Vitamin D is a steroid hormone which is naturally synthesized in the deep skin layers from 7-dehydrocholesterol when exposed to ultraviolet B wavelengths     . Ninety percent of the vitamin D in non-supplemented individuals’ bodies arises from natural synthesis in response to sunlight   . Vitamin D is found in a limited number of foods including fortified milk or juice, egg yolk, liver, and fatty fish       . Based upon results of the National Health and Nutrition Examination Survey (NHANES) 2005-2006, estimated total vitamin D availability in adult females ranges from 144 to 276 IU/day  resulting in low serum levels of 25-hydroxyvitamin D (25(OH)D)  While vitamin D has historically been thought to be primarily responsible for calcium homeostasis and bone health, new research is showing that vitamin D may also be important in immunomodulation, regulation of cell growth, and cardiovascular health  . Although D3 is the preferred form, as it is more effective in increasing serum 25(OH) D levels, vitamin D2 is often used in fortified foods.
There is a lack of consensus on vitamin D levels of sufficiency, insufficiency, and deficiency as a scientific committee process has not been established to define these levels   . This is exacerbated by studies finding that levels once considered “normal” may actually be insufficient based upon current evidence  . The Institute of Medicine (IOM), based on a review of data, set the following guidelines for these levels:  (Table 1).
Several studies reported that 25(OH)D levels must exceed 32 ng/mL to maximize skeletal integrity. The Endocrine Society issued clinical practice guidelines in 2011 stating that the desirable serum concentration of 25(OH)D is greater than 75 nmol/L (30 ng/mL) to maximize the effect on calcium, bone, and muscle metabolism  .
Using the IOM definitions, results from the National Health and Nutrition Examination Surveys (NHANES) 2001-2006 showed that two-thirds of the
Table 1. Institute of Medicine definition of vitamin D status.
population had sufficient vitamin D; one-quarter were insufficient; and eight percent were deficient  . A South Korean study assessed vitamin D deficiency among infants and noted that 48.7% had insufficient 25(OH)D levels  . Among breastfed infants, vitamin D deficiency prevalence was significantly higher at 90.4% than formula fed infants, who had a prevalence of 15.4%. Mean serum 25(OH)D of breastfed infants was significantly lower compared to formula fed infants, even when vitamin D was sufficient (p < 0.001).
Vitamin D deficiency rates have increased with the modern lifestyle characterized by decreased sun exposure related to urbanization, increased indoor activities, and use of sunscreen  . Caucasian individuals require 10 - 12 minutes of full body sun exposure to synthesize 10,000-20,000 IU of vitamin D; in comparison, an individual with dark skin would need 60 - 72 minutes of exposure to synthesize the same amount of vitamin D   . During the winter months, individuals living at a latitude above 40˚ do not synthesize an adequate amount of vitamin D, regardless of exposure  .
Only small amounts of 25(OH)D pass from the maternal circulation to human milk  . Instead, vitamin D is passed into human milk as cholecalciferol (vitamin D3, the parent form of 25(OH)D)  . It has been shown that 20% - 30% of maternal vitamin D is expressed in human milk compared to only 1% of maternal 25(OH)D    .
The Institute of Medicine (IOM) and the American College of Obstetrics and Gynecologists (ACOG) recommends 600 IU of vitamin D daily for all adults, including pregnant and lactating women (increased in 2008 from the Adequate Intake of 200 IU per day, which was previously arbitrarily set)    . The Endocrine Society shares this recommendation but suggest that 1500 to 2000 IU may be necessary to maintain adequate serum 25(OH)D  . The Recommended Dietary Allowance (RDA) for infants 0 - 12 months, as set by the Food and Nutrition Board (FNB) at the IOM of the National Academies, is 400 IU  .
The Tolerable Upper Intake Level for vitamin D has been set at 4000 IU in pregnancy and lactation, and at 1000 IU for infants  . However, most reports suggest a toxicity threshold for vitamin D between 10,000 and 40,000 IU per day   . Symptoms of toxicity are unlikely at daily intakes below 10,000 IU per day; however, the Food and Nutrition Board suggests that even lower vitamin D intakes may have adverse health effects over time, based on national survey data, observational studies, and clinical trials  . The No Observed Adverse Effect Level (NOAEL) has been set at 10,000 IU daily by the IOM   .
Setting adequacy levels is difficult since the relationship between serum 25(OH)D levels and vitamin D intake is non-linear, for reasons that are not entirely clear  . Doses greater than or equal to 1000 IU per day are associated with a 1 nmol/L rise in serum 25(OH)D per 40 IU of intake  . Doses up to 600 IU per day are associated with a 2.3 nmol/L increase in serum 25(OH)D per 40 IU consumed  .
Traditionally, it has been thought that the vitamin D content of human milk was inadequate to meet infant vitamin D needs    . Vitamin D activity in human milk has been found to range between five to 80 IU/L    . However, research completed in the 1980s showed that vitamin D content of human milk is directly related to maternal serum vitamin D levels and, therefore has the potential to be adequate when mothers’ vitamin D levels are sufficient      .
In infants and young children, long-term vitamin D deficiency leads to rickets  . While rickets were thought to be eradicated in the United States, recent reports show that prevalence is rising once again   . A review from the United States included 166 patients with nutritional rickets during the time period of 1986-2003; 96% of the children with rickets were breastfed. Therefore, the American Academy of Pediatrics (AAP) recommends that all breastfed infants receive 400 IU of oral vitamin D daily starting at birth        . This recommendation is a result of poor transfer of vitamin D from the maternal circulation into human milk in addition to the guidance to minimize sun exposure to infants up to six months of age       .
Many parents do not supplement their infants according to the AAP recommendation   . Of 1140 exclusively breastfed infants, only 15.9% received any vitamin D supplements  . Other reports show compliance rates between 2% - 36%    . Only 5% of those with rickets are reported to have received vitamin D supplementation   . This guideline may undermine breastfeeding success by implying that human milk is an inadequate source of complete nutrition for infants  . In fact, 36.4% of surveyed pediatricians in Seattle, WA chose not to recommend vitamin D supplements to breastfed infants due to concerns that parents would cease breastfeeding  . There is also some concern surrounding direct supplementation of the infant. Risks are currently unclear, but may include allergic reactions to the ingredients, aspiration pneumonia, accidental overdose, and changes in intestinal flora and pH may compromise. Given these concerns, the purpose of this review was to evaluate the impact of high-dose maternal vitamin D during lactation on human milk content and infant vitamin D status.
2.1. Data Source and Searches
Medline and Google scholar were searched from January 2003 through May 2018. Earlier studies were excluded due to the unacceptably large variance in vitamin D assays prior to this time  . The key words and MESH terms used to gather and classify the research both independently and in combination were maternal, breastfeed(ing) or lactation, Vitamin D, dietary supplements, milk/human, cholecalciferol, infants/physiologic phenomena. References found in articles were examined to identify any additional articles but no studies within the search delimitations were identified. The search was delimited to English language and humans. The search strategies for breastfeed* and vitamin D (or Vitamin D status or cholecalciferol) and milk/human and infant/physiologic phenomena yielded 173 articles. Substitution of lactation instead of breastfeed* yielded 184 articles; all but 11 were duplicates. Adding maternal to the search strategy did not alter the number of titles retrieved. Adding the term dietary supplements did not improve the final yield.
Abstracts were reviewed for eligibility. Complete articles of those accepted were independently reviewed by two investigators (KS and MAH) and accepted if they met the inclusion criteria of maternal supplementation during lactation, included data on vitamin D status of infants, evaluated the effect of maternal dose and exclusive breastfeeding or minimal, and documented, formula supplementation. Articles were excluded if they were not published in English, full text was unavailable (1 article), did not evaluate infant vitamin D status in relation to maternal supplementation with vitamin D, evaluated more than one micronutrient or were commentaries on previous studies. Articles were also excluded if they examined vitamin D supplementation during pregnancy since the association between vitamin D supplementation during pregnancy and offspring vitamin D status and outcomes has been the subject of a current systematic review and meta-analysis  .
2.2. Study Selection
Of the 184 articles which were identified, 86 articles titles were outside of the inclusion criteria, 98 were considered for review of abstracts. This review yielded 11 studies that met inclusion criteria with a variety of strengths, weaknesses, and overall contribution to the present research question. The PRISMA diagram is presented in Figure 1.
2.3. Assessment of Quality
Quality was evaluated using the Academy of Nutrition and Dietetics EAL Quality Assessment Checklists  .
An overall positive rating indicates that criteria were adequately addressed. A negative rating was assigned if the criteria were not adequately addressed and a neutral rating was assigned if indicators were mixed and the study design was neither exceptionally strong nor exceptionally weak. The checklists include 10 validity criteria: clearly stated research question (procedures, outcomes, inclusion/exclusion criteria) potential bias, comparability of groups, handling of withdrawn subjects, appropriate statistical analysis and conclusions supported by the results. The quality checklists are shown in Table 2 and Table 3.
A summary of study populations, interventions and outcomes and overall quality ratings are shown in Table 4.
3.1. Interventions with Vitamin D2
Vitamin D2 supplementation was chosen in several studies in an effort to better
Figure 1. PRISMA diagram.
Table 2. Quality criteria checklist intervention studies.
track the effects of supplementation, as contribution from other sources is unlikely   . Hollis and Wagner examined two groups of exclusively breastfeeding mother-infant pairs (n = 18)  . In the first intervention group, mothers
Table 3. Quality criteria checklist review studies.
Table 4. Summary of evidence.
Quality Rating Scale for Original Studies: Positive Quality―Most of the answers to the quality/validity questions were “yes” (including criteria 2, 3, 6, 7 and at least one other additional criterion); Neutral Quality―Answers on quality criteria 2, 3, 6, and 7 did not indicate that the study was exceptionally strong; Negative Quality―If six or more of answers on validity questions were no. Quality Rating Scale for Reviews: Positive Quality―if most of the answers on validity questions 1 - 4 were “yes”; Neutral Quality―if answers to any one of the first four validity questions was “no”; Negative Quality―if most (six or more) or the answers to validity questions are “no”.
were supplemented with 1600 IU vitamin D2 plus 400 IU vitamin D3 from a multivitamin. Mothers in this group exhibited decreased serum D3 (p < 0.02); a comparison group which received 3600 IU vitamin D2 plus 400 IU vitamin D3 demonstrated an increase in circulating total vitamin D species (p < 0.06). Both groups had increased vitamin D2 (p < 0.0001; p < 0.04, respectively). Total circulating 25(OH)D increased in group 1 mothers from 27.6 ± 3.3 to 36.1 ± 2.3 ng/mL (p < 0.05), compared to group 2 mothers whose circulating 25(OH)D increased from 32.9 ± 2.4 to 44.5 ± 3.9 ng/mL (p < 0.04). Between groups, maternal 25(OH)D3 did not differ; however, 25(OH)D2 was significantly higher in the group receiving the larger dose of D2 (p < 0.01). Milk ARA was doubled in the first group from 35.5 ± 3.5 to 69.7 ± 3.0 IU/L (p < 0.0001) compared with a much greater increase in those supplemented at the higher level in which the change was from 40.4 ± 3.7 to 134.6 ± 48.3 IU/L (p < 0.0001). In breastfed infants, vitamin D3 increased in both groups, from 7.9 ± 1.1 to 21.9 ± 4.7 ng/mL (p < 0.02) with the lower level supplement from 12.7 ± 3.4 to 18.8 ± 4.1 ng/mL (p < 0.2). Vitamin D2 increased in both groups, the increase being twice as great in the infants whose mothers were supplemented at the higher level (p < 0.0001). Total circulating 25(OH)D concentrations increased to the same extent in both groups. Although milk ARA never reached concentrations high enough to meet the DRI of 400 IU per day, infants in this study did not show signs of deficiency after 3 months of supplementation.
Basile et al. completed a similar randomized controlled trial using the same vitamin D2 and vitamin D3 intervention  . Researchers enrolled 25 fully breastfeeding mother-infant pairs. In the group receiving the lowest level of D2, maternal serum D2 increased from 22.4 ± 8.8 to 33.9 ± 6.5 (p = 0.002); and with the higher level of supplement, increased from 28.5 ± 8.6 to 43.0 ± 11.6 (p = 0.0008). Between groups, the highest level supplement resulted in significantly higher serum 25(OH)D levels (p = 0.03). Serum 25(OH)D levels increased in both groups of infants but infants breastfed by mothers given the highest level of D2 had significantly greater increases in serum 25(OH)D. Milk ARA was not measured in this study.
3.2. Daily Vitamin D Supplementation Interventions
In 2006, Wagner et al. studied 10 fully breastfeeding mother-infant pairs  . Mothers in Group 1 were supplemented with 400 IU vitamin D3 and infants were supplemented with 300 IU vitamin D3. In the second group, mothers were supplemented with 6400 IU vitamin D3 and infants received a placebo. Of note, only four mother-infant pairs completed the entire study, but using an intent-to-treat approach, all 10 pairs were included in analysis. Total circulating 25(OH)D levels decreased in mothers in group 1 through visit 5; there was a slight improvement at visits 6 and 7 that corresponded to increased outdoor activities and sun exposure. Mothers in group 2 had an immediate increase in 25(OH)D levels that was sustained throughout the study period (significance not provided). Compared to group 1, the mean milk ARA in group 2 significantly increased to 873 IU/L (p < 0.0003), which resulted in a dramatic rise in infant circulating 25(OH)D levels. This rise in infant 25(OH)D was almost identical to that in the infants receiving 300 IU per day vitamin D3 directly via oral supplementation.
Hollis et al. conducted a study of 95 fully breastfeeding mother-infant pairs in 2015  . The intervention began with three groups: 400 IU vitamin D3 for mothers plus 400 IU vitamin D3 for infants; 2400 IU vitamin D3 to mothers, placebo for infant; and 6400 IU vitamin D3 for mothers, placebo to infants. Partway through the study, the 2400 IU group was discontinued due to safety concerns regarding low infant 25(OH)D levels. Therefore, only the 400 IU group (group 1) and the 6400 IU group (group 2) were included in analysis. Maternal 25(OH)D decreased −10.5 nmol/L from visit 1 (V1) to visit 7 (V7) in group 1 (p = 0.02) and increased +51.3 nmol/L in group 2 between V1 and V7 (p < 0.0001). At V1, maternal 25(OH)D was similar between groups (p = 0.17); at V4, group 2 had significantly higher 25(OH)D (150.5 ± 47.1 nmol/L compared to 83.0 ± 29.1 nmol/L, p < 0.0001). At V7, group 2 continued to have significantly higher maternal 25(OH)D (151.2 ± 51.3 nmol/L compared to 79.0 ± 31.3 nmol/L, (p < 0.0001). Infant 25(OH)D increased in both groups from V1 to V7 (Group 1: 36.0 ± 26.1 nmol/L to 109.1 ± 31.8 nmol/L, significance not provided; Group 2: 41.0 ± 25.6 nmol/L to 108.5 ± 38.0 nmol/L, significance not provided). Between groups, there was no statistical significance at any time point (V1 p = 0.35; V4 p = 0.10; V7 p = 0.94). This suggests that high-dose maternal supplementation of 6400 IU vitamin D3 daily is as effective as direct infant supplementation of 400 IU daily.
3.3. Monthly Vitamin D Supplementation Interventions
Wheeler, et al. completed a randomized controlled trial in 2016 with 87 exclusively breastfeeding mother-infant pairs in New Zealand  . Group 1 mothers received a placebo; Group 2 mothers received 50,000 IU vitamin D3 per month; and Group 3 mothers received 100,000 IU vitamin D3 per month. Vitamin D supplemented mothers had a significantly higher serum 25(OH)D concentrations at study end compared to placebo (p = 0.043 in group 2; p = 0.001 in group 3). Infant change in serum 25(OH)D was not significantly different between groups (p = 0.67 in group 2; p = 0.13 in group 3). In group 1, 26% of mothers exhibited vitamin D deficiency, compared to 4% of women in group 2, and 0% in group 3 (p = 0.002). Deficiency rates between infants in all groups were similar (27%, 29%, and 19%, respectively; p = 0.65).
3.4. Daily versus Monthly Vitamin D Supplementation
Oberhelman, et al. conducted a randomized controlled trial of 40 exclusively breastfed mother-infant pairs to evaluate the difference between monthly supplementation with 150,000 IU vitamin D and daily supplementation with 5000 IU vitamin D per day for 1 month  . The monthly supplement was provided to mothers at study visits. Compliance with daily supplementation was monitored and found to be excellent (98%). Those receiving the once monthly dose had significantly higher 25(OH)D concentrations than on days 1, 3, and 7, but not on days 14 and 28 (significance not provided). With monthly supplementation, maternal 25(OH)D levels peaked on day 3, and the maximum value observed in any mother was 72 ng/mL. By day 28, the increase in 25(OH)D between baseline and day 28 was 11.9 ± 4.2 ng/mL in those receiving the once monthly dose and 15.0 ± 5.7 ng/mL with daily dosing. (p = 0.06). None of the mothers’ serum 25(OH)D concentrations remained < 20 ng/mL on day 28 (group 1 values: 43.9 ± 11.8 ng/mL, range 22 - 71; group 2 values: 41.2 ± 8.9 ng/mL, range 26 - 60 ng/mL). Human milk cholecalciferol concentrations mirrored serum 25(OH)D concentrations, with peak values approximately 25% of serum values. By day 28, serum 25(OH)D levels increased in the infants of both groups and all infants achieved a serum 25(OH)D concentration >20 ng/mL. However, increases in infant’s 25(OH)D levels did not directly correlate with maternal increases in maternal 25(OH)D (r = 0.07, p = 0.64). A recent study (Nalk, 2017) evaluated effectiveness of maternal supplementation with 600,000 IU vitamin D in divided doses over 10 days in the early postpartum period on maternal and infant serum 25(OH) D levels and measures of rickets in exclusively breastfed infants. Maternal 25(OH)D levels after 6 months of study were greater 40.3 ± 21.6 ng/mL in supplemented vs 22.9 ± 20.18 ng/mL in controls. Infant levels were also higher (29.29 ± 14.67 ng/mL) in those fed by supplemented mothers and 15.73 ± 17.73 ng/mL in controls. After 6 months of exclusive breastfeeding, only one mother and four infants had serum 25 (OH) D levels < 11 ng/ml in the supplemented group (n = 53) vs 9 mothers and 25 infants (n = 57) in the control. Although 2 infants in each group developed clinical rickets. This study suggests that large divided doses may be effective in preventing vitamin D deficiency in mothers and infants during the first six months of breastfeeding  .
Healthy people 2020 set a breastfeeding target of 81.9% of infants “ever being breastfed”  . Breast milk is considered to be the gold-standard for infant nutrition, providing all the nutrients the young infant requires. It is therefore important to evaluate whether the adequacy of Vitamin D in breast milk can be insured through maternal intake.
The results of the available studies demonstrate that high-dose daily maternal vitamin D supplementation (ranging from 4000 IU to 6000 IU/day) during lactation is effective in improving both maternal and infant vitamin D status. Monthly supplementation at levels of 150,000 IU corrected both maternal and infant deficiency but not at lower levels (50,000 IU and 100,000 IU). This is likely because cholecalciferol is quickly converted to 25(OH)D by the mother; cholecalciferol is the main form of vitamin D transferred into human milk   . Therefore, daily supplementation may be preferred to optimize infant vitamin D status. There is some concern surrounding compliance. Mothers who received monthly vitamin D supplementation in the reviewed studies were administered their dose on-site; mothers who received daily supplementation were largely responsible for their own administration. An unrelated study in Japan found that monthly dosing of bisphosphonates for osteoporosis had higher compliance compared to daily and weekly dosing  . Additional studies assessing compliance in relation to maternal self-administered vitamin D supplementation need to be conducted to confirm these results. Previous studies demonstrate that vitamin D3 is more bioavailable than vitamin D2;    this needs to be confirmed in the current context as studies with varying types of vitamin D did not assess similar doses.
Limitations of the research to support the question of vitamin D supplementation is a general lack of standards for acceptable levels of serum 25(OH)D, the diagnosis of vitamin D insufficiency and vitamin D deficiency. The studies were heterogeneous with respect to intervention protocol, timing and levels of supplementation. Additional studies assessing safety of long-term, high-dose vitamin D supplementation are warranted. The longest study follow-up was 7 months, the effects of longer term supplementation on infant outcomes and maternal and infant toxicity are lacking.
Seven randomized controlled trials and four reviews demonstrate that maternal supplementation of vitamin D may be an effective method of improving breastfed infants’ vitamin D status. However, there is no consensus as to which dose is ideal, both in timing (monthly versus daily) and quantity. This is further compounded by the fact that there is no agreement on the classification of vitamin D deficiency on the basis of serum 25(OH) D levels. Furthermore, the dose apparently necessary to correct both the mother’s and infant’s vitamin D status is well above the Adequate Intake of 600 IU per day. In fact, the Tolerable Upper Limit is set at 4000 IU during lactation, which appears to be the minimum dose necessary to meet the nursing infant’s vitamin D needs. The Food and Nutrition Board of the National Academies of Science, Engineering, and Medicine should consider revising the Dietary Reference Intakes for vitamin D, given the evidence that higher doses show no signs of adverse events and are necessary to meet the vitamin D needs of both the mother and infant. Additionally, the AAP should likewise consider revising their statement on infant supplementation of vitamin D to include high-dose maternal supplementation of approximately 4000 IU per day as adequate in lieu of direct infant supplementation. Ideally, further studies evaluating the ideal dose and confirming other studies’ results should be conducted. It appears that daily dosing is the most effective method for improving both maternal and infant vitamin D status. Based on this evidence, practitioners may confidently recommend maternal-only supplementation of vitamin D at a minimum dose of 4000 IU per day.