Traumatic injury to the knee joint (such as ligament disruption or meniscal damage) predisposes to secondary osteoarthritis (OA), a debilitating disease of the joints characterized by the focal deterioration and eventual full-thickness loss of articular cartilage and by periarticular bone and soft tissue adaptations. Radiographic evidence of OA includes narrowing of the joint space after cartilage damage, with accompanying changes in periarticular trabecular bone and/or osteophyte formation at the joint margins and ligament entheses. Accordingly, periarticular bony adaptations may be useful as progressive prognostic indicators (and as potential therapeutic drug targets) of OA, particularly during the early stages of OA pathogenesis (1, 2).
Osteophytes arise on bone at the synovial joint margins, from activated cells of the periosteum, initially appearing as cartilaginous growths that later calcify via endochondral ossification (3). Mediators of inflammation secreted by the synovium, such as interleukin-1, tumor necrosis factor α, and transforming growth factor β, are believed to be involved in osteophytogenesis (4, 5). Osteophytes are known to cause joint stiffness by limiting the movement of the joint (6) and leading to fixed bony deformity, and thus are permanent bony changes that increase the severity of OA, and are strong predictors of OA-related pain (7). Therefore, the reduction of osteophyte incidence using drugs known to influence bone mass (such as the bisphosphonates), and/or drugs capable of blocking the potential costimulation of osteophytogenesis by mediators of inflammation (such as nonsteroidal antiinflammatory drugs [NSAIDs]), remains a viable disease-modifying goal of OA therapy.
Our study focused on the reported ability of alendronate to block osteophytosis (8), a process that the closely related bisphosphonate drug risedronate has not been shown to influence (9). The aim of our study was to determine whether the use of bisphosphonate drugs at an early stage of OA pathogenesis reduced periarticular bony adaptations (such as osteophyte severity) and/or cartilage degradation in OA. We also examined whether the addition of an NSAID to that regimen would further reduce osteophytic OA joint adaptations. The NSAID meloxicam is a selective cyclooxygenase 2 (COX-2) inhibitor (10), with known antiinflammatory and antiangiogenic effects (11–13). NSAIDs are prescribed primarily for the management of OA pain (14, 15); however, they have also been found to have positive effects on cartilage health in OA by increasing proteoglycan and hyaluronan synthesis (16). Thus, we hypothesized that the addition of an NSAID to the bisphosphonate drug regimen may further mediate the reduction of the inflammatory stimulus and prove beneficial in preventing or slowing osteophytes and cartilage degradation. Our study further tested the utility of 2 different multimodal noninvasive imaging technologies (high-resolution in vivo microfocal computed tomography [micro-CT] and in vivo micro–magnetic resonance imaging [micro-MRI]), with subsequent histologic validation, to assess mineralized and soft tissue adaptations indicative of OA.
- Top of page
- MATERIALS AND METHODS
- AUTHOR CONTRIBUTIONS
Despite the prevalence of OA in the aging population, there are no currently approved DMOADs (22). The analgesic acetaminophen is the first-line therapy indicated for OA treatment and is used for pain management, with no known disease-modifying benefits. Several bisphosphonate drugs have previously been shown to have disease-modifying treatment outcomes in OA in animal models (23–25) and humans (26–28). However, those studies had contradictory results regarding the disease-modifying efficacy of bisphosphonate drugs, particularly since most studies in humans involved patients with end-stage OA, rather than individuals at an earlier stage of OA pathogenesis.
In this study, we evaluated the disease-modifying effects of the bisphosphonate drugs risedronate and alendronate and the NSAID meloxicam on the progression of posttraumatic OA in an established rat model (25, 29, 30). We found that alendronate significantly, and risedronate to some degree, reduced trabecular bone volume loss in the medial femoral compartment compared with untreated rats early after a knee triad injury. These findings were consistent with those of previous studies examining posttraumatic OA in several other animal models (9, 31) and were also supported by the results of investigations of risedronate in later-stage human OA (26).
With respect to osteophytosis, risedronate did not reduce or retard bony osteophyte development in rats with knee triad injury, which was consistent with previous results in a rabbit model of OA (9). In contrast, alendronate showed a trend toward reducing the number and severity of osteophytes in the rat knee triad injury model, although the results were not statistically significant compared with the untreated rats with knee triad injury. Our findings are consistent with those of an earlier study (8) and suggest that alendronate inhibits osteophytogenesis by an alternative mechanism. This mechanism by which alendronate exerts its antiosteophytic activity remains to be determined and is a potential direction for further research.
In the second part of our study, we found that the NSAID meloxicam not only preserved bone volume after a knee triad injury, but also, surprisingly, led to increased bone volume compared with all other treatment groups. As a selective inhibitor of the inducible mediator of inflammation COX-2, meloxicam was previously shown to negatively impact bone healing (32), possibly since COX-2 regulates genes that are critical for osteoblast differentiation (33). This reduced periosteal osteoblastogenesis would therefore likely also reduce subsequent endochondral bone formation. However, the increased epiphyseal trabecular bone volume observed in our study is supported by the observation that disruption of COX-2 gene expression by meloxicam also results in reduced RANKL and osteoprotegerin expression in osteoblasts (34). Thus, a reduction in RANKL levels may have contributed to the maintenance of bone mass in the group of rats with knee triad injury treated with meloxicam. On the other hand, the reduced remodeling observed in all bisphosphonate-treated groups can be attributed to the effects of the potent bisphosphonate drugs. In that scenario, the combination therapy likely did not increase the amount of trabecular bone since bone formation was suppressed by the action of the potent N-containing bisphosphonate drug; this is a well-known consequence of the use of bisphosphonate drugs (35).
It is also important to note that in the present study bone volume percentages appeared to have increased in all groups between the first and second stage of the study (data not shown). These changes could represent differences in the ages of the rat test populations in the first and second parts of the study, since they were supplied based on body mass. Nonetheless, relative to the untreated rats with knee triad injury and untreated uninjured rats in the second part of the study, the rats with knee triad injury treated with meloxicam had a significantly higher bone volume percentage. Interestingly, rats treated with both risedronate and meloxicam did not display any synergistic effects in increased bone volume, but had less bone volume than that observed in rats treated with meloxicam alone. The addition of risedronate may have reduced metaphyseal osteoblastic activity and resulted in reduced bone volume compared with rats treated with meloxicam alone.
Beyond the changes observable on radiography, OA may present pathophysiology that can be detected using physiologic imaging modalities such as MRI. One such phenomenon is a bone marrow lesion (historically termed bone marrow edema), which presents as a high-intensity, diffuse fluid signal within the epiphyses of the femur and tibia of many patients with knee OA (19, 20). A bone marrow lesion is identified by increased signal on fat-suppressed, T2-weighted scans, with lower signal contrast to background in T1-weighted scans of the same subchondral bone marrow region. The histologic characteristics of bone marrow lesions in OA remain a subject of controversy. Necrotic bone marrow, bone marrow fibrosis, and occasionally bone marrow edema have been associated with bone marrow lesions in OA (36), and bone marrow lesions are frequently observed in association with the cartilage degradation seen in OA (37). Patients who exhibit bone marrow lesions are more likely to develop severe OA over the subsequent 15–30 months (38), and thus, the early detection of bone marrow lesions with cartilage deterioration may improve the management and treatment of OA.
A major finding of our study was the diffuse, bone marrow lesion–like fluid phase we detected on fat-suppressed, spin-echo, T2-weighted micro-MRIs of the distal femoral epiphyseal marrow space, 8 weeks after injury, in untreated rats with knee triad injury. In stark contrast, a very thin and discrete fluid signal was evidenced in the epiphyses of normal control rats with no surgical injury, suggesting a quiescent basal level of fluid content. In rats with knee triad injury treated with alendronate and rats with knee triad injury treated with risedronate, the diffuse bone marrow lesion–like fluid phase remained 8 weeks after knee triad injury surgery. Conversely, the rats with knee triad injury treated with meloxicam showed a reduced bone marrow lesion–like signal, suggestive of a treatment-based antiinflammatory outcome. The bone marrow lesion–like phenomenon identified using micro-MRI may have a potential application as a measure of OA disease progression and drug efficacy in this posttraumatic rat OA model, and warrants further detailed investigation.
Histologic assessment of the knee joint using a modified Mankin scoring system revealed that the bisphosphonates alendronate and risedronate tended to reduce the progression of OA and maintain cartilage health in this short-term study, although the results were not statistically significant compared with findings in untreated controls with knee triad injury. Those modest chondroprotective effects were consistent with the previously demonstrated positive effects of alendronate therapy in OA (8), and were also supported by the results of investigations of risedronate in later-stage human OA (27, 28, 39).
In conclusion, our data suggest that bisphosphonates and meloxicam potentially reduce the progression of OA following joint trauma in the rat knee triad injury model and could have additional clinical applications regarding the management of secondary OA. Alendronate and meloxicam were effective in maintaining and increasing bone volume, respectively, while alendronate also displayed some positive effects in inhibiting osteophytosis. Meloxicam alone, or in combination with risedronate, reduced the bone marrow fluid signal at the site of knee triad injury when quantified and compared with signal in sham-operated normal controls. Further research is necessary to elaborate on the precise mechanisms by which bisphosphonates and meloxicam operate in inhibiting the periarticular bone– and fluid-based changes associated with joint damage secondary to trauma. The findings of our study may be applicable in the pharmacologic management of disease following joint damage, secondary to traumatic rupture of the ACL and/or medial meniscus, as is often encountered in sports injuries. In those situations, it may prove feasible to conservatively manage and preserve periarticular bone microarchitecture and retard osteophyte formation by using bisphosphonates and NSAIDs as combination therapy for a discrete therapeutic window immediately after the injury and prior to subsequent surgical joint reconstruction.
- Top of page
- MATERIALS AND METHODS
- AUTHOR CONTRIBUTIONS
All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Doschak had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study conception and design. Zernicke, Doschak.
Acquisition of data. Jones, Tran, Li, Doschak.
Analysis and interpretation of data. Jones, Maksymowych, Zernicke, Doschak.