotential risk factor (such as age, sex, BMI, time spent out-doors, season, sunscreen use, multiparity, breastfeeding, edu-cational level, physical activity). Only those with p<0.05wereincluded in a multivariate stepwise logistic regression analy-sis. The results are presented as odds ratios (OR) and 95 %confidence intervals (CI).

Bivariate associations between continuous variables weretested by Pearson’s correlation coefficient or Spearman’s rhotest, as appropriate.

Results

A total of 372 apparently healthy participants with nearlyequal numbers from both genders (males, 184; females, 188)were studied; 183 blood samples were drawn during thewinter months and 189 during the summer months. The mean(SD) age of the participants was 34.09 (9.98) years, rangingfrom 18 to 62 years. Subjects >50 years old constituted 3.8 %of the sample. The main clinical and biochemical characteris-tics of the whole study sample and of each gender are pre-sented in Table 1.

Vitamin D status and its determinants

Regardless of season, almost all study subjects (99.2 %) fellbelow the recommended level of 25OHD ≥75 nmol/L(≥30 ng/mL) (Fig. 2). Only three males (21, 25, and 32 yearsold) in this study reached this recommended level; measure-ment of their vitamin D status was performed in January, May,and September, respectively. About 11.8 % of 25OHD valueswere equal to or below the lower limit of detection and wereassigned 7.5 nmol/L (3 ng/mL). The discrepancy between themean and the median was due to positively skewed distribu-tion of 25OHD values (Table 1).

As shown in Table 1, significant gender-specific differ-ences (p<0.001) have been demonstrated in terms of age, timespent outdoors, physical activity, total alkaline phosphatase,and albumin. In addition, females were on average older andhad significantly lower median 25OHD and higher medianserum iPTH levels than males (p<0.001; Table 1). All studyparticipants consumed Middle Eastern diet, including dairy

products; thus, their answers about consumingmore than threeservings a week were “yes.” Overall, 61 % of study subjectsreported regular physical activity (Table 1); the principalphysical activity practiced among our sample was walking;about half of the participants (50.5 %) walked 30–60 mindaily. The use of sunscreen was reported by 24.2 % of sub-jects: 45.2 % of women and only 2.7 % of men.

In the whole sample, median (IQR) 25OHD levels weresignificantly lower in the subjects older than 35 years (n=164;44%) compared to younger participants ((n=208; 55.9%) [16.7(10.0–32.5) vs. 22.5 (12.3–35.4) nmol/L; p=0.004] [6.7 (4.0–13) vs. 9.0 (4.9–14.1) ng/mL]). Unlike 25OHD, median (IQR)iPTH levels were significantly higher in the older group [6.4(5.1–8.2) vs. 5.2 (4.2–6.7) pmol/L; p<0.001] [60.6 (48.1–77.9)vs. 49.3 (40.3–63.3) pg/mL]. Creatinine clearance was inverselycorrelated with age (Pearson’s coefficient r=−0.435, p<0.001).No significant relationships were observed between age and theother biochemical variables (data not shown).

One hundred thirteen (60.1 %) of the studied women wereveiled. Serum 25OHD level in veiled women was slightlylower than that in non-veiled women (median (IQR) 13.2(10.0–21.6) nmol/L vs. 17.1 (10.0–33.1) nmol/L [5.3 (4.0–8.6) ng/mL vs. 6.8 (4.0–13.2) ng/mL]), and the difference wasstatistically significant (p=0.042). No significant differenceswere found in other clinical and biochemical characteristicsbetween the two women groups (data not shown).

Seasonal differences in serum 25OHD concentrationdepended mainly on sex; 25OHD was significantly higher inmen than women in both seasons, especially during the sum-mer months (p<0.001); that is, among studied men, medianlevel in summer was significantly higher than in winter(p<0.001). In contrast, these seasonal differences were notobserved among women (p=0.54).

When we used a cutoff of 25OHD <25 nmol/L, 227 (61 %)of our subjects fell below this cutoff point (referred to here asmoderate-to-severe vitamin D deficiency) (Fig. 2). The sig-nificant predictors of moderate-to-severe vitaminD deficiencyin univariate analysis were female sex (p<0.001), winterseason (p<0.001), subjects over 35 years old (p=0.011),practicing sport less than three times a week (p=0.022), andeducational level up to secondary (12 grades) (p=0.014). Inthe multivariate model, female sex [adjusted odds ratio (95 %CI) 3.7 (2.4–6.0)] and winter season [adjusted odds ratio(95 % CI) 2.8 (1.8–4.4)] remained significant factors, butphysical activity, age group, and educational level were nolonger significant predictors of moderate-to-severe vitamin Ddeficiency after adjustment for sex and season (Table 2). BMI,whether numerical or categorized into two groups below vs.equal to or greater than 30 kg/m2, had no predictive power ofmoderate-to-severe vitamin D deficiency.

When females were analyzed separately, wearing the hijabwas the only significant predictor of moderate-to-severe vita-min D deficiency in univariate analysis, as 65.3 % of non-

176, Page 4 of 10 Arch Osteoporos (2014) 9:176

Table 1 Participant characteris-tics by gender

Creatinine clearance was stan-dardized for a body surface areaof 1.73 m2 . Body surface area(BSA) was calculated as follows:BSA (m2 )=square root ([height(cm)×weight (kg)]/3,600)

NA not applicable, SDstandard deviation, 25OHD25-hydroxyvitamin D, iPTH in-tact parathyroid hormone, BMIbody mass index, IQR interquar-tile range, M male, F femalea For normally distributed values,the means are compared usingStudent’s t test; when distributionis skewed, medians are reportedand compared using Mann–Whit-ney U test, with p values reportedin the corresponding table rowb Pearson chi-squarec The manufacturer’s normalvalues as reported in the kitd Normal values for creatinine clear-ance=120–130 mL/min/1.73 m2

Parameter(reference values)c

All (n=372) Males(n=184; 49.5 %)

Females(n=188; 50.5 %)

pa

Age (years)

Mean (SD) 34.09 (9.98) 32.20 (9.0) 35.95 (10.55)

Median (IQR) 31 (26–43) 29 (26–40) 37 (26–46) 0.001

Corrected serum calcium (2.15–2.55 mmol/L)

Mean (SD) 2.28 (0.08) 2.28 (0.08) 2.28 (0.08) 0.579

Median (IQR) 2.29 (2.23–2.34) 2.28 (2.23–2.33) 2.29 (2.22–2.35)

Phosphorus (0.81–1.45 mmol/L)

Mean (SD) 1.16 (0.16) 1.15 (0.16) 1.17 (0.15) 0.163

Median (IQR) 1.16 (1.06–1.28) 1.16 (1.04–1.25) 1.19 (1.06–1.29)

Alkaline phosphatase (M 80–270 U/L; F <240 U/L)

Mean (SD) 180.30 (51.58) 189.33 (50.86) 171.42 (50.86)

Median (IQR) 170 (144–209) 182 (151–215) 161 (137–196) <0.001

Albumin (35–52 g/L)

Mean (SD) 47.2 (3.0) 48.5 (2.7) 46.0 (2.8) <0.001

Median (IQR) 47.0 (45.0–49.0) 48.5 (47.0–50.0) 46.0 (44.0–48.0)

Creatinine (M 62–106 μmol/L, F 44–80 μmol/L)

Mean (SD) 77.8 (13.3) 87.5 (9.7) 67.2 (8.0) NA

Median (IQR) 76.9 (66.3–87.5) 87.5 (80.4–96.4) 67.2 (62.8–72.5)

Creatinine clearance (mL/min/1.73 m2)d

Mean (SD) 106.22 (16.43) 107.66 (16.35) 104.83 (16.44)

Median (IQR) 104.03 (94.63–115.35) 104.58 (96.38–116.35) 103.86 (93.21–115.26) 0.120

25OHD (nmol/L)

Mean (SD) 24.7 (16.9) 30.2 (18.3) 19.3 (13.4)

Median (IQR) 20.0 (10.1–33.7) 27.2 (14.7–40.9) 14.4 (10.0–24.3) <0.001

25OHD (nmol/L) in May–October

Mean (SD) 28.7 (18.5) 36.8 (18.6) 20.2 (14.1)

Median (IQR) 25.0 (13.0–40.8) 33.3 (23.2–50.2) 14.4 (10.0–29.8) <0.001

25OHD (nmol/L) in November–April

Mean (SD) 20.5 (13.9) 22.8 (14.8) 18.4 (12.8)

Median (IQR) 16.5 (10.0–26.5) 19.3 (10.7–32.2) 14.4 (9.3–22.5) 0.018

iPTH (pmol/L)

Mean (SD) 6.2 (2.4) 5.3 (1.5) 7.1 (2.7)

Median (IQR) 5.8 (4.5–7.2) 5.0 (4.3–6.3) 6.5 (5.2–8.4) <0.001

Magnesium (0.70–1.05 mmol/L)

Mean (SD) 0.87 (0.12) 0.88 (0.13) 0.84 (0.1) 0.014

Median (IQR) 0.86 (0.8–0.94) 0.87 (0.8–0.95) 0.82 (0.79–0.92)

BMI (kg/m2)

Mean (SD) 26.50 (4.77) 26.68 (4.37) 26.33 (5.15)

Median (IQR) 25.93 (23.04–29.70) 26.15 (12.77–30.07) 25.62 (22.51–29.52) 0.153

Physical activity ≥3 or more times per week

Yes (%) 227 (61) 131(71.2) 96 (51.1) <0.001b

No (%) 145 (39) 53 (28.8) 92 (48.9)

Time spent outdoors (h/week)

≤3 h/week (%) 161 (43.3) 46 (25.0) 115 (61.2) <0.001b

>3 h/week (%) 211 (56.7) 138 (75.0) 73 (38.8)

Educational level

Secondary school orless (%)

99 (28.0) 37 (20.6) 62 (35.6) 0.02b

Higher than secondary(%)

255 (72.0) 143 (79.4) 112 (64.4)

Arch Osteoporos (2014) 9:176 Page 5 of 10, 176

veiled women vs. 82.3 % of veiled women had moderate-to-severe vitamin D deficiency (p=0.008, odds ratio=2.5 (95 %CI 1.3–4.9)).

Prevalence of secondary hyperparathyroidismand its determinants

Secondary hyperparathyroidism [iPTH >6.9 pmol/L(>65 pg/mL)] was observed in 30.4 % of our subjects:17.9 % of men and 42.6 % of women (p<0.001). Otherabnormal biochemical parameters such as hypocalcemia[serum-corrected calcium <2.15 mmol/L (<8.6 mg/dL)] andhypophosphatemia [phosphorus <0.81 mmol/L (<2.5 mg/dL)]were observed only in 23 (6.2 %) and 7 (1.9 %) subjects,respectively. Total alkaline phosphatase was elevated in 19(5.1 %) subjects.

Median (IQR) iPTH levels were higher in women than men(p<0.001, Table 1). In both sexes, there was a trend toward anincrease in iPTH concentration with age (Spearman’s rho=0.266; p<0.001). There was a weak but statistically significantinverse correlation between iPTH and total serum-correctedcalcium (Spearman’s rho=−0.104; p=0.045). No significantassociations were observed between iPTH and the other bio-chemical variables. It should be noted that 25OHD was notsignificantly correlated with any of these biomarkers.

As expected, there was a significant negative correlationbetween iPTH and 25OHD concentrations (Spearman’srho=−0.423, p<0.001).

When serum iPTH values were plotted by 2.5 nmol/L(1 ng/mL) 25OHD strata, to establish whether there are thresh-old values of 25OHD concentration under which iPTH startsto rise, we found by inspecting the plot that iPTH levels

Fig. 2 Distribution of the studypopulation according to the extentof their vitamin D status

Table 2 Predictor factors formoderate-to-severe (25OHD<25 nmol/L) deficiency vs.25OHD ≥25 nmol/L

25OHD 25-hydroxyvitamin D,CI confidence intervala Adjusted for sex and seasonb Sixteen missing values

Serum 25OHD concentration <25 nmol/L=227 cases

Factor n (%) Raw odds ratio(95 % CI)

Adjusted oddsratio (95 % CI)

p

Sex

Females 142/188 (75.5 %) 3.6 (2.3–5.6) <0.001

Male 85/184 (46.2 %)

Season

Winter 133/183 (72.7 %) 2.7 (1.7–4.1) 2.8 (1.8–4.4) <0.001

Summer 94/189 (49.7 %)

Age

≥35 years 112/164 (68.3 %) 1.7 (1.1–2.7) 1.3 (0.8–2.1)a 0.235

<35 years 115/208 (55.3 %)

Educational levelb

Secondary school or less 72/101 (71.3 %) 1.9 (1.1–3.0) 1.5 (0.9–2.5)a 0.149

Higher than secondary 146/255 (57.3 %)

Physical activity

≥3 times/week 128/227 (56.4 %) 1.7 (1.1–2.6) 1.3 (0.8–2.1)a 0.3

<3 times/week 99/145 (68.3 %)

176, Page 6 of 10 Arch Osteoporos (2014) 9:176

increased substantially for 25OHD levels of <27.5 nmol/L(<11 ng/mL) (Fig. 3). This value is close to the breakpointof 25 nmol/L (10 ng/mL), so we decided to keep the divisionat this value to make comparisons with other studies possible.Consequently, the proportion of secondary hyperparathyroid-ism was significantly higher among subjects with serum25OHD concentrations <25 nmol/L compared to those withserum concentration ≥25 nmol/L (41.9 vs. 12.4 %, p<0.001),with median (IQR) iPTH levels significantly higher in thegroup with lower 25OHD concentration (6.4 (5.1–8.2) vs.4.8 (4.0–6.0) pmol/L [60.7 (48.8–78.0) pg/mL vs. 45.3(38.1–57.0) pg/mL], p<0.001).

To determine factors associated with hyperparathyroidism(iPTH >6.9 pmol/L), a binary logistic regression model wasbuilt. The candidate explanatory variables were 25OHD, agegroup, sex, season, BMI, physical activity, time spent outdoors,and serum calcium. In univariate analysis, the significant pre-dictors of hyperparathyroidism were 25OHD <25 nmol/L(p<0.001), sex (p<0.001), age group (p<0.001), time spentoutdoors (p=0.006), and BMI ≥30 kg/m2 (p=0.011). In multi-variate analysis, only moderate-to-severe vitamin D deficiency,sex, and age group remained statistically significant factors foriPTH elevation. The odds ratio (95 % CI) of iPTH elevation insubjects with moderate-to-severe vitamin D deficiency was 3.9(2.2–7.1; p<0.001), in females 2.2 (1.4–3.6; p=0.001), and insubjects over 35 years old 2.1 (1.2–3.7; p=0.006).

Discussion

Subclinical vitamin D deficiency, defined as a serum 25OHDconcentration of <50 nmol/L, was highly prevalent (90.1 %)in our sample, raising a health concern of great importance

among this group of apparently healthy and relatively well-educated Syrian adults living in Damascus at latitude 33.25°north, with sufficient sunshine throughout the year. Mean±SDvitamin D value of our study subjects [24.7±16.9 nmol/L (9.8±6.7 ng/mL)], corresponded to moderate deficiency as classi-fied by Lips [34], which was close to comparable studiesconducted in neighboring countries, e.g., Lebanon withmean±SD levels of 25OHD ranging from 24.2±17.7 to 28±35 nmol/L [35, 36]; similar findings were reported in SaudiArabia where 80 % of studied population had 25OHD below25 nmol/L [23, 24].

In contrast, a recent population-based study from Jordan[37] found surprisingly low prevalence of vitamin D deficien-cy, as only 1.5 % of men and about 14 % of women hadvitamin D levels below 50 nmol/L. But another population-based study of women of child-bearing age in Jordan gaveresults that were more consistent with our results and those ofother studies from the region (96 % of women had vitamin Dlevels below 50 nmol/L) [38].

Compared with data from the USA where vitamin D-fortified foods are available, as reported by the Third NationalHealth and Nutrition Examination Survey [39], or with Euro-pean countries reports, the prevalence of vitamin D deficiencywas substantially higher in our sample using the same cutoffpoint, which reaffirms the belief that vitamin D deficiency ispandemic in our sunnier region [6, 7, 25, 26].

Female gender was one of the most frequently reported riskfactors for hypovitaminosis D [22–24, 35–44]. Indeed, wefound significant differences between genders in vitamin Dlevels in both seasons. In agreement with previous studiesfrom the Middle East and elsewhere, wearing the hijab was asignificant predictor of vitamin D deficiency (<25 nmol/L)among the studied women [22–24, 35, 37, 38, 43–45]. In

Fig. 3 Relation of iPTH to25-hydroxyvitamin D

Arch Osteoporos (2014) 9:176 Page 7 of 10, 176

contrast to some studies, we did not find any effect of otherfemale-specific factors such as multiparity, breastfeeding, andsunscreen use on the prevalence of vitamin D deficiency[22–24, 44]. However, we excluded women who were preg-nant or breastfeeding within the last 3 years.

In agreement with a comparable study from Lebanon, wedid not observe a significant association betweenhypovitaminosis D and increased BMI [35]. Likewise,Binkley et al. did not find such correlation [46], while suchrelationship has been observed in subjects of older age than inour sample by Holick et al. and Rucker et al. [20, 21].However, Snijder et al. [47] reported that total body fat has astrong inverse association with vitamin D, while with anthro-pometric measures, such as BMI, this association is weaker.This may be one reason why we did not find an associationbetween BMI and hypovitaminosis D.

Seasonal differences in serum 25OHD in healthy popula-tions have been documented in several studies worldwide [21,39–43]. Likewise, we found significant seasonal influences onvitamin D status in men but not in women. The greater sexdifference in serum 25OHD concentration in both seasonsimplicates differences in skin synthesis of vitamin D, as menreported spending more time outdoors and had more physicalactivity than women. Lifestyle behaviors, such as shadow-seeking because of the heat, cultural, and religious norms suchas conservative clothing styles are the most likely explanationfor these findings. Actually, in our community, even womenwithout hijab (uncovered head) adopt relatively conservativedress style and wear long sleeves outdoors.

Despite lack of symptoms and the rarity of abnormal bio-chemical parameters such as hypocalcemia andhypophosphatemia in this study, it is likely thathypovitaminosis D has deleterious effects on the skeleton asevidenced by increasing serum iPTH levels, especially inthose with vitamin D deficiency (<25 nmol/L); 41.9 % ofthe latter displayed iPTH values above normal levels. Similarfindings were reported by Saliba et al., Malabanan et al., andSahota et al. [18, 19, 48]. In addition to vitamin D deficiency(<25 nmol/L), female sex and subjects over 35 years oldsignificantly contributed to PTH elevation. This was expectedin view of the reduction of cutaneous production of vitamin Dand increasing iPTH level with aging [2, 3, 34, 41, 46].Therefore, older people need more vitamin D to overcomethe secondary hyperparathyroidism associated with the de-cline in creatinine clearance with aging [13, 41].

The serum 25OHD levels that induce an increase in iPTHsecretion differ among studies, reflecting different demo-graphic and clinical characteristics of the studied subjects,especially in terms of age, sex, genetic factors, renal function,physical activity, calcium intake, phosphate and magnesiumstatus and lifestyle, as well as differences in assay methodol-ogy and in statistical models chosen for curve fitting and itsassumptions [9, 10, 14, 18, 46]. In our study, the limited

number of subjects with serum 25OHD concentrations ex-ceeding 75 nmol/L has not allowed us to find an associationbetween vitamin D and iPTH at higher values of vitamin Dbecause it was not reasonable to expect an iPTH plateau atvitamin D concentrations encountered in our sample. On theother hand, the visual inspection in Fig. 3 reveals that anincrease of iPTH levels at 25OHD concentrations below25 nmol/L supports the division point between mild andmoderate-to-severe vitamin D deficiency as suggested by Lips[34]. Other researchers did not find an association between25OHD and iPTH [21, 46]. Furthermore, the existence of suchplateau is not supported by the literature review conducted bySai et al. and Willett et al. [15, 16]. Aloia et al. argued even ifan inflection point in PTH values does exist, its clinical utilityas a way of identifying optimal values of vitamin D is doubtful[17]. Apparently, the serum concentrations of 25OHD cannotbe predicted by serum iPTH levels in individual subjects withvitamin D deficiency; thus, biochemical evidence of second-ary hyperparathyroidism may not be a good indicator ofvitamin D inadequacy [13, 16, 20, 41, 46, 48].

Low vitamin D synthesis in the skin seems to be the majorrisk factor in our study subjects of both sexes, probably due toinsufficient sun exposure. Atmospheric conditions (such as airpollution) and darker skin which is common in Mediterraneancountries have also been suggested as causes of reducedvitamin D synthesis in the skin [2, 22, 35]. Additionally, it iscrucial to mention that vitamin D inadequacy and the highprevalence of hyperparathyroidism may be due to the fact thatthe average dietary intake of calcium and vitamin D does notmeet the recommended daily level as suggested by previousstudies among the adult Arab population in Middle East [27].The estimated average daily calcium intake per person inSyria, based on 11,566 households, was 769.2±428.0 mg/day, ranging from 131.7 to 3,328.9 mg/day [49], which islower than the recommended daily calcium intake of1,000 mg/day by the Institute of Medicine (IOM) [8]. How-ever, considerable uncertainty exists regarding the healthiestor safest dietary calcium intake between countries; for exam-ple, for adults 19–50 years old, the recommended daily intakeis 700 mg/day in the UK and 1,000 mg/day in the USA.However, further studies are needed to address this issue inour community.

To our knowledge, this is the first attempt at determining25OHD levels in apparently healthy adults in Syria in order toelucidate factors associated with vitamin D deficiency.

There were a number of limitations in our study. The studysubjects were not randomly selected from the general popula-tion but were volunteers who responded to the advertisement.The cross-sectional study design does not allow causal con-clusions to be drawn. An important limitation of our study isthat we did not assess accurately participants’ daily calciumand vitamin D intake.We did not directly measure exposure tosunlight nor the percentage of body uncovered. Finally, the

176, Page 8 of 10 Arch Osteoporos (2014) 9:176

degree of sun exposure and the daily intake of vitamin D andcalcium were obtained from self-reports of dairy productconsumption and might not reflect the true exposure or intake.We did not analyze here the relationship between 25OHD-PTH and bone mineral density, which will be the subject ofanother paper.

Conclusion

Our study highlights the magnitude of hypovitaminosis Damong our sample. In the whole sample, winter season andfemale gender were independent predictors of vitamin Ddeficiency (<25 nmol/L). The only predictor of vitamin Ddeficiency among women was concealing clothing (hijab).

In turn, vitamin D deficiency (25OHD <25 nmol/L), age,and female gender were independent predictors of iPTH ele-vation. While our results cannot be generalized to the Syrianpopulation, it may still offer an insight into vitamin D status inSyria.

Further studies are warranted to measure the magnitude ofthe problem and risk factors of hypovitaminosis D in differentage and social and ethnic groups of the Syrian population.This is especially important in early adulthoodwhere adequatevitamin D status is vital for optimizing peak bone mass andtherefore reducing the risk of osteoporosis later in life. Ifconducted, such community-based research may lead to pub-lic health measures such as food fortification and guidelineson vitamin D supplementation to correct vitamin D inadequa-cy in any vulnerable subgroups or the whole population, inorder to avoid the potentially harmful skeletal andextraskeletal consequences of this silent epidemic.

Acknowledgments The authors are grateful to Prof. Hyam Bashour forcritically reading the manuscript. We acknowledge that this work wasfunded by Damascus University.

Conflicts of interest None

References

1. Peterlik M, Boonen S, Cross HS, Lamberg-Allardt C (2009) VitaminD and calcium insufficiency-related chronic diseases: an emergingworld-wide public health problem. Int J Environ Res Public Health 6:2585–2607

2. Hossein-nezhad A, Holick MF (2013) Vitamin D for health: a globalperspective. Mayo Clin Proc 88:720–755

3. Holick M (2007) Vitamin D deficiency. N Engl J Med 357:266–2814. Souberbielle JC, Body JJ, Lappe JM, Plebani M, Shoenfeld Y, Wang

TJ, Bischoff-Ferrari HA, Cavalier E, Ebeling PR, Fardellone P,Gandini S, Gruson D, Guérin AP, Heickendorff L, Hollis B, Ish-Shalom S, Jean G, von Landenberg P, Largura A, Olsson T, Pierrot-Deseilligny C, Pilz S, Tincani A, Valcour A, Zittermann A (2010)Vitamin D and musculoskeletal health, cardiovascular disease,

autoimmunity and cancer: recommendations for clinical practice.Autoimmun Rev 9:709–715

5. Pearce SH, Cheetham TD (2010) Diagnosis and management ofvitamin D deficiency. BMJ 340:b5664

6. Rizzoli R, Boonen S, Brandi ML, Bruyère O, Cooper C, Kanis JA,Kaufman JM, Ringe JD, Weryha G, Reginster JY (2013) Vitamin Dsupplementation in elderly or postmenopausal women: a 2013 updateof the 2008 recommendations from the European Society for Clinicaland Economic Aspects of Osteoporosis and Osteoarthritis (ESCEO).Curr Med Res Opin 29:305–313

7. Lips P, Hosking D, Lippuner K, Norquist JM, Wehren L, Maalouf G,Ragi-Eis S, Chandler J (2006) The prevalence of vitamin D inade-quacy amongst women with osteoporosis: an international epidemi-ological investigation. J Intern Med 260:245–254

8. Ross AC, Manson JE, Abrams SA, Aloia JF, Brannon PM, ClintonSK, Durazo-Arvizu RA, Gallagher JC, Gallo RL, Jones G, KovacsCS, Mayne ST, Rosen CJ, Shapses SA (2011) The 2011 report ondietary reference intakes for calcium and vitamin D from the Instituteof Medicine: what clinicians need to know. J Clin Endocrinol Metab96:53–58

9. Holick MF, Binkley NC, Bischoff-Ferrari HA, Gordon CM, HanleyDA, Heaney RP, Murad MH, Weaver CM (2012) Guidelines forpreventing and treating vitamin D deficiency and insufficiencyrevisited. J Clin Endocrinol Metab 97:1153–1158

10. Bischoff-Ferrari HA, Giovannucci E, Willett WC, Dietrich T,Dawson-Hughes B (2006) Estimation of optimal serum concentra-tions of 25-hydroxyvitamin D for multiple health outcomes. Am JClin Nutr 84:18–28

11. Dawson-Hughes B, Heaney RP, Holick MF, Lips P, Meunier PJ,Vieth R (2005) Estimates of optimal vitamin D status. OsteoporosInt 16:713–716

12. Bischoff-Ferrari HA, Keller U, Burckhardt P, Quack Lötscher K,Gerber B, l’Allemand D, Laimbacher J, Bachmann M, Rizzoli R(2012) Recommandations de la Commission fédérale d’alimentationconcernant l’apport de vitamine D. Forum Med Suisse 12:775–778

13. Vieth R, Ladak Y, Walfish PG (2003) Age-related changes in the 25-hydroxyvitamin D versus parathyroid hormone relationship suggest adifferent reason why older adults require more vitamin D. J ClinEndocrinol Metab 88:185–191

14. Fuleihan GE, Rahme M, Bassil D (2013) Do desirable vitamin Dlevels vary globally? In: Burckhardt P, Dawson-Hughes B,Weaver C(eds) Influences on bone health. Springer, pp 273–299

15. Sai AJ, Walters RW, Fang X, Gallagher JC (2011) Relationshipbetween vitamin D, parathyroid hormone, and bone health. J ClinEndocrinol Metab 96:E436–E446

16. Willett AM (2005) Vitamin D status and its relationship with para-thyroid hormone and bone mineral status in older adolescents. ProcNutr Soc 64:193–203

17. Aloia JF, Talwar SA, Pollack S, FeuermanM,Yeh JK (2006) Optimalvitamin D status and serum parathyroid hormone concentrations inAfrican American women. Am J Clin Nutr 84:602–609

18. Saliba W, Barnett O, Rennert HS, Lavi I, Rennert G (2011) Therelationship between serum 25(OH)D and parathyroid hormonelevels. Am J Med 124:1165–1170

19. Malabanan A, Veronikis IE, HolickMF (1998) Redefining vitamin Dinsufficiency. Lancet 351:805–806

20. Holick MF, Siris ES, Binkley N, Beard MK, Khan A, Katzer JT,Petruschke RA, Chen E, de Papp AE (2005) Prevalence of vitamin Dinadequacy among postmenopausal North American women receiv-ing osteoporosis therapy. J Clin Endocrinol Metab 90:3215–3224

21. Rucker D, Allan JA, Fick GH, Hanley DA (2002) Vitamin D insuf-ficiency in a population of healthy western Canadians. Can MedAssoc J 166:1517–1524

22. Arabi A, El Rassi R, El-Hajj Fuleihan G (2010) Hypovitaminosis Din developing countries-prevalence, risk factors and outcomes. NatRev Endocrinol 6:550–561

Arch Osteoporos (2014) 9:176 Page 9 of 10, 176

23. Fuleihan GE (2009) Vitamin D deficiency in the Middle East and itshealth consequences for children and adults. Clinic Rev Bone MinerMetab 7:77–93

24. Mithal A, Wahl DA, Bonjour JP, Burckhardt P, Dawson-Hughes B,Eisman JA, El-Hajj Fuleihan G, Josse RG, Lips P, Morales-Torres J,IOF Committee of Scientific Advisors (CSA) Nutrition WorkingGroup (2009) Global vitamin D status and determinants ofhypovitaminosis D. Osteoporos Int 20:1807–1820

25. Lips P (2010) Worldwide status of vitamin D nutrition. J SteroidBiochem Mol Bio 121:297–300

26. Wahl DA, Cooper C, Ebeling PR, Eggersdorfer M, Hilger J,Hoffmann K, Josse R, Kanis JA, Mithal A, Pierroz DD, StenmarkJ, Stöcklin E, Dawson-Hughes B (2012) A global representation ofvitamin D status in healthy populations. Arch Osteoporos 7:155–172

27. Musaiger AO, Hassan AS, Obeid O (2011) The paradox of nutrition-related diseases in the Arab countries: the need for action. Int JEnviron Res Public Health 8:3637–3671

28. El-Hajj Fuleihan G, Baddoura R, Awada H, Salam N, Salamoun M,Rizk P (2002) Low peak bone mineral density in healthy Lebanesesubjects. Bone 31:520–528

29. Department of Health and Human Services (US) (2004) The 2004Surgeon General’s Report on bone health and osteoporosis: what itmeans to you. Washington: HHS, Office of the Surgeon General; p 7.http://www.niams.nih.gov/Health_Info/Bone/SGR/SGRBoneHealth_Eng.pdf

30. World Health Organization (WHO) (2003) GEMS/FOOD RegionalDiets: regional per capita consumption of raw and semi-processedagricultural commodities. Prepared by the Global EnvironmentMonitoring System/Food Contamination Monitoring andAssessment Programme (GEMS/Food). Rev. ed. http://www.who.int/foodsafety/publications/chem/regional_diets/en/

31. Figge J, Jabor A, Kazda A, Fencl V (1998) Anion gap and hypoal-buminemia. Crit Care Med 26:1807–1810

32. Cockcroft DW, Gault MH (1976) Prediction of creatinine clearancefrom serum creatinine. Nephron 16:31–41

33. Mosteller RD (1987) Simplified calculation of body surface area. NEngl J Med 317:1098

34. Lips P (2001) Vitamin D deficiency and secondary hyperparathy-roidism in the elderly: consequences for bone loss and fractures andtherapeutic implications. Endocr Rev 22:477–501

35. Gannagé-Yared MH, Chemali R, Yaacoub N, Halaby G (2000)Hypovitaminosis D in a sunny country: relation to lifestyle and bonemarkers. J Bone Miner Res 15:1856–1862

36. Fuleihan GE, DeebM (1999) Hypovitaminosis D in a sunny country.N Engl J Med 340:1840–1841

37. Batieha A, Khader Y, Jaddou H, Hyassat D, Batieha Z, Khateeb M,Belbisi A, Ajlouni K (2011) Vitamin D status in Jordan: dress styleand gender discrepancies. Ann Nutr Metab 58:10–18

38. Nichols EK, Khatib IM, Aburto NJ, Sullivan KM, Scanlon KS,WirthJP, Serdula MK (2012) Vitamin D status and determinants of defi-ciency among non-pregnant Jordanian women of reproductive age.Eur J Clin Nutr 66:751–756

39. Looker AC, Dawson-Hughes B, Calvo MS, Gunter EW, SahyounNR (2002) Serum 25-hydroxyvitamin D status of adolescents andadults in two seasonal subpopulations from NHANES III. Bone 30:771–777

40. van Der Mei IA, Ponsonby AL, Engelsen O, Pasco JA, McGrath JJ,Eyles DW, Blizzard L, Dwyer T, Lucas R, Jones G (2007) The highprevalence of vitamin D insufficiency across Australian populationsis only partly explained by season and latitude. Environ HealthPerspect 115:1132–1139

41. Ono Y, Suzuki A, Kotake M, Zhang X, Nishiwaki-Yasuda K,Ishiwata Y, Imamura S, Nagata M, Takamoto S, Itoh M (2005)Seasonal changes of serum 25-hydroxyvitamin D and intact parathy-roid hormone levels in a normal Japanese population. J Bone MinerMetab 23:147–151

42. Dawson-Hughes B, Harris SS, Dallal GE (1997) Plasma calcidiol,season, and serum parathyroid hormone concentrations in healthyelderly men and women. Am J Clin Nutr 65:67–71

43. Levis S, Gomez A, Jimenez C, Veras L, Ma F, Lai S, Hollis B, RoosBA (2005) Vitamin D deficiency and seasonal variation in an adultSouth Florida population. J Clin Endocrinol Metab 90:1557–1562

44. Meddeb N, Sahli H, Chahed M, Abdelmoula J, Feki M, Salah H,Frini S, Kaabachi N, Belkahia C, Mbazaa R, Zouari B, Sellami S(2005) Vitamin D deficiency in Tunisia. Osteoporos Int 16:180–183

45. Glerup H, Mikkelsen K, Poulsen L, Hass E, Overbeck S, Thomsen J,Charles P, Eriksen EF (2000) Commonly recommended daily intakeof vitamin D is not sufficient if sunlight exposure is limited. J InternMed 247:260–268

46. Binkley N, Novotny R, Krueger D et al (2007) Low vitamin D statusdespite abundant sun exposure. J Clin Endocrinol Metab 92:2130–2135

47. Snijder MB, van Dam RM, Visser M, Deeg DJ, Dekker JM, BouterLM, Seidell JC, Lips P (2005) Adiposity in relation to vitamin Dstatus and parathyroid hormone levels: a population-based study inolder men and women. J Clin Endocrinol Metab 90:4119–4123

48. Sahota O, Mundey MK, San P, Godber IM, Lawson N, Hosking DJ(2004) The relationship between vitamin D and parathyroid hor-mone: calcium homeostasis, bone turnover, and bone mineral densityin postmenopausal women with established osteoporosis. Bone 35:312–319

49. Wendler C, von Cramon-Taubadel S, de Haen H, Padilla Bravo CA,Jrad S (2011) Food security in Syria: preliminary results based on the2006/07 expenditure survey. Discussion papers. http://ageconsearch.umn.edu/bitstream/115537/2/Food%20security%20in%20Syria.pdf

176, Page 10 of 10 Arch Osteoporos (2014) 9:176