- Email: [email protected]
Vitamin D metabolism in insulin-dependent diabetes mellitus
0221-8747/83 $3.00 + .OO Copyright 0 1984 Pergamon Press Ltd.
Bone Dis. & Rel. Res. 5, 107-l 10 (1983) Printed in the USA. All rights reserved.
Metab.
VitaminD Metabolism in Insulin-dependent Diabetes Mellitus T.L. STORM, O.H. S@RENSEN,Bj. LUND, Bi. LUND, J.S. CHRISTIANSEN, A.R. ANDERSEN, I.B. LUMHOLTZ, and H.-H. PARVING
Departments of Medicine and Rheumatology, Sundby Hospital, Department of Orthopedic Surgery, Rigshospitalef, Steno Memorial Hospital, Department of Medicine F; Herlev Hospital, Copenhagen, Denmark. Address for correspondence Copenhagen S, Denmark.
and reprints: 0. Helmer S$rensen, Department of Medicine, Sundby Hospital, ltaliensvej
1, DK-2300
by the demonstration of reduced activity of renal 25(OH)D1cr-hydroxylase (Spencer et al., 1980) and reduced levels of circulating 1,25(OH),D (Schneider et al., 1977), both of which were returned to normal by insulin treatment. Our study wasaimed at elucidating possible derangements of vitamin D metabolism in adult IDDM patients during ordinary, poor, and good metabolic control. The influence of diabetic nephropathy on vitamin D metabolism was also evaluated.
Abstract A significant loss of bone has been observed in diabetes mellitus. The patogenesis is unknown, but an impairment of vitamin D metabolism might be involved. Consequently, we have studied vitamin D metabolism in five groups of insulindependent diabetic patients. Significantly reduced levels of serum 25(OH)D were seen only in patients with diabetic nephropathy. The serum levels of 1,25(OH),D were reduced only in diabetic ketoacidosis but normalized during recovery. It is concluded that vitamin D metabolism is largely normal in adult insulin-dependent diabetes, and it seems unlikely that a disturbance of the vitamin D metabolism can explain the bone loss in the ordinarily controled diabetics.
Patients and Methods The vitamin D metabolites were studied in five groups of insulindependent diabetic patients (IDD):
Key Words: Insulin-dependent Diabetes-25(0H)D1,25(0H),D_Diabetic Nephropathy-Ketoacidosis.
I. 32 patients (age 15-45) wlthout proteinuria (Albustlx) and normal serum creatinine. II. 13 pattents (age 20-41) with dlabetlc nephropathy and normal serum creatinine were studied over a 2-year period. Ill. 14 patients (age 34-73) with diabetic nephropathy and elevated serum creatinine (> 150 ~molll). IV 9 newly diagnosed, untreated patients (age 21-35), blood glucose from 15-25 mmolll, ketonuria. but normal standard bicarbonate. The patients were either strictly regulated by five or six dally injections of soluble Insulin or by an artificial betacell (Biostator). The glomerular filtration rate (GFR) was measured on admission and after 7-10 days of regulation. V. 6 patients admitted with acute ketoacidosis and treated with lowdose insulin. None of the patients had persistent proteinuria, and serum creatinine was normal before this episode.
Introduction A significant loss of bone has been observed in both insulindependent (IDDM) (Levin et al., 1976; Ringe et al., 1976; Rosenbloom et al., 1977; McNair et al., 1979; McNair et al., 1981) and noninsulin-dependent diabetes (Levin et al., 1976; Shore et al., 1981). The pathogenesis of the bone loss is unknown, but an impairment of the vitamin D metabolism might be involved. Christiansen et al. (1982) found slightly decreased levels of serum 25(OH)D and markedly reduced serum levels of 24,25(OH),D. Reduced serum levelsof the biologically most active vitamin D metabolite, 1,25-dihydroxyvitamin D [1,25(OH),D], have been found in diabeticchildren (Gertner et al., 1979; Fraser et al., 1981), whereas normal levels have been reported in insulin-treated adult patients (Gertner et al., 1980; Heath et al., 1979). Studies in streptozotocin diabetic rats have shown reduced intestinal calcium absorption, which was restored by administration of 1 ,25(OH)*D3. The authors concluded that the defect in vitamin D metabolism was probably at the 1cr-hydroxylation step in the kidney (Schneider et al., 1976). This was supported
None of the patients in groups I-III had ketonuna. Diabetic nephropathy was diagnosed clinically if the following criteria were fulfilled: persistent proteinuria of more than 0.5 g/24 h, duration of IDDM of more than 10 years, presence of diabetic retinopathy, and no clinical or laboratory evidence of disease of the kidneys or the renal tract other than diabetic glomerulosclerosis. Two pattents did not fulfill all these cntena. A kidney biopsy was therefore performed, and in both cases slight or moderate diffuse diabetic glomerulosclerosis was found. Assays for circulating vitamin D metabolites were as follows: 25(OH)D was measured bv a comoetitive orotein-bindina assav. narmal mean 28 -c 15 ng/ml (SD) (Lund and SQrensen~~&$ The 24,25(0H),D and 1,25(OH),D metabolites were separated by gel 107
108
T L. Storm et al.: Vitamin D metabolism in diabetes mellitus
chromatography followed by high pressure liquid chromatography. They were measured by ligand binding assay using rachitic rat kidney cytosol and rachitic chick intestinal cytosol, respectively, as binding proteins. The normal values were 1.5 + 0.9 nglml (SD) for 24,25(OH),D and 33.1 f 15.3 pglml (SD) for 1,25(OH),D (Lund etal., 1979). GFR was measured after a single i.v. injection of “0-EDTA at 9 AM
by studying the plasma disappearance for 4 h (Br@chner-Mortensen, 1972). This procedure was applied to all patients in group II. The patients in group IV had their GFR measured using a constant infusion technique with ‘251-thalamate(Mogensen, 1971). Serum creatinine and blood glucose were measured on an auto-analyzer. Statistical analyses were performed using the Wilcoxon test for paired and
unpaireddataand Spearmanrankcorrelation.
Results Serum 25/OH)D
Normal values were seen in the patients without proteinuria (groups I and IV), whereas significantly (P < 0.01) reduced levels were found when diabetic nephropathy was present (groups Ii and Ill) (Fig. 1). The serum concentrations did not change during recovery from ketoacidosis (group V). Serum 24,25(OH),D
A trend toward reduced values was found in the groups studied (normal GFR without proteinuria 0.9 + 0.5 nglml, normal GFR with proteinuria 1.O f 0.7 nglml, age-matched controls 1.5 ? 0.9 nglml). No significant changes were seen during recovery from ketoacidosis. Serum 24,25(OH),D were not measured in group IV. Serum 1,25(OH),D
This was measured in groups II-V. As shown in Figure 2, the mean levels were normal in groups II-IV, although a few low values were found. No correlations could be found between the serum levels and GFR or serum creatinine. The GFR fell significantly (P < 0.01) during the 2 years of observation in group II, but no significant changes occurred in the serum 1,25(OH)2D concentrations (Fig. 3). Serum 1,25(OH)D2 was in the upper normal range on admission in group IV. Significant falls were Serum 25-OHD (nglml) 601 50
-1 t
30
8
:
-08 _ ::.
pco.01
seen in both the GFR (P < 0.01) and the serum 1,25(OH),D (P < 0.01) during strict regulation (Fig. 4). Reduced (P < 0.02) levels of circulating 1,25(OH),D were found on admission in acute ketoacidosis, with rapid normalization following correction of the acidosis (Figs. 1 and 5). Discussion Streptozotocin diabetes in the rat isassociated with decreased serum concentrations of 1,25(OH)2D (Schneider et al., 1977) as a consequence of reduced activity of the renal 1cr-hydroxylase enzyme system (Schneider et al., 1976) due to lack of insulin (Spencer et al., 1980). These observations could, however, not be confirmed in the present human study, since all the young, nonobese, untreated, newly diagnosed patients (group IV) had normal serum 1,25(OH),D levels or even concentrations in the upper normal range on admission. The supernormal values might partly be explained by a hemoconcentration as indicated by a fall in hematocrit from 0.43 f 0.02 to 0.40 + 0.01 (mean + SEM) following treatment. Furthermore, the serum levels of several hormones are elevated during acute stress, for example, growth hormone, which is known to stimulate the 1ar-hydroxylase. There was no correlation between renal function and serum 1,25(OH),D. Mean serum 1,25(OH),D concentrations were normal in the patients with normal GFR + proteinuria(group II) and in those with reduced GFR + proteinuria (group Ill), indicating that only minor disturbances occur in serum levels of 1,25(OH),D in ordinarily controled IDDM. The mean concentrations of 24,25(OH)2D were slightly but not significantly reduced in the two groups of patients studied. The mean 25(OH)D concentrations were reduced in patients with impaired kidney function. There was no difference in dietary intake between the groups that could explain the difference in serum 25(OH)D levels, but this might be due to renal loss of 25(OH)D and its binding protein (Goldstein et al., 1981). There was, however, no correlation between the degree of renal insufficiency and the reduction of circulating 25(OH)D. Severely reduced serum concentrations of 25(OH)D (lower than 5 nglml where osteomalacia might occur) were seen only in two young men with moderate and severe renal insuffiency, respectively.
Serum 1.25-(OHbD (Pglmt) 60 -
pco.02
P
.
40 : .
.
-_8+ W” 0
.
. : :
30 2o-
. k
‘O-
: .
: .
. ..
“I_ .
.
.
. .
.
. .
i .
?? :_ -
??
-.-
-
i Proteinuria Renal insuff. GROUP
-
+-
I
II
+
-
+ III
IV
Controls
Fig. 1. Serum levels of 25(OH)D in insulin-dependent diabetic patients with and without renal impairment compared to normal controls.
Proteinuria Renal insuff. GROUP
+ II
;r
+ +
-
III
IV
A
-&% J& Ji ‘5
. . . V
Controls
Fig. 2. Serum levels of 1,25(OH),D in insulin-dependent diabetic patients with and without renal impairment compared to normal controls.
109
T. L. Storm et al.: Vitamin D metabolism in diabetes mellitus Serum 12%(OH)2D (pglmt)
;;25PH)2D
GFR mllmin
t
t
1 120 100 80 60
4oit-__---*
I
2 years
Start
mean
Start
2 years
Fig, 3. Renal function (GFR) and serum levels of
1,25(OH),Dduring 2 years of observation in insulin-dependent patients with renal involvement.
A similar renal loss of the dihydroxylated metabolites would also be expected in patients with impaired kidney function. The normal levels of circulating 1,25(OH),D in patients with proteinuria might be due to increased production, since this hormone is strictly physiologically regulated, in contrast to 25(OH)D. Furthermore, the more polar metabolites are less completely bound and, therefore, less prone to vary with the circulating vitamin D-binding protein concentration. Ketoacidosis was accompanied by significant decrease (P < 0.02) in the circulating 1,25(OH)$ levels. This is in agreement with animal experiments that have shown that metabolic nonketotic acidosis impairs the conversion of 25(OH)D to 1,25(OH),D (Lee et al., 1977; Baran et al., 1982; Sauveur et al., 1977). Since the lcr-hydroxylation of 25(OH)D is regulated in part by parathyroid hormone (PTH) (Hughes et al., 1975) and since renal response to PTH is impaired during acidosis (Beck et al., 1975) one might expect the low levels of 1,25 (OH),D seen initially in ketoacidosis. In serum-free medium, kidney cells require insulin in order to respond to PTH with increased 1,25(OH),D production (Henry, 1981) but since the patients in group IV had normal serum levels of 1,25(OH)*D, it seems unlikely that lack of insulin can explain the low values seen in the acidotic patients. Serum phosphorus levels in untreated patients with diabetic ketoacidosis are either normal or elevated, but with the GFR mllmin
24
7.3s2.5 8.99.5
17.7t3.1 19.2t36
48
72 Hours??
24.1+-09 244~3.5
Fig. 5. Serum levels of 1,25(OH),D during recovery from diabetic
ketoacidosis. institution of therapy the levels decline rapidly as phosphorus shifts back into the cellular compartment (Knochel, 1977). Since low serum phosphorus concentration stimulates renal synthesis of 1,25(OH),D in humans (Gray et al., 1977) hypophosphatemia might in part be responsible for the transitory hypernormal serum levels of 1,25(OH),D seen during recovery from ketoacidosis. In the present study, vitamin D metabolism was largely normal in adult insulin-dependent diabetes, which is in agreement with the results of Heath et al. (1979) and Gertner et al. (1980). Moderately reduced serum 25(OH)D levels were, however, encountered when renal function was affected. Circulating levels of 1,25(OH),D were significantly reduced only in ketoacidosis and normalized when the acidosis was corrected. We conclude that it is unlikely that a disturbance of vitamin D metabolism can explain the bone loss described in ordinarily controled diabetics.
Acknowledgement: This study was supported by grants from The
Danish Association against Rheumatic Diseases, Landsforeningenfor Sukkersyge, and the Danish Medical Research Council. The skillful technical assistanceof Merete Frandsen and Ulla M. Smidt is gratefully acknowledged.
References
1.25-(OH)2D pglmt
.
60 210 50 190 40 170 30 150 70 130 10 110 Aam1ssion
H(‘ + &_~~
(2113-311) &l/l)
..
After
regulation
.
. Admtssron
After regulation
Fig. 4. Renalfunction (GFR) and serum levels of 1,25(OH),D recovery frcxnnewly diagnosed insulin-dependent diabetes.
during
Baran D T, Lee S.W., Jo O.D. and AVIOII L.V.: Acqurred alterations in vitamin D metabolism in the acidotic state. Calcif. Tissue Int. 34:165-168, 1982. Beck N.. Kim H P. and Kim KS.: Effect of metabolic acidosis on renal action of parathyrord hormone Am. J. Physiol. 228:1483-1488, 1975. Brochner-Mortensen J.: A simple method for the determination of glomerular filtration rate. Stand. J. C/in. Lab. Invest. 30:271-274, 1972. Christiansen C., Christensen MS., McNair P., Nielsen B. ant Madsbad S.: Vitamtn D metabolites in diabetic patients: decreased serum concentration of 24,25-dihydroxyvitamin D. Stand. J. C/in. Lab. Invest. 42:487-491,1982. Frazer TE., Whrte N.H , Hough S.. Santiago J V., McGee B.R., Bryce G., Mallon J. and Avroli L.V.: Alterations in circulating vitamin D metabolites in the young insulin-dependent diabetic. J. C/in. Endocrinol. Metab. 53:1154-1159, 1981 Gertner J.M., HorstRL. andTambor1aneW.V.: MineralmetabolismandvitaminD status in fuvenile diabetics (JOD): changes following normalization of plasma glucose (PG) with a portable infusion pump. Diabetes 2’8:354, 1979. Gertner J.M., Tamborlane W.V.. Horst R.L., Sherwin R.S., Felig P and Gene1 M.: Mineral metabolism in diabetes mellitus: changes accompanying treatment with a portable subcutaneous insulin Infusion system. J. C/m. Endocnnol. Metab. 50:862-866. 1980.
110
T. L. Storm et al.: Vitamin D metabolism in diabetes mellitus
Goldstein D.A., Haldimann B., Sherman D., Norman A.W. and Massry S.G.: Vitamin D metabolites and calcium metabolism in patients with nephrotic syndrome and normal renal function. J. C/in. Endocrinol. Metab. 52:116121, 1981. Gray R.W., Wilz DR., Ca1dasA.E. and Lemann J.: The importanceof in regulating plasma 1,25(0H),-vitamin
phosphate
D levels in humans: studies in
healthy subjects, in calcium stone formers and in patients with primary hyperparathyroidism. J. C/in. Endocfinol. Metab. 45:299-306, 1977. Heath H., Lambert P.W., Service F.J. and Arnaud S.B.: Calcium homeostasis in diabetes mellitus. J. C/in. Endomino/. Metab. 49:462-466, 1979. Henry H.L.: Insulin permits parathyroid hormone stimulation of 1,25dihydroxyvitamin D, production in cultured kidney cells. Endocrinology 108:733-735, 1981. Hughes M.R., Brumbaugh P.F and Haussler M.R.: Regulation of serum la,25-dihydroxyvitamin D, by calcium and phosphate in the rat. Science 190:578-580, 1975. Knochel J.P.: The pathophysiology and clinical characteristics of severe hypophosphatemia. Arch. Intern. Med. 137:203-220, 1977. LeeS.W., Russell J. and Avioli L.V.: 25.Hydroxycholecalciferol to 1,25-dihydroxycholecalciferol: conversion impaired by systemic metabolic acidosis. Science 195:994-996, 1977. Levin M.F., Boisseau V.C. and Avioli L.V.: Effects of diabetes mellitus on bone mass in juvenile and adult-onset diabetes. N. Engl. J. Med. 294:241-245, 1976. Lund Bj.. Lund Bi. and SQrensen O.H.: Measurement of circulating 1,25dihydroxyvitamin D in man. Changes in serum concentrations during treatment with Ia-hydroxycholecalciferol. Acta Endocrinol. 91:338-350, 1979. Lund B. and SQrensen O.H.: Measurement of 25.hydroxyvitamin D in serum and its relation to sunshine, age and vitamin D intake in the Danish population. Stand. J. C/in. Lab. Invest. 39:23-30, 1979. McNair P,, Christiansen C., Christensen M.S., Madsbad S., Faber O.K., Binder C. and Transbel I.: Development of bone mineral loss in insulin-treated
diabetes: a 1l/zyears follow-up study in sixty patients. Eur. J. C/in. Invest. 11:55-59, 1981. McNair
P., Madsbad
S., Christiansen
C., Christensen
M.S., Faber O.K..
Binder C. and Transbel I.: Bone loss in diabetes: effects of metabolic state. Diabetologia 17:283-286, 1979. Mcgensen C.E.: Glomerularfiltration rateand renal plasma flow in short-termand long-term juvenile diabetes mellitus. Stand. J. C/in. Lab. Invest. 28:91-100, 1971. Ringe J.D., Kuhlencordt F and Kuhnau J.: Mineralgehalt des Skeletts bei Langzeitdiabetikern. Dtsch. Med. Wochenschr. 101:280-282, 1976. Rosenbloom A.L., Lezotte D.C., Weber FT, Gudat J., Heller D.R., Weber M.L., Klein S. and Kennedy B.B.: Diminution of bone mass in childhood diabetes. Diabetes 26:1052-1055, 1977. Sauveur B., Garabedian M., Fellot C.. Mongin P. and Balsan S.: The effect of induced metabolic acidosis on vitamin D, metabolism in rachitic chicks. Cal& Tissue Res. 23:121-124. 1977. Schneider L.E., Omdahl J. and Schedl H.P.: Effects of vitamin D and its metabolites on calcium transport in the diabetic rat. Endocrinology 99:793-799, 1976. Schneider L.E., Schedl H.P., McCain T. and Haussler M.R.: Experimental diabetes reduces circulating 1,25-dihydroxyvitamin D in the rat. Science 196:1452-1454,1977. Shore R.M., Chesney R.W., Mazess R.B., Rose P.G. and Bargman G.J.: Osteopenia in juvenile diabetes. Calcif. Tissue Int. 33:455-457,198l. Spencer E.M., Khalil M. and Tobiassen 0.: Experimental diabetes in the rat causes an insulin-reversible decrease in renal 25.hydroxyvitamin D,-lo-hydroxylase activity. Endocrinology 107:300-305, 1980.
Received: May 10, 1983 Revised: September 6, 1983 Accepted: September 23, 1983
RCSUMB Une perte osaeuse slgniflcativea ete observie dans Is diabete suck. Sa pathogenisest inconnue mals pourrait &entuelkmant lmptiquer une anomaha du mitabolisme de la vitamine D. Pour cette raison, nous avons etudie b m&abolisme da la dtamine D dans 5 groupes de sujets attelnts da diabete Insulino-dependant. Destauxslrtquessignigcatlvement raduitsde25.OH Dn’ont It~observesquechezlessujetsatteintsdenbphropathiediab(tique. Le taux sertque du 1,25(0H)2 D n’a 118trot& abaisse que chez tes diabetiques en actdocetose, mais it s’est normali~ apres dispar’ttionda celte-cl. On peut en conclure que te m&abollsma da la vitamlne D est pour I’essentiel normal dans te diab&e de I’adulte fnsullno-dependant, et II sembte peu probable qu’un trouble du m&abolisms de la vhamine D pulsse exptiquer la perte osseuse chsz tes diabetiques lnsullnes habituets.
Bone Dis. & Rel. Res. 5, 107-l 10 (1983) Printed in the USA. All rights reserved.
Metab.
VitaminD Metabolism in Insulin-dependent Diabetes Mellitus T.L. STORM, O.H. S@RENSEN,Bj. LUND, Bi. LUND, J.S. CHRISTIANSEN, A.R. ANDERSEN, I.B. LUMHOLTZ, and H.-H. PARVING
Departments of Medicine and Rheumatology, Sundby Hospital, Department of Orthopedic Surgery, Rigshospitalef, Steno Memorial Hospital, Department of Medicine F; Herlev Hospital, Copenhagen, Denmark. Address for correspondence Copenhagen S, Denmark.
and reprints: 0. Helmer S$rensen, Department of Medicine, Sundby Hospital, ltaliensvej
1, DK-2300
by the demonstration of reduced activity of renal 25(OH)D1cr-hydroxylase (Spencer et al., 1980) and reduced levels of circulating 1,25(OH),D (Schneider et al., 1977), both of which were returned to normal by insulin treatment. Our study wasaimed at elucidating possible derangements of vitamin D metabolism in adult IDDM patients during ordinary, poor, and good metabolic control. The influence of diabetic nephropathy on vitamin D metabolism was also evaluated.
Abstract A significant loss of bone has been observed in diabetes mellitus. The patogenesis is unknown, but an impairment of vitamin D metabolism might be involved. Consequently, we have studied vitamin D metabolism in five groups of insulindependent diabetic patients. Significantly reduced levels of serum 25(OH)D were seen only in patients with diabetic nephropathy. The serum levels of 1,25(OH),D were reduced only in diabetic ketoacidosis but normalized during recovery. It is concluded that vitamin D metabolism is largely normal in adult insulin-dependent diabetes, and it seems unlikely that a disturbance of the vitamin D metabolism can explain the bone loss in the ordinarily controled diabetics.
Patients and Methods The vitamin D metabolites were studied in five groups of insulindependent diabetic patients (IDD):
Key Words: Insulin-dependent Diabetes-25(0H)D1,25(0H),D_Diabetic Nephropathy-Ketoacidosis.
I. 32 patients (age 15-45) wlthout proteinuria (Albustlx) and normal serum creatinine. II. 13 pattents (age 20-41) with dlabetlc nephropathy and normal serum creatinine were studied over a 2-year period. Ill. 14 patients (age 34-73) with diabetic nephropathy and elevated serum creatinine (> 150 ~molll). IV 9 newly diagnosed, untreated patients (age 21-35), blood glucose from 15-25 mmolll, ketonuria. but normal standard bicarbonate. The patients were either strictly regulated by five or six dally injections of soluble Insulin or by an artificial betacell (Biostator). The glomerular filtration rate (GFR) was measured on admission and after 7-10 days of regulation. V. 6 patients admitted with acute ketoacidosis and treated with lowdose insulin. None of the patients had persistent proteinuria, and serum creatinine was normal before this episode.
Introduction A significant loss of bone has been observed in both insulindependent (IDDM) (Levin et al., 1976; Ringe et al., 1976; Rosenbloom et al., 1977; McNair et al., 1979; McNair et al., 1981) and noninsulin-dependent diabetes (Levin et al., 1976; Shore et al., 1981). The pathogenesis of the bone loss is unknown, but an impairment of the vitamin D metabolism might be involved. Christiansen et al. (1982) found slightly decreased levels of serum 25(OH)D and markedly reduced serum levels of 24,25(OH),D. Reduced serum levelsof the biologically most active vitamin D metabolite, 1,25-dihydroxyvitamin D [1,25(OH),D], have been found in diabeticchildren (Gertner et al., 1979; Fraser et al., 1981), whereas normal levels have been reported in insulin-treated adult patients (Gertner et al., 1980; Heath et al., 1979). Studies in streptozotocin diabetic rats have shown reduced intestinal calcium absorption, which was restored by administration of 1 ,25(OH)*D3. The authors concluded that the defect in vitamin D metabolism was probably at the 1cr-hydroxylation step in the kidney (Schneider et al., 1976). This was supported
None of the patients in groups I-III had ketonuna. Diabetic nephropathy was diagnosed clinically if the following criteria were fulfilled: persistent proteinuria of more than 0.5 g/24 h, duration of IDDM of more than 10 years, presence of diabetic retinopathy, and no clinical or laboratory evidence of disease of the kidneys or the renal tract other than diabetic glomerulosclerosis. Two pattents did not fulfill all these cntena. A kidney biopsy was therefore performed, and in both cases slight or moderate diffuse diabetic glomerulosclerosis was found. Assays for circulating vitamin D metabolites were as follows: 25(OH)D was measured bv a comoetitive orotein-bindina assav. narmal mean 28 -c 15 ng/ml (SD) (Lund and SQrensen~~&$ The 24,25(0H),D and 1,25(OH),D metabolites were separated by gel 107
108
T L. Storm et al.: Vitamin D metabolism in diabetes mellitus
chromatography followed by high pressure liquid chromatography. They were measured by ligand binding assay using rachitic rat kidney cytosol and rachitic chick intestinal cytosol, respectively, as binding proteins. The normal values were 1.5 + 0.9 nglml (SD) for 24,25(OH),D and 33.1 f 15.3 pglml (SD) for 1,25(OH),D (Lund etal., 1979). GFR was measured after a single i.v. injection of “0-EDTA at 9 AM
by studying the plasma disappearance for 4 h (Br@chner-Mortensen, 1972). This procedure was applied to all patients in group II. The patients in group IV had their GFR measured using a constant infusion technique with ‘251-thalamate(Mogensen, 1971). Serum creatinine and blood glucose were measured on an auto-analyzer. Statistical analyses were performed using the Wilcoxon test for paired and
unpaireddataand Spearmanrankcorrelation.
Results Serum 25/OH)D
Normal values were seen in the patients without proteinuria (groups I and IV), whereas significantly (P < 0.01) reduced levels were found when diabetic nephropathy was present (groups Ii and Ill) (Fig. 1). The serum concentrations did not change during recovery from ketoacidosis (group V). Serum 24,25(OH),D
A trend toward reduced values was found in the groups studied (normal GFR without proteinuria 0.9 + 0.5 nglml, normal GFR with proteinuria 1.O f 0.7 nglml, age-matched controls 1.5 ? 0.9 nglml). No significant changes were seen during recovery from ketoacidosis. Serum 24,25(OH),D were not measured in group IV. Serum 1,25(OH),D
This was measured in groups II-V. As shown in Figure 2, the mean levels were normal in groups II-IV, although a few low values were found. No correlations could be found between the serum levels and GFR or serum creatinine. The GFR fell significantly (P < 0.01) during the 2 years of observation in group II, but no significant changes occurred in the serum 1,25(OH)2D concentrations (Fig. 3). Serum 1,25(OH)D2 was in the upper normal range on admission in group IV. Significant falls were Serum 25-OHD (nglml) 601 50
-1 t
30
8
:
-08 _ ::.
pco.01
seen in both the GFR (P < 0.01) and the serum 1,25(OH),D (P < 0.01) during strict regulation (Fig. 4). Reduced (P < 0.02) levels of circulating 1,25(OH),D were found on admission in acute ketoacidosis, with rapid normalization following correction of the acidosis (Figs. 1 and 5). Discussion Streptozotocin diabetes in the rat isassociated with decreased serum concentrations of 1,25(OH)2D (Schneider et al., 1977) as a consequence of reduced activity of the renal 1cr-hydroxylase enzyme system (Schneider et al., 1976) due to lack of insulin (Spencer et al., 1980). These observations could, however, not be confirmed in the present human study, since all the young, nonobese, untreated, newly diagnosed patients (group IV) had normal serum 1,25(OH),D levels or even concentrations in the upper normal range on admission. The supernormal values might partly be explained by a hemoconcentration as indicated by a fall in hematocrit from 0.43 f 0.02 to 0.40 + 0.01 (mean + SEM) following treatment. Furthermore, the serum levels of several hormones are elevated during acute stress, for example, growth hormone, which is known to stimulate the 1ar-hydroxylase. There was no correlation between renal function and serum 1,25(OH),D. Mean serum 1,25(OH),D concentrations were normal in the patients with normal GFR + proteinuria(group II) and in those with reduced GFR + proteinuria (group Ill), indicating that only minor disturbances occur in serum levels of 1,25(OH),D in ordinarily controled IDDM. The mean concentrations of 24,25(OH)2D were slightly but not significantly reduced in the two groups of patients studied. The mean 25(OH)D concentrations were reduced in patients with impaired kidney function. There was no difference in dietary intake between the groups that could explain the difference in serum 25(OH)D levels, but this might be due to renal loss of 25(OH)D and its binding protein (Goldstein et al., 1981). There was, however, no correlation between the degree of renal insufficiency and the reduction of circulating 25(OH)D. Severely reduced serum concentrations of 25(OH)D (lower than 5 nglml where osteomalacia might occur) were seen only in two young men with moderate and severe renal insuffiency, respectively.
Serum 1.25-(OHbD (Pglmt) 60 -
pco.02
P
.
40 : .
.
-_8+ W” 0
.
. : :
30 2o-
. k
‘O-
: .
: .
. ..
“I_ .
.
.
. .
.
. .
i .
?? :_ -
??
-.-
-
i Proteinuria Renal insuff. GROUP
-
+-
I
II
+
-
+ III
IV
Controls
Fig. 1. Serum levels of 25(OH)D in insulin-dependent diabetic patients with and without renal impairment compared to normal controls.
Proteinuria Renal insuff. GROUP
+ II
;r
+ +
-
III
IV
A
-&% J& Ji ‘5
. . . V
Controls
Fig. 2. Serum levels of 1,25(OH),D in insulin-dependent diabetic patients with and without renal impairment compared to normal controls.
109
T. L. Storm et al.: Vitamin D metabolism in diabetes mellitus Serum 12%(OH)2D (pglmt)
;;25PH)2D
GFR mllmin
t
t
1 120 100 80 60
4oit-__---*
I
2 years
Start
mean
Start
2 years
Fig, 3. Renal function (GFR) and serum levels of
1,25(OH),Dduring 2 years of observation in insulin-dependent patients with renal involvement.
A similar renal loss of the dihydroxylated metabolites would also be expected in patients with impaired kidney function. The normal levels of circulating 1,25(OH),D in patients with proteinuria might be due to increased production, since this hormone is strictly physiologically regulated, in contrast to 25(OH)D. Furthermore, the more polar metabolites are less completely bound and, therefore, less prone to vary with the circulating vitamin D-binding protein concentration. Ketoacidosis was accompanied by significant decrease (P < 0.02) in the circulating 1,25(OH)$ levels. This is in agreement with animal experiments that have shown that metabolic nonketotic acidosis impairs the conversion of 25(OH)D to 1,25(OH),D (Lee et al., 1977; Baran et al., 1982; Sauveur et al., 1977). Since the lcr-hydroxylation of 25(OH)D is regulated in part by parathyroid hormone (PTH) (Hughes et al., 1975) and since renal response to PTH is impaired during acidosis (Beck et al., 1975) one might expect the low levels of 1,25 (OH),D seen initially in ketoacidosis. In serum-free medium, kidney cells require insulin in order to respond to PTH with increased 1,25(OH),D production (Henry, 1981) but since the patients in group IV had normal serum levels of 1,25(OH)*D, it seems unlikely that lack of insulin can explain the low values seen in the acidotic patients. Serum phosphorus levels in untreated patients with diabetic ketoacidosis are either normal or elevated, but with the GFR mllmin
24
7.3s2.5 8.99.5
17.7t3.1 19.2t36
48
72 Hours??
24.1+-09 244~3.5
Fig. 5. Serum levels of 1,25(OH),D during recovery from diabetic
ketoacidosis. institution of therapy the levels decline rapidly as phosphorus shifts back into the cellular compartment (Knochel, 1977). Since low serum phosphorus concentration stimulates renal synthesis of 1,25(OH),D in humans (Gray et al., 1977) hypophosphatemia might in part be responsible for the transitory hypernormal serum levels of 1,25(OH),D seen during recovery from ketoacidosis. In the present study, vitamin D metabolism was largely normal in adult insulin-dependent diabetes, which is in agreement with the results of Heath et al. (1979) and Gertner et al. (1980). Moderately reduced serum 25(OH)D levels were, however, encountered when renal function was affected. Circulating levels of 1,25(OH),D were significantly reduced only in ketoacidosis and normalized when the acidosis was corrected. We conclude that it is unlikely that a disturbance of vitamin D metabolism can explain the bone loss described in ordinarily controled diabetics.
Acknowledgement: This study was supported by grants from The
Danish Association against Rheumatic Diseases, Landsforeningenfor Sukkersyge, and the Danish Medical Research Council. The skillful technical assistanceof Merete Frandsen and Ulla M. Smidt is gratefully acknowledged.
References
1.25-(OH)2D pglmt
.
60 210 50 190 40 170 30 150 70 130 10 110 Aam1ssion
H(‘ + &_~~
(2113-311) &l/l)
..
After
regulation
.
. Admtssron
After regulation
Fig. 4. Renalfunction (GFR) and serum levels of 1,25(OH),D recovery frcxnnewly diagnosed insulin-dependent diabetes.
during
Baran D T, Lee S.W., Jo O.D. and AVIOII L.V.: Acqurred alterations in vitamin D metabolism in the acidotic state. Calcif. Tissue Int. 34:165-168, 1982. Beck N.. Kim H P. and Kim KS.: Effect of metabolic acidosis on renal action of parathyrord hormone Am. J. Physiol. 228:1483-1488, 1975. Brochner-Mortensen J.: A simple method for the determination of glomerular filtration rate. Stand. J. C/in. Lab. Invest. 30:271-274, 1972. Christiansen C., Christensen MS., McNair P., Nielsen B. ant Madsbad S.: Vitamtn D metabolites in diabetic patients: decreased serum concentration of 24,25-dihydroxyvitamin D. Stand. J. C/in. Lab. Invest. 42:487-491,1982. Frazer TE., Whrte N.H , Hough S.. Santiago J V., McGee B.R., Bryce G., Mallon J. and Avroli L.V.: Alterations in circulating vitamin D metabolites in the young insulin-dependent diabetic. J. C/in. Endocrinol. Metab. 53:1154-1159, 1981 Gertner J.M., HorstRL. andTambor1aneW.V.: MineralmetabolismandvitaminD status in fuvenile diabetics (JOD): changes following normalization of plasma glucose (PG) with a portable infusion pump. Diabetes 2’8:354, 1979. Gertner J.M., Tamborlane W.V.. Horst R.L., Sherwin R.S., Felig P and Gene1 M.: Mineral metabolism in diabetes mellitus: changes accompanying treatment with a portable subcutaneous insulin Infusion system. J. C/m. Endocnnol. Metab. 50:862-866. 1980.
110
T. L. Storm et al.: Vitamin D metabolism in diabetes mellitus
Goldstein D.A., Haldimann B., Sherman D., Norman A.W. and Massry S.G.: Vitamin D metabolites and calcium metabolism in patients with nephrotic syndrome and normal renal function. J. C/in. Endocrinol. Metab. 52:116121, 1981. Gray R.W., Wilz DR., Ca1dasA.E. and Lemann J.: The importanceof in regulating plasma 1,25(0H),-vitamin
phosphate
D levels in humans: studies in
healthy subjects, in calcium stone formers and in patients with primary hyperparathyroidism. J. C/in. Endocfinol. Metab. 45:299-306, 1977. Heath H., Lambert P.W., Service F.J. and Arnaud S.B.: Calcium homeostasis in diabetes mellitus. J. C/in. Endomino/. Metab. 49:462-466, 1979. Henry H.L.: Insulin permits parathyroid hormone stimulation of 1,25dihydroxyvitamin D, production in cultured kidney cells. Endocrinology 108:733-735, 1981. Hughes M.R., Brumbaugh P.F and Haussler M.R.: Regulation of serum la,25-dihydroxyvitamin D, by calcium and phosphate in the rat. Science 190:578-580, 1975. Knochel J.P.: The pathophysiology and clinical characteristics of severe hypophosphatemia. Arch. Intern. Med. 137:203-220, 1977. LeeS.W., Russell J. and Avioli L.V.: 25.Hydroxycholecalciferol to 1,25-dihydroxycholecalciferol: conversion impaired by systemic metabolic acidosis. Science 195:994-996, 1977. Levin M.F., Boisseau V.C. and Avioli L.V.: Effects of diabetes mellitus on bone mass in juvenile and adult-onset diabetes. N. Engl. J. Med. 294:241-245, 1976. Lund Bj.. Lund Bi. and SQrensen O.H.: Measurement of circulating 1,25dihydroxyvitamin D in man. Changes in serum concentrations during treatment with Ia-hydroxycholecalciferol. Acta Endocrinol. 91:338-350, 1979. Lund B. and SQrensen O.H.: Measurement of 25.hydroxyvitamin D in serum and its relation to sunshine, age and vitamin D intake in the Danish population. Stand. J. C/in. Lab. Invest. 39:23-30, 1979. McNair P,, Christiansen C., Christensen M.S., Madsbad S., Faber O.K., Binder C. and Transbel I.: Development of bone mineral loss in insulin-treated
diabetes: a 1l/zyears follow-up study in sixty patients. Eur. J. C/in. Invest. 11:55-59, 1981. McNair
P., Madsbad
S., Christiansen
C., Christensen
M.S., Faber O.K..
Binder C. and Transbel I.: Bone loss in diabetes: effects of metabolic state. Diabetologia 17:283-286, 1979. Mcgensen C.E.: Glomerularfiltration rateand renal plasma flow in short-termand long-term juvenile diabetes mellitus. Stand. J. C/in. Lab. Invest. 28:91-100, 1971. Ringe J.D., Kuhlencordt F and Kuhnau J.: Mineralgehalt des Skeletts bei Langzeitdiabetikern. Dtsch. Med. Wochenschr. 101:280-282, 1976. Rosenbloom A.L., Lezotte D.C., Weber FT, Gudat J., Heller D.R., Weber M.L., Klein S. and Kennedy B.B.: Diminution of bone mass in childhood diabetes. Diabetes 26:1052-1055, 1977. Sauveur B., Garabedian M., Fellot C.. Mongin P. and Balsan S.: The effect of induced metabolic acidosis on vitamin D, metabolism in rachitic chicks. Cal& Tissue Res. 23:121-124. 1977. Schneider L.E., Omdahl J. and Schedl H.P.: Effects of vitamin D and its metabolites on calcium transport in the diabetic rat. Endocrinology 99:793-799, 1976. Schneider L.E., Schedl H.P., McCain T. and Haussler M.R.: Experimental diabetes reduces circulating 1,25-dihydroxyvitamin D in the rat. Science 196:1452-1454,1977. Shore R.M., Chesney R.W., Mazess R.B., Rose P.G. and Bargman G.J.: Osteopenia in juvenile diabetes. Calcif. Tissue Int. 33:455-457,198l. Spencer E.M., Khalil M. and Tobiassen 0.: Experimental diabetes in the rat causes an insulin-reversible decrease in renal 25.hydroxyvitamin D,-lo-hydroxylase activity. Endocrinology 107:300-305, 1980.
Received: May 10, 1983 Revised: September 6, 1983 Accepted: September 23, 1983
RCSUMB Une perte osaeuse slgniflcativea ete observie dans Is diabete suck. Sa pathogenisest inconnue mals pourrait &entuelkmant lmptiquer une anomaha du mitabolisme de la vitamine D. Pour cette raison, nous avons etudie b m&abolisme da la dtamine D dans 5 groupes de sujets attelnts da diabete Insulino-dependant. Destauxslrtquessignigcatlvement raduitsde25.OH Dn’ont It~observesquechezlessujetsatteintsdenbphropathiediab(tique. Le taux sertque du 1,25(0H)2 D n’a 118trot& abaisse que chez tes diabetiques en actdocetose, mais it s’est normali~ apres dispar’ttionda celte-cl. On peut en conclure que te m&abollsma da la vitamlne D est pour I’essentiel normal dans te diab&e de I’adulte fnsullno-dependant, et II sembte peu probable qu’un trouble du m&abolisms de la vhamine D pulsse exptiquer la perte osseuse chsz tes diabetiques lnsullnes habituets.