INTRODUCTION: Lumbar disc degeneration (DD) has been widely investigated but its association with degeneration in cervical and thoracic regions and the exact difference in prevalence between patients and controls in whole spine has not been fully documented.
METHODS: Whole spine MRI of 299 individuals (198 patients, 101 controls) was analyzed. Lumbar discs were analyzed for total endplate damage score (TEPS), DD, herniation, annular tears and Modic changes. Thoracic and cervical regions were assessed for the presence of disc degeneration. All the observations were made by two independent observers (κ=0.9). Hierarchical clustering program was used to group degenerated discs. Binomial logistic regression analysis was used to find the predictors of disc degeneration.
RESULTS: Patients and controls were age matched (38+/−9, 33.5+/− 10, p=NS) but only 4.3% of patients had no DD at any level compared to 70% of controls (p=0.000). Of the 30 controls who showed any DD, 27 (90.3%) were confined to lower two lumbar segments. In contrast 42 (21.6%) patients had DD extending to the thoracic and cervical spine also (p=0.005). Similarly DD involving all lumbar segments was only in 1.4% of controls compared to 10.8% in patients (p=0.005). Only 5 discs in controls had a TEPS of 6 or more compared to 172 discs in patients (p
Cluster analysis showed that DD of lower lumbar segments have a high tendency of association with lower cervical degeneration and upper lumbar DD had a higher association with thoracic DD.
DISCUSSION: Endplate damage rather than annular tear was the primary cause for DD. Patients had a significantly higher incidence of DD and wider distribution in all regions of spine whereas controls had DD restricted to lower two lumbar levels. This could imply a different etiopathogenesis of DD in the two groups.