A 1-Year Randomized, Double-Blind, Placebo-Controlled Study of Intravenous Ibandronate on Bone Loss Following Renal Transplantation


Knut T. Smerud, knut.smerud@smerud.com


The clinical profile of ibandronate as add-on to calcitriol and calcium was studied in this double-blind, placebo-controlled trial of 129 renal transplant recipients with early stable renal function (≤ 28 days posttransplantation, GFR ≥ 30 mL/min). Patients were randomized to receive i.v. ibandronate 3 mg or i.v. placebo every 3 months for 12 months on top of oral calcitriol 0.25 mcg/day and calcium 500 mg b.i.d. At baseline, 10 weeks and 12 months bone mineral density (BMD) and biochemical markers of bone turnover were measured. The primary endpoint, relative change in BMD for the lumbar spine from baseline to 12 months was not different, +1.5% for ibandronate versus +0.5% for placebo (p = 0.28). Ibandronate demonstrated a significant improvement of BMD in total femur, +1.3% versus −0.5% (p = 0.01) and in the ultradistal radius, +0.6% versus −1.9% (p = 0.039). Bone formation markers were reduced by ibandronate, whereas the bone resorption marker, NTX, was reduced in both groups. Calcium and calcitriol supplementation alone showed an excellent efficacy and safety profile, virtually maintaining BMD without any loss over 12 months after renal transplantation, whereas adding ibandronate significantly improved BMD in total femur and ultradistal radius, and also suppressed biomarkers of bone turnover. Ibandronate was also well tolerated.


analysis of covariance


bone-specific alkaline phosphatase


twice daily


bone mineral density




dual-energy X-ray absorptiometry


ethylenediaminetetraacetic acid


estimated glomerular filtration rate


enzyme immunoassay


end-stage renal disease


good clinical practice


International Conference of Harmonization


intact Parathyroid hormone






Kidney Disease: Improving Global Outcomes






Markov Chain Monte Carlo


N-terminal telopeptide of type-I collagen


propeptide of type I procollagen


parathyroid hormone




serious adverse event


standard deviation


standard error of the mean


selective estrogen receptor modulator


Bone disease is a common complication in patients with end-stage renal disease (ESRD), both prior to and after renal transplantation (1–4). Following renal transplantation, most of the bone loss occurs within the first 6 months, especially from areas rich in trabecular bone (5–8). As much as 10% loss of bone mass has been reported during the first year after renal transplantation (9–11). Increased posttransplantation bone loss may be explained by several factors such as established bone loss at the time of renal transplantation, effects of immunosuppressive therapy on the already compromised skeleton after transplantation and persisting chronic renal failure (generally in stage ≥3) also after transplantation (12).

Interventions for preventing and treating bone disease following renal transplantation have been reviewed in several recent papers (13–15), and the Kidney Disease: Improving Global Outcomes (KDIGO) recently published practice guidelines (16). Intervention studies have not shown a significant reduction of fracture risk. Most of these studies are however suboptimally designed with small sample sizes and inclusion of patients long time after transplantation (14). Thus, adequately powered trials are crucial to evaluate the effect of antiresorptive therapy for prevention of bone loss in this population.

Bisphosphonates are primarily excreted via the kidney and clearance is consequently reduced with reduced renal function, giving a higher systemic drug exposure, even if this may not necessarily translate into better skeletal effect. Most bisphosphonate prescription labelings (summary of product characteristics) advise against use in severe renal impairment because of lack of clinical safety data. Treatment appears to be well tolerated in renal transplant recipients, as no significant adverse effects have been reported in randomized controlled trials (14). However, bisphosphonates may have the potential to induce low bone turnover and adynamic bone disease in renal transplant recipients (17).

Given the paucity of trials on efficacy and safety of bisphosphonates in renal transplant patients we designed a single center, prospective, randomized, double-blind, placebo-controlled study. These patients normally have a large intake of tablets every day, so to avoid potential problems with oral drug interactions and lack of compliance, we chose to use a long-acting parenteral bisphosphonate. Ibandronate was administered intravenously every 3 months and compared with placebo, i.v. isotonic saline. In addition all patients received 0.25 mcg calcitriol once daily and 500 mg calcium twice daily.

The main aim was to study relative changes in lumbar spine (L2-L4) BMD from baseline to 12 months between the treatment groups. Secondary aims were to study loss of BMD in other compartments and changes in biochemical markers of bone turnover. In addition thorough assessment of the general and renal-specific tolerability of ibandronate as add-on treatment to oral calcium and calcitriol during the first year following renal transplantation was addressed.

Materials and Methods


Altogether 525 patients received a renal transplant at our center during the inclusion period. The patient disposition is shown in Figure 1. One hundred and twenty nine patients gave written informed consent and were randomized to receive study drugs or matching placebo. Inclusion criteria were patients of either sex over 18 years receiving a single live or deceased renal transplant. Exclusion criteria were previous parathyroidectomy and use of bisphosphonates during the last year. Not acceptable concomitant medication also included sodium fluoride, calcitonin, strontium, parathyroid hormone (PTH), selective estrogen receptor modulators (SERMs), growth hormone or anabolic steroids. Pregnant or lactating women and those of childbearing potential not using adequate contraception were also excluded as were patients with hypersensitivity to bisphosphonates. Patients were not eligible for randomization until they had obtained a stable and adequate graft function with an estimated glomerular filtration rate (eGFR; according to Cockcroft–Gault (18)) of at least 30 mL/min and had a total plasma calcium below 2.55 mmol/L during the last 2 weeks. Patients were to be included in the study no later than 4 weeks after transplantation. Adynamic bone disease was considered an exclusion criterion, but no bone biopsies were taken in this study. There was no selection bias in terms of differences in recipient age or gender distribution in our population compared to those patients who were not recruited into the trial for logistic reasons (data not shown).

Figure 1.

Patient flow diagram. LTFU = lost to follow-up.

Study design

Patients were randomized according to a computer-generated list of random numbers to allocate patients in a 1:1 ratio using a block size of 8 to receive either ibandronate 3 mg i.v. given as an infusion over 3 min every 3 months, or to receive placebo (isotonic saline water with matching appearance to the active medication); both regimens lasted 12 months. Eligible and consenting patients were sequentially allocated a randomization number, and given treatment from the corresponding vials at the study site. All study personnel were blinded to the allocation for the duration of the study. The code was not broken for any patient until the complete data set had been entered into the study database and the database had been locked.

All patients received underlying treatment in the form of daily oral intake of 0.25 mcg/day of calcitriol (Rocatrol®, Roche) and calcium carbonate 1260 mg, equivalent to 500 mg of calcium twice daily (Weifa-Kalsium®, Weifa, Oslo, Norway). Standard immunosuppressive protocol consisted of a calcineurin inhibitor (cyclosporine or tacrolimus), prednisolone and mycophenolate mofetil (MMF). All patients were given i.v. basiliximab induction (20 mg on day 0 and 4 relative to transplantation). One-year cumulative prednisolone exposure was similar in the two groups, 5435 (±2321) mg for the ibandronate group and 5315 (±1837) mg for the placebo group. Biopsy-verified rejection episodes were treated with repeated i.v. boluses of methylprednisolone, in a total dose of at least 1250 mg. Steroid-resistant rejections were treated with antithymocyte globulin (Thymoglobuline®, Genzyme).

According to standard local practice, the patients were followed at the transplant unit for 10–12 weeks after transplantation, and then admitted to further follow-up by a local nephrologist. The first two i.v. ibandronate/placebo study doses were given while the patients were in our center, the two last quarterly doses were given by the local nephrologist, who was also responsible for assessing any change in concomitant medication and to actively ask for adverse events in this period. All patients who had not discontinued the study were seen at Oslo University Hospital (Rikshospitalet) also for the final (month 12) assessment visit.

Dual-energy X-ray absorptiometry (DXA) was used to measure BMD of the whole body as well as selected parts of the skeleton such as the lumbar spine (L2 to L4), the left and right proximal femur, the nondominant forearm (ultradistal as well as the proximal third of radius). For the DXA examination, a narrow fan-beam GE Lunar Prodigy densitometer (GE Medical Systems, Lunar Corp., Madison, WI, USA) with software versions 11.20, 12.10 and 12.20 (from the same manufacturer) was used throughout the entire study.

A standard lateral x-ray of the thoraco-lumbar spine was performed at inclusion and after 1 year at the Department of Radiology, Oslo University Hospital, Rikshospitalet, Norway. Vertebral fractures were defined as a morphometric compression of at least 20%, or a decrease in height of 4 mm or more in two or more vertebrae between the exams. Patients were also asked at every clinic visit if they had any fracture since the last visit. Any fracture was classified as a low energy or traumatic fracture.


Standard biochemical and hematological safety variables were analyzed in the accredited laboratory at Oslo University Hospital, Rikshospitalet, Norway. Fasting blood samples were collected on the days of the DXA measurements. To standardize collection, serum and EDTA plasma samples for bone-related biochemistry were taken between 0800 h and 1000 h after an overnight fast and isolated and stored at −80°C for later, simultaneous analysis. All samples to be compared were analyzed in the same run to eliminate interassay variation.

Serum levels of intact N-terminal propeptide of type I procollagen (PINP) were measured by radioimmunoassay (RIA), (Orion Diagnostica Oy, Espoo, Finland). Serum concentrations of bone-specific alkaline phosphatase (bALP) and osteocalcin levels were analyzed by an enzyme immunoassay (EIA) provided by Quidel Corp., San Diego, CA, USA. Osteocalcin assay measures intact osteocalcin 1–49. Degradation products of the N-terminal telopeptides of type-I collagen (NTX-1) were measured with an EIA (Serum Osteomark®, Wampole Laboratories Inc., Princeton, NJ, USA). Plasma intact PTH (iPTH) was analyzed from EDTA-plasma samples, and we used the third-generation assay (Scantibodies Laboratory Inc, Santee, CA, USA). Serum 25-hydroxy-calcitriol levels (25(OH)D) were assessed by competitive RIA using a kit from DiaSorin Inc., Stillwater, MN, USA. The intra- and interassay coefficients of variation were < 10% for all assays.

Safety and tolerability measures were assessed from adverse events reported spontaneously by the patient or observed by the investigator, and classified by seriousness (unexpected, nonserious or serious), severity (mild, moderate or severe) and relationship to study medication (unrelated, unlikely, possible, probable or not assessable). Of specific interest were posttransplantation complications such as treated and biopsy-verified transplant rejections, any cytomegalovirus infections, as well as hypercalcemia leading to withdrawal or reduction in the dose of oral calcium or calcitriol.

Statistical analysis

Sample size:  Based on an expected difference of 4% (SD 6%) on the primary endpoint between the ibandronate (expected to lose a mean of 3% in BMD) and placebo group (expected to lose 7%), a total of 49 patients per group was required to reach a power of 90%. The calculation was based on an independent samples t-test with two-sided 5% significance level. With an expected dropout rate of up to 25%, we aimed to randomize about 130 patients. These calculations were performed using the software SamplePower2™.

Statistical methods:  All analyses and background summarizations were done on the intention-to-treat (ITT) population, consisting of all randomized patients. All applied statistical methods were approved by the study steering committee before unblinding the data. The primary endpoint with respect to efficacy was the difference in percent lumbar spine (L2-L4) BMD change from baseline to 12 months between the treatment groups. Analyses of BMD data at 3 and 12 months were based on analysis of covariance (ANCOVA) models adjusting for baseline. Bone biomarker analyses were based on log(e) transformed values using ANCOVA models adjusted for baseline. These analyses provided an estimate of the relative treatment difference in geometric mean. Missing values were handled using the multiple imputation technique. We used 20 imputations for each missing value from a multivariate normal distribution combining Markov Chain Monte Carlo (MCMC) and regression methods, and combining the results in accordance with standard methods (19). For comparing adverse events the chi-square or Fisher's exact test was employed. We used a significance level of 5%, two-sided p-values and 95% confidence limits. All models were checked for assumption violations. All calculations were performed using software SPSS SamplePower version 2 (SPSS Inc., Chicago, IL, USA).


The protocol and associated documents were approved by the Regional Ethics Committee for Medical Research in Southern Norway, the Norwegian Directorate of Health, the Norwegian Data Inspectorate as well as the Norwegian Medicines Control Agency. The trial is registered with ClinicalTrials.gov, number NCT00423384, and conducted under the EudraCT number 2006-003884-30. The study was conducted in compliance with the International Conference of Harmonisation (ICH) guidelines for good clinical practice (GCP). All participants provided written informed consent.


Between January 2007 and May 2009, 129 patients were included (ibandronate n = 66/placebo n = 63, Figure 1). All participants received at least two doses of study medication covering the total dose for the first 6 months of the study. The first three doses were taken by 94%, and all four doses by 89% of the patients in the ibandronate group versus 87% and 86%, respectively for the placebo group. A total of 16 patients (7 in the ibandronate group, 9 in the placebo group) did not receive all infusions. Ten patients stopped study medication early but were followed by the intention-to-treat principle and all data were obtained. Six patients withdrew prematurely from the study (five in the placebo group). Two patients were lost to follow-up (one in each group). A single patient (placebo) withdrew due to early fractures in the thoracic spine and was consequently prescribed regular bisphosphonate treatment. Table 1 shows the demographics and baseline characteristics of our study population. Overall the groups appeared to have fairly similar demographics and baseline characteristics.

Table 1.  Demographics and baseline characteristics
   Treatment group
Ibandronate N = 66Placebo N = 63Total N = 129
  1. SD = standard deviation; BMD = bone mineral density; Tx = transplantation; Rnd = randomization (and 1st study dose).

GenderMaleN (%)48 (72.7%)51 (81.0%)99 (76.7%)
 FemaleN (%)18 (27.3%)12 (19.0%)30 (23.3%)
AgeYearsMean (SD)50.2 (13.5)52.6 (14.0)51.4 (13.8)
Previous renal transplantation(s) N (%)12 (18.2%)7 (11.1%)19 (14.7%)
Prestudy useCalcitriolN (%)37 (56.1%)51 (81.0%)88 (68.2%)
 CalciumN (%)5 (7.6%)5 (7.9%)10 (7.8%)
BMDLumbar spine L2-L4g/cm2 (SD)1.175 (0.175)1.193 (0.167)1.184 (0.171)
 T-score (SD) −0.56 (1.38)−0.44 (1.35)−0.50 (1.36)
 Z-score (SD) −0.25 (1.29)−0.23 (1.52)−0.24 (1.40)
FracturesClinically reportedN (%)0 (0%)0 (0.0%)0 (0%)
 Determined by LS X-rayN (%)2 (3.0%)4 (6.3%)6 (4.7%)
Time Tx-to-RndDaysMean (SD)17.7 (6.4)19.3 (7.1)18.5 (6.8)
Cumulative prednisolone Tx-to-RndDose-equivalent in mgMean (SD)1087 (335)1174 (382)1129 (360)
eGFRmL/min/1.73 m2Mean (SD)64.45 (19.9)64.15 (18.2)64.30 (19.0)

Bone mineral density

For the primary endpoint, relative change in BMD in the lumbar spine (L2–L4) from baseline to 12 months, we found an increase in both groups, +1.5% for ibandronate and +0.5% for placebo, respectively (Figure 2A). However, this difference of 1.0% between groups in relative change was not significant (p = 0.33). For other compartments where BMD was measured (Table 2, Figure 2), we found a significant effect of ibandronate versus placebo.

Figure 2.

Change in BMD (in%) for ibandronate and placebo at lumbar spine L2–L4 (panel A), total femur (panel B), ultradistal radius (panel C), proximal third of radius (panel D) and total body (panel E). Data are represented as mean (±SEM).

Table 2.  Bone mineral density
CompartmentUnitTimepointIbandronate (Mean ± SD) n = 66Placebo (Mean ± SD) n = 63Est. treatment dif. (95% CI)p-Value
  1. SD = standard deviation; CI = confidence interval. *Significant at 0.05 level = not corrected for multiple testing.

Lumbar spineg/cm2Baseline1.175 (0.18)1.193 (0.17)  
(L2-L4) Week 101.172 (0.19)1.186 (0.18)  
  Week 521.194 (0.19)1.189 (0.16)  
  Δabs week 0–520.017 (0.06)0.004 (0.08)0.013 (−0.01,0.04)0.28
 T-scoreBaseline−0.56 (1.38)−0.44 (1.36)  
  Week 10−0.60 (1.48)−0.49 (1.42)  
  Week 52−0.42 (1.46)−0.46 (1.30)  
  Δabs week 0–520.13 (0.45)0.05 (0.62)0.067 (−0.13, 0.26)0.50
 Z-scoreBaseline−0.25 (1.29)−0.23 (1.52)  
  Week 10−0.32 (1.40)−0.32 (1.57)  
  Week 52−0.22 (1.36)−0.37 (1.46)  
  Δabs week 0–520.01 (0.43)−0.03 (0.68)0.043 (−0.16, 0.25)0.67
Total femurg/cm2Baseline0.911 (0.13)0.913 (0.16)  
  Week 100.908 (0.13)0.910 (0.15)  
  Week 520.925 (0.12)0.905 (0.14)  
  Δabs week 0–520.011 (0.04)−0.007 (0.04)0.017 (0.004, 0.030)0.013*
 T-scoreBaseline−1.19 (0.96)−1.23 (1.21)  
  Week 10−1.23 (0.96)−1.26 (1.17)  
  Week 52−1.09 (0.93)−1.32 (1.11)  
  Δabs week 0-520.07 (0.28)−0.07 (0.32)0.135 (0.03, 0.239)0.012*
 Z-scoreBaseline−0.74 (0.89)−0.81 (1.21)  
  Week 10−0.80 (0.87)−0.85 (1.18)  
  Week 52−0.70 (0.81)−0.96 (1.08)  
  Δabs week 0-520.01 (0.31)−0.10 (0.35)0.114 (0.004, 0.224)0.043*
Radius,g/cm2Baseline0.381 (0.07)0.394 (0.07)  
Ultradistal Week 100.377 (0.07)0.393 (0.07)  
  Week 520.384 (0.08)0.385 (0.07)  
  Δabs week 0-520.002 (0.03)−0.008 (0.03)0.010 (0.001, 0.019)0.039*
 T-scoreBaseline−0.75 (1.57)−0.49 (1.52)  
  Week 10−0.89 (1.60)−0.54 (1.50)  
  Week 52−0.72 (1.70)−0.78 (1.52)  
  Δabs week 0-520.01 (0.69)−0.24 (0.66)0.237 (−0.005, 0.478)0.055
 Z-scoreBaseline−0.46 (1.51)−0.12 (1.60)  
  Week 10−0.58 (1.56)−0.22 (1.56)  
  Week 52−0.38 (1.62)−0.43 (1.53)  
  Δabs week 0-520.06 (0.69)−0.22 (0.64)0.260 (0.022, 0.498)0.032*
Radius,g/cm2Baseline0.726 (0.10)0.738 (0.08)  
Proximal 1/3 Week 100.727 (0.10)0.743 (0.08)  
  Week 520.731 (0.10)0.739 (0.08)  
  Δabs week 0-520.005 (0.03)0.000 (0.03)0.004 (−0.005, 0.014)0.35
 T-scoreBaseline−0.68 (1.16)−0.62 (0.88)  
  Week 10−0.69 (1.11)−0.57 (0.94)  
  Week 52−0.64 (1.16)−0.63 (0.91)  
  Δabs week 0-520.04 (0.41)−0.01 (0.33)0.046 (−0.086, 0.178)0.49
 Z-scoreBaseline−0.37 (1.14)−0.25 (1.01)  
  Week 10−0.38 (1.10)−0.22 (1.04)  
  Week 52−0.30 (1.11)−0.28 (0.99)  
  Δabs week 0-520.07 (0.43)0.01 (0.37)0.046 (−0.093, 0.185)0.51
Total bodyg/cm2Baseline1.136 (0.11)1.153 (0.09)  
  Week 101.136 (0.11)1.145 (0.09)  
  Week 521.146 (0.10)1.151 (0.09)  
  Δabs week 0-520.008 (0.03)−0.001 (0.03)0.007 (−0.003, 0.182)0.17
 T-scoreBaseline−0.71 (1.28)−0.56 (1.08)  
  Week 10−0.72 (1.27)−0.67 (1.12)  
  Week 52−0.59 (1.17)−0.63 (1.08)  
  Δabs week 0-520.08 (0.40)−0.04 (0.34)0.098 (−0.003, 0.230)0.14
 Z-scoreBaseline−0.51 (1.21)−0.54 (1.15)  
  Week 10−0.58 (1.18)−0.69 (1.19)  
  Week 52−0.55 (1.04)−0.75 (1.11)  
  Δabs week 0-52−0.08 (0.46)−0.13 (0.40)0.057 (−0.091, 0.206)0.45

Other efficacy variables

There was a consistent reduction of bone turnover markers with ibandronate, both for bone formation markers and bone resorption markers. As shown in Table 3, ibandronate consistently reduced bone formation markers, whereas in the placebo group, there was no change or even an increase over the 12 months study period. Low-energy vertebral fractures, reported through a clinical questionnaire, were reported by two patients in the ibandronate group and one in the placebo group. Thoraco-lumbar x-ray could only verify one fracture in each of the study groups.

Table 3.  Biochemical measures of bone variables
BiomarkerUnitTimepointIbandronate (Mean ± SD)Placebo (Mean ± SD)Est. rel. dif. (95% CI)p-Value
  1. SD = standard deviation; CI = confidence interval. *Significant at 0.05 level, not corrected for multiple testing; Δabs= absolute difference between week 0 (baseline) and week 52; Est. rel. dif. = estimated relative difference (between study groups); BCE = bone collagen equivalent.

Type I procollagen (N-terminal) PINP
 μg/LBaseline74.3 (59.2)56.7 (42.0)  
  Week 1064.2 (60.2)56.2 (31.7)  
  Week 5259.4 (57.1)78.6 (48.7)  
  Δabs week 0-52−13.1 (56.4)21.6 (60.5)0.65 (0.51, 0.81)0.0003*
 U/LBaseline16.7 (29.0)13.2 (18.4)  
  Week 1012.7 (19.8)10.6 (15.5)  
  Week 529.7 (11.3)14.9 (19.9)  
  Δabs week 0-52−5.5 (21.5)1.1 (22.2)0.66 (0.53, 0.83)0.0004*
Bone-specific alkaline phosphatase (bALP)
 U/LBaseline37.8 (31.5)31.7 (12.3)  
  Week 1035.1 (27.9)34.3 (21.3)  
  Week 5227.9 (19.2)33.5 (15.7)  
  Δabs week 0-52−9.5 (19.0)1.7 (16.0)0.79 (0.69, 0.91)0.0011*
Collagen type 1 cross-linked N-telopeptide (NTX)
 nM BCEBaseline36.4 (41.1)30.2 (11.3)  
  Week 1029.3 (69.3)24.5 (9.1)  
  Week 5223.8 (19.7)24.4 (10.7)  
  Δabs week 0-52−12.5 (31.9)−6.2 (12.7)0.85 (0.72, 1.00)0.0558
 pg/mLBaseline79.6 (65.5)91.7 (64.1)  
  Week 1052.3 (44.1)48.5 (38.3)  
  Week 5256.1 (55.9)45.7 (23.1)  
  Δabs week 0–52−21.5 (35.8)−43.4 (54.5)1.14 (0.96, 1.35)0.14
25-OH vitamin D
 nMBaseline60.7 (27.1)59.4 (27.8)  
  Week 5265.9 (31.6)69.1 (36.3)  
  Δabs week 0–524.7 (22.1)9.1 (26.2)0.93 (0.83, 1.05)0.28

Safety and tolerability

The adverse event profile (Table 4) was similar in both groups, and the overall picture is that there were fewer and less serious adverse events (SAEs) reported for the ibandronate group. Especially, when analyzing how investigators assessed the relationship between adverse event and study drug, only adverse events occurring in the placebo group were considered possibly or probably related to study drugs (bone pain, hypercalcemia). Serious adverse events were reported by 33.3% and 58.7% of the patients for ibandronate and placebo, with a total of 29 and 71 individual SAEs, respectively. Three patients died during the study, all in the placebo group. No patients withdrew from the study because of adverse events. Biopsy-verified and treated transplant rejections occurred in 18 patients (27.3%) receiving ibandronate versus 22 (34.9%) on placebo. Cytomegalovirus (CMV) infections were reported for 14 (21.2%) and 21 (33.3%) patients, respectively.

Table 4.  Adverse events and renal safety
 Ibandronate (N = 66)Placebo (N = 63)p-Value
  1. Number (%) of subjects experienced any such event, unless otherwise specified, with p-value for difference between groups analyzed with a chi-square test (1Fishers exact test). 2Classified by the investigator as adverse event if serum creatinine was above normal lab range, and if the observation was clinically relevant. SD = standard deviation.

 Any adverse events51 (77.3%)59 (93.7%)NA
 Any serious adverse event22 (33.3%)37 (58.7%)NA
 Total number of serious adverse events2971NA
 Death0 (0.0%)3 (4.8%)0.1141
 Discontinuation from study due to adverse event0 (0.0%)0 (0.0%)NA
Other adverse events occurring in ≥ 4% of patients in either group
 Cytomegalovirus infection14 (21.2%)21 (33.3%)0.122
 Urinary tract infection13 (19.7%)12 (19.0%)0.926
 Lymphocele1 (1.5%)6 (9.5%)0.0581
 Chest pain1 (1.5%)3 (4.8%)0.3581
 Gastroenteritis1 (1.5%)6 (9.5%)0.0581
 Pneumonia2 (3.0%)3 (4.8%)0.6751
Renal safety
 Biopsy verified transplant rejections18 (27.3%)22 (34.9%)0.348
 Elevated serum creatinine25 (7.6%)10 (15.9%)0.142
 Hypercalcemia leading to withdrawal of calcium11 (16.7%)13 (20.6%)0.563
 Withdrawal of calcitriol8 (12.1%)9 (14.3%)0.716
 Mean (SD) change in serum creatinine from baseline to month 12 (mM)−2.1 (27.7)4.2 (21.6)0.151
Vertebral fractures
 Clinically reported2 (3.0%)1 (1.6%)1.0001
 X-ray verified1 (1.5%)1 (1.6%)1.0001


This study was not able to demonstrate any significant effect of adding ibandronate to calcium and calcitriol in terms of preventing loss of bone mineral density in the lumbar spine when treating for 12 months after renal transplantation. However, rather surprisingly, none of the study groups showed a decline in lumbar spine BMD over 12 months. Ibandronate added a significantly protective effect against bone loss in total femur and ultradistal radius, even if the small changes may implicate little clinical relevance. The study was not powered to determine any effect on fracture risk. Ibandronate appeared to be safe and well tolerated for 12 months treatment in renal transplant recipients. To our knowledge, this study is the largest, prospective, randomized, double-blind and controlled clinical trial of bisphosphonate treatment following renal transplantation.


Based on previously reported BMD loss of up to 10% during the first year after transplantation, we conservatively assumed that the placebo group (receiving calcium and calcitriol, only) would lose 7% of lumbar spine BMD over a year, and that a fair expectation for the ibandronate group would be a difference of at least 4%-points. Our results showed that both groups improved BMD during the study period (0.5% for placebo and 1.5% for ibandronate). More recent publications of prospective trials have indeed reported bone loss of only 0.1–5.7% in the lumbar spine (20–22).

We observed that our study population might have been somewhat ‘healthier’ than expected with respect to presence of bone disease at transplantation. At baseline, our population had a mean lumbar spine BMD Z-score of −0.24, whereas a recent previous, similarly sized cohort of renal transplant recipients from our own center presented with a Z-score of −0.46 (3). The modern immunosuppressive regimens with lower rejection rates and reduced cumulative glucocorticoid exposure may be another reason why we could not find a significant additional benefit of bisphosphonates above and beyond active vitamin D and calcium supplements. It could be speculated that bisphosphonate therapy was initiated too late, as most bone loss is to be expected within the first few weeks following transplantation, explained by an enlargement of the remodeling space rather than a depression of bone formation by steroids. However, since bisphosphonates currently are contraindicated with GFR below 30 mL/min very few patients may be eligible for immediate post transplant treatment. Moreover, our patients were randomized and treated only 18 days after transplantation on average (range 5–37 days).

Despite fairly good results with calcium and calcitriol alone also for other compartments, the present study showed that ibandronate had significant add-on effects and appeared not only to prevent bone loss, but indeed increased BMD during the study. There was a trend which may suggest a continued positive effect of ibandronate over time, and we therefore aim to follow-up this study population for long-term evaluation of bone health (protocol specified). Long-term prospective studies have demonstrated that the annual mean loss of BMD remains stable and ranges between 1% and 3% (23), which is a more pronounced loss than the normal population (24,25). The recent metaanalysis of Stein et al. (26) also supports the need for long-term fracture data from well-designed clinical trials, even if each trial by itself may not be powered to detect differences in fracture rates.

As DXA does not provide specific information on bone turnover, the BMD results should be interpreted together with clinical assessment, and bone turnover biomarkers. Despite some uncertainty of the value of BMD measurements in the evaluation of bone status in renal transplant recipients, the KDIGO guidelines for posttransplant bone disease recommend measurement of BMD in the first 3 months after renal transplantation if risk factors for osteoporosis are present (16).

For biochemical biomarkers, we observed a consistent and robust lowering for all bone formation markers (PINP, osteocalcin and bALP) for ibandronate as opposed to no change and even an increase for the control group. It appears that this change in bone formation markers occurs already within 10 weeks following initiation of therapy and continues to decline throughout the study period for ibandronate. Our findings are consistent with previous work from our group in that serum osteocalcin levels normally increase during the first 10 weeks after transplantation in patients not receiving bisphosphonates, calcium, or calcitriol (27). In agreement with other investigators, we also observed that the effect on the bone resorption marker (NTX) was delayed and reached borderline significance only after month 12, apparently consistent with slow recoupling of bone formation and resorption after transplantation (21).

In this study, ibandronate decreased biochemical markers of bone turnover, but the net result on BMD was positive only for total femur and distal radius. Our subgroup analyses, albeit with small numbers, demonstrated that patients with the higher levels of bone resorption markers (NTX) responded better to bisphosphonate therapy. Our findings are indirectly in accordance with studies in postmenopausal osteoporosis showing that patients with the largest reduction in bone resorption during the early period of bisphosphonate treatment had the most pronounced gain in bone mass (28).

Of particular interest was the impact of bisphosphonates on renal safety. Ibandronate demonstrated an excellent safety profile in our study; the type, frequency and severity of adverse events did not differ from the control group. There were no more transplant rejections nor were there more kidney function related adverse events in the ibandronate group. We found no differences in the mean serum creatinine levels at any of the time points, nor in the number of hypercalcemic episodes, or number of patients needing to reduce calcium or calcitriol dosing during the study. No patients dropped out of the study due to adverse events.


The study was well matched to the demographic background of the general Norwegian renal transplant population, with the exception of the somewhat better Z-score than previous cohorts from our center. We were surprised to learn that both study groups gained rather than lost BMD over 12 months, and in order to explain the results in the ‘placebo’ group, i.e. patients taking calcium and calcitriol alone, we would in retrospect have liked to follow a noninterventional control arm. With the current design, we were not able to determine whether the protection of BMD was due to effect of calcium and calcitriol, or if this protection would have occurred anyway. A recent metaanalysis (26) hinted that rates of bone loss from recent studies were much lower than those from decades ago, possibly due to a decrease in glucocorticoid use. Other plausible explanations include modern management of calcium homeostasis prior to transplantation, shorter duration of chronic renal failure and/or dialysis before transplantation, and possibly a generally better control of vitamin D status following transplantation. Our study was not designed to assess these impacts on the outcome.

Only some 60% of potentially eligible recipients were recruited. However, lack of selection in 40% was mainly due to staff availability and there was no difference in age or sex between the two populations.

Furthermore, we would ideally have followed the patients for more than 12 months, but due to the perceived possible risk of inducing adynamic bone disease in renal transplant recipients, our protocol was designed to treat these patients for only one year until the safety profile had been documented properly in this population. Because low bone turnover is a well-acknowledged complication after renal transplantation, the use of bisphosphonates in this setting has been questioned. Bisphosphonates reduce bone turnover and may therefore induce adynamic bone disease in patients at risk (12,17). Coco et al. (29) demonstrated that pamidronate combined with calcitriol and low-dose calcium preserved bone mass at 6 months, but also that all patients receiving pamidronate developed low bone turnover. It is a limitation of our study that we could not assess bone morphology.


In summary, adding i.v. ibandronate administered every 3 months during the first transplant year to calcium and calcitriol, did not demonstrate any benefit in terms of preventing loss of bone mineral density in the lumbar spine. Calcium and calcitriol supplementation alone showed an excellent efficacy and safety profile in this study, virtually maintaining BMD without any loss over 12 months after renal transplantation. Ibandronate added a significantly protective, although clinically modest, effect against bone loss in total femur and ultradistal radius, and the reduction in bone biomarkers was more pronounced in the ibandronate group. We have demonstrated that ibandronate as a representative of the bisphosphonate class can be given safely to renal transplant recipients.


Personal acknowledgments: The authors wish to thank Kirsten Lund for excellent study and patient coordination at site; Gunhild A. Isaksen for managing all DXA operations; Kajendran Mohanathas and Claudia Grasnick for data management; and Karl Johan Sundt and Hilde Kloster Smerud for managing the blinding of the study medication and for regulatory affairs consultation. In addition to the clinical investigators listed as authors, a large number of nephrologists in local Norwegian hospitals were instrumental in recording study data and for treating the study patients after being discharged from Oslo University Hospital, Rikshospitalet.

Role of the funding source: The main and formal sponsor of the study was Smerud Medical Research International (CRO) AS, a contract research organization which covered salary of two of the authors (ICO and KTS), and costs for underlying study medication (calcium, calcitriol) and for the bone biomarker laboratory kits. Cosponsors were the University of Oslo and the Rikshospitalet-Radiumhospitalet Medical Center (now Oslo University Hospital, Rikshospitalet, Norway) which covered salaries for those investigators and study support staff being affiliated there. Lastly, the investigator group received an unrestricted grant to the research group and free study medication and matching placebo from Roche Norge AS, Oslo, Norway, which is the license holder and manufacturer of ibandronate. The main sponsor provided monitoring (quality control), data management and biostatistics services to the study as well as randomization, labeling and distribution of the study drugs. All authors have full access to all data in the study.


The authors of this manuscript have no conflicts of interest to disclose as described by the American Journal of Transplantation.