The effects of altering intervertebral stiffness on passive movement testing of the cervical spine: A computer-based simulation

Abstract

Purpose: To determine how alterations in segmental intervertebral stiffness at different locations affect passive posteroanterior movements (PAs). Relevance: Passive movements such as PAs are common techniques intended to assess stiffness of individual intervertebral motion segments. PAs were originally considered to produce isolated accessory glides between vertebrae and were described as if the PA movement occurred entirely at the target location. It is now known that PAs involve both local and regional structures, but it is not known, how PAs might be affected by altered mobility in different structures. Methods: A simplified two dimensional computer-based model of the cervical spine was constructed using size and stiffness values from the literature. The model consisted of eight bodies corresponding to the head and cervical vertebrae linked by seven uniaxial joints. Stiffness-displacement curves were plotted for simulations of PA movements to C4/5 with the joints first modelled as constant stiffness (linear springs). The simulations were repeated with the joints modelled as minimal stiffness in the neutral zone and linear interpolations between the known data points at 0.33, 0.66 and 1.00 Nm (non-linear springs). For the linear model, the stiffness of C4/5 (target level) and C3/4 (adjacent level) were separately increased by 50%. For the non-linear model the intervertebral stiffness was increased in two ways for both the target and adjacent intervertebral levels; the size of the neutral zone was decreased and the stiffness within the neutral zone was increased. For each simulated PA movement, the stiffness was plotted against displacement. Results: For the linear model, altering the intervertebral stiffness at either location simply altered the constant stiffness of PA movements. The PA stiffness from the non-linear model consisted of an area of constant stiffness followed by successive 'steps' to levels of greater stiffness. Decreasing the size of the neutral zone resulted in all of the steps occurring earlier in the PA movement, but the height of the steps remaining unchanged. Increasing the stiffness within the neutral zone on the other hand resulted in an increase in the height of the steps rather than a change in their location. For both the linear and non-linear simulations, the target level had a much greater impact on PA stiffness than the adjacent level. Conclusions: If intervertebral stiffness is modelled as linear, it is not possible to differentiate between differences in PAs resulting from increased stiffness in the target and adjacent levels as both have a uniform effect throughout the PA movement. If intervertebral stiffness is modelled as non-linear, reducing the intervertebral mobility in two ways each results in characteristic differences in the pattern of PA stiffness. Implications: The results of this study are consistent with pilot data we have collected where differences in PA mobility of the cervical spine in painful and less painful locations were suggestive of both of the patterns described from the non-linear simulations. Further studies are underway to clarify the relationship between symptoms and PA mobility. Recognising the non-linear behaviour of intervertebral stiffness may result in greater accuracy in the assessment and interpretation of both manual and instrumented assessment of passive movements.