View Full Version : Functional methods for the ankle complex

Steve Piazza
07-11-2007, 07:46 AM
Thanks to everyone for this very interesting discussion.

Our laboratory has had similar difficulties in implementing kinematic fitting techniques to accurately locate the axes of the ankle joint complex. One important point is that the fitted joint axes or centers should be compared to their "true" counterparts (such as the radiographically-determined femoral head center in the case of the hip joint), rather than to previously-reported average axes. This is especially true in the case of the talocrural and subtalar joint axes, which are known to vary widely across subjects. Finding an axis that lies within the range of previously-reported subtalar axis orientations (such as the range reported by Inman) is not sufficient, because the range is so large.

In our own tests (Lewis et al., ASME J Biomechanical Eng 128:596-603, 2006), we compared the axes found during a simultaneous fitting of the two ankle axes to the axes found using helical axis decomposition of tibia-talus and talus-calcaneus motions. We studied cadaver specimens with markers mounted on bone pins, so skin movement was not an issue. We still saw errors in axis orientation that were between 20 and 40 degrees, errors that did not result when we tested an anthropomorphic mechanical linkage. This suggested to us that the cadaver joints exhibited deviation from their mechanical analogues that made the problem ill-posed. Looking into the reasons for this ill-posedness, we ran computer simulations that showed us that, as Mike Schwartz suggested earlier in this discussion, nearly-identical marker trajectories can produce very different ankle-complex axes following kinematic fitting of a two-revolute model.

Steve Piazza
Greg Lewis
Penn State University