Abstract
Background: Dairy food intake has been associated with infertility; however, little is known with regard to associations with reproductive hormones or anovulation.
Objective: We investigated whether intakes of dairy foods and specific nutrients were associated with reproductive hormone concentrations across the cycle and the risk of sporadic anovulation among healthy women.
Methods: We prospectively measured serum reproductive hormones ≤8 times/menstrual cycle for 2 cycles from 259 regularly menstruating women (mean age: 27.3 y). Dairy food intake was assessed via 24-h dietary recalls 4 times/cycle. Dairy food intakes were assessed by 1) total and low- and high-fat dairy products; 2) dairy nutrients, including fat, lactose, calcium, and phosphorus; and 3) dairy foods, including milk, cheese, butter, cream, yogurt, and ice cream categories. Weighted linear mixed models were used to evaluate associations between dairy nutrients or food intakes and hormone concentrations. Modified Poisson regression models with robust error variance were used to evaluate anovulation. Models were adjusted for age, body mass index, race, physical activity, Mediterranean diet score, total energy, protein, fiber, caffeine, and other hormones.
Results: Each serving increase in total and low- and high-fat dairy foods and all increases in amounts of all dairy nutrients tested were associated with an ∼5% reduction in serum estradiol concentrations but were not associated with anovulation. Total and high-fat dairy food intakes were positively associated with serum luteinizing hormone concentrations. We observed associations between intakes of >0 servings of yogurt (RR: 2.1; 95% CI: 1.2, 3.7) and cream (RR: 1.8; 95% CI: 1.0, 3.2) and a higher risk of sporadic anovulation compared with no intake.
Conclusions: Our study showed associations between increasing dairy food and nutrient intakes and decreasing estradiol concentrations as well as between cream and yogurt intakes and the risk of sporadic anovulation. These results highlight the potential role of dairy in reproductive function in healthy women.
Keywords: dairy, reproductive hormones, estradiol, menstrual cycle, ovulation
Introduction
Dairy products (e.g., milk, cheese, butter, and yogurt) are major sources of energy and nutrients among adults in the United States (1). In particular, milk and milk products are the largest contributors of SFAs, calcium, potassium, and vitamin D in the American diet (1). The 2010 Dietary Guidelines for Americans recommend a higher intake of fat-free or low-fat milk and milk products, rather than full-fat products, as part of a healthy eating pattern (2). Previous studies have identified associations between intakes of total dairy foods and a decreased risk of endometriosis (3) and between low-fat dairy food intake and a decreased risk of delayed menopause (4). In addition, a reduced risk of uterine leiomyomata was associated with increasing dairy food intake (5, 6), which suggests a positive role of dairy food consumption on reproductive health. However, high intakes of low-fat milk have been associated with increased risks of ovarian cancer (7) and a high consumption of total dairy products was associated with a greater risk of endometrial cancer among postmenopausal women (8). These studies suggest that dairy food consumption plays a role in influencing women’s reproductive health, especially in conditions that involve hormonal dysregulation (9).
Studies reported inconsistent associations between dairy food intake and infertility. A case-control study in women in an agricultural region found that drinking ≥3 glasses (1 glass = 8 ounces) of milk/d was protective for female factor infertility (10). On the other hand, a more recent study detected a positive association between low-fat dairy food intake and an increased risk of anovulatory infertility, and likewise, an inverse association between high-fat dairy food intake and anovulatory infertility (11). Another recent study in women undergoing infertility treatment reported a positive association between dairy food intake and live birth in women aged ≥35 y (12). The biological mechanisms that support an association between dairy food intake and ovulatory function are largely unknown. However, impaired ovarian function was observed in rats fed excessive amounts of lactose (13) and galactose (14), which points to potential biological pathways. These results highlight the need for further study of dairy food intake and reproductive function, particularly among healthy women. Our aim was to investigate how intakes of dairy foods and nutrient components are associated with reproductive hormone concentrations and sporadic anovulation among regularly menstruating women.
Methods
Study design and sample collection.
Details of the study design and sample collection are described elsewhere (15). In brief, the BioCycle Study was designed to determine the association of oxidative stress levels with endogenous reproductive hormone and antioxidant concentrations across the menstrual cycle in a prospective cohort of premenopausal women (15–17). The study prospectively recruited 259 healthy, regularly menstruating women with self-reported cycle lengths of 21–35 d for ≥6 mo before recruitment from western New York in 2005–2007. Women aged 18–44 y with a self-reported BMI (in kg/m2) of >18 or <35 at screening were eligible for the study. We excluded women who used hormonal contraceptives in the past 3 mo before screening, who were pregnant or breastfeeding in the past 6 mo, who had a history of ovulatory disorders or uterine abnormalities (e.g., uterine fibroids), or who were unwilling to stop regular intake of vitamin, mineral, or herbal antioxidant supplements during the study. We obtained information on age, race, lifestyle, and reproductive and health history through questionnaires that were completed at baseline. To calculate BMIs used in analyses, trained research staff measured weight and height according to standardized protocols. Each participant completed the 2002 long-form International Physical Activity Questionnaire. High, moderate, and low physical activity categories were calculated on the basis of standard International Physical Activity Questionnaire cutoffs (18).
Each participant provided fasting blood specimens during ≤8 visits/cycle at specific phases of the menstrual cycle: menstruation, mid- and late follicular phase; luteinizing hormone (LH)7 surge; ovulation; and early, mid-, and late luteal phases. The timing of visits was supported by the use of fertility monitors (Clearblue Easy Fertility Monitor; Inverness Medical) and personal cycle-length history (19), and visits were scheduled between 0700 and 0830 to collect fasting blood specimens and to reduce diurnal variation (15). In total, 250 women provided data for 2 cycles and 9 women for 1 cycle; 94% of the participants completed ≥7 clinic visits/cycle. The University at Buffalo Health Sciences Institutional Review Board approved the study and served as the institutional review board designated by the NIH under a reliance agreement. All of the study participants provided written informed consent.
Dietary assessment.
Participants completed a 24-h dietary recall ≤4 times/cycle at the visits corresponding to the following phases: menstruation, mid-follicular, ovulation, and mid-luteal. A total of ≤8 dietary recalls were conducted per woman with ∼87% of women completing 4 dietary recalls/cycle. The recalls were conducted by trained and certified staff either by telephone or in person. These trained dietary interviewers used both open-ended questions and visual amount-estimation tools to capture food and beverage consumption. Food and beverage intakes were analyzed for nutrient components and dairy items by using the Nutrition Data System for Research software version 2005, developed by the Nutrition Coordinating Center, University of Minnesota. Dairy products are major contributors of dietary calcium and phosphorus, and lactose is the main sugar in milk products. Thus, we considered the percentage of energy from fat, lactose, calcium, and phosphorus obtained via dairy food consumption as dairy nutrient components, averaged per cycle. Intakes of dairy products were grouped into specific dairy food categories (i.e., milk, cheese, butter, cream, yogurt, and ice cream) and dichotomized as any intake or no intake. Dairy products were also grouped into low-fat or high-fat dairy foods. Low-fat dairy foods included skim and <2%-fat milks, cottage cheese, and any products with nonfat or low-fat dairy, whereas high-fat dairy foods included the rest of the dairy products. Intakes of desserts, snacks, and mixed dishes containing dairy products were excluded from the analysis due to the difficulty in precisely estimating dairy components that are included in the corresponding recipes; our post hoc analysis that included these categories in the total dairy food intake variable did not change our results substantially.
Reproductive hormone analysis.
Total estradiol, follicle-stimulating hormone (FSH), LH, progesterone, and sex hormone–binding globulin (SHBG) were measured by using solid-phase competitive chemiluminescent enzymatic immunoassays by Specialty Laboratories, Inc., on a DPC Immulite 2000 analyzer (Siemens Medical Solutions Diagnostics) at the Kaleida Health Center for Laboratory Medicine, Buffalo, New York. Details of reproductive hormonal analysis were previously reported (15). Total testosterone concentration (ng/dL) was measured by LC–tandem MS by using a Shimadzu Prominence Liquid Chromatograph (Shimadzu Scientific Instruments, Inc.) with an ABSciex 5500 tandem mass spectrometer (AB Sciex) at the University of Minnesota. Increased sensitivity was obtained with the use of Mobile Phase B (100% acetonitrile) with a low standard of 4 ng/dL added to the standard curve. Free (i.e., bioavailable) estradiol and testosterone and the free androgen index were calculated (20) according to standardized methods (21, 22). The CV was <10% for estradiol and SHBG, <5% for LH and FSH, <14% for progesterone, and <7% for total testosterone. A sporadic anovulatory cycle was defined as a cycle with peak progesterone concentrations ≤5 ng/mL and no observed serum LH peak during the mid- or late luteal phase cycle visit (23). In total, 42 of the 509 cycles (8.3%) in the study were considered anovulatory.
Statistical analysis.
The distribution of demographic variables and dairy nutrient components was compared by quartiles of percentage of energy from dairy fat intake averaged over the study period (Table 1). Chi-square tests and ANOVA were used to test associations between demographic variables and quartiles of mean percentage of energy from dairy fat intake. The results are presented as means ± SDs or percentages.
TABLE 1.
Characteristics of healthy premenopausal women participating in the BioCycle Study by quartile of mean percentage of energy from dairy fat intake across the study period1
| Quartile (% of energy from dairy fat intake) |
||||||
| Overall | 1 (0.0–2.9) | 2 (2 .9–4.7) | 3 (4.7–6.4) | 4 (6.4–15.4) | P | |
| Women, n | 259 | 64 | 65 | 65 | 65 | |
| Demographic characteristics | ||||||
| Age, y | 27.3 ± 8.2 | 25.2 ± 7.1 | 27.3 ± 8.0 | 27.6 ± 8.9 | 29.0 ± 8.5 | 0.07 |
| BMI, kg/m2 | 24.1 ± 3.9 | 24.3 ± 4.5 | 24.2 ± 3.6 | 23.6 ± 3.2 | 24.2 ± 4.1 | 0.78 |
| Physical activity level,2 n (%) | ||||||
| Low | 25 (10) | 6 (9) | 9 (14) | 4 (6) | 6 (9) | 0.29 |
| Moderate | 92 (35) | 28 (44) | 25 (38) | 21 (32) | 18 (28) | |
| High | 142 (55) | 30 (47) | 31 (48) | 40 (62) | 41 (63) | |
| Race, n (%) | ||||||
| White | 154 (59) | 26 (41) | 36 (55) | 37 (57) | 55 (85) | <0.0001 |
| Black | 51 (20) | 19 (30) | 16 (25) | 10 (15) | 6 (9) | |
| Other | 54 (21) | 19 (30) | 13 (20) | 18 (28) | 4 (6) | |
| High school education or less, n (%) | 33 (13) | 8 (13) | 8 (12) | 8 (12) | 9 (14) | 1.00 |
| Current smoker, n (%) | 10 (4) | 1 (2) | 3 (5) | 4 (6) | 2 (3) | 0.71 |
| Married, n (%) | 66 (25) | 10 (16) | 18 (28) | 14 (22) | 24 (37) | 0.04 |
| Nulliparous, n (%) | 187 (72) | 51 (80) | 46 (71) | 50 (77) | 40 (62) | 0.15 |
| Past OC3 use, n (%) | 140 (54) | 29 (45) | 43 (66) | 34 (52) | 34 (52) | 0.15 |
| Nutrient components | ||||||
| Total dairy fat, % of energy | 5.0 ± 3.0 | 1.6 ± 0.9 | 3.9 ± 0.5 | 5.5 ± 0.5 | 8.9 ± 2.3 | <0.0001 |
| Lactose, g/d | 6.7 ± 5.7 | 5.0 ± 6.2 | 5.7 ± 3.9 | 6.4 ± 5.0 | 9.7 ± 6.3 | <0.0001 |
| Calcium, mg/d | 251 ± 169 | 152 ± 162 | 216 ± 117 | 249 ± 124 | 385 ± 177 | <0.0001 |
| Phosphorus, mg/d | 204 ± 138 | 125 ± 132 | 175 ± 94 | 205 ± 11 | 312 ± 145 | <0.0001 |
Values are means ± SDs unless otherwise indicated; n = 259.
Physical activity was calculated on the basis of the standard 2002 long-form International Physical Activity Questionnaire.
OC, oral contraceptive.
We used linear mixed models with random intercepts to evaluate the associations between mean intakes of dairy foods and nutrient components and log-transformed reproductive hormones, including estradiol, free estradiol, FSH, LH, progesterone, SHBG, testosterone, free testosterone, and the free androgen index across the cycle; for progesterone, only measurements during the luteal phase (i.e., early, mid-, and late luteal phases) were included. First, dairy food intake was assessed as total, low-fat, and high-fat dairy food intakes and considered as continuous variables. Mean intakes of total, low-fat, and high-fat dairy products were also examined for their associations with reproductive hormone concentrations at specific phases of the cycle that correspond to menstrual, follicular, periovulatory, and midluteal phases. Second, dairy food intake was evaluated by specific dairy nutrient components, including dairy fat, lactose, calcium, and phosphorus, in quartiles and trends were tested by using the median intake of each nutrient component as a continuous variable. Finally, dairy consumption was examined by specific dairy food categories, such as milk, cheese, butter, cream, yogurt, and ice cream. Milk and cheese were categorized as 0, <1, 1–2, and ≥2 servings/d, whereas intakes of butter, cream, yogurt, and ice cream were examined as a dichotomous variable (i.e., no intake or >0 servings/d) due to a substantial number of women who reported no servings over the cycle and insufficient numbers for further categorization. Models were adjusted for age, BMI, race, physical activity, Mediterranean diet score as a measure of a healthy dietary pattern (24), and total energy, protein, fiber, and caffeine intakes; we did not adjust for education or parity because there was no difference by intake of dairy products. Because concentrations of endogenous reproductive hormones change over the cycle through the complex feedback mechanisms with other hormones, we conducted models that were additionally adjusted for concurrent hormone concentrations by using inverse probability weights (25, 26). Results are presented as percentage changes (Table 2, Supplemental Table 1) or percentage differences (Tables 3 and 4, Supplemental Tables 2 and 3) with 95% CIs.
TABLE 2.
Associations between total, low-fat, and high-fat dairy food intakes and serum reproductive hormone concentrations measured ≤8 times/menstrual cycle for 2 cycles (n = 4072) in healthy premenopausal women1
| Change (95% CI), % |
||
| Model 1 | Model 2 | |
| Total dairy | ||
| Estradiol, pg/mL | −4.6 (−6.9, −2.3) | −4.1 (−6.5, −1.5) |
| Free estradiol, pg/mL | −4.0 (−6.2, −1.7) | −4.0 (−6.3, −1.5) |
| FSH, mIU/mL | 0.2 (−1.9, 2.2) | 0.1 (−2.1, 2.3) |
| LH, ng/mL | 2.9 (0.2, 5.7) | 3.5 (0.5, 6.5) |
| Progesterone,2 ng/mL | −3.1 (−9.2, 3.3) | −1.1 (−7.6, 5.9) |
| SHBG, nmol/L | −0.6 (−2.2, 1.1) | −0.4 (−2.0, 1.2) |
| Testosterone, ng/dL | −0.5 (−1.8, 0.8) | −0.4 (−1.8, 1.0) |
| Free testosterone, ng/dL | −0.4 (−1.8, 1.0) | −0.2 (−1.6, 1.3) |
| Free androgen index | 0.4 (−1.7, 2.5) | 0.3 (−1.8, 2.5) |
| Low-fat dairy3 | ||
| Estradiol, pg/mL | −4.6 (−10.3, 1.5) | −3.2 (−9.4, 3.4) |
| Free estradiol, pg/mL | −2.6 (−8.1, 3.3) | −2.2 (−8.1, 4.1) |
| FSH, mIU/mL | −0.5 (−5.5, 4.7) | 0.6 (−4.8, 6.2) |
| LH, ng/mL | 0.7 (−5.7, 7.5) | 1.5 (−5.4, 9.0) |
| Progesterone,2 ng/mL | 1.3 (−13.5, 18.6) | 5.9 (−9.7, 24.2) |
| SHBG, nmol/L | −1.8 (−6.2, 2.8) | −1.1 (−5.4, 3.3) |
| Testosterone, ng/dL | −3.3 (−6.6, 0.1) | −3.1 (−6.5, 0.5) |
| Free testosterone, ng/dL | −3.2 (−6.7, 0.5) | −2.9 (−6.5, 0.8) |
| Free androgen index | −0.2 (−5.6, 5.6) | −3.4 (−8.9, 2.4) |
| High-fat dairy4 | ||
| Estradiol, pg/mL | −4.7 (−7.2, −2.1) | −4.4 (−7.1, −1.7) |
| Free estradiol, pg/mL | −4.3 (−6.8, −1.8) | −4.3 (−6.9, −1.7) |
| FSH, mIU/mL | 0.3 (−1.9, 2.6) | 0.1 (−2.3, 2.5) |
| LH, ng/mL | 3.4 (0.4, 6.5) | 4.2 (0.9, 7.6) |
| Progesterone,2 ng/mL | −4.1 (−10.7, 3.0) | −2.7 (−9.7, 4.8) |
| SHBG, nmol/L | −0.4 (−2.1, 1.5) | −0.4 (−2.1, 1.4) |
| Testosterone, ng/dL | −0.1 (−1.4, 1.3) | 0.2 (−1.3, 1.6) |
| Free testosterone, ng/dL | 0.1 (−1.4, 1.6) | 0.4 (−1.1, 2.0) |
| Free androgen index | 0.5 (−1.8, 2.8) | 0.9 (−1.4, 3.2) |
Values are % changes in hormone concentrations per 1-serving increase in dairy food intake; n = 259. Model 1 adjusted for total energy intake; age; BMI; race; physical activity; Mediterranean diet score; and fiber, protein, and caffeine intake. Model 2 adjusted for all covariates included in model 1 as well as other reproductive hormone concentrations by using inverse probability weights. FSH, follicle-stimulating hormone; LH, luteinizing hormone; SHBG, sex hormone–binding globulin.
Limited to measurements of progesterone during the luteal phase.
Low-fat dairy includes any dairy products that are nonfat, low-fat, or <2% fat.
High-fat dairy includes all remaining dairy products.
TABLE 3.
Associations between dairy nutrients in the highest quartile compared with the lowest quartile and reproductive hormone concentrations measured ≤8 times/menstrual cycle for 2 cycles (n = 4072) in healthy premenopausal women1
| Model 1 |
Model 2 |
|||
| Dietary nutrients | Difference (95% CI), % | P-trend | Difference (95% CI), % | P-trend |
| Total dairy fat, % of energy | ||||
| Estradiol, pg/mL | −11.0 (−17.4, −4.2) | <0.01 | −11.5 (−18.0, −4.4) | <0.01 |
| Free estradiol, pg/mL | −9.8 (−16.1, −3.0) | <0.01 | −10.9 (−17.3, −3.9) | <0.01 |
| FSH, mIU/mL | −0.5 (−6.5, 5.9) | 0.74 | −0.8 (−7.0, 5.8) | 0.66 |
| LH, ng/mL | 3.9 (−4.2, 12.7) | 0.43 | 6.6 (−2.2, 16.2) | 0.16 |
| Progesterone,2 ng/mL | −10.2 (−26.8, 10.1) | 0.15 | −8.1 (−25.1, 12.6) | 0.25 |
| SHBG, nmol/L | −2.0 (−6.6, 2.7) | 0.61 | −2.7 (−7.0, 1.8) | 0.30 |
| Testosterone, ng/dL | −1.5 (−5.1, 2.1) | 0.45 | −1.4 (−5.0, 2.5) | 0.42 |
| Free testosterone, ng/dL | −1.5 (−5.4, 2.6) | 0.40 | −0.9 (−4.8, 3.1) | 0.51 |
| Free androgen index | 0.9 (−5.0, 7.3) | 0.96 | 1.0 (−4.8, 7.1) | 0.93 |
| Lactose, g/d | ||||
| Estradiol, pg/mL | −10.1 (−17.4, −2.2) | <0.01 | −7.2 (−15.1, 1.4) | 0.06 |
| Free estradiol, pg/mL | −7.6 (−14.8, 0.3) | 0.04 | −6.7 (−14.3, 1.6) | 0.10 |
| FSH, mIU/mL | 4.1 (−2.9, 11.7) | 0.27 | 3.1 (−4.2, 10.9) | 0.49 |
| LH, ng/mL | 3.8 (−5.2, 13.6) | 0.54 | 5.0 (−4.6, 15.7) | 0.40 |
| Progesterone,2 ng/mL | −3.2 (−22.7, 21.1) | 0.72 | 5.6 (−15.3, 31.7) | 0.64 |
| SHBG, nmol/L | −4.0 (−9.5, 1.8) | 0.09 | −4.2 (−9.4, 1.2) | 0.06 |
| Testosterone, ng/dL | −3.7 (−7.9, 0.7) | 0.17 | −2.8 (−7.2, 1.7) | 0.31 |
| Free testosterone, ng/dL | −2.3 (−6.9, 2.6) | 0.55 | −1.6 (−6.2, 3.3) | 0.74 |
| Free androgen index | 3.8 (−3.5, 11.7) | 0.14 | 1.1 (−6.1, 8.9) | 0.22 |
| Calcium, mg/d | ||||
| Estradiol, pg/mL | −11.9 (−19.2, −4.0) | <0.01 | −8.4 (−16.3, 0.2) | 0.02 |
| Free estradiol, pg/mL | −9.4 (−16.6, −1.6) | 0.02 | −7.9 (−15.6, 0.5) | 0.04 |
| FSH, mIU/mL | 3.8 (−3.4, 11.5) | 0.21 | 3.7 (−3.8, 11.7) | 0.23 |
| LH, ng/mL | 6.1 (−3.4, 16.4) | 0.20 | 8.2 (−2.0, 19.5) | 0.08 |
| Progesterone,2 ng/mL | −2.1 (−22.1, 23.1) | 0.62 | 3.4 (−17.5, 29.6) | 0.98 |
| SHBG, nmol/L | −3.6 (−9.1, 2.2) | 0.14 | −4.0 (−9.1, 1.4) | 0.09 |
| Testosterone, ng/dL | −0.3 (−4.7, 4.2) | 0.91 | 0.7 (−3.9, 5.5) | 0.62 |
| Free testosterone, ng/dL | 1.2 (−3.6, 6.2) | 0.47 | 1.3 (−3.5, 6.4) | 0.45 |
| Free androgen index | 7.0 (−0.6, 15.1) | 0.03 | 5.4 (−1.9, 13.2) | 0.09 |
| Phosphorus, mg/d | ||||
| Estradiol, pg/mL | −12.4 (−19.7, −4.4) | <0.01 | −9.5 (−17.4, −0.9) | 0.02 |
| Free estradiol, pg/mL | −9.8 (−17.1, −1.9) | 0.02 | −8.4 (−16.1, 0.0) | 0.04 |
| FSH, mIU/mL | 2.6 (−4.6, 10.3) | 0.36 | 2.1 (−5.3, 10.1) | 0.35 |
| LH, ng/mL | 6.1 (−3.4, 16.6) | 0.19 | 7.7 (−2.6, 19.0) | 0.08 |
| Progesterone,2 ng/mL | −3.7 (−23.7, 21.6) | 0.62 | 1.2 (−19.6, 27.3) | 0.96 |
| SHBG, nmol/L | −5.0 (−10.5, 0.8) | 0.09 | −4.5 (−9.6, 0.9) | 0.10 |
| Testosterone, ng/dL | −1.4 (−5.8, 3.1) | 0.64 | −1.0 (−5.5, 3.8) | 0.79 |
| Free testosterone, ng/dL | 0.0 (−4.8, 5.1) | 0.88 | 0.0 (−4.8, 5.1) | 0.88 |
| Free androgen index | 6.8 (−0.8, 15.1) | 0.06 | 2.0 (−5.5, 10.1) | 0.37 |
n = 259. Model 1 adjusted for total energy intake; age; BMI; race; physical activity; Mediterranean diet score; and fiber, protein, and caffeine intakes. Results for all quartiles for model 1 are reported in Supplemental Table 2. Model 2 adjusted for all covariates included in model 1 as well as for other reproductive hormone concentrations by using inverse probability weights. Results for all quartiles for model 2 are reported in Supplemental Table 3. P-trend was calculated with the median intake of nutrients in each quartile as a continuous variable. FSH, follicle-stimulating hormone; LH, luteinizing hormone; SHBG, sex hormone–binding globulin.
Limited to measurements of progesterone during the luteal phase.
TABLE 4.
Associations between servings of dairy food and reproductive hormone concentrations measured ≤8 times/menstrual cycle for 2 cycles (n = 4072) in healthy premenopausal women1
| Difference (95% CI), % |
||
| Dairy food intake | Model 1 | Model 2 |
| Milk,2 servings/d | ||
| Estradiol, pg/mL | ||
| <1 | −3.1 (−8.7, 3.0) | −0.8 (−6.8, 5.5) |
| ≥1 to <2 | −6.7 (−15.5, 2.9) | −2.7 (−12.2, 7.8) |
| ≥2 | −12.7 (−32.0, 12.0) | −17.1 (−37.7, 10.2) |
| Free estradiol, pg/mL | ||
| <1 | −1.9 (−7.4, 4.0) | −0.9 (−6.6, 5.2) |
| ≥1 to <2 | −4.6 (−13.2, 4.9) | −2.9 (−12.1, 7.3) |
| ≥2 | −4.5 (−24.9, 21.6) | −9.2 (−31.2, 19.9) |
| FSH, mIU/mL | ||
| <1 | −2.6 (−7.3, 2.3) | −2.1 (−7.0, 3.0) |
| ≥1 to <2 | −7.4 (−14.6, 0.4) | −6.9 (−14.6, 1.4) |
| ≥2 | −1.6 (−19.9, 20.9) | 0.2 (−21.9, 28.4) |
| LH, ng/mL | ||
| <1 | −6.1 (−11.9, 0.1) | −4.3 (−10.4, 2.3) |
| ≥1 to <2 | 3.2 (−7.0, 14.5) | 4.1 (−6.8, 16.3) |
| ≥2 | −5.3 (−27.3. 23.4) | −3.1 (−29.4, 33.1) |
| Progesterone,3 ng/mL | ||
| <1 | 7.6 (−8.5, 26.4) | 7.5 (−8.4, 26.1) |
| ≥1 to <2 | −10.1 (−31.3, 17.5) | −4.7 (−27.1, 24.5) |
| ≥2 | 16.7 (−40.0, 127) | 11.8 (−44.9, 127) |
| SHBG, nmol/L | ||
| <1 | −3.6 (−7.3, 0.3) | −2.0 (−5.6, 1.8) |
| ≥1 to <2 | −5.2 (−11.2, 1.2) | −5.5 (−11.3, 0.6) |
| ≥2 | −13.7 (−26.7, 1.6) | −10.3 (−23.9, 5.8) |
| Testosterone, ng/dL | ||
| <1 | −1.6 (−4.5, 1.4) | −0.5 (−3.5, 2.6) |
| ≥1 to <2 | −3.5 (−8.2, 1.3) | −2.6 (−7.4, 2.5) |
| ≥2 | −12.9 (−23.0, −1.4) | −13.9 (−24.9, −1.2) |
| Free testosterone, ng/dL | ||
| <1 | −0.6 (−3.8, 2.7) | 0.2 (−2.9, 3.5) |
| ≥1 to <2 | −2.4 (−7.4, 3.0) | −1.5 (−6.6, 3.9) |
| ≥2 | −18.9 (−29.1, −7.3) | −17.8 (−28.8, 5.2) |
| Free androgen index | ||
| <1 | 2.3 (−2.6, 7.5) | 2.2 (−2.6, 7.3) |
| ≥1 to <2 | 3.1 (−5.0, 11.8) | 2.7 (−5.2, 11.2) |
| ≥2 | −10.3 (−27.0, 10.2) | −8.5 (−26.3, 13.6) |
| Cheese,2 servings/d | ||
| Estradiol, pg/mL | ||
| <1 | 1.0 (−5.4, 7.7) | 0.7 (−5.8, 7.6) |
| ≥1 to <2 | −6.4 (−18.4, 7.3) | −9.9 (−21.6, 3.6) |
| ≥2 | −8.1 (−37.2, 34.3) | −11.7 (−41.5, 33.2) |
| Free estradiol, pg/mL | ||
| <1 | 1.0 (−5.2, 7.5) | 0.3 (−6.1, 7.0) |
| ≥1 to <2 | −8.9 (−20.3, 4.1) | −11.8 (−23.1, 1.2) |
| ≥2 | −3.6 (−33.4, 39.6) | −5.3 (−37.4, 43.3) |
| FSH, mIU/mL | ||
| <1 | −1.3 (−6.4, 4.2) | −1.4 (−6.7, 4.3) |
| ≥1 to <2 | −3.5 (−7.5, 16.1) | 2.9 (−8.6, 16.0) |
| ≥2 | −6.8 (−32.1, 27.9) | −12.4 (−37.8, 23.5) |
| LH, ng/mL | ||
| <1 | −3.1 (−9.5, 3.9) | −4.0 (−10.7, 3.3) |
| ≥1 to <2 | 12.6 (−2.9, 30.6) | 9.4 (−6.4, 28.0) |
| ≥2 | 59.9 (5.9, 141) | 54.6 (−0.5, 140) |
| Progesterone,3 ng/mL | ||
| <1 | 1.3 (−15.0, 20.8) | 0.0 (−16.0, 19.0) |
| ≥1 to <2 | −14.8 (−40.9, 23.0) | −14.6 (−40.7, 23.0) |
| ≥2 | −3.2 (−64.2, 162) | −2.6 (−69.1, 207) |
| SHBG, nmol/L | ||
| <1 | −1.6 (−5.6. 2.5) | −1.5 (−5.2, 2.3) |
| ≥1 to <2 | 8.3 (−0.6, 18.1) | 8.6 (0.1, 17.8) |
| ≥2 | −3.6 (−24.1, 22.4) | −4.1 (−25.5, 23.4) |
| Testosterone, ng/dL | ||
| <1 | −2.4 (−5.4, 0.7) | −3.2 (−6.3, 0.0) |
| ≥1 to <2 | 0.5 (−6.0, 7.3) | −0.1 (−6.7, 7.0) |
| ≥2 | 10.0 (−8.4, 32.2) | 10.6 (−9.8, 35.6) |
| Free testosterone, ng/dL | ||
| <1 | −2.4 (−5.7, 1.0) | −3.2 (−6.5, 0.2) |
| ≥1 to <2 | −0.7 (−7.7, 6.8) | −1.7 (−8.7, 5.7) |
| ≥2 | 13.6 (−7.2, 39.0) | 12.3 (−9.7, 39.7) |
| Free androgen index | ||
| <1 | −1.1 (−6.2, 4.2) | −1.8 (−6.7, 3.5) |
| ≥1 to <2 | −4.3 (−14.3, 7.0) | −5.4 (−15.3, 5.6) |
| ≥2 | 17.1 (−14.1, 59.6) | 15.3 (−17.2, 60.5) |
| Butter4 | ||
| Estradiol, pg/mL | −0.8 (−6.2, 4.9) | −0.7 (−6.3, 5.2) |
| Free estradiol, pg/mL | −1.6 (−6.8, 3.9) | −1.3 (−6.7, 4.4) |
| FSH, mIU/mL | −2.4 (−6.8, 2.2) | −2.4 (−7.0, 2.4) |
| LH, ng/mL | 0.7 (−5.2, 6.9) | 0.1 (−6.1, 6.6) |
| Progesterone,3 ng/mL | −4.7 (−18.1, 10.9) | −4.0 (−17.5, 11.7) |
| SHBG, nmol/L | 1.6 (−2.0, 5.4) | 0.7 (−2.8, 4.3) |
| Testosterone, ng/dL | 0.4 (−2.3, 3.2) | 0.8 (−2.0, 3.8) |
| Free testosterone, ng/dL | 0.1 (−2.9, 3.2) | 0.9 (−2.2, 4.0) |
| Free androgen index | −1.1 (−5.5, 3.6) | 0.1 (−4.4, 4.8) |
| Cream4 | ||
| Estradiol, pg/mL | 0.9 (−4.9, 6.9) | 5.6 (−0.6, 12.1) |
| Free estradiol, pg/mL | 0.7 (−4.8, 6.6) | 4.9 (−1.0, 11.3) |
| FSH, mIU/mL | −2.9 (−7.5, 1.9) | −2.8 (−7.6, 2.2) |
| LH, ng/mL | −3.3 (−9.2, 2.9) | −0.3 (−6.7, 6.5) |
| Progesterone,3 ng/mL | 6.9 (−9.1, 25.7) | 6.4 (−9.3, 24.9) |
| SHBG, nmol/L | −1.3 (−4.9, 2.4) | −0.9 (−4.4, 2.6) |
| Testosterone, ng/dL | 0.1 (−2.7, 3.0) | 0.7 (−2.2, 3.8) |
| Free testosterone, ng/dL | 0.0 (−3.0, 3.2) | 0.9 (−2.2, 4.1) |
| Free androgen index | 0.7 (−3.9, 5.6) | 1.6 (−3.0, 6.5) |
| Yogurt4 | ||
| Estradiol, pg/mL | 1.3 (−4.7, 7.6) | 0.2 (−5.9, 6.7) |
| Free estradiol, pg/mL | 2.0 (−3.8, 8.2) | 1.1 (−4.9, 7.4) |
| FSH, mIU/mL | 0.3 (−4.6, 5.4) | 0.6 (−4.5, 6.0) |
| LH, ng/mL | −0.2 (−6.5, 6.5) | −1.1 (−7.6, 5.9) |
| Progesterone,3 ng/mL | 12.1 (−4.8, 32.0) | 9.9 (−6.6, 29.4) |
| SHBG, nmol/L | −2.4 (−6.2, 1.6) | −2.5 (−6.1, 1.3) |
| Testosterone, ng/dL | −0.8 (−3.7, 2.3) | −0.6 (−3.7, 2.6) |
| Free testosterone, ng/dL | 0.6 (−2.6, 4.1) | 1.0 (−2.3, 4.5) |
| Free androgen index | 3.2 (−1.9, 8.5) | 3.9 (−1.2, 9.2) |
| Ice cream4 | ||
| Estradiol, pg/mL | 3.6 (−2.2, 9.8) | 2.9 (−3.1, 9.2) |
| Free estradiol, pg/mL | 0.2 (−5.3, 6.0) | 0.4 (−5.3, 6.4) |
| FSH, mIU/mL | 3.6 (−1.2, 8.7) | 1.5 (−3.4, 6.7) |
| LH, ng/mL | 3.5 (−2.8, 10.1) | 2.2 (−4.3, 9.2) |
| Progesterone,3 ng/mL | −10.1 (−23.3, 5.4) | −12.4 (−25.0, 2.5) |
| SHBG, nmol/L | 3.1 (−0.7, 6.9) | 3.6 (0.0, 7.3) |
| Testosterone, ng/dL | −0.5 (−3.3, 2.3) | −1.7 (−4.5, 1.2) |
| Free testosterone, ng/dL | −1.6 (−4.6, 1.4) | −2.6 (−5.6, 0.5) |
| Free androgen index | −3.4 (−7.9, 1.2) | −4.0 (−8.4, 0.6) |
n = 259. Model 1 adjusted for total energy intake; age; BMI; race; physical activity; Mediterranean diet score; and fiber, protein, and caffeine intake. Model 2 adjusted for other hormones and covariates that were included in model 1. FSH, follicle-stimulating hormone; LH, luteinizing hormone; SHBG, sex hormone–binding globulin.
Categorized as 0 (reference), <1, ≥1 to <2, or ≥2 servings/d.
Limited to measurements of progesterone during the luteal phase.
Categorized as no intake (reference) vs. >0 servings/d.
We used Poisson regression with robust error variance to estimate RRs and 95% CIs for the associations between cycle-averaged intakes of dairy foods and nutrient components and sporadic anovulation. Intakes of total and high-fat dairy were evaluated as <1, ≥1 to <2, and ≥2 servings, whereas intakes of low-fat dairy products and milk were examined as 0, <1, and ≥1 servings. Other specific dairy food categories (i.e., cheese, butter, cream, yogurt, and ice cream) were treated as dichotomous variables (i.e., no intake compared with >0 servings/d), and nutrient components were evaluated in quartiles. Stratified analysis was performed separately by low- compared with high-fat products for total dairy foods, milk, and cheese. Anovulation models were also adjusted for age, BMI, race, physical activity, Mediterranean diet score, and total energy, protein, fiber, and caffeine intakes. We further compared models that additionally adjusted for sugar intake. SAS version 9.4 (SAS Institute) was used for all statistical analyses.
Results
Mean ± SD values for age and BMI among the 259 women in our study were 27.3 ± 8.2 y and 24.1 ± 3.9, respectively (Table 1). Most participants were physically active (90%), either white (59%) or black (20%), high school graduates (87%), and nonsmokers (96%). Total dairy fat intake as a percentage of calories was associated with race, in that white women were more likely to be in the upper quartile of total dairy fat intake than other races. Women in the highest quartile of calories from dairy fat also had significantly higher mean lactose, calcium, and phosphorus intakes than did women in the lower quartiles. The median number of servings of mean daily total dairy food intakes was 1.3 (range: 0–7.5 servings). Approximately 81.1%, 89.2%, 56.4%, 48.6%, 42.1%, and 43.2% of women reported the consumption of milk, cheese, butter, cream, yogurt, and ice cream, respectively, ≥1 time during the dietary data collection.
For every serving increase, the intake of dairy food was associated with a 4.6% (95% CI: −6.9%, −2.3%) decrease in serum estradiol and a 4.0% (95% CI: −6.2%, −1.7%) decrease in free estradiol concentrations (Table 2). Total dairy food intake was associated with a 2.9% (95% CI: 0.2%, 5.7%) increase in LH concentrations across the cycle. For associations of intakes of low- and high-fat dairy products and reproductive hormones, results were similar, although they were significant only for high-fat dairy products. Associations between total, low-fat, and high-fat dairy food intakes and reproductive hormones were consistent at specific phases of the cycle (i.e., menstrual, follicular, periovulatory, and midluteal phases; Supplemental Table 1). We found that an increase in dairy nutrients, which include dairy fat, lactose, calcium, and phosphorus, was associated with an ∼10% reduction in concentrations of estradiol and free estradiol (Table 3, Supplemental Table 2). Although the associations between dairy nutrients and hormones were attenuated in general, results were consistent after adjusting for all other hormones in addition to the aforementioned covariates (Table 3, Supplemental Table 3).
Compared with no intake, intakes of ≥2 servings milk/d were inversely associated with testosterone and free testosterone concentrations, although estimates were imprecise (Table 4). Intakes of ≥2 servings cheese/d were associated with increases in LH concentrations in models that adjusted for both confounders and other hormones, relative to no intake. When comparing intakes of >0 servings/d to no intake, ice cream was associated with SHBG (3.6% difference; 95% CI: 0.0%, 7.3%) after additionally adjusting for other hormones. We did not detect any significant associations between other specific dairy food categories, such as butter, cream, and yogurt, and reproductive hormones.
The intake of cream was associated with an 83% increased risk of sporadic anovulation (RR: 1.8; 95% CI: 1.0, 3.2) and yogurt was associated with a higher risk of sporadic anovulation (RR: 2.1; 95% CI: 1.2, 3.7) compared with no intake (Table 5). Even after adjusting for total sugar intake, we identified an increased risk of anovulation associated with both cream (RR: 1.8; 95% CI: 1.0, 3.3) and yogurt (RR: 2.3; 95% CI: 1.3, 4.3) intakes. No associations were found with total, low- and high-fat dairy products, dairy nutrient components, or other types of dairy food and sporadic anovulation. We did not detect significant associations between intakes of milk and cheese with sporadic anovulation, even after stratification by low- and high-fat dairy products (data not shown).
TABLE 5.
Associations between dairy nutrient and food intakes in quartiles and risks of sporadic anovulation among 509 cycles in healthy premenopausal women with the use of generalized linear models in the BioCycle Study (2005–2007)1
| Cycle, n |
RR (95% CI) |
|||
| Quartile | Ovulatory | Anovulatory | Model 1 | Model 2 |
| Dietary nutrients | ||||
| Total dairy fat, % of energy | ||||
| 1 (reference) | 117 | 10 | 1.0 (1.0, 1.0) | 1.0 (1.0, 1.0) |
| 2 | 116 | 11 | 1.3 (0.6, 3.1) | 1.4 (0.6, 3.2) |
| 3 | 117 | 11 | 1.4 (0.6, 3.3) | 1.4 (0.6, 3.3) |
| 4 | 117 | 10 | 1.5 (0.6, 3.6) | 1.3 (0.6, 3.7) |
| Lactose, g/d | ||||
| 1 (reference) | 114 | 13 | 1.0 (1.0, 1.0) | 1.0 (1.0, 1.0) |
| 2 | 117 | 10 | 0.9 (0.5, 1.9) | 1.0 (0.5, 2.1) |
| 3 | 119 | 9 | 0.7 (0.3, 1.6) | 0.8 (0.3, 1.7) |
| 4 | 117 | 10 | 0.7 (0.3, 1.8) | 0.8 (0.3, 2.2) |
| Calcium, mg/d | ||||
| 1 (reference) | 116 | 11 | 1.0 (1.0, 1.0) | 1.0 (1.0, 1.0) |
| 2 | 117 | 10 | 1.0 (0.4, 2.2) | 1.0 (0.5, 2.3) |
| 3 | 112 | 16 | 1.5 (0.7, 3.1) | 1.6 (0.7, 3.4) |
| 4 | 122 | 5 | 0.5 (0.2, 1.4) | 0.5 (0.2, 1.6) |
| Phosphorus, mg/d | ||||
| 1 (reference) | 116 | 11 | 1.0 (1.0, 1.0) | 1.0 (1.0, 1.0) |
| 2 | 116 | 11 | 1.0 (0.5, 2.2) | 1.1 (0.5, 2.4) |
| 3 | 115 | 13 | 1.3 (0.6, 2.9) | 1.4 (0.6, 3.2) |
| 4 | 120 | 7 | 0.7 (0.3, 1.7) | 0.8 (0.3, 2.0) |
| Dairy foods, servings/d | ||||
| Total dairy | ||||
| <1 (reference) | 174 | 16 | 1.0 (1.0, 1.0) | 1.0 (1.0, 1.0) |
| ≥1 to <2 | 172 | 15 | 1.1 (0.6, 2.0) | 1.2 (0.7, 2.1) |
| ≥2 | 121 | 11 | 1.4 (0.6, 3.3) | 1.4 (0.6, 3.4) |
| Low-fat dairy2 | ||||
| 0 (reference) | 147 | 18 | 1.0 (1.0, 1.0) | 1.0 (1.0, 1.0) |
| >0 to <1 | 253 | 19 | 0.7 (0.4, 1.4) | 0.7 (0.4, 1.5) |
| ≥1 | 67 | 5 | 0.7 (0.3, 1.9) | 0.8 (0.3, 2.2) |
| High-fat dairy3 | ||||
| <1 (reference) | 262 | 24 | 1.0 (1.0, 1.0) | 1.0 (1.0, 1.0) |
| ≥1 to <2 | 130 | 14 | 1.3 (0.7, 2.3) | 1.3 (0.7, 2.3) |
| ≥2 | 75 | 4 | 1.0 (0.3, 3.1) | 0.9 (0.3, 2.9) |
| Milk | ||||
| 0 (reference) | 143 | 17 | 1.0 (1.0, 1.0) | 1.0 (1.0, 1.0) |
| >0 to <1 | 279 | 23 | 0.8 (0.4, 1.4) | 0.8 (0.4, 1.5) |
| ≥1 | 45 | 2 | 0.4 (0.1, 1.8) | 0.4 (0.1, 1.9) |
| Cheese | ||||
| 0 (reference) | 99 | 13 | 1.0 (1.0, 1.0) | 1.0 (1.0, 1.0) |
| >0 | 368 | 29 | 0.8 (0.4, 1.6) | 0.8 (0.4, 1.7) |
| Butter | ||||
| 0 (reference) | 274 | 28 | 1.0 (1.0, 1.0) | 1.0 (1.0, 1.0) |
| >0 | 193 | 14 | 0.9 (0.4, 1.7) | 0.9 (0.4, 1.6) |
| Cream | ||||
| 0 (reference) | 310 | 25 | 1.0 (1.0, 1.0) | 1.0 (1.0, 1.0) |
| >0 | 157 | 17 | 1.8 (1.0, 3.2) | 1.8 (1.0, 3.3) |
| Yogurt | ||||
| 0 (reference) | 335 | 22 | 1.0 (1.0, 1.0) | 1.0 (1.0, 1.0) |
| >0 | 132 | 20 | 2.1 (1.2, 3.7) | 2.3 (1.3, 4.3) |
| Ice cream | ||||
| 0 (reference) | 335 | 29 | 1.0 (1.0, 1.0) | 1.0 (1.0, 1.0) |
| >0 | 132 | 13 | 1.2 (0.6, 2.2) | 1.2 (0.7, 2.2) |
n = 259. Model 1 adjusted for energy intake; age; BMI; race; physical activity; Mediterranean diet score; and fiber, protein, and caffeine intakes. Model 2 adjusted for covariates that were included in model 1and sugar intake.
Low-fat dairy includes any dairy products that are nonfat, low-fat, or <2% fat.
High-fat dairy includes all remaining dairy products.
Discussion
We detected inverse associations between dairy food intakes and concentrations of estradiol and free estradiol. Increasing intakes of selected nutrients via dairy products was also associated with moderate reductions in concentrations of estradiol and free estradiol, although these changes did not translate into subsequent associations with sporadic anovulation in our study. Nevertheless, we found a significant association between intakes of cream (i.e., >0 servings/d; range: 0–8 servings/d) and yogurt (i.e., >0 servings/d; range: 0–2 servings/d) relative to no intake and sporadic anovulation, which suggested the potential importance of specific dairy food intake on ovulatory function among healthy, regularly menstruating women. Although a confirmatory study is needed, these findings add to a growing body of evidence that indicate a role of dairy food intake in reproductive function in healthy women, with potential implications for fertility.
Although a lack of similar studies on dairy food intake and reproductive hormones in healthy women limits direct comparison, our observations are in line with the results observed in experimental studies. An animal study found significantly lower concentrations of estradiol and elevated FSH and LH concentrations in rats fed high amounts of galactose, a breakdown component of lactose, compared with controls (14). In addition, progesterone concentrations were significantly decreased in rats after lactose treatment, whereas no differences were found in serum concentrations of estradiol (13). Here, we did not detect any associations between dairy food intake and lactose and progesterone, but our overall results imply modest changes in reproductive hormones, particularly estradiol and LH, with higher consumption of these foods.
We previously detected reductions of ∼5% in estradiol concentrations and a subsequent increase in the risk of anovulation with increasing fiber intake (27). Although we also observed reductions of ∼4–5% in estradiol concentrations with increasing dairy food intake, we did not find such associations with sporadic anovulation, which may imply that the changes in reproductive hormone concentrations due to dairy food intake are not sufficient to affect ovulatory function in healthy women with regular menstrual cycles. The specific biological mechanisms linking dairy food intake and reproductive hormones or anovulation are still unknown; yet, such an inverse association may support findings for conditions that are related to changes in estradiol concentrations, such as uterine fibroids (6). Thus, further work is needed to better understand whether such biological mechanisms are at play in healthy women.
Results from previous studies that explored dairy food intakes and ovulatory function are inconsistent. In a small intervention study in which 5 women drank 500 mL milk/d for 21 d, menstruation and ovulation occurred regularly in 4 women, although anovulation was detected in 1 woman who suffered from oligomenorrhea (9). The consumption of ≥3 glasses of milk was protective for female fertility in a case-control study in 322 infertile women and 322 controls (10), but no association between dairy food intake and anovulatory infertility was reported in a more recent study that used the Nurses’ Health Study II data (11). However, when stratified by dairy fat, the latter study found that the consumption of >2 servings low-fat dairy food/d compared with <1 serving/wk increased the risk of anovulatory infertility (RR: 1.9; 95% CI: 1.2, 2.8), whereas the intake of ≥1 serving high-fat dairy food/d was protective of anovulatory infertility (RR: 0.7; 95% CI: 0.5, 1.0) (11). In another recent study, a positive association between total dairy food intake and live birth was found among women aged ≥35 y who were undergoing infertility treatment (12). Although these studies highlight inconsistent associations between dairy food intake and fertility, all suggest a potential role for dairy food intake in influencing female reproductive health. Previous findings from our data suggest that dietary fat was not associated with anovulation (28), and we did not detect any associations between dairy fat and sporadic anovulation in this analysis. Even after dairy consumption was stratified by low-fat and high-fat dairy, no significant associations were detected. Different results across studies may be due to differing study populations and outcomes of interest. In addition, each study used different types of dietary assessment methods (e.g., FFQ or 24-h diet recall) or adjusted for different covariates in the models.
In 2000, the American Heart Association recommended increasing intakes of nonfat or low-fat milk and milk products to reduce the consumption of SFAs and subsequently reduce the risk of cardiovascular disease in the general population (29). We found significant associations between total dairy and high-fat dairy food intake and concentrations of estradiol, free estradiol, and LH, and our observations for low-fat dairy products were similar, although imprecise. This suggests that there were no differences in associations based on dairy fat consumption with regard to reproductive hormones or ovulatory function in women with regular menstrual cycles.
Interestingly, we detected an association between intakes of cream and yogurt and an increased risk of anovulation. This was consistent with findings from the Nurses’ Health Study II in which high intakes of yogurt were associated with anovulatory infertility (11). Previously, we found that sweetened beverages were associated with changes in reproductive hormone concentrations, although their consumption did not appear to influence ovulatory function (30). Our results remained after we adjusted for total sugar intake or added sugar (data not shown), which suggests that the risk of anovulation was independent of the sugar content included in many flavored yogurt products. It is not clear if our findings were by chance or based on complex, yet unknown biological mechanisms. One possible explanation could be related to the yogurt-manufacturing process, which involves standardizing the fat and solid nonfat content to improve yogurt texture (31). During this process, specific FAs or fat-soluble components that could affect ovulation may be added or removed. This may be the case for cream as well, which is also a substantially processed dairy product. Differences in eating patterns and sporadic anovulation by demographic and lifestyle factors could be related to our observations. However, there were no substantial differences in age, BMI, race, education, and physical activity between women by ovulatory status in our data (32), and our results remained after adjustment for potential confounding factors.
Our study has several strengths. To minimize potential measurement error, we used multiple 24-h dietary recalls, which are considered to be valid measures of dietary exposures (33, 34), although there may be seasonal variations in diet. All of the participants consumed ≥1 of the specific types of dairy products; thus, dairy intolerance or allergy was likely not present in this population. In addition, only women who were not planning to undertake a special diet or were taking supplements (e.g., vitamins and minerals) during the study period were included in the study, which enabled the generalizability of our findings. The prospective study design in the BioCycle Study strengthened the ability to draw inferences by reducing the potential for bias from known risk factors for sporadic anovulation among healthy women with regular menstrual cycles. We did not include women with irregular menses or anovulation history in the BioCycle Study, and therefore, our results are generalizable only to regularly menstruating women. Our measurement of multiple serum hormone concentrations, including progesterone, at specific time points across the menstrual cycle was determined by fertility monitors, which improved our ability to adequately capture hormonal variability across the cycle and the luteal phase.
Nevertheless, there are limitations in our study. Although we were able to identify and group desserts, snacks, and mixed dishes containing dairy products, we did not include them in the analysis due to the difficulty in estimating the proportion of dairy components within specific recipes. Because some dishes could contain a substantial amount of dairy products, some misclassification is also possible. However, our post hoc analysis, which added desserts, snacks, and mixed dishes into the total dairy food intake, did not change our results substantially. Selected nutrient components, particularly calcium and phosphorus, are also abundant in nondairy foods and are often added to processed foods. Because we quantified intakes of selected nutrients obtained from dairy foods only, our results should not be generalized to those nutrients from all sources. Although the major estrogen in milk is estrone rather than estradiol (35), there is a possibility that our results could have been affected by an exposure to exogenous steroid hormones through the intake of commercial milk and milk products from pregnant cows (35, 36). It is tempting to speculate the contribution of exogenous hormones to our results; however, values of naturally occurring hormones in the dairy products consumed by our study participants were unavailable to us.
To our knowledge, this is the first study investigating associations between dairy food intake and reproductive hormones and sporadic anovulation among regularly menstruating women. Our study results showed modest changes in reproductive hormones, particularly estradiol, with increasing total dairy food intake, but these changes were not associated with sporadic anovulation. However, we confirmed a previously reported association, specifically of yogurt consumption, with increased sporadic anovulation. This may necessitate further research to better understand the biological roles of types of dairy food on ovulatory function in women.
Acknowledgments
JW-W and SLM designed and conducted the research and provided data; KK performed the statistical analyses; KK and SLM wrote the manuscript and had primary responsibility for final content; and KK, JW-W, KAM, TCP, ENC, LAS, and SLM interpreted the data and critically revised the manuscript for important intellectual content. All authors read and approved the final manuscript.
Footnotes
Abbreviations used: FSH, follicle-stimulating hormone; LH, luteinizing hormone; SHBG, sex hormone–binding globulin.
References
- 1.O’Neil CE, Keast DR, Fulgoni VL, Nicklas TA. Food sources of energy and nutrients among adults in the US: NHANES 2003–2006. Nutrients 2012;4:2097–120. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. US Department of Agriculture; US Department of Health and Human Services. Dietary guidelines for Americans, 2010. Washington (DC): US Government Printing Office; 2010. [Google Scholar]
- 3.Harris HR, Chavarro JE, Malspeis S, Willett WC, Missmer SA. Dairy-food, calcium, magnesium, and vitamin D intake and endometriosis: a prospective cohort study. Am J Epidemiol 2013;177:420–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Carwile JL, Willett WC, Michels KB. Consumption of low-fat dairy products may delay natural menopause. J Nutr 2013;143:1642–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Wise LA, Radin RG, Palmer JR, Kumanyika SK, Rosenberg L. A prospective study of dairy intake and risk of uterine leiomyomata. Am J Epidemiol 2010;171:221–32. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Wise LA, Palmer JR, Ruiz-Narvaez E, Reich DE, Rosenberg L. Is the observed association between dairy intake and fibroids in African Americans explained by genetic ancestry? Am J Epidemiol 2013;178:1114–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Fairfield KM, Hunter DJ, Colditz GA, Fuchs CS, Cramer DW, Speizer FE, Willett WC, Hankinson SE. A prospective study of dietary lactose and ovarian cancer. Int J Cancer 2004;110:271–7. [DOI] [PubMed] [Google Scholar]
- 8.Ganmaa D, Cui X, Feskanich D, Hankinson SE, Willett WC. Milk, dairy intake and risk of endometrial cancer: a 26-year follow-up. Int J Cancer 2012;130:2664–71. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Maruyama K, Oshima T, Ohyama K. Exposure to exogenous estrogen through intake of commercial milk produced from pregnant cows. Pediatr Int 2010;52:33–8. [DOI] [PubMed] [Google Scholar]
- 10.Greenlee AR, Arbuckle TE, Chyou PH. Risk factors for female infertility in an agricultural region. Epidemiology 2003;14:429–36. [DOI] [PubMed] [Google Scholar]
- 11.Chavarro JE, Rich-Edwards JW, Rosner B, Willett WC. A prospective study of dairy foods intake and anovulatory infertility. Hum Reprod 2007;22:1340–7. [DOI] [PubMed] [Google Scholar]
- 12.Afeiche MC, Chiu YH, Gaskins AJ, Williams PL, Souter I, Wright DL, Hauser R, Chavarro JE. Dairy intake in relation to in vitro fertilization outcomes among women from a fertility clinic. Hum Reprod 2016;31:563–71. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Liu G, Shi F, Blas-Machado U, Duong Q, Davis VL, Foster WG, Hughes CL. Ovarian effects of a high lactose diet in the female rat. Reprod Nutr Dev 2005;45:185–92. [DOI] [PubMed] [Google Scholar]
- 14.Bandyopadhyay S, Chakrabarti J, Banerjee S, Pal AK, Goswami SK, Chakravarty BN, Kabir SN. Galactose toxicity in the rat as a model for premature ovarian failure: an experimental approach readdressed. Hum Reprod 2003;18:2031–8. [DOI] [PubMed] [Google Scholar]
- 15.Wactawski-Wende J, Schisterman EF, Hovey KM, Howards PP, Browne RW, Hediger M, Liu A, Trevisan M. BioCycle Study: design of the longitudinal study of the oxidative stress and hormone variation during the menstrual cycle. Paediatr Perinat Epidemiol 2009;23:171–84. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Schisterman EF, Gaskins AJ, Mumford SL, Browne RW, Yeung E, Trevisan M, Hediger M, Zhang C, Perkins NJ, Hovey K, et al. Influence of endogenous reproductive hormones on F2-isoprostane levels in premenopausal women: the BioCycle Study. Am J Epidemiol 2010;172:430–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Mumford SL, Browne RW, Schliep KC, Schmelzer J, Plowden TC, Michels KA, Sjaarda LA, Zarek SM, Perkins NJ, Messer LC, et al. Serum antioxidants are associated with serum reproductive hormones and ovulation among healthy women. J Nutr 2016;146:98–106. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Craig CL, Marshall AL, Sjostrom M, Bauman AE, Booth ML, Ainsworth BE, Pratt M, Ekelund U, Yngve A, Sallis JF, et al. International Physical Activity Questionnaire: 12-country reliability and validity. Med Sci Sports Exerc 2003;35:1381–95. [DOI] [PubMed] [Google Scholar]
- 19.Howards PP, Schisterman EF, Wactawski-Wende J, Reschke JE, Frazer AA, Hovey KM. Timing clinic visits to phases of the menstrual cycle by using a fertility monitor: the BioCycle Study. Am J Epidemiol 2009;169:105–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Sjaarda LA, Schisterman EF, Schliep KC, Plowden T, Zarek SM, Yeung E, Wactawski-Wende J, Mumford SL. Dietary carbohydrate intake does not impact insulin resistance or androgens in healthy, eumenorrheic women. J Clin Endocrinol Metab 2015;100:2979–86. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Sartorius G, Ly LP, Sikaris K, McLachlan R, Handelsman DJ. Predictive accuracy and sources of variability in calculated free testosterone estimates. Ann Clin Biochem 2009;46:137–43. [DOI] [PubMed] [Google Scholar]
- 22.Södergård R, Bäckström T, Shanbhag V, Carstensen H. Calculation of free and bound fractions of testosterone and estradiol-17 beta to human plasma proteins at body temperature. J Steroid Biochem 1982;16:801–10. [DOI] [PubMed] [Google Scholar]
- 23.Lynch KE, Mumford SL, Schliep KC, Whitcomb BW, Zarek SM, Pollack AZ, Bertone-Johnson ER, Danaher M, Wactawski-Wende J, Gaskins AJ, et al. Assessment of anovulation in eumenorrheic women: comparison of ovulation detection algorithms. Fertil Steril 2014;102:511–8, e2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Trichopoulou A, Costacou T, Bamia C, Trichopoulos D. Adherence to a Mediterranean diet and survival in a Greek population. N Engl J Med 2003;348:2599–608. [DOI] [PubMed] [Google Scholar]
- 25.Robins JM, Hernan MA, Brumback B. Marginal structural models and causal inference in epidemiology. Epidemiology 2000;11:550–60. [DOI] [PubMed] [Google Scholar]
- 26.Cole SR, Hernan MA. Constructing inverse probability weights for marginal structural models. Am J Epidemiol 2008;168:656–64. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Gaskins AJ, Mumford SL, Zhang C, Wactawski-Wende J, Hovey KM, Whitcomb BW, Howards PP, Perkins NJ, Yeung E, Schisterman EF. Effect of daily fiber intake on reproductive function: the BioCycle Study. Am J Clin Nutr 2009;90:1061–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Mumford SL, Chavarro JE, Zhang C, Perkins NJ, Sjaarda LA, Pollack AZ, Schliep KC, Michels KA, Zarek SM, Plowden TC, et al. Dietary fat intake and reproductive hormone concentrations and ovulation in regularly menstruating women. Am J Clin Nutr 2016;103:868–77. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Krauss RM, Eckel RH, Howard B, Appel LJ, Daniels SR, Deckelbaum RJ, Erdman JW, Kris-Etherton P, Goldberg IJ, Kotchen TA, et al. AHA dietary guidelines—revision 2000: a statement for healthcare professionals from the Nutrition Committee of the American Heart Association. Circulation 2000;102(18):2284–99. [DOI] [PubMed] [Google Scholar]
- 30.Schliep KC, Schisterman EF, Mumford SL, Pollack AZ, Perkins NJ, Ye A, Zhang CJ, Stanford JB, Porucznik CA, Hammoud AO, et al. Energy-containing beverages: reproductive hormones and ovarian function in the BioCycle Study. Am J Clin Nutr 2013;97:621–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Sfakianakis P, Tzia C. Conventional and innovative processing of milk for yogurt manufacture; development of texture and flavor: a review. Foods 2014;3:176–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Hambridge HL, Mumford SL, Mattison DR, Ye A, Pollack AZ, Bloom MS, Mendola P, Lynch KL, Wactawski-Wende J, Schisterman EF. The influence of sporadic anovulation on hormone levels in ovulatory cycles. Hum Reprod 2013;28:1687–94. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Schatzkin A, Kipnis V, Carroll RJ, Midthune D, Subar AF, Bingham S, Schoeller DA, Troiano RP, Freedman LS. A comparison of a food frequency questionnaire with a 24-hour recall for use in an epidemiological cohort study: results from the biomarker-based Observing Protein and Energy Nutrition (OPEN) Study. Int J Epidemiol 2003;32:1054–62. [DOI] [PubMed] [Google Scholar]
- 34.Subar AF, Kipnis V, Troiano RP, Midthune D, Schoeller DA, Bingham S, Sharbaugh CO, Trabulsi J, Runswick S, Ballard-Barbash R, et al. Using intake biomarkers to evaluate the extent of dietary misreporting in a large sample of adults: the OPEN study. Am J Epidemiol 2003;158:1–13. [DOI] [PubMed] [Google Scholar]
- 35.Malekinejad H, Scherpenisse P, Bergwerff AA. Naturally occurring estrogens in processed milk and in raw milk (from gestated cows). J Agric Food Chem 2006;54:9785–91. [DOI] [PubMed] [Google Scholar]
- 36.Hartmann S, Lacorn M, Steinhart H. Natural occurrence of steroid hormones in food. Food Chem 1998;62:7–20. [Google Scholar]