Purpose of the study: How does a total knee arthroplasty (TKA) function? Do all prostheses provide the excellent results reported in the literature? This gait analysis compared patients with a TKA versus normal subjects in order to obtain a 3D quantification of the kinematic and dynamic differences between patients with a very good functional result and controls.

Material and methods: Twenty patients who had a TKA for less than one year and whose functional outcome was scored very good (KSS knee > 85/100, VAS ≤1/10) were compared in a double blind study with 20 normal controls. The knees were masked so that the investigators were unaware of the type of subject (operated or not), the side operated, or the type of implant. The analysis as performed on an AMTI platform with six infrared cameras which followed the displacements of 36 reflectors. Motion Analysis software was applied. The gait parameters recorded were: speed, step length, flexion angle, duration of weight bearing/oscillation phases, and dynamic variables: flexion-extension moment, varus-valgus moment, internal/external rotation moment.

Results: Adjusted for age and height, step length, walking speed, and duration of the weight bearing phase were identical in the operated and control populations. Kinematic and dynamic variables demonstrated significant differences. At lift-off, all of the TKA subjects were in functional permanent flexion (m=10); the flexion moment of the quadriceps was less than in the non-operated subjects. In the frontal plane, the weight-bearing phase was identical between the operated subjects and controls, but with a varus dynamic (m=4) during the oscillating phase. In the horizontal plane, there was an external rotation of the tibia (m=+5) during weight bearing.

Discussion: Gait analysis provides quantitative information which is not perceptible at physical examination nor with videoscopic explorations. Even patients with an excellent KSS score exhibit important anomalies despite the fact that the physical exam finds a normal range of motion and normal muscle force. The degree by degree 3D gait analysis reveals the difference.

Conclusion: Despite a clinical score considered to be very good, patients with a TKA have a functional deficit of the extensor system during take-off, even when the knee has complete active extension; the weight-bearing phase of the step is in external rotation and the oscillating phase exhibits varus laxity.

This study aims to compare the gait pattern of patients operated with a TKA versus a normal population in order to evaluate if the excellent function of TKA reported in the literature corresponds to objectively measured parameters.

20 patients operated of TKA with a follow up > 1 year, all patients rated with a very good functional result (Knee Society Knee score > 85/100 – VAS < = 1/10) were compared with a group of 20 “ normal” persons.

The study was blind: the examiner did not know whether the person was a normal, or which knee was operated.

The test consisted in an 11 meters walk, on an AMTI force platform; the movements of the body were recorded with 6 IR cameras and analysed with the “Motion Analysis” software.

The implant was a mobile bearing AP stabilised knee.

The measured parameters were kinematic : speed, step length, flexion angle, duration of stance /WB phase and dynamic : flexion/extension, varus/valgus, internal/ external moments and resultant force direction. When matched with age and BMI, all kinematic parameters of the TKA group are equal to that of the normals.

However, dynamic parameters differ significantly between both groups: At the end of stance phase and heel strike the operated patients have a lack of extension of 10° despite a clinical measurement of full extension (0°) In the frontal plane, all patients exhibited a valgus walking pattern but the mechanical axis measured on long standing radiographs was 180°+/−1°. In the horizontal plane, all operated patients had an external rotation of +8° compared to the normals.

Despite excellent clinical scores and radiologic positioning, gait analysis demonstrates important dynamic differences between the TKA and the normal group. The extension lag at heel strike may be related to either quad weakness, or an insufficient extension gap at surgery; The valgus resultant pattern occurs despite a perfect alignment of the mechanical axis (180°) on the operated patients: this rises the question whether this alignment is the goal or if an undercorrection would be more physiologic. External rotation is superior to the normal group : it is in relation with the external rotation of the femoral and tibial components.

Conclusion. Gait analysis of the TKA group of patients compared to normal demonstrates important dynamic differences in relation with the surgical positioning of the implant.