Abstract
Factors affecting bone strength have not been elucidated in patients with rheumatoid arthritis (RA) undergoing silicone implant arthroplasty. In this observational study, we aimed to investigate factors affecting bone strength. We measured toughness in terms of bone strength, bone mineral density, and bone quality of 105 resected metacarpal heads (I:32, II:36, III:33, IV:2, V:2) obtained during silicone implant arthroplasty at the metacarpophalangeal (MP) joint in patients with RA. Bone strength was associated with all parameters in micro-computed tomography (µCT) and all parameters in bone histomorphometry, other than osteoid volume and osteoid surface. Clinical factors associated with bone strength included prednisolone (PSL) dose at the time of surgery, % young adult mean, T-score and the Rooney score. Bone volume fraction by µCT only influenced bone strength per multiple linear regression analysis. Compared to joints with higher bone strength, those with lower bone strength had a higher magnitude of implant subsidence, which assumed to the risk of reduced mobility and implant fracture after the operation. Suppressing the degradation of subchondral bone structures by continuously ameliorating disease activity without the use of PSL appeared to be crucial to maintain the radiological appearance of silicone implant arthroplasty at the MP joint.
Supplementary Information
The online version contains supplementary material available at 10.1038/s41598-025-31762-z.
Keywords: Bone strength, Rheumatoid arthritis, Bone mineral density, Micro-computed tomography, Bone histomorphometry
Subject terms: Medical research, Rheumatology
Introduction
Rheumatoid arthritis (RA) is a chronic inflammatory autoimmune disease that primarily affects the joints1. Preventing joint destruction is crucial, as it directly impacts daily life due to functional impairment. Advances in therapy, including biological disease-modifying anti-rheumatic drugs (bDMARDs) like tumor necrosis factor-alfa (TNF-ɑ) inhibitors and interleukin-6 (IL-6) receptor antibodies, have shifted the therapeutic goal toward achieving both clinical and structural remission2,3. However, RA can lead to both joint destruction and secondary osteoporosis, which increases fracture risk4.
The pathophysiology of bone loss and secondary osteoporosis in RA involves both primary osteoporosis factors—due to the typically older, female RA population—and RA-specific factors. Elevated cytokines like TNF-α and IL-6 impact osteoclast and osteoblast differentiation, while glucocorticoid use in patients with RA induces a bone metabolic state marked by increased resorption and decreased formation5,6. Despite recent therapeutic advancements, patients with RA still face a higher risk of fragility fractures compared to the general population4,7,8, highlighting the need for strategies to reduce fracture risk and enhance quality of life.
Bone mineral density (BMD) can predict bone strength but cannot adequately estimate fracture risk because many fractures occur in the osteopenic range of BMD9. Bone strength is affected not only by BMD but also by bone quality10. To effectively predict fragility fracture risk in RA and optimize preventive strategies, it is crucial to identify factors reflecting bone strength through bone quality indicators (e.g., bone microstructure, morphometry, and clinical background).
Recent evidence shows that whole hand BMD is significantly lower in patients with RA compared with healthy controls and is significantly associated with lumbar spine BMD in patients with RA11. The average age of this RA cohort is 52.7 years old, suggesting that RA itself would be the risk factor of secondary osteoporosis.
We aimed to investigate factors affecting bone strength in patients with RA by directly measuring bone strength and analyzing its relationship with clinical characteristics and bone quality data (e.g., bone microarchitecture and histomorphometry). Additionally, we assessed whether measured bone strength influenced subsidence of the implant after silicone implant arthroplasty at the MP joint.
Results
Patient characteristics
Metacarpal heads were taken from 32 specimens from the thumb, 36 from the index finger, 33 from the middle finger, 2 from the ring finger, and 2 from the little finger (Supplementary Fig. 1a, b, c). Demographic and disease-specific characteristics are shown in Table 1. The mean age of patients was 63.6 years, 91.7% were female, and 51% were taking osteoporosis drugs.
Table 1.
Patient characteristics at the time of the metacarpophalangeal joint arthroplasty.
| Characteristic | Arthroplasty = 51 (patients with RA = 48; metacarpophalangeal joints = 105) |
|---|---|
| Age (years), mean (SD) | 63.6 (13.4) |
| Females, n (%) | 44 (91.7) |
| Disease duration (years), mean (SD) | 22.3 (8.7) |
| MTX use, n (%) | 30 (58.8) |
| MTX dose (mg/week), mean (SD) | 7.9 (2.3) |
| PSL use, n (%) | 27 (52.9) |
| PSL dose (mg/day), mean (SD) | 2.7 (1.3) |
| PSL dose in the past 5 years (mg/day), mean (SD) | 2.4 (1.3) |
| Usage of bDMARDs, n (%) | 27 (52.9) |
| Infliximab | 1 |
| Adalimumab | 1 |
| Golimumab | 5 |
| Certolizumab pegol | 2 |
| Tocilizumab | 9 |
| Abatacept | 2 |
| Tofacitinib | 4 |
| Usage of csDMARDs, n (%) | 45 (93.8) |
| Salazosulfapyridine | 14 |
| Iguratimod | 11 |
| Tacrolimus | 9 |
| Bucillamine | 7 |
| Mizollibine | 3 |
| Gold sodium thiocyanate | 1 |
| Usage of osteoporosis drugs, n (%) | 26 (51.0) |
| Bisphosphonates | 14 |
| Risedronate | 6, 6.0 ± 4.1 yrs, 1.8–11.4 yrs |
| Weekly | 3, 7.0 ± 4.9 yrs, 1.8–11.4 yrs |
| Monthly | 3, 4.9 ± 4.0 yrs, 1.3–9.1 yrs |
| Alendronate | 4, 5.7 ± 5.8 yrs, 0.1–11.8 yrs |
| Minodronate | 4, 2.0 ± 3.5 yrs, 0.2–7.3 yrs |
| RANK ligand inhibitor | 2 |
| parathyroid hormone analog | 2 |
| Others | 10 |
| DAS28-ESR, mean (SD) | 3.3 (1.1) |
| DAS28-ESR in the past 5 years, mean (SD) | 3.2 (0.8) |
| Rooney score, mean (SD) | 7.5 (8.2) |
| Serum TRACP-5b (mU/dL), mean (SD) | 326.7 (166.1) |
| Serum BAP (µg/L), mean (SD) | 11.6 (6.7) |
| YAM (%), mean (SD) | 76.5 (14.5) |
| T-score, mean (SD) | -1.9 (1.2) |
| Larsen grade, n (%) | |
| II | 10 (9.5) |
| III | 39 (37.1) |
| IV | 40 (38.1) |
| V | 16 (15.2) |
Continuous data are expressed as the mean (standard deviation), while categorical data are shown as numbers (percentages).
BAP, bone-specific alkaline phosphatase; bDMARDs, biological disease-modifying antirheumatic drugs; DAS28-ESR, disease activity score; 28, erythrocyte sedimentation rate; MTX, methotrexate; PSL prednisolone, TRACP-5b; tartrate-resistant acid phosphatase-5b; YAM, Young Adult Mean.
Related to bisphosphonates, risedronate was used in 6 cases and mean exposed duration was 6.0 ± 4.1 years (range; 1.8–11.4 years). Of these, 17.5 mg/week of risedronate was used in 3 cases, and 75 mg/month of risedronate was used in 3 cases. Next, alendronate was used in 4 cases and mean exposed duration was 5.7 ± 5.8 years (range 0.1–11.8 years). Then, minodronate was used in 4 cases and mean exposed duration was 2.0 ± 3.5 years (range, 0.2–7.3 years) (Table 1).
Most patients had low-to-moderate disease activity, and about half of patients were using steroids or biologic disease-modifying anti-rheumatic drugs (bDMARDs).
Cases under 8 mg/week of methotrexate (MTX) were totally 39 cases (0 mg/week in 18, 4 mg /week in 4, 6 mg/week in 4, 7 mg/week in 1, and 8 mg/week in 12 cases).
The average young adult mean (YAM) and T-score were 76.5% and − 1.9, and 75.2% of MP joints were Larsen grade III or IV (Table 1).
Bone strength of metacarpal heads
In this study, toughness parameters derived from compression tests were operationally defined as bone strength. Bone strength results were obtained for 98 of the 105 specimens, with seven specimens excluded due to collapse during testing. The mean ± standard deviation (SD) bone strength was 83.1 ± 73.6 N/mm (Supplementary Fig. 2).
Differences in bone strength between patients taking and not taking drugs
Bone strength did not differ significantly between patients treated with or without MTX or bDMARDs (Table 2). In contrast, bone strength was significantly lower in patients treated with PSL at the time of surgery than in those not treated with PSL (Table 2). In addition, bone strength was significantly lower in patients taking osteoporosis drugs at the time of surgery than in those who did not. There was no significant difference in bone strength between those who used bisphosphonates and those who did not (Table 2).
Table 2.
Comparison of bone strength between drug usage and non-usage groups.
| Drugs | Bone strength (N.mm), mean (SD) | P value | ||
|---|---|---|---|---|
| With drug | Without drug | |||
| PSL | 67.7 (52.4) | 100.5 (89.3) | 0.032* | |
| MTX | 89.4 (80.9) | 76.0 (64.4) | 0.370 | |
| bDMARDs | 87.1 (83.4) | 78.7 (61.7) | 0.571 | |
| Osteoporosis drugs | 61.8 (56.2) | 101.2 (81.8) | 0.006* | |
| Bisphosphonates | with BP | without BP | 0.396 | |
| 55.1 (53.9) | 69.5 (59.1) | |||
Continuous data are expressed as the mean (standard deviation). Data were analyzed using the Student’s t-test (MTX and BP) or Welch’s t-test (PSL, bDMARDs, and osteoporosis drugs), and P values were calculated.
* Statistically significant correlation (P < 0.05).
PSL, prednisolone; MTX, methotrexate; bDMARDs, biological disease-modifying antirheumatic drugs; BP, bisphosphonates.
Correlation between bone strength and clinical characteristics by simple regression analysis
The YAM and T-scores of BMD were positively correlated with bone strength (Table 3, Supplementary Figs. 3k, l). The daily PSL dose at the time of surgery showed a significant negative correlation with bone strength, whereas the Rooney score showed a positive correlation (Table 3, Supplementary Figs. 3d, g). Bone strength was compared after categorizing patients into those receiving less than 2.5 mg/d (low), 2.5 to less than 5 mg/d (mid), and 5 mg/d or more (high) of prednisolone. However, no significant differences were detected between the groups. In contrast, other clinical parameters, such as disease duration and disease activity of RA, showed no significant correlation with bone strength (Table 3, Supplementary Figs. 3a-c, e, f, h-j).
Table 3.
Correlations between bone strength and clinical characteristics at the time of surgery as analyzed using simple regression analysis.
| Variable | Correlation coefficient (r) | P value |
|---|---|---|
| Disease duration | −0.1451 | 0.1541 |
| DAS28-ESR | −0.0217 | 0.8320 |
| Mean DAS28-ESR in the past 5 years | −0.0002 | 0.9981 |
| PSL dose | −0.2125 | 0.0356* |
| Mean PSL dose in the past 5 years | −0.1575 | 0.1214 |
| MTX dose | 0.0833 | 0.4147 |
| Rooney score | 0.2259 | 0.0253* |
| Larsen grade | −0.0811 | 0.4273 |
| Serum TRACP-5b | 0.0251 | 0.8254 |
| Serum BAP | −0.0478 | 0.6738 |
| YAM | 0.5952 | < 0.0001* |
| T-score | 0.5946 | < 0.0001* |
Data are analyzed using Pearson’s correlation analysis, and correlation coefficients (r values) and P values are calculated.
* Statistically significant correlation (P < 0.05).
BAP, bone-specific alkaline phosphatase; DAS28-ESR, disease activity score; 28, erythrocyte sedimentation rate; MTX, methotrexate; PSL prednisolone, TRACP-5b; tartrate-resistant acid phosphatase-5b; YAM, Young Adult Mean.
Correlation between bone strength and µCT variables by simple regression analysis
Each µCT parameter was significantly positively or negatively correlated with bone strength (Table 4). The most significant positive correlation (r = 0.8237) was detected between bone volume fraction (bone volume/total volume) (BV/TV) and bone strength (Table 4; Supplementary Fig. 4). The µCT images of the specimens are shown in Fig. 1.
Table 4.
µCT data and correlations between bone strength and µCT parameters as analyzed using simple regression analysis.
| Variable | mean (SD) | 95% CI | Correlation coefficient with bone strength (r) | P value |
|---|---|---|---|---|
| vBMD (mg/cm3) | 78.18 (50.30) | 68.09, 88.26 | 0.7557 | < 0.0001* |
| BS/BV (1/mm2) | 14.06 (3.87) | 13.29, 14.84 | −0.5921 | < 0.0001* |
| BV/TV (%) | 24.95 (11.59) | 22.62, 27.27 | 0.8237 | < 0.0001* |
| Tb.Th (µm) | 215.08 (48.90) | 205.28, 224.89 | 0.4915 | < 0.0001* |
| Tb.N (1/mm) | 0.68 (0.20) | 0.64, 0.72 | 0.6706 | < 0.0001* |
| Tb.Sp (µm) | 466.75 (128.19) | 441.05, 492.45 | −0.5616 | < 0.0001* |
| Conn.D (1/mm3) | 3.12 (1.60) | 2.80, 3.44 | 0.5821 | < 0.0001* |
| TBPf (1/mm) | 2.61 (2.78) | 2.05, 3.16 | −0.6759 | < 0.0001* |
| SMI | 1.59 (0.61) | 1.47, 1.71 | −0.7553 | < 0.0001* |
| V* m.space (mm3) | 1.73 (1.43) | 1.44, 2.02 | −0.5432 | < 0.0001* |
| DA | 1.62 (0.21) | 1.57, 1.66 | 0.2153 | 0.0332* |
| N.Nd/TV (1/mm3) | 5.32 (2.13) | 4.89, 5.75 | 0.4843 | < 0.0001* |
| TSL/TV (1/mm2) | 5.36 (1.85) | 4.99, 5.73 | 0.6285 | < 0.0001* |
Continuous data are expressed as the mean (standard deviation) and 95% confidence intervals (95% CI). Data are analyzed using Pearson’s correlation analysis, and correlation coefficients (r values) and P values are calculated.
* Statistically significant correlation (P < 0.05).
BS/BV, bone surface density (bone surface/bone volume); BV/TV, bone volume fraction (bone volume/tissue volume); Conn.D., connectivity density; DA, degree of anisotropy; N. Nd/TV, number of nodes/tissue volume; SMI, structure model index; Tb.N., trabecular number; TBPf, trabecular bone pattern factor; Tb.Sp.; Trabecular separation; Tb.Th., trabecular thickness; TSL/TV, total strut length/tissue volume; vBMD, volumetric bone mineral density; V* m. space, marrow space star volume.
Fig. 1.
µCT images in specimens.
Full and half-cut images by µCT in the specimen of (a) the strongest bone strength, (b) the weakest bone strength. The bone strength values for the strongest specimen and the weakest specimen were 373.02 N.mm and 3.4 N.mm, respectively.
Correlation between bone strength and bone histomorphometry variables by simple regression analysis
For bone structural parameters bone volume/ tissue volume (BV/TV), trabecular thickness (Tb.Th), and trabecular number (Tb.N) were positively correlated with bone strength, whereas trabecular space (Tb.Sp) was negatively correlated (Table 5, Supplementary Fig. 5a-d). Regarding the bone formation (osteoid) parameters, neither osteoid volume/ bone volume (OV/BV) nor osteoid surface/ bone surface (OS/BS) was significantly correlated with bone strength (Table 5, Supplementary Fig. 5e, f). However, osteoid thickness (O.Th) and osteoblast surface/ bone surface (Ob.S/BS) were positively correlated with bone strength (Table 5, Supplementary Fig. 5 g, h). For bone resorption parameters, both eroded surface/ bone surface (ES/BS) and osteoclast surface/ bone surface (Oc.S/BS) were positively correlated with bone strength (Table 5, Supplementary Fig. 5i, j).
Table 5.
Bone histomorphometry data and correlations between bone strength and bone histomorphometry parameters as analyzed using simple regression analysis.
| Variable | mean (SD) | 95% CI | Correlation coefficient with bone strength (r) | P value |
|---|---|---|---|---|
| BV/TV (%) | 17.71 (8.00) | 15.76, 19.67 | 0.6985 | < 0.0001* |
| Tb.Th (µm) | 135.04 (39.09) | 125.51, 144.57 | 0.6923 | < 0.0001* |
| Tb.N (/mm) | 1.28 (0.34) | 1.20, 1.36 | 0.4181 | 0.0004* |
| Tb.Sp (µm) | 700.84 (249.96) | 639.87, 761.81 | −0.4834 | < 0.0001* |
| OV/BV (%) | 1.42 (1.46) | 1.06, 1.77 | 0.0106 | 0.9322 |
| OS/BS (%) | 9.22 (8.64) | 7.11, 11.33 | 0.1196 | 0.3352 |
| O.Th (µm) | 9.27 (2.78) | 8.59, 9.94 | 0.4252 | 0.0003* |
| Ob.S/BS (%) | 0.38 (0.96) | 0.15, 0.62 | 0.4721 | < 0.0001* |
| ES/BS (%) | 0.73 (0.88) | 0.51, 0.94 | 0.4816 | < 0.0001* |
| Oc.S/BS (%) | 0.14 (0.28) | 0.07, 0.21 | 0.4179 | 0.0004* |
Continuous data are expressed as the mean (standard deviation) and 95% confidence intervals (95% CI). Data are analyzed using Pearson’s correlation analysis, and correlation coefficients (r values) and P values are calculated.
* Statistically significant correlation (P < 0.05).
BV/TV, bone volume/tissue volume; ES/BS, eroded surface/bone surface; Ob.S/BS, osteoid surface/bone surface; Oc.S/BS, osteoclast surface/bone surface; O.Th, osteoid thickness; OS/BS, osteoid surface/bone surface; OV/BV, osteoid volume/bone volume; Tb.N, trabecular number; Tb.Sp, trabecular separation; Tb.Th, trabecular thickness.
Extraction of variables affecting bone strength using multiple linear regression analysis
To solve the problem of multicollinearity, several variables were omitted, and multiple linear regression analysis was performed. The multiple linear regression model could explain 83% of the bone strength (R² = 0.83, adjusted R² = 0.72). BV/TV on µCT (β = 4.43, SE = 0.67, t = 6.63, p < 0.0001) was the only significant positive predictor of bone strength (Table 6).
Table 6.
Variables affecting bone strength as analyzed using multiple regression analysis.
| Variables | β | SE | T | VIF | P value |
|---|---|---|---|---|---|
| Disease duration | −0.45 | 1.11 | −0.40 | 2.64 | 0.6911 |
| DAS28-ESR | 3.75 | 12.03 | 0.31 | 3.93 | 0.7578 |
| Mean DAS28-ESR in the past 5 years | 6.83 | 16.03 | 0.43 | 4.35 | 0.6733 |
| PSL dose | -3.92 | 4.63 | −0.85 | 2.59 | 0.4050 |
| MTX dose | −1.14 | 1.98 | −0.57 | 2.61 | 0.5711 |
| Rooney score | 0.59 | 1.05 | 0.56 | 4.52 | 0.5805 |
| Larsen grade | 0.84 | 9.40 | 0.09 | 2.48 | 0.9298 |
| Serum TRACP-5b | −0.01 | 0.06 | −0.18 | 3.65 | 0.8615 |
| Serum BAP | 3.61 | 2.76 | 1.31 | 2.59 | 0.2016 |
| YAM | −0.34 | 0.67 | −0.50 | 2.45 | 0.6182 |
| Usage of bDMARDs | −5.89 | 16.50 | −0.36 | 1.90 | 0.7238 |
| Usage of osteoporosis drugs | −14.48 | 13.68 | −1.06 | 1.92 | 0.2992 |
|
BV/TV (µCT) |
4.43 | 0.67 | 6.63 | 2.20 | < 0.0001* |
|
DA (µCT) |
32.81 | 31.06 | 1.06 | 1.65 | 0.3001 |
|
OS/BS (bone histomorphometry) |
0.22 | 0.81 | 0.27 | 2.22 | 0.7896 |
|
O.Th (bone histomorphometry) |
−1.23 | 3.34 | −0.37 | 3.83 | 0.7164 |
|
Ob.S/BS (bone histomorphometry) |
5.10 | 9.73 | 0.52 | 4.67 | 0.6046 |
|
ES/BS (bone histomorphometry) |
−0.26 | 9.62 | −0.03 | 3.57 | 0.9787 |
The variables include the clinical characteristics and µCT and bone histomorphometry parameters.
Data are analyzed using the least squares method and unstandardized regression coefficients (β), standard error (SE), t-values, variance inflation factor (VIF), and P values are calculated.
* Statistically significant correlation (P < 0.05).
Implant subsidence
In the ten strongest bone strength group (ST group), the mean ± SD bone strength at the time of surgery was 248.8 ± 68.5 N/mm, and the mean ± SD magnitude of implant subsidence over a mean period of 723.8 days was 1.3 ± 0.9 mm. The magnitude of the implant subsidence was mild (Fig. 2a). In the ten weakest bone strength group (WE group), the mean ± SD bone strength at the time of surgery was 9.4 ± 3.5 N/mm, and the mean ± SD magnitude of implant subsidence over a mean period of 1021.1 days was 2.9 ± 1.0 mm. In one case in the WE group, subsidence was severe, and the implant stem perforated the bone (Fig. 2b). The magnitude of implant subsidence was significantly lower in the ST group than in the WE group (Fig. 2c). Although the period during which implant subsidence was evaluated differed between the ST and WE groups, the magnitude of implant subsidence was significantly smaller in the ST group than in the WE group, even when it was recalculated as the magnitude per year (Fig. 2d). The magnitude of subsidence per year was 0.680 ± 0.395 mm/year for the ST group and 1.139 ± 0.381 mm/year for the WE group (P < 0.05). The results of the ANCOVA adjusted for the follow-up period are shown in Supplementary Table 1.
Fig. 2.
Implant subsidence after implant arthroplasty.
Radiographs before and after the implant arthroplasty at the MP joint of the thumb in (a) the ST group and (b) the WE group. (c) Comparison of the magnitude of implant subsidence during the entire observation period, (d) Implant subsidence per year between the ST and WE groups. The bottom, middle, and upper lines in the box plots correspond to the 25th, 50th, and 75th percentiles, respectively. Data were analyzed using Student’s t-test. * Statistically significant difference (P < 0.05).
Discussion
Periarticular osteoporosis progresses in RA because of persistent joint inflammation, and BMD in the whole body is significantly lower in patients with RA than in healthy controls12. This study revealed that BMD-related factors (YAM and T-score) were associated with bone strength of the metacarpal head. This suggests that the bone strength of inflamed sites can be predicted to some extent by measuring the BMD of the whole body.
Disease activity score 28 joints- erythrocyte sedimentation rate (DAS28-ESR) at the time of surgery was not associated with bone strength in the metacarpal head. And the Rooney score positively correlated to the bone strength. Even though the BV/TV of the metacarpal head was high, local inflammation was concentrated in the MP joint, and the specimen may have been taken at a time when pain was particularly severe. A previous cohort study reported that the incidence of non-vertebral fractures remained stable13, while the rate of patients achieving DAS28 remission increased over 10 years irrespective of local inflammation. These results suggest that systemic and local inflammation at one point in time or within the last five years did not necessarily reflect bone strength severity. Accumulation of systemic and local inflammation over five years might affect bone strength.
PSL is generally believed to reduce BMD in patients with RA14–16. In this study, PSL dose at the time of surgery showed a significant negative correlation with bone strength, but the mean PSL dose over five years prior to surgery was not significantly associated with bone strength. Because PSL itself is known to not strongly contribute to the loss of systemic BMD in situations of well-controlled disease activity during remission16,17, it is possible that the accumulation of PSL did not affect bone strength in this study. In addition, the use of MTX or bDMARDs at the time of surgery also did not influence metacarpal bone strength. However, in a previous study using a mouse arthritis model, prophylactic treatment with an anti-IL-6 receptor antibody prevented deterioration of bone structure and strength in these arthritic mice18. Therefore, tight control of RA with bDMARDs, including at least anti-IL-6 receptor antibodies from the early onset of RA, may help maintain bone structure and strength. In this study, patients treated with drugs for osteoporosis showed significantly lower bone strength than those not treated with drugs for osteoporosis. The mean BV/TV of the patients treated with these drugs (n = 49) was 41%, and that of those without these drugs (n = 56) was 63.5%. The decreased bone strength in these patients is primarily attributable to the low BV/TV, and drugs for osteoporosis was used. Alternatively, the long-term use of bisphosphonates might have reduced bone strength even if bone mass was maintained19, but this could not be verified in this study.
Bone quality can be defined as the accumulation of bone aspects that contribute to bone strength but are not measured by bone density20 and can include parameters such as architecture, bone remodeling, damage accumulation, and mineralization10. In this study, we assessed bone architecture using µCT and bone remodeling through histomorphometry. A previous study found that bone formation and resorption parameters were significantly higher in patients with RA. In the present study, bone strength was positively correlated not only with the osteoblast surface but also with the osteoclast surface in the bone histomorphometry results. In particular, when the metacarpal bone showed erosive changes (Supplementary Fig. 6a), many osteoclasts adhered to the trabecular bone (Supplementary Fig. 6b, c, and d). This event indicates a hyper-state (promoting) of bone turnover due to local inflammation. Bone remodeling progressed following persistent synovitis, and secondary bone formation may have occurred in the subchondral bone Supplementary Fig. 6e and f). In our previous report, destructive hip due to RA demonstrated significantly increased parameters in both bone formation and resorption compared with non-destructive hip due to RA, suggesting that bone remodeling was hyper state in response to synovitis due to RA disease itself21,
The progression of joint lesions in RA induces or promotes osteoarthritis (OA)22,23. A previous study reported that compared to whole bone BV/TV, local BV/TV more accurately reflects bone strength24 and BV/TV explained 87% of the variation in bone elastic properties25. Although our study was the first investigation using the metacarpal head, it was also consistent with previous reports suggesting that local BV/TV accounts for most of the bone strength.
Implant subsidence is an unfavorable phenomenon after silicone implant arthroplasty. In the present study, the magnitude of subsidence was more severe in the WE group than in the ST group, resulting in bone perforation in one patient in the WE group. This suggests that the strength of the subchondral bone is significantly related to loss of mobility and possible implant fracture in clinical practice.
This study has several limitations. Ideally, a prospective study involving healthy individuals or a non-RA trauma control group would have been designed to investigate subchondral bone fragility in patients with rheumatoid arthritis. However, this was not feasible due to ethical concerns surrounding the use of human bone specimens. Additionally, factors such as a history of drug use, complications, and lifestyle may influence metacarpal head bone strength, but were not fully accounted for.
In conclusion, inhibiting structural degradation at subchondral bone without the use of PSL appeared to be crucial to maintain the effect of silicone implant arthroplasty for patients with RA.
Methods
Patients and specimens
A total of 105 metacarpal heads were obtained from 48 consecutive Japanese patients with RA undergoing silicone implant arthroplasty of MP joint using Swanson Finger Joint Implant (Stryker Co.,Ltd., MI, USA) at our center of rheumatology between December 2012 and June 202126,27. The operation was performed when destructive lesions with ulno-palmar subluxation appeared at the MP joint, and aesthetic and /or functional impairment of the hand persisted for six months or more. The operation was not indicated if there was only swelling and/or redness, but it was considered if there was persistent pain and /or dexterous movement of the hand. All patients met the 1987 revised criteria of the American College of Rheumatology (ACR) for the classification of RA or the 2010 ACR/European League Against Rheumatism classification criteria28,29. When multiple MP joints were operated in the same patient, the most inflamed MP joint was taken as the specimen. Bone specimens were immersed in 10% formalin-neutral buffer fixative immediately after extraction and preserved until the experiments. The cases of juvenile idiopathic arthritis or those diagnosed as malignancy were excluded.
Recruitment sample collection for this study began on December 25, 2012 and ended on June 8, 2021. Participants provided written informed consent and minor were not included. The data were started to access for research purposes after June 8, 2020. The authors had access to information that could identify individual participants after data collection.
This study was approved by the Ethics Committees of the Niigata Rheumatic Center (Approval No.: 2020-003) and the Ethical Review Committee of Chugai Pharmaceutical Co. Ltd. (Approval No.: E20037) and was carried out in accordance with the Declaration of Helsinki. Informed consent was obtained from all the patients.
Clinical assessment
The following background and clinical characteristics were obtained from medical records: DAS28-ESR, MTX dose, PSL dose, Rooney score30, serum tartrate-resistant acid phosphatase 5b (TRACP-5b) level, serum bone specific alkaline phosphatase (BAP) level, % YAM and T-score in BMD, Larsen grade at the operated MP joint31, use of bDMARDs, and use of drugs for osteoporosis. For DAS28-ESR and PSL doses, data at the time of surgery and the mean values five years before surgery were obtained.
To solely assess the activity of local inflammation, Rooney score was used. It is a histopathological examination of the synovium to see if there are characteristic findings of RA. It consists of six items (synoviocyte hyperplasia, fibrosis, proliferating blood vessels, perivascular infiltrates of lymphocytes, focal aggregates of lymphocytes, and diffuse infiltrates of lymphocytes), each of which is given a score of 0–10, and the maximum total score is thus 60 points. In this study, fibrosis and blood vessel count were excluded, because they are also seen in the chronic phase of inflammation.
Biomechanical test
To measure the bone material strength, bone blocks of the metacarpal head were prepared using IsoMet-1000 (BUEHLER, Lake Bluff, IL, USA). The bone blocks were cut to dimensions of 5 × 7 × 7 mm (width × depth × height), with the distal end of the metacarpal head containing articular cartilage as the top. The prepared bone blocks were subjected to compression testing against an axis perpendicular to the bone surface at a loading rate of 2 mm/min using a bone strength measuring device AGS-5kNX (SHIMADZU, Kyoto, Japan). From the obtained load-displacement curve, toughness parameters, defined as energy values up to a displacement of 2 mm, were used as an index of bone material strength in this study.
Measurement of bone structure
Bone specimens were analyzed using micro-computed tomography (µCT) with an R_mCT2 (Rigaku, Tokyo, Japan) (X-ray tube voltage, 90 kV; X-ray tube current, 200 µA). The specimens were imaged at a voxel size of 40 μm × 40 μm and scanning time of 2 min. Bone structures were analyzed using TRI/3D-BON-FCS software (Ratoc System Engineering, Tokyo, Japan), and the region of interest (ROI), consisting of a 3,000 μm × 3,000 μm × 4,000 μm (width × depth × height) area starting from 1,500 μm proximal to the distal end of the metacarpal head, was selected (Supplementary Fig. 7). If a height of 4,000 μm could not be selected for the ROI of the metacarpal head, the maximum height that could be selected was set. The analyzed parameters included the following: volumetric BMD (vBMD, mg/cm3), bone surface density (bone surface/bone volume, BS/BV, 1/mm2), bone volume fraction (bone volume/total volume, BV/TV, %), trabecular thickness (Tb.Th, µm), trabecular number (Tb.N, 1/mm), trabecular separation (Tb.Sp, µm), connectivity density (Conn.D, 1/mm3), trabecular bone pattern factor (TBPf, 1/mm), structure model index (SMI), marrow space star volume (V* m.space, mm3), degree of anisotropy (DA), number of nodes per tissue volume (N.Nd/TV, 1/mm3), and total strut length per tissue volume (TSL/TV, 1/mm2).
Bone histomorphometry
Metacarpal heads were subjected to bone histomorphometric analysis. The specimens for bone morphometry were created from the parts not used in the compression test, which were cut when preparing bone blocks for the compression test. The specimens were fixed in 10% formalin-neutral buffer, washed with water, dehydrated, defatted, and embedded in methyl methacrylate (MMA) without decalcification. The prepared MMA block was shaped using an IsoMet-1000, sectioned into 6-µm-thick sections with a Leica RM2255 microtome (Leica Biosystems, Nußloch, Germany), and stained with Villanueva-Goldner stain. The following parameters were measured using a histomorphometric system with a histometry RT CAMERA (System Supply, Nagano, Japan): BV/TV (%), Tb.Th (µm), Tb.N (/mm), Tb.Sp (µm), osteoid volume (OV/BV, %), osteoid surface (OS/BS, %), osteoid thickness (O.Th, µm), osteoblast surface (Ob.S/BS, %), eroded surface (ES/BS, %), and osteoclast surface (Oc.S/BS, %).
Measurement of implant subsidence
We investigated 10 specimens with the highest bone strength (“strong” bone strength group; ST group) and 10 specimens with the lowest bone strength (“weak” bone strength group; WE group). The magnitude of implant subsidence in the bone was measured by Bass’s method32 using radiographs (front view) at two time points: approximately 1 month and 2.4 years after surgery. The magnitude of subsidence was measured using a routine X-ray imaging and were confirmed by the two authors (R.K.and H.I.) (Supplementary Fig. 8). The magnitude of subsidence per year was calculated and the results were compared. Also, the results of the ANCOVA adjusted for the follow-up period are analyzed (Supplementary Table 1).
Statistical analysis
Clinical data were presented as mean (standard deviation) or number (ratio). Student’s t-test or Welch’s t-test was used to examine mean differences. Correlations between the two variables and simple regression analyses were performed using Pearson’s correlation analysis. Multiple linear regression analysis was performed after reducing the number of variables so that the variance inflation factor (VIF) was less than five, considering multicollinearity. Since the objective of this study was to identify key parameters that may influence the mechanical properties of bone, variables showing stronger correlations with bone strength were retained during the elimination process. Specifically, during variable reduction, two variables with an absolute value of correlation coefficient of ≥ 0.7 were selected, and variables with a low correlation coefficient for correlation with bone strength were reduced (Supplementary Fig. 9). All statistical analyses were performed using JMP 15.0.0 (SAS Institute Japan Ltd., Tokyo, Japan). Statistical significance was set at P < 0.05.
Data Availability
Data is provided within the manuscript or supplementary information files.
Supplementary Information
Below is the link to the electronic supplementary material.
Abbreviations and formal nomenclatures.
All the abbreviations and their formal nomenclatures were shown in order of the origin in Supplementary Table 2.
Acknowledgements
We would like to thank Kureha Special Laboratory for biomechanical testing, preparation of histopathological slides, and bone histomorphometry analysis.
Author contributions
Rika Kakutani: Data curation, Investigation, Formal analysis, Writing – original draft, Writing – review & editing. Hiroto Yoshida: Conceptualization, Methodology, Investigation, Formal analysis, Writing – original draft, Writing – review & editing. Naoki Kondo: Investigation, Formal analysis, Writing – original draft, Writing – review & editing. Mariko Noguchi-Sasaki: Resources, Supervision, Funding acquisition, Project administration, Writing – review & editing. Hajime Ishikawa: Conceptualization, Methodology, Data curation, Supervision, Writing – original draft, Writing – review & editing.
Funding
This study was conducted using research funding from Chugai Pharmaceutical Co. Ltd.
Data availability
Data is provided within the manuscript or supplementary information files.
Declarations
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Smolen, J. S. et al. Rheumatoid arthritis. Nat. Rev. Dis. Primers. 4, 18001 (2018). [DOI] [PubMed] [Google Scholar]
- 2.Smolen, J. S. et al. Treating rheumatoid arthritis to target: recommendations of an international task force. Ann. Rheum. Dis.69, 631–637 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Smolen, J. S. et al. Treating rheumatoid arthritis to target: 2014 update of the recommendations of an international task force. Ann. Rheum. Dis.75, 3–15 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.van Staa, T. P., Geusens, P., Bijlsma, J. W., Leufkens, H. G. & Cooper, C. Clinical assessment of the long-term risk of fracture in patients with rheumatoid arthritis. Arthritis Rheum.54, 3104–3112 (2006). [DOI] [PubMed] [Google Scholar]
- 5.Tsukasaki, M. & Takayanagi, H. Osteoimmunology: evolving concepts in bone-immune interactions in health and disease. Nat. Rev. Immunol.19, 626–642 (2019). [DOI] [PubMed] [Google Scholar]
- 6.Adachi, J. D., Bensen, W. G. & Hodsman, A. B. Corticosteroid-induced osteoporosis. Semin Arthritis Rheum.22, 375–384 (1993). [DOI] [PubMed] [Google Scholar]
- 7.Kim, D. et al. Incidence and risk factors of fractures in patients with rheumatoid arthritis: an Asian prospective cohort study. Rheumatol. Int.36, 1205–1214 (2016). [DOI] [PubMed] [Google Scholar]
- 8.Nyhall-Wahlin, B. M., Ajeganova, S., Petersson, I. F. & Andersson, M. Increased risk of osteoporotic fractures in Swedish patients with rheumatoid arthritis despite early treatment with potent disease-modifying anti-rheumatic drugs: a prospective general population-matched cohort study. Scand. J. Rheumatol.48, 431–438 (2019). [DOI] [PubMed] [Google Scholar]
- 9.Pasco, J. A. et al. The population burden of fractures originates in women with osteopenia, not osteoporosis. Osteoporos. Int.17, 1404–1409 (2006). [DOI] [PubMed] [Google Scholar]
- 10.NIH Consensus Development Panel on Osteoporosis Prevention. Diagnosis, and therapy. Osteoporosis Prevention, Diagnosis, and therapy. JAMA285, 785–795 (2001). [DOI] [PubMed] [Google Scholar]
- 11.Brance, M. L. et al. Whole-hand and regional bone mineral density involvement in rheumatoid arthritis. Rheumatol. Clin. (Engl Ed). 19, 555–559 (2023). [DOI] [PubMed] [Google Scholar]
- 12.McInnes, I. B. & Schett, G. The pathogenesis of rheumatoid arthritis. N Engl. J. Med.365, 2205–2219 (2011). [DOI] [PubMed] [Google Scholar]
- 13.Ochi, K. et al. Ten-year incidences of self-reported non-vertebral fractures in Japanese patients with rheumatoid arthritis: discrepancy between disease activity control and the incidence of non-vertebral fracture. Osteoporos. Int.26, 961–968 (2015). [DOI] [PubMed] [Google Scholar]
- 14.Adler, R. A. & Rosen, C. J. Glucocorticoids and osteoporosis. Endocrinol. Metab. Clin. North. Am.23, 641–654 (1994). [PubMed] [Google Scholar]
- 15.Berris, K. K., Repp, A. L. & Kleerekoper, M. Glucocorticoid-induced osteoporosis. Curr. Opin. Endocrinol. Diabetes Obes.14, 446–450 (2007). [DOI] [PubMed] [Google Scholar]
- 16.De Nijs, R. N. J. Glucocorticoid-induced osteoporosis: a review on pathophysiology and treatment options. Minerva Med.99, 23–43 (2008). [PubMed] [Google Scholar]
- 17.Yoshii, I., Chijiwa, T. & Sawada, N. Rheumatoid arthritis in tight disease control is no longer risk of bone mineral density loss. Osteoporos. Sarcopenia. 6, 75–81 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Yoshida, H. et al. Anti-interleukin-6 receptor antibody prevents loss of bone structure and bone strength in collagen-induced arthritis mice. Scand. J. Rheumatol.47, 384–391 (2018). [DOI] [PubMed] [Google Scholar]
- 19.Mashiba, T. et al. Effects of suppressed bone turnover by bisphosphonates on microdamage accumulation and Biomechanical properties in clinically relevant skeletal sites in Beagles. Bone28, 524–531 (2001). [DOI] [PubMed] [Google Scholar]
- 20.Ott, S. M. Bone strength: more than just bone density. Kidney Int.89, 16–19 (2016). [DOI] [PubMed] [Google Scholar]
- 21.Kijima, Y., Kondo, N., Okumura, G. & Endo, N. Bone histomorphometry of femoral head cancellous bone in patients who underwent total hip arthroplasties due to destructive hip in rheumatoid arthritis. Acta Med. Okayama. 75, 125–131 (2021). [DOI] [PubMed] [Google Scholar]
- 22.Sanchez-Lopez, E., Coras, R., Torres, A., Lane, N. E. & Guma, M. Synovial inflammation in osteoarthritis progression. Nat. Rev. Rheumatol.18, 258–275 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Motta, F., Barone, E., Sica, A. & Selmi, C. Inflammaging and osteoarthritis. Clin. Rev. Allergy Immunol.64, 222–238 (2023). [DOI] [PubMed] [Google Scholar]
- 24.Perilli, E. et al. Dependence of mechanical compressive strength on local variations in microarchitecture in cancellous bone of proximal human femur. J. Biomech.41, 438–446 (2008). [DOI] [PubMed] [Google Scholar]
- 25.Masquers, G., Musy, S. N., Wandel, J., Gross, T. & Zysset, P. K. Bone volume fraction and fabric anisotropy are better determinants of trabecular bone stiffness than other morphological variables. J. Bone Min. Res.30, 1000–1008 (2015). [DOI] [PubMed] [Google Scholar]
- 26.Funamura, K. et al. More than 10 years’ follow-up of the Swanson metacarpophalangeal joint arthroplasty for the rheumatoid hand. J. Orthop. Sci.29, 788–794 (2024). [DOI] [PubMed] [Google Scholar]
- 27.Nemoto, T. et al. Metacarpophalangeal joint arthroplasty using flexible hinge toe implant with grommets for boutonnière deformity of rheumatoid thumb. J. Hand Surg. Asian Pac.23, 66–70 (2018). [DOI] [PubMed] [Google Scholar]
- 28.Arnett, F. C. et al. The American rheumatism association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum.31, 315–324 (1988). [DOI] [PubMed] [Google Scholar]
- 29.Aletaha, D. et al. Rheumatoid arthritis classification criteria: an American College of Rheumatology/European League Against Rheumatism collaborative initiative. Arthritis Rheum. 62, 2569–2581 (2010). (2010). [DOI] [PubMed]
- 30.Rooney, M. et al. Analysis of the histologic variation of synovitis in rheumatoid arthritis. Arthritis Rheum.31, 956–963 (1988). [DOI] [PubMed] [Google Scholar]
- 31.Larsen, A. A radiological method for grading the severity of rheumatoid arthritis. Scand. J. Rheumatol.4, 225–233 (1975). [DOI] [PubMed] [Google Scholar]
- 32.Bass, R. L., Stern, P. J. & Nairus, J. G. High implant fracture incidence with Sutter silicone metacarpophalangeal joint arthroplasty. J. Hand Surg. Am.21, 813–818 (1996). [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Data is provided within the manuscript or supplementary information files.
Data is provided within the manuscript or supplementary information files.