For personal use. Only reproduce with permission from The Lancet Publishing Group. Osteoarthritis has traditionally been regarded as a disease of articular cartilage, with cartilage degeneration and, finally, erosion being the main identifying features. Whether degeneration and erosion is the primary pathological process or a secondary effect of the disease is unclear because there are equally prominent changes in other articular tissues—notably bone—as orthopaedic surgeons will testify. Any hypothesis about the cause and pathology of osteoarthritis has to account for all features of the disease. Because research has focused almost exclusively on articular cartilage and mechanical loading, we believe that this work offers no true explanation for the diversity of effects throughout the body. We think that a widely recognised form of the disease exists, in which changes occur in many joints and in sites remote from joints, which is well described by the term primary generalised osteoarthritis. Hypothesis We propose that primary generalised osteoarthritis is a metabolic disorder in which systemic factors induce changes in skeletal tissues by modification of the formation and biosynthetic activity of cells derived from mesenchymal precursors. Once tissue homoeostasis has been disturbed and maintenance of skeletal integrity impaired, normal wear and tear might lead to cartilage damage that has traditionally been the focus of osteoarthritis. In this case, the erosion of articular cartilage would be caused by mechanical forces, but the disease itself would not. However, this erosion is only a small part of the disease process. Our hypothesis also offers explanations for the changes in bone, even at sites remote from the joint, in other tissues such as muscle, and for the link with obesity. Changes in articular cartilage in osteoarthritis have been extensively studied. In the early stages of this disease HYPOTHESIS 1118 THE LANCET • Vol 357 • April 7, 2001 Lancet 2001; 357: 1118–20 Department of Orthopaedic Surgery, University of Aberdeen, Aberdeen AB25 2ZD, UK (Prof R M Aspden DSc, B A A Scheven PhD, J D Hutchison PhD) Correspondence to: Prof R M Aspden (e-mail: r.aspden@abdn.ac.uk) the tissue softens and the surface undergoes fibrillation before full-thickness cartilage loss. Multiple cell clones appear within their enclosing chondrons, suggesting an attempt by the chondrocytes to proliferate. The two possibilities widely believed to be the cause of osteoarthritis are abnormal loads on normal cartilage or normal loads on abnormal cartilage. Findings in some animals and secondary osteoarthritis in man provide clear evidence of a role for abnormal loading. However, such loading causes degeneration principally in only one joint, and no generalised changes take place. 1 The evidence for abnormalities in cartilage resulting in tissue breakdown and for osteoarthritis with normal loading is sparse, partly because of the impossibility of obtaining tissue samples from patients in the early stages of disease. Spontaneous osteoarthritis in animals offers little help because of uncertainty whether animals are true models for the pathological process or merely for the final outcome. Although some genetic factors that result in cartilage matrix modifications have been identified, they are rare and do not explain the high frequency of the disease. We believe that a new approach is required. Osteoarthritic changes are not limited to cartilage. Changes in juxta-articular bone are also well described, with sclerosis of the subchondral bone, the formation of so-called cysts within the bone, and the development of osteophytes. This involvement of bone is widely recognised, and is generally believed to be a secondary consequence of the disease, although some investigators disgree. 2 Reduced mineralisation, increased formation of woven bone, and evidence for enhanced osteoclastic activity have been shown in bone matrix, 3 and in osteoblast behaviour, 4 and are difficult to explain as secondary changes. Patients with osteoarthritic changes seen on hip radiographs were seen to have a higher than average bone-mineral density, not only in the hip but also in the distal radius, vertebrae, and calcaneus, 5 and altered mineralisation and increased concentrations of growth factors were recorded in the iliac crest of patients with osteoarthritis of the hand. 6 These matrix changes go with cellular changes. Osteoblasts from patients with osteoarthritis proliferate in vitro more rapidly than those from healthy individuals and express different concentrations of markers. 7 Clearly, the amount of bone, Osteoarthritis as a systemic disorder including stromal cell differentiation and lipid metabolism Richard M Aspden, Ben A A Scheven, James D Hutchison Hypothesis For many years articular cartilage has been the focus of research aimed at improving understanding of and treatment for osteoarthritis. Although much is known about the tissue, research has had little success in elucidating the pathogenesis of generalised osteoarthritis. A new hypothesis is required. Substantial changes in many tissues, including bone, muscle, ligaments, and joint capsule, as well as cartilage, are increasingly recognised in this disease, and not all these changes are localised to the affected joints. There is also a well established link with obesity. These observations, the common origins of the mesenchymal cells that maintain these tissues, and the possible role of neuroendocrine factors that can regulate bone mass, result in the hypothesis that systemic factors that include altered lipid metabolism could explain the diversity of physiological changes in generalised osteoarthritis. If proven, this hypothesis could have important implications for a new approach to pharmacological intervention in the early stages of the disease.