A Longitudinal Study of the Effect of Spinal Degenerative Disease on Bone Density in the Elderly

Objective: To describe the relationship between spinal degenerative disease and bone density in the elderly both cross sectionally and longitudinally.

Methods: Random population based sample of 113 men and 187 women over 60 yrs of age participating in the Dubbo Osteoporosis Epidemiology Study (a longitudinal population based study of fracture risk factors) who had bone density measured on 2 occasions (average interval 2.5 yrs) and spinal radiographs on one occasion (performed according to a standardized approach).

Results: Spinal degenerative disease, of varying severity, was common in this population (osteophytes 69%, disc narrowing 67%, posterior element disease 99%). Scores for osteophytosis, disc narrowing, and posterior element disease (together with age and body mass index) independently explained 43% of the variation in spinal bone density in men (p < 0.00001) and 26% in women (p < 0.00001). The rate of change at the spine increased with increasing severity of osteophytosis in both men (p = 0.03) and women (p = 0.05), but not the other measures, and the total amount of variation explained by these measures was modest. In comparison, severity of spinal degenerative disease had only a modest but significant relationship with femoral neck bone density in both sexes, but not its rate of change. Subjects with any degree of osteophytosis had higher bone density compared to those without osteophytes at both the lumbar spine (men 21% higher, 95% CI 12-31; women 16% higher, 95% CI 9-23) and, to a lesser extent, femoral neck (men 12% higher, 95% CI 4-20; women 6% higher, 95% CI 1-13). Vascular calcification had no relationship with bone density at either spine or hip.

Conclusion: Spinal bone density measurement and its sequential followup may be erroneous in the elderly due to concomitant degenerative disease. Bone density at the femoral neck was much less affected by spinal degenerative disease, which suggests that this site may be more efficacious for both monitoring response to treatment and determining fracture risk in the elderly.