Clinical Endocrinology (1989), 31, 31-38

CORRECTION OF LOW CIRCULATING LEVELS OF

DURING REVERSAL OF HYPOMAGNESAEMIA 1,25-DIHYDROXYVITAMIN D BY 25-HYDROXYVITAMIN D

M. FUSS, P. BERGMANN, A. BERGANS, J. BAGON, E. COGAN, T. PEPERSACK, M. VAN GOSSUM AND J. CORVILAIN

Services de Mkdecine Interne et de Biologie Unique , H6pital Universitaire Brugmann, UniversitP Libre de Bruxelles, and DPpartement de Ngphrologie, H6pital Saint Jean,

Brussels. Belgium

(Received 21 November 1988: returned for reuision 27 January 1989;.finalIy revised 6 February 1989; accepted 28 February 1989)

S U M M A R Y

The effect of 25-hydroxyvitamin D (250HD), given orally during the reversal of hypomagnesaemia, was studied in five patients with hypomagnesaemic hypo- calcaemia and low serum levels of 250HD and 1,25-dihydroxyvitamin D (1,25(OH)2D). The results were compared to those obtained in five other patients with similar initial levels of magnesium, calcium, 250HD and 1,25(OH)*D who did not receive 250HD. Serum levels of 1,25(OH)zD in the ten hypomagnesaemic patients were lower than in ten control subjects with low serum levels of 250HD. The reversal of hypomagnesaemia was similar in the two groups of patients and elicited a similar increase of circulating iPTH levels. The expected increase of circulating 250HD was observed in patients supplemented with 250HD; their circulating 1,25(OH)2D rose within 48 h to normal levels, contrasting with the delayed and poor increase of 1 ,25(OH)2D in patients receiving no 250HD. The evolution of serum calcium was however identical in the two groups. Our results suggest that vitamin D deficiency was a significant factor leading to low circulating levels of 1 ,25(OH)2D in hypomagne- saemic hypocalcaemic patients. The biological consequences of low serum 1,25(OH)2D in these patients remain unclear, but clearly, normal levels of 1,25(OH)2D are not essential for the correction of hypomagnesaemic hypocal- caemia.

In a previous work (Fuss et al., 1985) we found low serum levels of 25-hydroxyvitamin D(250HD) and 1,25-dihydroxyvitamin D( 1,25(OH)zD) in patients with hypocalcaemia secondary to hypomagnesaemia. Magnesium administration induced a rapid increase of serum concentrations of parathyroid hormone and calcium. However, the rise of circulating 1,25(OH)2D levels was slow and delayed as compared with that of calcium.

Correspondence: Dr M. Fuss, Service de Medecine Interne, Hapita1 Universitaire Brugmann, Place Van Gehuchten, 4, 1020 Bruxelles, Belgium.

31

32 M . Fuss et al.

Whether higher serum concentrations of the substrate, 250HD, would permit a more rapid and greater increase of serum 1,25(OH)2D and calcium concentrations is open to conjecture. The aim of the present work was to answer this question by giving 250HD to hypomagnesaemic hypocalcaemic patients during the reversal of their hypomagnesaemia and to compare their circulating levels of 1,25(0H)zD and calcium to the results obtained in the previous work.

MATERIALS AND METHODS

Study group

Five patients (one man, four women, aged 46 to 85, mean 65-2 years) with hypomagnesae- mic hypocalcaemia composed the present study group. Hypomagnesaemia could be attributed to severe alcoholism, to alcoholism and diuretics, to cisplatin toxicity, or to diabetic ketoacidosis. The diabetic patient received magnesium and 250HD 5 days after the correction of ketoacidosis; serum level of magnesium had remained unchanged during the 5 days. None of the patients exhibited clinical or radiological features of bone disease. Three of them had some degree of renal failure (serum creatinine levels 1 15-327 pmolll). All patients ate the normal hospital diet throughout the study.

Control subjects

Ten patients (nine men, one woman, aged 29 to 69, mean 51.0 years), without disease or treatment known to alter calcium and magnesium metabolism and with serum 250HD concentrations similar to those of the five hypomagnesaemic patients, were selected as controls for pretreatment serum levels of 1,25(OH)zD.

Protocol of study

Magnesium sulphate (MgS04.7H20, 0.51 mEq of magnesium/kg a day) was injected intramuscularly in divided doses for 6 days from day zero, except in patient 5 in whom serum magnesium spontaneously increased during the survey. All patients received pharmacological doses of 250HD orally, in order to insure immediate availability of large amounts of the substrate: 5 pg/kg at day zero, then 2 pg/kg on each of the following 6 days. The following quantities were measured in serum before and during treatment: creatinine, magnesium, calcium, phosphate, immunoreactive PTH (iFTH), 250HD, 1,25(OH)2D and vitamin D binding protein (DBP). Informed consent to the treatment was obtained from all patients.

Analytical procedures

Serum magnesium was measured by atomic absorption spectrophotometry, serum creatinine, calcium and phosphate by using a Technicon Auto Analyzer and serum albumin by electrophoresis. Serum calcium concentration was corrected for serum albumin (Pain et al., 1975) before and at the end of the study, using the following formula: serum corrected calcium = serum calcium (mmo 1/1) + 0.2 (4.1 - serum albumin), serum albumin being expressed in g/dl.

EHect of vitamin D in hypomagnesaemia 33

Table 1. Blood data (means+SE) during reversal of hypomagnesaemia in hypomagnesaemic hypocalcaemic patients

Days after starting therapy Before therapy 1-2 3-5 6-7

(a) Present study. Patients receiving 250HD supplements Magnesium (mmol/l) 0.41 0.78 0.82 (0.74-1 '00) 0.07 0.07 0.15 Mid-mol iPTH(pg/ml) 1 142 1535 1049 (290-8 50) 436 459 272 Intact iPTH (pg/ml) 66 95 58 (10-55) 20 20 18 25 OH D (nmolil) 19.5 36.0 43.5

1,25 (OH)2D (pmo1/1) 35.0 74.4 81.1

DBP(pmo 1 /I) 5.1 5.3 5.7

(30-75) 8.8 12.5 17.0

(48- 168) 9.1 12.5 14.2

(3.9-8.2) 0. 4 0.5 0.5 Phosphate (mmol/l) 1.07 1.20 1.46 (0.80- I ,451 0.10 0.16 0.13 Calcium (mmol/l) I .83 I .98 2.13 (2.25-2.63) 0.13 0.03 0.07

(b) Previous study. Patients receiving no 250HD supplements Magnesium (mmol/l) 0.39 0.74 0.86 (0.74- 1 '00) 0.04 0.08 0.07 Nt-iPTH (pgiml) 530 743 680 (300-620) 48 92 47 25 OH D (nmoI/l) 10.0 9.0 8.0 (30-75) 3.0 2.0 1.8 1,25(OH)zD (pmoI/l) 28.8 27.4 30.7 (48-168) 7.0 6.5 8.2 Phosphate (mmol/l) 0.84 1.07 1.17 (040-1.45) 0.13 0.19 0.25 Calcium (mmol/l) 1.80 I .93 2.10 (2.25-2.63) 0.20 0.20 0.18

0.78 0.13

994 237 37 11 88.5 28.0 86.4 9.6 5.8 0.4 1.53 0.19 2.23 0.04

0.92 0.09

665 55 8.0 3.5

45.6 3.4 1.20 0.23 2.25 0.05

SE, Standard error of the mean; iPTH, immunoreactive parathyroid hormone; 250HD. 25-hydroxyvitamin D; 1,25(OH)2D, 1,25-dihydroxy- vitamin D; DBP, vitamin D-binding protein. Normal values are indicated in parentheses.

Serum iPTH level was determined by using two commercial assays from ImmunoNuc- lear Corporation, a radioimmunoassay which recognizes mid-molecule fragments of PTH and an immunoradiometric assay which measures the intact hormone.

Serum 250HD (limit of detection 5 nmo1/1 of serum) and 1,25(0H)zD (limit of detection 12 pmo1/1) were measured respectively by the methods of Belsey et al. (1974) and of Bouillon et al. (1980). Undetectable serum concentrations of 250HD and 1,25(OH)*D were arbitrarily assigned the numerical value of the assay sensitivity, in order to permit calculations. Serum DBP level was measured by R. Bouillon (Bouillon et al.,

34 M . Fuss et al.

1977). All measurements of serum iPTH, 250HD, 1,25(OH)2D and DBP in each patient were performed in the same assays in order to avoid interassay variations.

Twenty-seven normal subjects from the medical staff (1 5 men, 12 women, aged 24 to 43, mean 43 years) served to define the normal range of serum data, except for serum concentrations of iPTH, 1,25(OH)zD and vitamin D-binding protein (DBP), where the data from 33, 42 and 180 normal subjects were used respectively.

Unpaired Student’s t-test was used to compare patients and control subjects and paired Student’s t-test to appreciate the significance of changes in the patients. Serial measurements of serum DBP levels were submitted to analysis of variance.

RESULTS

Biochemical data before reversal of hypomagnesaemia

Present study The initial serum data from the present study are given in Table l(a) and/or shown in

Fig. 1. Serum magnesium and calcium concentrations were significantly (P < 0-01) lowered. Serum calcium concentrations corrected for low serum albumin (3.5 g/dl k 0.4 SE. Normal range: 3.5-4.7 g/dl) was also low (1.93 f 0.13 mmol /l. N: 2-30-2.52 mmo 1/1, P< 0.02, results not shown). Serum phosphate concentrations were normal.

Three patients (individual data not shown) had normal serum levels in the mid- molecule specific assay and low circulating levels of intact PTH; the two other patients had high serum levels using the two assays; both had renal failure. The five patients showed lower serum values of 250HD (P<O.Ol) and of 1,25(OH)zD (P<O.OOl) than normal subjects (Table 1 (a)). Serum DBP concentration was normal in hypomagnesae- mic patients. No correlation appeared between the serum concentrations of 1,25(OH)*D and those of 250HD, iPTH, calcium, phosphate or magnesium, nor between serum concentrations of calcium and iPTH.

Comparison with the data from the previous work Patients from the present and from the previous study were not significantly different in

age (mean f SD : 65.2 & 6.5 us 57.6 f 8.0). As indicated in Tables l(a) and (b), their initial serum concentrations of magnesium, calcium, phosphate, 250HD and 1,25(OH)2D were not different.

Comparison of serum I,2S(OH)2D levels in hypomagnesaemic patients and in control subjects with low circulating levels of 250HD

The 10 patients pooled from our present and previous studies (Table 2) had lower circulating levels of 1,25(OH)2D (P < 0.002) than the 10 normomagnesaemic control subjects with similar low levels of 250HD. Four patients from the two studies had some degree of renal insufficiency, a cause of low circulating 1,25(OH)2D levels; when these four patients were excluded, hypomagnesaemic patients still showed lower circulating levels of 1,25(OH)2D than the 10 control subjects (Table 2).

Eflect of vitamin D in hypomagnesaemia Treatment

35

t I

L I Mid-mol AS

0 5 Ti m e (days)

Fig. I , Effect of recovery from hypomagnesaemia on serum data (meanf SE) from five patients receiving *, 250HD supplementation compared to the data from a previous study in five patients who 0, did not receive 25OHD. Mg, Magnesium; Ca, calcium; P, phosphate; iPTH, immunoreactive parathyroid hormone, percentage of starting value, using assays recognizing either mid-molecule (mid-mol AS) or amino-terminal fragments (Nt AS) of F'TH.

Metabolic changes during 250HD supplementation and reversal of hypomagnesaemia

Present study Metabolic changes during treatment are given in Table l(a) and/or shown in Fig. 1

(closed circles). Serum magnesium level had increased (P < 0.03) after 1-2 days. Serum

36 M . Fuss et al.

Table 2. Serum concentrations (meanskSE) of 250HD and 1,25(OH)zD in patients pooled from two studies and in control subjects with low

circulating levels of 250HD

Hypomagnesaemic patients

Control subjects with low serum

250HD All cases

n 10 10 250HD (nmol/l) 13.0 14.8 (30-75) 3.5 4.8 1,25(OH)2D (pmol/l) 84.2 32.6 (48-1 68) 9.1 5.3

Patients with normal serum

creatinine levels

6 12.0 3.5

32.4 7.0

250HD, 25-hydroxyvitamin D; 1,25(OH)zD, 1,25-dihydroxyvitamin D.

mid-molecule and intact iPTH levels rose (Pc0.02) after 1-2 days, using both assays, even in patients in whom circulating iPTH level was elevated before treatment. Serum concentrations of phosphate also increased. Serum 250HD slowly increased during the administration of pharmacological doses of this metabolite. Circulating 1,25(OH)2D level rose (P=O.Ol) as early as 1-2 days after administrating 250HD, to levels not different from normal values. Variance analysis of serial serum DBP measurements disclosed no significant change.

Serum calcium concentrations increased progressively and were different from pretreatment values after 6-7 days (P<0.05), though remaining lower than normal (P c 0.001) throughout the study, even when serum calcium was corrected for serum albumin (2.32 k 0.03 mmo1/1, P < 0.001).

Comparison with the data from the previous work As reported in Tables l(a) and (b), and shown in Fig. 1, the increase of serum levels of

magnesium and phosphate showed a similar time-course in the two studies. The relative increment of serum iPTH levels using the two assays in the present study was not significantly different from that obtained in the previous study by using a N-terminal- specific assay.

As expected, serum 250HD concentration increased in patients supplemented with 250HD and remained unchanged in patients receiving no 250HD. Serum 1,25(OH)2D concentrations rose earlier and to higher levels (P < 0.01) during 250HD supplemen- tation than in the patients who were not supplemented with 250HD. The time-course of serum calcium (not corrected for serum albumin) did not differ in the two studies.

DISCUSSION

The aim of the present work was to investigate the effect of 250HD administration on serum 1,25(OH)2D and calcium concentrations in magnesium and vitamin D-deficient patients, during the recovery from their hypomagnesaemia. For that purpose we

Efect of vitamin D in hypomagnesaemia 37

compared the results obtained in patients supplemented with 250HD to the results obtained in previous patients who were not supplemented with 250HD. A prerequisite for such a comparison was that the groups of patients would not differ with respect to the other factors known to influence vitamin D metabolism, such as serum creatinine, calcium, phosphate, magnesium, 250HD, 1,25(OH)2D and PTH. This appeared to be the case, except that it was not possible to compare the initial parathyroid status in the patients from the two studies since different assays were used. The diversity of results obtained in our two studies, depending on the individual patients and on the assay used, has been reported in the literature. Indeed, serum concentrations of iPTH have been found low (Anast et al., 1976; Bar et a]., 1975; Ralston et al., 1983), normal (Chase & Slatopolsky., 1974; Suh et al., 1973) or high (Medalle et al., 1976; Allgrove et al., 1984; Selby et al., 1984) in hypocalcaemic magnesium-deficient patients. Nevertheless, the correction of hypomagnesaeniia elicited a comparable transient increase of serum iPTH levels in our two groups, whatever the assay used. Therefore it is very likely that the earlier and greater increase of serum 1,25(OH)2D levels in the present work can be ascribed to 250HD supplementation. The free fraction of circulating 1,25(OH)2D also increased under treatment since serum DBP did not change significantly.

Our results thus suggest that vitamin D deficiency was a significant factor in the slow and poor increase of circulating 1,25(OH)2D in hypomagnesaemic hypocalcaemic patients receiving magnesium salts without 250HD supplements.

However, 250HD depletion was not the sole cause of the alterations of serum 1 ,25(OH)2D levels in magnesium-deficient patients, since, before 250HD supplemen- tation, no correlation appeared between the circulating concentrations of the two metabolites and since they showed lower serum concentrations of 1 ,25(OH)2D than control subjects for comparably low 250HD serum levels. Magnesium deficiency by itself could induce low 1,25(OH)2D serum levels in two ways, either through low serum levels of iPTH, when this is the case, or through a depression of the 250HD- 1 alpha hydroxylase system. Indeed, experimental data show that magnesium supports 250HD-la-hydroxy- lase activity (Ghazarian & De Luca, 1974; Fraser & Kodicek, 1979).

The biological consequences of low serum 1,25(OH)2D concentrations in hypomagne- saemia are not yet well defined; the rapid restoration of normal circulating 1,25(OH)2D remained without detectable benefit on the correction of hypocalcaemia. This observa- tion re-emphasizes the conclusion of our earlier report and of other authors (Rude et a/., 1985) that normal circulating levels of 1,25(OH)2D are not needed for the reversal of hypomagnesaemic hypocalcaemia.

ACKNOWLEDGEMENTS

This work was supported by Grant 3.4543.79 of the FRSM of Belgium. The authors acknowledge the nurses of the Endocrine and Metabolic Unit, and Mrs Soffers for illustrative assistance. They are grateful to Professor R. Bouillon for providing us with its ‘H’ antiserum against 1 ,25(OH)2D and for serum DBP measurements.

REFERENCES

ALLGROVE, J., ADAMI, S., FRAHER, L., REUBEN, A. & O’RIORDAN, J.L.H. (1984) Hypomagnesaemia: studies of

ANAST, C.S., WINNACKER, J., FORTE, L. & BURNS, T.W. (1976) Impaired release of parathyroid hormone in parathyroid hormone secretion and function. Clinical Endocrinology, 21,435449.

magnesium deficiency. Journal of Clinical Endocrinology and Metabolism. 42,707-7 17.

38 M . Fuss et al.

BAR, R.A., WILSON, H.E. & MAZZAFERRI, E.L. (1975) Hypomagnesaemic hypocalcaemia secondary to renal magnesium wasting. A possible consequence of high-dose gentamicin therapy. Annals of Internal Medicine, 82, 646-649.

BELSEY, R.E., DE LUCA, H.F. & POTTS, T.T. JR (1974) A rapid assay for 25-OH-vitamin D3 without preparative chromatography. Journal of Clinical Endocrinology and Metabolism, 38, 1046- 105 1.

BOUILLON, R., DE MOOR, P., BAGCLIONI, E.G. & USKOKOVIC, M.R. (1980) A radioimmunoassay for 1.25- dihydroxycholecalciferol. Clinics/ Chemistry, 26, 562-567.

BOUILLON, R., VAN BAELEN, H. & DE MOOR, P. (1977) The measurement of the vitamin D binding protein in human serum. Journal of Clinical Endocrinology and Metabolism, 45,225-23 1.

CHASE, L. & SLATOPOLSKY, E. (1974) Secretion and metabolic efficacy of parathyroid hormone in patients with severe hypomagnesemia. Journal of Clinical Endocrinology and Metabolism. 38, 363-37 I.

FRASER, D.R. & KODICEK, E. (1970) Unique biosynthesis by kidney of a biolocally active vitamin D metabolite. Nature, 228, 764-766.

Fuss, M., COGAN, E., GILLET, C., KARMALI, R., GEURTS, J., BERGANS, A., BRAUMAN, H., BOUILLON, R. & CORVILAIN, J. (1985) Magnesium administration reverses the hypocalcaemia secondary to hypomagnesae- mia despite low circulating levels of 25-hydroxy-vitamin D and 1,25-dihydroxyvitamin D. Clinical Endocrinology, 22,807-8 15.

GHAZARIAN, J.G. & DE LUCA, H.F. (1974) 25-Hydroxycholecalciferol- 1-hydroxylase: a specific requirement for NADPH and a hemoprotein component in chick kidney mitochondria. Archives of Biochemistry and Biophysics, 160, 63-72.

MEDALLE, R., WATERHOUSE, C. & HAHN, T.J. (1976) Vitamin D resistance in magnesium deficiency. American Journal of Clinical Nutrition, 29, 558-558.

PAIN, R.W., ROWLAND, K.M., PHILIPS, P.J. & DUNCAN B.McL. (1975) Current ‘corrected’ calcium concept challenged. British Medical Journal, 4,617-619.

RALSTON, S., BOYLE, L.T., COWAN, R.A., CREAN, G.P., JENKINS, A. & THOMSON, W.S. (1983) PTH and vitamin D responses during treatment of hypomagnesaemic hypoparathyroidism. Acta Endocrinologica, 103,535- 538.

RUDE, R.K., ADAMS, J.S., RYZEN, E., ENDRESS, D.B., NIIMI, H., HORST, R.L., HADDAD, J.G. & SINGER, F.R. (1985) Low serum concentrations of 1,25-dihydroxyvitamin D in human magnesium deficiency. Journal of Clinical Endocrinology and Metabolism, 61,933-940.

SELBY, P.E., PEACOCK, M. & BAMBACH C.P. (1984) Hypomagnesemia after small bowel resections. Treatment with 1, alpha-hydroxylated vitamin D metabolite. British Journal of Surgery, 71,334-337.

SUH, S.M., TASHJIAN, A., MATSUO A., PARKINSON D. & FRASER D. (1973) Pathogenesis of hypocalcemia in primary hypomagnesemia: normal end-organ responsiveness to parathyroid hormone, impaired para- thyroid gland function. Journal of Clinical Investigation, 52, 153-160.