Article Text

Extended report
Zoledronate for prevention of bone erosion in tophaceous gout: a randomised, double-blind, placebo-controlled trial
  1. Nicola Dalbeth1,
  2. Opetaia Aati1,
  3. Gregory D Gamble1,
  4. Anne Horne1,
  5. Meaghan E House1,
  6. Mark Roger2,
  7. Anthony J Doyle2,3,
  8. Ashika Chhana1,
  9. Fiona M McQueen4,
  10. Ian R Reid1
  1. 1Department of Medicine, University of Auckland, Auckland, New Zealand
  2. 2Department of Radiology, Auckland District Health Board, Auckland, New Zealand
  3. 3Department of Anatomy with Radiology, University of Auckland, Auckland, New Zealand
  4. 4Department of Molecular Medicine, University of Auckland, Auckland, New Zealand
  1. Correspondence to Dr Nicola Dalbeth, Bone and Joint Research Group, Department of Medicine, Faculty of Medical and Health Sciences, University of Auckland, 85 Park Rd, Grafton, Auckland 1023, New Zealand; n.dalbeth{at}


Objectives The osteoclast has been implicated in development of bone erosion in gout. The aim of this study was to determine whether zoledronate, a potent antiosteoclast drug, influences bone erosion in people with tophaceous gout.

Methods This was a 2-year, randomised, double-blind, placebo-controlled trial of 100 people with tophaceous gout. Participants were randomised to annual administration of 5 mg intravenous zoledronate or placebo. The primary endpoint was change in the foot CT bone erosion score from baseline. Secondary endpoint was change in plain radiographic damage scores. Other endpoints were change in bone mineral density (BMD), bone turnover markers and the OMERACT-endorsed core domains for chronic gout studies.

Results There was no change in CT erosion scores over 2 years, and no difference between the two treatment groups at Year 1 or 2 (p(treat)=0.10, p(time)=0.47, p(treat*time)=0.23). Similarly, there was no change in plain radiographic scores over 2 years, and no difference between the two groups at Year 1 or 2. By contrast, zoledronate increased spine, neck of femur, total hip and total body BMD. Zoledronate therapy also reduced the bone turnover markers P1NP and β-CTX compared with placebo. There was no difference between treatment groups in OMERACT-endorsed core domains.

Conclusions Despite improvements in BMD and suppression of bone turnover markers, antiosteoclast therapy with zoledronate did not influence bone erosion in people with tophaceous gout. These findings suggest a disconnect between responses in the healthy skeleton and at sites of focal bone erosion in tophaceous gout.

  • Gout
  • Osteoporosis
  • Arthritis

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Bone erosion is a frequent manifestation of chronic tophaceous gout.1–3 These areas of matrix destruction are usually found in close proximity to intra-articular tophi,4 and may lead to joint damage with subsequent disability. To date, no clinical trials have assessed structural damage as an endpoint in gout. Therefore, optimal strategies to prevent and treat bone erosion in people with gout are currently unknown.

Osteoclasts are responsible for bone resorption in the healthy skeleton. The osteoclast has been implicated in the pathogenesis of gouty erosions.5 ,6 High numbers of preosteoclasts are present in the circulation and synovial compartment in people with erosive gout. Within the tophus, there are numerous osteoclast-like cells surrounding monosodium urate (MSU) crystal deposits. Osteoclast-like cells are also present at the tophus-bone interface within the joints of people with erosive gout. Taken together, these data provide a rationale for assessing antiosteoclast therapies for bone erosion in this disease.

Bisphosphonates are widely available agents that inhibit osteoclast function, and are used for treatment of bone diseases such as osteoporosis, Paget's disease and metastatic bone lesions. In animal models of inflammatory arthritis, bisphosphonates markedly reduce the progression of focal articular bone erosions.7–9 Zoledronate, a potent bisphosphonate, may be effective in suppressing the progression of bone erosions in people with rheumatoid arthritis (RA).10

The aim of this study was to determine whether administration of zoledronate influences bone erosion in people with tophaceous gout.

Patients and methods

Study design

This was a 2-year, randomised, double-blind, placebo-controlled single-centre trial of zoledronate for erosive gout. This trial was approved by the Northern X Regional Ethics Committee, and all participants provided written informed consent. The study was registered as a clinical trial with the Australian New Zealand Clinical Trials Registry (ACTRN12608000463370).


Participants were recruited from rheumatology clinics and inpatient databases throughout Auckland, and through advertising to general practitioners and the public. Participants were recruited between 16 December 2008 and 2 August 2010, and the final study visit was 26 September 2012. Inclusion criteria were: diagnosis of gout according to the 1977 American Rheumatism Association classification criteria,11 at least one subcutaneous tophus confirmed by a rheumatologist, on stable therapy for gout (no change in the last 3 months), age over 18 years, and able to provide informed consent. Exclusion criteria were: prolonged treatment with potent bisphosphonates in the preceding 2 years, impaired renal function (creatinine clearance <30 mL/min), serum calcium <2.1 mmol/L, serum 25-hydroxy Vitamin D <15 ng/mL, allergy to bisphosphonate, pregnancy or breastfeeding, or unstable systemic medical condition. One hundred and five patients were screened for inclusion into the study. Five patients did not meet the inclusion criteria and were not recruited into the study (two with creatinine clearance <30 mL/min, three withdrew consent after screening) (figure 1).

Figure 1

Flow of participants through the study.


All visits took place at a clinical research facility in a tertiary medical centre. The study included 100 participants; 50 randomly assigned to annual administration of intravenous zoledronate 5 mg, and 50 assigned to annual administration of placebo (100 mL 0.9%NaCl identically packaged to zoledronate and administered in an identical fashion). Participants had clinical assessment, laboratory testing and imaging assessment of bone erosion (by CT and plain radiography (XR)) at baseline, and after 1 and 2 years. Treatment allocations were randomised by the study statistician prior to the commencement of the study, using a random block randomisation algorithm, based on computer-generated (Excel 2003) random numbers. To ensure masking, only the statistician and the staff member who labelled the bottles for intravenous infusions had access to treatment allocation, and neither had contact with participants. All the other study personnel and participants were blinded to treatment allocation throughout the study. Adherence to the study products was complete, as the study research assistant directly administered the product. However, the Year 1 infusions were not administered to four participants as they declined a second dose due to an acute-phase response following the baseline visit infusion (all these participants were in the zoledronate group).

The primary endpoint was change in the foot CT bone erosion score from baseline at Years 1 and 2. CT is considered the gold standard method of bone erosion measurement, with greater sensitivity than MRI and XR.12–14 The secondary endpoints were change from baseline at Years 1 and 2 in XR damage score and erosion score. Other exploratory endpoints were change from baseline at Years 1 and 2 in the following measures: bone mineral densities (BMD) of the spine, mean total hip and total body, the serum bone turnover markers procollagen type-1 N-terminal propeptide (P1NP) and β-C-terminal telopeptide of type I collagen (β-CTX), urinary C-terminal telopeptide of type II collagen (CTX-II), swollen (/66) and tender (/68) joint counts and hand grip strength using a hand dynamometer (Surgical Synergies, New Zealand). Core domains for chronic gout studies as endorsed by Outcome Measures in Rheumatology (OMERACT)15 were measured at each timepoint; flare frequency (in the preceding 3 months), subcutaneous tophus count, serum urate concentration (SU), health assessment questionnaire (HAQ)-II score,16 100 mm pain visual analogue score (VAS), and patient global assessment (0–10 Likert scale).

Computed tomography

CT images of both feet were obtained at the beginning of the study, and at the Years 1 and 2 visits. The participants were positioned feet first in a supine position with the feet in a plantar flexion position. Ankles and feet were scanned axially in one helical acquisition as previously described.17

CT bone erosion volume in both feet was scored using a CT bone erosion scoring method, based on the RA MRI score (RAMRIS) for erosion,18 as validated for gout in a separate group of patients.17 The gout CT bone erosion scoring system includes the following bones for erosion on a semiquantitative scale from 0 to 10 in each foot: 1st metatarsal (MT) head, 2nd–4th MT base, cuboid, middle cuneiform, distal tibia (maximum total score 140). All CT scans were scored at the end of the study (paired scans from baseline, 1 and 2 years, with order known) by two musculoskeletal radiologists (AD and MR), who were blinded to treatment allocation and XR scores. Mean scores from both readers were used in the analysis.

Plain radiography

XR of the hands and feet were scored by a rheumatologist (ND) and musculoskeletal radiologist (MR) who were blinded to treatment allocation and CT scores. The films were scored for erosion and joint space narrowing using a modification of the Sharp-van der Heijde scoring method,19 validated for gout.20 All XR were analysed at the end of the study (paired scans from baseline, 1 and 2 years, with order known). Mean scores from both readers were used in the analysis.

Bone mineral density

BMD was measured at baseline, year 1 and year 2 at the lumbar spine (L1-L4), dual proximal femur, and total body using a Lunar Prodigy dual-energy X-ray absorptiometer (GE Lunar, Madison, Wisconsin, USA).

Bone and cartilage turnover markers

At baseline, year 1 and year 2, fasting blood and urine samples were collected and stored at −80°C until they were batch analysed at the end of the study. Serum P1NP and β-CTX were measured using the Roche Elecsys 2010 platform (Roche Diagnostics, Indianapolis, Indiana, USA), and urine CTX-II by Urine CartiLaps ELISA (Immunodiagnostic System, Tyne and Wear, UK). Urine CTX-II was corrected for urine creatinine (mmol/L).

Adverse events

Adverse events were recorded at each study visit. Acute-phase response (APR) was assessed using a questionnaire administered by telephone 1 week after administration of study medication.21 One participant in the placebo group did not complete the postinfusion questionnaire after the Year 1 infusion.

Study power and statistical analysis

Prospective assessment of study power was not possible, as at the time of commencing the study there were no published observational studies or clinical trials quantifying progression of bone erosion in gout over time or in response to treatment. We explored the reproducibility of erosion scores in foot XR from 26 patients with gout and at least one tophus.20 In this analysis, the mean XR score was 34 (SD 27). Using these data, two groups of 45 participants had adequate power (80%) at the two-tailed 5% significance level to detect a difference of 50% between the groups in total erosion score (ie, 34 vs 18). This constituted a medium effect size in respect of Cohen's dictum21a and was considered clinically relevant. To preserve the power of this study, the effective sample size was 100 (ie +10% allowance for loss to follow-up). As CT has superior sensitivity to XR,12 ,13 it was possible that smaller differences may have been apparent in our study which used CT erosion assessment as the primary endpoint.

Analysis was undertaken using the SAS statistical package (V.9.2 SAS Institute). All dependent variables were normal or rendered normal by transformation. Treatment and control groups were compared by a mixed-models approach to repeated measures employed to model the difference in the change in each of the bone erosion parameters between groups (with baseline level included as covariate (ANCOVA). Significant main (treatment or time) and interaction (treatment by time) effects were examined using the method of Tukey to preserve the overall 5% significance level. Missing at random data were imputed using maximum likelihood methods within the mixed procedure of SAS. In sensitivity analyses, missing data were imputed using Markov chain Monte Carlo approach and, alternatively, using a standard carry-forward analysis, and the results compared. In additional analyses, baseline SU concentration was included in the ANCOVA models. Fisher's exact test was used to test for differences in numbers of participants who experienced at least one event or to compare categorical variables at baseline. Student t test was used to compare treatment groups at baseline. Least squares adjusted means and SEs of the means are presented. All tests were two-tailed and p<0.05 was considered significant. Primary data analyses were performed on an intention-to-treat basis.


Baseline characteristics

The flow of participants through the study is shown in figure 1. There were 50 participants randomised to each treatment group. At Year 1, there were 46 participants in both groups. At Year 2, there were 36 participants in the zoledronate group and 38 in the placebo group.

The baseline characteristics of the two groups were similar (table 1). Participants had a mean disease duration of more than 20 years, and mean subcutaneous tophus count of 7. The majority of participants (92%) were taking urate-lowering therapy, with mean SU 0.37 mmol/L.

Table 1

Baseline characteristics of participants

Primary endpoint: CT bone erosion score

CT bone erosion scores did not differ between the two groups at baseline (table 1). Over the 2-year study period, there was no change in CT erosion scores over time, and no difference between the two treatment groups at Year 1 or 2 (p(treat)=0.10, p(time)=0.47, p(treat*time)=0.23) (figure 2). Similar results were observed after including mean SU in the model (p(treat)=0.13, p(time)=0.49, p(treat*time)=0.22), and using various methods of imputation to account for missing data including carry-forward imputation (p(treat)=0.10, p(time)=0.47, p(treat*time)=0.20) and Markov chain Monte Carlo Imputation (p(treat)=0.18, p(time)=0.44, p(treat*time)=0.29).

Figure 2

Primary and secondary endpoints. (A) CT) erosion scores, (B) Plain radiographic (XR) damage scores, (C) XR erosion scores. Left panel shows total scores and right panel shows change scores. Data are presented as mean (SEM).

Secondary endpoint: XR damage scores

XR damage scores and erosion scores did not differ between the two groups at baseline (table 1). Over the 2-year study period, there was no change in XR damage scores or erosion scores, and no difference between the two groups at Year 1 or 2 (figure 2).

Exploratory endpoints: bone density and bone turnover markers

Baseline bone density was excellent with total body T-scores of 1.32 in the placebo group and 1.38 in the zoledronate group (table 1). Zoledronate treatment increased spine, neck of femur, total hip and total body BMD (figure 3). Similarly, zoledronate led to significant reductions in the bone turnover markers P1NP and β-CTX compared with placebo (figure 3).

Figure 3

Change in bone mineral density (BMD) and bone turnover markers. (A) Change in neck of femur BMD, (B) Change in total hip BMD, (C) Change in anteroposterior (AP) spine BMD, (D) Change in total body BMD, (E) Change in procollagen-1 N-terminal propeptide (P1NP), (F) Change in β-C-terminal telopeptide of type I collagen (β-CTX). Data are presented as mean (SEM).

Exploratory endpoints: gout outcome measures

In the entire group, there was no significant change in subcutaneous tophus count or SU concentration over the 2-year period (figure 4). However, there was a reduction in gout flare frequency and HAQ-II, and increase in pain visual analogue scale and patient global assessment (figure 4). No difference was observed between treatment groups over time in OMERACT-endorsed core domains including flare frequency, subcutaneous tophus count, SU, HAQ-II, pain VAS, patient global assessment (see online supplementary table S1 and figure S4). Similarly, there was no difference between groups in tender joint count, swollen joint count, grip strength, or urine CTX-II (see online supplementary table S1).

Figure 4

Change in OMERACT endorsed core domains for chronic gout studies. (A) Change in gout flare frequency (number of flares in preceding three months), (B) Change in subcutaneous tophus count, (C) Change in serum urate concentration (mmol/L), (D) Change in Health Assessment Questionnaire (HAQ)-II score, (E) Change in pain visual analogue scale (mm), (F) Change in patient global assessment of disease activity (0–10 Likert score). Data are presented as mean (SEM).

Adverse events

Adverse events were reported in 38 participants receiving zoledronate and 38 participants receiving placebo (table 2). There were 48 serious adverse events reported in 19 zoledronate-treated participants and 35 serious adverse events in 18 placebo-treated participants. There were two deaths in the zoledronate group (both due to cardiac causes) and no deaths in the placebo group.

Table 2

Adverse events

One serious adverse event was an overnight hospital admission for observation and pain relief due to an acute-phase response in a zoledronate-treated patient. Fractures were reported in one zoledronate-treated patient (two fractures) and one placebo-treated patient (one fracture). Cardiac arrhythmias occurred in two zoledronate-treated participants (five events), and four participants (six events) in the placebo group. Osteonecrosis of the jaw or atypical femoral fractures were not reported in either group.

Hospital admission due to gout flare was reported as a serious adverse event in four zoledronate-treated participants (seven events) and no placebo-treated participants (p=0.12). Gout flares were reported as adverse events in 11 zoledronate-treated participants (21 events) and 11 placebo-treated participants (16 events).

Following the baseline infusion, symptoms consistent with an APR were observed in 25/50 (50%) zoledronate-treated participants and 18/50 (36%) placebo-treated participants (table 2 and see online supplementary table S2). In both groups, symptoms of APR were less frequent following the second infusion.


This 2-year randomised, double-blind, placebo-controlled trial has not demonstrated a beneficial effect for zoledronate on bone erosion in joints of people with tophaceous gout. Although zoledronate had a clear effect on BMD and bone turnover markers, similar improved outcomes were not observed in measures of structural joint disease or other chronic gout outcome measures. In those receiving zoledronate, APR symptoms, such as muscle pain, occurred more frequently and there was a trend towards worse patient global assessment. Our data do not support the use of antiosteoclast therapy in the management of bone erosion in people with longstanding and tophaceous gout.

At present, optimal treatment strategies for bone erosion in gout are unknown. This study is the first clinical trial to systematically analyse bone erosion in people with tophaceous gout. The participants in this study had a mean disease duration of more than 20 years and were on stable urate-lowering therapy at the study entry and throughout the study, with mean SU of 0.38 mmol/L (6.3 mg/dL). No significant differences were observed in subcutaneous tophus counts or bone erosion scores over time. The findings of this study contrast with our recent observational study of people with severe gout treated with pegloticase,22 a drug that leads to rapid regression of subcutaneous tophi.23 In pegloticase, responders who achieved very low SU (<0.06 mmol/L (1 mg/dL)), XR erosion scores improved over 1 year, with evidence of filling-in of bone erosions at sites of tophus regression.22 The lack of radiographic progression in the current study raises a number of interesting questions about the processes of bone erosion in gout, particularly about the natural history of bone erosion in gout, whether bone erosions can heal in those with long disease duration, and whether erosions progress at all when SU are held below 0.36 mmol/L.

Recent laboratory studies indicate that, in addition to effects on osteoclastogenesis, MSU crystals have profound effects on osteoblast viability and function.24 ,25 Therefore, in the presence of persistent intra-articular MSU crystals, the deleterious effects on osteoblasts may prevent erosion healing, even with potent antiosteoclast therapy. Together, the laboratory and clinical studies suggest that in order to achieve prevention or healing of bone erosion, intensive urate-lowering therapy should be the focus. It is conceivable that targeting the osteoclast in the context of very intensive urate-lowering may have benefits in promoting healing of erosions. However, in the current study, addition of SU into the analysis models did not alter our findings, suggesting that addition of antiosteoclast therapy may not have benefit over urate-lowering therapy alone.

The findings of this study can be compared with therapeutic studies of other erosive arthropathies. Many in vivo preclinical studies of autoimmune inflammatory arthritis have demonstrated that, despite causing transient increases in proinflammatory cytokines, such as TNF-α, IL-6 and IL-1β,26 zoledronate has a beneficial effect on development of bone erosion and bone marrow oedema (BMO).7–9 ,27 In a study of 39 people with early RA on methotrexate, treatment with zoledronate reduced development of new erosions and BMO on MRI over 26 weeks.10 More recently, a study of 22 people with psoriatic arthritis showed that zoledronate treatment reduced MRI BMO, but did not alter bone erosion over 52 weeks.28 In RA and psoriatic arthritis, BMO is a preerosive lesion.29–31 The osteopathology of gout appears to differ from these other forms of inflammatory arthritis, with MRI studies showing that BMO is a minor feature of chronic gouty arthropathy, and that bone erosion is strongly associated with tophus, rather than BMO or synovitis.32 ,33 Thus, the stability of tophus burden in this study may account for the lack of change in bone erosion over the 2-year period, even following treatment with zoledronate.

This study has provided new insights into the effects of potent bisphosphonate therapy on BMD. The participants were predominantly middle-aged men with well-preserved BMD at baseline. Over half the participants were of Polynesian ancestry, which is associated with increased BMD compared with people of European ancestry.34 Even with excellent bone density at baseline, further increases in BMD were observed following zoledronate therapy, with suppression of bone turnover markers. The percentage change in BMD observed in this study was similar to studies of zoledronate in women with postmenopausal osteoporosis.35 Taken with the joint imaging data, our BMD data indicate a disconnect between therapeutic responses in the healthy skeleton and at sites of focal bone erosion in the context of MSU crystal deposition and tophus formation.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 21a.
  23. 22.
  24. 23.
  25. 24.
  26. 25.
  27. 26.
  28. 27.
  29. 28.
  30. 29.
  31. 30.
  32. 31.
  33. 32.
  34. 33.
  35. 34.
  36. 35.

Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Files in this Data Supplement:


  • Handling editor Tore K Kvien

  • Contributors ND (the guarantor) accepts full responsibility for the work and the conduct of the study, had access to the data, and controlled the decision to publish. ND conceived of the study, scored the radiographs, contributed to the data interpretation, and drafted the manuscript. OA and AH recruited participants and coordinated study visits. GDG analysed the data. MEH contributed to administration of study products. MR and AJD scored the images. AC completed laboratory testing. FMMQ assisted with protocol development and interpretation of the results. IRR conceived of the study, contributed to the data interpretation, and drafted the manuscript. All authors read and approved the final manuscript.

  • Funding This study was funded by the Health Research Council of New Zealand (grant number 09-111D).

  • Competing interests Novartis provided study drug to the investigators but had no influence over development of the study design, conduct of the study, data analysis or manuscript preparation. Ian Reid has received consulting and speaker fees from Novartis. Nicola Dalbeth has received speaker fees from Novartis. The other authors have no conflicts to declare.

  • Patient consent Obtained.

  • Ethics approval Northern X Regional Ethics Committee.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Linked Articles