Abstract
Summary
Physical activity monitoring using a single accelerometer works reliably in clinical practice and is of added value as clinical outcome tool, as it provides objective and more precise information about a patient's activity compared to currently used questionnaires.
Introduction
Standard clinical outcome tools do not comply with the new generation of patients who are younger and more active. To capture the high functional demands of these patients, current outcome scales have been optimised (e.g. New-Knee Society Score: New-KSS), new outcome scales have been developed (e.g. Knee disability and Osteoarthritis Outcome score: KOOS). Also objective measurement tools (e.g. activity monitors) have become increasingly popular. This study evaluates the pre- and postoperative TKA status of patients using such optimised and new outcome tools.
Patients and Methods
Physical activity of 18 preoperative (68 ± 6yrs) and 16 postoperative (72 ± 7yrs, follow up range 9–10 yrs) TKA patients was assessed using a most recommended patient reported questionnaire, SQUASH (high value=more active) and a body-fixed 3D-accelerometer based activity monitor (AM). The AM was worn for 4–7 successive days at the lateral side of the nonaffected upper leg. Activity parameters (e.g. # steps, # transfers, # walking bouts of short/long duration, cadence) were derived using validated algorithms. Function was measured using patient reported questionnaires: KOOS-PS (range 0–100=worse) and New-KSS (higher score=better). Independent t-test, Mann-Whitney test and Pearson's r were used to compare groups and to investigate correlations (p<0.05).
Results
All AMs and questionnaires were returned (response rate 100%) and showed similar or higher completion rates for the AM (100%) than questionnaires (range 82–100%). At 10yrs follow-up, the function of TKA patients was significantly improved with regards to preoperative showing lower KOOS-PS and higher New-KSS subscores. Also physical activity was higher at 10yrs showing significantly more steps/day (+39%). Other activity parameters like walking bouts and transfers were also higher in the postoperative group (resp.8% and 21%), but not significantly. Most walking bouts were short taking 10–30s (pre: 63%, post: 59% of the bouts) and consisting of 10–15 steps (pre: 78%, post: 75% of the bouts). The only correlation found between activity and functional outcomes was between AM data (amount steps, sitting, walking time) and the New-KSS Function walking & standing subscale (r-range 0.37–0.54). No correlation was found between AM data and SQUASH. Moderate to high correlations were found between functional outcomes (KOOS-PS vs. New-KSS, r-range −0.56 – −0.81).
Discussion/Conclusion
At 10yr follow up, TKA patients continue to have higher functional scores and also maintain higher activity levels than preoperative, as is mainly shown in steps/day. The fact that only the New-KSS Function Walking & Standing subscale correlated with AM data indicates that function and activity are two widely independent outcome dimensions. This suggests that patients are active largely independent of their functional limitations (e.g. high activity, worse KOOS). The lack of correlation between objectively measured and patient reported physical activity indicates patients are less reliable in estimating their actual activity. The correlations between functional outcomes indicate redundancy. AM meets and exceeds response and completion rates of questionnaires. AM seems to be more objective, precise and sensitive to measure physical activity than questionnaires and adds a largely independent outcome dimension to clinical assessments.
