Subchondral bone of the human knee in aging and osteoarthritis

Objective: Although most research investigating the pathogenesis of osteoarthritis (OA) has focused on cartilage, it has been suggested that the subchondral bone (SCB) plays an important role in the development of OA. The relationships between aging, severity of OA change and the SCB thickness and density in the human knee joint specimens from a wide range of ages were examined. Methods: One hundred forty knee joints from 72 individuals (25 females, 45 males and 2 unknowns; average age 54.8 years, range 17 to 91 years) were obtained. The surface of the articular cartilage of both the femur and tibia was evaluated for gross morphological changes with a 4-point grading scale. The lateral and medial femoral condyles were cut along a sagittal plane and the tibia along a coronal plane to make bone and cartilage strip specimens. The strips were X-rayed onto mammography film and then scanned into a computer for assessment of SCB thickness and density using image analysis software. Results: Medial tibial SCB thickness was significantly lower among the elderly (age>69 years) than among the young (age<40) or the middle-aged (40 to 69) (P<0.001 via ANOVA). Lateral tibial SCB thickness also showed the same trend of decreasing thickness with increasing age, but differences between age groups were not statistically significant. Tibial SCB thicknesses were significantly lower in arthritic grades compared to normal grades (P=0.008 in lateral and 0.017 in medial via ANOVA); in contrast, no significant differences between normal and arthritic were found in femoral SCB thicknesses. The arthritic group tended to have lower SCB densities than the normal group, but this was statistically significant in only the lateral femoral condyle. Conclusions: The results obtained in the present study are not consistent with generally accepted notions of the relationship between subchondral bone thickness or density and OA. Subchondral bone changes are not etiologic for OA but, more likely, are secondary to loss of articular cartilage which precedes the appearance of subchondral sclerosis. (C) 2002 Published by Elsevier Science Ltd on behalf of OsteoArthritis Research Society International.