Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2019 Jun 13.
Published in final edited form as: Appl Physiol Nutr Metab. 2017 Oct 20;43(3):259–265. doi: 10.1139/apnm-2017-0334

Investigation of the C-3-epi-25(OH)D3 of 25-hydroxyvitamin D3 in urban schoolchildren

Samantha E Berger 1, Maria I Van Rompay 2, Catherine M Gordon 3, Elizabeth Goodman 4, Misha Eliasziw 5, Michael F Holick 6, Jennifer M Sacheck 7
PMCID: PMC6563605  NIHMSID: NIHMS1032003  PMID: 29053944

Abstract

The physiological relevance C-3 epimer of 25-hydroxyvitamin D (3-epi-25(OH)D) is not well understood among youth. The objective of this study was to assess whether demographic/physiologic characteristics were associated with 3-epi-25(OH)D3 concentrations in youth. Associations between 3-epi-25(OH)D3 and demographics and between 3-epi-25(OH)D3, total 25-hydroxyvitamin (25(OH)D) (25(OH)D2 + 25(OH)D3), total cholesterol, high-density lipoprotein, low-density lipoprotein, and triglycerides were examined in racially/ethnically diverse schoolchildren (n = 682; age, 8–15 years) at Boston-area urban schools. Approximately 50% of participants had detectable 3-epi-25(OH)D3 (range 0.95−3.95 ng/mL). The percentage of 3-epi-25(OH)D3 of total 25(OH)D ranged from 2.5% to 17.0% (median 5.5%). Males were 38% more likely than females to have detectable 3-epi-25(OH)D3 concentrations. Both Asian and black race/ethnicity were associated with lower odds of having detectable 3-epi-25(OH)D3 compared with non-Hispanic white children (Asian vs. white, odds ratio (OR) 0.28, 95% confidence interval (CI) 0.14–0.53; black vs. white, OR 0.38, 95%CI 0.23–0.63, p < 0.001). Having an adequate (20–29 ng/mL) or optimal (>30 ng/mL) 25(OH)D concentration was associated with higher odds of having detectable 3-epi-25(OH)D3 than having an inadequate (<20 ng/mL) concentration (OR 4.78, 95%CI 3.23–6.94 or OR 14.10, 95%CI 7.10–28.0, respectively). There was no association between 3-epi-25(OH)D3 and blood lipids. However, when considering 3-epi-25(OH)D3 as a percentage of total 25(OH)D, total cholesterol was lower in children with percent 3-epi-25(OH)D3 above the median (mean difference −7.1 mg/dL, p = 0.01). In conclusion, among schoolchildren, sex, race/ethnicity, and total serum 25(OH)D concentration is differentially associated with 3-epi-25(OH)D. The physiological relevance of 3-epi-25(OH)D3 may be related to the 3-epi-25(OH)D3 as a percentage of total 25(OH)D and should be considered in future investigations.

Keywords: 25-hydroxyvitamin D3, C-3 epimer, vitamin D, Daily D Health Study

Introduction

Low concentrations of 25-hydroxyvitamin D (25(OH)D), the standard biomarker of vitamin D status, have been associated with increased cardiovascular disease risk in adolescents and adults (Giovannucci 2008; Reis et al. 2009). According to the National Health and Nutrition Examination Survey (NHANES) 2001–2006, approximately 6 million children in the United States were vitamin D deficient, defined as a serum 25(OH)D concentration less than 20 ng/mL (Looker et al. 2011). Given the high prevalence of vitamin D deficiency in children and that children with cardiovascular risk factors are at higher risk for cardiovascular disease in adulthood, it is critical to understand the relationships that exist between serum 25(OH)D and cardiovascular risk factors among youth when early prevention is possible (Bao et al. 1997).

Vitamin D status is characterized by serum concentrations of 25(OH)D3 because it has a long half-life, is not tightly regulated compared with 1,25-dihydroxyvitamin D (1,25(OH)2D) (the active form), and is considered an accurate representation of vitamin D intake from diet, supplements, and synthesis in the skin from exposure to ultraviolet B radiation (Institute of Medicine (IOM) 2011; van den Ouweland et al. 2013). It is now known that 25(OH)D undergoes epimerization during metabolism, and the 3-epi-25(OH)D3 of 25(OH)D3 is most prevalent, forming 3-epi-25-hydroxyvitamin D3 (3-epi-25(OH)D) (Singh et al. 2006). Significant blood concentrations of the 3-epi-25(OH)D3 have been found in infants (Singh et al. 2006; Stepman et al. 2011; Granado-Lorencio et al. 2012), and, more recently, in adults (Lensmeyer et al. 2012; Engelman et al. 2014; Cashman et al. 2014; Lutsey et al. 2015). However, the 3-epi-25(OH)D3 is not routinely measured, and without chromatographic resolution, it could be mistakenly reported as 25(OH)D3 because of their identical molecular weights (Lensmeyer et al. 2006).

Laboratory studies have found potential physiological relevance of 3-epi-25(OH)D3 with some binding capacity to the vitamin D receptor (Kamao et al. 2005), but limited data exist on its relationship with anthropometrics and blood lipids. Studies in adults found that being female (Cashman et al. 2014; Lutsey et al. 2015) and having a higher waist circumference (Engelman et al. 2014; Cashman et al. 2014) were negative determinants of 3-epi-25(OH)D3 (Cashman et al. 2014). One study examining 3-epi-25(OH)D, anthropometrics, and blood lipids, separately among black and white adults, found higher concentrations of 3-epi-25(OH)D3 were associated with a better cardiovascular risk profile (Lutsey et al. 2015). Furthermore, the relationship among 3-epi-25(OH)D3, anthropometrics, and blood lipids in children is especially unclear. Research is needed to understand the potential physiological relevance of the 3-epi-25(OH)D3 to determine if it is necessary to consider the 3-epi-25(OH)D3 when characterizing vitamin D status. To understand the significance of the 3-epi-25(OH)D3 in children, we examined the 3-epi-25(OH)D3 cross-sectionally among a diverse sample of schoolchildren in the Boston area. Our goals were to (i) determine if demographic characteristics were associated with 3-epi-25(OH)D3 concentrations; (ii) examine associations between 3-epi-25(OH)D3 concentration and blood lipids; and (iii) identify the impact of considering 3-epi-25(OH)D3 on vitamin D status classification.

Materials and methods

Participants

This cross-sectional study investigated 3-epi-25(OH)D3 using data drawn from 682 children enrolled in the Daily D Health Study. A detailed description of the recruitment and design of the Daily D Health Study has been published elsewhere (Sacheck et al. 2015). Briefly, the Daily D Health Study was a randomized, double-blind trial assessing the impact of 6-month vitamin D supplementation on serum 25(OH)D and cardiometabolic risk factors in Boston-area schoolchildren (8–15 years). The Daily D study protocol was approved by Tufts University’s Institutional Review Board. The ClinicalTrials.gov registration number is NCT01537809.

Demographics, anthropometrics, and vitamin D intake

Birth date, race/ethnicity, and eligibility for free or reduced-price school meals, as a proxy measure of socioeconomic status, were collected from parents during informed consent procedures. The mean age was 11.2 ± 1.4 years, and age was categorized into a binary variable (≥12 years old vs. <12 years old) for analysis. Racial/ethnic categories were grouped as non-Hispanic white, Hispanic/Latino, black/African American, Asian/Asian American, and multiracial/other. Height and weight were measured and body mass index (BMI) calculated as previously reported (Centers for Disease Control and Prevention (CDC) 2013, Sacheck et al. 2015). Weight status was categorized into obese (≥95th percentile) and nonobese (<95th percentile) categories. Pubertal status was assessed using a brief pubertal questionnaire designed and validated for this age group as previously described (Carskadon and Acebo 1993; Sacheck et al. 2015). Dietary intake of vitamin D was assessed using the 2004 Block Food Frequency Questionnaire for Children, which inquired about intake over the past week (NutritionQuest, Berkeley, Calif., USA).

25(OH)D and 3-epi-25(OH)D3 measurements and blood lipids

Blood was drawn from the antecubital vein following an overnight fast. Total serum 25(OH)D was measured by validated liquid chromatography-mass spectrometry (LC-MS/MS) method including fractionation of 25(OH)D2, 25(OH)D3, and the 3-epi-25(OH)D3 of 25(OH)D3 (Holick et al. 2005). Serum samples of 25(OH)D were prepared and analyzed through a turbulent flow LC system (Cohesive Technologies, Franklin, Mass., USA) followed by traditional laminar flow chromatography and analyzed relative to the National Institute of Standards and Technology vitamin D standard for detection and quantification of 25(OH)D3, 25(OH)D2, and 3-epi-25(OH)D3. The analysis was performed using a Triple Stage Quadrupole Quantum Ultra triple mass-spectrometer (Thermo Finnigan Corp., San Jose, Calif., USA). The intra-assay coefficient of variation is 6.0%. Concentrations of 25(OH)D <20 ng/mL were classified as inadequate, 20–29 ng/mL as adequate, and ≥30 as optimal (Dawson-Hughes et al. 2005; IOM 2011). Concentrations of 3-epi-25(OH)D3 were considered detectable at ≥0.95 ng/mL based on assay sensitivity. Total cholesterol, low-density lipoprotein (LDL) cholesterol, high-density lipoprotein (HDL) cholesterol, and triglycerides and were measured as previously described (Van Rompay et al. 2015).

Statistical analysis

Pearson’s χ2 test was used to compare the distribution of participant characteristics across groups by 3-epi-25(OH)D3 status (detectable vs. nondetectable). Statistically significant differences between groups in continuous variables were determined with Student’s independent t test. For variables with skewed distributions, the Kruskal–Wallis test was used. To calculate 3-epi-25(OH)D3 as a percent of total 25(OH)D, we added concentrations of 25(OH)D2 and 25(OH)D3 and the 3-epi-25(OH)D3 of 25OHD3, divided 3-epi-25(OH)D3 concentration by the summed total 25(OH)D, and multiplied the result by 100. Logistic regression was used to examine the relationship between participant characteristics and the odds of having detectable 3-epi-25(OH)D3, as well as the odds of being above or below the median of percent 3-epi-25(OH)D3 (5.5%). Multiple linear regression was used to examine linear relationships between 3-epi-25(OH)D3 concentration and blood lipids, adjusting for participant characteristics (age, sex, race, weight status, puberty status, and free/reduced lunch eligibility). The distribution of 3-epi-25(OH)D3 and triglycerides were right-skewed and logarithmically transformed for this analysis.

Generalized linear models were used to examine the mean difference in blood lipids between children with and without detectable 3-epi-25(OH)D3 and children above or at or below median of percent 3-epi-25(OH)D3 and each was adjusted for participant characteristics and vitamin D status. Least-square means and standard errors were reported for total, LDL, and HDL cholesterol. P values from multiple comparisons were adjusted with Tukey’s Honestly Significant Difference test. Because triglyceride data were skewed, results are reported as geometric mean and standard error, as well as ratio of geometric means and 95% confidence interval.

To assess the impact of including 3-epi-25(OH)D3 in calculating total 25(OH)D concentration on vitamin D status classification, we compared the proportions of children with inadequate, adequate, and optimal 25(OH)D concentrations with and without consideration of 3-epi-25(OH)D3. Agreement between the 2 methods was evaluated with weighted kappa statistics. We also examined the relationship between 3-epi-25(OH)D3 and 25(OH)D as a percent of total 25(OH)D. Spearman’s rank correlation was used to determine the correlation between 3-epi-25(OH)D3 and percent 3-epi-25(OH)D3 and 25(OH)D3 concentrations. A P value of <0.05 was considered statistically significant for all analyses. Data were analyzed using SAS version 9.3 (SAS Institute, Cary, N.C., USA).

Results

Participant characteristics

Overall, 55% of children were <12 years old; 49% were male; 41% were non-Hispanic white, 15% black/African American, 22% Hispanic/Latino, 8% Asian/Asian American, and 14% multiracial/other; 64% had reached puberty; 67% were eligible for free or reduced/priced lunch; 25% were classified as obese. Mean 25(OH)D concentration was 21.9 ± 6.8 ng/mL. Of 682 children in this study, 42% had inadequate, 47% had adequate, and 11% had optimal 25(OH)D concentrations. Mean total cholesterol was 156.5 ± 26.8 mg/dL, mean HDL cholesterol was 50.5 ± 12.2 mg/dL, mean LDL cholesterol was 88.0 ± 23.8 mg/dL, and median triglycerides were 65.0 ± 40.0 mg/dL.

Detectable versus nondetectable 3-epi-25(OH)D3

Half of the children (337 of 682) had detectable 3-epi-25(OH)D3 concentrations. Of those with detectable 3-epi-25(OH)D3, the median 3-epi-25(OH)D3 value was 1.4 ng/mL (range 0.95–3.95 ng/mL; Fig. 1A). Participant characteristics by detectable 3-epi-25(OH)D3 status are summarized in Table 1. The proportion of children classified within each level of vitamin D status was significantly different between those with and without detectable 3-epi-25(OH)D3 (p < 0.001). Among children with inadequate vitamin D concentration, 75% did not have detectable 3-epi-25(OH)D3. Among children with adequate and optimal vitamin D concentrations, 62% and 83% had detectable 3-epi-25(OH)D3, respectively. Dietary vitamin D was greater in children with detectable 3-epi-25(OH)D3 (median 120 IU vs. 103 IU, p < 0.005). Among children with detectable 3-epi-25(OH)D3 there was a significantly higher proportion of males (p = 0.02), prepubertal children (p = 0.008), and non-Hispanic whites (p < 0.0001), and fewer black/African American and Asian/Asian American children than whites. There were no significant differences in age, weight status, or free or reduced-price lunch eligibility between children with and without detectable 3-epi-25(OH)D3 concentrations.

Fig. 1.

Fig. 1.

Open in a new tab

(A) Distribution of 3-epimer of 25-hydroxyvitamin D (3-epi-25(OH)D3) among racially/ethnically diverse schoolchildren. (B) Distribution of percentage of 3-epi(25(OH)D3. One child not included in the figure had a 3-epimer concentration of 12.3 ng/mL and a percentage of 3-epi-25(OH)D of 37.6%.

Table 1.

Subject characteristics by detectable epimer status and logistic regression for characteristics associated with odds of having detectable 3-epimer (n = 682).

Detectable versus nondetectable Odds of having detectable 3-epimer
Subject characteristics Detectable 3-epimer, N = 337; n (row %) Nondetectable 3-epimer, N = 345; n (row %) P Model 1a; OR (95% CI) Model 2b; OR (95% CI)
Vitamin D status (ng/mL) <0.001
 Inadequate (<20) 70 (25) 209 (75) 1.00 1.00
 Adequate (20–29) 203 (62) 123 (38) — 4.78 (3.23, 6.94)
 Optimal (≥30) 64(83) 13 (17) — 14.10 (7.10, 28.0)
Vitamin D intake (IU)* 120 (70,199) 103 (56, 163) 0.005† — —
Age (y) 0.07
 ≥12 139 (46) 166 (54) 1.00 1.00
 <12 198 (53) 179 (47) 1.15 (0.81, 1.64) 1.08 (0.74, 1.58)
Sex 0.02
 Female 156 (45) 192 (55) 1.00 1.00
 Male 181 (54) 153 (46) 1.38 (1.01, 1.89) 1.18 (0.79, 1.57)
Race/ethnicity <0.001
 White/Caucasian 162 (57) 120 (43) 1.00 1.00
 Black/African American 33 (34) 65 (66) 0.38 (0.23, 0.63) 0.72 (0.41, 1.27)
 Hispanic/Latino 74 (49) 77 (51) 0.73 (0.53, 0.99) 1.05 (0.66, 1.67)
 Asian/Asian American 15 (27) 41 (73) 0.28 (0.14, 0.53) 0.45 (0.22, 0.91)
 Multiracial/other 53 (56) 42 (44) 0.96 (0.60, 1.56) 1.38 (0.81, 2.35)
Puberty status 0.008
 Prepubertal 233 (53) 205 (47) 1.00 1.00
 Pubertal 104 (43) 140 (57) 1.31(0.91,1.89) 1.25 (0.83, 1.86)
Weight status 0.25
 Nonobese 243 (48) 262 (52) 1.00 1.00
 Obese 94 (53) 83 (47) 1.09 (0.76,1.55) 1.38 (0.93, 2.03)
Free or reduced-price lunch eligibility 0.48
 Not eligible 113 (51) 107 (49) 1.00 1.00
 Eligible 224 (49) 238 (51) 1.17 (0.82, 1.67) 1.26 (0.86, 1.86)
Open in a new tab

Note: Bold p values and ORs are statistically significant (p < 0.05). CI, confidence interval; OR, odds ratio.

a

Model 1: Adjusted for age, sex, race, weight status, puberty status, and free/reduced lunch eligibility.

b

Model 2: Model 1 + vitamin D status.

*

Median (interquartile range) is reported for vitamin D intake.

†

Kruskal–Wallis test used for statistical significance between group medians.

In the multivariate analysis, male participants (odds ratio (OR) 1.38, 95% confidence interval (CI) 1.01−1.89) had significantly higher odds of having detectable 3-epi-25(OH)D3 concentrations than female participants (Table 1). Asian or black participants had lower odds of having detectable 3-epi-25(OH)D3 (Asian/Asian American vs. white, OR 0.28, 95%CI 0.14–0.53; black/African American vs. white, OR 0.38, 95%CI 0.23−0.63). After adjustment for vitamin D status, the associations between detectable 3-epi-25(OH)D3 and sex as well as race were attenuated. Only vitamin D status was significantly associated with higher odds of having detectable 3-epi-25(OH)D3 (adequate vs. inadequate: OR 4.78, 95%CI 3.23–6.94; optimal vs. inadequate: OR 14.10, 95%CI 7.10–28.0).

Among those with detectable 3-epi-25(OH)D3, continuous 3-epi-25(OH)D3 was not associated with age, sex, weight status, or puberty status. Median 3-epi-25(OH)D3 was significantly higher in children without free or reduced-price lunch eligibility compared with children with free or reduced-price lunch eligibility (median [interquartile range]: 1.50 [0.61] ng/mL vs. 1.35 [0.63] ng/mL, p = 0.049). 3-epi-25(OH)D3 was also associated with race/ethnicity, in that white/Caucasian children had higher median 3-epi-25(OH)D3 compared with all other race/ethnicity groups (1.60 [0.61] ng/mL vs. 1.17−1.36 [0.29−0.69] ng/mL, p = 0.001).

Median percent 3-epi-25(OH)D3 of 25(OH)D

Among children with detectable 3-epi-25(OH)D3, the median percent 3-epi-25(OH)D3 of 25(OH)D was 5.5%, ranging from 2.5%–17.0% (Fig. 1B). There were differences in percent 3-epi-25(OH)D3 by vitamin D status. Specifically, among children with inadequate 25(OH)D, 79% were above the median percent 3-epi-25(OH)D3, while among those with optimal 25(OH)D, only 31% were above the median. The proportion of females above the median was significantly higher than males (54% vs. 38%, respectively, p = 0.002), but no other differences with participant characteristics were found. Multivariate logistic regression results were similar, with males being 50% less likely to have percent 3-epi-25(OH)D3 above the median compared with females (OR 0.50, 95%CI 0.32−0.78), and those with adequate or optimal 25(OH)D less likely to be above the median (adequate vs. inadequate: OR 0.22, 95%CI 0.11–0.42; optimal vs. inadequate: OR 0.10, 95%CI 0.05−0.24). With additional adjustment for vitamin D status, sex retained significance. Similarly to the median analysis, continuous percent 3-epi-25(OH)D3 was associated with sex. Females had significantly higher percent 3-epi-25(OH)D3 than males (mean (SD): 6.16 (2.0) % vs. 5.77 (1.9) %, p = 0.033).

3-epi-25(OH)D3, percent 3-epi-25(OH)D3 and blood lipids

We did not find a linear relationship between 3-epi-25(OH)D3 and blood lipids (data not shown). In comparing blood lipid concentrations between children with detectable 3-epi-25(OH)D3 and nondetectable 3-epi-25(OH)D3, we found no statistically significant differences in mean total cholesterol, LDL, HDL, or triglycerides. However, the adjusted mean difference in total cholesterol between children above the median percent 3-epi-25(OH)D3 was 7 mg/dL lower thanthoseatorbelowthemedian(149.6±2.8mg/dLvs.156.7±2.6;mean difference: −7.1, 95%CI −12.7−1.5 mg/dL; p = 0.01).

Relationship between 3-epi-25(OH)D3 and total 25(OH)D

The relationship between 3-epi-25(OH)D3 and total 25(OH)D did not appear linear, with large variation in 3-epi-25(OH)D3, particularly at higher concentrations of total 25(OH)D; nonetheless, 3-epi-25(OH)D3 concentration was positively correlated with 25(OH)D3 (Spearman’s correlation coefficient, r = 0.40; p < 0.05; Fig. 2A). The relationship between percent 3-epi-25(OH)D3 and total 25(OH)D was also nonlinear, with stable variability in percent 3-epi-25(OH)D3 across increasing concentrations of total 25(OH)D (Fig. 2B). Percent 3-epi-25(OH)D3 of total 25(OH)D was negatively correlated with total 25(OH)D (Spearman’s correlation coefficient, r = −0.40; p < 0.05).

Fig. 2.

Fig. 2.

Open in a new tab

(A) Scatter plot of 3-epimer of 25-hydroxyvitamin D (3-epi-25(OH)D3) and 25-hydroxyvitamin D (25(OH)D3). (B) Scatterplot of percentage of 3-epi-25(OH)D3 and 25(OH)D3. Individual with 3-epimer of 12.3 ng/mL was not included in graph.

3-epi-25(OH)D3 and vitamin D status

Table 2 illustrates reclassification of children when considering the 3-epi-25(OH)D3 in total 25(OH)D. Good agreement was found between the 2 vitamin D status classification methods ([H9260] = 0.78, 95%CI 0.72–0.84). However, among the 70 children with inadequate vitamin D without 3-epi-25(OH)D3 in total 25(OH)D, 33% were reclassified from inadequate to adequate after addition of 3-epi-25(OH)D3 to total 25(OH)D. Similarly, among the 203 children with adequate vitamin D levels without addition of 3-epi-25(OH)D3, 14% were reclassified from adequate to optimal. Overall, 15% of children with detectable 3-epi-25(OH)D3 concentrations were reclassified to a higher vitamin D status when 3-epi-25(OH)D3wasaddedto total 25(OH)D.

Table 2.

Total number of children with detectable 3-epimer classified as inadequate, adequate, or optimal for serum 25(OH)D status with and without inclusion of the 3-epimer of 25(OH)D (N = 337).*

Total 25(OH)D without 3-epimer Inclusion of 3-epimer in total 25(OH)D
Inadequate Adequate Optimal
Inadequate, n = 70 47 (67%) 23 (33%) 0 (0%)
Adequate, n = 203 0(0%) 175 (86%) 28 (14%)
Optimal, n = 64 0(0%) 0 (0%) 64 (100%)
Open in a new tab

Note: Inadequate <20 ng/mL adequate 20–29 ng/mL, optimal ≥30 ng/mL. Bold indicates change in status with consideration of 3-epimer. Percentages are indicative of row percent. 25(OH)D, 25-hydroxyvitamin D.

*

Weighted Kappa = 0.78.

Discussion

Evidence is limited on the physiological relevance of 3-epi-25(OH)D3 in children; thus a clearer understanding of the 3-epi-25(OH)D3 can help determine the necessity of considering 3-epi-25(OH)D3 clinically and in studies examining vitamin D. With half of the children in the Daily D Health Study having detectable 3-epi-25(OH)D3, we were able to determine that male, prepubertal, and non-Hispanic white children were more likely to have detectable 3-epi-25(OH)D3 concentrations, which suggests that having detectable 3-epi-25(OH)D3 is not a random phenomenon.

To our knowledge, there is no literature to compare these findings with other studies in children, but several of our findings coincide with existing literature on the 3-epi-25(OH)D3 in adult population studies. The range of 3-epi-25(OH)D3 concentration in the Daily D Health Study was similar to adults in the Survey of the Health of Wisconsin (SHOW), a cross-sectional population-based study in non-Hispanic white adults (0−4.9 ng/mL) (Engelman et al. 2014) and in the Atherosclerosis Risk in Communities (ARIC) study, which provided analyses of 3-epi-25(OH)D3 separately by white and black adults (2.12 ng/mL and 2.16 ng/mL, respectively) (Lutsey et al. 2015). The concentration of 3-epi-25(OH)D3 in infants has been shown to be much higher, with concentrations as high as 188 ng/mL (Singh et al. 2006), suggesting that 3-epi-25(OH)D3 concentrations drop from infancy to childhood, but do not drop from childhood to adulthood.

The correlation between 3-epi-25(OH)D3 and 25(OH)D3 was lower in our study than in 2 adult studies (r = 0.40 vs. r = 0.66, Engelman et al. 2014; r = 0.78, Cashman et al. 2014, respectively) but similar to the ARIC study (r = 0.54 in whites and r = 0.36 in blacks). This is surprising given that the SHOW and the National Adult Nutrition Survey (NANS) studies were observational, some participants were using supplements, and there was variation in season of data collection. Whereas, in the Daily D Health Study, our analysis was cross-sectional at the beginning of the trial prior to vitamin D supplementation and during the same season. Our lower correlation could be explained, however, by racial/ethnic differences, in that the SHOW and NANS studies were primarily non-Hispanic white and the Daily D Health Study had a diverse population.

We found that males were more likely to have detectable 3-epi-25(OH)D3, but females were more likely to be above the median percent 3-epi-25(OH)D3 of 5.5%. The ARIC study and a national survey in Thailand found that males had higher 3-epi-25(OH)D3 concentrations (Chailurkit et al. 2015; Lutsey et al. 2015), thus sex seems to impact having detectable 3-epi-25(OH)D3 concentrations and the magnitude of concentrations, but the relationship requires further investigation. Our findings also indicate that non-Hispanic white children were more likely to have detectable 3-epi-25(OH)D3, which is similar to the findings from the ARIC study, in which 33% of whites and 15% of blacks had detectable 3-epi-25(OH)D3 concentrations (Lutsey et al. 2015) and NHANES, in which 91% of non-Hispanic whites and 64% of non-Hispanic blacks had detectable 3-epi-25(OH)D3 (Schleicher et al. 2011, 2016). In these studies, higher vitamin D intake was also associated with higher 3-epi-25(OH)D3 concentrations. Among our findings, dietary vitamin D was significantly higher in children with detectable 3-epi-25(OH)D3, but dietary vitamin D intake was not associated with higher 3-epi-25(OH)D3 concentrations or higher percent 3-epi-25(OH)D3. The NHANES and SHOW studies found lower waist circumference to be a predictor of higher 3-epi-25(OH)D3 concentrations (Cashmanetal.2014;Engelmanetal.2014), which also does not coincide withourfindingsofnorelationshipbetween3-epi-25(OH)D3 and weight status. Nonetheless, a study of healthy-term infants aged newborn to 1 year also found that the 3-epi-25(OH)D3 was not predictive of lean mass, in contrast with the normal 25(OH)D3 variant (Hazell et al. 2014).

When evaluating the relationship between 3-epi-25(OH)D3 and blood lipids, 3-epi-25(OH)D3 was associated with blood lipids only when examining 3-epi-25(OH)D3 as a percentage of total 25(OH)D. Total cholesterol was 7 ng/mL lower in children with percent 3-epi-25(OH)D3 above the median (5.5%). This suggests that there may be a physiological relevance of the 3-epi-25(OH)D3, but not independent of 25(OH)D; rather the ratio of the 2 values could be important. Scant evidence exists on associations between the 3-epi-25(OH)D3 and blood lipids. In the ARIC study, among both whites and blacks, HDL concentrations were higher with increasing tertiles of 3-epi-25(OH)D3 concentration, but no associations were found with LDL or triglycerides, and associations with total cholesterol were not reported (Lutsey et al. 2015). In general, higher concentrations of 3-epi-25(OH)D3 were associated with a better cardiovascular risk profile, including lower BMI, lower prevalence of diabetes, and lower C-reactive protein concentration, but this study did not examine these relationships using percent of 3-epi-25(OH)D3 for total 25(OH)D (Lutsey et al. 2015). Additional studies investigating potential physiologic relevance of the 3-epi-25(OH)D3 are needed, particularly in children.

Currently, an individual’s vitamin D status is determined by 25(OH)D3 concentrations, and whether 3-epi-25(OH)D3 is included as part of 25(OH)D3 is dependent on the assay used. The consideration of 3-epi-25(OH)D3 in total 25(OH)D concentrations increased the proportion of children classified with a higher vitamin D status. There is potential practical relevance in 15% (51 out of 337 children with detectable 3-epi-25(OH)D3) of children moving from inadequate to adequate status or adequate to optimal status. Other studies have reported smaller percentages of individuals being reclassified to adequacy status with inclusion of the 3-epi-25(OH)D3 in total 25OHD, such as 2%–3% of adults in ARIC (Lutsey et al. 2015), and 7% (APrON) (Aghajafari et al. 2016) to 9% of newborns (Strathmann et al. 2012). This suggests that, particularly in younger age groups, regardless of the small serum concentrations, 3-epi-25(OH)D3 could impact the identification of an individual being considered deficient or not, and contribute to determining supplementation needs. This may also be particularly important in individuals with lower 25(OH)D where a change in classification might be clinically meaningful (Bailey et al. 2013), such as in individuals living in northern latitudes during the winter, those with darker skin, and those who are over-weight or obese. Our data support the general consensus that the 3-epi-25(OH)D3 should be quantified in infant/pediatric populations to avoid complicating interpretation of total 25(OH)D concentrations.

This was the first study to our knowledge to examine the 3-epi-25(OH)D3 in a diverse group of school-aged children, which provided a distribution of 3-epi-25(OH)D3 detectability to make a useful comparison between children with and without detectable 3-epi-25(OH)D3. Due to known differences in vitamin D by race/ethnicity, it may be appropriate to examine the association between 3-epi-25(OH)D3 and cardiometabolic risk factors separately by racial/ethnic group, but small sample sizes in the present study precluded these analyses.

In conclusion, we have shown that there are demographic characteristics that can impact 3-epi-25(OH)D3 concentration, including sex and race/ethnicity. We have also demonstrated certain characteristics that are associated with having detectable 3-epi-25(OH)D3. Even small concentrations of 3-epi-25(OH)D3 can impact how individuals are classified based on vitamin D status. Our findings show that considering the 3-epi-25(OH)D3 as a percent of 25(OH)D provides alternative but important insight for potential physiological relevance. Finally, further research is needed to clearly elucidate the physiological relevance of 3-epi-25(OH)D3, particularly its relationship to cardiovascular risk factors.

Acknowledgements

The authors would like to thank the Daily D Health Study participants, research coordinators, and Tufts University graduate research assistants who helped with data collection. We would also like to thank Daily D study co-investigators Christina Economos and Virginia Chomitz. Finally, we gratefully acknowledge financial support from the National Institutes of Health Office of Dietary Supplements. J.M.S., C.M.G., E.G., and M.F.H. designed research; S.E.B., J.M.S., C.M.G., E.G., and M.F.H. conducted research; S.E.B., M.I.V.R., and M.E. analyzed data; and S.E.B., M.I.V.R., and J.M.S. wrote the paper. S.E.B. had primary responsibility for final content. All authors read and approved the final manuscript.

Conflict of interest statement and funding disclosure

This work was supported by the National Institutes of Health, National Heart, Lung, And Blood Institute under Award Number R01HL106160. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Contributor Information

Samantha E. Berger, Friedman School of Nutrition Science and Policy, Tufts University, Boston, MA 02111, USA.

Maria I. Van Rompay, Friedman School of Nutrition Science and Policy, Tufts University, Boston, MA 02111, USA.

Catherine M. Gordon, Divisions of Adolescent and Transition Medicine, University of Cincinnati College of Medicine, Cincinnati, OH 45267, USA.

Elizabeth Goodman, MassGeneral Hospital for Children and Harvard Medical School, Boston, MA 02114, USA..

Misha Eliasziw, Department of Public Health and Community Medicine, Tufts University, Boston, MA 02111, USA..

Michael F. Holick, Endocrinology, Diabetes and Nutrition Section, Department of Medicine, Boston University Medical Center, Boston, MA 02118, USA.

Jennifer M. Sacheck, Friedman School of Nutrition Science and Policy, Tufts University, Boston, MA 02111, USA.

References

  1. Aghajafari F, Field CJ, Rabi D, Kaplan BJ, Maggiore JA, O’Beirne M, et al. 2016. Plasma 3-epi-25-hydroxycholecalciferol can alter the assessment of vitamin D status using the current reference ranges for pregnant women and their newborns. J. Nutr 146(1): 70–75. doi: 10.3945/jn.115.220095. [DOI] [PubMed] [Google Scholar]
  2. Bailey D, Veljkovic K, Yazdanpanah M, and Adeli K 2013. Analytical measurement and clinical relevance of vitamin D(3) C3-epimer. Clin. Biochem 46(3): 190–196. doi: 10.1016/j.clinbiochem.2012.10.037. [DOI] [PubMed] [Google Scholar]
  3. Bao W, Srinivasan SR, Valdez R, Greenlund KJ, Wattigney WA, and Berenson GS 1997. Longitudinal changes in cardiovascular risk from childhood to young adulthood in offspring of parents with coronary artery disease: the Bogalusa Heart Study. JAMA J. Am. Med. Assoc 278(21): 1749–1754. doi: 10.1001/jama.1997.03550210047037. [DOI] [PubMed] [Google Scholar]
  4. Carskadon MA, and Acebo C 1993. A self-administered rating scale for pubertal development. J. Adolesc. Health, 14(3): 190–195. doi: 10.1016/1054-139X(93)90004-9. [DOI] [PubMed] [Google Scholar]
  5. Cashman KD, Kinsella M, Walton J, Flynn A, Hayes A, Lucey AJ, et al. 2014. The 3 epimer of 25-hydroxycholecalciferol is present in the circulation of the majority of adults in a nationally representative sample and has endogenous origins. J. Nutr 144(7): 1050–1057. doi: 10.3945/jn.114.192419. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. CDC. 2013. BMI Percentile Calculator for Child and Teen: English Version Centers for Disease Control and Prevention, Atlanta, Ga., USA: Available from http://www.cdc.gov/nccdphp/dnpa/healthyweight/assessing/bmi/00binaries/bmi-tables.pdf. [Google Scholar]
  7. Chailurkit L, Aekplakorn W, and Ongphiphadhanakul B 2015. Serum C3 epimer of 25-hydroxyvitamin D and its determinants in adults: a national health examination survey in Thais. Osteoporosis Int 26(9): 2339–2344. doi: 10.1007/s00198-015-3125-y. [DOI] [PubMed] [Google Scholar]
  8. Dawson-Hughes B, Heaney RP, Holick MF, Lips P, Meunier PJ, and Vieth R 2005. Estimates of optimal vitamin D status. Osteoporosis Int 16(7): 713–716. doi: 10.1007/s00198-005-1867-7. [DOI] [PubMed] [Google Scholar]
  9. Engelman CD, Bo R, Zuelsdorff M, Steltenpohl H, Kirby T, and Nieto FJ 2014. Epidemiologic study of the C-3 epimer of 25-hydroxyvitamin D(3) in a population-based sample. Clin. Nutr. (Edinb. Scotl.), 33(3): 421–425. doi: 10.1016/j.clnu.2013.06.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Giovannucci E 2008. 25-Hydroxyvitamin D and risk of myocardial infarction in men: a prospective study. Arch. Intern. Med 168(11): 1174. doi: 10.1001/archinte.168.11.1174. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Granado-Lorencio F, Blanco-Navarro I, Pérez-Sacristán B, Donoso-Navarro E, and Silvestre-Mardomingo R 2012. Serum levels of 3-epi-25-OH-D3 during hypervitaminosis D in clinical practice. J. Clin. Endocrinol. Metab 97(12): E2266–E2270. doi: 10.1210/jc.2012-2627. [DOI] [PubMed] [Google Scholar]
  12. Hazell TJ, Gallo S, Berzina L, Vanstone CA, Rodd C, and Weiler HA 2014. Plasma 25-hydroxyvitamin D, more so than its epimer, has a linear relationship to leaner body composition across infancy in healthy term infants. Appl. Physiol. Nutr. Metab 39(10): 1137–1143. doi: 10.1139/apnm-2013-0586. [DOI] [PubMed] [Google Scholar]
  13. Holick MF, Siris ES, Binkley N, Beard MK, Khan A, Katzer JT, et al. 2005. Prevalence of Vitamin D inadequacy among postmenopausal North American women receiving osteoporosis therapy. J. Clin. Endocrinol. Metab 90(6): 3215–3224. doi: 10.1210/jc.2004-2364. [DOI] [PubMed] [Google Scholar]
  14. IOM 2011. Dietary Reference Intakes for Calcium and Vitamin D Edited by Ross AC, Taylor CL, Yaktine AL, and Del Valle HB. National Academies Press (US), Washington, DC, USA. doi: 10.17226/13050. [DOI] [PubMed] [Google Scholar]
  15. Kamao M, Hatakeyama S, Sakaki T, Sawada N, Inouye K, Kubodera N, et al. 2005. Measurement and characterization of C-3 epimerization activity toward vitamin D3. Arch. Biochem. Biophys 436(1): 196–205. doi: 10.1016/j.abb.2005.01.017. [DOI] [PubMed] [Google Scholar]
  16. Lensmeyer GL, Wiebe DA, Binkley N, and Drezner MK 2006. HPLC method for 25-hydroxyvitamin D measurement: comparison with contemporary assays. Clin. Chem 52(6): 1120–1126. doi: 10.1373/clinchem.2005.064956. [DOI] [PubMed] [Google Scholar]
  17. Lensmeyer G, Poquette M, Wiebe D, and Binkley N 2012. The C-3 epimer of 25-hydroxyvitamin D(3) is present in adult serum. J. Clin. Endocrinol. Metab 97(1): 163–168. doi: 10.1210/jc.2011-0584. [DOI] [PubMed] [Google Scholar]
  18. Looker AC, Johnson CL, Lacher DA, Pfeiffer CM, Schleicher RL, and Sempos C 2011. Vitamin D status: United States, 2001–2006 NCHS Data Brief, Institute of Medicine; U.S. Department of Health and Human Services; Centers for Disease Control; National Institutes of Health. [PubMed] [Google Scholar]
  19. Lutsey PL, Eckfeldt JH, Ogagarue ER, Folsom AR, Michos ED, and Gross M 2015. The 25-hydroxyvitamin D3 C-3 epimer: distribution, correlates, and reclassification of 25-hydroxyvitamin D status in the population-based Atherosclerosis Risk in Communities Study (ARIC). Clin. Chim. Acta, 442: 75–81. doi: 10.1016/j.cca.2014.12.036. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Reis JP, von Mühlen D, Miller ER, Michos ED, and Appel LJ 2009. Vitamin D status and cardiometabolic risk factors in the United States adolescent population. Pediatrics, 124(3): e371–e379. doi: 10.1542/peds.2009-0213. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Sacheck JM, Van Rompay MI, Olson EM, Chomitz VR, Goodman E, Gordon CM, et al. 2015. Recruitment and retention of urban schoolchildren into a randomized double-blind vitamin D supplementation trial. Clin. Trials (Lond.), 12(1): 45–53. doi: 10.1177/1740774514556551. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Schleicher RL, Encisco SE, Chaudhary-Webb M, Paliakov E, McCoy LF, and Pfeiffer CM 2011. Isotope dilution ultra performance liquid chromatography-tandem mass spectrometry method for simultaneous measurement of 25-hydroxyvitamin D2, 25-hydroxyvitamin D3 and 3-epi-25-hydroxyvitamin D3 in human serum. Clin. Chim. Acta, 412(17–18): 1594–1599. doi: 10.1016/j.cca.2011.05.010. [DOI] [PubMed] [Google Scholar]
  23. Schleicher RL, Sternberg MR, Looker AC, Yetley EA, Lacher DA, Sempos CT, et al. 2016. National estimates of serum total 25-hydroxyvitamin D and metabolite concentrations measured by liquid chromatography-tandem mass spectrometry in the US population during 2007–2010. J. Nutr 146(5): 1051–1061. doi: 10.3945/jn.115.227728. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Singh RJ, Taylor RL, Reddy GS, and Grebe SKG 2006. C-3 epimers can account for a significant proportion of total circulating 25-hydroxyvitamin D in infants, complicating accurate measurement and interpretation of vitamin D status. J. Clin. Endocrinol. Metab 91(8): 3055–3061. doi: 10.1210/jc.2006-0710. [DOI] [PubMed] [Google Scholar]
  25. Stepman HCM, Vanderroost A, Stöckl D, and Thienpont LM 2011. Full-scan mass spectral evidence for 3-epi-25-hydroxyvitamin D3 in serum of infants and adults. Clin. Chem. Lab. Med 49(2): 253–256. doi: 10.1515/CCLM.2011.050. [DOI] [PubMed] [Google Scholar]
  26. Strathmann FG, Sadilkova K, Laha TJ, LeSourd SE, Bornhorst JA, Hoofnagle AN, and Jack R 2012. 3-epi-25 hydroxyvitamin D concentrations are not correlated with age in a cohort of infants and adults. Clin. Chim. Acta, 413(1–2): 203–206. doi: 10.1016/j.cca.2011.09.028. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Van Rompay MI, McKeown NM, Goodman E, Eliasziw M, Chomitz VR, Gordon CM, et al. 2015. Sugar-sweetened beverage intake is positively associated with baseline triglyceride concentrations, and changes in intake are inversely associated with changes in HDL cholesterol over 12 months in a multi-ethnic sample of children. J. Nutr 145(10): 2389–2395. doi: 10.3945/jn.115.212662. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. van den Ouweland JMW, Vogeser M, and Bächer S 2013. Vitamin D and metabolites measurement by tandem mass spectrometry. Rev. Endocr. Metab. Disord 14(2): 159–184. doi: 10.1007/s11154-013-9241-0. [DOI] [PubMed] [Google Scholar]

RESOURCES