Aim: Accurate soft tissue balance in total knee arthroplasty (TKA) is not only technically challenging but also difficult to teach to trainees; we believe that computer navigation provides a very useful tool for objective and reproducible soft tissue balance.

Methods: We studied 52 patients (31 females and 21 males) with knee osteoarthritis and recorded the change of the Medial (MCL) and Lateral Collateral Ligament (LCL) length at full extension and at 90o flexion. Pre- and post-operative results were compared. The assessment was performed by consultant orthopaedic surgeons using trackers and navigation knee replacement software. Data was analysed using the student t-test

Results: The navigation software programme was used to measure the change of the collateral ligament length. Ligament laxity is represented by a negative number and a positive number is used to represent stretching and apparent elongation of the ligament.

The medial collateral (MCL) length at full extension ranged from −9mm to 11mm and post-operatively was reduced to −16mm and 8mm, (p=0.042). At 90o flexion the length ranged from −3mm to 9mm and postoperatively was reduced to −8mm and 10mm (p=0.025).

The lateral collateral (LCL) length at full extension changed from −10mm to 9mm pre-operatively to −13mm and 6mm post-operatively (p=0.011). At 90o flexion the range from −8mm and 9mm pre-operatively changed to − 5mm and 11mm post-operatively (p=0.005).

All the above changes correspond to improvement in the post-operative axial alignment.

Conclusion: Our results demonstrate that computer navigation provides a useful adjunct to the accurate and reproducible soft tissue balance in knee arthroplasty which can be used to evaluate results and for training purposes.

Aim: The difficulty in accurately assessing coronal alignment of a total knee prosthesis (TKR) is widely accepted in the literature yet standard practice in the UK is to obtain AP and lateral knee views only; we compared standard AP knee films with long leg views of TKR in order to determine the most optimal way of assessment of the prosthetic knee alignment.

Methods: We included all patients who underwent TKR between January and September 2005 at Kings College Hospital under the care of one orthopaedic consultant. We excluded 11 patients with revision surgery, augmented prosthesis, high tibial osteotomies or severe tibiotalar joint arthritis.

We included 50 sets of radiographs from 48 patients (17 men and 31 women). The prostheses used were PFC (40) and Scorpio (10) and six of them were navigated and 44 were standard TKR.

We compared the difference between the angle of the tibial component with the mechanical axis of the tibia in the long leg image and the angle of the prosthesis with the midline of the visualised tibia in a standard antero-posterior knee view. Statistical analysis was carried out using the student t-test.

Results: The mean difference between the two views was 5.34o (range 1.9o – 12o) (p< 0.001). We did not find any difference between the Scorpio and PFC knees or between navigated and non navigated prostheses.

Conclusion:We concluded that the long leg views compared with the standard antero-posterior knee views provide more accurate information on the position and alignment of the tibial component of a TKR.

Aims: Kinematics of the arthritic knee joint is to date not very well understood, yet this is a significant parameter affecting the results of knee arthroplasties; we studied the axial rotation of the tibia during knee flexion in osteoarthritic knees in order to understand better the kinematics of the arthritic joint.

Methods: Tibial rotation and the screw home mechanism were studied in 55 consecutive patients (31 females and 24 males) with diagnosed knee OA. The assessment was performed by consultant orthopaedic surgeons using the trackers and the software of a navigation system, prior to any soft tissue release. The Student t-test was used for the statistical analysis.

Results: We identified 3 different patterns of tibial rotation during knee flexion.

26 knees had normal tibial rotation pattern with the tibia rotating internally during knee flexion (mean rotation: 15.5°).

We also recorded that most of the tibial rotation occurs in the first 0–30° of flexion (70%) p< 0.001.

Conclusion: The screw home mechanism and the normal tibial rotation upon knee flexion were absent or distorted in the majority of osteoarthritic knees. We found three distinctive patterns of the tibial rotation (normal, erratic and reversed) during knee flexion.