Degeneration of the intervertebral disc results in patent cracks [1] and a decrease in osmotic pressure associated with loss of fixed charges. The relationship between mechanical load and damage in the disc is very poor [2]. This finding is at odds with physical intuition. The subject of this study is relationship between the development of patent cracks and the decrease in osmotic pressure in the degenerating disc in the light of the physics of swelling [3–7]. We restrict the experimental part of this study to hydrogel, thus avoiding complications associated with biological variability. The finite element modelling [6,7] used in this study catches salient features of stress profiles measured by Mc Nally and al.

Thin hydrogel samples with a crack of 5 mm are used. The crack opens as a result of decreasing osmotic pressure in the experiments and in the simulation. The initial uniform stress distribution turns into a distribution with a decreased average stress level and a high stress around the crack tip. A decrease in osmotic pressure opens an existing crack in swelling materials independently from external mechanical load. Hence, disc degeneration causes the overall stress to decrease, while local stress around a crack tip increases. This mechanism may explain why damage in the disc is so poorly correlated with mechanical load [3] and why the degenerated disc is characterized by patent cracks [1]. The process of crack opening in the degenerating disc is comparable to the crack development in an aging oaken beam, while loosing its turgor.

Introduction: In vivo measurements of intradiscal stresses are difficult. McNally measured stress profiles in human discs. It is unclear why some exhibit stress peaks in posterior annulus while others do not. Therefore finite element (FE) models are useful to improve the knowledge of stress distribution in the disc. We compared experimental and numerical stress in discs under axial loading, in non degenerated and degenerated disc.

Methods: The FE disc model resembles one fourth of a full disc. The annulus contains both matrix and fibers, while the nucleus only consists of matrix. Similar load profiles were applied and model predictions of matrix stress were compared to experiments (stress profilometry).

Results: Both experimental data and numerical simulations exhibit a peak of axial stress in posterior annulus and lower peaks in anterior annulus. Simulating a “normal” disc results in a uniform matrix stress profile from posterior to anterior. By reducing the fixed charged density (FCD) to 50% in both nucleus and annulus, stress profiles become non-uniform. Stresses in the nucleus decrease. Axial annulus stresses exhibit peaks on anterior and posterior side. Stress peaks increase when FCD decrease under the same loading.

Discussion: The size of the peaks computationally depends on the FCD in discs. Decreasing the FCD shows development of stress peaks in the annulus. A uniform stiffness is seen in nucleus region, but not in annulus. The hydrostatic pressure, due to the FCD, is not high enough to evenly distribute the load over the whole disc. The posterior stress peaks may explain why hernia develops particularly in the posterior annulus.