The Journal of Bone & Joint Surgery

The Journal of Bone & Joint Surgery, Volume 87, Issue 2

Scientific Articles | February 01, 2005

Effect of Bisphosphonates on Periprosthetic Bone Mineral Density After Total Joint Arthroplasty: A Meta-Analysis

Mohit Bhandari, MD, MSc, FRCSC¹; Sohail Bajammal, MD¹; Gordon H. Guyatt, MD, MSc¹; Lauren Griffith, MSc¹; Jason W. Busse, DC, MSc¹; Holger Schünemann, MD, PhD¹; Thomas A. Einhorn, MD²

¹ Department of Clinical Epidemiology and Biostatistics, McMaster University Health Sciences Center, Room 2C9, 1200 Main Street West, Hamilton, ON L8N 3Z5, Canada. E-mail address for M. Bhandari: bhandam@mcmaster.ca
² Department of Orthopaedic Surgery, Boston University Medical Center, 720 Harrison Avenue, Suite 808, Boston, MA 02118

View Disclosures and Other Information

In support of their research or preparation of this manuscript, one or more of the authors received grants or outside funding from a Canadian Institutes of Health Research Fellowship Award. In addition, one or more of the authors received payments or other benefits or a commitment or agreement to provide such benefits from commercial entities (Merck and Novartis). No commercial entity paid or directed, or agreed to pay or direct, any benefits to any research fund, foundation, educational institution, or other charitable or nonprofit organization with which the authors are affiliated or associated.

Investigation performed at the Department of Clinical Epidemiology and Biostatistics, McMaster University, Hamilton, Ontario, Canada, and the Department of Orthopaedic Surgery, Boston University Medical Center, Boston, Massachusetts

The Journal of Bone and Joint Surgery, Incorporated

J Bone Joint Surg Am, 2005 Feb 01;87(2):293-301. doi: 10.2106/JBJS.D.01772

5 Recommendations (Recommend) | 3 Comments | Saved by 3 Users Save Case

Article

Comments (2)

text A A A

Abstract

Background: Periprosthetic bone loss following total joint arthroplasty may threaten the survival of the implant. Bisphosphonates are effective in reducing bone loss in conditions associated with accelerated bone turnover. To determine the current understanding of the effect of bisphosphonates on periprosthetic bone mineral density after total joint arthroplasty, we conducted computerized searches for randomized controlled trials evaluating the use of bisphosphonates in patients treated with primary total joint arthroplasty.

Methods: We searched MEDLINE, EMBASE, the Cochrane Central Register of Controlled Trials, the Cochrane Database of Systematic Reviews, the web site of the United Kingdom National Research Register, and the archives of the American Academy of Orthopaedic Surgeons annual meetings (1989 through 2003), and we conducted hand searches of the bibliographies of relevant articles. We assessed methodological quality and abstracted relevant data. When necessary, we contacted authors to provide additional information.

Results: Of 386 citations that were initially identified, six (five complete papers and one abstract), which included a total of 290 patients, met our inclusion criteria. Those papers showed that significantly less periprosthetic bone loss had occurred in the bisphosphonate-treated patients than in the control patients at three months (152 patients; weighted mean difference, 3.3%; 95% confidence interval, 1.9% to 4.7%; p < 0.01), six months (248 patients; weighted mean difference, 4.5%; 95% confidence interval, 1.6% to 7.4%; p < 0.001), and twelve months (197 patients; weighted mean difference, 4.2%; 95% confidence interval, 1.5% to 6.9%; p = 0.03). Bisphosphonates appeared to have a larger effect on bone loss following arthroplasties with cement than on bone loss following arthroplasties without cement (difference, 0.1%, 5%, and 5.4% at three, six, and twelve months; significant difference [p < 0.001] at one year only) and a larger effect on bone loss following total knee arthroplasties than on bone loss following total hip arthroplasties (difference, 4.1%, 11.5%, and 7.1% at three, six, and twelve months; significant difference [p < 0.001] at six months only). None of the studies related the effects of bisphosphonates on bone mineral density to clinically relevant outcomes.

Conclusions: A meta-analysis of six randomized controlled trials suggested that bisphosphonates have a beneficial effect with regard to maintaining more periprosthetic bone mineral density than that in controls. However, the limitations of the available studies and the lack of analyses of clinically relevant outcomes (functional outcomes, revision rates, and quality of life) necessitate the planning and conduct of a sufficiently sized, methodologically sound study with clinically relevant end points. Until this has been done, the current evidence regarding the beneficial effects of bisphosphonates on periprosthetic bone after total joint arthroplasty should be interpreted with caution.

Level of Evidence: Therapeutic Level I. See Instructions to Authors for a complete description of levels of evidence.

Figures in this Article

Periprosthetic bone loss is common following hip and knee arthroplasties, and some studies have suggested that it may adversely affect the longevity of the im-plants^1-3. The factors that lead to periprosthetic bone loss in the first year after an arthroplasty are not clearly understood but are thought to be associated with an increase in osteoclast activity, resulting from stress-shielding, and the generation of wear particles^3,4.

Bisphosphonates effectively inhibit bone resorption in both animals and humans⁵. A meta-analysis of randomized trials showed that postmenopausal women treated with either alendronate or risedronate have a reduction in the prevalence of vertebral and nonvertebral fractures that are associated with a decrease in bone density⁶. Randomized trials have also demonstrated that bisphosphonates reduce skeletal fragility in patients with proven malignant disease and bone metastases⁷.

Recently, there have been a number of case-control randomized trials of the use of bisphosphonates in patients treated with total hip or knee arthroplasty^8-15. These trials have overcome the limitations of earlier studies by decreasing bias through randomization, concealment of allocation, and blinding. However, small sample sizes (range, thirteen to ninety-six patients) and wide confidence intervals of the results have limited the clinical relevance of the findings.

A systematic approach to the literature, with the potential of statistical pooling, theoretically allows stronger inferences and more secure guides to clinical practice. We conducted a systematic review and meta-analysis of randomized controlled trials to determine the effect of bisphosphonates on bone mineral density following total joint arthroplasty.

Materials and Methods

Eligibility Criteria

We identified articles in which (1) the target population consisted of patients with hip or knee arthritis undergoing total joint arthroplasty, (2) the intervention was the use of oral or intravenous bisphosphonates following total joint arthroplasty, (3) the outcome measure was bone mineral density monitored with dual-energy x-ray absorptiometry, and (4) the study was a published or unpublished randomized controlled trial.

Study Identification

We conducted, in duplicate, a computerized search of the electronic databases EMBASE (1980 to 2003) and OVID MEDLINE and PubMed MEDLINE (1966 to August 2003) with the following terms and Boolean operators: (alendronate OR pamidronate OR bisphosphonate) AND (arthroplasty OR hip arthroplasty OR knee arthroplasty OR joint replacement OR joint prosthesis). We also searched the Cochrane Database of Systematic Reviews, the Cochrane Central Register of Controlled Trials, the web site of the United Kingdom National Research Register (), and abstracts of archives of the annual meetings of the American Academy of Orthopaedic Surgeons () for 1989 through 2003 to identify additional studies. One of us reviewed the bibliography of each article that met our inclusion criteria for additional relevant studies. Corresponding authors and field experts provided additional potentially relevant studies. We reviewed the titles, and, if the title suggested any possibility that the article might meet eligibility criteria, we retrieved the abstract. Two of us reviewed the abstracts and chose potentially eligible studies for retrieval. Only the methods section was reviewed to determine whether the complete article was eligible; thus, the reviewer was blinded to the author, institution, journal, and results.

Assessment of Study Quality

Two of us independently assessed each published study for the quality of the study design with use of a 21-point scale¹⁶. This scale grades the reporting of studies on the basis of eligibility criteria, adequacy of randomization, description of therapies, assessment of outcomes, and statistical analysis. Two of us also graded the methodological quality of each study with respect to randomization (present and concealed); blinding of patients, clinicians, and those assessing outcomes; and proportion of patients lost to follow-up. In both cases, the reviewers resolved disagreements by discussion to achieve consensus.

Data Extraction

For each eligible study, one of us extracted relevant data and checked the accuracy. Specifically, we abstracted the type of arthroplasty, sizes of the intervention and control groups, demographic data (age, sex, and body mass index), baseline bone mineral density measured with dual-energy x-ray absorptiometry, intervention protocol, duration of the study, loss to follow-up, study outcomes measured (percentage changes in bone mineral density at three, six, and twelve months compared with baseline bone mineral density, when available), and source of funding. Only one article was written in a language other than English (German), and it was translated by a medical translator. If the mean bone mineral density was not reported in the text or a table in the article, it was extrapolated from accompanying graphs.

Accuracy of Data Abstraction

We asked the first or corresponding author of each eligible study to verify the accuracy of our data abstraction as well as our scores for the methodology of their study. Authors provided additional information when necessary. All of the authors confirmed the accuracy of the data extraction and provided additional information that had not been reported in the published papers.

Data Analysis

For each study, we calculated the mean differences in bone mineral density (and 95% confidence intervals) between treatment and control groups at three, six, and twelve months. We used the total bone mineral density (in grams per square centimeter) of the proximal femoral metaphysis of the patients treated with hip arthroplasty and the total distal femoral and proximal tibial metaphyseal bone mineral density of those treated with knee arthroplasty. We pooled data across studies by using weighted mean differences (and 95% confidence intervals). In instances in which a standard error for a mean difference in bone mineral density was not reported, we calculated the standard error by converting the p value to a z-score and solving for the standard error with the formula: z = mean difference/standard error. The methods of Hedges and Olkin¹⁷ were used to test for significance and homogeneity. To assess for publication bias, we conducted funnel plots to examine sample size versus treatment effect across included studies. A funnel plot depicts the relationship between precision in estimating the underlying treatment effect and the sample size of component studies. Results from small studies scatter widely at the bottom of the graph, whereas the spread narrows for larger studies. In the absence of bias, the plot resembles a symmetrical inverted funnel. Conversely, if there is bias, funnel plots are often skewed and asymmetrical¹⁸.

Evaluation of Heterogeneity

Prior to analyzing the data, we developed hypotheses regarding potential sources of heterogeneity. We defined study heterogeneity as a significant test of heterogeneity (p < 0.1) and differences in the treatment effects across studies. We used p < 0.1 for tests of heterogeneity to maintain a lower-than-conventional threshold for exploring sources of heterogeneity, given that such tests are typically underpowered. We hypothesized that heterogeneity may be due to differences in the type of arthroplasty (with or without cement), region (knee or hip), type of bisphosphonate (oral alendronate or intravenous pamidronate), and study quality score (<70 or =70 points). For each potential determinant of heterogeneity, weighed mean differences in bone mineral density were compared.

Results

Study Identification

Our literature search identified 385 potentially relevant citations, 210 from MEDLINE, forty-four from EMBASE, five from the Cochrane Central Register of Controlled Trials, one from the Cochrane Database of Systematic Reviews, four from the archives of the annual meetings of the American Academy of Orthopaedic Surgeons, two from the web site of the United Kingdom National Research Register, and 119 from the bibliographies of relevant articles (Table I). The application of the eligibility criteria eliminated all but nine studies. Two of those studies proved to be duplicate presentations (abstracts) at international meetings, and one article reported data on the same subset of patients, with longer follow-up, as had been reported on in another study. Thus, six randomized controlled trials proved eligible for our study (Table II and Appendix). The weighted kappa on overall agreement based on the application of study eligibility criteria to study titles was 0.85 (95% confidence interval, 0.74 to 0.97).

View Larger

TABLE I Search Results

Database	Results (no. of articles)	Search Process	Reference No. of Articles Accepted
OVID MEDLINE (1966 to wk 2 of August 2003)	15	Hand search for inclusion criteria	8-10,12
PubMed MEDLINE (1966 to August 2003)	195	Electronic limit to human and clinical trials, hand search for inclusion criteria	8-10,12
EMBASE (1980 to wk 34 of 2003)	44	Hand search for inclusion criteria	9,10,12
Cochrane Central Register of Controlled Trials (2nd quarter of 2003)	5	All met criteria	9-13
Cochrane Database of Systematic Reviews	1	Hand search for inclusion criteria	None
Archives of annual meetings of American Academy of Orthopaedic Surgeons (1989-2003)	4	Hand search for inclusion criteria	11,13
United Kingdom National Research Register web site (to August 2003)	2	Hand search for inclusion criteria	12
Hand search of bibliography of included articles	119	Hand search for inclusion criteria	11

Add to My JBJS

TABLE I Search Results

Database	Results (no. of articles)	Search Process	Reference No. of Articles Accepted
OVID MEDLINE (1966 to wk 2 of August 2003)	15	Hand search for inclusion criteria	8-10,12
PubMed MEDLINE (1966 to August 2003)	195	Electronic limit to human and clinical trials, hand search for inclusion criteria	8-10,12
EMBASE (1980 to wk 34 of 2003)	44	Hand search for inclusion criteria	9,10,12
Cochrane Central Register of Controlled Trials (2nd quarter of 2003)	5	All met criteria	9-13
Cochrane Database of Systematic Reviews	1	Hand search for inclusion criteria	None
Archives of annual meetings of American Academy of Orthopaedic Surgeons (1989-2003)	4	Hand search for inclusion criteria	11,13
United Kingdom National Research Register web site (to August 2003)	2	Hand search for inclusion criteria	12
Hand search of bibliography of included articles	119	Hand search for inclusion criteria	11

View Larger

TABLE II Study Results

Mean Change in Bone Mineral Density Compared with Baseline† (%)

Baseline Bone Mineral Density^*(g/cm²)

3 mo

6 mo

12 mo

Study

Parameter

Intervention

Control

Intervention

Control

% Difference Between Intervention and Control‡

Intervention

Control

% Difference Between Intervention and Control‡

Intervention

Control

% Difference Between Intervention and Control‡

Wang et al.¹³, 2003 (n = 96: 48 with intervention, 48 controls)

Net bone mineral density, proximal part of tibia

0.90 ± 0.16

0.85 ± 0.18

9.4

-6.5

15.9 (5.5-26.3)

5.4

-3.6

9.0 (1.3-16.7)

Wilkinson et al.¹², 2001 (n = 54: 25 with intervention, 29 controls)

Net bone mineral density, proximal part of femur

1.77 ± 0.15

1.77 ± 0.27

-1.1

-4.0

2.9 (1.1-4.7)

-1.1§

-2.82§

1.72 (-0.34-3.8)

Venesmaa et al.¹⁰, 2001 (n = 13: 8 with intervention, 5 controls)

Total bone mineral density in Gruen³⁸ femoral regions of interest

1.60 ± 0.25

1.58 ± 0.12

-3.13

-9.49

6.4 (0.13-12.7)

-2.5

-9.49

7.0 (0.4-13.6)

Soininvaara et al.⁸, 2002 (n = 19: 8 with intervention, 11 controls)

Total metaphyseal bone mineral density

1.45 ± 0.17

1.39 ± 0.22

-3.45

-10.79

7.3 (0.03-14.6)

-3.45

-15.11

11.7 (0.12-23.2)

-4.83

-20.14

15.3 (0.16-30.5)

Hennigs et al.⁹, 2002 (n = 66: 42 with intervention, 24 controls)

Group 1 (n = 21; alendronate 10 mg/day, 10 wk)

Total bone mineral density in Gruen³⁸ femoral regions of interest

1.25 ± 0.18

1.20 ± 0.16

-1.77§

-4.77§

3.0 (0.3-5.7)

-1.9% ± 2.8

-5.4

3.5 (-0.66-7.66)

-1.42§

-3.7

2.3 (0.22-4.3)

Group 2 (n = 21; alendronate 20 mg/day, 5 wk)

Total bone mineral density in Gruen³⁸ femoral regions of interest

1.27 ± 0.14

1.20 ± 0.16

-1.23§

-4.77§

3.5 (-0.8-7.9)

-2.54§

-5.4

2.9 (-0.49-6.3)

-1.85§

-3.7

1.9 (-0.26-4.1)

Lyons¹¹, 1999 (n = 16#: 8 with intervention, 8 controls)

Total bone mineral density in Gruen³⁸ femoral regions of interest

1.49

1.50

1.96

-4.79

6.8 (3.1-10.4)

Weighted mean difference

3.3 (1.9-4.7); p < 0.01

4.5 (1.6-7.4); p < 0.001

4.2 (1.5-6.9); p = 0.02

Heterogeneity

p = 0.67

p = 0.06

p = 0.03

^* The values are given as the mean and standard deviation. †NA = not available. ‡The 95% confidence interval is shown in parentheses. §The values were extrapolated from graphs. #The sixteen patients represent a subgroup evaluated soon after a total hip arthroplasty ("acute" group).

Add to My JBJS

TABLE II Study Results

Mean Change in Bone Mineral Density Compared with Baseline† (%)

Baseline Bone Mineral Density^*(g/cm²)

3 mo

6 mo

12 mo

Study

Parameter

Intervention

Control

Intervention

Control

% Difference Between Intervention and Control‡

Intervention

Control

% Difference Between Intervention and Control‡

Intervention

Control

% Difference Between Intervention and Control‡

Wang et al.¹³, 2003 (n = 96: 48 with intervention, 48 controls)

Net bone mineral density, proximal part of tibia

0.90 ± 0.16

0.85 ± 0.18

9.4

-6.5

15.9 (5.5-26.3)

5.4

-3.6

9.0 (1.3-16.7)

Wilkinson et al.¹², 2001 (n = 54: 25 with intervention, 29 controls)

Net bone mineral density, proximal part of femur

1.77 ± 0.15

1.77 ± 0.27

-1.1

-4.0

2.9 (1.1-4.7)

-1.1§

-2.82§

1.72 (-0.34-3.8)

Venesmaa et al.¹⁰, 2001 (n = 13: 8 with intervention, 5 controls)

Total bone mineral density in Gruen³⁸ femoral regions of interest

1.60 ± 0.25

1.58 ± 0.12

-3.13

-9.49

6.4 (0.13-12.7)

-2.5

-9.49

7.0 (0.4-13.6)

Soininvaara et al.⁸, 2002 (n = 19: 8 with intervention, 11 controls)

Total metaphyseal bone mineral density

1.45 ± 0.17

1.39 ± 0.22

-3.45

-10.79

7.3 (0.03-14.6)

-3.45

-15.11

11.7 (0.12-23.2)

-4.83

-20.14

15.3 (0.16-30.5)

Hennigs et al.⁹, 2002 (n = 66: 42 with intervention, 24 controls)

Group 1 (n = 21; alendronate 10 mg/day, 10 wk)

Total bone mineral density in Gruen³⁸ femoral regions of interest

1.25 ± 0.18

1.20 ± 0.16

-1.77§

-4.77§

3.0 (0.3-5.7)

-1.9% ± 2.8

-5.4

3.5 (-0.66-7.66)

-1.42§

-3.7

2.3 (0.22-4.3)

Group 2 (n = 21; alendronate 20 mg/day, 5 wk)

Total bone mineral density in Gruen³⁸ femoral regions of interest

1.27 ± 0.14

1.20 ± 0.16

-1.23§

-4.77§

3.5 (-0.8-7.9)

-2.54§

-5.4

2.9 (-0.49-6.3)

-1.85§

-3.7

1.9 (-0.26-4.1)

Lyons¹¹, 1999 (n = 16#: 8 with intervention, 8 controls)

Total bone mineral density in Gruen³⁸ femoral regions of interest

1.49

1.50

1.96

-4.79

6.8 (3.1-10.4)

Weighted mean difference

3.3 (1.9-4.7); p < 0.01

4.5 (1.6-7.4); p < 0.001

4.2 (1.5-6.9); p = 0.02

Heterogeneity

p = 0.67

p = 0.06

p = 0.03

Study Characteristics

The eligibility criteria, outcomes, and methodologies of the included studies are presented in the Appendix. All studies were randomized trials, and two of them were blinded (placebo). Calcium carbonate was utilized in the control group in three studies, and no medication was used in the control group in the other three. Follow-up was complete in four studies. The reviewers achieved excellent agreement, and the assessment of the study quality was excellent (intraclass correlation, 0.88; 95% confidence interval, 0.11 to 0.99). The kappa values for the various components of the study design (such as randomization and blinding of patients, clinicians, and those assessing outcomes; conduct of the statistical analysis; and follow-up) ranged from 0.45 to 1.0.

Four studies assessed the effect of a bisphosphonate on patients treated with total hip arthroplasty, and two assessed the effect on those treated with total knee arthroplasty. Two studies involved uncemented hip components, one involved cemented hip components, one study involved hybrid total hip arthroplasty (cemented and uncemented components), and two involved cemented knee components. The sample sizes of the trials ranged from thirteen to ninety-six patients. Alendronate (10 mg/day) was used in five studies, and pamidronate (a one-time infusion of 90 mg in 500 mL of normal saline solution on the fifth postoperative day) was used in one study. Investigators measured periprosthetic bone mineral density with dual-energy x-ray absorptiometry at baseline (within one to two weeks after the surgery). Four studies provided bone mineral density data at three months, five studies provided it at six months, and four studies provided it at twelve months. Follow-up was nearly complete across the studies (range, 96% to 100%).

Bone Mineral Density

All studies demonstrated an overall decline in bone mineral density in both the treatment and the control groups at three months following the surgery; however, the periprosthetic bone mineral density maintained at three months in the bisphosphonate-treated patients was significantly greater than that in the controls (152 patients; weighted mean difference, 3.3%; 95% confidence interval, 1.9 to 4.7, p < 0.01) (Table II) (Fig. 1). The significantly greater periprosthetic bone mineral density associated with the bisphosphonate treatment remained at six months (248 patients; weighted mean difference, 4.5%; 95% confidence interval, 1.6 to 7.4; p < 0.001) (Fig. 2). With regard to the bone mineral density at twelve months, two of the five studies demonstrated an overall increase in patients with bisphosphonate treatment compared with that in the controls. The overall treatment effect across studies at twelve months favored bisphosphonate treatment over the control groups (197 patients; weighted mean difference, 4.2%; 95% confidence interval, 1.5 to 6.9; p = 0.02) (Fig. 3).

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 1: Weighted mean difference (WMD) in bone mineral density between bisphosphonate-treatment and control groups at three months after joint arthroplasty. In the study by Hennigs et al.⁹, one group (Hennigs 1) was treated with 10 mg/day of alendronate for ten weeks, and another (Hennigs 2) was treated with 20 mg/day of alendronate for five weeks.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 2: Weighted mean difference (WMD) in bone mineral density between bisphosphonate-treatment and control groups at six months after joint arthroplasty. In the study by Hennigs et al.⁹, one group (Hennigs 1) was treated with 10 mg/day of alendronate for ten weeks, and another (Hennigs 2) was treated with 20 mg/day of alendronate for five weeks.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 3: Weighted mean difference (WMD) in bone mineral density between bisphosphonate-treatment and control groups at twelve months after joint arthroplasty. In the study by Hennigs et al.⁹, one group (Hennigs 1) was treated with 10 mg/day of alendronate for ten weeks, and another (Hennigs 2) was treated with 20 mg/day of alendronate for five weeks.

Neither the type of bisphosphonate (oral or intravenous) nor the study quality score significantly altered the effect of bisphosphonate on bone mineral density as compared with the bone density in the controls (see Appendix). However, at one year, the difference in the bone mineral density between the bisphosphonate and control groups was significantly greater (p < 0.001) after the arthroplasties with cement (weighted mean difference, 7.5%; 95% confidence interval, 4.3 to 10.7) than it was after the cementless arthroplasties (weighted mean difference, 2.1%; 95% confidence interval, 0.61 to 3.6). However, the weighted mean age of the patients who underwent arthroplasty with cement (68.0 years) was greater than that of the patients who underwent cementless arthroplasty (55.0 years). The effect of bisphosphonates on bone mineral density was significantly greater (p < 0.001) six months after the total knee arthroplasties (weighted mean difference, 14.0%; 95% confidence interval, 6.3 to 21.7) than it was six months after the hip arthroplasties (weighted mean difference, 2.5%; 95% confidence interval, 0.96 to 4.1); however, the difference was not significant at one year.

Funnel plots did not demonstrate evidence of publication bias at three, six, or twelve months.

Discussion

The primary finding of the meta-analysis, based on the six studies that met the eligibility criteria, is that administration of bisphosphonates for twenty-four to fifty-two weeks postoperatively probably reduces periprosthetic bone loss up to one year after total hip or total knee arthroplasty. Our review suggests that the effect of bisphosphonates on bone mineral density may be greater in patients who have undergone knee arthroplasty than in those who have undergone hip arthroplasty and greater in those treated with cemented components than in those treated with uncemented components.

Strengths and Limitations of the Study

The current overview met most of the methodological criteria for research overviews¹⁹. Specifically, we developed explicit inclusion and exclusion criteria, assessed the methodological quality of the studies, demonstrated the reproducibility of selection and assessment criteria, performed a quantitative analysis, and explored possible reasons for differences between the results of the studies. Authors of the individual studies verified our data abstraction. Our decision to include a trial published in a non-English-language journal improved the generalizability of our findings. We limited any potential selection bias^20,21 by rigorously searching many databases, bibliographies, and programs of annual orthopaedic meetings for all eligible published and unpublished articles. We further limited selection bias by conducting all aspects of the selection process in duplicate.

Despite our comprehensive search strategy, we cannot exclude the possibility that there was a publication bias²⁰ against studies that did not demonstrate a significant difference in effect between two treatments. However, our funnel plots did not suggest that this was the case. It also should be noted that the strict eligibility criteria used for this study may limit the generalizability of our findings to randomized trials in which dual-energy x-ray absorptiometry was used to monitor bone density.

Regardless of the rigor with which studies are identified, assessed, and analyzed, the results from statistical pooling are only as valid as the studies on which the analysis is based. Bias in clinical trials can be limited by randomization, concealment of allocation, blinding, and complete follow-up. The patients or surgeons were not blinded in four of the six trials in the current review^8-10,13. The reported estimates of treatment effect were smaller in the blinded studies than they were in the unblinded studies. Follow-up was complete in four of the six trials^8-10,13.

Critics of meta-analysis cite differences between metaanalyses of small trials and single large trials^22,23. LeLorier et al. evaluated the consistency of the results of meta-analyses with the results of single, large randomized trials on the same topics²². Among forty comparisons, they noted five inconsistent findings (13%). Thus, 87% of meta-analyses provided an accurate reflection of the results of single large trials. Egger et al. showed that, compared with single large trials on the same topic, meta-analyses tend to overestimate treatment effects¹⁸. They advocated the use of funnel plots to limit publication bias and discrepancies between meta-analyses and large trials. A simple analysis of funnel plots is a useful test for the likely presence of bias in meta-analyses, but the results from such an analysis should be interpreted with caution because its capacity to detect bias is limited when meta-analyses are based on a limited number of small trials. In the absence of a single large trial evaluating bisphosphonate treatment of patients treated with joint arthroplasty, our best estimate of effect was based on a meta-analysis of small randomized trials. Although funnel plots did not suggest publication bias, we interpret those plots with caution.

The association between industry funding and the results of medical and surgical randomized trials has been the topic of much debate^24-26. In the current review, one trial was funded by a commercial entity¹¹ and, in another trial, the drugs were given free by a commercial entity¹². The industryfunded trial demonstrated a significant benefit of the commercial entity's product, but Wilkinson et al.¹² reported no significant effects of pamidronate at six months.

The efficacy of bisphosphonates in limiting bone loss and preventing fractures in postmenopausal women has been well documented. Cranney et al. conducted a meta-analysis of eleven trials in which postmenopausal women had been randomly assigned to receive either alendronate or a placebo and bone density was measured for at least one year⁶. The pooled relative risk for vertebral fractures in patients given =5 mg of alendronate was 0.52 (95% confidence interval, 0.43 to 0.65), and the pooled relative risk of nonvertebral fractures in patients given =10 mg of alendronate was 0.51 (95% confidence interval, 0.38 to 0.69). Thus, alendronate reduced the risk of fracture by 48% and 49%, respectively, compared with the risk in controls. After three years of treatment with =10 mg of alendronate, the pooled estimate of the difference in the percentage change between alendronate and the placebo was 7.48% (95% confidence interval, 6.12 to 8.85) for the lumbar spine (at two to three years), 5.60% (95% confidence interval, 4.80 to 6.39) for the hip (at three to four years), 2.08% (95% confidence interval, 1.53 to 2.63) for the forearm (at two to four years), and 2.73% (95% confidence interval, 2.27 to 3.20) for the total body (at three years).

Ross et al. evaluated the efficacy of bisphosphonates in a meta-analysis that revealed a significant decrease in skeletal morbidity (fractures, the need for radiation therapy, and hypercalcemia) and an increase in the time to the first skeletal-related event in patients with metastatic bone disease⁷. In studies with follow-up of six months or more, bisphosphonates, compared with placebos, significantly reduced the odds ratio for vertebral fractures (0.69 [95% confidence interval, 0.57 to 0.84], based on seven studies), nonvertebral fractures (0.65 [95% confidence interval, 0.54 to 0.79], based on nine studies), combined fractures (0.65 [95% confidence interval, 0.55 to 0.78], based on seven studies), radiation therapy (0.67 [95% confidence interval, 0.57 to 0.79], based on eight studies), and hypercalcemia (0.54 [95% confidence interval, 0.36 to 0.81], based on eleven studies).

Alendronate's seemingly greater beneficial effect on bone mineral density after arthroplasties with cement may have been confounded by patient age. The weighted mean age for patients undergoing cemented arthroplasty was greater than that for those undergoing arthroplasty without cement (68.0 compared with 55.0 years). It is also possible that the cement is responsible, in part, for our findings. In vitro studies have shown that cement particles co-cultured with macrophages result in an increase in pro-resorptive cytokine production compared with that in cultures of macrophages alone²⁷. In addition, cement particles can induce differentiation of preosteoclastic cells to osteoclasts²⁸. Etidronate, a bisphosphonate, has been reported to inhibit cement-induced bone resorption in vitro²⁹. These experimental findings have been supported by clinical series in which bone mineral density was found to be decreased following total hip arthroplasties with cement^30,31.

Another possible explanation for the difference between the effect of alendronate after arthroplasties with cement and the effect after cementless arthroplasties is that, with cementless arthroplasties, there may be improved stress transfer from the loaded prosthesis to the surrounding bone, resulting in increased bone formation³². It is plausible that this increase in bone density could mask any protective biochemical effect offered by a bisphosphonate. However, the amount of bone loss and/or bone formation around an uncemented stem may be related to the stem design^33-35.

It is more difficult to explain why alendronate treatment may provide more benefit after total knee arthroplasties than after total hip arthroplasties. It is possible that, with total knee arthroplasties, a greater proportion of the surface area of contact between the prosthesis and the skeleton occurs in trabecular bone, and trabecular bone is more susceptible to the effects of osteoclastic activity than is cortical bone³⁶.

Our subgroup analyses should be interpreted with caution. These analyses are inherently underpowered, and differences may be the result of chance alone³⁷. Investigators have suggested that tests of interaction between different subgroups can limit erroneous conclusions; however, tests of interaction are often underpowered³⁷. Thus, our findings that bisphosphonates exert a differential effect on total knee arthroplasties and arthroplasties with cement should be considered exploratory.

While the randomized trials in this meta-analysis provide current estimates of the effect of bisphosphonates on patients treated with joint arthroplasty, the beneficial effect on bone mineral density is not easily extrapolated to other clinically meaningful outcomes. Whether improvements in bone mineral density improve quality of life and function and decrease the risk of revision surgery remains unknown. The current meta-analysis does, however, provide important hypotheses that suggest a biological rationale for the role of bisphosphonates in this patient population.

In conclusion, our meta-analysis suggests that bisphosphonates have a beneficial effect with regard to preservation of periprosthetic bone for up to one year after total joint arthroplasty. However, the limitations of the studies and the fact that they did not address clinically relevant outcomes (functional outcomes, revision rates, and quality of life) indicate the need for a sufficiently sized, methodologically sound study to assess clinically relevant end points. Until such a study is performed, the current evidence regarding the use of bisphosphonates after total joint arthroplasty should be considered preliminary.

Appendix

Tables presenting the characteristics of the six randomized trials and the results of the sensitivity analysis are available with the electronic versions of this article, on our web site at (go to the article citation and click on "Supplementary Material") and on our quarterly CD-ROM (call our subscription department, at 781-449-9780, to order the CD-ROM). ?

Acknowledgements

Note: The authors thank Dr. J. Mark Wilkinson, Dr. Petri Venesmaa, Dr. T. Soininvaara, Debra Freedholm, and Dr. Albert Leung (Merck) for providing additional information about their studies and in some instances supplying the raw data of their studies. They thank Dr. T. Hennigs for providing additional data in a face-to-face meeting to discuss his study.

References

Malchau H, Herberts P, Ahnfelt L. Prognosis of total hip replacement in Sweden. Follow-up of 92,675 operations performed 1978-1990. Acta Orthop Scand.1993;64: 497-506.64497 1993 [PubMed][CrossRef]

Haddad FS, Masri BA, Garbuz DS, Duncan CP. The prevention of periprosthetic fractures in total hip and knee arthroplasty. Orthop Clin North Am.1999;30: 191-207.30191 1999 [PubMed][CrossRef]

van Loon CJ, de Waal Malefijt MC, Buma P, Verdonschot N, Veth RP. Femoral bone loss in total knee arthroplasty. A review. Acta Orthop Belg.1999;65: 154-63.65154 1999 [PubMed]

Willert HG, Buchhorn GH. The biology of the loosening of hip implants. In: Jakob RP, Fulford P, Horan F, editors. European instructional course lectures. Volume 4. London: The British Editorial Society of Bone and Joint Surgery; 1999. p 58-82.458 1999

Russell RG, Rogers MJ. Bisphosphonates: from the laboratory to the clinic and back again. Bone.1999;25: 97-106.2597 1999 [PubMed][CrossRef]

Cranney A, Wells G, Willan A, Griffith L, Zytaruk N, Robinson V, Black D, Adachi J, Shea B, Tugwell P, Guyatt G; Osteoporosis Methodology Group and The Osteoporosis Research Advisory Group. Meta-analyses of therapies for postmenopausal osteoporosis. II. Meta-analysis of alendronate for the treatment of postmenopausal women. Endocr Rev.2002;23: 508-16.23508 2002 [CrossRef]

Ross JR, Saunders Y, Edmonds PM, Patel S, Broadley KE, Johnston SR. Systematic review of role of bisphosphonates on skeletal morbidity in metastatic cancer. BMJ.2003;327: 469.327469 2003 Erratum in: BMJ.2004;328: 384.328384 2004 [PubMed][CrossRef]

Soininvaara TA, Jurvelin JS, Miettinen HJ, Suomalainen OT, Alhava EM, Kroger PJ. Effect of alendronate on periprosthetic bone loss after total knee arthroplasty: a one-year, randomized, controlled trial of 19 patients. Calcif Tissue Int.2002;71: 472-7.71472 2002 [PubMed][CrossRef]

Hennigs T, Arabmotlagh M, Schwarz A, Zichner L. [Dose-dependent prevention of early periprosthetic bone loss by alendronate]. Z Orthop Ihre Grenzgeb.2002;140: 42-7. German.14042 2002 [PubMed][CrossRef]

Venesmaa PK, Kroger HP, Miettinen HJ, Jurvelin JS, Suomalainen OT, Alhav EM. Alendronate reduces periprosthetic bone loss after uncemented primary total hip arthroplasty: a prospective randomized study. J Bone Miner Res.2001;16: 2126-31.162126 2001 [PubMed][CrossRef]

Lyons A. Effects of alendronate in total hip arthroplasty. Proceedings of the South African Orthopaedic Association. J Bone Joint Surg Br.1999;81 Suppl 3: 313.81313 1999

Wilkinson JM, Stockley I, Peel NF, Hamer AJ, Elson RA, Barrington NA, Eastell R. Effect of pamidronate in preventing local bone loss after total hip arthroplasty: a randomized, double-blind, controlled trial. J Bone Miner Res.2001;16: 556-64.16556 2001 [PubMed][CrossRef]

Wang CJ, Wang JW, Weng LH, Hsu CC, Huang CC, Chen HS. The effect of alendronate on bone mineral density in the distal part of the femur and proximal part of the tibia after total knee arthroplasty. J Bone Joint Surg Am.2003;85: 2121-6.852121 2003 [PubMed]

Venesmaa PK, Kroger HP, Jurvelin JS, Miettinen HJ, Suomalainen OT, Alhava EM. Periprosthetic bone loss after cemented total hip arthroplasty: a prospective 5-year dual energy radiographic absorptiometry study of 15 patients. Acta Orthop Scand.2003;74: 31-6.7431 2003 [PubMed][CrossRef]

Hilding M, Ryd L, Toksvig-Larsen S, Aspenberg P. Clodronate prevents prosthetic migration: a randomized radiostereometric study of 50 total knee patients. Acta Orthop Scand.2000;71: 553-7.71553 2000 [CrossRef]

Detsky AS, Naylor CD, O'Rourke K, McGeer AJ, L'Abbe KA. Incorporating variations in the quality of individual randomized trials into meta-analysis. J Clin Epidemiol.1992;45: 255-65.45255 1992 [PubMed][CrossRef]

Hedges LV, Olkin I. Statistical methods for meta-analysis. Orlando: Academic Press; 1985. p 108-38.108 1985

Egger M, Davey Smith G, Schneider M, Minder C. Bias in meta-analysis detected by a simple, graphical test. BMJ.1997;315: 629-34.315629 1997 [PubMed]

Bhandari M, Morrow F, Kulkarni AV, Tornetta P 3rd. Meta-analyses in orthopaedic surgery. A systematic review of their methodologies. J Bone Joint Surg Am.2001;83: 15-24.8315 2001 [PubMed]

Bailar JC 3rd. The promise and problems of meta-analysis. N Engl J Med.1997;337: 559-61.337559 1997 [PubMed][CrossRef]

Naylor CD. Meta-analysis and the meta-epidemiology of clinical research. BMJ.1997;315: 617-9.315617 1997 [PubMed]

LeLorier J, Gregoire G, Benhaddad A, Lapierre J, Derderian F. Discrepancies between meta-analyses and subsequent large randomized, controlled trials. N Engl J Med.1997;337: 536-42.337536 1997 [PubMed][CrossRef]

Ioannidis JP, Cappelleri JC, Lau J. Issues in comparisons between metaanalyses and large trials. JAMA.1998;279: 1089-93.2791089 1998 [PubMed][CrossRef]

Kjaergard LL, Als-Nielsen B. Association between competing interests and authors' conclusions: epidemiological study of randomised clinical trials published in the BMJ. BMJ.2002;325: 249.325249 2002 [PubMed][CrossRef]

Clifford TJ, Barrowman NJ, Moher D. Funding source, trial outcome and reporting quality: are they related? Results of a pilot study. BMC Health Serv Res.2002;2: 18.218 2002 [PubMed][CrossRef]

Bekelman JE, Li Y, Gross CP. Scope and impact of financial conflicts of interest in biomedical research: a systematic review. JAMA.2003;289: 454-65.289454 2003 [PubMed][CrossRef]

Ingham E, Green TR, Stone MH, Kowalski R, Watkins N, Fisher J. Production of TNF-alpha and bone resorbing activity by macrophages in response to different types of bone cement particles. Biomaterials.2000;21: 1005-13.211005 2000 [PubMed][CrossRef]

Sabokbar A, Fujikawa Y, Brett J, Murray DW, Athanasou NA. Increased osteoclastic differentiation by PMMA particle-associated macrophages. Inhibitory effect by interleukin 4 and leukemia inhibitory factor. Acta Orthop Scand.1996;67: 593-8.67593 1996 [PubMed][CrossRef]

Sabokbar A, Fujikawa Y, Murray DW, Athanasou NA. Bisphosphonates in bone cement inhibit PMMA particle induced bone resorption. Ann Rheum Dis.1998;57: 614-8.57614 1998 [PubMed][CrossRef]

Wilkinson JM, Hamer AJ, Rogers A, Stockley I, Eastell R. Bone mineral density and biochemical markers of bone turnover in aseptic loosening after total hip arthroplasty. J Orthop Res.2003;21: 691-6.21691 2003 [CrossRef]

Jasty M, O'Connor DO, Henshaw RM, Harrigan TP, Harris WH. Fit of the uncemented femoral component and the use of cement influence the strain transfer the femoral cortex. J Orthop Res.1994;12: 648-56.12648 1994 [CrossRef]

Brodner W, Bitzan P, Lomoschitz F, Krepler P, Jankovsky R, Lehr S, Kainberger F, Gottsauner-Wolf F. Changes in bone mineral density in the proximal femur after cementless total hip arthroplasty. A five-year longitudinal study. J Bone Joint Surg Br.2004;86: 20-6.8620 2004

Ohta H, Kobayashi S, Saito N, Nawata M, Horiuchi H, Takaoka K. Sequential changes in periprosthetic bone mineral density following total hip arthroplasty: a 3-year follow-up. J Bone Miner Metab.2003;21: 229-33.21229 2003

Cohen B, Rushton N. Bone remodelling in the proximal femur after Charnley total hip arthroplasty. J Bone Joint Surg Br.1995;77: 815-9.77815 1995

Einhorn TA. Osteoporosis and metabolic bone disease. Adv Orthop Surg.1984;8: 175-84.8175 1984

Assmann SF, Pocock SJ, Enos LE, Kasten LE. Subgroup analysis and other (mis)uses of baseline data in clinical trials. Lancet.2000;355: 1064-9.3551064 2000 [PubMed][CrossRef]

Gruen TA, McNeice GM, Amstutz HC. "Modes of failure" of cemented stemtype femoral components: a radiographic analysis of loosening. Clin Orthop.1979;141: 17-27.14117 1979 [PubMed]

Topics

alendronate ; bone mineral density ; bisphosphonates ; arthroplasty ; hip replacement arthroplasty ; knee replacement arthroplasty

Username

Password

Access for Patients

Forgot Password?

Have a subscription to the print edition?

Not a Subscriber?

Buy this Article

Get online access for 30 days for $35

New to JBJS?

Subscribe to the online edition for 30 days for $75

Register for a FREE limited account to get full access to all CME activities, to comment on public articles, or to sign up for alerts.

Have a subscription to the print edition?

Current subscribers to The Journal of Bone & Joint Surgery in either the print or quarterly DVD formats receive free online access to JBJS.org.

Activate your online account now

Forgot your password?

Enter your username and email address. We'll send you a reminder to the email address on record.

Username:*

Email Address:*

Forgot your username or need assistance? Please contact customer service at subs@jbjs.org. If your access is provided
by your institution, please contact you librarian or administrator for username and password information. Institutional
administrators, to reset your institution's master username or password, please contact subs@jbjs.org

References

Accreditation Statement

These activities have been planned and implemented in accordance with the Essential Areas and policies of the Accreditation Council for Continuing Medical Education (ACCME) through the joint sponsorship of the American Academy of Orthopaedic Surgeons and The Journal of Bone and Joint Surgery, Inc. The American Academy of Orthopaedic Surgeons is accredited by the ACCME to provide continuing medical education for physicians.

CME Activities Associated with This Article

Submit a Comment

Please read the other comments before you post yours. Contributors must reveal any conflict of interest.
Comments are moderated and will appear on the site at the discretion of JBJS editorial staff.

* = Required Field

Comment Author(s) * (if multiple authors, separate names by comma)

Example: John Doe

Affiliation & Institution*

Comment Title*

Comment*

Cancel

Mohit Bhandari, M.D.

Posted on May 26, 2005

Dr. Bhandari and colleagues respond to Dr Shetty et al

McMaster University

ID: 87/2/293

We read with great interest the comments raised by Shetty and colleagues on our paper â€œâ€œEffect of bisphosphonates on periprosthetic bone mineral density after total joint arthroplasty. A meta-analysisâ€ by Bhandari et al (2005; 87(2): 293-301).

We have indicated in the discussion section of the paper that â€œOur subgroup analyses should be interpreted with caution. These analyses are inherently underpowered, and differences may be the result of chance alone. Investigators have suggested that tests of interaction between different subgroups can limit erroneous conclusions; however, tests of interaction are often underpowered. Thus, our findings that bisphosphonates exert a differential effect on total knee arthroplasties and arthroplasties with cement should be considered exploratory.â€ We agree, with Shetty and colleagues that the effect of cementation could be confounded by the site of arthroplasty. Meta-regression analysis could potentially answer this question. However, the limited number of studies make such an analysis difficult.

Regarding the comment on the effect of pamidronate, we reported that in the context of discussing the potential influence of industry funding on study results. We stated â€œIn the current review, one trial was funded by a commercial entity(1) and, in another trial, the drugs were given free by a commercial entity(2). The industry- funded trial demonstrated a significant benefit of the commercial entityâ€™s product, but Wilkinson, et al, reported no significant effects of pamidronate at six months(2).â€ We indicated that pamidronate had no significant effects, eliminating the risk of bias of industry funding, without elaborating on the explanation of absence of effects, whether fall in bone mass in the pamidronate group or recovery in bone mass in the placebo group. We thank the authors for highlighting this point.

Finally, we thank the authors for emphasizing our suggestion of the importance of further work in this field. As we indicated in the paper, â€œWhether improvements in bone mineral density improve quality of life and function and decrease the risk of revision surgery remains unknown. The current meta-analysis does, however, provide important hypotheses that suggest a biological rationale for the role of bisphosphonates in this patient populationâ€. We look forward to hearing about a sufficiently sized, methodologically sound study to assess clinically relevant end points in this topic.

Sincerely yours,

Mohit Bhandari, M.D., Sohail Bajammal, M.D., Thomas Einhorn, M.D.

References:

1. Lyons A. Effects of alendronate in total hip arthroplasty (Abstractt) Journal of Bone and Joint Surgery-British Volume. Vol.81 Suppl 3, pp.313, 1999.

2. Wilkinson JM, Stockley I, Peel NF, Hamer AJ, Elson RA, Barrington NA, Eastell R. Effect of pamidronate in preventing local bone loss after total hip arthroplasty: a randomized, double-blind, controlled trial. J.Bone Miner Res. 2001;16:556-64.

Nitin. R Shetty

Posted on May 05, 2005

Effect of bisphosphonates on periprosthetic bone mineral density after total joint arthroplasty.

Department of Orthopaedics, Northern General Hospital, Sheffied, U.K.

To The Editor:

We read with interest the paper, â€œEffect of bisphosphonates on periprosthetic bone mineral density after total joint arthroplasty. A meta-analysisâ€ by Bhandari et al (1). We were intrigued by the finding that the effect size of bisphosphonate therapy was greater for cemented versus uncemented implants, and in total knee arthroplasty (TKA) versus total hip arthroplasty (THA). We note that both of the TKA studies were made using cemented implants, but that in half of the THA studies uncemented implants were used. Could the authors clarify whether cementation and joint site were independent predictors of drug efficacy, or might the greater effect in cemented implants be due to the common variable effect of TKA.

The authors also state that in our study the effect of a single post- operative dose of pamidronate on femoral bone mineral density was lost at 6 months.(2) We feel that it is important to comment that this late loss of effect was due largely to a partial recovery in bone mass in the placebo group at this time-point, rather than a late fall in bone mass in the pamidronate group.

Finally, we agree with the authorsâ€™ comments that studies to date have used small samples and not included clinically relevant outcome measures and that the clinical relevance of the results are, as yet, unclear. As bone mineral density is a major determinant of bone strength it seems plausible that pharmacological preservation of bone mass might influence the risk of periprosthetic fracture. Direct evidence for a relationship between regional bone loss, focal osteolysis, and aseptic loosening is lacking. We believe that further work is required to determine whether regional bone loss is an independent predictor of implant failure. If such a relationship were confirmed, this would justify the establishment of larger scale trials using clinically relevant outcomes such as development of osteolytic lesions or rates of implant survival.

References:

(1) Mohit Bhandari, Sohail Bajammal, Gordon H. Guyatt, Lauren Griffith, Jason W. Busse, Holger SchÃ¼nemann, and Thomas A. Einhorn Effect of Bisphosphonates on Periprosthetic Bone Mineral Density After Total Joint Arthroplasty. A Meta-Analysis J Bone Joint Surg Am 2005; 87: 293-301

(2) Wilkinson JM, Stockley I, Peel NFA, Hamer AJ, Elson RA, Barrington NA et al. Effect of pamidronate in preventing local bone loss after total hip arthroplasty: A randomized, double-blind, controlled trial. J Bone Miner Res 2001;16:556 -64.

Print
PDF
Email

Recipient(s) will receive an email with a link (good for 5 days) to 'Effect of Bisphosphonates on Periprosthetic Bone Mineral Density After Total Joint Arthroplasty: A Meta-Analysis'.
Your Name:*
Example: John Doe

Email Address:* CC Me:
Enter your valid email address. Example: jdoe@example.com

Recipient's Email Address:*

Separate multiple email address with semi-colons (up to 5).

Subject:* 's The Journal of Bone & Joint Surgery: 'Effect of Bisphosphonates on Periprosthetic Bone Mineral Density After Total Joint Arthroplasty: A Meta-Analysis'
Subject for your email.

Message:

(Optional, message will truncate at 1000 characters)

Processing your request... Please Wait...

Copyright © in the material you requested is held by The Journal of Bone & Joint Surgery, Inc. (unless otherwise noted). This email ability is provided as a courtesy, and by using it you agree that that you are requesting the material solely for personal, non-commercial use, and that it is subject to Journal of Bone & Joint Surgery, Inc.'s Terms of Use. The information provided in order to email this topic will not be used to send unsolicited email, nor will it be furnished to third parties. Please refer to The Journal of Bone & Joint Surgery Inc.'s Privacy Policy for further information.

Copyright © The Journal of Bone & Joint Surgery Inc. All rights reserved.
Share
Get Citation

Mohit Bhandari, Sohail Bajammal, Gordon H. Guyatt, Lauren Griffith, Jason W. Busse, Holger Schünemann, Thomas A. Einhorn; Effect of Bisphosphonates on Periprosthetic Bone Mineral Density After Total Joint ArthroplastyA Meta-Analysis. The Journal of Bone & Joint Surgery. 2005 Feb;87(2):293-301.

Download citation file:

RIS (Zotero)

EndNote

BibTex

Medlars

ProCite

RefWorks

Reference Manager

Copyright © The Journal of Bone and Joint Surgery, Inc.
Rights & Permissions
Article Alerts

Processing your request... Please Wait.
Submit a Comment
Data Supplements