Defects of the vertebral end plate: implications for disc degeneration depend on size

Bony vertebral end plates must be porous to allow metabolite transport into the disc, and yet strong to resist high intradiscal pressure (IDP). End plate defects may therefore have nutritional and mechanical consequences for the disc, depending on their size and type. We hypothesize that broad, diffuse defects are more closely associated with disc decompression and degeneration than are focal Schmorl's node-type defects.

This study aimed to determine how the size and type of end plate defects are related to decompression and degeneration in the adjacent intervertebral disc.

Mechanical, histologic, and micro-computed tomographic investigations were carried out in cadaver spines.

The study involved 40 motion segments (T8-T9 to L4-L5) dissected from 23 cadavers aged 48-98 years. Intradiscal stresses were measured, under 1 kN compression, by pulling a pressure transducer along the disc's midsagittal diameter. The resulting "stress profiles" revealed nucleus pressure (IDP) and maximum stresses in the anterior and posterior annulus. Micro-computed tomography was then used to examine all 40 discs, with 5 mm of adjacent bone on either side, so that end plate defects could be characterized at a resolution of 35 µm. Cross-sectional area (in the transverse plane), volume, location, and morphologic type were determined for all bony defects in the 80 end plates. Finally, discs from each motion segment (with hyaline cartilage and bone attached) were sectioned (undecalcified) at 7 µm for histology to allow degeneration to be assessed.

Substantial defects were identified in 24 of 40 specimens (35 of 80 end plates). Of these, 83% was centrally located, and 17% was laterally located. Defects occurred more frequently in male than female specimens (p=.043), and were more common in thoracic than lumbar end plates (p=.002), although lumbar defects were greater in volume (p=.05). Defect area and volume increased with decreasing IDP, with decreasing peak stress in the annulus, and with increasing tissue degeneration. Stepwise multiple regression showed that average defect area depended most strongly on IDP, whereas maximum defect area and volume depended most strongly on peak stress in the anterior annulus. Multiple end plate defects were associated with lower values of IDP and higher degeneration scores when compared with erosions and Schmorl's nodes.

Disc degeneration has a stronger association with large or multiple end plate defects than with small or single defects (of any type). Large end plate defects probably allow greater volume changes within the disc, leading to greater nucleus decompression.