Statistics from Altmetric.com
Subchondral bone density as a possible pathogenetic factor contributing to articular cartilage degeneration in osteoarthritis (OA) has been receiving increasing attention.1 The Kellgren-Lawrence system for grading joint OA emphasised subchondral bone sclerosis and osteophyte formation more than joint space narrowing, a view, however, that has recently been questioned.2 The importance of subchondral bone density in contributing to cartilage degeneration has been emphasised for years in the work of Radin and colleagues.3 Observations that dense subchondral bone may indeed precede cartilage degeneration and thereby be an initiating factor, or that bone density may be involved in the progression of OA, have been advanced by studies on animal models.4 5 However, little is known about the basis for subchondral bone density in OA, or its consequences.
Colony stimulating factor-1 (CSF-1) is the primary regulator of mononuclear phagocyte production and has important effects on bone.6 7 Among other cell types, CSF-1 is secreted by osteoblasts, and is required for the differentiation of osteoclast progenitors to osteoclasts.6 Our interest in CSF-1 in relation to bone density in OA arose from studies on a mouse model of osteopetrosis (op/op)8 where an inactivating mutation in the gene encoding CSF-1 impairs osteoclast development; the bone formed cannot be resorbed. We considered that increased bone density in the joint in OA is most probably caused by low CSF-1 concentrations in the subchondral bone; however, it is also possible that low concentrations of synovial fluid CSF-1 could contribute to the increased bone density. We used a radioimmunoassay to measure CSF-1 concentrations in synovial fluid and plasma from subjects with severe knee OA, and compared the results with similar studies on a “control” group with acute traumatic or athletic injury to the knee.
Table 1 summarises aspects of the clinical condition of patients in group A with the diagnosis of OA of the knee,9 and those patients in group B who sustained injuries of the knee and underwent arthroscopy. Circulating CSF-1 was measured in plasma and synovial fluid by a modified10 radioimmunoassay (RIA) that specifically detects biologically active CSF-1 in the range of 0.02–3.60 ng/tube. Synovial fluid samples were incubated with one tenth volume of bovine testicular hyaluronidase (Sigma) (1500 units per ml in RIA buffer) for 10 minutes at 37°C before RIA. Normal rabbit serum (20 μl) was added to 20–100 μl samples of the hyaluronidase treated synovial fluid to control for protein concentration in the RIA before assay. Differences between the groups with regard to standardised CSF-1 concentrations in blood and synovial fluid, as well as age, were tested for significance using t tests for independent samples. All tests of significance were two tailed with an α level of 0.05. No power analysis was performed.
In this study there were nine patients in group A with knee complaints extending over many years and radiographically severe OA. Subsequently, five of these patients had a total knee replacement of the joint that was aspirated during the course of the study. None of the eight patients with knee injury in group B had complaints of more than six months’ duration and none had radiographic evidence of OA. Table 2shows that age was significantly different (p<0.001) between the two groups, as would be anticipated based on the late life clinical manifestations of OA.11
The studies reported here seem to be the first to use a RIA specific for CSF-1, and to compare plasma and synovial fluid samples simultaneously. As the half life of circulating CSF-1 is only about 10 minutes,12 and most of the CSF-1 target cells are not in the joint,7 it seemed unlikely that a significant difference in the circulating CSF-1 concentration (rather than a difference in local concentration) would be observed in OA. This opinion is supported by the lack of significant difference in the plasma CSF-1 concentrations of patients with severe OA and the control group (table 2). In this pilot study, the number of subjects with OA was small, although earlier studies on synovial fluids in OA patients were also done on only two,13 eight,14 and 1315 subjects, respectively. No statistically significant differences in the synovial fluid CSF-1 concentrations were found between the two groups in this study (table 2). Thus, these results indicate that patients with severe OA do not have a decreased synovial fluid CSF-1 concentration.
It is possible to conceive of OA initially as a “limited form of osteopetrosis” restricted to the subchondral plate, with ensuing cartilage degeneration. Our findings that the concentrations of CSF-1 in synovial fluids of subjects with severe OA were not appreciably lower than those from subjects with acute knee injury indicates that OA may not be related to low synovial fluid CSF-1 concentrations. These studies, however, do not eliminate the possibility that the reduced synthesis or recruitment of CSF-1 at the local subchondral site is important for the increased bone density of the joint in OA.
This work was supported in part by NIH grants CA 32551 (to ERS) and AG 14442 (to DH).
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.