Intervention Protocol

You have free access to this content

Bisphosphonates for osteoarthritis

  1. Wei Yang1,
  2. Qi Zhuo2,
  3. Wei Chai2,
  4. Jiying Chen2,
  5. Cheng Sun3,
  6. Yan Wang2,*

Editorial Group: Cochrane Musculoskeletal Group

Published Online: 3 NOV 2013

DOI: 10.1002/14651858.CD010805

How to Cite

Yang W, Zhuo Q, Chai W, Chen J, Sun C, Wang Y. Bisphosphonates for osteoarthritis (Protocol). Cochrane Database of Systematic Reviews 2013, Issue 11. Art. No.: CD010805. DOI: 10.1002/14651858.CD010805.

Author Information

  1. 1

    Chinese PLA General Hospital, Department of Endocrinology, Beijing, Beijing, China

  2. 2

    Chinese PLA General Hospital, Department of Orthopaedic Surgery, Beijing, China

  3. 3

    Peking University Third Hospital, Department of Orthopaedic Surgery, Beijing, China

*Yan Wang, Department of Orthopaedic Surgery, Chinese PLA General Hospital, 28 Fuxing Road, Haidian district, Beijing, 100853, China.

Publication History

  1. Publication Status: New
  2. Published Online: 3 NOV 2013




  1. Top of page
  2. Background
  3. Objectives
  4. Methods
  5. Acknowledgements
  6. Appendices
  7. Contributions of authors
  8. Declarations of interest
  9. Sources of support

Description of the condition

In 1986, the American College of Rheumatology defined osteoarthritis (OA) as “a heterogeneous group of conditions that lead to joint symptoms and signs, which are associated with defective integrity of articular cartilage, in addition to related changes in the underlying bone at the joint margins" (Altman 1986). OA is by far the most prevalent form of all joint disorders and a major cause of pain and functional disability that affected 151 million individuals globally in 2000 (Symmons 2006), and the number is expected to double by the year 2020 (Lawrence 2004). The prevalence of symptomatic OA (pain on most days of a recent month and radiographic evidence of OA in the affected joint) is approximately 13% in adults aged ≥ 60 years (Dunlop 2001; Lawrence 2004), 6% of all adults aged 30 years for the knee (Longo 2011), 4% in those aged between 55 and 74 years for the hip (Lawrence 2004), and 10% of people aged > 70 years for the hand (Zhang 2002). Because OA is a chronic, progressively degenerative disease of synovial joints lacking disease-modifying medication, operative interventions, primarily in the form of joint arthroplasty have become increasingly popular but also posed a heavy economic and societal burden. By 2030, the demand for primary total hip and knee arthroplasties is estimated to grow by 174% to 572,000 and by 673% to 3.48 million procedures, respectively, in US alone (Kurtz 2007).

The key and most remarkable characteristic of OA is the destroyed integrity of articular cartilage, its abnormalities, however, are not as limited. Furthermore, cartilage itself is both avascular and aneural, OA symptoms, such as pain and stiffness, logically could not be directly generated by it. Therefore, there must exist other important features that are responsible for the pain of OA such as alterations in subchondral bone architecture (increased volume, trabecular thickening and separation, sclerosis), joint margin osteophytes outgrowth on radiograph, and changes in the juxta-articular bone marrow known as bone marrow lesions on magnetic resonance imaging (MRI), all of which have been shown to correlate with the presence of increased bone mineral density (Felson 2007; Li 1997; Lo 2005; Messent 2005; Spector 2003). Increased bone mineral density could produce a stress shielding effect, leading to localised weakening of trabecular micro-structure and enhanced bone turnover (Tissakht 1996). In other words, bone is an ignored important player in the development process of OA. Changes in different joint tissues are not independent with each other. Indeed, not only are they spatially linked, they take place simultaneously and in concert. The correlation between degenerations of cartilage and alterations in subchondral bone as well as bone marrow are suggested and supported by a number of studies (Felson 2003; Guermazi 2003; Muraoka 2007), and provides a possibility that drugs affecting bone metabolism may have benefits in altering the progression of OA. The nature of their exact relationship, however, remained to be elucidated.

Although the aetiology of OA remains elusive, multiple risk factors have been found to be associated with osteoarthritis in epidemiology studies, with biomechanics playing a central role and a complex interplay of other important factors such as age, female sex, previous injury, genetic predisposition, metabolic elements, and so on (Creamer 1997).

The association between symptoms and radiographic examinations of OA is weak (Conaghan 2011). OA is most often diagnosed by an overall clinical impression coming from the patient's age, history, physical examination findings, and radiographic presentation. The most frequently affected joints are weight-bearing joints such as knees and hips, as well as the cervical and lumbosacral spine, and hands. By contrast, the wrist, elbow, and ankle are usually spared (Longo 2011). In recent years, no longer misconceived as a uniform disease which affects only joint cartilage, the theory that OA is a disease of the whole synovial joint has already become widely accepted, meaning that OA may represent failure of the joint as an organ, analogous to renal or cardiac failure (Arden 2006), resulting from an imbalance in the dynamic equilibrium between the breakdown and repair of joint tissues. This new understanding and other discoveries relating to its pathophysiology have driven the division of OA into distinguishable phenotypes, comprising post­traumatic, ageing­related, genetic, pain­related and metabolic forms (Zhuo 2012).

Although chondrocytes are probably the most important cells involved in the development of osteoarthritis (Longo 2011), the process, generally initiated either by damage to normal articular cartilage or by defective cartilage failure under normal joint loading, is an active interaction between degradation and repair processes of cartilage, bone, and synovium. Under either circumstance, an essentially similar pathway follows, involving a multitude of factors such as various proteolytic enzymes, cytokines and growth factors secreted by chondrocytes as well as synovium (Zhuo 2012), finally leading to the degradation of collagen and proteoglycans (which comprise over 90% of the cartilage macromolecules), synovial inflammation, bone remodeling, and muscle weakness.


Description of the intervention

Similar to many musculoskeletal disorders, the management of osteoarthritis includes physical measures (physical therapy and exercise programs) (Bartels 2007; Fransen 2008; Fransen 2009, Verhagen 2007), medical treatment (acetaminophen, non-steroidal anti-inflammatory drugs, opioids, glucosamine and chondroitin, intra-articular steroid or hyaluronate injections) (Bellamy 2006a; Bellamy 2006b; Cepeda 2006; Nüesch 2009; Towheed 2006; Towheed 2009), and surgical therapies (osteotomy, arthroscopic surgery, and total joint replacement) (Brouwer 2007; Jacobs 2001; Jacobs 2005; Jolles 2006; Laupattarakasem 2008; Nakama 2012; Reichenbach 2010; Zhang 2010). However, unlike diseases such as osteoporosis and rheumatoid arthritis, where dramatic success had been achieved in preservation of function and structure as well as symptom relief by means of pharmacological interventions, current mainstream medical treatment for OA is in essence symptomatically palliative with only modest efficacy and yet frequent or serious adverse effects, especially with long-term use (ACR 2008; Hochberg 2012; Hunter 2011; Zhang 2008).

Inspired by disorders such as rheumatoid arthritis, for which the development of the disease-modifying anti-rheumatic drugs has successfully altered the disease course and clinical management, an array of new etiologically targeted agents, collectively named as disease-modifying anti-OA drugs after their potential via a variety of mechanisms and targets including subchondral bone, cartilage and synovium, to slow, halt, or even revert the disease progression as well as to provide symptomatic benefit, have been emerging, and this may well be the preferable approach that is in accordance with the concept of OA as a disease entity involving the whole joint (Hunter 2011; Verbruggen 2006).To link the whole joint tissue to the pathophysiology of OA offers the possibility of introducing new etiologically targeted pharmacological interventions to the treatment regimen. Clinical studies both in animals and humans have suggested that a wide array of such agents are promising candidates of possessing the ability to modify OA progression as well as provide concomitant improvement in OA symptoms (Hunter 2011). Some of these candidate drugs include those targeting cartilage (inhibitors of matrix metalloproteinases (MMPs), inducible nitric oxide synthase, and cathepsin K, glucosamine, chondroitin, diacerein, avocado soybean unsaponifiables, growth factors, doxycycline, vitamin D and vitamin E), systemic agents inhibiting inflammatory pathways (licofelone, cytokines, antibodies of tumour necrosis factor, steroids), and disease-modifying anti-OA drugs targeting subchondral bone (calcitonin, bisphosphonates, and strontium ranelate) (Bondeson 2011; Fajardo 2005; Hunter 2011; Zhuo 2012).


How the intervention might work

Bisphosphonates are a class of drugs commonly used for the prevention and treatment of osteoporosis (Wells 2008a; Wells 2008b; Wells 2008c) as well as to treat similar bone diseases that feature bone fragility, such as Paget's disease, bone metastasis, multiple myeloma, osteogenesis imperfecta, hypercalcaemia, and so on (FDA 2011).

Bisphosphonates are named based on the fact that they have two phosphonates, and their structural similarity to pyrophosphate, a byproduct of cellular metabolism and the mechanism by which bone mineralization is regulated (Drake 2008). They work by directly impairing osteoclast activity, and thus decrease physiological bone resorption, via mechanisms of either increasing cellular apoptosis or influencing their metabolic pathway (Fleisch 1998). In the meantime, bisphosphonates also appear to have a positive effect on osteoblasts (Plotkin 2008). Bisphosphonates are, however, differential both in structure, in effect, and in mechanism of action (Rogers 2004). Early developed bisphosphonates do not contain nitrogen group (etidronate, clodronate, and tiludronate), while the second- and third-generation ones are nitrogen-containing (alendronate, neridronate, ibandronate, pamidronate, risedronate, and zoledronic acid,). The nitrogen-containing bisphosphonates work primarily by inhibiting activation of enzymes that utilize pyrophosphate, such as farnesyl pyrophosphate synthase, a key regulatory enzyme and determinant in the regulation of core osteoclast cellular activities and anti-resorptive capacity, and are more potent than the simple not-containing-nitrogen bisphosphonates, which induce apoptosis of osteoclasts by disrupting the use of adenosine triphosphate (ATP) as the source of energy (Beek 2003; Kimmel 2007).

The primary reason that bisphosphonates hold the OA therapeutic potential is their established pharmacology of altering bone remodeling through a direct inhibitory effect on the osteoclast activity (function and number), occurring at the subchondral bone level of affected joints, where altered bone mineral quality and density as well as bone turnover similar to that of osteoporosis was observed (Spector 2003). This is further demonstrated by a clinical study (Bingham 2006) in which patients with progressive knee OA were associated with higher urinary levels of C-terminal telopeptide of type II collagen (CTX-II), a marker of cartilage degradation, as well as by a number of animal studies, in which the classic spontaneous model of Duncan-Hartley guinea pig (Bendele 2005; Meyer 2001a; Meyer 2001b) and other models were used to investigate the anti-resorptive effects of bisphosphonates on OA (Hayami 2004; Muehleman 2002), and displayed that when measured by size and severity of cartilage damage and osteophyte outgrowth, risedronate, a bisphosphonate, slowed OA progression by as much as 30 to 40%. In addition, it was demonstrated that bisphosphonates may exert immuno-modulation, and thus, inhibition of pro-inflammatory cytokines, which is also a major participant in the osteoarthritic process (Frith 2001). Data suggested that inflammation may also play an important role in the pain of OA (Felson 2003; Haynes 2002), and both knee effusions and synovial thickening are more common in patients with knee pain than those without knee pain (Hill 2001). Bisphosphonates may thus relieve knee pain by way of antiinflammatory effects (Frith 2001).


Why it is important to do this review

OA is the most common and prevalent form of arthritis and a leading cause of disability; however, a remarkable gap exists with respect to desirable pharmacologic interventions, considering that current pharmacological treatment options have been burdened with issues such as low to moderate efficacy, long term safety concerns, and inability to alter the course of the disease.

The field of disease-modifying osteoarthritis drugs is rapidly evolving, in which bisphosphonates is one important class of candidates. Cochrane reviews have already analysed some possible agents such as diacerein (Fidelix 2009), doxycycline (da Costa 2012), glucosamine (Towheed 2009), and chondroitin in the treatment of OA (Singh 2006). Based on the fact that there have been several published and on-going prospective, double-blind randomised controlled trials (RCTs) involving bisphosphonates, we believe it is clinically relevant to systematically review and discuss the current evidence of effects of bisphosphonates in the treatment of OA.



  1. Top of page
  2. Background
  3. Objectives
  4. Methods
  5. Acknowledgements
  6. Appendices
  7. Contributions of authors
  8. Declarations of interest
  9. Sources of support

To assess the benefits and harms of bisphosphonates for the pharmacological treatment of the hip, knee and hand OA. Both symptomatic and structural outcomes will be evaluated.



  1. Top of page
  2. Background
  3. Objectives
  4. Methods
  5. Acknowledgements
  6. Appendices
  7. Contributions of authors
  8. Declarations of interest
  9. Sources of support

Criteria for considering studies for this review


Types of studies

RCTs and quasi-RCTs (method of allocating participants to a treatment which is not strictly random: e.g. by date of birth, hospital record number, alternation) evaluating bisphosphonates for treating the hip, knee and hand OA.


Types of participants

Adult patients aged 45 years and above, of either sex, with clinically and/or radiologically confirmed primary OA of the knee, hip, or hand. Studies testing conditions such as secondary OA, and inflammatory ones such as rheumatoid arthritis will be excluded. If mixed population are tested, we will include them if the subgroup data for primary OA are available.


Types of interventions

RCTs, or quasi-RCTs comparing bisphosphonates with placebo, no intervention and active comparator. There will be no restrictions on route of administration or dosage. We will include all durations of treatment.


Types of outcome measures


Major outcomes

  • Pain with a hierarchy of 11 levels

    • Pain overall
    • Pain on walking
    • Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) pain subscale
    • Pain on activities other than walking
    • WOMAC global scale
    • Lequesne osteoarthritis index global score
    • Other algofunctional scale
    • Patient's global assessment
    • Physician's global assessment
    • Other outcome
    • No continuous outcome reported

When more than one is reported, we will take the highest on the list e.g. choose pain overall over pain on walking.

  • Physical function with a hierarchy of eight levels

    • Global disability score
    • Walking disability
    • WOMAC disability subscore
    • Composite disability scores other than WOMAC
    • Disability other than walking
    • WOMAC global scale
    • Lequesne osteoarthritis index global score
    • Other algofunctional scale

When more than one is reported, take the highest on the list e.g. choose global disability score over walking disability.

  • Radiographic joint structure changes according to the given hierarchy

    • Minimum joint space width
    • Median joint space width
    • Semi-quantitative measurement
  • Quality of Life
  • Number of patients experiencing any adverse event
  • Patients who withdrew because of adverse events
  • Patients experiencing any serious adverse events


Minor outcomes

  • Percentage of "progressors" (i.e. number of severe joint space narrowing, which is most often defined as a narrowing of greater than 0.5mm (Ravaud 1998))
  • Joint arthroplasty
  • Bone marrow lesions

If a study provided more than one pain or function scale, we will refer to a previously described hierarchy list and extract those highest on it (e.g. global Visual Analog Scale (VAS) pain takes precedence over WOMAC osteoarthritis index pain subscale, and WOMAC disability subscale over Lequesne osteoarthritis index score etc) (McAlindon 2000).

In order to map safety profile of each drug, we intend to present specific adverse events for each disease modifying anti-OA drugs candidate included; meanwhile, whenever possible, the following three outcomes will be analysed as well: patients who withdrew because of adverse events, patients experiencing serious adverse events, and patients experiencing any adverse event.

We will extract data at the end of the trial. If multiple time-points should be provided, we will extract data based on short-, intermediate-, and long-term follow-up, which are pre-defined as ≤ 6 months, 7 to 24 months, and ≥ 24 months. We expect different clinical effects in the short versus long term. In the short term, pain-related outcomes may be more obvious, whereas structural changes may only be detected in the long term using X-ray method. However, with the use of more sensitive MRI, structural changes may be quantitated at an earlier time, and data concerning MRI outcomes may be combined with joint space width (JSW) on radiograph using the standardised mean difference (SMD).


Search methods for identification of studies


Electronic searches

We will search the Cochrane Central Register of Controlled Trials (The Cochrane Library, current issue), MEDLINE (1946 to present), and EMBASE (1980 to present). No language or publication restrictions will be applied.

We will combine the MEDLINE (subject-specific strategy with the "Cochrane Highly Sensitive Search Strategy for identifying randomized trials in MEDLINE: sensitivity-maximizing version (2008 revision); Ovid format" as illustrated in the Cochrane Handbook for Systematic Reviews of Interventions (Lefebvre 2011). We will modify the MEDLINE search strategy (Appendix 1) for use in other databases.

In order to identify recently completed, unpublished and ongoing trials, we will search Current Controlled Trials ( and the World Health Organization (WHO) International Clinical Trials Registry Platform (ICTRP) search portal (


Searching other resources

We will check the reference lists of published studies to identify additional trials. Where necessary, we will contact authors of identified studies and companies who produce relevant products for additional information. Furthermore, we ill screen the websites of regulatory authorities such as the U.S. Food and Drug Administration (FDA) ( and the European Medicines Agency (EMA) ( for information on adverse events. We will search for errata or retractions from included studies published in full-text on PubMed ( and report the date this was done within the review.


Data collection and analysis


Selection of studies

Two authors (ZQ, YW) will assess the results of the searches. To determine which studies are to be assessed further, we will independently scan the titles, abstracts and key-words of every record retrieved. We will retrieve full-text articles for further assessment. Where necessary, we will contact the original authors of identified studies for missing information. Differences in opinion will be resolved by discussion, or a third author will be consulted. If bisphosphonates were included in any multi-armed RCTs, we will extract all relevant data and include them in our analyses.


Data extraction and management

Two authors (ZQ, YW) will independently assess each trial and extract information on trial design, study population, interventions and outcomes using a data extraction form specifically designed for this review. Differences will be resolved by referring back to the original article and reaching a consensus. When necessary, we will seek further information from the trial authors.

During the data extraction process, if a mixture of change-from-baseline scores (change scores) and final measurement (final values) are to be identified, we will put priority on the change scores on the ground that change scores remove a component of between-person variability, and was also considered to have a less skewed distribution (Deeks 2011). However, if final values are explicitly reported and change scores need to be imputed, we will include final values, considering that change scores and final values could be combined in a meta-analysis when using the (unstandardised) mean difference method.

We also intend to extract data at all periods of follow-up for each outcome in all trials so as to make the most use of the information available, and present a summary effect over different time points predefined as short, medium, and long-term.


Assessment of risk of bias in included studies

Two authors (ZQ, YW) will independently assess risk of bias using the criteria described in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011a). The following sources of bias will be assessed.

  1. Randomization process: assessment for selection bias
  2. Allocation concealment process: assessment of selection bias
  3. Blinding of participants and care personnel: assessment for performance bias
  4. Blinding of outcome assessors: assessment for detection bias
  5. Incomplete data bias
  6. Selective reporting bias
  7. Major imbalances in key baseline characteristics

We will grade each potential source of bias as 'high risk', 'low risk' or 'unclear risk' and provide a quote from the study report together with a justification for our judgment in the 'Risk of bias' table. We will summarise the risk of bias judgements across different studies for each of the domains listed. We will consider blinding separately for different key outcomes where necessary (e.g. for unblinded outcome assessment, risk of bias for all-cause mortality may be very different than for a patient reported pain scale). As well, we will consider the impact of missing data by key outcomes.

Where information on risk of bias relates to unpublished data or correspondence with a trialist, we will note this in the 'Risk of bias' tables.

When considering treatment effects, we will take into account the risk of bias for the studies that contribute to that outcome.

We will present the figures generated by the risk of bias tool to provide summary assessments of the risk of bias. Where necessary, disagreements will be resolved by discussion.


Measures of treatment effect

We expect to find both event (dichotomous) data (e.g. adverse events, and percentage of 'progressors') and continuous data (e.g. pain and function related outcomes, and joint space width). We will use risk ratios (RRs) with 95% confidence intervals (CIs) for reporting dichotomous data. We intend to calculate the risk difference (RD) and convert the RD into the number needed to treat for an additional beneficial outcome (NNTB) or the number needed to treat for an additional harmful outcome (NNTH) if the RR is statistically significant. We will express continuous data as mean differences (MDs) with 95% CIs. When different scales are used to measure the same conceptual outcome (e.g. disability), standardised mean differences (SMDs) will be calculated instead, with corresponding 95% CIs. We will back-translate SMDs to a typical scale (e.g. 0 to 10 for pain) by multiplying the SMD by a typical among-person standard deviation (e.g. the standard deviation of the control group at baseline from the most representative trial).


Unit of analysis issues

We expect the unit of analysis in this review to be individual patients, but will be alert to other potential unit of analyses issues, such as the inclusion of patients undergoing more than one intervention, repeated observation from more than one time-point, and multiple observations for the same outcome (e.g. total adverse events). Regarding multiple time points observation, we aim to additionally perform analysis based on different periods of follow-up. Should any such issues arise, we will seek further advice from methodological experts. Where multiple trial arms are reported in a single trial, we will include only the relevant arms. If two comparisons (e.g. drug A versus placebo and drug B versus placebo) are combined in the same meta-analysis, we will halve the control group to avoid double-counting.


Dealing with missing data

When necessary, we will seek missing data, particularly denominators and standard deviations, from the authors of the primary studies. We will perform intention-to-treat (ITT) analyses wherever possible. We will not impute missing standard deviations (SDs) unless they can be inferred from standard error (SE), CI, t values and P values. If missing data is imputed, we will undertake a sensitivity analysis to explore the effect of including the imputed data.


Assessment of heterogeneity

Besides the visual inspection of forest plot analyses, we will examine statistical heterogeneity using the Chi2 statistic with significance set at P < 0.1. We will also use the I2 statistic to estimate total variation across studies that is due to heterogeneity rather than chance or sampling error where 0% to 40% might not be important, 30% to 60% may represent moderate levels, 50% to 75% may represent substantial heterogeneity and 75% to 100% considerable heterogeneity (Higgins 2003; Deeks 2011). We will assess possible sources of substantial heterogeneity (I2 > 50%) by sensitivity and subgroup analyses.


Assessment of reporting biases

Provided data can be pooled from 10 or more trials, we will examine potential publication biases by producing funnel plots (Egger 1997). For studies published after 1st July 2005, we will screen the International Clinical Trials Registry Platform (ICTRP) search portal of the World Health Organisation (WHO) ( for the a priori trial protocol. We will evaluate whether selective reporting of outcomes is present (outcome reporting bias).


Data synthesis

We intend to perform all the meta-analyses using random-effects model, because it was suggested that this model is the most appropriate for the majority of meta-analyses when studies included in the analyses were gathered from the published literature, in which circumstance, the trial data typically would not be in accordance with the assumptions of the fixed-effect model, i.e., all studies share a common true effect and all observed variation only reflects sampling error. Therefore, while producing the same point estimate as fixed-effect model, random-effects model tends to produce more conservative confidence intervals better reflecting the uncertainty around the point estimate (Borenstein 2009; Schmidt 2009).


Subgroup analysis and investigation of heterogeneity

We will assess possible sources of heterogeneity using subgroup analyses as follows, which may be conducted based on:

  1. first generation versus second and third generation bisphosphonates;
  2. dosage of bisphosphonates (total amount taken per day, e.g. 5mg/day and 15 mg/day);
  3. location of the study joint (hands may have a different response in that they are not weight-bearing joints as the hip and knee);
  4. duration of OA (≤ 6 months, 7 to 24 months, and ≥24 month).


Sensitivity analysis

We plan to perform sensitivity analyses in order to explore the influence of the following factors on effect size, by repeating the analysis and:

  1. excluding unpublished studies;
  2. excluding studies with unclear or high risk of bias relating to non concealment of allocation, non blinding of outcome assessors, non-ITT analysis, incomplete outcome data;
  3. using different measures of calculating treatment effects (e.g. MDs versus SMDs) and different statistical models (fixed-effect versus random-effects models).


'Summary of findings' table

We will illustrate key information in terms of the quality of evidence, the magnitude of treatment effect of interest, and an overall grading of the evidence on the basis of the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach using 'Summary of findings' tables (Schünemann 2011), which may include the following outcomes.

  1. Pain relief
  2. Physical function
  3. Quality of life
  4. Radiographic joint structure
  5. Number of patients with serious adverse events
  6. Number of patients experiencing any adverse event
  7. Patients who withdrew because of adverse events

Two authors (ZQ and ZDT) will independently assess the quality of the evidence. We will use the five GRADE considerations (study limitations, consistency of effect, imprecision, indirectness and publication bias) to assess the quality of a body of evidence as it relates to the studies which contribute data to the meta-analyses for the prespecified outcomes. We will use methods and recommendations described in the Cochrane Handbook for Systematic Reviews of Interventions (Schünemann 2011b). We will justify all decisions to down- or up-grade the quality of studies using footnotes and we will make comments to aid reader's understanding of the review where necessary.

We will provide the absolute percent difference, the relative percent change from baseline, and the number needed-to-treat (NNT) (the NNT will be provided only when the outcome shows a statistically significant difference).

For dichotomous outcomes, such as serious adverse events, we will calculate the NNT from the control group event rate and the relative risk using the Visual Rx NNT calculator (Cates 2008). For continuous measures, we will calculate the NNT using the Wells calculator (available at the Cochrane Musculoskeletal Group Editorial office).

For dichotomous outcomes, we will calculate the absolute risk difference using the 'risk difference statistic in the Cochrane Collaboration's statistical software, Review Manager 2013, and express the result as a percentage. For continuous outcomes, we will calculate the absolute benefit as the improvement in the intervention group minus the improvement in the control group, in the original units.

We will calculate the relative percent change for dichotomous data as 'Risk ratio - 1' and express it as a percentage. For continuous outcomes, we will calculate the relative difference in the change from baseline as the absolute benefit divided by the baseline mean of the control group.


Interpreting results and reaching conclusions

We will follow the guidelines in the Cochrane Handbook for Systematic Reviews of Interventions for interpreting results and will be aware of distinguishing a lack of evidence of effect from a lack of effect (Higgins 2011b). We will base our conclusions only on findings from the quantitative or narrative synthesis of included studies for this review. We will avoid making recommendations for practice and our implications for research will suggest priorities for future research and outline what the remaining uncertainties are in the area.


Assessment of bias in conducting the systematic review

We will conduct the review according to this published protocol and report any deviations from it in the 'Differences between protocol and review' section of the systematic review.



  1. Top of page
  2. Background
  3. Objectives
  4. Methods
  5. Acknowledgements
  6. Appendices
  7. Contributions of authors
  8. Declarations of interest
  9. Sources of support

Our gratitude goes to Elizabeth Ghogomu, Lara Maxwell, Laura Laslett for valuable comments and help about the protocol. We also acknowledge the help of Renea Johnston along the way.



  1. Top of page
  2. Background
  3. Objectives
  4. Methods
  5. Acknowledgements
  6. Appendices
  7. Contributions of authors
  8. Declarations of interest
  9. Sources of support

Appendix 1. MEDLINE search strategy

  2. osteoarthr$.tw.
  3. (degenerative adj2 arthritis).tw.
  4. or/1-3
  5. exp Diphosphonates/
  6. biphosphonate$.mp.
  7. bisphosphanate$.mp.
  8. diphosphonate$.mp.
  9. diphosphanate$.mp.
  10. exp Etidronic Acid/
  11. etidronate$.mp.
  12. exp Clodronic Acid/
  13. clodronate$.mp.
  14. pamidronate$.mp.
  15. exp Alendronate/
  17. risedronate$.mp.
  18. tiludronate$.mp.
  19. ibandronate$.mp.
  20. zoledronate$.mp.
  21. incadronate$.mp.
  22. olpadronate$.mp.
  23. neridronate$.mp.
  24. exp Zoledronic Acid/
  25. 5 or 6 or 7 or 8 or 9 or 10 or 11 or 12 or 13 or 14 or 15 or 16 or 17 or 18 or 19 or 20 or 21 or 22 or 23 or 24
  26. 4 and 25
  27. Randomized controlled
  28. Controlled clinical
  29. randomized.ab.
  30. placebo.ab.
  31. Drug therapy.fs.
  32. randomly.ab.
  33. trial.ab.
  34. groups.ab.
  35. or/28-33
  36. exp Animals/ not Humans/
  37. 34 not 35
  38. 26 and 37


Contributions of authors

  1. Top of page
  2. Background
  3. Objectives
  4. Methods
  5. Acknowledgements
  6. Appendices
  7. Contributions of authors
  8. Declarations of interest
  9. Sources of support

Wei Yang, Wei Chai, Datong Zeng, Jiying Chen, Cheng Sun, Qi Zhuo and Yan Wang contributed equally to the preparation of the protocol. Yan Wang will be the guarantor of the review.


Declarations of interest

  1. Top of page
  2. Background
  3. Objectives
  4. Methods
  5. Acknowledgements
  6. Appendices
  7. Contributions of authors
  8. Declarations of interest
  9. Sources of support

None known.


Sources of support

  1. Top of page
  2. Background
  3. Objectives
  4. Methods
  5. Acknowledgements
  6. Appendices
  7. Contributions of authors
  8. Declarations of interest
  9. Sources of support

Internal sources

  • Chinese PLA General Hospital, China.


External sources

  • National Natural Science Foundation of China, China.


Additional references

  1. Top of page
  2. Abstract
  3. Background
  4. Objectives
  5. Methods
  6. Acknowledgements
  7. Appendices
  8. Contributions of authors
  9. Declarations of interest
  10. Sources of support
  11. Additional references
ACR 2008
Altman 1986
Arden 2006
Bartels 2007
Beek 2003
  • van Beek ER, Cohen LH, Leroy IM, Ebetino FH, Löwik CW, Papapoulos SE. Differentiating the mechanisms of antiresorptive action of nitrogen containing bisphosphonates. Bone and Mineral Research 2003;33(5):805-11.
Bellamy 2006a
  • Bellamy N, Campbell J, Welch V, Gee TL, Bourne R, Wells GA. Intraarticular corticosteroid for treatment of osteoarthritis of the knee. Cochrane Database of Systematic Reviews 2006, Issue 2. [DOI: 10.1002/14651858.CD005328.pub2]
Bellamy 2006b
  • Bellamy N, Campbell J, Welch V, Gee TL, Bourne R, Wells GA. Viscosupplementation for the treatment of osteoarthritis of the knee. Cochrane Database of Systematic Reviews 2006, Issue 2. [DOI: 10.1002/14651858.CD005321.pub2]
Bendele 2005
Bingham 2006
  • Bingham CO 3rd, Buckland-Wright JC, Garnero P, Cohen SB, Dougados M, Adami S, et al. Risedronate decreases biochemical markers of cartilage degradation but does not decrease symptoms or slow radiographic progression in patients with medial compartment osteoarthritis of the knee: Results of the two-year multinational knee osteoarthritis structural arthritis study. Arthritis & Rheumatism 2006;54(11):3494–507.
Bondeson 2011
Borenstein 2009
  • Borenstein M, Hedges LV, Higgins JPT, Rothstein HR. Fixed-effect versus random-effects models. Introduction to Meta-Analysis. Wiley, 2009:61-102.
Brouwer 2007
Cates 2008
  • Cates C. Visual Rx. 3. Available from, 2008.
Cepeda 2006
Conaghan 2011
  • Conaghan PG,  Hunter DJ,  Maillefert JF,  Reichmann WM,  Losina E. Summary and recommendations of the OARSI FDA osteoarthritis Assessment of Structural Change Working Group. Osteoarthritis Cartilage 2011;19(5):606-10.
Creamer 1997
da Costa 2012
Deeks 2011
  • Deeks JJ, Higgins JPT, Altman DG (editors). Chapter 9: Analysing data and undertaking meta-analyses. In: Higgins JPT, Green S (editors). Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 (updated March 2011). The Cochrane Collaboration, 2011. Available from
Drake 2008
Dunlop 2001
Egger 1997
Fajardo 2005
FDA 2011
  • U.S. Food, Drug Administration. Bisphosphonates (marketed as Actonel, Actonel+Ca, Aredia, Boniva, Didronel, Fosamax, Fosamax+D, Reclast, Skelid, and Zometa) Information. (accessed 26 March 2013).
Felson 2003
  • Felson DT, McLaughlin S, Goggins J, LaValley MP, Gale ME, Totterman S, et al. Bone marrow edema and its relation to progression of knee osteoarthritis. Annals of Internal Medicine 2003;139(5 Pt 1):330-6.
Felson 2007
Fidelix 2009
  • Fidelix TS, Soares B, Fernandes Moça Trevisani V. Diacerein for osteoarthritis. Cochrane Database of Systematic Reviews 2009, Issue 1. [DOI: 10.1002/14651858.CD005117.pub2]
Fleisch 1998
Fransen 2008
Fransen 2009
Frith 2001
Guermazi 2003
Hayami 2004
Haynes 2002
Higgins 2003
Higgins 2011a
  • Higgins JPT, Altman DG, Sterne JAC (editors). Chapter 8: Assessing risk of bias in included studies. In: Higgins JPT, Green S (editors). Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 (updated March 2011). The Cochrane Collaboration, 2011. Available from
Higgins 2011b
  • Higgins JPT, Green S (editors). Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 [updated March 2011]. The Cochrane Collaboration, 2011. Available from
Hill 2001
  • Hill CL, Gale DG, Chaisson CE, Skinner K, Kazis L, Gale ME, et al. Knee effusions, popliteal cysts, and synovial thickening: association with knee pain in osteoarthritis. Journal of Rheumatology 2001;28(6):1330-7.
Hochberg 2012
Hunter 2011
Jacobs 2001
  • Jacobs W, Anderson PG, van Limbeek J, Wymenga AAB. Mobile bearing vs fixed bearing prostheses for total knee arthroplasty for post-operative functional status in patients with osteoarthritis and rheumatoid arthritis. Cochrane Database of Systematic Reviews 2001, Issue 2. [DOI: 10.1002/14651858.CD003130.pub2]
Jacobs 2005
Jolles 2006
Kimmel 2007
  • Kimmel DB. Mechanism of action, pharmacokinetic and pharmacodynamic profile, and clinical applications of nitrogen-containing bisphosphonates. Journal of Dental Research 2007;86(11):1022-33.
Kurtz 2007
  • Kurtz S, Ong K, Lau E, Mowat F, Halpern M. Projections of primary and revision hip and knee arthroplasty in the United States from 2005 to 2030. Journal of Bone & Joint Surgery 2007;89(4):780-5.
Laupattarakasem 2008
Lawrence 2004
Lefebvre 2011
  • Lefebvre C, Manheimer E, Glanville J. Chapter 6: Searching for studies. In: Higgins JPT, Green S (editors). Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 (updated March 2011). The Cochrane Collaboration, 2011. Available from
Li 1997
  • Li B, Aspden RM. Mechanical and material properties of the subchondral bone plate from the femoral head of patients with osteoarthritis or osteoporosis. Annals of Rheumatic Diseases 1997;56(4):247-54.
Lo 2005
Longo 2011
  • Longo DL, Fauci AS, Kasper DL, Hauser SL, Jameson JL, Loscalzo J. Osteoarthritis. Harrison's Principles of Internal Medicine. 18th Edition. McGraw-Hill Education, 2011:Chapter 332.
McAlindon 2000
Messent 2005
  • Messent EA, Buckland-Wright JC, Blake GM. Fractal analysis of trabecular bone in knee osteoarthritis (OA) is a more sensitive marker of disease status than bone mineral density (BMD). Calcified Tissue International 2005;76(6):419-25.
Meyer 2001a
  • Meyer JM,  Dansereau SM,  Farmer RW,  Jeans GL,  Prenger MC. Bisphosphonates structurally similar to risedronate (actonel) slow disease progression in the guinea pig model of primary osteoarthritis [abstract]. Arthritis & Rheumatism 2001;44 Supplement 9:1527.
Meyer 2001b
  • Meyer J,  Farmer R,  Prenger MC. Risedronate but not alendronate slows disease progression in the guinea pig model of primary osteoarthritis [abstract]. Journal of Bone & Mineral Research 2001;16(Supplement 1):SA472.
Muehleman 2002
  • Muehleman C,  Green J,  Williams JM,  Kuettner KE,  Thonar EJ,  Sumner DR. The effect of bone remodeling inhibition by zoledronic acid in an animal model of cartilage matrix damage. Osteoarthritis and Cartilage  2002;10(3):226-33.
Muraoka 2007
Nakama 2012
  • Nakama GY, Peccin MS, Almeida GJM, Lira Neto ODA, Queiroz AAB, Navarro RD. Cemented, cementless or hybrid fixation options in total knee arthroplasty for osteoarthritis and other non-traumatic diseases. Cochrane Database of Systematic Reviews 2012, Issue 10. [DOI: 10.1002/14651858.CD006193.pub2]
Nüesch 2009
  • Nüesch E, Rutjes AWS, Husni E, Welch V, Jüni P. Oral or transdermal opioids for osteoarthritis of the knee or hip. Cochrane Database of Systematic Reviews 2009, Issue 4. [DOI: 10.1002/14651858.CD003115.pub3]
Plotkin 2008
Ravaud 1998
  • Ravaud P, Giraudeau B, Auleley GR, Drape JL, Rousselin B, Paolozzi L, et al. Variability in knee radiographing: implication for definition of radiological progression in medial knee osteoarthritis. Annals of Rheumatic Diseases 1998;57(10):624-9.
Reichenbach 2010
Review Manager 2013
  • The Nordic Cochrane Centre, The Cochrane Collaboration. Review Manager (RevMan). 5.2. Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration, 2013.
Rogers 2004
Schmidt 2009
Schünemann 2011
  • Schünemann HJ, Oxman AD, Higgins JPT, Vist GE, Glasziou P, Guyatt GH. Chapter 11: Presenting results and ‘Summary of findings' tables. In: Higgins JPT, Green S (editors), Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 (updated March 2011). The Cochrane Collaboration, 2011. Available from
Schünemann 2011b
  • Schünemann HJ, Oxman AD, Vist GE, Higgins JPT, Deeks JJ, Glasziou P, et al. Chapter 12: Interpreting results and drawing conclusions. In: Higgins JPT, Green S (editors), Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 (updated March 2011). The Cochrane Collaboration, 2011. Available from
Singh 2006
Spector 2003
Symmons 2006
  • Symmons D, Mathers C, Pfleger B. Global burden of osteoarthritis in the year 2000. (accessed 26 March 2013).
Tissakht 1996
Towheed 2006
Towheed 2009
Verbruggen 2006
Verhagen 2007
Wells 2008a
  • Wells GA, Cranney A, Peterson J, Boucher M, Shea B, Welch V, et al. Alendronate for the primary and secondary prevention of osteoporotic fractures in postmenopausal women. Cochrane Database of Systematic Reviews 2008, Issue 1. [DOI: 10.1002/14651858.CD001155.pub2]
Wells 2008b
  • Wells GA, Cranney A, Peterson J, Boucher M, Shea B, Welch V, et al. Etidronate for the primary and secondary prevention of osteoporotic fractures in postmenopausal women. Cochrane Database of Systematic Reviews 2008, Issue 1. [DOI: 10.1002/14651858.CD003376.pub3]
Wells 2008c
  • Wells GA, Cranney A, Peterson J, Boucher M, Shea B, Welch V, et al. Risedronate for the primary and secondary prevention of osteoporotic fractures in postmenopausal women. Cochrane Database of Systematic Reviews 2008, Issue 1. [DOI: 10.1002/14651858.CD004523.pub3]
Zhang 2002
  • Zhang Y, Niu J, Kelly-Hayes M, Chaisson CE, Aliabadi P, Felson DT. Prevalence of symptomatic hand osteoarthritis and its impact on functional status among the elderly: The Framingham Study. American Journal of Epidemiology 2002;156(11):1021-7.
Zhang 2008
  • Zhang W, Moskowitz RW, Nuki G, Abramson S, Altman RD, Arden N, et al. OARSI recommendations for the management of hip and knee osteoarthritis, Part II: OARSI evidence-based, expert consensus guidelines. Osteoarthritis and Cartilage 2008;16(2):137-62.
Zhang 2010
  • Zhang W, Nuki G, Moskowitz RW, Abramson S, Altman RD, Arden NK, et al. OARSI recommendations for the management of hip and knee osteoarthritis: Part III: changes in evidence following systematic cumulative update of research published through January 2009. Osteoarthritis and Cartilage 2010;18(4):476-99.
Zhuo 2012