Aims

Whilst short and mid-term activity levels after a Total Knee Replacement (TKR) have been reported in the literature; there is an over simplification of the reporting and longer-term activity levels are unknown. The aim of this study was to map the long-term trajectories of patients’ physical activity levels postoperatively to identify meaningful subgroups and explore associations with preoperative variables.

Methods

This was a secondary analysis of a single centre longitudinal cohortstudy using group-based trajectory modelling (GBTM) of the University of California, Los Angeles (UCLA) physical activity score over ten years.

Multinomial logistic regression models (both adjusted and unadjusted) were used to test associations between preoperative variables and trajectory group membership.

Results

266 of the 904 eligible patients were recruited (29%). Data from 260 patients was available for analysis. Four trajectory groups were identified with good fit of the model (average posterior probability 0.79 to 0.93). Of the four groups, the two more active groups had a peak activity level between two-three years postoperatively; the less active groups had a peak activity level at between three months and one year. Preoperative UCLA, participation in sports in the three years prior to surgery and male gender showed trends towards association with a higher activity group.

Conclusions

Four trajectory groups were identified giving a more detailed understanding of temporal trends in physical activity levels post TKR. There was weak evidence to show an association between patient expectation, preoperative UCLA score and participation in sports in the prior three years and group membership.