Effects of Calcitriol or Calcium on Bone Mineral Density, Bone ...

0013-7227/01/$03.00/0 Printed in U.S.A.

The Journal of Clinical Endocrinology & Metabolism 86(9):4098 – 4103 Copyright © 2001 by The Endocrine Society

Effects of Calcitriol or Calcium on Bone Mineral Density, Bone Turnover, and Fractures in Men with Primary Osteoporosis: A Two-Year Randomized, Double Blind, Double Placebo Study PETER R. EBELING, JOHN D. WARK, STELLA YEUNG, CATHY POON, NOURIA SALEHI, GEOFFREY C. NICHOLSON, AND MARK A. KOTOWICZ Departments of Diabetes and Endocrinology (P.R.E., J.D.W., S.Y., C.P.) and Nuclear Medicine (N.S.), and Medicine, University of Melbourne Hospital, and Department of Medicine, Royal Melbourne Hospital (P.R.E., J.D.W.), Melbourne, Parkville 3050, Australia; and Department of Medicine, Geelong Hospital (G.C.N., M.A.K.), Geelong, Victoria 3220, Australia Osteoporosis in men is an emerging public health problem. As calcitriol reduces the rate of vertebral fractures in osteoporotic postmenopausal women, we conducted a prospective study of this treatment in men with primary osteoporosis. Our study was a 2-yr, randomized, double masked, double placebocontrolled trial of calcitriol (0.25 ␮g twice daily) or calcium (500 mg twice daily) in 41 men with primary osteoporosis and at least 1 baseline fragility fracture. Thirty-three men (85%) completed the study. There were no differences in baseline characteristics. Spinal and femoral neck bone mineral densities at 2 yr were

O

STEOPOROSIS IN men is already a public health concern (1–3). Over the next 15 yr, about 30% of all hip fractures will occur in men (4). Several recent populationbased studies report that the prevalence of spinal fractures is similar or only slightly less in men compared with women (5– 8). Vertebral fractures occur earlier in men, and in middleaged men the prevalence of vertebral fractures is higher than in women (9). Mortality after hip and spine fractures is also higher in men than in women (10). Longevity is increasing the absolute number of elderly men and women in the community. In addition, there is evidence that the age-specific incidence of fractures is increasing in men. Despite this, there have only been two studies examining the antifracture efficacy of any drug in men with osteoporosis (11, 12). A recent, large prospective study showed that alendronate increases bone density and reduces vertebral fractures in hypo- or eugonadal men with osteoporosis (12). Until now, the choice of antiosteoporotic therapy in men had been based on small observational studies of the effects of drugs on bone mineral density (BMD) in men and on inferences from studies in women with osteoporosis. Further studies in men are needed to provide additional drug therapy that has proven antifracture efficacy in men.

Abbreviations: BMC, Bone mineral content; BMD, bone mineral density; FCA, fractional calcium absorption; ITT, intention to treat; NTx, N-telopeptide cross-links; OC, osteocalcin; 25OHD, 25-hydroxyvitamin D; 1,25-(OH)2D3, 1,25-dihydroxyvitamin D3.

unchanged in both groups. Serum osteocalcin decreased in both groups by 30% (P < 0.05), whereas urine N-telopeptide cross-links decreased only in the calcium group by 30% (P < 0.05). After 2 yr, fractional calcium absorption increased by 34% (P < 0.01) in the calcitriol group. Nineteen incident fragility fractures occurred (14 vertebral and 5 nonvertebral) in 7 men. Over 2 yr, the number of men with vertebral fractures (6 vs. 1; P ⴝ 0.097) was similar in both groups. In conclusion, the efficacy of calcitriol remains unproven as a single agent for the treatment of osteoporosis in men. (J Clin Endocrinol Metab 86: 4098 – 4103, 2001)

Dietary calcium and vitamin D supplementation with 700 IU/d prevented age-related bone loss in men in one study (13), but not in another study of younger men with adequate dietary calcium intake (14). Dietary calcium and vitamin D supplementation results in small increases in spinal BMD in elderly men (13). Neither study had adequate power to detect changes in fracture rate adequately. Active intestinal fractional calcium absorption (FCA) declines with age in men, particularly after 80 yr, and may contribute to age-related secondary hyperparathyroidism and bone loss in both genders (15). Higher FCA values are associated with slower rates of femoral bone loss, but this may not be independent of 1,25-dihydroxyvitamin D [1,25-(OH)2D] concentrations. Men with spinal fractures also have lower FCA (16). In these men, low serum 1,25-(OH)2D concentrations only explain half of the deficit in FCA. In postmenopausal women with osteoporosis, calcitriol increases intestinal calcium absorption and prevents bone loss (17–19). In two studies, calcitriol has been reported to reduce the incidence of spine fractures (20, 21) in women with postmenopausal osteoporosis. To test the hypothesis that calcitriol might reverse the deficit in FCA in men with osteoporosis and increase BMD, we compared the effects of calcitriol alone with those of calcium in men with primary osteoporosis. We used changes in BMDs and total body bone mineral content (BMC) as primary end points, whereas biochemical bone turnover markers, active calcium absorption, and the numbers of new fractures were secondary end points.

4098

Ebeling et al. • Treatment of Osteoporosis in Men

The Journal of Clinical Endocrinology & Metabolism, September 2001, 86(9):4098 – 4103 4099

Subjects and Methods Subjects The study comprised 41 Caucasian men with primary osteoporosis, aged 27–77 yr, recruited consecutively from hospital clinics and the participating physicians’ private practices. Each patient had at least 1 fragility fracture; 39 had low trauma vertebral fractures, and 2 young men had low trauma proximal femoral fractures. One of the men with a prior hip fracture sustained a vertebral fracture during the study. In 4 men (3 in the calcium group and 1 in the calcitriol group), all evaluable lumbar vertebrae were fractured, and their LS-BMD data were excluded from the spinal BMD analysis (Table 1). The presence of a low trauma fracture was the main criterion for entry. Although low BMD was not required, the majority of men had a baseline spinal or hip t score of less than ⫺2.5. No men had disease known to affect bone or mineral metabolism, and all had normal baseline 25-hydroxyvitamin D (25OHD) and T concentrations. The 2-yr prospective study was double masked with a double placebo. Patients were randomized to receive either calcitriol (0.25 ␮g twice daily) and placebo calcium tablets or calcium (500 mg twice daily) and placebo calcitriol capsules. BMD and BMC were measured at baseline and every 6 months; biochemical measurements of bone turnover were measured at baseline and every 3 months; thoracicolumbar spinal x-rays were performed at baseline and annually, and FCA was measured at baseline and at study completion. All patients gave informed consent, and the human research and ethics advisory committees of The Royal Melbourne Hospital Research Foundation and Geelong Hospital approved the research protocol.

Bone densitometry and biochemical markers of bone turnover BMDs of the spine (second to fourth lumbar vertebrae) and the femoral neck, and total body BMC were measured by dual x-ray absorptiometry using QDR-2000 (Hologic, Inc., San Francisco, CA; Royal Melbourne Hospital; n ⫽ 35) and DPX-L (Lunar Corp., Madison, WI; Geelong Hospital; n ⫽ 6) densitometers. The in vitro and in vivo coefficients of variation were 0.38% and 1% at the lumbar spine and 0.38% and 1.7% at the femoral neck, respectively (22). Standardized BMD was determined by comparison of the individual BMD with the appropriate North American reference data and was expressed as number of sd (z-score) different from the age- and sex-specific mean BMD. For measurement of biochemical bone turnover markers, blood samples were taken, and 2-h urine specimens were collected between 0700 – 0900 h after an overnight fast. All samples were stored at ⫺70 C until

analysis. The urine total pyridinium cross-links, pyridinoline and deoxypyridinoline were measured by monitoring fluorescence of eluates from HPLC (23). All pyridinium cross-link values were corrected for the individual measured recovery of the internal standard, isodesmosine, and urinary creatinine concentrations. Intra- and interassay coefficients of variation were each 8% and 10% for pyridinoline and deoxypyridinoline, respectively. Urinary N-telopeptide (NTx) cross-links were measured by duplicate ELISAs (Osteomark, Seattle, WA). The intra- and interassay coefficients of variation were 8% and 9%, respectively. Serum bone alkaline phosphatase was measured by duplicate immunoradiometric assays (24) using two monoclonal antibodies directed toward the bone isoenzyme of alkaline phosphatase (MetraBiosystems, Mountain View, CA). The intraassay coefficient of variation was 8%, and cross-reaction with other alkaline phosphatases was 6%. Serum osteocalcin (OC) was measured by duplicate immunoradiometric assays using antibodies raised against human OC (Immutopics, Palo Alto, CA). The intra- and interassay coefficients of variation were 7% and 9%, respectively.

Assessment of FCA Active, or vitamin D-dependent, intestinal calcium absorption was assessed by a modification of the method described by Need et al. (16). Radiocalcium (45Ca) was purchased from the Australian Atomic Energy Commission (Lucas Heights, Australia) and was administered orally with 20 mg calcium carrier (as CaCl2) in 200 ml demonized water. Heparinized blood samples were obtained at baseline, 30 and 60 min after treatment, instead of at 60 min alone, and FCA was calculated as the fraction absorbed per h. FCA was not corrected for age or dietary calcium intake. Dietary calcium intakes were calculated from 4-d diet diaries at baseline and 2 yr.

Thoracicolumbar spine radiographs Roentgenograms of the thoracic and lumbar spine were obtained using two x-ray machines. The heights of the anterior, mid, and posterior margins of each vertebral body were measured to the nearest 0.1 mm; all spinal measurements were made with a pair of calipers by two observers, masked to the treatment assignment (P.R.E. and M.A.K.). One of the observers (P.R.E.) remeasured vertebral heights measured at the second site. A vertebral fracture was defined as a decrease of 20% in the anterior, mid, or posterior height of the body of any vertebra from T4 to L4. The changes in vertebral heights at yr 1 and 2 were compared with baseline to detect new vertebral. All incident vertebral fractures represented changes from a normal vertebra, and the majority of fractures

TABLE 1. Baseline variables (mean ⫾ SD) relating to age, bone density (BMD), total body bone mineral content (TBBMC), prevalent vertebral fractures, fractional calcium absorption (FCA), daily dietary calcium intake (diet Ca), biochemical bone turnover markers [osteocalcin; bone alkaline phosphatase (BAP); deoxypyridinoline (D-Pyr); and N-telopeptide (NTx)], serum hormone concentrations, creatinine clearance, and urinary calcium excretion for 39 men with primary osteoporosis Variable

Age (yr) LS-BMD (g/cm2) FN-BMD (g/cm2) TBBMC (g) Fractures (no.) FCA [fraction/h (0.26 – 0.88)] Diet Ca [mg/d (800 mg/d)] Osteocalcin [ng/mL (12.1–18.9)] BAP [IU/L (15– 41.3)] D-Pyr [nmol/mmol creatinine (8 –14)] NTx [pmol BCE/␮mol Cr (19.6 – 62.6)] T [nmol/ml (7–28)] PTH [pmol/liter (1– 6.5)] 25OHD [nmol/liter (25–108)] 1,25-(OH)2D3 [pmol/liter (34 –134)] Creatinine clearance, ml/sec (1.5–2.5) Urinary Calcium [mmol/d (2.5– 6.2)] Normal ranges are in parentheses where indicated.

Calcium (n ⫽ 19)

Calcitriol (n ⫽ 20)

60.5 ⫾ 12.4 0.840 ⫾ 0.176 (n ⫽ 16) 0.636 ⫾ 0.121 2148 ⫾ 376 4.5 ⫾ 3.6 0.57 ⫾ 0.26 697 ⫾ 287 6.2 ⫾ 2.6 12.1 ⫾ 3.0 12.0 ⫾ 5.0 62 ⫾ 36 19.3 ⫾ 8.1 3.5 ⫾ 1.2 86 ⫾ 27 75 ⫾ 26 1.9 ⫾ 0.9 4.7 ⫾ 2.6

57.5 ⫾ 11.3 0.868 ⫾ 0.191 (n ⫽ 19) 0.688 ⫾ 0.103 2375 ⫾ 408 3.2 ⫾ 2.3 0.70 ⫾ 0.20 675 ⫾ 375 5.2 ⫾ 1.7 11.9 ⫾ 2.7 10.5 ⫾ 3.6 50 ⫾ 24 16.9 ⫾ 5.7 4.3 ⫾ 2.0 91 ⫾ 42 83 ⫾ 17 1.9 ⫾ 0.6 4.5 ⫾ 2.6

P

0.43 0.66 0.16 0.08 0.16 0.10 0.84 0.15 0.82 0.29 0.60 0.27 0.15 0.69 0.27 0.99 0.75

4100

The Journal of Clinical Endocrinology & Metabolism, September 2001, 86(9):4098 – 4103

were also clinically apparent. Clinical nonvertebral fractures were ascertained by questioning the patient, and incident worsening of prevalent baseline vertebral fractures was also noted.

Statistical analysis Preliminary power calculations showed that we had 88% power (␣ ⫽ 0.05) to detect a difference of 3% in spinal BMD between groups at 2 yr, assuming an sd of 3% in the percent change in BMD. Two-sample t tests or their nonparametric equivalent, e.g. the Wilcoxon rank sum test if data were found not to be normally distributed, were performed to test for baseline differences and to compare differences between treatment groups in responses of bone density, biochemical bone turnover markers, and FCA. Changes from baseline were defined as the study time point value minus the baseline value, expressed as a percentage. To assess these changes, a one-sample t test and or its nonparametric equivalent, e.g. Wilcoxon sign test, was performed for each treatment group. The differences between groups were compared using two-sample t tests or their nonparametric equivalent. Repeated measures ANOVA was also used to test for differences between treatment groups and changes over time within treatment groups for the BMD parameters and the biochemical bone marker data. The differences in incident vertebral fractures were tested by Fisher’s exact test. The analysis was by intention to treat (ITT), and where there were missing data, the last value entered was carried forward to each visit up to month 24 for analysis of BMD data and repeated measures analyses. All analyses were performed with the SAS software program (25).

Results

Of the 41 men enrolled, 39 men were evaluable for ITT analysis. Thirty-seven men (17 in the calcium group and 20 in the calcitriol group) completed 12 months, and 33 men (16 in the calcium group and 17 in the calcitriol group) completed 24 months. The overall study retention rate was 85%. Baseline variables for the 39 men evaluable for ITT analysis are shown in Table 1. There were no differences in baseline characteristics. The men had an overall mean of 3.8 prevalent baseline vertebral fractures (range, 0 –12). Regarding baseline bone turnover, mean serum OC and urinary bone resorption markers were in the upper part of the normal range for men. The mean baseline FCA was decreased by more than 1 sd in


18% compared with controls (16). Baseline serum T, 25OHD, and 1,25-(OH)2D concentrations were all within the normal range in both groups. Changes in BMD, biochemical bone turnover markers, and FCA

The changes in regional BMDs and total body BMC are shown in Table 2 and Fig. 1. Although there was a transient increase in femoral neck BMD in the calcium-treated group, by 2 yr there were no significant changes in BMD in either group relative to baseline or relative to each other. If spinal BMD data of the men with four lumbar vertebral fractures were included, BMD appeared to increase by 8.8% and 7.8% at 1 and 2 yr, respectively, in the calcium group. There were also nonsignificant increases of 1.4% and 1.6%, respectively, in the calcitriol group. There were no differences between treatment groups in bone turnover markers (Table 2 and Fig. 2). Mean serum bone alkaline phosphatase concentrations decreased during the first 6 months of treatment in both groups and then increased above baseline at 24 months; however, the percent increases in each group were not significant. Urinary NTx/creatinine decreased significantly at 6 months in both groups; however, at 24 months it had decreased by 30% (P ⬍ 0.05) in the calcium group and by 6% (P ⫽ 0.75) in the calcitriol group. The decrease in NTx/creatinine after calcium treatment is greater than the decrease seen with calcium (500 mg) and vitamin D (400 – 450 IU) in men with osteoporosis, where NTx decreased by 9% (12). However, it is less than the 59% decrease seen in men treated with alendronate in addition to calcium and vitamin D in the same study. Serum osteocalcin decreased by 35% in the calcium group and by 25% in the calcitriol group (P ⬍ 0.01 and P ⬍ 0.05, respectively) at 12 months, remained below baseline in both groups, and was significant in the calcium group (P ⬍ 0.05) at 24 months. The mean daily dietary calcium increased by 24% to 729

TABLE 2. Percentage changes in variables (mean ⫾ SD) relating to bone density (BMD), total body bone mineral content (TBBMC), prevalent vertebral fractures, fractional calcium absorption (FCA), daily dietary calcium intake (diet Ca), biochemical bone turnover markers [osteocalcin; bone alkaline phosphatase (BAP); deoxypyridinoline (D-Pyr); and N-telopeptide (NTx)], serum hormone concentrations, and creatinine clearance at 1 and 2 yr for 39 men with primary osteoporosis Variable

LS-BMD (g/cm2)a FN-BMD (g/cm2)b TBBMC (g)b FCA (fraction/h)c Dietary Ca (mg/d) Osteocalcin (ng/ml) BAP (IU/liter) D-Pyr (nmol/pmol creatinine) NTx (nmol BCE/pmol creatinine) T (ng/ml) PTH (pmol/liter) 25OHD (nmol/liter) 1,25-(OH)2D3 (pmol/liter) Ca (corr) (mmol/liter) 24-h urinary Ca (mmol/d) Creatinine clearance (ml/sec)

Calcium 12 months (n ⫽ 17)

3.1 ⫾ 5.5 2.7 ⫾ 5.3 1.3 ⫾ 4.0 ⫺35 ⫾ 22c ⫺5.5 ⫾ 21 ⫺17 ⫾ 44 ⫺19 ⫾ 47d ⫺13 ⫾ 55 2.0 ⫾ 4.4 6.2 ⫾ 45 ⫺10 ⫾ 31

Calcitriol

24 months (n ⫽ 16)

12 months (n ⫽ 20)

24 months (n ⫽ 19)

1.9 ⫾ 5.7 2.0 ⫾ 6.0 ⫺0.9 ⫾ 5.7 14 ⫾ 48 24 ⫾ 71 ⫺18 ⫾ 26d 23 ⫾ 26 8 ⫾ 56 ⫺30 ⫾ 55d ⫺2.0 ⫾ 21 ⫺2.6 ⫾ 44 ⫺3 ⫾ 48 ⫺9.2 ⫾ 12.1c 1.6 ⫾ 5.1 0.7 ⫾ 44 ⫺6 ⫾ 26

1.4 ⫾ 6.9 0.2 ⫾ 4.0 0.6 ⫾ 2.7

1.6 ⫾ 10.1 ⫺0.05 ⫾ 4.3 ⫺0.9 ⫾ 4.8 34 ⫾ 55c 40 ⫾ 97 ⫺6.9 ⫾ 39 41 ⫾ 46 42 ⫾ 112 ⫺6 ⫾ 83d ⫺2.9 ⫾ 32 ⫺26 ⫾ 42 ⫺11 ⫾ 27 3.3 ⫾ 19.3 4.6 ⫾ 5.3c 140 ⫾ 231d ⫺0.3 ⫾ 25

⫺25 ⫾ 33d ⫺4.0 ⫾ 28 0.4 ⫾ 54 ⫺16 ⫾ 28d ⫺23 ⫾ 41d 2.7 ⫾ 3.6d 115 ⫾ 285d ⫺0.1 ⫾ 19

P12

0.44 0.10 0.47 0.33 0.86 0.30 0.81 0.53 0.76 0.009 0.29

P24

0.90 0.23 1.0 0.26 0.93 0.34 0.16 0.26 0.33 0.92 0.12 0.55 0.032 0.10 0.003 0.53

For differences from baseline: d, P ⬍ 0.05; c, P ⬍ 0.01, differences between groups at 12 and 24 months, respectively: P12 and P24. Numbers evaluable by ITT analysis: a, (n ⫽ 16, n ⫽ 19; b, n ⫽ 19, n ⫽ 20 at both time points for calcium and calcitriol groups, respectively, and c, n ⫽ 18 for calcitriol group at 24 months.



FIG. 2. Percent changes (mean ⫾ SEM) in biochemical markers of bone turnover in 39 men with primary osteoporosis treated with calcitriol or calcium for 2 yr. Data are the mean ⫾SEM for serum osteocalcin (A) and urinary NTx (B). To allow for multiple comparisons, only differences from baseline of P ⬍ 0.01 (*) are identified.

FIG. 1. Percent changes (mean ⫾ SEM) in lumbar spine (A) and femoral neck (B) BMD and total body BMC (C) in 39 men with primary osteoporosis treated with calcitriol or calcium for 2 yr. *, P ⬍ 0.05 vs. baseline.

mg/d (P ⫽ 0.19), and by 40% to 757 mg/d (P ⫽ 0.08) in the calcium and calcitriol groups, respectively (Table 2). At 2 yr, the total daily calcium intake in the calcium group was 1729 mg. FCA increased by 34% (P ⬍ 0.01) at 2 yr in the calcitriol group, but did not change significantly in the calcium group (14%). Sixteen men (10 in the calcium group and 6 in the calcitriol group) had low baseline fractional calcium absorption (⬍0.55/h). They did not have greater increases in spinal, femoral neck, or total hip BMDs with either calcitriol or calcium therapy than men with normal baseline FCA. Fractures

A total of 19 new fragility fractures (14 vertebral and 5 nonvertebral) occurred in 7 men over 2 yr. Three incident vertebral fractures occurred in 3 men, and 10 incident vertebral fractures occurred in 5 men in the calcitriol group during the first and second years of the study, respectively. Two men in the calcitriol group had incident vertebral fractures in both years of the study. Only 1 incident vertebral fracture occurred in the calcium group in the first year of the study. Thus, over 2 yr, 1 of 16 patients (6%) in the calcium group and 6 of 19 patients (32%) in the calcitriol group had at least 1 new fracture. There was a nonsignificant trend for

more men in the calcitriol group to sustain vertebral fractures (6 vs. 1; P ⫽ 0.097) over 2 yr. All 5 nonvertebral fractures (2 ribs, distal radius, and superior and inferior pubic rami) occurred in the calcitriol group. The fractured pubic rami were sustained after a fall down steps, whereas all other fractures were low trauma fractures. Incident worsening of a prevalent vertebral fracture occurred only in 1 patient in the calcium group, who also sustained an incident new vertebral fracture. Changes in calciotropic hormone and T concentrations

Neither serum 25OHD nor total T concentrations differed significantly from baseline (Table 2). Serum 1,25-(OH)2D3 decreased by 9% in the calcium group (P ⬍ 0.01), but increased slightly (P ⫽ 0.89) in the calcitriol group, with a significant difference between treatment groups (P ⫽ 0.03). Venous sampling was performed at inconstant times after the calcitriol treatment and may have been too late to detect increases in serum 1,25-(OH)2D3 concentrations related to calcitriol therapy. Serum PTH concentrations in the calcitriol group decreased by 23% and 26% at 1 and 2 yr (P ⫽ 0.01 and P ⫽ 0.12, respectively), but did not change significantly in the calcium group. Safety

Two patients, one in each treatment group, were included in the safety population, although they were excluded from the ITT population, one because a repeat serum T concentration was below the normal range, and the second because

4102

The Journal of Clinical Endocrinology & Metabolism, September 2001, 86(9):4098 – 4103

prostatic carcinoma was diagnosed after the baseline visit. The withdrawal rate was similar in both groups, with another three patients withdrawing in each group. The reasons for withdrawal in the three patients who withdrew from the calcitriol group were multiple vertebral fractures, myalgia, and nausea. In the three patients who withdrew from the calcium group, the reasons for withdrawal were fatigue, diagnosis of a malignant fibrous histiocytoma, and an interstate move. Serum calcium levels rose within the normal range, and urinary calcium excretion increased by 140% from baseline during calcitriol therapy. Although urinary calcium excretion was increased above 10 mmol/d in 2 of the patients taking calcitriol, it was asymptomatic. However, there were 22 of 35 patients (63%) with urinary calcium that increased by more than 0.1 mmol/d above baseline. No man complained of passing urinary gravel or of renal colic. In 1 of these patients, a decreased creatinine clearance did not return to normal. Serum calcium (corrected for the serum albumin concentration) increased above normal to 2.67 mmol/ liter in only 1 man in the calcium group. No hypercalcemia occurred in the calcitriol group. Discussion

We found that men with primary osteoporosis treated with calcitriol and calcium therapy had transient increases in spinal and femoral neck BMDs; however, in both groups BMD returned to baseline at both sites by 2 yr. No lasting increases were seen in the total body calcium with calcitriol or calcium treatment. FCA was increased by calcitriol, whereas serum PTH concentrations were decreased. There are no previous data on the use of calcitriol therapy in men with osteoporosis. In a 3-yr study of postmenopausal women with osteoporosis, calcitriol therapy resulted in stabilization of the vertebral fracture rate compared with a 3-fold increase in the vertebral fracture rate over the last 2 yr of the study in women receiving calcium (20). In the first year of another 3-yr study, vertebral fracture rates were also reduced by calcitriol therapy; the latter 2 yr of this study were uncontrolled (21). Differences between studies in the effects of calcitriol on BMD may be related to either differing doses of calcitriol used or differing habitual dietary calcium intakes. Aloia et al. (17) showed significant increases in spinal and distal radius BMDs and total body calcium with calcitriol at an average dose of 0.8 ␮g/d. Vertebral fracture rates tended to be higher in the placebo group, and there was a high incidence of hypercalciuria and hypercalcemia in the calcitriol group. Gallagher and Goldgar (18), using an average calcitriol dose of 0.62 ␮g/d, showed an increase in spinal BMD and stable total body calcium compared with decreases in subjects taking placebo, without an adverse effect on serum calcium concentrations. However, there were no differences in fracture rates. Ott and Chestnut (26) used the lowest final average calcitriol dose (0.43 ␮g/d), and the dietary calcium intake of their placebo group was 400 mg/d higher. Changes in BMD at most sites were similar in each group. The incidence of vertebral fractures was 10% higher in the calcitriol group, but


this difference was not significant. When subjects in this study were subdivided according to average daily calcitriol doses, those receiving more than 0.6 ␮g/d had the greatest increases in BMD (27). Only one study using low average daily calcitriol doses (0.42 ␮g/d) has shown vertebral height loss (28), which was not seen in women receiving hormone replacement therapy alone or placebo. The goal of osteoporosis treatment is to prevent further fragility fractures. Although fractures were a secondary end point of our study, and we had limited power to detect differences between groups, we detected a trend for an increased number of new fractures in men receiving calcitriol. Only one man receiving calcium had a new vertebral fracture. Neither treatment group showed a decrease in BMD after 2 yr of treatment. Although the men receiving calcitriol had an average dietary calcium intake below the recommended daily allowance, dietary calcium intakes were similar in men with and without fractures in the calcitriol group (857 ⫾ 359 and 711 ⫾ 402 mg/d; P ⫽ 0.45). An elevated bone resorption rate is an independent risk factor for fracture (29). However, there was no evidence of increased bone resorption in the calcitriol group in our study. It is also possible that there is a narrow therapeutic window for calcitriol, and the response to a fixed dose in a population is heterogeneous (30), or that there are gender differences in its effects on bone and mineral metabolism, including bone turnover. In women treated with calcitriol, bone turnover is decreased, and this may partially result from suppression of the activation of bone turnover by PTH. However, high doses of calcitriol (2 ␮g/d) given to normal men over 7 d increased bone formation markers without increasing bone resorption (31). The incident fracture rate in our study is high, with a 32% 2-yr incidence of fractures in the calcitriol group. However, the main limitation of our study is its small sample size and limited power to detect differences in fracture rates. Although our study had only 38% probability to detect a difference in fracture rates, a 26% difference in fracture rates between treatment groups is of biological concern. Nevertheless, we cannot confidently distinguish fracture rates between groups. However, a larger study adequately powered to examine differences in fracture rates, would be unwise given that our study may indicate a safety problem with the use of calcitriol alone in men with severe osteoporosis. By comparison, our power to detect differences in BMD based on our current data were even more limited. In conclusion, there were no differences between calcitriol and calcium with respect to their effects on BMD and bone turnover, but there was a trend for calcitriol therapy to be associated with an increased number of vertebral fractures. The latter raises concern, but our study had limited power to detect a difference in fracture rates. In conclusion, the efficacy of calcitriol remains unproven as a single agent for the treatment of osteoporosis in men. Acknowledgments We acknowledge the study participants for their time and enthusiasm in participating in the study. We also thank Bahtiyar Kaymakci for his technical expertise in bone densitometry. We are grateful to Profs. B. E. Christopher Nordin, Ego Seeman, and John Hopper for their critical



comments regarding our study. The statistical analyses were performed by Datapharm Australia and were checked by Mr. Andrew Martin and Dr. Philip McCloud at Roche Products (Australia) Pty. Ltd. Received July 20, 2000. Accepted May 16, 2001. Address all correspondence and requests for reprints to: Dr. Peter R. Ebeling, Department of Diabetes and Endocrinology, The Royal Melbourne Hospital, Melbourne, Victoria 3050, Australia. E-mail: p.ebeling@ medicine.unimelb.edu.au. This work was supported by the D. W. Keir Fellowship, The Royal Melbourne Hospital, and Roche Products (Australia) Pty. Ltd. Presented in part at the European Congress on Osteoporosis, Berlin, Germany, September 1998.

References 1. Orwoll ES, Klein RF 1995 Osteoporosis in men. Endocr Rev 16:87–116 2. Seeman E 1993 Osteoporosis in men: epidemiology, pathophysiology and treatment possibilities. Am J Med 95:225–285 3. Ebeling PR 1998 Osteoporosis in men. New insights into aetiology, pathogenesis, prevention and management. Drugs Aging 13:421– 434 4. Cooper C, Campion G, Melton LJ 1992 Hip fractures in the elderly: a worldwide projection. Osteop Int 2:285–289 5. Cooper C, Atkinson EJ, O’Fallon MW, Melton III LJ 1992 Incidence of clinically diagnosed vertebral fractures: a population-based study in Rochester, Minnesota 1985–1989. J Bone Miner Res 7:221–227 6. Jones G, Nguyen T, Sambrook PN, Kelly PJ, Eisman JA 1994 Symptomatic fracture incidence in elderly men and women: the Dubbo osteoporosis epidemiology study (DOES). Osteop Int 4:277–282 7. Melton III LJ, Thamer M, Ray NF, Chan JK, Chestnut III CH, Einhorn TA, et al. 1997 Fractures attributable to osteoporosis: report from the National Osteoporosis Foundation. J Bone Miner Res 12:16 –23 8. Burger H, Van Daele PLA, Grasluis K, et al. 1997 Vertebral deformities and functional impairment in men and women. J Bone Miner Res 12:152–157 9. Jones G, White C, Nguyen T, Sambrook PN, Kelly PJ, Eisman JA 1996 Prevalent vertebral deformities: relationship to bone mineral density and spinal osteophytosis in elderly men and women. Osteop Int 6:233–239 10. Poor G, Atkinson EJ, Lewallen DG, O’Fallon WM, Melton III LJ 1995 Agerelated hip fractures in men: clinical spectrum and short-term outcome. Osteop Int 5:419 – 426 11. Ringe JD, Dorst A, Kipshoven C, Rovati LC, Setnikar I 1998 Avoidance of vertebral fractures in men with idiopathic osteoporosis by a three year therapy with calcium and low-dose intermittent monofluorophosphate. Osteop Int 8:47–52 12. Orwoll E, Ettinger M, Weiss S, et al. 2000 Alendronate for the treatment of osteoporosis in men. N Engl J Med 343:604 – 610 13. Dawson-Hughes B, Harris SS, Krall EA, et al. 1997 Effect of calcium and

14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31.

vitamin D supplementation on bone density in men and women 65 years of age or older. N Engl J Med 337:670 – 676 Orwoll ES, Oviatt SK, McClung MR, Deftos LJ, Sexton G 1990 The rate of bone loss in normal men and the effects of calcium and cholecalciferol supplementation. Ann Intern Med 112:29 –34 Wolf RL, Zmuda JM, Charron M, Heaney RP, Cauley JA 1999 Calcium absorption efficiency in older men: Relationships to age and bone loss. J Bone Miner Res 14:A1193 Need AG, Morris HA, Horowitz M, Scopacasa F, Nordin BEC 1998 Intestinal calcium absorption in men with spinal osteoporosis. Clin Endocrinol (Oxf) 48:163–168 Aloia JF, Vazswani A, Yeh JK, Ellis K, Yasumura S, Cohn SH 1988 Calcitriol in the treatment of postmenopausal osteoporosis. Am J Med 84:401– 408 Gallagher JC, Goldgar D 1990 Treatment of postmenopausal osteoporosis with high doses of synthetic calcitriol. A randomized controlled study. Ann Intern Med 113:649 – 655 DeLuca HF, Riggs BL 1982 1,25-Dihydroxyvitamin D3: short and long-term effects on bone and calcium metabolism in patients with postmenopausal osteoporosis. Proc Natl Acad Sci USA 79:3325–3329 Tilyard MW, Spears GF, Thomson J, Dovey S 1992 Treatment of postmenopausal osteoporosis with calcitriol or calcium. N Engl J Med 326:357–362 Gallagher JC, Riggs BL, Recker RR, Goldgar D 1989 The effect of calcitriol on patients with postmenopausal osteoporosis with special reference to fracture frequency. Proc Soc Exp Biol Med 191:287–291 Khan KM, Henzell SL, Broderick C, et al. 1997 Instrument performance in bone density testing at five Australian centres. Aust NZ J Med 27:526 –530 Ebeling PR, Atley LM, Guthrie JR, et al. 1996 Bone turnover markers and bone density across the menopausal transition. J Clin Endocrinol Metab 81:3366 – 3371 Gomez Jr B, Ardakani S, Ju J, et al. 1995 Monoclonal antibody assay for measuring bone-specific alkaline phosphatase activity in serum. Clin Chem 41:1560 –1566 SAS Institute 1999 SAS system for windows, version 6.12. Cary: SAS Institute Ott SM, Chestnut CH 1989 Calcitriol treatment is not effective in postmenopausal osteoporosis. Ann Intern Med 110:267–274 Ott SM, Chestnut CH 1990 Tolerance to dose of calcitriol is associated with improved density in women with postmenopausal osteoporosis. J Bone Miner Res 5(Suppl 2):S186 Jensen GF, Meinecke B, Boesen J, Transbol I 1985 Does 1,25-(OH)2D3 accelerate spinal bone loss? Clin Orthop 192:215–221 Melton III LJ, Khosla S, Atkinson EJ, O’Fallon WM, Riggs BL 1997 Relationship of bone turnover to bone density and fractures. J Bone Miner Res 12:1083–1091 Kanis JA, McCloskey EV, de Takats D, Bernard J, Zhang DM 1997 Treatment of osteoporosis with vitamin D. Osteop Int 7(Suppl 3):S140 –S146 Bollerslev J, Gram J, Nielsen HK, et al. 1985 Effect of a short course of 1,25-dihydroxyvitamin D3 on biochemical markers of bone remodeling in adult male volunteers. J Clin Endocrinol Metab 60:615– 617