Adductor canal blocks offer an alternative to femoral nerve block for postoperative pain relief in knee arthroplasty. They may reduce the risk of quadriceps weakness, allowing earlier mobilisation of patients postoperatively. However, little is known about the effect of a tourniquet on the distribution of local anaesthetic in the limb.

Ultrasound-guided adductor canal blocks were performed on both thighs of five human cadavers. Left and right thighs of each cadaver were randomised to tourniquet or no tourniquet for one hour. Iohexol radio-opaque contrast (Omnipaque 350) was substituted for the local anaesthetic for X-Ray imaging. All limbs underwent periodic flexion and extension during this hour to simulate positioning during surgery. The cadavers were refrozen. Fiducial markers were inserted into the frozen tissue. X-rays were obtained in 4 planes (AP, lateral 45° oblique/medial oblique, lateral). University Research Ethics Approval was obtained and cadavers were all pre-consented for research, imaging and photography according to the Anatomy Act (1984).

Analysis of radiographs showed contrast distribution in all thighs to be predominantly on the medial aspect of the thighs. The contrast margins were entire and well circumscribed, strongly suggesting it was largely contained within the aponeurosis of the adductor canal. Tourniquets appeared to push the contrast into a narrower and more distal spread along the length of the thigh compared to a more diffuse spread for those without. Proximal spread towards the femoral triangle was reduced in limbs without tourniquets.

The results suggest that contrast material may remain within the adductor canal structures during adductor canal blocks. Tourniquets may cause greater distribution of contrast proximally and distally in the thigh, but this does not appear to be clinically significant. Further studies might include radio-stereo photometric analysis using the fiducial markers in the limbs and in vivo studies to show the effect of haemodynamics on distribution.

Several previous studies have examined the mechanical environment in the femur using computational modelling. In particular the proximal femur has been extensively studied using finite element (FE) analyses. This study considers the issues associated with modelling with special interest in the distal femur. FE models require appropriate input on the geometry of the system being considered, material properties of different components, loading regimes and boundary conditions (i.e. the manner in which the system is supported). This study focuses on the last two of the above. A number of models with variable levels of complexity; and different boundary and loading conditions were considered. The simplest loading and boundary conditions considered comprised load applications at the tibio-femoral joint with the proximal femur artificially restrained. More complex models had the femur fully supported on muscles and ligaments. In each case the stress-strain environment in the femur was examined. The results show that the sophistication of the model needs to be based on the answers being sought from the analysis. Some good predictions on the mechanical environment can be made with relatively crude models. For example the stress-strain behaviour in the vicinity of the knee joint was found to be reasonably well predicted by the model that was artificially restrained in the mid-femoral region. Further while different models can be used for comparing different scenarios (e.g. forces during the gait cycle) true quantitative measures are strongly dependent on experimental loading data. The study also shows that it is important to generate and evaluate models of increasing complexity in order to maintain transparency with respect to the influence of different parameters associated with loading and boundary conditions.