Sex steroid hormones modulate serum ionized magnesium and calcium levels throughout the menstrual cycle in women

FERTILITY AND STERILITY VOL. 69, NO. 5, MAY 1998 Copyright 1998 American Society for Reproductive Medicine Published by Elsevier Science Inc. Printed on acid-free paper in U.S.A. Sex steroid hormones modulate serum ionized magnesium and calcium levels throughout the menstrual cycle in women Ozgul Muneyyirci-Delale, M.D.,* Vijaya L. Nacharaju, Ph.D.,* Burton M. Altura, Ph.D. and Bella T. Altura, Ph.D. State University of New York, Health Science Center at Brooklyn, New York, New York Received August 21, 1997; revised and accepted December 22, 1997. Financial support: B.M.A. was supported in part by National Institutes of Health grant AA08674. Reprint requests: Bella T. Altura, Ph.D., Box 31, SUNY Health Science Center, 450 Clarkson Avenue, Brooklyn, NY 11203 (FAX: 718-270- 3103). * Department of Obstetrics and Gynecology. Department of Physiology. Department of Medicine. The Center for Cardiovascular and Muscle Research. 0015-0282/98/$19.00 PII S0015-0282(98)00053-3 Objective: To determine the serum concentrations of the sex steroid hormones with respect to the concentrations of the biologically active fractions of magnesium and calcium during the different phases of the menstrual cycle. Design: Controlled clinical study. Setting: An academic research environment. Patient(s): Six parous and four nulliparous healthy cycling female volunteers. Main Outcome Measure(s): Concentrations of the sex steroid hormones estrogen, progesterone, and testosterone as well as the ionized Ca and Mg levels were measured in the serum of normal cycling women during five different stages: the menstrual, early follicular, late follicular, ovulatory (ovulatory/early luteal), and luteal phases. Result(s): In each woman, there was a comparatively high ionized Mg level coincident with the early follicular phase, a statistically significant decrease in ionized Mg around the time of ovulation, a significant decrease in ionized and total Mg when the serum progesterone concentration peaked, and a significant increase in the serum Ca 2 /Mg 2 ratio at both the ovulatory and luteal phases. In addition, a decrease in ionized Mg was found with increased testosterone levels. Conclusion(s): Healthy women of reproductive age demonstrate recurring cycling of ionized Mg and cyclic alterations in the ionized Ca to Mg ratio in their serum. The changes in serum concentrations of these important physiologically active cations, in the range at which they occur, can affect such entities as the vasculature, synaptic transmission, and excitation-secretion coupling and thus can produce the well-known premenstrual syndromes during the luteal phase in women who are somewhat deficient in Mg or in those who have an unusually increased Ca 2 /Mg 2 ratio. (Fertil Steril 1998;69:958 62. 1998 by American Society for Reproductive Medicine.) Key Words: Estrogen, progesterone, testosterone, ionized magnesium, ionized calcium, PMS, menstrual cycle, ionized Ca/ionized Mg ratio The divalent cations magnesium and calcium have been implicated in premenstrual problems such as premenstrual migraines, premenstrual syndrome (PMS) in general, disorders during pregnancy (e.g., preeclampsia or eclampsia), menopausal cardiovascular and bone problems (loss and susceptibility to fracture), and pathophysiology after contraceptive use (1). Both ions are also involved in intermediary metabolism, neuromuscular activity, secretion, excitation-secretion coupling, cardiovascular health, and bone metabolism, all entities and activities that are regulated by various hormones, including the sex steroids. Several previous studies have attempted to follow changes in the concentrations of these cations and their relation to the menstrual cycle. As early as the 1970s, it was reported that serum total Mg levels are decreased at ovulation in normal women and shortly after taking contraceptive estrogen (2). In 1972, a midcycle drop in total serum Mg was reported in normal women, but a slight increase was observed when women took progesterone as a contraceptive (3). In a study of healthy women by Pitkin et al. (4), an increase in serum ionized Ca was seen before ovulation, lasting until the end of the luteal phase and then falling, whereas total Ca, Mg, and phosphorus exhibited no such pattern. 958

Deuster et al. (5) divided the menstrual cycle into four phases and reported that plasma total Mg concentrations were highest during menses, gradually declined thereafter, reached a low point during the ovulatory phase, and finally rose in the luteal phase, whereas total Mg in the red blood cells (RBCs) remained constant. Facchinetti et al. (6) found an increase in total intracellular Mg in both lymphocytes and polymorphonuclear neutrophils (PMN) in normal asymptomatic women during the luteal phase, but they could find no change in serum total Mg or RBC total Mg content. More recently, Facchinetti et al. (7) showed that women with complaints of PMS had reduced total intracellular Mg in their lymphocytes and PMNs compared with asymptomatic controls. Although no cyclic effects were seen, a statistically significant deficiency of intracellular total Mg was observed in RBCs and mononuclear cells by Rosenstein et al. (8) at each sampling time in women with complaints of PMS as compared with control women. These results confirmed the findings reported previously by Abraham and Lubran for RBCs (9). It is of interest that in an ad lib nutrient-intake sampling of young women, as related to their menstrual cycle, it was found that intakes of vitamin D and Mg, among other ions, were significantly higher during the midluteal phase (10). None of the studies simultaneously measured the biologically active fraction of serum Mg (i.e., ionized Mg 2 ) and the plasma sex steroids; nor were the interrelation of the cations with testosterone measured. Magnesium is found in the blood in four forms: total Mg, protein-bound Mg, Mg complexed to small anion ligands, and the biologically active free ionized Mg (11). The latter makes up about 60% to 70% of the total serum Mg (11). The ionized fraction of Mg has been shown recently to be labile and to correlate with the levels of intracellular free Mg 2 in RBCs of patients with several types of chronic vascular diseases (11, 12). In view of the above findings and of some findings from our laboratory, relating extracellular increases in the Ca 2 to Mg 2 ratio to an adverse effect on blood vessel tone (and cardiovascular diseases in general) (13), we were interested in studying the relation, if any, between ionized and total Mg and ionized Ca and the hormones estrogen, progesterone, and testosterone during one menstrual cycle in normal women. MATERIALS AND METHODS Patients Six parous and four nulliparous female volunteers were included; the subjects had no medical problems and were not taking any medications or drugs. The study was approved by the institutional review board. Informed consent was obtained from all volunteers. All of the subjects were in good health and had a history of regular menstrual cycles before the study. All were nonsmokers. Blood samples were drawn from the women under anaerobic conditions between days 2 4, 6 9, and 11 14 after the menstrual period. The volunteers were asked to use the Ovuquick LH kit to determine the LH surge. Blood samples were then drawn between days 1 3 and 6 9 after the LH surge. Serum was separated from whole blood and stored at 10 C to 20 C for cation analysis and at 79 C for hormone assays. Concentrations of the sex steroid hormones estrogen, progesterone, and testosterone as well as the ionized Ca and Mg levels were measured in the serum of normal cycling women during five different stages: the menstrual, early follicular, late follicular, ovulatory (ovulatory/early luteal), and luteal phases. Hormone and Cation Assays Estradiol, progesterone, and testosterone levels were determined using Coat-a-Count RIA kits (Diagnostic Products Corporation, Los Angeles, CA). The LH kit was Ovuquick (Quidel, San Diego, CA). We measured Ca 2 and Mg 2 with NOVA Stat Profile 8 ion-selective electrodes for Ca 2 and Mg 2, respectively (11). Total Mg was measured with a Kodak DT 60 colorimetric instrument (Rochester, NY), which uses a formazan dye complex (11). The Ca 2 /Mg 2 ratios were calculated (i.e., the measured serum concentration of Ca 2 divided by the measured serum concentration of Mg 2 ), as was the percentage of ionized Mg 2. The normal reference ranges for our laboratory are as follows: Ca 2, 1.09 1.33 mm/l; Mg 2, 0.54 0.64 mm/l; total Mg, 0.74 0.96 mm/l; and % Mg 2, 61 75 (11, 12). Statistical Analysis Where appropriate, data are presented as means SD. The paired Student s t-test was used to ascertain relation between two groups. A P 0.05 was defined as statistically significant. The relations between two variables were analyzed by simple linear regression (with 95% confidence limit) using Sigmaplot graphic software. Correlation analysis was accomplished with Pearson s two-tailed significance test. Multiple regression analysis was performed using SPSS 7.5 software where necessary. RESULTS Table 1 shows the baseline characteristics of the subjects enrolled in the study. We first ascertained that age had no confounding influence on the results. Table 2 presents the mean serum levels of estrogen, progesterone, and testosterone for four different phases of the menstrual cycle. Our data showed a comparatively elevated serum Mg 2 level and a low Ca 2 /Mg 2 ratio early in the follicular phase. Paired t-tests showed a significant decrease in ionized Mg 2 around the time of ovulation, as indicated by the estrogen peak and by urine testing for the LH surge (Fig. 1). We also found a significant decrease in ionized as well as total Mg (paired t-test) when the progesterone concentration peaked (luteal phase) (Fig. 2). Serum levels of total Mg, FERTILITY & STERILITY 959

TABLE 1 Baseline characteristics of the 10 healthy female volunteers at enrollment. Characteristic Mean SD (range) FIGURE 1 Relation between low serum estrogen level (early follicular phase) and high serum estrogen level (ovulation) and serum ionized Mg 2 levels (mm/l) (A) and serum ionized Ca 2 / Mg 2 ratios (B). Values are means SD. Age (y) 30.6 4.8 (23 35) Height (m) 1.6 0.07 (1.5 1.7) Weight (kg) 62.0 7.0 (50 75) Body mass index (kg/m 2 ) 23.4 2.8 (21 27) Cycle length (d) 29.1 3.0 (25 34) Note. None of the subjects were smokers. however, did not differ significantly from one another in the early follicular, late follicular, ovulatory, or menstrual phases, confirming the studies of Facchinetti et al. (6). In addition, we saw a tendency for the Ca 2 level to rise during the luteal phase, confirming the work of Pitkin et al. (4). Although this increase in Ca 2 did not reach statistical significance in our study, the increase in the Ca 2 to Mg 2 ratio was highly significant (paired t-tests), and the latter was already present at the time of ovulation. That is, a significant elevation in the Ca 2 /Mg 2 ratio was seen with the peak in estrogen concentration (Fig. 1B) and at the time of the increase in progesterone (Fig. 3). Finally, we found a statistically significant decrease in Mg 2 (paired t-test), but not in total Mg, and an increase in the Ca 2 /Mg 2 ratio (paired t-test) at the peak serum testosterone level. DISCUSSION We have shown a significant decrease of serum ionized Mg (the biologically active blood fraction of Mg) at about the time of ovulation (Fig. 1) compared with the follicular phase of the menstrual cycle in normal cycling women. The serum total Mg became significantly decreased only during the luteal phase, as demonstrated in Figure 2B. At the same time that the ionized Mg was decreased, it is evident that the serum Ca 2 to Mg 2 ratio was increased significantly, as TABLE 2 Serum hormone levels in different phases of the menstrual cycle. Menstrual phase Estrogen (pg/ml) Hormone Progesterone (ng/ml) Testosterone (ng/dl) Early follicular 47.36 7.67 0.69 0.37 15.52 6.62 Late follicular 77.30 40.58 0.46 0.18 14.35 5.11 Ovulation 184.79 74.76 1.23 0.98 23.72 15.35 Luteal 102.2 36.99 12.35 6.74 19.76 15.33 Note. Values are means SD. shown by the present data. Pitkin et al. (4) had found an elevation of Ca 2 that started just before ovulation and lasted through the luteal phase, whereas it declined during the follicular phase. The fact that the Ca 2 /Mg 2 ratio is increased simultaneously with the peak of estrogen and with the increase in progesterone would confirm that this effect is present throughout the premenstrual period and suggests that this ratio may be related to the PMS complaints that some women have during this period. We have demonstrated recently that an elevated Ca 2 /Mg 2 ratio is associated with the onset of migraine and tension headaches (14 16). Our clinical trials also indicate that these headaches are relieved 960 Muneyyirci-Delale et al. Hormones, divalent cations, and menstruation Vol. 69, No. 5, May 1998

FIGURE 2 Relation between low (early follicular phase) and high (luteal phase) serum progesterone levels and serum ionized Mg 2 (mm/l) (A) and serum total Mg (mm/l) (B). Values are means SD. in most women (who show an elevated Ca 2 /Mg 2 ratio) after IV administration of Mg (15, 16). It is possible that the decrease in serum Mg 2 observed herein at the time of ovulation may result from an intracellular movement of this ion, as suggested by Facchinetti et al. (6), or there may be other effects as well, such as changes in parathyroid hormone or calcitonin, as suggested by Pitkin et al. (4) and other investigators. There seems to be a delay in the decrease of total Mg because it became statistically significant only with the progesterone peak. Thus, it is possible that this decrease is a homeostatic effect that takes time, which may be why some researchers have found it and others have not, because researchers have performed their measurements on slightly different days of the menstrual cycle (2, 6). The decrease in ionized Mg coincident with the peak serum testosterone levels cannot be distinguished from that produced by the estrogen peak because the two hormones peaked at the same time in our study. It is known that the two hormones are interconverted (17), and the resulting testosterone is the sum of that released from the ovaries, adrenals, and peripheral tissues (18). Premenstrual syndrome is defined as the repeated occurrence of either irritability or depression and fatigue during the luteal phase of the menstrual cycle, accompanied by bloated sensations in the abdomen or extremities, breast tenderness, or headache. According to Mortola (19), the term PMS should be reserved for women who demonstrate at least one of the behavioral and one of the physical symptoms. Freeman et al. (20) stated that PMS is characterized by a cluster of negative mood, behavioral, and physical symptoms that consistently occur in the premenstrual week to a degree that disrupts normal functions. The symptoms diminish with the onset of menses. This complex psychoneuroendocrine disorder is a significant health problem for 2% to 10% of cycling women (21, 22). Despite this frequency, rigorous scientific examination of suggested treatments has occurred only recently (20, 23). Because several investigators have found that the intracellular total Mg level is decreased throughout the menstrual cycle in women with complaints of PMS (7 9), it seems reasonable to assume that the presently demonstrated normal cycling levels of Ca 2 and Mg 2 reach critical values in FIGURE 3 Relation between low (early follicular phase) and high (luteal phase) serum progesterone levels and serum ionized Ca 2 / Mg 2 ratios. Values are means SD. FERTILITY & STERILITY 961

these women. These women are therefore susceptible to vascular constriction both in the cerebral and the abdominal circulation during the luteal period. It is well known that low Mg 2 levels result in constriction of these blood vessels (13). It is thus also possible that the water retention (bloating) that occurs during the luteal phase results, in part, from a slight increase in constriction of the arterial renal vessels. We have demonstrated previously, in in vitro and in vivo studies on cerebral and peripheral blood vessels, as well as in in vitro umbilical-placental blood vessels, that an elevated Ca 2 /Mg 2 ratio induces spasms (13, 24). We have been able to reduce varied PMS symptoms associated with low Mg 2 and elevated Ca 2 /Mg 2 levels with the use of magnesium infusions during the second week of the luteal phase (25). Others previously used oral administration of Mg salts, with some success (26, 27). The nervous system may become sensitized and more labile under these conditions, producing increased irritability, depression, and other mood-altering effects (28), as these ions are most important in synaptic transmission (29) and excitationsecretion coupling mechanisms (30). Low-Mg environments have been shown to exacerbate spontaneous discharges from cortical and hippocampal cells in the brain (31). High doses of estrogen that are given as a contraceptive can induce hypercoagulability and thromboembolic disease (18, 32). Indigent pregnant women who are poorly nourished have a much higher risk of preeclampsia or eclampsia than those who are well nourished. All of this is understandable when considering the high doses of the hormones with a virtual loss of blood Mg 2, so that the Ca 2 /Mg 2 ratio is increased. It has been shown repeatedly that this ratio is important for blood vessel tone, excitation-secretion coupling, synaptic transmission, and bone metabolism (1, 13, 30). On the other hand, it is known that low doses of estrogen are protective and beneficial to the cardiovascular system, protective against osteoporosis, and protective to the central nervous system, including retarding the onset of Alzheimer s disease. It would seem that the sex steroids can act differently, depending on the amounts circulating, entraining two prime regulators of key physiologic mechanisms (i.e., Mg 2 and Ca 2 ). In conclusion, our results provide the first rational basis for the simultaneous physiologic changes in cations and sex steroids in the menstrual cycle. These ionic changes probably result in the various PMS symptoms, provided that there is an underlying deficiency of the biologically active Mg or an excess of ionized Ca. These data may also lead to new methods for better diagnosing and managing women with PMS-related complaints and symptoms. References 1. Seelig MS. Interrelationship of magnesium and estrogen in cardiovascular and bone disorders, eclampsia, migraine and premenstrual syndrome. J Am Coll Nutr 1993;12:442 58. 2. Goldsmith NF, Pace N, Baumberger JP, Ury H. Magnesium and citrate during the menstrual cycle: effect of an oral contraceptive on serum magnesium. Fertil Steril 1970;21:292 9. 3. Dale E, Simpson G. Serum magnesium levels of women taking an oral or long-term injectable progestational contraceptive. Obstet Gynecol 1972;39:115 9. 4. Pitkin RM, Reynolds WA, Williams GA, Harris GK. Calcium-regulating hormones during the menstrual cycle. J Clin Endocrinol Metab 1978;47:626 32. 5. Deuster PA, Dolev E, Bernier LL. Magnesium and zinc status during the menstrual cycle. Am J Obstet Gynecol 1987;157:964 8. 6. Facchinetti F, Borella P, Valentini M, Fioroni L, Genazzani AR. Premenstrual increase of intracellular magnesium levels in women with ovulatory, asymptomatic menstrual cycles. Gynecol Endocrinol 1988;2:249 56. 7. Facchinetti F, Borella P, Pironti T, Genazzani AR. Reduction of monocyte magnesium in patients affected by premenstrual syndrome. J Psychosom Obstet Gynaecol 1990;11:221 9. 8. Rosenstein DL, Elin RJ, Hosseini JM, Grover G, Rubinow DR. Magnesium measures across the menstrual cycle in premenstrual syndrome. Biol Psychiatry 1994;35:557 61. 9. Abraham GE, Lubran M. Serum and red cell magnesium levels in patients with premenstrual tension. Am J Clin Nutr 1981;34:2364 6. 10. Martini MC, Lampe JW, Slavin JL, Kurzer MS. Effect of the menstrual cycle on energy and nutrient intake. Am J Clin Nutr 1994;60:895 9. 11. Altura BT, Shirey TL, Young CC, Dell Orfano K, Hiti J, Welsh R, et al. Characterization of a new ion selective electrode for ionized magnesium in whole blood, plasma, serum and aqueous samples. Scand J Clin Lab Invest Suppl 1994;54 (Suppl 217):21 36. 12. Altura BT, Burack JL, Cracco RQ, Galland L, Markell MS, Mauskop A, et al. Clinical studies with the NOVA ISE for IMg 2. Scand J Clin Lab Invest Suppl 1994;54 (Suppl 217):53 68. 13. Altura BM, Altura BT. Magnesium in cardiovascular biology. Sci Am Sci Med 1995;2:28 37. 14. Mauskop A, Altura BT, Cracco RQ, Altura BM. Deficiency in serum ionized Mg but not total Mg in patients with migraine: possible role of Ca/Mg ratio. Headache 1993;33:135 8. 15. Mauskop A, Altura BT, Cracco RQ, Altura BM. Intravenous magnesium sulfate relieves migraine attacks in patients with low serum ionized magnesium levels. Clin Sci 1995;89:633 6. 16. Mauskop A, Altura BT, Cracco RQ, Altura BM. Intravenous magnesium sulfate rapidly alleviates headaches of various types. Headache 1996;36:154 60. 17. Yen SSC. Chronic anovulation caused by peripheral endocrine disorders. In: Yen SSC, Jaffe RB, editors. Reproductive endocrinology. Philadelphia: Lippincott-Raven, 1991:580. 18. Adashi EY, Rock JA, Rosenwalks Z. Reproductive endocrinology, surgery, and technology. Vol 2. Philadelphia: Lippincott-Raven, 1996. 19. Mortola JF. Issues in the diagnosis and research of premenstrual syndrome. Clin Obstet Gynecol 1992;35:587 98. 20. Freeman EW, Rickels K, Sondheimer JJ, Polansky M. A double-blind trial of oral progesterone, alprazolam, and placebo in treatment of severe premenstrual syndrome. J Am Med Assoc 1995;274:51 7. 21. Woods NF, Most A, Dery EK. Prevalence of perimenstrual symptoms. Am J Public Health 1982;72:1257 64. 22. Logue AM, Moos RH. Perimenstrual symptoms: prevalence and risk factors. Psychosom Med 1986;48:388 414. 23. Moline MI. Pharmacologic strategies for managing premenstrual syndrome. Clin Pharmacol 1993;12:181 96. 24. Altura BM, Carella A, Altura BT. Magnesium deficiency-induced spasms of umbilical vessels: relation to preeclampsia, hypertension, growth retardation. Science 1983;221:376 8. 25. Mauskop A, Altura BT, Cracco RQ, Altura BM. An open trial of magnesium supplementation for the treatment of migraines and symptoms of premenstrual syndromes in premenopausal women. Effect on serum ionized Mg levels. Neurology 1997;48(Suppl 2):A261 2. 26. Facchinetti F, Sances G, Genazzani AR, Nappi G. Magnesium prophylaxis of menstrual migraine: effects on intracellular magnesium. Headache 1991;31:298 301. 27. Facchinetti F, Borella F, Sances G, Fiorini I, Nappi RE, Genazzani R. Oral magnesium successfully relieves premenstrual mood changes. Obstet Gynecol 1991;78:177 84. 28. Backström T. Neuroendocrinology of premenstrual syndrome. Clin Obstet Gynecol 1992;35:612 28. 29. Chutkow JG. Magnesium and the peripheral nervous system: metabolism, neurophysiological functions and clinical disorders. In: Sigel H, Sigel A, editors. Metal ions in biological system. Vol 26. New York: Marcel Dekker, 1990. 30. Aikawa JK. Magnesium: its biological significance. Boca Raton (FL): CRC Press, 1981. 31. Walther H, Lambert JDC, Jones RSG, Heinemann U, Hamon B. Epileptiform activity in combined slices of the hippocampus, rubiculum and entorhinal cortex during perfusion with low magnesium medium. Neurosci Lett 1986;69:156 61. 32. Goldzieher JW, Dozier TS. Oral contraceptives and thromboembolism: a reassessment. Am J Obstet Gynecol 1975;123:878 914. 962 Muneyyirci-Delale et al. Hormones, divalent cations, and menstruation Vol. 69, No. 5, May 1998