Risk factors for lumbar intervertebral disc height narrowing: a population-based longitudinal study in the elderly

Abstract

Background: The progression of disc degeneration is generally believed to be associated with low back pain and/or degenerative lumbar diseases, especially in the elderly. The purpose of this study was to quantitatively evaluate changes in lumbar disc height using radiographic measurements and to investigate risk factors for development of disc height narrowing of the elderly.

Methods: From 1997 to 2007, 197 village inhabitants at least 65 years-old who participated in baseline examinations and more than four follow-up examinations conducted every second year were chosen as subjects for this study. Using lateral lumbar spine radiographs of each subject, L1-L2 to L5-S1 disc heights were measured. The subjects were divided into two groups according to the rate of change in disc height: mildly decreased (≤20 % decrease) and severely decreased (>20 % decrease). A stepwise multiple logistic regression analysis was used to select those factors significantly associated with disc height narrowing.

Results: Disc height at each intervertebral disc (IVD) level decreased gradually over ten years (p < 0.01, an average 5.8 % decrease of all disc levels). There was no significant difference in the rate of change in disc height among the IVD levels. Female gender, radiographic knee osteoarthritis and low back pain at baseline were associated with increased risk for disc height narrowing.

Conclusions: We conducted the first population-based cohort study of the elderly that quantitatively evaluated lumbar disc height using radiographic measurements. The risk factors identified in this study would contribute to a further understanding the pathology of disc degeneration.

Fig. 1

Disc height measurement. Ha: anterior disc height, Hp: posterior disc height, Ds: superior disc depth, Di: inferior disc depth. Disc height is expressed as the disc height index (DHI), which was calculated as: [(Ha + Hp)/(Ds + Di)] × 100

Fig. 2

Rate of change in disc height index (∆DHI). The ∆DHI shows a significant gradual decrease over the ten-year period (p < 0.0001) (Fig. 3). An analysis of variance revealed no significant disc level and time-point interactions (p = 0.63). The ∆DHI at each level shows similar changes over ten years that did not differ significantly among the disc levels (p = 0.78)

Fig. 3

Percentage of discs classified as severely and mildly decreased at each disc level by rate of change in disc height index (∆DHI). At each disc level, the severely decreased group was approximately 30% and the mildly decreased group was approximately 70%

Fig. 4

Number of subjects with severely decreased disc height by disc level. The graph shows the number of subjects whose lumbar discs was identified as severely decreased at each level. The illustration below the graph is a representative image of the severely decreased degeneration pattern by number of subjects