Combined Oral Contraceptives: Association with Serum 25-Hydroxyvitamin D and Calcium and Bone Homeostasis

Abstract

Background:

Use of combined oral contraceptives (COCs) has been found to increase serum 25-hydroxyvitamin D [25(OH)D] but effects on calcium and bone homeostasis are unclear.

Materials and Methods:

Serum 25(OH)D, parathyroid hormone (PTH), alkaline phosphatase (ALK) and estradiol, dietary intake of bone-related nutrients and foods, bone mineral density (BMD), and body fat were compared in adult women (20–35 years; body mass index 21.5 ± 2.3 kg/m²) users (+COC, n = 32) and nonusers (–COC, n = 20) of COC. Biochemical markers were measured by automated assays. BMD at total body (TB), lumbar spine (LS), femoral neck (FN) and trochanter (TR), and body fat, were measured by dual-energy X-ray absorptiometry. Dietary intake was assessed by a food frequency questionnaire.

Results:

Intake of calcium, dairy foods, and fruits and vegetables, were adequate and did not differ by COC. Mean 25(OH)D was 35% higher in +COC (110.4 ± 27.3 nmol/L, 44.2 ± 1.8 ng/mL) compared with –COC (81.7 ± 22.8 nmol/L, 32.7 ± 2.3 ng/mL; p < 0.001). Mean PTH, ALK, and estradiol were 28%, 12%, and 62% lower, respectively, in +COC compared with –COC (p ≤ 0.05). Mean BMD z-scores (all sites) were adequate and did not differ by COC. There were no correlations between 25(OH)D and dietary, biochemical, and body composition variables. PTH was inversely correlated with TR-BMD z-score in –COC (r = −0.47; p = 0.04), and ALK was inversely correlated with TB-, TR-, and LS-BMD z-scores in –COC (r ≤ −0.43; p ≤ 0.04), but not in +COC.

Conclusions:

Increased serum 25(OH)D with COC use was paralleled by expected physiologic adjustments in calcium and bone homeostasis, and adequate bone mass status, in nonobese young adult women consuming bone-healthy diets.

Introduction

Hormonal oral contraceptives are used by women of reproductive age worldwide.¹ The most commonly used oral contraceptives are those combining an estrogen component with a progestogen component, usually identified as combined oral contraceptive (COC) in contrast to progesterone-only oral contraceptives.² Besides the prevention of pregnancy, the use of COC has been linked to noncontraception health effects, some of concern such as increased risk of stroke and myocardial infarction for COC pills containing ≥50 μg estrogen, and others beneficial to health such as risk reduction of endometrial and ovarian cancer.² Use of COC appears neutral or slightly beneficial to bone health in adult premenopausal women,^3,4 although it may inhibit bone accrual in adolescent girls and may limit peak bone mass in very young women, especially if started <2 years after menarche.⁵ The effects of COC on bone metabolism have been extensively examined indicating consistent associations with reduced markers of bone turnover.⁵

Moreover, some studies have shown that women users of COC have higher concentration of plasma/serum 25-hydroxivitamin D [25(OH)D] than women nonusers,^6

–10 although it is not known if this increase translates into health benefits, such as improved bone health¹¹ and reduced risk of ovarian cancer,¹² in COC users.

Vitamin D in its active form [1,25-dihydroxyvitamin D or 1,25(OH)₂D] functions primarily in the regulation of calcium and phosphorus intestinal absorption, homeostasis, and deposition in bone, and also in the regulation of several nonskeletal systems.¹¹ The concentration of 25(OH)D circulating in serum/plasma is considered the best biomarker of vitamin D status.¹¹ Concentrations higher than 75 nmol/L (30 ng/mL) are indicative of an optimal adequate vitamin D status, whereas those <50 nmol/L (20 ng/mL) and 30 nmol/L (12 ng/mL) indicate insufficiency and deficiency, respectively.^11,13 The Endocrine Society guidelines define vitamin D deficiency as a 25(OH)D concentration ≤50 nmol/L (20 ng/mL), and vitamin D insufficiency as a 25(OH)D concentration in the range from >50 to 72.5 nmol/L (21–29 ng/mL).¹⁴ In addition to exposure to ultraviolet B (UVB) radiation, vitamin D supplements, and exogenous estrogen, 25(OH)D levels may be affected by the diet and individual characteristics such as age, body mass index (BMI) and body fat.¹⁵ Therefore, age, habitual diet, and body fat composition need to be taken into account when examining concentrations of circulating 25(OH)D in relation to COC use.

In women users of COC, the magnitude of increase in circulating 25(OH)D compared with nonusers varied between studies within a wide range, from barely detectable to >50%^{6

–10,16

–19} and in some studies this increase did not reach statistical significance.^18,19 Few of these studies included assessment of the major calciotropic hormone parathyroid hormone (PTH)^8–9,17 and other markers of calcium and bone metabolism,⁹ with inconclusive results. For instance, circulating PTH was found decreased¹⁷ or not changed⁹ in women COC users compared with nonusers. Therefore, it remains unclear if the increased concentration of 25(OH)D in COC users elicit physiologic adjustments in calcium and bone homeostasis. Different results between studies may be owing in part to differences in the age of the women, BMI, body fat, bone mass status, and habitual diet particularly dietary intake of nutrients and food groups affecting bone health.²⁰

We hypothesize that in healthy young adult women with no vitamin D deficiency and with sufficient calcium intake, a higher concentration of circulating 25(OH)D with COC use is associated with physiologic changes in markers of calcium and bone homeostasis that preserve bone mineral density (BMD).

The objective of this study was to compare serum 25(OH)D, biochemical measurements of calcium and bone homeostasis, dietary intake of bone-related nutrients and food groups, body fat, BMD, and their relationship, in young adult women users (+COC) and nonusers (–COC) of COC.

Materials and Methods

Subjects and sample collection

Fifty-two young adult women (20–35 years), apparently healthy, nonobese, nonsmokers, not engaged in moderate/heavy physical activity, nonpregnant and nonbreastfeeding, with no history of metabolic disease, no or low alcohol consumption (<5 g/day), and with no use of medication (except for COC) and no use of any vitamin or mineral supplements including no vitamin D supplementation, during the prior 6 months, participated in the study. The women were nonusers (n = 20, –COC) and users (n = 32, +COC) of estrogen-containing oral contraceptives for the past 4 months or longer. Based on standard deviation values from an early study,⁷ it was estimated that this sample size would be sufficient to detect a difference in serum 25(OH)D between –COC and +COC of 16.9 nmol/L at 5% significance (α = 0.05) and 75% power(β = 0.25).

The study protocol was approved by the Ethical Committee of Facultad de Medicina, Universidad de la República, Uruguay (Identification No.: 070153-000253-14). The research was completed in accordance with the Declaration of Helsinki as revised in 2013. After signing the written informed consent, information on habitual diet and COC use and composition, was obtained by a structured questionnaire. Samples of fasting blood (10 mL) were obtained for biochemical analyses. Anthropometry and dual-energy X-ray-absorptiometry (DXA) measures were used for assessment of body fat and bone mass. All sampling and measurements were carried out in the period of October to December 2014 (late-spring to early-summer season in Uruguay).

Habitual dietary intake of nutrients and food groups

Information on habitual dietary intake was obtained by a food frequency questionnaire consisting of local foods, with special attention to dairy products and fruits and vegetables, as previously described.²¹ Emphasis on these food groups was based on their recognized impact on bone health.²⁰ Serving sizes were according to the Uruguayan Dietary Guidelines²² and were identified with the aid of pictures of commonly used home food measures. Typical serving sizes were as follows: milk, 250 mL; cheese, 30–50 g; yogurt, 200 mL; fruits, 150 g; raw vegetables, 200 g. Nutrient intakes were estimated using the program SARA version 1.2.22 (Subsecretaria de Estrategias Sanitarias, Ministerio de Salud, Argentina) adapted to locally available foods. Results were expressed as average daily intake of servings of specific food groups, and average daily intake of macro- and micronutrients.

Biochemical measurements

Serum calcium, phosphorus, and ALK activity were determined by conventional methods using an automatic analyzer (Lisa 200, Hycel, France). Serum 25(OH)D, intact PTH, and estradiol were measured using chemiluminescent immunometric assays (Automated Immunoassay Analyzer AIA 900, TOSOH Bioscience, Japan). Intra-assay coefficients of variation were <6%. Assay performances were monitored by using the low- and high-concentration quality control sets provided by the manufacturer and were within the expected range.

Anthropometric and body composition measurements

Body weight and standing height were measured by using a calibrated electronic scale (Seca) and a stadiometer (Seca), respectively. Bone mineral content (BMC) and BMD of the total body (TB), lumbar spine (LS; L1–L4), and hip (trochanter [TR] and femoral neck [FN]), and body fat of the TB, android, and gynoid regions, were assessed by DXA (GE Lunar Prodigy Advance 9.0, software enCore v18). All DXA scanning and calibration was performed by the same operator. Quality control of the DXA measurements followed standard procedures. According to the International Society of Clinical Densitometry recommendations²³ for women before menopause, BMD results were expressed as z-scores using the age- and sex-matched reference database included in the equipment software. BMD z-scores lower than −2 are considered as “below expected range for age.” Body fat measures were expressed as body weight percentage.

Statistical analyses

Complete dietary and biochemical data were obtained for all women. One woman in the +COC group did not show up for DXA measurements. All variables were tested for normality and those with non-normal distribution were described and analyzed using nonparametric methods. Differences in dietary, biochemical, and body composition variables between –COC and +COC groups were assessed by t-test or Mann–Whitney test, as appropriate. Differences in categorical variables between –COC and +COC were assessed by the chi-square test. Associations between continuous variables (biochemical, dietary, and body composition) were examined by simple correlation analyses, Pearson or Spearman as appropriate, in all women and in COC subgroups. The statistical analyses were performed using Statgraphics Centurion, version XVI.II. p- ≤ 0.05 were considered significant.

Results

The young adult women of the –COC and +COC groups did not differ in age, BMI, and habitual dietary intake of major food groups, macro- and micronutrients (p ≥ 0.12; Table 1). All women were self-identified as of European descendance. BMI (overall mean, 21.5 kg/m²) was mostly within the normal range.²⁴ Overweight (BMI >25 kg/m²) was present in 3 (15%) and 4 (13%) women in the –COC and +COC groups, respectively (p = 0.80). Overall median dietary calcium intake (979 mg/day) was slightly lower than the 1000 mg/day dietary reference intake of calcium for adult women.¹³ Habitual intake of dairy foods was 2 servings/day or more (overall median, 2.8 servings/day), and habitual intake of fruits and vegetables was 4 servings/day or more (overall mean, 4.6 servings/day) in all women, approaching the local dietary guidelines for dairy foods (3 servings/day) and for fruits and vegetables (5 servings/day).²²

Table 1.

Dietary and Biochemical Characteristics of the Women Users and Nonusers of Combined Oral Contraceptives

	Nonusers, –COC (n = 20)	Users, +COC (n = 32)	p ^a
Age (years)	25.6 (24.5–26.7)	24.8 (24.0–25.7)	0.45
Body mass index (kg/m²)	21.9 (21.2–22.6)	21.2 (20.6–21.8)	0.31
Habitual dietary intake
Energy (kcal/day)	1998 (1864–2132)	2072 (1969–2176)	0.54
Protein (g/day)	84.7 (77.3–92.0)	92.0 (86.3–97.6)	0.20
Fiber (g/day)	27.0 (17.7–33.5)	23.5 (18.0–26.0)	0.24^b
Calcium (mg/day)	1055 (858–1161)	976 (868–1106)	0.58^b
Phosphorous (mg/day)	1501 (1390–1612)	1482 (1396–1567)	0.58
Potassium (mg/day)	3971 (2708–4514)	3396 (3087–3971)	0.41^b
Dairy foods (servings/day)	2.7 (2.1–3.4)	2.9 (2.2–3.4)	0.74^b
Fruits/vegetables (servings/day)	5.0 (4.5–5.6)	4.3 (3.9–4.7)	0.12
Biochemical serum measurements
Estradiol (pg/mL)	107.9 (68.6–196.4)	40.7 (32.9–59.9)	<0.001^b
Calcium (mg/dL)	8.44 (8.30–8.58)	8.56 (8.44–8.67)	0.35
Phosphorous (mg/dL)	4.18 (3.82–4.55)	4.21 (3.92–4.50)	0.94
Alkaline phosphatase (U/L)	145 (135–155)	127 (119–135)	0.05
Intact PTH (pg/mL)	46.4 (32.7–60.0)	33.5 (28.8–38.3)	0.04
25(OH)D (nmol/L)	81.8 (73.8–90.0)	110.5 (104–117)	<0.001
(ng/mL)	32.7 (29.5–36.0)	44.2 (41.6–46.8)

Values are given as mean (95% CI) or median (95% CL) (dietary fiber, dietary calcium, dietary potassium, dairy foods, serum estradiol).

Comparison between users and nonusers of COC by t-test except when indicated as “b.”

Comparison between users and nonusers of COC by Mann–Whitney test.

25(OH)D, 25-hydroxyvitamin D; CI, confidence interval; COC, combined oral contraceptives; CL, confidence limits of the median; PTH, parathyroid hormone.

The women in the +COC group used combined formulations of oral contraceptives containing ethinyl estradiol (20–30 μg) plus a progestogen (drospirenone, gestodene, or dienogest; 0.1–3 mg). The period of COC use varied from 4 months to 15 years (median 4 years) with 28 of the 32 women (87.5%) having used COC during 2 years or more.

Women of the –COC and +COC groups differed in some of the biochemical measurements (Table 1). Median serum estradiol was 62% lower and mean serum intact PTH was 28% lower in +COC compared with –COC (p < 0.001 and p = 0.03, respectively). Serum ALK was 12% lower in +COC compared with –COC (p = 0.05). Serum calcium and serum phosphorus did not differ between groups (p ≥ 0.35). Serum 25(OH)D was on average 35% higher in +COC compared with –COC (p < 0.001). In the +COC group, serum 25(OH)D concentration was not associated with the period of COC use (p = 0.94), and not associated with the estradiol/progestogen dose ratio in the oral contraceptives (p = 0.90). Serum 25(OH)D concentration was <75 nmol/L (30 ng/mL) in 7 (35%) and 3 (9%) women of the –COC and +COC groups, respectively (p = 0.02). Only one woman, in the –COC group, had serum 25(OH)D concentration <50 nmol/L (20 ng/mL).

Total percent body fat was not different between COC groups, although the median was nonsignificantly higher in the +COC compared with the –COC group, 34.7% and 32.3%, respectively (p = 0.10; Table 2). Moreover, among women with total percent body fat above the median, most (n = 20, 77%) were in the +COC group (p = 0.02). Percent fat in the android and gynoid regions did not differ by COC use (p ≥ 0.25). Mean android/gynoid ratio was 0.74 and 0.77, in the –COC and +COC groups, respectively (p = 0.39). Similarly, mean z-scores of BMC and BMD at TB, FN, TR, and LS, did not differ between women in the –COC and +COC groups (p ≥ 0.19; Table 2). Mean z-scores of all bone measurements were within the normal age range irrespective of COC use. Four women (–COC, n = 2; +COC, n = 2) had BMD z-score at the TR lower than −2.0.

Table 2.

Body Fat and Bone Mineral Density Measures of the Women Users and Nonusers of Combined Oral Contraceptives

	Nonusers, –COC (n = 20)	Users, +COC (n = 31)	p ^a
Body fat (%)
Total, %	32.3 (27.6 to 35.3)	34.7 (32.3 to 36.6)	0.10^b
Android, %	32.9 (30.1 to 35.7)	35.9 (33.6 to 38.1)	0.25
Gynoid, %	44.0 (42.2 to 45.8)	45.7 (44.3 to 47.2)	0.28
Bone measures (z scores)
Total body BMC	0.520 (0.249 to 0.790)	0.816 (0.599 to 1.033)	0.23
Total body BMD	0.445 (0.210 to 0.680)	0.442 (0.253 to 0.631)	0.99
Femoral neck BMD	−0.175 (−0.427 to 0.077)	−0.206 (−0.409 to 0.004)	0.89
Trochanter BMD	−0.635 (−0.936 to −0.334)	−0.529 (−0.771 to −0.287)	0.70
Lumbar BMD	−0.135 (−0.426 to 0.156)	0.213 (−0.021 to 0.447)	0.19

Values are given as mean (95% CI) except for total body fat (median, 95% CL).

Comparison between users and nonusers of COC (t-test) except when indicated as “b.”

Comparison between users and nonusers of COC by Mann–Whitney test.

BMC, bone mineral content; BMD, bone mineral density.

Simple correlations between biochemical indices of calcium homeostasis and BMD z-scores were examined in all women and separately in –COC and +COC groups (Table 3). There were no significant correlations between serum 25(OH)D and, serum PTH (p ≥ 0.10), serum ALK (p ≥ 0.11), and BMD z-scores (p ≥ 0.55), in all women and COC subgroups. Also, there were no significant correlations between serum 25(OH)D and, BMI, body fat measures, and dietary intake of specific nutrients or food groups (results not shown). Serum PTH was inversely correlated with TR-BMD z score in the pooled women (r = −0.291, p = 0.04), and in the –COC women (r = −0.470, p = 0.04) but not in the +COC women (p = 0.85).

Table 3.

Pearson Correlations Between Biochemical Indices of Calcium Homeostasis and Bone Mineral Density in the Women Users and Nonusers of Combined Oral Contraceptives

Related variables	All women (n = 52)		Nonusers, –COC (n = 20)			Users, +COC (n = 32)^a
Related variables	r	p	R	p		r	p
Serum 25(OH)D and PTH	−0.233	0.10	0.274		0.24	−0.187		0.30
Serum 25(OH)D and ALK	−0.224	0.11	0.001		0.98	−0.188		0.31
Serum iPTH and ALK	0.171	0.23	0.259		0.27	−0.153		0.40
Serum 25(OH)D and TB BMD z score	−0.008	0.96	0.135		0.57	−0.090		0.63
Serum 25(OH)D and TR BMD z score	0.081	0.57	0.145		0.54	0.006		0.97
Serum 25(OH)D and FN BMD z score	−0.018	0.90	0.110		0.64	−0.089		0.64
Serum 25(OH)D and LS BMD z score	0.060	0.67	0.088		0.71	−0.112		0.55
Serum PTH and TB BMD z score	−0.271	0.05	−0.424		0.06	−0.106		0.57
Serum PTH and TR BMD z score	−0.291	0.04	−0.470		0.04	0.036		0.85
Serum PTH and FN BMD z score	−0.191	0.18	−0.130		0.58	−0.339		0.07
Serum PTH and LS BMD z score	−0.202	0.16	−0.254		0.28	−0.018		0.93
Serum ALK and TB BMD z score	−0.284	0.04	−0.616		<0.01	−0.020		0.92
Serum ALK and TR BMD z score	−0.287	0.04	−0.528		0.02	−0.028		0.88
Serum ALK and FN BMD z score	−0.226	0.11	−0.431		0.07	−0.051		0.79
Serum ALK and LS BMD z score	−0.347	0.01	−0.470		0.04	−0.171		0.36

Units of measurements: 25(OH)D, nmol/L; iPTH, pg/mL; serum ALK, U/L.

n = 31 for BMD variables.

ALK, alkaline phosphatase; FN, femoral neck; iPTH, intact PTH; LS, lumbar spine; TB, total body; TR, trochanter.

An inverse correlation between serum PTH and TB-BMD z score was observed in all women (r = −0.27, p = 0.05). Serum ALK was inversely correlated with TB-, TR-, and LS-BMD z scores in all women (r = −0,284, r = −0.287, r = −0.347, p ≤ 0.04, respectively) and in the –COC women (r = −0,616, r = −0.528, r = −0.470, respectively, p ≤ 0.04), but not in the +COC group (p ≥ 0.36).

Discussion

In this study of nonobese young adult women with adequate vitamin D status and consuming diets providing nutrients and food groups that favor bone health, it was shown that COC users had a substantial increase in serum 25(OH)D, suppression of serum PTH and biochemical indication of reduced bone turnover, compared with nonusers. BMD status was adequate irrespective of COC use. Findings of this and previously published studies that examined the use of estrogen-containing contraceptives in relation to circulating 25(OH)D levels and other measurements are summarized in Table 4. This study is among the few studies combining biochemical, dietary, and bone mass data in the problem approach, and the one providing evidence that the COC-related increase in serum 25(OH)D is paralleled by the expected physiologic response of calcium and bone homeostasis that preserve bone mass status in healthy young women.

Table 4.

Summary of Studies Examining Use of Estrogen-Containing Contraceptives and Concentration of Circulating 25-Hydroxyvitamin D in Adult Women

Source	Study site and population of women	Age of women	Type of study and contraceptive examined	Mean/median serum/plasma 25OHD in women nonusers	Change in serum/plasma 25(OH)D in women users compared with nonusers	Other measurements	Additional findings
Sowers et al.⁶	Iowa/United StatesEuropean heritage	20–80 years	Cross-sectionalOC	60 nmol/L (n = 20)^a (premenopausal)	↑ 15% (n = 66)	Vitamin D intake (diet/supplements)Sunlight exposureSmoking practicePhysical activityBMD	25(OH)D levels negatively related to age, and positively related to use of exogenous estrogen and premenopausal state independent of age25(OH)D levels not associated to BMD
Harris and Dawson-Hughes⁷	Boston/United StatesCaucasian	20–40 years	Longitudinal1-year comparison at winter seasonOC	59 nmol/L (n = 40)	↑ 41% (n = 26)	Vitamin D intake (diet/supplements)Smoking practices	No association between 25(OH)D levels and ethynyl estradiol dose, type of OC and duration of useOver 1 year, users of OC maintained higher 25(OH)D levels compared with nonusers
Garcia-Bailo et al.¹⁶	Toronto/CanadaEthnicity: CaucasianEast AsianSouth Asian	20–29 years	Cross-sectionalHC	51.1 nmol/L (n = 695)	↑ 53% (n = 280)	Vitamin D intake (diet/supplements)Physical activityBMI, waist circumferenceC-reactive protein	Women HC users had higher 25(OH)D than women non-HC across ethnic groupsNo association between 25(OH)D and CRP when HC use was accounted for
Shirazi et al.⁸	Malmo/SwedenWomen born northern European countries	41–73 years	Cross-sectionalOC	86.0 nmol/L (n = 366)	↑ 3% (n = 361)	BMIDietary intake of foods and nutrientsVitamin D supplementsSmoking statusAlcohol consumptionSerum PTH, calcium, phosphate, albumin, creatinine	In all women, 25(OH)D levels positively associated with age, vitamin D intake, moderate alcohol consumption, and serum calcium and phosphate; negatively associated with serum PTH. Associations not informed separately in women OC users and nonusers
Hedlund et al.¹⁷	Gothenburg/SwedenFair-skinned women	25–40 years	Cross-sectionalNot specified	63.9 nmol/L (n = 69)	↑ 16% (n = 15) (ns)	Sun exposureVitamin D intake (diet/supplements)Physical activitySmoking statusSerum PTH	Almost 30% of variation in 25(OH)D levels explained by sun exposure, vitamin D supplements and OC use.Serum PTH lower in women OC users compared with nonusers
Møller et al.⁹	Aarhus/DenmarkCaucasian	25–35 years	Cross-sectionalHC	70 nmol/L (n = 23)	↑ 20% (n = 52)	Dietary calciumVitamin D supplementsSmoking habitsBMD at total body, hip, lumbar spinePlasma 1,25 (OH)₂D, VDBPPlasma calcium, phosphate, calcitonin, iPTH, BS-ALK, and urinary NTx	Plasma 1,25 (OH)₂D and VDBP levels higher in women HC users compared with nonusersNo difference by HC use in indices of calcium homeostasis and bone metabolism, and in BMD
Harmon et al.¹⁰	Detroit/United StatesAfrican-American	23–34 years	Cross-sectionalHC	37.5 nmol/L (n = 1193)^a	↑ 20% (n = 226)(regression model adjusted for covariates)	Vitamin D intake (diet/supplements)Education levelSmoking statusParityBMIAlcohol useMelanin index, sunny vacation, time outside	No increase in 25(OH)D levels in women past-users of exogenous estrogen but not current usersIn all women, 25(OH)D levels positively associated with vitamin D suppl. dose and negatively associated with BMI
Wyon et al.¹⁸	Walsall/UKUniversity students	18–45 years	Prospective double-blind vitamin D_3/placebo supplementationOC	43.5 nmoL/L (n = 19)(baseline)	↑ 11% (n = 19) ns(baseline)	BMI	Greater increase in 25(OH)D levels following vitamin D suppl. in OC users compared with nonusers (estrogen-OC effect on 25(OH)D magnified with vitamin D supplementation)
Palaniswamy et al.¹⁹	Oulu and Lapland/FinlandNorthen European women	31 years	Cross-sectionalanalysis (baseline) of Northern Finland Birth Cohort 1966OC	64.6 nmoL/L (n = 1777)	↑ 1% (n = 607)	DaylightAnthropometrySmokingAlcohol intakePhysical activityUnhealthy diet scoreSex and gender	OC use increased both 25(OH)D₃ and 25(OH)D₂ levels
This study	Montevideo/UruguayEuropean descendance	20–35 years	Cross-sectionalOC	81.8 nmol/L (n = 20)	↑ 35% (n = 32)	Dietary intake of foods and nutrients (calcium, protein, fiber, dairy foods, fruits and vegetables)Serum calcium, phosphorous, ALK, iPTH, estradiolBody fatBMD at total body, TR, FN, LS	Serum estradiol, iPTH and ALK lower in women OC users compared with nonusersBMD status adequate in OC users and in nonusersSerum iPTH inversely related to TR-BMD z score only in nonusersSerum ALK inversely related to total body, TR- and LS-BMD z scores only in nonusersSerum 25(OH)D not associated with BMD, body fat and dietary intake

Calculated using 2.5 as the conversion factor from ng/mL to nmol/L.

BS-ALK, bone-specific alkaline phosphatase; FN, femoral neck; HC, hormonal contraceptives, all types and delivery methods (oral, subdermal, spiral); LS, lumbar spine; ns, not significant; NTx, cross-linked N-terminal telopeptide of type I collagen; OC, oral contraceptive; TR, trochanter; VDBP, vitamin D binding protein.

The overall vitamin D status of the women studied was sufficient [serum 25(OH)D >50 nmol/L or 20 ng/mL] in all women except one^11,13 most likely because the measurement was carried out during the late-spring early-summer season in the southern hemisphere when endogenous vitamin D synthesis related to UVB exposure is favored.¹³ Dietary vitamin D intake was not assessed in this study although it possibly had a low contribution to serum 25(OH)D because of the poor vitamin D food sources present in the habitual dietary patterns in Uruguay.²⁵ Serum 25(OH)D was not associated with BMI or with body fat, in contrast to other studies,^26,27 possibly because the women were nonobese in addition to being nonvitamin D deficient. BMI and body fat did not differ by COC use, as expected, because estrogen-containing hormonal contraceptives have little or no effect on body weight and body fat composition in young adults.^28,29

Even in the absence of vitamin D deficiency, the 35% increase in serum 25(OH)D with COC use translated into improvement in vitamin D status, with >90% of the women COC users having serum 25(OH)D at optimal level [serum 25(OH)D >75 nmol/L or 30 ng/mL]^11,13,14 compared with only 65% among the women non–oral contraceptive (OC) users. These results indicate that OC use further improves vitamin D status in already vitamin D sufficient women.

In contrast to serum 25(OH)D, serum estradiol was markedly lower (62%) in women COC users compared with nonusers. This was an expected result because exposure to exogenous estrogen from contraceptives is known to blunt the circulating levels of endogenous estradiol consistent with suppressed ovulation.^30,31 Moreover, exposure to exogenous estrogen from contraceptives causes widespread systemic metabolic effects among which increased sex hormone-binding globulin³² and vitamin D binding protein (VDBP).^9,33 The mechanisms by which exogenous estrogen in contraceptives increases serum 25(OH)D, still under study, include the increased concentration in circulating VDBP with no change in free 25(OH)D^9,33 and possibly the increased activity in the 25-hydroxylation of cholecalciferol in liver.³⁴ Serum 1,25(OH)₂D, the active metabolite of vitamin D, was also found increased in adult⁹ and adolescent³³ women users of estrogen-containing contraceptives compared with nonusers, consistent with the possible upregulating effect of estrogen on 1-alpha-hydroxylase activity.¹⁶

Serum PTH was reduced in the COC-user young adult women of our study compared with those of nonusers, as also seen in a previous study¹⁷ but not in another study,⁹ of women of similar age range. It should be noted that mean dietary calcium intake was 800 mg/day in the latter study,⁹ whereas it was close to the dietary reference intake (1000 mg/day)¹³ in this study, a calcium intake more adequate for physiologic homeostatic adaptation. On the contrary, we did not find a significant inverse correlation between serum PTH and 25(OH)D in the whole group of women, or in the COC subgroups, in contrast to other studies.^8,17,35 This may be owing to the fact that serum 25(OH)D concentrations in the women of our study were mostly >75 nmol/L (30 ng/mL) with a small range of variation within COC subgroups.

Although we did not measure 1,25(OH)₂D, the blunting of the circulating PTH with COC use is consistent with the expected physiologic response of calcium homeostasis to the COC-related 1,25(OH)₂D increase described in other studies.^9,33 Moreover, the negative correlations found between serum PTH and BMD z-scores at TB and TR in the whole group and in women nonusers of COC, but not in those of COC users, are consistent with the notion that lower PTH is indicative of reduced bone turnover with COC use, and possibly that a low plateau in serum PTH is reached being no longer associated to BMD. Reduced bone turnover is a well-described effect of hormonal contraceptives in adult women.⁵

Serum ALK was lower in COC users compared with those of nonusers, and it was negatively correlated to BMD z-scores at different sites in all women and particularly in those of nonusers, consistent with reduced bone turnover with COC use. Although measurement of bone-specific ALK rather than total ALK is the preferred marker of bone turnover,⁵ reduced serum total ALK is often used as proxy of reduced serum bone ALK because the latter together with liver ALK represent ∼90% of total ALK in a 1:1 ratio.³⁶ Serum total ALK has been found to be inversely correlated with LS BMD in young adults of a population-based analysis,³⁷ in line with our interpretation of results. Moreover, the lack of association observed between serum ALK and BMD z-scores at TB and TR in COC users could suggest downregulation of different ALK isoforms, not just bone ALK, by exposure to exogenous estrogen present in COC.³⁶

The metabolic effect of COC on ALK isoforms and related health implications merit further study given the established relationship between increased serum ALK and inflammation, metabolic syndrome, and cardiovascular/renal disease.³⁶

Bone mass status was mostly adequate in the women studied irrespective of COC, consistent with their adequate vitamin D status and the characteristics of their diets with frequent consumption of dairy foods and fruits and vegetables, and adequate intakes of calcium, potassium, protein, and fiber, known to benefit bone health.^20,38 BMD z-scores at TB and at trabecular bone sites did not differ on average by COC use. This was an expected result because most studies found no harm, and some found a benefit of estrogen-containing oral contraceptive use on bone health of adult women.^5,39

In contrast, use of hormonal contraception is of concern for bone health of adolescent girls because the alteration of endogenous estrogen and IGF-1 concentrations may slow down bone mass acquisition and impair attaining peak bone mass particularly if oral contraceptive use is initiated at a very young age.^5,39,40 In our study of young adult women, although most of them had used COC for over 2 years, and up to 15 years, COC use did not seem to impair bone health.

Taking together the results from this study and those of other studies,^{5,9,10,17,33,39,40} it appears that in healthy young women the exposure to exogenous estrogen by oral contraceptive use blunts endogenous serum estradiol, increases serum 25(OH)D, and suppresses serum PTH, by interrelated mechanisms not yet completely elucidated that result in reduced bone turnover (Fig. 1). Changes in bone turnover markers usually precede detectable changes in bone mass and hence are valuable tools in clinical practice.⁴¹ Bone turnover is typically increased during adolescence associated with a rapid bone accrual that translates into peak bone mass attainment in young adults.⁴² On the contrary, after peak bone mass is achieved in adulthood, increased bone turnover indicates increased bone remodeling (bone resorption followed by bone formation) that, if not adequately balanced, may result in bone mass loss.⁴¹

FIG. 1.

Exposure to exogenous estrogen by oral contraceptive use and calcium and bone homeostasis in healthy young women. 1,25 (OH)₂D, 1,25-dihydroxyvitamin D; 25(OH)D, 25-hydroxyvitamin D; VDBP, vitamin D binding protein.

Therefore, reduced bone turnover may be harmful for bone mass status when the bone mass acquisition process is still active (adolescent women), but it may preserve bone mass status if peak bone mass has been already attained (adult women).

A limitation of our study was the small sample size that precluded covariate adjustments when examining the associations between serum 25(OH)D and biochemical and body composition variables. Information on dietary vitamin D intake, detailed physical activity, and other biochemical markers of calcium, vitamin D, and bone metabolism such as specific markers of bone formation and degradation, could have been useful for a more thorough analysis. A major strength was the group of healthy women studied that was homogeneous for different conditions (age, lifestyle, diet, ethnicity) that may affect vitamin D status beyond COC use.

In conclusion, findings from this study indicate that the increased serum 25(OH)D with the use of estrogen-containing oral contraceptives is accompanied by expected physiologic adjustments in markers of calcium and bone homeostasis, and adequate bone mass status, in nonobese young adult women with adequate vitamin D status and consuming a bone-healthy diet. These results may not be applicable to overweight/obese women, older adult women, and women of different ethnicities, lifestyles, and dietary patterns.

Footnotes

Acknowledgment

The authors appreciate the cooperation and motivation of the participating women.

Authors' Contributions

C.M.D.: Conceptualization, methodology, supervision, formal analysis, writing—original draft, writing—review and editing, funding acquisition. R.C.: Investigation, methodology, writing—review and editing, funding acquisition. C.S.: Investigation, data curation. F.F.B.: Conceptualization, writing—review and editing. All authors read and approved the final version of the article.

Data Availability

Data described in the article and analytic code will be made available by the corresponding author upon request, pending application and approval.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

Supported by Cooperativa Nacional de Productores de Leche (Conaprole, Uruguay) in agreement with Fundación para el Progreso de la Química (FUNDAQUIM, Uruguay) (Contract No. 009550-000387); and, Programa de Desarrollo de las Ciencias Básicas (PEDECIBA, Uruguay) to C.M.D.

References

United Nations, Department of Economics and Social Affairs. Contraceptive Use by Method 2019: Data Booklet. Available from: https://digitallibrary.un.org/record/3849735 [Last accessed: August 3, 2023 ].

Cooper

, Patel

, Mahdy

. Oral Contraceptive Pills. [Updated November 24, 2022]. In: StatPearls [Internet]. Treasure Island (FL): StatPearls Publishing; 2022. Available from: https://www.ncbi.nlm.nih.gov/books/NBK430882/ [Last accessed: August 3, 2023 ].

Nappi

, Bifulco

, Tommaselli

, et al. Hormonal contraception and bone metabolism: A systematic review. Contraception, 2012; 86(6):606–621; doi: 10.1016/j.contraception.2012.04.009

Rocca

, Palumbo

, Bitonti

, et al. Bone health and hormonal contraception. Minerva Obstet Gynecol, 2021; 73(6):678–696; doi: 10.23736/S2724-606X.20.04688-2

Herrmann

, Seibel

. The effects of hormonal contraceptives on bone turnover markers and bone health. Clin Endocrinol, 2010; 72:571–583; doi: 10.1111/j.1365-2265.2009.03688.x

Sowers

MFR

, Wallace

, Hollis

, et al. Parameters related to 25-OH-D levels in a population-based study of women. Am J Clin Nutr, 1986; 43:621–628.

Harris

, Dawson-Hughes

. The association of oral contraceptive use with plasma 25-hydroxyvitamin D levels. J Am Coll Nutr, 1998; 17:282–284; doi: 10.1080/07315724.1998.10718760

Shirazi

, Almquist

, Malm

, et al. Determinants of serum levels of vitamin D: A study of life-style, menopausal status, dietary intake, serum calcium, and PTH. BMC Womens Health, 2013; 13:33; doi: 10.1186/1472-6874-13-33

Møller

, Streym

, Jensen

, et al. Increased plasma concentrations of vitamin D metabolites and vitamin D binding protein in women using hormonal contraceptives: A cross-sectional study. Nutrients, 2013; 5:3470–3480; doi: 10.3390/nu5093470

10.

Harmon

, Umbach

, Baird

. Use of estrogen-containing contraception is associated with increased concentrations of 25-hydroxy vitamin D. J Clin Endocrinol Metab, 2016; 101(9):3370–3377; doi: 10.1210/jc.2016-1658

11.

Whiting

, Calvo

. Vitamin D: Nutrition information brief. Adv Nutr, 2021; 12(5):2037–2039; doi: https://doi.org/10.1093/advances/nmab051

12.

Jung

, Jin

, Je

. Vitamin D intake, blood 25-hydroxyvitamin D, and risk of ovarian cancer: A meta-analysis of observational studies. J Womens Health (Larchmt), 2023; 32(5):561–573; doi: 10.1089/jwh.2022.0432

13.

Ross

, Manson

, Abrams

, et al. The 2011 report on Dietary Reference Intakes for calcium and vitamin D from the Institute of Medicine: What clinicians need to know. J Clin Endocrinol Metab, 2011; 96(1):53–58; doi: 10.1210/jc.2010-2704

14.

Holick

, Binkley

, Bischoff-Ferrari

, et al. Evaluation, treatment, and prevention of vitamin D deficiency: An Endocrine Society Clinical Practice Guideline. J Clin Endocrinol Metab, 2011; 96:1911–1930; doi: 10.1210/jc.2011-0385

15.

Mazahery

, von Hurst

. Factors affecting 25-hydroxyvitamin D concentration in response to vitamin D supplementation. Nutrients, 2015; 7:5111–5142; doi: 10.3390/nu7075111

16.

Garcıa-Bailo

, Josse

, Jamnik

, et al. Positive association between 25-hydroxyvitamin D and C-reactive protein is confounded by hormonal contraceptive use. J Womens Health (Larchmt), 2013; 22(5):417–425; doi: 10.1089/jwh.2012.4046

17.

Hedlund

, Brembeck

, Olausson

. Determinants of vitamin D status in fair-skinned women of childbearing age at northern latitudes. PLoS One, 2013; 8(4):e60864; doi: 10.1371/journal.pone.0060864

18.

Wyon

, Wolman

, Nevill

, et al. The influence of hormonal contraception on vitamin D supplementation on serum 25(OH)D3 status in premenopausal women: A prospective double-blind placebo random controlled trial. J Endocrinol Metab, 2017; 7(4):117–121; doi: 10.14740/jem441w

19.

Palaniswamy

, Hyppönen

, Williams

DM et al

. Potential determinants of vitamin D in Finnish adults: A cross-sectional study from the Northern Finland birth cohort 1966. BMJ Open, 2017; 7:e013161; doi: 10.1136/bmjopen-2016-013161

20.

Weaver

. Diet, gut microbiome, and bone health. Curr Osteoporos Rep, 2015; 13:125–130; doi: 10.1007/s11914-015-0257-0

21.

Cornes

, Sintes

, Peña

, et al. Daily intake of a functional synbiotic yogurt increases calcium absorption in young adult women. J Nutr, 2022; 152:1647–1654; doi: 10.1093/jn/nxac088

22.

MInisterio de Salud Pública. Dirección General de la Salud. Área Programática de Nutrición. Guia alimentaria para la población uruguaya. MSP, Uruguay, 2016. Available from: https://www.gub.uy/ministerio-salud-publica/comunicacion/publicaciones/guia-alimentaria-para-poblacion-uruguaya [Last accessed: August 4, 2023 ].

23.

ISCD Official Positions – Adult – International Society for Clinical Densitometry (ISCD). Iscd.org. Available from: https://www.iscd.org/official-positions/2019-iscd-official-positions-adult/ [Last accessed: June 14, 2022 ].

24.

World Health Organization. Body Mass Index. Available from: https://www.who.int/data/gho/data/themes/topics/topic-details/GHO/body-mass-index [Last accessed: August 4, 2023 ].

25.

Moliterno

, Donangelo

, Borgarello

, et al. Salt-related practices and its association with 24-hour urinary sodium excretion in an Uruguayan population cohort. Rev Panam Salud Publica, 2022; 46; doi: 10.26633/RPSP.2022.180

26.

Kremer

, Campbell

, Reinhardt

, et al. Vitamin D status and its relationship to body fat, final height, and peak bone mass in young women. J Clin Endocrinol Metab, 2009; 94:67–73; doi: 10.1210/jc.2008-1575

27.

Dominoni

LAC

, Gabiatti

, Piazza

FRG

, et al. Vitamin D is associated with body composition and fat intake, but not with cardiometabolic parameters in adults with obesity. Nutr Res, 2022; 105:97–104; doi: 10.1016/j.nutres.2022.06.006

28.

Reubinoff

, Grubstein

, Berry

, et al. Effects of low-dose estrogen oral contraceptives on weight, body composition, and fat distribution in young women. Fertil Steril, 1995; 63:516–521; doi: 10.1016/s0015-0282(16)57419-6

29.

Mayeda

, Torgal

, Westhoff

. Weight and body composition changes during oral contraceptive use in obese and normal weight women. J Womens Health (Larchmt), 2014; 23(1):38–43; doi: 10.1089/jwh.2012.4241

30.

Gaspard

, Remus

, Gillain

et al. Plasma hormone levels in women receiving new oral contraceptives containing ethinyl estradiol plus levonorgestrel or desogestrel. Contraception, 1983; 27:577–590; doi: 10.1016/0010-7824(83)90023-9

31.

Reape

, DiLiberti

, Hendy

, et al. Effects on serum hormone levels of low-dose estrogen in place of placebo during the hormone-free interval of an oral contraceptive. Contraception, 2008; 77:34–39; doi: 10.1016/j.contraception.2007.09.010

32.

Wang

, Wurtz

, Auro

, et al. Effects of hormonal contraception on systemic metabolism: cross-sectional and longitudinal evidence. Intern J Epidemiol, 2016; 1445–1457; doi: 10.1093/ije/dyw147

33.

Öberg

, Jorde

, Figenschau

, et al. Combined hormonal contraceptives and vitamin D metabolism in adolescent girls. Endocr Connect, 2022; 11:e210395; doi: 10.1530/EC-21-0395

34.

Saarem

, Pedersen

. Sex differences in the hydroxylation of cholecalciferol and of 5fl-cholestane-3a,7x,12a-triol in rat liver. Biochem J, 1987; 247:73–78; doi: 10.1042/bj2470073

35.

Khosla

, Atkinson

, Melton III

, et al. Effects of age and estrogen status on serum parathyroid hormone levels and biochemical markers of bone turnover in women: A population-based study. J Clin Endocrinol Metab, 1997; 82:1522–1527; doi: 10.1210/jcem.82.5.3946

36.

Haarhaus

, Cianciolo

, Barbuto

et al. Alkaline phosphatase: An old friend as treatment target for cardiovascular and mineral bone disorders in chronic kidney disease. Nutrients, 2022; 14; 2124; https://doi.org/10.3390/nu14102124

37.

Shu

, Tan

, Li

, et al. The correlation between serum total alkaline phosphatase and bone mineral density in young adults. BMC Musculoskelet Disord, 2022; 23:46; doi: 10.1186/s12891-022-05438-y

38.

Institute of Medicine. Dietary Reference Intakes: The Essential Guide to Nutrient Requirements. Washington, DC: The National Academic Press; 2006.

39.

Bachrach

. Hormonal contraception and bone health in adolescents. Front Endocrinol, 2020; 11:603; doi: 103389/fendo.2020.00603

40.

Caldeirão

, Orsolini

, da Silva

, et al. Effect of two combinations of low dose oral contraceptives on adolescent bone mass: A clinical trial with 2 years follow-up. Medicine, 2022; 101:37:e30680; doi: 10.1097/MD.0000000000030680

41.

Schini

, Vilaca

, Gossiel

, et al. Bone turnover markers: Basic biology to clinical applications. Endocr Rev, 2023; 44:417–473; doi: 10.1210/endrev/bnac031

42.

Ladang

, Rauch

, Delvin

, et al. Bone turnover markers in children: From laboratory challenges to clinical interpretation. Calcif Tissue Int, 2023; 112:218–232; doi: 10.1007/s00223-022-00964-2