The measurement of collateral knee laxity using non-invasive computer-assisted technology

Computer-assisted technology has provided surgeons with intra-operative quantitative measurement tools that have led to the development of soft-tissue balancing algorithms based on surgeon-applied varus-valgus stress. Unfortunately these forces tend not to be standardised and the resultant algorithms may at best be surgeon-specific. Furthermore, these techniques are only available intra-operatively and rely on the rigid fixation of trackers to bone. The aim of this study was to develop a non-invasive computer-assisted measurement technique and assess the variation in collateral knee laxity measurements between different clinicians.An image-free navigation system was adapted for non-invasive use by developing external mountings for active infrared trackers. A leg model with rigid tracker mountings was designed and manufactured for comparison. Multiple kinematic registrations of alignment were made for both the model and the right leg of a volunteer to quantify the soft tissue artefacts. Repeatability of the system was assessed by performing two registration processes on eight volunteers. Collateral knee laxity was assessed on a single volunteer by 16 participants of varying experience each applying a maximum varus and valgus knee stress. Two surgeons performed repeated examinations to assess intra-observer variation.For repeated registrations of alignment, the SD of the non-invasive mounting (0.8°) was only a third higher than the leg model (0.6°) and the actual range was only 1° larger. The repeated alignment measurements on the volunteers showed a high level of agreement with an intraclass correlation coefficient of 0.93. Varus-valgus stress values showed poor inter-observer variation with a wide range of angles for both varus (1° to 7°) and valgus stress (0.5° to 5°). A Mann-Whitney test between the two sets of repeated tests showed that both varus stress and overall laxity were significantly different (p