Full text of DOI:10.1359/jbmr.2001.16.3.565

JOURNAL OF BONE AND MINERAL RESEARCH
Volume 16, Number 3, 2001
© 2001 American Society for Bone and Mineral Research
Protective Effect of Short-Term Calcitriol or Cyclical
Etidronate on Bone Loss After Cardiac
or Lung Transplantation
KATHY HENDERSON,1,2 JOHN EISMAN,1 ANNE KEOGH,3 PETER MACDONALD,3
ALLAN GLANVILLE,3 PHILLIP SPRATT,3 and PHILIP SAMBROOK4
ABSTRACT
Bone loss is most rapid in the immediate period after cardiac or lung transplantation. This randomized study
compared the efficacy of 6 months of treatment with either calcitriol (1,25-dihydroxyvitamin D3; 0.5 ␮g/day)
or two cycles of etidronate plus calcium in preventing bone loss in 41 patients undergoing cardiac or lung
transplantation. Patients were followed for 18 months after cessation of treatment. Bone mineral density
(BMD) was measured by dual-energy X-ray absorptiometry (DXA). There were no significant differences
between groups with respect to age or cumulative dose of prednis(ol)one or cyclosporin over the 2 years. Bone
loss did not differ between groups after 6 months and, despite 6 months prophylaxis with either agent, bone
loss was significant in both groups at 6 months and 12 months. However, compared with an untreated
reference group, both therapies offered significant protection at 6 months and etidronate provided significant
protective carryover after therapy had been discontinued. These data suggest short-term prophylaxis with
calcitriol or cyclical etidronate is partially effective in reducing bone loss after cardiac or lung transplantation
but treatment needs to be continued for a longer term. (J Bone Miner Res 2001;16:565–571)
Key words: corticosteroid, osteoporosis, transplantation
comparative efficacy of calcitriol and bisphosphonates in
posttransplantation osteoporosis. Furthermore, although
bone loss is known to be most rapid during the first 6
months after transplantation, no study has investigated
whether short-term treatment early in the posttransplanta-
tion period provides longer-term skeletal protection.
This study compared the efficacy of calcitriol and cyclical
etidronate, administered for the first 6 months after surgery,
in patients undergoing cardiac or lung transplantation. The
INTRODUCTION
ATIENTS WHO have undergone heart transplantation lose
bone rapidly and are at increased risk of developing
P
osteoporotic fractures.(1) However, bone loss is most rapid
during the first 6 months after transplantation, and by the
third year after transplantation the majority of patients are
no longer losing bone from the lumbar spine(2,3) or femoral
neck.(3)
Prophylactic treatment with either cyclical etidronate or primary outcomes of the study were change in bone density
calcitriol has been shown to prevent lumbar bone loss in at the end of the treatment phase (6 months after transplan-
patients starting corticosteroids.(4,5) However, at the time tation) and after a further 6 months without preventative
this study was initiated, there were no data available on the therapy. A secondary endpoint was whether either of the
1Bone and Mineral Research Program, Garvan Institute of Medical Research, Sydney, Australia.
2Present address: School of Physiotherapy, Curtin University of Technology, Perth, Western Australia.
3Department of Cardiopulmonary Transplantation, St. Vincent’s Hospital, Sydney, Australia.
4University of Sydney, Sydney, Australia.
565

566
HENDERSON ET AL.
TABLE 1. BASELINE CHARACTERISTICS OF STUDY AND REFERENCE PATIENTS
Calcitriol
Etidronate
Reference
Number
Sex (M/F)*
21
15/6
20
15/5
25
21/4
Type of transplant (Heart/Lung)*
Age (years)
Height (cm)
16/4
15/6
25/0
49.9 Ϯ 9.5
172.1 Ϯ 10.4
1.1 Ϯ 0.5
72.6 Ϯ 13.1
48.8 Ϯ 12.3
168.5 Ϯ 8.2
1.1 Ϯ 0.5
69.2 Ϯ 11.0
47.6 Ϯ 14.1
172.4 Ϯ 9.9
N/A
Dietary calcium intake (g/day)
Weight (kg)
69.3 Ϯ 10.1
BMD (g/cm2)
Lumbar spine
Femoral neck
Whole body
1.14 Ϯ 0.20
0.88 Ϯ 0.17
1.16 Ϯ 0.17
1.09 Ϯ 0.13
0.88 Ϯ 0.12
1.15 Ϯ 0.08
1.19 Ϯ 0.18
N/A
1.18 Ϯ 0.10
Values are mean Ϯ SD.
N/A, not measured.
* Significant difference between groups, ␹ ( p Ͻ 0.05).
2
therapies conferred any long-term carryover protective ef- written informed consent to participate. For ethical reasons
fect during the subsequent 12 months. Efficacy versus no relating to the well-known rapid bone loss during the early
treatment was estimated in comparison with an untreated posttransplantation period, a randomized placebo-control
reference group of patients from the same facility.
group was not included in this study. However, reference
data from a historical group of comparable patients from the
same transplantation unit were available. Hence, the pri-
mary endpoints of this study were limited to comparisons
between the two therapies of interest while reference data
provided an indication of outcome relative to no preventa-
tive therapy.
The reference group of 25 patients had been recruited
between May 1991 and July 1992 for a longitudinal obser-
vational study of bone mineral density (BMD) after cardiac
transplantation.(2) Six patients from this group subsequently
commenced on calcitriol, because of concern about rapid
bone loss during the first 12 months after transplantation,
and were excluded from analyses unless otherwise stated.
One patient from the reference group died during the first
year, while another was not available for 2-year follow-up
measurements.
MATERIALS AND METHODS
Patients and study design
Forty-one patients were recruited from St. Vincent’s Hos-
pital, Sydney, Australia, between August 1995 and Decem-
ber 1996, all within 1 week of heart or lung transplantation.
Patients with preexisting diseases known to affect bone and
mineral metabolism; significant renal impairment or liver
disease; or prior treatment with chronic corticosteroids,
calcitonin, calcitriol, fluoride, or vitamin D Ͼ50,000
U/week were excluded. The study was randomized with
stratification according to age (Յ40 years or Ͼ40 years),
sex, and type of transplant.(6) Twenty-one patients were
assigned to receive calcitriol, 0.5 ␮g/day (Rocaltrol; Roche,
Basel, Switzerland), for 6 months and 20 patients were to
receive two cycles of etidronate, 400 mg/day, for 14 days
followed by 76 days of calcium carbonate, 1.25 g/day
(Didrocal; Procter and Gamble Pharmaceuticals, North Nor-
Immunosuppression
All patients received initial and maintenance immunosup-
wich, NY, USA). Baseline dietary calcium intake, assessed pression with corticosteroids, cyclosporin, and azathioprine.
by food frequency questionnaire,(7) did not differ between Corticosteroids were commenced perioperatively with in-
groups (Table 1) and patients from the calcitriol group did travenous methylprednisolone, 2 g, followed by 125 mg
not receive supplemental calcium.
every 8 h for three doses and then oral prednisolone at 1
During the first posttransplant year, 6 patients died (4 mg/kg per day reducing to 0.15–0.18 mg/kg per day by day
patients from the calcitriol group and 2 patients from the 14 and 0.10–0.15 mg/kg per day by 6 months. An intrave-
etidronate group), 3 patients within the first 6 months after nous dose of cyclosporin (2–3 mg/kg) was administered
transplantation. No deaths appeared to be related to the trial preoperatively where serum creatinine was Յ0.14 mmol/
medication. One patient withdrew from the study at 3 liter and oral cyclosporin was commenced 24 h postopera-
months without ever taking any trial medication and another tively. Dosage was adjusted according to blood levels, the
on etidronate developed avascular necrosis of the femoral target range varying according to time posttransplantation.
head, which required bilateral hemiarthroplasty 17 months Azathioprine, orally, was given to all patients at a dose of 2
posttransplantation. Adherence was monitored by tablet mg/kg per day. Rejection was managed by high-dose intra-
counts at 3 months and 6 months.
venous or oral corticosteroids followed by a rapid taper, the
The study was approved by the Research Ethics Commit- precise regimen according to protocol, dependent on sever-
tee of the St. Vincent’s Hospital and all patients provided ity of the episode and time since surgery. Each patient kept

EFFECT OF CALCITRIOL OR ETIDRONATE ON TRANSPLANTATION BONE LOSS
567
a diary of all corticosteroids, including larger amounts given
during periods of rejection, and cyclosporin taken.
TABLE 2. CUMULATIVE TOTAL CORTICOSTEROID (g) AND
CYCLOSPORIN (g) USE
During the course of the study the form of cyclosporin
routinely used in the transplant clinic was switched from
Sandimmune (Novartis, Basel, Switzerland) to microemul-
sified cyclosporin (Neoral; Novartis). During the transition
period, cyclosporin levels and serum creatinine were mon-
Calcitriol
Etidronate
Reference
After 6 months
Corticosteroid*
Cyclosporin
7.6 Ϯ 3.4
59 Ϯ 21
8.7 Ϯ 3.1
67 Ϯ 17
10.4 Ϯ 4.5
60 Ϯ 22
itored and dosage was adjusted as necessary. Both study After 12 months
Corticosteroid†
10.2 Ϯ 3.9
105 Ϯ 47
11.1 Ϯ 3.3
111 Ϯ 29
13.5 Ϯ 4.6‡
122 Ϯ 45
groups were effected equally by this change in medication
Cyclosporin
After 24 months
Corticosteroid†
Cyclosporin
formulation.
14.3 Ϯ 4.7
182 Ϯ 78
15.6 Ϯ 4.2
187 Ϯ 51
19.0 Ϯ 4.7‡
227 Ϯ 79
BMD
* NS ANOVA, p
ϭ
0.06; † significant difference between
BMD (g/cm2) of the lumbar spine (L2–L4), femoral neck,
and total body was measured using dual-energy X-ray ab-
sorptiometry (DXA) (Lunar DPX-L; LUNAR Corp., Mad-
ison, WI, USA). Measurements in duplicate were made on
average 15 Ϯ 6 days (mean Ϯ SD) after transplantation
(baseline) and at 6 months and 12 months after transplan-
tation. Two years after transplantation a single follow-up
measurement was made at each of these sites. Where du-
plicate measurements were available, the mean was used for
analyses. One patient from the calcitriol group (193 cm tall)
could not have a whole body scan.
The reference group had single BMD measurements of
the lumbar spine and total body at baseline, 6 months, 12
months, and 2 years after transplantation on the same DXA
machine as the study subjects.
Symptomatic fractures confirmed by radiographs were
recorded at 6, 12, and 24 months. However, surveillance
spine radiographs were not performed because previous
data from the unit suggested there would be insufficient
power to detect a difference between groups in fracture
incidence.
groups, one-way ANOVA ( p Յ 0.03); ‡ significantly different to
calcitriol ( p Ͻ 0.05).
Statistical analysis
Differences between treatment groups (“group”) in post-
transplantation bone loss were determined using repeated
measures analysis of covariance (ANCOVA) using baseline
BMD as the covariate. In the first instance only the two
treatment groups were entered into the model with baseline,
6-month, and 12-month BMD data as the within (“time”)
factor.
To obtain an indication of whether the treatments atten-
uated bone loss and whether there was a carryover protec-
tive effect, secondary models including both treatment
groups plus the historical reference data also were gener-
ated. These one-way ANCOVA models included baseline
BMD as a covariate and percent change in BMD at each
time point, including 2 years, as the dependent variables.
Games-Howell post hoc tests were used for pairwise com-
parisons where the ANCOVA models showed significant
between-group effects.
Biochemistry
All data are presented as means and SDs unless otherwise
stated and a level of p Յ 0.05 was accepted as significant.
Because all comparisons were selected a priori, no adjust-
ments were made for multiple comparisons.
Blood for measurement of osteocalcin and calcium, fast-
ing second-void urine specimen for measurement of hy-
droxyproline excretion, and a 24-h urine sample for mea-
surement of urinary calcium were collected at baseline, 6
months, and 12 months from all patient groups. Baseline
blood samples were collected preoperatively in 31 patients
but postoperatively in 35 patients. Baseline urine samples
were all collected in the early postoperative period. In the
RESULTS
The two treatment groups were comparable at baseline
calcitriol treatment group, plasma and urinary calcium were (Table 1). However, the reference group included relatively
measured again at 4 weeks and 12 weeks. No patients fewer female patients and did not include any patients with
experienced hypercalcemia, but the calcitriol dose was re- lung transplantation. Adherence to therapy was excellent
and did not differ between groups (overall median and
interquartile range (IQR), 97 Ϯ 10%; Mann-Whitney U test,
tied-p ϭ 0.57). Twenty patients took every dose allocated.
duced prophylactically to 0.25 ␮g in 1 patient with persis-
tently elevated creatinine. Serum osteocalcin was deter-
mined as previously described(8); other serum chemistry
was determined by automated methods. Urinary calcium
was measured using a Corning Calcium Analyser 940
(CIBA Corning Diagnostics Ltd., Halstead, UK) and hy-
droxyproline was quantified using a Technicon autoanalyzer
Corticosteroid and cyclosporin doses
Cumulative doses of corticosteroid and cyclosporin (Ta-
ble 2) were comparable between the two treatment groups at
AAI (Technicon Instruments Corp., Tarrytown, NY, USA). 6 months, 12 months, and 2 years (p Ͼ 0.2). However, when
Hydroxyproline was expressed as a ratio relative to the the reference group was included in the analysis of variance
urinary creatinine.
(ANOVA) models, there was a significant difference in

568
HENDERSON ET AL.
TABLE 3. BMD (g/cm2) FOR CALCITRIOL AND ETIDRONATE
TREATMENT GROUPS AFTER TRANSPLANTATION
Baseline
6 months
12 months
Lumbar spine
Calcitriol (17)
1.15 Ϯ 0.20 1.10 Ϯ 0.20 1.07 Ϯ 0.20
Etidronate (17) 1.09 Ϯ 0.13 1.06 Ϯ 0.14 1.03 Ϯ 0.14
Femoral neck
Calcitriol (17)
0.90 Ϯ 0.15 0.86 Ϯ 0.16 0.84 Ϯ 0.14
Etidronate (17) 0.88 Ϯ 0.12 0.84 Ϯ 0.11 0.82 Ϯ 0.11
Whole body
Calcitriol (16)
1.18 Ϯ 0.12 1.16 Ϯ 0.13 1.15 Ϯ 0.13
Etidronate (17) 1.16 Ϯ 0.08 1.15 Ϯ 0.09 1.14 Ϯ 0.08
Values mean Ϯ SD, numbers in each group in parentheses.
cumulative corticosteroid dose at 12 months and 2 years.
The reference group had the greatest cumulative corticoste-
roid dose but the only significant pairwise difference was
between the calcitriol and reference groups (Games-Howell
post hoc test, p Ͻ 0.05) at both time points.
BMD
There were no significant differences between groups in
baseline BMD at any site (p Ͼ 0.07; Table 1). Mean T
scores did not differ between groups (p Ͼ 0.2). For all
groups combined, they were Ϫ0.82 for the lumbar spine,
Ϫ0.53 for the whole body, and Ϫ1.64 for the femoral neck;
the latter was only determined in the calcitriol and eti-
dronate groups.
Change in BMD during the first 12 months after trans-
plantation did not differ between the calcitriol and eti-
dronate treatment groups at any site and there was a signif-
icant decrease during the 12-month study period in lumbar
spine (group x time, p ϭ 0.41; time, p ϭ 0.0001), femoral
neck (group x time, p ϭ 0.85; time, p ϭ 0.0001), and whole
body BMD (group x time, p ϭ 0.41; time, p ϭ 0.0001;
Table 3).
In the lumbar spine, the reference group lost significantly
more bone during the first 6 months than either of the two
treatment groups (Fig. 1). However, both treatment groups
lost a significant amount of bone during the intervention
phase of the study (time, p ϭ 0.001). There also was a
significant difference between groups at 12 months and 2
years (one-way ANCOVA, p Յ 0.05). Post hoc analysis
showed that at both 12 months and 2 years, the only sig-
nificant pairwise difference was between the etidronate
group and the reference group (Games-Howell, p Յ 0.05).
Percentage change in whole body BMD did not differ
between groups at any time point (p Ͼ 0.34). Femoral neck
FIG. 1. Percentage change from baseline BMD of (A) lumbar spine
(LS), (B) whole body (WB), and (C) femoral neck (FN). Data shown
from all surviving patients at each time point. Number of patients is
indicated at the base of each bar. Error bars indicate SD.
BMD was not measured in the reference group so these Fractures
comparisons cannot be made for this site. Percentage
Six patients had symptomatic fractures during the 2-year
change in femoral neck BMD for the two treatment groups
is shown in Fig. 1. Addition of corticosteroid dose as a
covariate did not change the result of these models and there
was no relationship between corticosteroid or cyclosporin
dose and rate of change in BMD at any of the sites mea-
sured.
study period, 3 patients from the etidronate group (all ver-
tebral), 2 patients from the calcitriol group (one neck of
femur, one shaft of humerus), and 1 patient from the refer-
ence group (vertebral). These numbers did not differ signif-
icantly between groups.

EFFECT OF CALCITRIOL OR ETIDRONATE ON TRANSPLANTATION BONE LOSS
569
TABLE 4. BIOCHEMICAL INDICES
Baseline
6 months
12 months
Plasma calcium (mmol/liter)
Calcitriol
Etidronate
Reference
2.29 Ϯ 0.10
2.30 Ϯ 0.16
2.33 Ϯ 0.36
2.34 Ϯ 0.06
2.31 Ϯ 0.11
2.38 Ϯ 0.13
2.29 Ϯ 0.06
2.35 Ϯ 0.07
2.39 Ϯ 0.10
Serum creatinine (mmol/liter)
Calcitriol
Etidronate
Reference
0.11 Ϯ 0.05
0.11 Ϯ 0.04
0.11 Ϯ 0.03
0.13 Ϯ 0.03
0.13 Ϯ 0.04
0.12 Ϯ 0.03
0.13 Ϯ 0.03
0.14 Ϯ 0.04
0.12 Ϯ 0.02
Serum osteocalcin (␮g/liter)
Patients with preoperative serum collection
Calcitriol
8.2 Ϯ 4.7
7.6 Ϯ 4.3
12.3 Ϯ 7.3
7.2 Ϯ 4.9
18.6 Ϯ 8.6
13.1 Ϯ 7.3
Etidronate
Patients with postoperative serum collectiona
Calcitriol
Etidronate
Reference
2.6 Ϯ 3.3
1.0 Ϯ 0
2.2 Ϯ 2.7
13.4 Ϯ 9.2
7.8 Ϯ 5.1
13.0 Ϯ 11.8
16.5 Ϯ 8.3
13.1 Ϯ 7.9
15.7 Ϯ 10.8
Urinary hydroxyproline/creatinine (␮mol/mmol)
Calcitriol
Etidronate
Reference
27.1 Ϯ 15.3
24.4 Ϯ 8.6
32.0 Ϯ 13.8
17.9 Ϯ 7.6
18.3 Ϯ 9.5
20.4 Ϯ 10.1
16.8 Ϯ 5.5
15.7 Ϯ 6.8
14.6 Ϯ 4.5
24-h Urinary calcium (mmol/24 h)
Calcitriol
Etidronate
4.0 Ϯ 2.4
2.9 Ϯ 1.5
2.6 Ϯ 2.0
5.6 Ϯ 2.9b
3.8 Ϯ 1.8
2.9 Ϯ 1.7
4.0 Ϯ 2.2
3.3 Ϯ 1.5
3.2 Ϯ 1.8
Reference
Values mean Ϯ SD.
a Six month and 12 month data include all patients.
b Significantly greater than reference group.
purposes, these were assigned the lowest detectable value of
1.0. Among the patients who had samples taken preopera-
tively, there was a significant increase in serum osteocalcin
during the study period (time, p ϭ 0.002) but this change
did not differ significantly between the two treatment
groups (group x time, p ϭ 0.31). Although serum osteocal-
cin values appeared to be lower in the etidronate group,
particularly at 6 months, between group differences were
not significant (p ϭ 0.16 and p ϭ 0.51 for 6 months and 12
months, respectively).
Biochemistry data
Serum calcium did not differ between groups or change
significantly during the course of the study (group x time,
p ϭ 0.78; time, p ϭ 0.76) and no patients developed
hypercalcemia. Serum creatinine did not differ between
groups but increased significantly in all groups during the
study (group x time, p ϭ 0.93; time, p ϭ 0.02). Urinary
hydroxyproline/creatinine values (measured soon after
transplantation) were increased in all groups at baseline but
decreased significantly in all groups by 12 months (time,
p ϭ 0.0001). Overall, patients from the calcitriol group had
higher 24-h urinary calcium levels (group, p ϭ 0.049) but
post hoc analyses found the difference between groups was
only significant at the end of the 6-month period of treat-
ment (p Ͻ 0.05 for calcitriol vs. reference groups). After 6
DISCUSSION
This randomized, prospective study shows that low-dose
months of treatment, 7 patients had hypercalciuria, 6 pa- calcitriol and cyclical etidronate for the first 6 months after
tients from the calcitriol group and 1 patient from the heart or lung transplantation are equivalent in protection
etidronate group. Changes with time were similar between against posttransplantation bone loss. However, although
groups and not significant (group x time, p ϭ 0.32; time, these therapies attenuated the rate of bone loss, neither was
p ϭ 0.08).
completely effective.
Five patients from each of the treatment groups and all
It had been anticipated that, because the most rapid bone
reference patients had baseline blood for serum osteocalcin loss occurs during the first 6–12 months after transplanta-
collected postoperatively (6–24 days). As expected, serum tion,(2,3) preventative therapy during this time would confer
osteocalcin was reduced immediately posttransplantation at least some long-term skeletal advantage. However, com-
(Table 4). Of the 35 patients who had blood for osteocalcin parison between the two treatment groups and the reference
collected postoperatively, 21 patients had levels below the group at 12 months and 2 years after transplantation indi-
detectable range of the assay. For analysis and presentation cated that treatment with calcitriol did not confer a signif-

570
HENDERSON ET AL.
icant carryover protective effect after it was ceased. Al- in a group of patients taking lower doses of glucocorticoids
though the change in BMD did not differ significantly
between the two treatment groups at any of the sites mea-
sured, the group treated with etidronate lost somewhat less
bone than the calcitriol group at the spine and in the whole
body, such that at 12 months and 2 years, bone loss in the
etidronate group was significantly less than the reference
group. However, bone loss at the femoral neck was similar
in both treatment groups and no data from the reference
group were available for comparison.
for a variety of mainly chronic rheumatic disorders, bone
markers decreased and BMD increased in response to alen-
dronate, a more potent bisphosphonate.(12) This may relate
to the lower doses of glucocorticoids or the greater potency
of alendronate. Moreover, cyclosporin, which is used in
transplant recipients, is known to increase bone turnover
(e.g., osteocalcin), which is associated with greater bone
loss.(13)
The optimal duration of therapy is yet to be established.
To date, most studies of osteoporosis prevention in trans-
plant recipients have been of 1 year or 2 years duration, and
few have been designed to evaluate the efficacy of limited
duration therapy. The results of this study suggest that the
pattern of initially rapid and then slower bone loss can be
modified and that net bone loss may be preventable. There
was some evidence of a protective carryover effect in the
etidronate group consistent with its adsorption to bone min-
eral surfaces and long half-life. It remains to be seen
whether more potent bisphosphonates can abrogate the bone
loss.
Another recent randomized study of heart and lung trans-
plant recipients compared treatment with calcitriol plus cal-
cium supplementation taken for 1 year or 2 years with
calcium supplementation alone.(14) In that study overall
change in femoral neck BMD at 2 years was Ϫ5% for
patients treated with calcitriol throughout, Ϫ7.4% for pa-
tients treated for only 1 year, and Ϫ8.2% for the control
group treated with calcium alone. Lumbar spine bone loss
was comparable in all groups. Interestingly, change in lum-
bar spine BMD in the calcium-supplemented control group
was only Ϫ3% compared with Ϫ6.7% in the present study,
despite slightly higher cumulative corticosteroid doses, sug-
gesting that calcium supplementation alone may attenuate
posttransplantation bone loss in the lumbar spine.
An important, but ethically unavoidable, limitation of this
study was the absence of a randomized placebo-control
group for comparison. Although the historical reference
group was comparable with the treated groups in many
ways, the reference group included fewer women and no
lung transplant recipients and, because females and lung
transplant recipients are recognized to be more at risk of
osteoporosis, this difference may have biased the study
against identifying a treatment benefit. However, it could be
proposed that an optimal osteoporosis treatment regimen
should be sufficient to prevent bone loss in all transplant
recipients. On these grounds, neither of the therapies was
adequately effective.
A few studies of similar design have been published
subsequent to the commencement of this study. One ran-
domized study of 41 heart transplant recipients that com-
pared cyclical etidronate with alfacalcidol plus calcium
found alfacalcidol therapy was more effective than cyclical
etidronate but that neither therapy completely prevented
bone loss.(9) In fact, despite ongoing treatment and lower
cumulative doses of corticosteroids, bone loss in the patients
from that study treated with cyclical etidronate consistently
exceeded the bone loss observed in analogous patients in the
present study (Ϫ7.7% vs. Ϫ2.9% at 6 months and Ϫ10.3%
vs. Ϫ6.3% at 12 months for lumbar spine). By contrast, the
patients treated with the active vitamin D compound expe-
rienced bone loss comparable with that in the present study
(Ϫ4.6% vs. Ϫ4.2% at 6 months and Ϫ7.0% vs. Ϫ6.7% at
12 months for lumbar spine). The cyclical etidronate regi-
men in that study was described as calcium carbonate for 11
weeks followed by etidronate for 2 weeks, suggesting that
etidronate therapy was not commenced until approximately
3 months after transplantation. Patients in the present study
were administered the etidronate component of the cycle
first so the antiresorptive effects were active while immu-
nosuppressive therapy was at its peak. This difference in
etidronate-calcium regimen could explain the major differ-
ence in effect and suggests that bisphosphonate therapy
should be commenced early in the posttransplantation pe-
riod to be most effective.
Another observational study of combined therapy(10)
comprising cyclical etidronate plus alfacalcidol in liver
transplant recipients reported median decreases of 6% in
lumbar spine and 7% in hip BMD at 12 months. These
changes are comparable with mean changes of Ϫ5.4% in
lumbar spine BMD and Ϫ6.3% in femoral neck BMD
observed in the present study. However, heart transplant
recipients treated with a more complex regimen including a
single intravenous infusion of pamidronate, followed by
cyclic etidronate plus a small dose of calcitriol, lost little
bone and had fewer fractures than a historical control
group.(11) Although the patients treated with the more com-
plex regimen did have a lower cumulative dose of cortico-
steroids than the treated patients in the present study (6.1 Ϯ
0.7 g vs. 8.2 Ϯ 0.5 g, mean Ϯ SEM, respectively at 6
months), adjustment for a cumulative corticosteroid dose
did not change the result of ANOVA models in either study.
Temporal changes in biochemical markers of bone turn-
over did not differ between the treated groups and the
reference group in the present study, again suggesting that
neither therapy in the doses prescribed in this study was
adequate to suppress the changes in bone turnover elicited
Both cyclical etidronate and relatively low-dose calcitriol
were partially effective against bone loss after heart or lung
transplantation. However, because neither provided com-
plete or prolonged protection, regimens that are more effec-
tive should be sought. Although etidronate did provide
some protective carryover effect, the data suggest that pre-
ventive treatment for 6 months after transplantation is not
sufficient to provide optimal protection against medium-
term bone loss. More effective and longer-term regimens
should be investigated to determine the optimal type and
by the combined immunosuppressive therapy. By contrast, duration of prophylactic therapy.

EFFECT OF CALCITRIOL OR ETIDRONATE ON TRANSPLANTATION BONE LOSS
571
9. Van Cleemput J, Daenen W, Guesens P, Dequeker J, Van de
Werf F, Vanhaecke J 1996 Prevention of bone loss in cardiac
transplant recipients. Transplantation 61:1495–1499.
10. Riemens SC, Oostdijk A, van Doormaal JJ, Thijn CJP, Drent
G, Piers DA, Groen EWJ, Meerman L, Slooff MJH, Haagsma
EB 1996 Bone loss after liver transplantation is not prevented
by cyclical etidronate, calcium and alfacalcidol. Osteoporos
Int 6:213–218.
ACKNOWLEDGMENTS
We acknowledge assistance with data collection from
study coordinators and densitometry technicians S. Hunt, G.
Marshall, E. Ball, and V. Wills. Study medications were
provided by Roche Australia, Pharmacia and Upjohn Pty
Limited, and Procter & Gamble Australia Pty, Ltd.
11. Shane E, Rodino MA, McMahon DJ, Addesso V, Staron RB,
Seibel MJ, Mancini D, Michler RE, Lo SH 1998 Prevention of
bone loss after heart transplantation with antiresorptive ther-
apy: A pilot study. J Heart Lung Transplant 17:1089–1096.
12. Saag KG, Emkey R, Schnitzer TJ, Brown JP, Hawkins F,
Goemaere S, Thamsborg G, Liberman UA, Delmas PD, Mal-
ice MP, Czachur M, Daifotis AG 1998 Alendronate for the
prevention and treatment of glucocorticoid-induced osteopo-
rosis. Glucocorticoid-Induced Osteoporosis Intervention Study
Group. N Engl J Med 339:292–299.
13. Thie´baud D, Krieg MA, Gillard-Berguer D, Jacquet AF, Goy
JJ, Burckhardt P 1996 Cyclosporine induces high bone turn-
over and may contribute to bone loss after heart transplanta-
tion. Eur J Clin Invest 26:549–555.
14. Sambrook P, Henderson N, Keogh A, Macdonald P, Glanville
A, Spratt P, Bergin P, Ebeling P, Eisman J 2000 Effect of
calcitriol on bone loss after cardiac or lung transplantation.
J Bone Miner Res 15:1818–1824.
REFERENCES
1. Shane E, Rivas M, Staron R, Silverberg S, Seibel M, Kuiper J,
Mancini D, Addesso V, Michler R, Factor-Litvak P 1996
Fracture after cardiac transplantation: A prospective longitu-
dinal study. J Clin Endocrinol Metab 81:1740–1746.
2. Henderson NK, Sambrook PN, Kelly PJ, Macdonald P, Keogh
AM, Spratt P, Eisman JA 1995 Bone mineral loss and recovery
after cardiac transplantation. Lancet 346:905 (letter).
3. Shane E, Rivas M, McMahon DJ, Staron RB, Silverberg SJ,
Seibel MJ, Mancini D, Michler RE, Aaronson K, Addesso V,
Lo SH 1997 Bone loss and turnover after cardiac transplanta-
tion. J Clin Endocrinol Metab 82:1497–1506.
4. Adachi JD, Bensen WG, Brown J, Hanley D, Hodsman A,
Josse R, Kendler DL, Lentle B, Olszynski W, Ste-Marie L-G,
Tenenhouse A, Chines AA 1997 Intermittent etidronate ther-
apy to prevent corticosteroid-induced osteoporosis. N Engl
J Med 337:382–387.
5. Sambrook P, Birmingham J, Kelly P, Kempler S, Nguyen T,
Pocock N, Eisman J 1993 Prevention of corticosteroid osteo-
porosis. A comparison of calcium, calcitriol, and calcitonin.
N Engl J Med 328:1747–52.
6. Signorini DF, Leung O, Simes RJ, Beller E, Gebski VJ 1993
Dynamic balanced randomization for clinical trials. Stat Med
12:2343–2350.
Address reprint requests to:
Dr. Kathy Henderson
School of Physiotherapy
Curtin University of Technology
Selby Street
7. Angus RM, Sambrook PN, Pocock NA, Eisman JA 1989 A
simple method for assessing calcium intake in Caucasian
women. J Am Diet Assoc 89:209–214.
Shenton Park WA 6008, Australia
8. Kelly PJ, Pocock NA, Sambrook PN, Eisman JA 1989 Age
and menopause-related changes in indices of bone turnover. Received in original form March 13, 2000; in revised form Sep-
J Clin Endocrinol Metab 69:1160–1165. tember 28, 2000; accepted October 31, 2000.