Clinical and Laboratory Pearl
Calcium and Phosphorus in Milk of Brazilian Mothers
Oral Using Contraceptives
Jose G. Dorea, PhD, FACN, and Edina Myazaki, PhD Departamento de Nutric?a? o e Departamento de Estatistica, Universidade de Brasilia, Brasilia DF, BRAZIL Key words: human milk, lactation, calcium, phosphorus, hormonal contraceptives Objective: Oral contraceptives (OC) are the most efficient method of contraception and it is the
most prescribed by doctors in developing countries. Therefore we studied the effects of combination pill and
mini-pill on calcium and phosphorus in milk of breast-feeding mothers at different stages of lactation.
Methods: Fifty-four breast-feeding mothers made up three study groups: 33 mothers who had been advised
by their doctors to use either combination pill (12), or mini-pill (21), as well as a control group of 21 mothers
that used no hormonal contraceptives. All mothers completed a questionnaire and provided samples of
milk before and after a measured period of observation. Mean duration of study was 76, 120, and
101 days, respectively for users of mini-pill, combination pill, and controls. Determination of calcium and
phosphorus was done by inductively coupled plasma—atomic absorption spectrometry.
Results: Overall the decrease in milk concentrations of phosphorus (6%) and calcium (26.3%) during the
study period was not influenced by OC treatment. Regression analyses which took into consideration length of
treatment, socioeconomic status, number of children, duration of previous lactation, type of contraceptive, and
age of mothers and repeated measurements (before and after OC) showed that milk calcium was significantly
affected by stage of lactation (p50.0013).
Conclusion: The use of hormonal contraceptive such as the combination pill (levonorgestrel 0.15
mg+ethynilestradiol 0.03 mg) and mini-pill (norethindone 0.35 mg) does not seem to affect the secretion of
calcium and phosphorus in milk of mothers.
During human lactation, a higher demand for Ca and INTRODUCTION
P secretion in milk alters the metabolism of these elements.
Changes in the quantity and composition of mother’s milk Due to an increase in bone turnover  serum Ca and P are during the period of fast growth for exclusively breast-fed elevated [7,8] and less urinary Ca is excreted . Besides infants is of concern to health professionals. The nutritional being the more efficient, oral contraceptives (OC) are the benefits of exclusive breast feeding for growth, protection and most widely used method of contraception. Even during breast well being of infants are well recognized, especially for feeding, the estrogen-progestin combination pill is widely low income classes in less developed societies. In these prescribed (63%) by doctors in developing countries and it is societies high birth rate is both a threat to the well being of estimated that 10% of breast feeding mothers around the mothers and an economic burden for the social development. world use OC .
Birth control and exclusive breast feeding must be in balance The inhibitory effect of estrogen on bone resorption to guarantee the survival and social improvement of low has been known and used for treatment of osteoporosis income groups. . Women taking OC had higher mean bone density  The concentrations of calcium (Ca) and phosphorus (P) in and less urinary calcium  than nonusers.
human milk varies between communities and between individ- The nutritional consequences of contraceptive practices on uals . During extended lactation a decrease in milk calcium the nursing human are still unsettled. The longterm use of OC concentration is observed as lactation progresses [2– 4] al- before pregnancy does not affect calcium and phosphorus con- though phosphorus concentrations may vary in some mothers, centrations in human milk . A review of early publications
while maintaining stable in others .of OC use during lactation , showed a decrease in milk
Address reprint requests to: Jose Dorea, PhD, FACN, C. P. 04322, Universidade de Brasilia, 70919-970 Brasilia DF, BRAZIL.
Journal of the American College of Nutrition, Vol. 17, No. 6, 642–646 (1998)
Published by the American College of Nutrition
OC and Minerals in Human Milk
calcium concentrations associated with combination pills. by cleaning the nipples with distilled and deionized water, and However, hormonal treatment based on progestins, such as by using properly cleaned glassware (washed in acid and rins- mini-pills or implants do not show consistent results. Some ing with EDTA solution followed by distilled and studies [15,16] showed a decrease on calcium concentration deionized water). After collection, samples were divided in while other  showed no significant effects. Only Toddy- aliquots and stored at —20?C until analysis.
walla et al  showed an observable effect of hormonal Before analysis samples were thawed and thoroughly ho-
mogenized in a vortex. An aliquot of 2 mL of milk was treatment on milk phosphorus concentration. This paper aims to
accurately weighed to the nearest decimal point with an ana- study the effects of OC use during lactation on the concentra-
lytical scale, and mixed with 7 mL of HNO(Suprapur, Merck, tion of calcium and phosphorus in milk, taking into consider- 3
ation confounding constitutional variables such as maternal Darmstadt, Germany). The mixture was heated at 110?C until it age, previous lactation, length of breast feeding, and variables reached a volume of approximately 1 mL. After cooling, 2 mL
associated with contraception, such as type and length of use of concentrated HClO(Suprapur, Merck, Darmstadt, 4 of OC. Ger- many) were added to the digests and again heated until the liquid turned a dark brown color. The digests were cooled and 2 mL of HO(30%) (Suprapur, Merck, Darmstadt, Germany) 22 MATERIALS AND METHODS were added. The mixture was heated to 70?C until the digests turned clear and had a volume of approximately 1 mL. After Breast feeding mothers were recruited from private and cooling, the digests were quantitatively transferred to a public clinics after explaining the objectives and obtaining their cali- brated test tube and brought to a volume of 10 mL with consent. The mothers from private clinics were from middle double distilled and deionized water. class whereas the ones from public hospitals were usually slum The determination of calcium and phosphorous was done in dwellers mainly from the squatter village of Paranoa, city a Inductively Coupled Plasma (ICP) Spectroflame model of Brasilia (Table 1). FVM03 with a focal distance of 75 cm (Kleve, Germany) in the Records of variables such as stage of lactation (age of emission line of 317.93 and 214.91 nm, respectively. A child), initiation and length of study were registered for adjust- con- centric Meihard (Kleve, Germany) nebulizer was ing the statistical model of analysis. The choice of contracep- operating under the following conditions: pressure, 38 psi; flux tion was decided between the mothers and their physician. of cooling gas, 13 L/minute; flux of auxiliary gas, 0.5 Because we had no interference on choice of contraceptive or L/minute; sample injector flow, 1 mL/minute, and forward time of its initiation, we kept records of these variables power of 1.1 kW. for statistical adjustments and corrections. Standard Reference Material 1577 Bovine Liver (National Milk was sampled before and during treatment, at a planned Institute of Standards and Technology, Washington, DC) was interval of at least 2 weeks, between 7:00 and 11:30 a.m. The used for quality assurance using an identical procedure on milk samples (2 to 10 mL) were collected by manual expres- a sion. Preventive measures to avoid contamination were ensured 0.25 g of sample. All samples were processed in duplicate and the final value was taken as the mean of running each sample Table 1. Background Information on Characteristics of three times against standards prepared from solutions for Mothers and Contraceptives atomic absorption spectroscopy (Merck, Darmstadt, Germany). The effects of contraceptives on changes in concentrations Control Combination Mini-pill of calcium and phosphorus in milk were studied by a multiple Mothers, mean?SD regression analysis taking into account, age of mother, number (range) of children, socioeconomic status, day of lactation on first day N 21 12 21
of treatment, length of treatment, and type of contraceptive .
SD and ranges. Sta- The data were summarized as means?
tistical computations were done with a SAS computer program
Age (years) 29.4 ? 5.4 ? 3.5 ? 5.6 20.5 24.6 for PC (SAS Institute, Cary, NC). A p value less than 0.05 was (22–41) (16–29) (16–37) considered statistically significant. Number of children 2.2 ? 1.2 1.8 ? 1.0 1.8 ? 0.9
(1–5) (1–4) (1–4) Socioeconomic status* 8/12 12/1 19/2 Treatment, median RESULTS (range) Stage of lactation, 102 180 74 The characteristics of mothers and lactation information as days (28–254) (34–380) (28–190) well as type of contraceptives are shown in Table 1. Although Duration of study,21 (14–35) 19 (14–46) 19 (14–43) some mothers had started taking oral contraceptives as early as days 2 weeks post-partum others started treatment after the third Combination pilllevonorgestrel 0.15 mgethylnilestradiol 0.03 mg. 5+month. Mean stage of lactation at start of treatment for the three Minipill5norethindone 0.35 mg.
* Socioeconomic status: low/high. groups ranged from 2.5 to 4 months. We had scheduled
JOURNAL OF THE AMERICAN COLLEGE OF NUTRITION 643
OC and Minerals in Human Milk
second sample (post-treatment) to take place at least 2 weeks been shown that milk production is diminished by combined after initiation of treatment. Certain difficulties, in oral contraceptives  but it is not affected by progestins .
particular with the mothers in keeping appointments, caused In spite of reports that the hormonal contraception can affect the variation in length of treatment between mothers. lactation , both quantitatively (milk yield), as well as qual- These sources of variability were accounted for in the itatively (specific nutrients such as calcium and phosphorus),
regression analysis. we could not confirm the decrease in calcium seen for
long- term use of high estrogen dose of combination pills Table 2 shows the summary of concentration of milk cal-
cium and phosphorus as a function of contraceptive type. For [15–17] and progestins [15–16]. Our results however, agreed
with the study of Toddywalla et al  with regards to the both minerals there was a decline in concentration with
effect of low dose estrogen-pill on milk calcium concentration. time that is confounded with OC treatment. The statistical
analysis is shown in Table 3. Overall no significant The consis- tent decrease in milk calcium observed by others difference was seen for calcium concentration due to OC (Table 4) in association with combination pill could be treatment, only stage of lactation per se was a significant explained by the elevated doses of estrogen used in those source of variation for cal- cium concentration. studies. Compara- tively, the dose of estrogen (0.01 mg of
ethinylestradiol) stud- ied by Toddywalla et al  was The distribution of calcium and phosphorus concentrations
in milk as a function of length of lactation for all mothers in smaller than others (0.05 to
Fig. 1, shows a decline in calcium concentration as 0.1 mg of estrogen) and similar to our study. Other differences lactation progresses. For comparative purposes Table 4 shows in studies were the length of hormonal treatment, and statistical summary of studies dealing with hormonal contraceptive analysis employed. The shorter hormonal treatment and the effects on milk concentrations of calcium and phosphorus. smaller dose in our study could explain in part why we found
no significant effect on calcium due to estrogen treatment. It is noteworthy that statistical analysis used by early inves- tigators did not fully account for important intervening DISCUSSION vari- ables. A summary of statistical shortcomings is shown in
Table Results of our study are in good agreement with world wide 4. In most studies where statistical methods were described the reports on calcium and phosphorus concentrations . A effect of contraceptives was measured within each group review of the calcium concentration in milk before 1978 was as differences in absolute or relative changes in mineral made by Jelliffe and Jelliffe  which shows results concen- trations over 1 to 10 month periods. Indeed, from Brazil (204 gIg), Bantu, Africa (208 to 287 gIg), UK iicontrol groups used and analyzed independently, also showed (299 gIg), New Hibrides (258 gIg), US (340 gIg), India iiisignificant de- creases in milk components associated with (342 gIg), and Australia (286 to 307 gIg). In all these iistage of lactation [15,24]. studies, range of variations are comparable between affluent The statistics of simple mean comparison within each group and devel- oping countries. As observed previously [2– 4], we may be adequate when there are no other intervening variables. confirm the decline in calcium due to progress of lactation. In the case of nutrient composition of human milk, The calcium demand of lactation is met by the favorable specially calcium and phosphorus it has been suggested that hormonal balance in postpartum, where women are hyperpro- diet, eth- nicity and behavior [25,26] may affect the lactemic and hypoestrogenic. Under such circumstances cal- incorporation of these elements into breast milk. Besides, the cium and phosphorus depots are easily mobilized by PTH. type of contracep- tive chosen by the women was of a Later in lactation with resumption of menses and elevation of different composition with different concentrations of estrogen, markers of bone mobilization are lowered . It has estrogens and progestins. These hormones known to either affect lactation  or calcium metabolism (estrogens) in particular, could have a cumulative effect that needed Table 2. Summary of Results as Mean and (SD) of evaluation in combination with other vari- ables. Since none of Concentrations (gIg) of Calcium and Phosphorus in Milk of ithe studies employed a statistical model which could handle Mothers Using Oral Contraceptives analysis that included intervening vari- ables, mainly time dependent decrease in milk calcium or rates of changes in Calcium Phosphorus Groups concentrations of the minerals during the OC treatment, Before OC After OC Before OC After OC their findings could be questioned. Control 249.9 (53.5) 236.2 (44.6) 121.7 (24.6) 121.9 (20.1) Statistical selection of limited capability or lack of proper Combination description has reduced the general applicability of findings in pill 224.2 (61.2) 212.9 (35.4) 119.3 (28.3) 134.8 (31.5) some of the papers summarized in Table 4. Worst of all, some Mini-pill 227.0 (48.9) 242.8 (54.7) 121.5 (31.9) 117.8 (19.0) of the papers that presented results claiming an effect of OC on OC5Oral contraceptives. calcium and phosphorus lacked valid statistical conclusions. If Combination pilllevonorgestrel 0.15 mgethynilestradiol 0.03 mg. 5+definite conclusions are to be drawn regarding OC effects on Minipill5norethindone 0.35 mg.
milk concentrations of nutrients, future research must avoid the
errors of the past. In the present study, a multivariate statistical
644 VOL. 17, NO. 6
OC and Minerals in Human Milk
Table 3. Summary of results (p) of statistics of multiple regression analysis Source of Age of Stage of Socioeconomic Number of (TC) Type of (PT) Period ofTC3PT variation mother lactation status children contraceptive treatment
Ca 0.8068 0.0013 0.0789 0.8853 0.8958 0.6986 0.1608
P 0.4710 0.7920 0.4302 0.3992 0.5022 0.4684 0.2262 2The regression coefficients (R) for calcium and phosphorus were 0.08 and 0.27, respectively. method of analysis, did not show that short term use of OC- containing estrogen affected calcium and phosphorus concen- trations in milk.
This work was supported in part by a grant from the Bra-
zilian National Research Council (CNPq). We thank the moth-
ers that contributed to the study and the interest and coopera-
tion of Drs. Lucila Mota, Iraci Oliveira, and H. Borato,
and nurses Sonia Silveira and Celina Santana from
University of Brasilia Hospital, and Posto de Saude 15-FHDF.
We also thank Dr. Teresa Costa for collection of samples
and Dr. Geraldo Boaventura and Paulo Roque Silva for Concentrations of calcium and phosphorus in milk according laboratory work, as well as Dr. Connie McManus for to length of lactation. redactorial suggestions.
1. Prentice A, Dibba B, Jarjur LMA, Larskey MA, Paul AA: Is breast
milk calcium concentration influenced by calcium intake
during pregnancy? Lancet 344:422–412, 1994.
Table 4. Summary of Results of Studies Relating Hormonal Contraceptive Effects During Lactation on Milk Concentration of
Calcium and Phosphorus and Respective Statistical Analysis of Data Length of Statistical analysis Reference Contraceptive, mg [Ca] [P] study 1 Abdel Kader et al, 1969  Lynestrenol 2.5+mestrenol 0.075 0 4 mComparison of percent average decrease. 2  Lynesterol 1.0mestrenol 0.1 0 Stage of lactation accounted in the +2  Dihydroxyprogesterone (150 monthly) 22design.  Lynestrenol 0.5Iday 22Barsivala & Virkar, 1973  Norgestrel 0.5+ethynil estradiol 0.05 0 3 m No indication of statistical test used. 2
0 0 1 m  Megestrol acetate 0.5 0 0 3 m
0 0 1 m Toddywalla et al, 1977  Norethisterone 0.35+ethynil estradiol 0.01 0 0 10 m t-test on differences of means within  Norethisterone 0.35 0 2each group.
 Medroxyprogesterone 150mgI3 months 0 0
2 0  300mgI6 months
Seth et al, 1977 Norethindone acetate 40:  implant at 6 w 0 0 8 m No indication of statistical test used.
 implant at 6 w 0 2
This study  Levongestrel 0.15+ethynil estradiol 0.01 0 0 3.5 w* Multiple regression with covariate Norethindone including stage of lactation in all
5number of mothers; 25decrease in concentration; m5increase in concentration; 05no change in concentration. 1 Month.
* Mean, weeks (range50.5–3.0 m).
JOURNAL OF THE AMERICAN COLLEGE OF NUTRITION 645
OC and Minerals in Human Milk
2. Finnley DA, Lonnerdal B, Dewey KG, Grivetti LE: Inorganic lactation: Current findings, methodological considerations and fu- constituents of breast milk from vegetarian and non ture priorities. Wld Hlth Quart 36:168–200, 1983.
vegetarian women: Relationships with each other and with organic 15. Abdel-Kader MMA, Hay AA, El-Safouri S, Aziz MTA, el-Safouri constitu- ents. J Nutr 115:772–781, 1985. S, Saad el-Din J, Kamal I, Hefnawi F, Abdalla M: Clinical,
biochemical and experimental studies on lactation. IV. 3. Allen JC, Keller RP, Archer P, Neville MC: Studies in
human lactation: milk composition and daily secretion rates of Quantita- tive and qualitative changes induced in goat’s milk by
macronu- trients in the first year of lactation. Am J Clin Nutr gestagens. Am J Obstet & Gynecol 105:978–985, 1969.
54:69–80, 1991. 16. Toddywalla VS, Joshi L, Virkar K: Effect of contraceptive steroid
on human lactation. Am J Obstet Gynecol 127:245–249, 1977. 4. Neville MC, Allen JC, Archer PC, Casey CE, Seacat J, Keller RP,
Lutes V, Rasbach J, Neifert M: Studies in human lactation: milk 17. Barsivala VM, Virkar KD: The effect of oral contraceptives on volume and nutrient composition during weaning and lactogenesis. concentrations of various components of human milk. Contracep- Am J Clin Nutr 54:81–92, 1991. tion 7:307–312, 1973.
5. Laskey MA, Dibba B, Prentice A: Low ratios of calcium to 18. Diggle PJ, Liang K-Y, Zeger SL: ―Analysis of Longitudinal Data.‖
phosphorus in breast milk of rural Gambian mothers. Acta Paediatr Oxford, UK: Oxford Science Publications, pp 55–77, 1996.
Scand 80:250–251, 1991. 19. Parr RM, Demayer EM, Iyengar VG, Byrne AR, Kirkbright GF,
Schoch G, Niinisto L, Pineda O, Vis HL, Hofvander Y: Minor and 6. Sowers M, Randolph J, Shapiro B, Jannausch M: A
prospective study of bone density after pregnancy. Obstet trace elements in human milk from Guatemala, Hungary, Nigeria, Gynecol 85:285– Philippines, Sweden and Zaire. Results from a WHOIIAEA Joint
Project. Biol Trace Elem Res 29:51–75, 1991. 289, 1995.
7. Specker BL, Tsang RC, Ho ML: Changes in calcium homeostasis 20. Jellife D, Jelliffe EFP: The volume and composition of human milk over the first year postpartum: effect of lactation and in a poorly nourished communities. A review. Am J Clin weaning. Obstet Gynecol. 78:56–62, 1991. Nutr
8. Kalkwarf HJ, Specker BL, Heubi JE, Vieira NE, Yergey AL: 31:492–515, 1978.
Intestinal calcium absorption of women during lactation and after 21. Zinaman MJ, Hickey M, Tomai TP, Albertson BD, Simon weaning. Am J Clin Nutr 63:526–531, 1996. JA: Calcium metabolism in postpartum lactation: the effect of
estrogen status. Fertil Steril 54(3):465–469, 1990. 9. Klein CJ, Moser-Villon PB, Dopuglass LW, Ruben KA, Troeki O:
A longitudinal study of urinary calcium, magnesium and zinc 22. Saarikoski M: Contraception during lactation. Ann Med 25:181–
excretion in lactating and non-lactating postpartum women. Am J 184, 1993.
Clin Nutr 61:779–786, 1995. 23. McCann MF, Moggia AV, Higgins JE, Potts M, Becker C: The
effects of a progestin-only contraceptive (levonorgestrel 0.03 mg) 10. Strauss LT, Speckhard M, Rochat RW, Senanayake P: Oral con-
traceptive during lactation: A global survey of physician practice. on breast-feeding. Contraception 40:635–648, 1989.
Int J Gynaecol Obstet 19:169–175, 1981. 24. Seth U, Yadava HS, Agarwal N, Laumas KR, Hingorani V: Effect
of sub dermal silastic implant containing norethindrone acetate on 11. Kitai E, Blum M, Kaplan B: The bone sparing effect of oral
contraceptive use in non-smoking women. Clin Exp Obstet human lactation. Contraception 16:383–398, 1977.
Gy- necol. 19:30–33, 1992. 25. Hillman LS: Mineral and vitamin D adequacy in infants fed human
milk or formula between 6 and 12 months of age. J Pediatr 12. Wolman RL, Clark P, McNally E, Harries MG, Reeve J: Dietary
calcium as a statistical determinant of spinal trabecular bone den- 117:S134–S142, 1990.
sity in amenorrhoeic and oestrogen-replete athletes. Bone 26. Prentice A, Barclay DV: Breast-milk calcium and phosphorus Miner Res 17:415–423, 1992. concentrations of mothers in rural Zaire. Eur J Clin Nutr 45:611–
13. Kirksey A, Ernst JA, Roepke JL, Tsai TL: Influence of 617, 1991.
mineral intake and use of oral contraceoptive on the mineral
content of human colostrum and more mature human milk. Am J
Clin Nutr Received September 1997; revision accepted May 1998.32:30–39, 1979.
14. Hull VJ: Research on the effects of hormonal contraceptives on
646 VOL. 17, NO. 6