Effects of oral calcitriol on bone mineral density in patients with end-stage renal failure

Kidney International, Vol. 45 (1994), PP. 245—252

Effects of oral calcitriol on bone mineral density in patients with end-stage renal failure PATRICK RUEDIN, RENE RIzzoLI, DANIEL SLOSMAN, MICHEL LESKI, and JEAN-PHILIPPE BONJOUR Divisions of Nephrology, of Clinical Pathophysiology, World Health Organization Collaborating Center for Osteoporosis and Bone Disease, and of Nuclear Medicine, University Hospital, Geneva, Switzerland

Effects of oral calcitriol on bone mineral density in patients with end-stage renal failure. Evolution of bone mineral density (BMD) at various skeletal sites and the influence of calcitriol on BMD are still poorly documented in patients with terminal renal failure. Using dual photon absorptiometry, we investigated the changes in BMD at the

ous histological forms of bone disease have been demonstrated depending on dietary intakes, aluminum exposure, or dialysis-

related factors, osteitis fibrosa represents the most frequent form, the severity of which appears to be proportional to the

degree of secondary hyperparathyroidism [11—15]. The treatment of this form of renal osteodystrophy involves the control 0.02 sg/day) and compared them to 25 patients with of excess PTH. Over the last two decades, most studies (very SEM: 0.21 ESRF but not treated with calcitriol (control group) over a period of 20.3 1.5 and 17,2 1.2 months, respectively. Lumbar spine BMD few were double-blind placebo-controlled) have provided eviincreased by 7.7 3.2%/year in the treated group and decreased by 2.5 dence for the beneficial effects of calcitniol on biochemical, 1.3%/year in the control group (P < 0.005). Femoral shaft BMD radiological and histological manifestations of hyperparathyincreased more in treated than in control group (+ 6.7 2.3 vs. + 1.4 roidism [16—22]. Indeed, calcitniol directly inhibits PTH secre2.0%/year; P < 0.05) and femoral neck BMD remained stable. PTH tion [23], causing declines in PTH levels of greater magnitude levels increased by 92 121 and 1033 254 pmollyear (P < 0.01) in the treated group and the controls, respectively. Osteocalcin changes were than those achieved with similar increases of calcemia. Furtherlevels of lumbar spine, femoral neck and midfemoral shaft in 21 patients with end-stage renal failure (ESRF) treated with calcitriol (mean dosage

—2.7

3.7 and +20.1

11.7 pg/liter (P < 0.05) per year in the same

more, adequate suppression of PTH in uremia may require

groups. These results indicate that low doses of oral calcitriol in high, potentially toxic plasma calcium levels. patients with end-stage renal failure were associated with an increase in

BMD at the levels of lumbar spine and femoral shaft, and with a stabilization of serum PTH and osteocalcin concentrations.

Thus, if calcitriol therapy is largely recognized as exerting favorable influence on osteodystrophy encountered in the frame of end-stage renal failure, the effects of its chronic administra-

tion on bone mineral density of various skeletal sites particularly prone to osteoporotic fractures, such as lumbar spine and femoral neck, have not yet been clearly documented. In the treatment of postmenopausal or senile osteoporosis, the efficacy of calcitriol is still largely debated [24-27]. In the present study, we prospectively monitored the evolution of bone mineral density at the levels of lumbar spine, femoral neck and midfemoral shaft in patients with end-stage renal failure. The patients were divided into two groups, with or without low doses of oral calcitriol. The results indicated that oral calcitniol was associated with significant increases in lumbar spine and midfemoral shaft bone mineral density and with a stabilization of serum PTH and osteocalcin concentrations.

Bone mass is a major determinant of the skeletal resistance to fracture [1—3]. Several studies have evaluated bone mineral density in patients under long-term dialysis. Either appendicutar bone, mostly composed of cortical bone, or axial skeleton, comprising a substantial proportion of trabecular bone, were monitored with contrasting results [4—10]. This probably reflects various underlying pathological processes, different lengths or types of dialysis programs, variable compliance with dietary or pharmacological interventions, or preferential sensitivity of cortical or trabecular bone to excess of PTH. In a recent cross sectional study performed in 106 patients attending the same dialysis center, we demonstrated, using dual photon absorptiometry, a particularly marked decrease in bone mineral Methods density at the levels of lumbar spine, femoral neck and of a Subjects cortical weight-bearing bone, the midfemoral shaft [10]. Osteodystrophy is a prominent feature in patients with terBone mineral density (BMD) was prospectively assessed in minal renal failure undergoing dialysis therapy. Although van- patients with end-stage renal failure attending the University Hospital of Geneva dialysis center. The date of inclusion in the study was the date of the first BMD measurement. To evaluate the effects of oral calcitniol therapy on BMD and parathyroid Received for publication February 24, 1993 and in revised form August 2, 1993 status, subjects were divided into two groups: one (N = 21) Accepted for publication August 2, 1993

© 1994 by the International Society of Nephrology

which was receiving oral calcitniol (Rocaltrol®) at doses ranging from 0.25 g thrice a week to 0.5 jg daily at the beginning of the

245

246

Ruedin et a!: Calcitriol and bone mineral density in CRF

study. The goal of this treatment, which lasted 3 to 24 months in 16 patients, was to achieve a plasma calcium level averaging 2.30 to 2.50 mmollliter. In the control group (N = 25), prevention of osteodystrophy was limited to oral phosphate binder therapy. All patients had a second BMD measurement performed 20.3 1.5 and 17.2 1.2 months later for the calcitrioltreated and the control groups, respectively. Phosphate binder therapy consisted of either aluminum hydroxide or calcium carbonate, or a combination of both agents; the goal was to maintain plasma phosphate level lower than 2.0 mmollliter. All patients were advised to take the same phosphate-restncted diet. In controls, 17 patients were never given calcitriol; in 8 patients, calcitriol had been discontinued more than six months before the beginning of the study. The cessa-

Table 1. Patient characteristics Control group

N=25 2.9

Group with calcitriol therapy

N=21

3.6 Age years (13—83) (22—74) (Range) 14/11 13/8 Sex male/female 100 100 Menopause % Underlying renal diseases: 8 10 Chronic glomerulonephritis 6 7 Nephrosclerosis 3 S Interstitial nephritis 2 2 Polycystic kidney disease 0 2 Diabetic nephropathy 1 0 Reflux nephropathy tion of calcitriol therapy was justified only in two cases by Dialysis duration before 32.1 8.2 48.9 10.2 the first BMD moderate hypercalcemia. We excluded one patient with multiassessment months ple myeloma in the treated group, and one of the controls who (Range) (0—141) (0—136) underwent parathyroidectomy during the course of the study. Results are given as absolute numbers or as means SEM. There One patient in the control group had a subtotal parathyroidectomy four years before the beginning of the study. Four patients were no significant differences between groups in any parameters using the Mann Whitney test or chi-square test when appropriate. 56.8

56.2

in each group had a renal transplant which was functional for less than three months, at least two years before inclusion in the study. Dialysis therapy was hemodialysis for three to four hours thrice weekly in 27 patients, continuous ambulatory pentoneal dialysis in 13 patients and both methods of dialysis successively '"Gd radioactive source was not older than nine months. For in four patients. Only two patients were still in pre-dialysis the patients on continuous ambulatory peritoneal dialysis, dialysate was drained out prior to lumbar spine BMD measurestage during the course of the study. ments. X-ray was regularly performed to rule out compression fractures during the follow-up period. Because the program to Methods Determinations of plasma calcium, phosphate and alkaline analyze BMD at the level of femoral neck was not available at the beginning of the study, this measurement was performed in phosphatase were performed monthly by atomic absorption spectrophotometry and colorimetric methods, respectively. 25 patients only (15 patients in control group and 10 patients in Values represent the means of five monthly determinations group with calcitriol therapy). The results were expressed in done around the time of each BMD measurement. Plasma absolute values, in standard deviations from the mean of agemagnesium and aluminum were only assessed at the time of and sex-matched normal subjects (Z-score), or in deviation in BMD measurements. Determination of aluminum was per- percent from peak bone mass of young normal sex-matched formed by flameless atomic absorption after appropriate dilu- adults (% PBM) [30]. tion; magnesium was measured by atomic absorption spectroStatistical analysis photometry. For plasma aluminum determination, results lower Values are means SEM. Mann-Whitney's test was used to than 1.2 moI/liter were given a value of 1.2 pmolJliter. To detect aluminum overload, a deferoxamine test was done yearly compare intergroup differences. Paired Student's t-test was by infusing 40 mg/kg body wt at the end of a dialysis session used to compare values in the same group between the first and [28]. An increase in plasma aluminum higher than 5.6 rmol/liter second BMD measurements. The chi-square test was used at 24 hours or 48 hours after deferoxamine administration was where appropriate. Correlation was performed by Spearman considered as positive. In that case a course of deferoxamine method. A P value (two tailed) < 0.05 was interpreted as therapy was started (40 mg/kg body wtlweek for 6 months). significant. Serum PTH level was measured by radioimmunoassay using Results antibodies directed against a mid molecule epitope (amino Patient characteristics acids: 44-68; Endocrine Metabolic Center; interassay coeffiThe two groups were similar with respect to age, sex districient of variation: 5.8 to 9%). During the course of the study, the PTH assay changed and only 28 values were available at the bution, menopausal status and prevalence of underlying renal second BMD measurement to be compared with the first mid diseases (Table 1). Mean dialysis duration appeared to be longer molecule determinations. Serum osteocalcin level was mea- in the calcitriol-treated group but this difference was not statissured by radioimmunoassay (Cis Oris, Gif sur Yvette, France; tically significant. The dialysis parameters and the medical treatment during the study are presented in Table 2. The mean interassay coefficient of variation: 6.6%). Dual photon absorptiometry was performed using a Norland follow-up periods were similar in the two groups as were the

types of dialysis therapy and the mean dialysate calcium oral shaft and femoral neck. Coefficients of variation were concentrations. The mean daily dose of oral calcitriol was 0.21 0.02 p.g for the 21 treated patients. Mild to moderate 2.1%, 1.4% and 2.0% for lumbar spine, femoral shaft and 2600 instrument at the levels of lumbar spine (L2—L4), midfem-

femoral neck BMD measurements, respectively [29]. The I Ci

hypercalcemia (2.70 to 2.90 mmol/liter) occurred in four cases

Ruedin et a!: Calcitriol and bone mineral density in CRF

Table 2. Dialysis parameters and medical treatment Control group

N=25 Mean follow-up period months Type of dialysis therapy during the study: HD/CAPD/mixed therapy! no dialysis Dialysate calcium concentration for HD and CAPD patients mmo!/liter Number of patients with dialysate calcium concentration of 1.75 mmol/liter 1.50 mmol!liter 1.25 mmol!liter Daily dose of oral calcitriol .g/day Type of phosphate binders therapy

17.2 1.2 13/9/1/2

0.03

1.70

Group with calcitriol therapy

N=21

20.3

1.5 14/4/3/0

0.02

1.71

A1(OH)3 g!day

cant at the end of the follow-up. During the study, PTH levels nearly doubled in the control group whereas the mean value remained stable in the group treated with oral calcitriol. The results for osteocalcin level displayed the same pattern, but the increase of mean osteocalcin level in the control group did not reach statistical significance. However, as depicted in Figure 1, when expressed as percent change of initial value per year, the calcitnol-treated group displayed increases in osteocalcin and PTH levels of lower amplitudes than those of the control group (osteocalcin: +2.1 8.2%/year vs. +43.9 22.4%/year; P < 0.05. PTH: +21.5 15.3%/year vs. + 129.7 51.2%/year; P < 0.05).

20

18

1

3 0

2 0 6/2/8/4

0.21 0.02 5/6/5/1

(CaCO3!Al(OH)3/mixed/no

treatment) Daily dose of phosphate binders CaCO3 g!day

247

2.3

0.7

0.4 0.2

2.7

0.6

1.7

0.4k

Number of patients treated with 5 3 deferoxamine during the study Results are given as absolute numbers or as means SEM. Abbreviations are: HD, hemodialysis; CAPD, continuous ambulatory peritoneal dialysis. Differences between groups were evaluated using the Mann-Whitney test or chi-square test when appropriate. a P < 0.05 as compared with value in the control group

Evolution of bone mineral density The initial BMD values of all patients, expressed in Z-score, are presented on Table 4. Z-scores were significantly decreased at the three skeletal sites (P < 0.01). They appeared to be more severely depressed at the levels of femoral shaft (—1.19 0.20) and femoral neck (—1.13 0.16) than that of lumbar spine (—0.66 0.23), but differences were not statistically significant. Except for femoral neck, initial values of BMD were significantly lower in the group treated with calcitriol. Figures 2, 3 and

4 depict the evolution of individual BMD values of lumbar spine, femoral shaft and femoral neck. At the level of lumbar

spine, we observed a significant increase in BMD in the group treated with calcitriol (from 0.842 0.041 g/cm2 to 0.945 0.055 g/cm2; P < 0.01), whereas the mean value of BMD tended to decrease (from 0.993 0.033 g/cm2 to 0.956 0.037 g/cm2; P < 0.05) in the control group. At the midfemoral shaft level, a and necessitated the discontinuation of this therapy; neverthe- similar pattern with a significant increase of the mean BMD in less, duration of calcitnol therapy in these four patients was 61 the treated group (from 1.423 0.086 g/cm2 to 1.571 0.114 11% of the mean follow-up period. The mean dose of g/cm2; P < 0.01) and a stabilization in the control group (from aluminum hydroxide was lower in the control group as com- 1.696 0.053 g/cm2 to 1.737 0.064 g/cm2) was observed. The mean change in femoral neck BMD of neither the calcitriolpared with the group on calcitriol therapy (0.7 0.2 vs. 1.7 0.2 gfday). The number of patients treated with deferoxamine treated group (from 0.685 0.046 g/cm2 to 0.72 1 0.041 glcm2) during the follow-up period were similar in the two groups (3 nor the control group (from 0.742 0.035 g/cm2 to 0.726 and 5 patients in controls and calcitriol-treated group, respec- 0.027 g/cm2) reached statistical significance. For this skeletal tively).

site only ten paired values could be analyzed in the treated

group. Figure 5 shows the variations in BMD given as percent Evolution of biochemical variables changes per year from baseline values. Calcitriol therapy was There was no significant difference in the plasma levels of associated with an increase in BMD at the levels of lumbar calcium, phosphate and magnesium measured at the beginning spine (P < 0.005) and femoral shaft (P < 0.05). At the level of

and the end of the study, when all patients were analyzed

femoral neck, the increment was less pronounced and not

together (Table 3). An overall increase in alkaline phosphatase was observed, but this was not significant when subjects who developed transfusion-acquired hepatitis were excluded from analysis (N = 8). PTH level was significantly higher at the end of the study. The higher mean serum osteocalcin level at the

significantly different from non-treated patients (+ 1.3 2.3%/ year in the group treated with calcitriol and —0.1 0.9%/year in the control group). The same pattern of results was observed when postmenopausal women were separately evaluated (data not shown). second BMD measurement did not reach statistical signifiSince mean initial BMDs were lower in the treated group, we cance. The small decrease in aluminum level may be due to the intended to compare subjects with identical initial BMDs and

preferential use of calcium carbonate rather than aluminum

selected 12 pairs with matched initial lumbar spine values in

hydroxide adopted during the course of the study and to controls and treated patients (0.954 0.049 g/cm2 vs. 0.950 treatments of aluminum-overloaded patients with deferox- 0.048 g/cm2). Over the study period, a gain in the calcitriolamine. At the end of the study, only one patient had an treated group (+8.0 2.5%/year) and a decrease in the control aluminum increment after deferoxamine loading greater than 5.6 molJliter. Both groups had a mean basal plasma aluminum of 1.3 0.1 molJliter. When comparing the control group with the group treated with oral calcitriol, calcemia was lower at the start of study in the latter, but the difference was no longer statistically signifi-

1.7%/year; P < 0.001) were observed. In these pairs with matched initial lumbar spine BMD, PTH concentrations increased from 1656 547 to 2521 970 in controls (P < 0.025), but remained stable in the calcitriol-treated group from 1402 275 to 1474 288 pmol/liter (NS). When pairs (N = 14) with similar femoral shaft BMD were considered (1.611 0.061 group (—3.9

248

Ruedln et al: Calcitriol and bone mineral density in CRF

Table 3. Biochemical data

All patients

2.39

Ti

12 0.02

N= 21

N=25

Ti Plasma calcium mmollliter

Group with calcitriol therapy

Control group

N=46

Ti

12

2.38

0.02

2.44

0.02

1.9

0.1

1.8

0.1

1.8

0.04

1.00

0.03

1.02

0.06

1.04

9

154

181)

112 8

135

112±7

119

108±8

120±10

2.41

0.03

2.34

T2

0.02"

2.35

0.04

(2.25—2.60)

Plasma phosphate mmollliter

1.8

0.1

1.9

0.1

0.04

0.96

0.05

12b

138

0.1

2.0

0.1

(0.8—1.3)

Plasma magnesium mmol/liter

0.99

(0.65—0.82)

Alkaline phosphatase U/liter

124

(30—125)

Alkalinephosphatase U/liter

16

117±14

0.95

0.04 36

177

118±9

(30—125), without hepatitis

N

38

Serum osteocalcin smol/liter (3.0-13.5)

Serum PTH pmol/liter (60—135)

N

38

4.7

58.7

1653 256

2520

46.3

0.1

1.6

Number of patients with plasma aluminum> 5.6 pinol/liter

44.2

425°

1859

28

43

Plasma alummum i.unol/lirer

22

10.7

18,3

48.8

444

3377 713°

1417

4

1.3

0

16

7.8

42.6

208

1531

20

15

0.2

1.5

1

16

72.2

23

0.1°

1.3

9

22

5.8

0.1

5

199° 13

0.2

1.8

8.1

0.la

1.3 1

Results are means SEM. Abbreviations are: TI, values at the time of first bone mineral density measurement; T2, values at the time of the second bone mineral density measurement. For plasma calcium, phosphate and alkaline phosphatase, the mean was calculated with individual results, which themselves represent the mean of 5 monthly determinations (around the time of each BMD measurement). The normal ranges are in parentheses. a p < 0.05, b p < 0.025, P < 0.01 T2 vs. Tl using paired t-test dP < 0.01 vs. control group for Ti using the Mann-Whitney test e P = 0.06 vs. control group for T2 using the Mann-Whitney test

200

g/cm2 vs. 1.636 0.061 g/cm2), the evolution was similar to that of the whole group with an increase in the treated group (+9.1 2.9%/year) of higher amplitude than that of the controls (+0.4 2.6%/year; P < 0.025). These pairs did not differ in age, sex prevalence, type or duration of dialysis.

PTH

Osteocalcin

Over the whole study period only one fracture supervened (femoral neck) in the control group; in this case, femoral neck BMD Z-scores were —0.30 and —0.68 at the first and the second

measurements, respectively. Discussion

100

In this controlled longitudinal study, low doses of calcitriol appeared to exert a favorable influence on lumbar spine and midfemoral shaft bone mineral density. This was associated with a stabilization of serum PTH and osteocalcin concentra-

E 4) C,)

tions.

(P.cO.05)

Calcitrol

N=25

N=21

N=15

N=12

—

+

—

+

Fig. 1. Serum osteocalcin and PTH levels expressed in percent change of initial values per year. Values are means SEM. Differences are evaluated with a Mann-Whitney test. Symbols are: () control group;

() group with calcitriol therapy.

The pathogenesis of renal osteodystrophy is related, at least in part, to elevated PTH levels in association with decreased alcitriol production [11—15]. The relationship between hyperparathyroidism and bone mass has been reviewed recently, and there is some evidence for a preferential decrease in cortical bone induced by an excess of PTH [31, 32]. In the present study, we confirmed the low bone mineral density at the levels of midfemoral shaft and femoral neck in patients on dialysis. However, we cannot be certain that elevated PTH concentration was a main contributor to the decreased bone mass. The high prevalence of menopausal females should be emphasized, since all were amenorrheic in our cohort. Low bone mass could also be the reflection of impaired mineralization. Among the possible causes, aluminum could be invoked. However, the

249

Ruedin et a!: Calcitriol and bone mineral density in CRF Table 4. Bone mineral density values

N=46

TI 0.924

Z-score PBM % Femoral shaft

—0.66 —16.4

TI

T2

BMD g/cm2

1.572

Z-score PBM % Femoral neck

—1.19 —18.8

BMD g/cm2

0.719

N

0.028 0.23 2.4

—0.39

0.052 0.20 2.2

—0.93

0.028

Z-score

0.16

—1.13

N

—10.7 2.9

0.063C

0.053 0.21

1.696

0.956

1.737

—0.75

—16.6 2.9

—14.2 2.5

—0.54 —14.0

0.724

0.025

0.742

0.726

0.24

—0.90

—25.6 3.0

44

25

0.055C

0.46''

1.571

0.Il4C 0.46 4.9

—15.2 49

—24.4 35

0.027

0.685 0.046

0.721

—1.47 o.Ise 10

0.25

—1.19

19

10

—29.7 2.9

0.041 19

10

2.5

—27.4

0.086e

1.423

0.945 —0.42

—1.39 —15.2

25

15

0.842 O.O4l

—1.28 0.34e —23.1 34C

—1.72 O.32

0.18

—1.08

15

T2

0.064 0.22 3.3 25

15

—1.12 0.14

25

0.035

0.037

—0.38 0.33 —14.0 33a

0.24

—28.0 2.2

2.2

—27.2

N

1.662

TI

T2 0.033 0.30

—0.15

—14.6 2.8

44

25

PBM %

0.993

0.032 0.27

0.951

44

25

N=2l5

N—255

Lumbar spine BMD glcm2

Group with calcitriol therapy

Control group

All patients

3.8

—28.9

19

Results are given as means SEM. Abbreviations are: Ti, values at the time of the first bone mineral density measurement; T2, values at the time of the second bone mineral density measurement. Z-score represents the standard deviation from the mean of age- and sex-matched normal subjects. PBM % represents the deviation from the peak bone mass of young normal sex-matched adults expressed in percent. a p < 0.05, b p < 0.025, C p < 0.01 T2 vs. Ti using paired t-test e p < 0.05, P = 0.025 group with calcitriol therapy vs. control group for Ti using the Mann-Whitney test For lumbar spine and femoral shaft BMD measurements.

1.6

B___________

A

1.4 ____

1.2

::::...:::

1.0

0.8 0.6 E

0.4 (NS) 0.2 _______________________________

0 6 12 18 24 30 36

Duration of follow-up, months '9

2.5k

_

1.5

00

Duration of follow-up, months

''

_________

1.0 0.5

0 6 12 18 24 30 36

Fig. 2. Individual values of lumbar spine bone mineral density in the control group (A) and in the group with calcitriol therapy (B). Differences are evaluated using a two-tailed paired Student's-test. Symbol is: (—) mean SCM. bone mineral density

B___________

30A 2.0

(P<0.01) ________________________________

(NS)

0 6 12 18 24 30 36 Duration of follow-up, months

(P<0.01)

0 6 12 18 24 30 36 Duration of follow-up, months

Fig. 3. Individual values of midfemoral shaft bone mineral density in the control group (A) and in the group with calcitriol therapy (B). Differences are evaluated using a two-tailed paired Student's r-test. Symbol is: (—) mean bone mineral density SCM.

influence of aluminum overload would be negligible, since the increases intestinal absorption of aluminum, this did not seem use of phosphate chelators containing aluminum was progres- to have significant consequences in our patients [33]. The administration of calcitriol has been shown to reduce sively replaced during the study period and since the prevalence of a positive deferoxamine test was low, at least similar in both PTH concentrations and to lead to an improvement of bone groups of patients. Although it has been reported that calcitriol histological signs of secondary hyperparathyroidism in the

Ruedin et a!: Calcitrial and bone mineral density in CRF

250 1.2

A

S

.-....

1.0

08

I

0.2

(NS)

(NS)

061218243036

061218243036

Duration of follow-up, months

Duration of follow-up, months

FIg. 4. Individual values of femoral neck bone mineral density in control group (A) and in the group with calcitriol therapy (B). Differences are evaluated using a two-tailed paired Student's paired t-test. Symbol is: (—) mean bone mineral density SEM.

14

12 10

8 6 Cl)

C

a, V

4

Ce

a,

C E

0

2 0 —2

-4 —6

N=25

N=21

N=25

N=21

N=15

N=10

—

+

—

+

—

+

Fig. 5. Changes in bone mineral density, at the levels of lumbar spine, femora! shaft and femoral neck, expressed in percent change of initial values per year. Values are means SEM. Differences are evaluated with a Mann-

Calcitrol

frame of chronic renal failure [15—22]. Calcitriol decreases the proliferation rate of parathyroid epithelial cells, lowers the synthesis and secretion of PTH, through the restoration of the sensitivity of parathyroid cells to extracellular calcium concentrations. It was thus of interest to assess the influence of calcitriol on BMD of skeletal sites particularly prone to osteoporotic fractures, such as lumbar spine and femoral neck. Our

Whitney test. Symbols are: () control group; (I) group with calcitriol therapy.

Calcitriol therapy had been introduced before the first BMD measurement, and was maintained throughout the study period. This treatment was associated with increases in lumbar spine and midfemoral shaft BMDs. The relatively small number of measurements available and the variability of the BMD measurements at the femoral neck level possibly prevented significant effects to be detected there.

Clinical or biochemical signs of impaired mineralization could study was a longitudinal assessment of the evolution of BMD at these skeletal sites. The patients were not randomly allocated to have constituted arguments for the prescription of calcitriol. calcitriol therapy or control groups, but the grouping was Indeed, not only initial plasma calcium concentrations, but also performed retrospectively. Whereas phosphate chelators have lumbar spine and midfemoral shaft BMD values were slightly been routinely used, there was no uniform policy concerning lower in the calcitriol-treated group which had started treatment calcitriol therapy in our dialysis center. This could explain why well before BMD measurements. In contrast, alkaline phosthe two groups, which were determined retrospectively, did not phatase, osteocalcin, PTH and aluminum levels were not difmarkedly differ. Furthermore, they were studied over the same ferent between the two groups. Thus, to evaluate the effects of period of time ruling out temporal drift in BMD measurements. calcitriol in patients with similar initial BMDs, we selected pairs

Ruedin et a!: Calcitriol and bone mineral density in CRF

251

in the calcitriol-treated and the control groups with matched BMDs. This analysis indicates that the effects of calcitriol on lumbar spine or midfemoral BMDs in patients with identical

5. LINDERGARD B: Changes in bone mineral content evaluated by photon absorptiometry before the start of active uremia treatment.

initial BMDs were quite comparable to those found in the whole

patient on prolonged maintenance hemodialysis: A three year

ClinNephrol 16:1216—1230, 1981 6. RICKERS H, CHRISTENSEN M, RODBRO F: Bone mineral content in

cohort. This suggests that the positive influence of calcitriol follow-up study. Clin Nephrol 20:302—307, 1983 could not merely result from the correction of a subclinical 7. PIin.o B, CHEN T, COOPERSTEIN L, SEGRE G, PUSCHETT J: Fractures and vertebral bone mineral density in patients with renal osteomalacic component. osteodystrophy. Clin Nephrol 30:57-62, 1988 The increases of BMD induced by calcitriol in patients with 8. EECKHOUT E, VERBEELEN D, SENNESAEL J, KAUFMAN L, J0NcKend-stage renal failure contrasts with the effects of this drug in HEER MH: Monitoring of bone mineral content in patients on postmenopausal osteoporosis, where they are much less eviregular hemodialysis. Nephron 52:158—161, 1989 dent [24—27]. Although we did not measure calcitriol concen- 9. ASAKA M, IIDA H, ENTANI C, FUJITA M, IZUMINO K, TAKATA M, SETO H, SASAYAMA S: Total and regional bone mineral density by trations in the present study, they are classically very low in dual photon absorptiometry in patients on maintenance hemodialterminal renal failure, whereas they are mostly within the ysis. Clin Nephrol 38:149—153, 1992 normal range, though they might be slightly lower, in post- 10. GABAY C, RUEDIN P, SLOSMAN D, BONJOUR JP, LE5KI M, RIzzoLl R: Bone mineral density in patients with end-stage renal failure. Am menopausal osteoporosis [34]. Furthermore, elevated PTH is JNephrol 13:115—123, 1993 not a uniform feature of this latter condition. Thus, one could MALLUCHE HH, FAUGERE MC: Renal osteodystrophy. N EngI I expect to see more dramatic effects of calcitriol therapy in 11. Med 321:317—318, 1989 disease states characterized by a decreased level of this latter 12. MALLUCHE HH, FAUGERE MC: Renal bone disease 1990: An hormone and elevated PTH. unmet challenge for the nephrologist. Kidney mt 38:193—211, 1990 The low doses of calcitriol administered were remarkably 13. DELMEZ JA, SLATOPOLSKY E: Recent advances in the pathogenesis and therapy of uremic secondary hyperparathyroidism. J Clin well tolerated. Indeed, in only four cases was the treatment discontinued because of moderate hypercalcemia. Similarly, we did not detect any increase in plasma phosphate concentration,

nor were we obliged to increase phosphate binder therapy, despite the fact that calcitriol is known to stimulate intestinal absorption of phosphate [35].

As suggested for the prevention and treatment of osteitis

Endocrinol Metab 72:735—739, 1991

14. FEINFELD DA, SHERWOOD LM: Parathyroid hormone and 1 ,25(OH)2D3 in chronic renal failure. Kidney mt 33:1049—1058, 1988 15. RITz E, MATTHIAS S, SEIDEL A, REICHEL H, SZABO A, Hotu.. WH:

Disturbed calcium metabolism in renal failure—Pathogenesis and therapeutic strategies. Kidney mt 42(Suppl 38):S37—S42, 1992 16. QUARLES LD, DAVIDAI GA, SCHWAB SJ, BARTHOLOMAY DW,

LOBAUGH B: Oral calcitriol and calcium: Efficient therapy for

fibrosa, calcitriol could also be envisaged in the therapy of renal uremic hyperparathyroidism. Kidney mt 34:840—844, 1988 failure-associated osteoporosis. This possibility should be sup- 17. DELMEZ JA, TINDIRA C, GROOMS P, Dusso A, WINDU5 DW,

ported by results obtained in randomized double-blind longterm studies, which should assess whether positive influences on bone mass will be associated with commensurate decrease in

fracture rates.

SLATOPOLSKY E: Parathyroid hormone suppression by intravenous

1 ,25-dihydroxyvitamin D. A role for increased sensitivity to calcium. J Clin Invest 83:1349—1355, 1989 18. BAKER LRI, ABRAMS SML, ROE CJ, FAUGERE MC, FANTI P, SUBAYTI Y, MALLUCHE HH: 1,25(OH)2D3 administration in mod-

erate renal failure—A prospective double-blind trial. Kidney mt

Acknowledgments Parts of this work were presented at the Annual Meeting of the

35:662—669, 1989 19. ANDRESS DL, NoltRis KC, CoBuIu.i JW, SLATOPOLSKY EA, SHER-

RARD DJ: Intravenous calcitriol in the treatment of refractory

American Society of Nephrology in Washington, D.C., December 1990,

osteitis fibrosa of chronic renal failure. N Engi I Med 321:274—279,

grants from the Swiss National Science Foundation (3200-25.535.88) and from Cilag AG, Switzerland. Dr. R. Rizzoli is a recipient of a Max Cloetta career development award. We are grateful to the nursing staff of the Dialysis Center of Geneva for assistance, and particularly Mrs.

H, WE5TEEL PF, COHEN SOLAL M, FOURNIER A: Prevention of osteitis fibrosa, aluminum bone disease and soft-tissue calcification

and were published in abstract form in the Journal of the American 1989 Society of Nephrology 1: 573, 1990. This project was supported by 20. MORINIERE P, BOUDAILLIEZ B, HOCINE C, BELBRIK S, RENAUD

M.O. Levy. T. Whitehouse, B.Sc., is acknowledged for reading the manuscript, and Mrs. M. Buttex and M.-C. Brandt for typing it.

Reprint requests to Dr René Rizzoli, M.D., Division of Clinical Pathophysiology, Department of Medicine, University Hospital, 1211 Geneva 14, Switzerland.

References

in dialysis patients: A long-term comparison of moderate doses of oral calcium Mg(OH)2 vs. Al(OH)3 laOH Vitamin D3. Nephrol Dial Transplant 4:1045—1053, 1989

21. COBURN 1W: Use of oral and parenteral calcitriol in the treatment of renal osteodystrophy. Kidney mt 38(Suppl 29):S54—S61, 1990 22. MALLUCHE HH, FAUGERE MC: Effects of 1,25(OH)2D3 administration on bone in patients with renal failure. Kidney Int 38(Suppl 29):548—S53, 1990

23. SILVER J, NAVEH-MANY T, MAYER H, SCHMELZER HJ, PopOVTZER MM: Regulation by vitamin D metabolites of parathyroid

hormone gene transcription in vivo in the rat. J Clin Invest 78:1296—1301, 1986

1. Hui SL, SLEMENDA CW, JOHNSTON CC JR: Baseline measurement

24. JENSEN GF, MEINECKE B, BOESEN J, TRANSBOL I: Does

of bone mass predicts fracture in white women. Ann Intern Med 2. WASNICH RD. Ross PD, HELLBRUN LK, VOGEL JM: Prediction of

I ,25(OH)2D3 accelerate spinal bone loss? A controlled therapeutic trial in 70-year-old women. Clin Orthop 192:215—221, 1985 25. AL0IA iF, VA5wANI A, YEH JK, ELLIS K, YASUMURA S, COHN

postmenopausal fracture risk with use of bone mineral measure-

SH: Calcitriol in the treatment of postmenopausal oSteoporosis.

ments. Am J Obstet Gynecol 153:745—751, 1985 3. JOHNSTON CC Ja, SLEMENDA CW, MELTON U III: Clinical use of bone densitometry. N EngI J Med 324:1105—1109, 1991 4. PARFITT AM, OLIVER I, WLACZAK N, LEVIN N, SANTIAGO G,

26. OTT SM, CHE5NUT CH III: Calcitriol treatment is not effective in postmenopausal osteoporosis. Ann Intern Med 110:267—274, 1989

CRUZ G: The effect of chronic renal failure and maintenance

porosis with high doses of synthetic calcitriol. Ann Intern Med

111:355—361, 1989

hemodialysis on bone mineral content of the radius. Am J Roentgenol 126:1292—1293, 1976

Am J Med 84:401—408, 1988 27. GALLAGHER JC, GOLDGAR D: Treatment of postmenopausal osteo113:649—655, 1990

28. MILLINER DS, NEBEKER HG, On SM, ANDRESS DL, SHERRARD

252

Ruedin et a!: Calcitriol and bone mineral density in CRF

DJ, ALFREY AC, SLATOPOLSKY BA, COBURN JW: Use of the

deferoxamine infusion test in the diagnosis of aluminum-related osteodystrophy. Ann Intern Med 101:775—780, 1984

29. SLOSMAN DO, RIzzoLl R, BUCHS B, PIANA F, DONATH A, BONJOUR JP: Comparative study of the performances of X-ray and gadolinium 153 bone densitometers at the level of the spine, femoral neck and femoral shaft. Eur J Nuci Med 17:3—9, 1990 30. BUCHS B, RIzzoLl R, SLOSMAN D, NYDEGGER V, BONJOUR JP:

Densit6 minrale osseuse de Ia colonne lombaire, du col et de Ia diaphyse fmoraux d'un échantillon de la population genevoise.

32. SILVERBERG Si, SHANE E, DE LA CRUZ L, DEMPSTER DW, FELDMAN F, SELDIN D, JACOBS TP, Sras ES, CAFFERTY M, PARISIEN MV, LINDSAY R, CLEMENS TL, BILEZIICIAN JP: Skeletal

disease in primary hyperparathyroidism. J Bone Miner Res 4:283— 291, 1989 33. ADLER AJ, BERLYNE GM: Duodenal aluminum absorption in the rat. Effect of vitamin D. Am J Physiol 249:G209—G213, 1985 34. TILYARD MW, SPEARS GFS, THOMSON J, DOVEY S: Treatment of

postmenopausal osteoporosis with calcitriol or calcium. N Engi J

Schweiz Med Wochenschr 122:1129—1136, 1992 31. RICHARDSON ML, Pozzi-MucELLI RS, KANTER AS, KOLB FO,

Med 326:357—362, 1992 35. RIZZOLI R, FLEISCH H, BONJOUR JP: Role of 1,25-dihydroxyvita-

ETrINGER B, GENANT HK: Bone mineral changes in primary

mm D3 on intestinal phosphate absorption in rats with a normal

hyperparathyroidism. Skeletal Radio! 15:85—95, 1986

vitamin D supply, J Clin Invest 60:639—647, 1977