Background: Clinical scores are widely used to evaluate the outcome of total knee arthroplasty (TKA). However, a lack of uniformity, the use of different terminology, and the diversity of methods used to translate numerical data into clinical outcomes have been described as potential problems. Gait analysis is believed to provide more objective parameters and allow the ascertainment of functional performance after knee arthroplasty. The aim of the present study was to obtain information about the correlation between the outcome in terms of locomotion and the clinical knee score after TKA.

Methods: 29 consecutive patients waiting for total knee arthroplasty (TKA) were included in the study. The Hospital for Special Surgery Score (HSS), the Knee Society Score (KSS) and a gait analysis were conducted 1 day prior to surgery and 3 months postoperatively. The following kinematic and temporospatial gait parameters, whose relevance has been established in knee arthroplasty were analyzed: In the sagittal plane, the following variables were determined: maximum knee flexion stance, maximum knee flexion swing, minimum hip flexion (= maximum hip extension) and minimum ankle dorsiflexion (= maximum ankle plantarflexion). The maximum pelvic obliquity stance was determined for analysis in the frontal plane while stride length, double support and gait velocity were calculated for temporospatial analysis. Data from the KSS and HSS were analyzed for the subgroups named pain, knee (knee-specific parameter), function and total sum. Pearson’s correlation coefficients were calculated for the above mentioned gait parameters and for knee score subgroups pre–and postoperatively.

Results: Preoperatively, positive correlations of r > 0.5 (0.001 < p < 0.005) were ascertained for maximum knee flexion swing, maximum pelvic obliquity stance, gait velocity and stride length, and were mainly determined for the subscore of function and the total sum of KSS and HSS. A lower correlation (r = 0.388, p = 0.041) was determined for maximum knee flexion stance. Postoperatively, positive correlations of r > 0.5 (0.000 < p < 0.003) were determined for gait velocity, maximum pelvic obliquity stance and stride length, mainly for the subscore of function and the total sum of KSS and HSS. A negative correlation of r < −0.5 (0.001 < p < 0.009) between these score subgroups and double support was only ascertained postoperatively. No correlations were registered between pain subscores of KSS or HSS and any of the gait variables.

Interpretation: In the current study we established high correlations particularly between temporospatial parameters and functional and total scores of KSS and HSS pre-and postoperatively. It is concluded that the functional subscores of KSS and HSS are particularly suitable to assess the dynamic outcome of TKA.