The Relationship Between Physical Activity and Vitamin D Status in Postpartum Lactating and Formula-Feeding Women

Abstract

Existing research shows an association between physical activity levels and vitamin D status in the elderly, men, women, children, and adolescent populations. This association has not yet been investigated in postpartum women. We hypothesized that based on the relationship between vitamin D and physical activity found in other populations, greater physical activity levels in postpartum women will be associated with higher serum 25(OH)D levels. A post hoc analysis of 286 postpartum women with self reported physical activity data from the America on the Move survey, and measured circulating serum 25(OH)D (measured by RIA) as an indicator of vitamin D status, was gathered at baseline (4–6 weeks postpartum), 4 months, and 7 months postpartum. The data were analyzed using SAS 9.4 (Cary, NC). 39.9% of women at visit 1 (baseline), 52.8% of women at visit 4 (month 4), and 55.9% of women at visit 7 (month 7) were meeting the NIH recommendation of 150 minutes of moderate intensity (3–6 METs) physical activity per week. Significant differences were seen in physical activity by race (p=0.0407). Caucasians were more likely to meet the standard recommendation than African Americans or Hispanics. Using multiple regression models to examine associations between duration of physical activity and serum 25(OH)D concentration, controlling for race, BMI, feeding type, and METs, it was found that at visit 1, total minutes of physical activity performed per week was positively associated with serum 25(OH)D (p=.03) and achieving at least 2.5 hrs/wk of physical activity had a trending association with an increase in 25(OH)D of 7.23 nmol/L (p=.05). At visit 4 (also controlling for treatment group and sun exposure), achieving at least 1.5 hrs/wk of physical activity was associated with an increase in 25(OH)D of 11.73 nmol/L (p=.04). By visit 7 no association between physical activity and maternal 25(OH)D was observed. In a repeated measures, mixed model analysis predicting maternal 25(OH)D during the study, achieving at least the recommended 150 minutes per week of physical activity (>2.5 hours) was not significantly associated with vitamin D status (pNS). Postpartum physical activity levels differ across race/ethnicity. Consistent with existing data of other population groups, greater physical activity was associated with higher serum 25(OH)D levels at postpartum visits 1 and 4. While no definitive conclusions can be drawn regarding precise levels of physical activity influencing 25(OH)D levels in postpartum women, the data suggest that increased activity during the first 4 months after birth is associated with improved vitamin D status. Additional research is needed because of the inconsistency seen at visit 7.

Introduction

Sufficient maternal vitamin D levels in postpartum women are essential for transfer to breast milk for their infants. Breast milk is known to be the gold standard of nutrition for infants due to its essential nutrients and immune factors, however it is often naturally deficient in vitamin D (1). The parent compound, cholecalciferol, is the most abundant form of vitamin D to transfer to the infant through breast milk; however, maternal cholecalciferol is quickly converted to 25-hydroxyvitamin D (25(OH)D) (2). This is the basis for maternal vitamin D supplementation. According to a study published in 2015, maternal vitamin D supplementation at 6000 IU per day provides adequate levels of vitamin D to infants through breast milk (1).

Various factors such as race, sun exposure, and body mass index (BMI) are all associated with serum vitamin D levels (1). Research has also shown an association between physical activity levels and vitamin D in the elderly, men, women, and children and adolescent populations.

In a study conducted by Kluczynski et al, positive associations were found between serum 25(OH)D concentrations and duration of recreational activity and yard work in postmenopausal women (3). A German study reported significant associations were found between walking duration and 25(OH)D serum levels in all seasons except summer in the elderly population (4). A Tasmanian study found that after adjustment for sun exposure, higher levels of physical activity at baseline and greater levels of physical activity over the two year study period (assessed by a pedometer) was associated with increased 25(OH)D levels, suggesting that increases in 25(OH)D as a result of physical activity may be occurring by a mechanism other than sun exposure (5). A study conducted in Australia found that self reported exercise frequency was a significant predictor of serum 25(OH)D concentrations in adolescents, and that a lack of physical activity is a potential factor that may contribute to the relatively high level of vitamin D insufficiency in adults across Australia (6). While the data in women of child bearing age is sparse, one study conducted in Germany found that physical inactivity was a significant risk factor for maternal vitamin D deficiency after controlling for season, vitamin D supplementation, and time spent outdoors (7).

While studies have shown the effects of exercise on infant acceptance of breast milk, milk volumes and composition, and bone density in lactating women, none have examined the relationship between exercise and vitamin D in postpartum women(8–10). Because of the importance of vitamin D in breast milk, determining whether a woman can proactively increase or sustain their vitamin D levels while breastfeeding by achieving a certain level of physical activity is a vital subject that has not yet been investigated.

This study was designed to investigate the relationship between vitamin D levels and physical activity in postpartum women. We hypothesized that based on the relationship between vitamin D and physical activity found in other populations, greater physical activity levels in postpartum women will be associated with greater serum vitamin D levels.

Materials and Methods

Participants

Figure 1 shows the enrollment and post hoc analysis of 564 women from a randomized, double-blind, comparative effectiveness trial of 3 doses of vitamin D supplementation in lactating mothers and their breastfeeding infants (November 2005–August 2012). The women were in good general health, without any underlying chronic conditions. The study was conducted at the Medical University of South Carolina (MUSC) and the University of Rochester (U of R). Approval was granted by MUSC’s Institutional Review Board for Human Subjects HR 16536 and Clinical and Translational Research Center (CTRC; Protocol 752) and U of R’s Institutional Review Board (14460) and CTRC (Protocol 1129), and registered via ClinicalTrials.gov NCT00412074. Following written informed consent, mothers were randomized to one of three vitamin D treatment groups: 400 IU vitamin D3 per day, 2400 IU, or 6400 IU vitamin D3 per day for the duration of the study (1).

Figure 1.

Women included in post hoc analysis.

Of the 564 women who consented, 104 were excluded due to partial breastfeeding, health status, loss to follow-up, moves, loss of interest in study, and open label vitamin D supplementation. The remaining 460 were randomized, however, 174 women were excluded from this post hoc analysis due to either lack of 25(OH)D measurement, or lack of physical activity data, or both. 286 women were included in the final analysis (Figure 1). 121 women were in the control group, 38 women were in the 2400 IU group, and 127 women were in the 6400 IU group.

Measurements

The following measurements were recorded for all women included in the study.

Physical Activity

Women in the study were given the America on the Move self reported exercise survey at visits 1, 4, and 7 (America on the Move Foundation, Inc., Aurora, CO). This survey collected the type of activity performed, the number of times done per week and average duration of each session. Using the current Compendium of Physical Activities, the exercise activities were converted to metabolic equivalents (METs) as a measure of exercise intensity(11).

Vitamin D

Maternal blood samples were obtained at baseline (4–6 weeks postpartum; V1), month 4 (V4), and month 7 (V7). Circulating 25(OH)D was measured by radioimmunoassay (Diasorin, Stillwater, MN) as an indicator of vitamin D status (1).

Sun Exposure/Skin Pigmentation

The Smart Probe 400 (IMS, Inc., Milford, CT) was used at each visit to measure degree of skin pigmentation on the mother’s inner and outer lower arm as an index of sun exposure. Score had to decrease by a value of 5 to indicate the skin had darkened from sun exposure, from visit 1 to 4 and then 4 to 7.

BMI

Height and weight of each mother was recorded at the first outpatient visit to determine BMI (weight [kg]/height² [m²]). During subsequent outpatient visits, only the mother’s weight was recorded. BMI was calculated for each outpatient visit.

Other Variables

Latitude and season

The two study sites chosen for this clinical trial are at two extremes in terms of latitude and UV exposure throughout the year. Charleston is a southern U.S. city (Group A), latitude 32°78′ whereas Rochester (Group B) is one of the more northern U.S. cities, latitude 43°15′ with limited sunlight exposure during several months due to cloud cover. The season was recorded based on the date of birth of the infant.

Breastfeeding status

Each mother completed a log of breastfeeding sessions per day during the week prior to each monthly visit and reported when, if ever, during the study period the infant was weaned from the breast and/or fed formula. Mothers were categorized as exclusively breastfeeding, combination breast and formula feeding, or exclusively formula feeding at each visit.

Statistics

Chi-square analyses were used to test for differences in grouped data. Student’s t-test analyses were used to test for differences in continuous variables with a dichotomous grouping. Linear regression was used to examine associations between duration of physical activity and 25(OH)D concentration, controlling for treatment, BMI, sun exposure, METs, and feeding type. These variables were included in the model because they are historically associated with vitamin D status or were significant in univariate analyses. A repeated measures mixed linear model was used to observe different variable associations with serum 25(OH)D. All analyses were performed using SAS, Inc. v 9. (Cary, NC).

Results

Patient Characteristics

Using the National Institute of Health (NIH) recommendation of 150 minutes (2.5 hours) of moderate activity per week, we found that at visit 1 (baseline), no significant differences were seen in amount of physical activity performed by treatment group (Table 1). Significant differences were seen in physical activity by race (P=0.007). Caucasians were more likely to meet the recommendation than African Americans or Hispanics. 49% of Caucasian women met the recommendation for physical activity, versus 36% of Hispanics, and 26% of African American women at visit 1. Of those women who met the recommendation 59% were Caucasian, 25% were Hispanic, and 16% were African American. Significant differences were not seen in physical activity when looking at breastfeeding or formula feeding. Significant differences were seen in physical activity by education level (P=0.01). 23% of women who did not complete high school, 35% of those with a high school education, and 45% of those with a college education were meeting the standard recommendation for physical activity. Of those who met the national recommendation 76% had a college education, 13% completed high school, and 11% had an education level lower than a high school diploma. Insurance status and marital status did not have any significant association with physical activity level. Delivery type nor having preeclampsia were significantly associated with physical activity level. The average BMI of women who met the standard physical activity recommendation did not differ from women who did not. Exercise intensity level (METs) did not differ by amount of physical activity achieved per week. Maternal age did not significantly differ in those who met the national recommendation and those who did not.

Table 1.

Baseline characteristics by physical activity duration reported at Visit 1.

Baseline Characteristics	<2.5 hr/wk physical activity	≥2.5 hr/wk physical activity	P
	N % characteristic % activity	N % characteristic % activity
Treatment
400 IU	79 65.29 45.93	42 34.71 36.84	0.07
2400 IU	26 68.42 15.12	12 31.58 10.53
6400 IU	67 52.76 38.95	60 47.24 52.63
Race
African Americans	50 73.53 29.07	18 26.47 15.79	0.007
Hispanics	51 63.75 29.65	29 36.25 25.44
Caucasians	71 51.45 41.28	67 48.55 58.77
Feeding Type
Exclusive Breastfeeding	118 57.00 68.60	89 43.00 78.07	0.08
Formula or Combination	54 68.35 31.40	25 31.65 21.93
Maternal Education
<High School	39 76.47 22.67	12 23.53 10.53	0.01
High School	28 65.12 16.28	15 34.88 13.16
College	105 54.69 61.05	87 45.31 76.32
Insurance Status
Private	77 54.61 44.77	64 45.39 56.14	0.056
Medicaid or None	95 65.52 55.23	50 34.48 43.86
Marital Status
Married	89 55.28 51.74	72 44.72 63.16	0.057
Single	83 66.40 48.26	42 33.60 36.84
Preeclampsia
Preeclamptic	6 85.71 3.49	1 14.29 99.12	0.249
Not Preeclamptic	166 59.50 96.51	113 40.50 0.88
Delivery Type
Cesarean Section	51 68.92 29.65	23 31.08 20.18	0.073
Vaginal	121 57.08 70.35	91 42.92 79.82
BMI – M (SD)	28.81 (6.15)	27.88 (5.37)	0.2
METs – M (SD)	3.49 (0.27)	3.54 (1.14)	0.65
Age- M (SD)	29 (6.2)	29.3 (5.6)	0.69

As mentioned above, the NIH recommendation of physical activity is 150 minutes of moderate activity or 75 minutes of vigorous activity per week. Moderate activity requirements were used in this analysis because only 1 woman at visit 1 (0.35%), 6 women at visit 4 (2.1%), and 7 women at visit 7 (2.45%) exercised vigorously. Whereas, 39.9% of women at visit 1, 52.8% of women at visit 4, and 55.9% of women at visit 7 were meeting the NIH recommendation of 150 minutes of moderate intensity (3–6 METs) physical activity per week.

Physical Activity and Race

As seen in figure 2, Caucasians had a greater number of mean hours of physical activity per week at visit 1, 4, and 7 compared to Hispanics and African Americans with 3.2, 4.9, and 4.6 hours respectively. Hispanics reported on average 2.6, 3.5, and 3.7 hours of physical activity per week at visits 1, 4, and 7 respectively. African Americans reported on average 2.3, 3.5, and 4.1 hours of physical activity per week at visits 1, 4, and 7 respectively.

Figure 2.

Repeated measures mixed model analysis was used to show the differences in the mean number of hours per week of physical activity by race at visits 1, 4, and 7.

Physical Activity and Vitamin D

Using a repeated measures mixed linear model we found that race, treatment, sun exposure, BMI, and visit number were all associated with serum 25(OH)D, however exclusive breastfeeding, exercise intensity, and total minutes of physical activity were not associated.

Using a multiple regression model that predicts serum 25(OH)D concentration at each visit, controlling for race, BMI, exclusive breastfeeding, and METS, we found that at visit 1, total minutes of physical activity performed per week was positively associated with serum 25(OH)D. We observed an increase in 25(OH)D of 1.3 nmol/L (p=.03). Performing greater than 2.5 hours of physical activity per week had a trending association with an increase serum 25(OH)D of 7.23 nmol/L (p=.05). At visit 4, while also controlling for treatment group and sun exposure in the multiple regression model, performing greater than 1.5 hours of physical activity per week was associated with an increase in 25(OH)D of 11.73 nmol/L (p=.04). At visit 7 we did not find any association between physical activity and serum vitamin D.

When we examined exclusively breastfeeding women in a repeated measures mixed linear model we found that race, treatment, sun exposure, BMI, and visit number were all associated with serum 25(OH)D, however exercise intensity, and total minutes of physical activity were not associated.

When examining exclusively breastfeeding women using a multiple regression model by visit, we did not find any associations between serum 25(OH)D and total physical activity.

Discussion and Conclusion

Previous studies have found associations between physical activity levels and serum vitamin D in multiple populations, however the data on postpartum women is lacking. Because of the importance of maternal vitamin D status while breastfeeding, a clear understanding of the relationship between physical activity and vitamin D in women after childbirth is vital. We found that serum vitamin D levels were associated with levels of physical activity at visits 1 and 4, but not 7. The amount of weekly physical activity needed to warrant a significant relationship with serum 25(OH)D levels was not consistent, alternating from significance seen at at least 2.5 hours per week at visit 1 to at least 1.5 hours per week at visit 4. It is interesting to observe that at visit 7 the association was no longer present. At visits 1 and 4 our findings agreed with previously published research showing that physical activity is positively associated with vitamin D levels. We hypothesized that the inconsistency seen at visit 7 could be due to self reporting differences as a result of the lifestyle differences between a mother with a 7 month old and a newborn.

Based on our findings, early after delivery, more exercise was associated with higher vitamin D levels, implying that maintaining a certain level of exercise may help women proactively sustain a greater vitamin D status. The current guidelines for healthy Americans set by the NIH are 150 minutes of moderate or 75 minutes of vigorous physical activity each week. While these guidelines have been proven beneficial for prevention of obesity, cardiovascular disease, diabetes, some types of cancer, and more, they may not necessarily hold true for sustaining healthy vitamin D levels in postpartum women (U.S. Department of Health and Human Services).

There are a few proposed mechanisms as to why physical activity increases serum vitamin D levels. One study proposes that because physical activity recruits adipose tissue for energy, vitamin D is also mobilized during lipolysis, thus increasing serum 25(OH)D levels (14). It is known that physical activity increases bone mass in local areas of impact thus increasing calcium absorption efficiency, and decreasing calcium excretion (15). It is proposed that the increase in serum calcium could in turn cause sparing of serum vitamin D (14). In addition, the relationship between parathyroid hormone (PTH) and physical activity could play a role in the association seen between physical activity and vitamin D. Research has shown increased PTH secretion after prolonged moderate intensity physical activity. For example, one study reported that 2 hours of cycle ergometry at an intensity of 60–75% of ventilatory threshold caused a significant increase in PTH (16). Because PTH is normally secreted to defend against a low serum calcium concentration, it stimulates the production of 1α-hydroxylase, therefore increasing conversion of 25(OH)D to 1,25(OH)D(17).

A strength of this study was that it utilized subjects from a large randomized control trial of women with diverse ethnic and racial backgrounds, from two geographically distinct areas in the United States.

A limitation of the study is that it relied on self-reported exercise data. In the future, objective measurements of physical activity such as the use of a fitness band in addition to self-reported data should be considered. Further research on vitamin D levels in the breast milk of women achieving different amounts of physical activity may offer answers for the inconsistencies seen between our results from visit 7 and current published studies on other populations.

Additional research is warranted due to the lack of association of physical activity and serum 25(OH)D that was seen at visit 7. While no definitive conclusions can be drawn regarding precise levels of physical activity influencing 25(OH)D levels in postpartum women, the data suggest that increased activity during the first 4 months postpartum is associated with improved vitamin D status.

Highlights.

• More women were meeting the NIH recommendation of physical activity as time passed postpartum.

• Caucasians were more likely to meet physical activity recommendations than minority groups.

• Higher physical activity levels were associated with higher postpartum serum 25(OH)D levels.