Bisphosphonate use is associated with a decreased joint narrowing rate in the non-arthritic hip

Abstract

Aims: The preventive effects of bisphosphonates on articular cartilage in non-arthritic joints are unclear. This study aimed to investigate the effects of oral bisphosphonates on the rate of joint space narrowing in the non-arthritic hip.

Methods: We retrospectively reviewed standing whole-leg radiographs from patients who underwent knee arthroplasties from 2012 to 2020 at our institute. Patients with previous hip surgery, Kellgren-Lawrence grade ≥ II hip osteoarthritis, hip dysplasia, or rheumatoid arthritis were excluded. The rate of hip joint space narrowing was measured in 398 patients (796 hips), and the effects of the use of bisphosphonates were examined using the multivariate regression model and the propensity score matching (1:2) model.

Results: A total of 45 of 398 (11.3%) eligible patients were taking an oral bisphosphonate at the time of knee surgery, with a mean age of 75.8 years (SD 6.2) in bisphosphonate users and 75.7 years (SD 6.8) in non-users. The mean joint space narrowing rate was 0.04 mm/year (SD 0.11) in bisphosphonate users and 0.12 mm/year (SD 0.25) in non-users (p < 0.001). In the multivariate model, age (standardized coefficient = 0.0867, p = 0.016) and the use of a bisphosphonate (standardized coefficient = -0.182, p < 0.001) were associated with the joint space narrowing rate. After successfully matching 43 bisphosphonate users and 86 non-users, the joint narrowing rate was smaller in bisphosphonate users (p < 0.001).

Conclusion: The use of bisphosphonates is associated with decreased joint degeneration in non-arthritic hips after knee arthroplasty. Bisphosphonates slow joint degeneration, thus maintaining the thickness of joint cartilage in the normal joint or during the early phase of osteoarthritis.Cite this article: Bone Joint Res 2022;11(11):826-834.

Keywords: Bisphosphonate; Hip osteoarthritis; Joint cartilage; Joint space narrowing; Knee arthroplasty; Osteoarthritis (OA); Osteoporosis; arthritis hips; bisphosphonates; hips; joint degeneration; joint space narrowing; knee arthroplasties; propensity score matching; radiographs; rheumatoid arthritis.

Fig. 1

Flow chart showing inclusion and exclusion of patients. BHA, bipolar hip arthroplasty; RA, rheumatoid arthritis; OA, osteoarthritis; THA, total hip arthroplasty.

Fig. 2

Measurements of the hip joint space width in a representative case. A 65-year-old woman underwent total knee arthroplasty for osteoarthritis (OA) (bisphosphonate non-user). a) Anteroposterior standing whole-leg radiograph taken before surgery. b) Anteroposterior standing whole-leg radiograph taken at four years after surgery. c) Magnified image of the area including the right hip in panel a). d) Magnified image of the area including the right hip in panel b). The joint space width and the femoral head diameter were measured as indicated by the yellow lines. JSW1 and JSW2 are the joint space widths before surgery and at the final follow-up, respectively. D1 and D2 are the femoral head diameters measured on radiographs at three months after surgery and at the final follow-up, respectively.

Fig. 3

Box and whisker plots of the control and bisphosphonate groups. a) Comparison of the joint space narrowing rate (JSNR) before matching between the groups. b) Comparison of the JSNR after matching between the groups. c) Comparison of the normalized JSNR (nJSNR) before matching between the groups. d) Comparison of the nJSNR after matching between the groups. The top and bottom of the boxes indicate the interquartile range, the line within the box indicates the median, and the whiskers represent points within 1.5 times the width of the interquartile range. All p-values were calculated using Mann-Whitney U test.

Fig. 4

Box and whisker plots showing comparison of a) the joint space narrowing rate (JSNR) and b) the normalized JSNR (nJSNR) between ipsilateral and contralateral hips to the side of knee arthroplasty. Analyses were performed in 255 patients who underwent knee arthroplasty on only one side during the study period. Both p-values were calculated using Mann-Whitney U test.