| Osteoarthritis And Osteoporosis: Is There A Relationship? |

Category: Practitioners

Asgar Ali Kalla
Professor of Rheumatology
University of Cape Town

Interest in the relationship between osteoarthritis (OA) and osteoporosis (OP) goes back several decades. This first began when orthopaedic surgeons noticed that OA changes were generally absent in patients who needed operations for osteoporotic hip fractures. It was also found that OA protected against hip fractures and compressive vertebral fractures. In clinical practice, the co-existence of the two disorders in individual patients is considered to be rare. This inverse relationship between OA and OP continues to be debated, in spite of these important clinical observations.

In OA, the large joints are affected predominantly, but small joints of the hands and feet can also be affected. The hip, knee and spine are the commonest sites of pain in OA, but in OP the pain is usually due to a fracture. When the large joints are involved this may affect a single joint (hip or knee), but occasionally one may see bilateral disease. The hands characteristically develop bony swellings at the tip of the fingers (distal interphalangeal joints) and hand function generall remains preserved.

Studies of bone mineral density (BMD) of the hip in patients with generalised OA (hip, hands, spine) have shown controversial results, some suggesting increased BMD in the OA group and others refuting these results. The Rotterdam study (1) of 1996, was the first longitudinal evaluation of radiographic OA and femoral bone density and loss, adjusted for age and body mass index (BMI). They followed up 1723 subjects for two years and demonstrated a graded relationship of hip OA with higher BMD and an increased age-related bone loss in men and women aged 55 years or older, suggesting a more pronounced difference in BMD in earlier life. A case study of 574 female twin pairs confirmed the existence of an inverse relationship between OA and OP at the hip (2). This relationship was localised to the ipsilateral hip and there was no clear association of radiological hip OA features and BMD in the contralateral hip or in the spine. Severe osteophyte (bony protrusions) formation was associated with a 5-6% higher BMD at the ipsilateral hip.

The association of knee OA with bone density has also been extensively studied. In the Chingford study, women aged between 45-64 years with knee OA of at least grade 2 on x-ray, had higher BMD, 7.6% at the spine and 6.2% at the femoral neck, compared with controls (3). A subsequent longitudinal study by the same group showed that women who later developed incident OA had significantly higher baseline BMD (4). A cross-sectional study using radiographic knee OA as the entry criterion (Baltimore), found a moderate increase in adjusted levels of spinal BMD, statistically significant only in men (5). The presence of any osteophytes was strongly associated with higher spinal BMD in men and in women. There were no differences in adjusted mean BMD at the hip by the presence of any evidence of knee OA in either men or women. These findings have not been confirmed by other studies, which measured subchondral BMD in using the spine software in patients with knee OA. They found that BMD was significantly lower than normal in six subchondral bone regions of the knees of white females with mild knee OA. The overall average BMD was 13.3% lower in the OA knees. African American females with mild OA knee had 9.1% lower BMD in one subchondral region only.

A number of studies have been undertaken to evaluate BMD in patients with generalised OA of the hands. Hochberg analysed a large sample from the Baltimore Longitudinal study of Ageing and failed to show any association between increased appendicular bone mass and the presence and severity of hand OA, both in men and women (6). Appendicular bone mass was estimated as percent cortical area (PCA) of the second metacarpal and BMD of the left radius measured by single photon absorptiometry (SPA).

A radiographic hand arthritis score (HOA) was examined in relation to BMD measurements at several sites in 300 healthy women aged 75 years or older (7). The HOA score positively correlated with total body, spinal and limb BMD, but there was no significant association with femoral BMD.The osteophytosis score correlated with BMD at all sites. Women with high HOA combined scores had higher values of BMD in spite of being older. Also, BMD adjusted for age was significantly associated with the HOA combined score.

With respect to spinal OA, there is always the problem of osteophytes, intervertebral disc space narrowing and end-plate and facet joint sclerosis interfering with the BMD readings. In a cohort of 375 women, aged 50-85 years, it was found that BMD was increased in women with spinal OA, not only at the lumbar spine but also at the femoral neck and total body (8). This study also found a decrease in both resorption and formation markers, suggesting decreased bone turnover in women with spinal OA.

Some of the issues relating to these studies include an absence of consensus on how OA should be defined. There is a huge difference between studies of patients receiving treatment for OA and epidemiological studies of radiographic OA in non-clinic populations. Therefore, conflicting findings could reflect variations in the emphasis on osteophytes versus narrowing in the interpretation of radiological grades. Additionally, earlier studies used small samples with elderly subjects and failed to consider or to adjust for important confounders of the relationship between OA and OP. Different techniques were used at different sites in different studies, most of the earlier negative studies using crude technology. However, the newer dual absorptiometry techniques (DXA) have contributed widely in the accomplishment of large epidemiological studies.

The most important issue to consider is whether the increased BMD associated with OA is accompanied by a reduced risk of fracture. Two population-based cohort studies of hip fracture and self-reported AO (9,10) have shown that OA is associated with a reduced risk of hip fractures. Other researchers have corroborated these findings. On the other hand, in a study examining the association between self-reported validated fractures and radiographic OA at multiple sites, another group found that despite having 5% higher BMD, subjects with hip OA had a significantly increased risk of fracture compared to normal controls (11). Subjects with lumbar spine OA, however, had a significantly reduced risk of fracture, while no clear association was seen with fracture for hand and knee OA. The authors suggested that the increased risk of fracture in subjects with hip OA was most likely due to mechanical and locomotor factors, particularly the risk of falling.

It is clear that OA is a heterogeneous disease with different risk factors for different joints and anatomic areas within the joint. This heterogeneity can account for the conflicting results in the studies, but could also provide clues to the site-specific pathogenesis of OA. In general, the large population studies based on densitomtry techniques showed a relationship between OA of large joints such as hip, knee, spine and total body, lumbar spine and femoral neck BMD. However, the relationship weakens when bone density is measured at a peripheral site (e.g. radius), consisting primarily of cortical bone. A consistent feature throughout the studies is the relation of the presence of osteophytes with increased bone mass. No association with moderate or severe joint space narrowing in isolation was found. Spinal osteophytes can falsely increase the lumbar BMD, but it is unlikely that hip osteophytes affect the femoral BMD measurements because most hip osteophytes occur near the acetabulum and are outside the regions of interest used for BMD measurement.

In summary, an association between higher BMD and radiographic OA of knees and hips undoubtedly exists. The relationship at other sites is less clear, probably reflecting different mechanisms involved in the pathogenesis of OA. The presence and severity of osteophytes is strongly related to higher bone mass. However, subjects with OA do not appear to be at lower risk for fractures, suggesting that the extra bone is not mechanically useful. Recently published longitudinal studies confirmed the relationship by showing that a higher baseline BMD increases the risk of incident radiographic knee OA. These studies also showed that factors affecting progression may be different.

REFERENCES

Burger H, et al. Association of radiographically evident osteoarthritis with higher bone mineral density and increased bone loss with age. The Rotterdam study. Arthritis Rheum 1996; 39: 81-86

Antoniades L, et al. A cotwin control study of the relationship between hip osteoarthritis and bone mineral density. Arthritis Rheum 2000; 43: 1450-1455

Hart DJ, et al. The relationship between osteoarthritis and osteoporosis in the general population: The Chingford study. Ann Rheum Dis 1994; 53: 158-162

Hart DJ, et al. The relationship of bone density and fracture to incident and progressive radiographic osteoarthritis of the knee: the Chingford study. Arthritis Rheum 2002; 46: 92-99

Lethbridge-Cejku M, et al. Axial and hip bone mineral density and radiographic changes of osteoarthritis of the knee: data from the Baltimore Longitudinal study of Aging. J Rheumatol 1996; 23: 1943-1947

Sowers M, et al. The association of bone mineral density and bone turnover markers with osteoarthritis of the hand and knee in pre- and post-menopausal women. Arthritis Rheum 1999; 42: 483-489

Marcelli C, et al. The relationship between osteoarthritis of the hands, bone mineral density, and osteoporotic fractures in elderly women. Osteoporosis Int 1995; 5: 382-388

Arden NK, et al. Osteoarthritis and risk of falls, rates of bone loss, and osteoporotic fractures. Study of Osteoporotic Fracture Research Group. Arthritis Rheum 1999; 42: 1378-1385

Lane NE, et al. Osteoarthritis, bone mass and fractures: how are they related? Arthritis Rheum 2002; 46: 1-4

Moskowitz RW. Bone remodelling in osteoarthritis: subchondral and osteophytic responses. Osteoarthritis Cartilage 1999; 7: 323-324

Mansell JP, et al. Abnormal cancellous bone collagen metabolism in osteoarthritis. J Clin Invest 1998; 101: 1596-1603

Published on 2005-06-08