Effects of total knee arthroplasty on symptoms, function and activity over 5 years in knee osteoarthritis: A propensity-score matched study

Abstract

Purpose: To evaluate the effects of total knee arthroplasty (TKA) on symptoms, function and activity over 5 years in knee osteoarthritis (KOA) patients.

Methods: Data were from the Osteoarthritis Initiative (OAI). Participants who conducted the first TKA from (not before) enrolment to 48 months were propensity score matching (PSM) on their characteristics at the visit before surgery (treated as baseline) to those who did not conduct a TKA at 48 months (treated as baseline). Changes in knee pain and functional disability were evaluated using the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) pain and function subscales. Changes in physical activity, physical function and overall physical health and mental health were assessed using the Physical Activity Scale for the Elderly, the 20-m walk speed and the five times chair-to-stand tests and the 12-Item Short Form Survey, respectively.

Results: Eighty-two pairs of participants in the TKA (56% women, mean 64.8 ± 8.4 years) and non-TKA groups were matched. Knee symptoms were significantly improved in the TKA group and flatted from 24 months (Pain: β = -3.29, 95% confidence interval [CI] = [-4.59 to -1.99], p < 0.001; Function: β = -10.12, 95% CI = [-14.21 to -6.03], p < 0.001). Physical function and overall physical health but not physical activity or mental health (PASE: β = 5.72, 95% CI = [-15.46 to 26.90], p = 0.597; Mental: β = 0.04, 95% CI = [-2.47 to 2.54], p = 0.976) was improved in the TKA group over 24 months.

Conclusions: TKA substantially improved knee symptoms and physical function over 60 months and physical health over 48 months, compared to those who had a similar severity of KOA but did not have a TKA, but this did not translate into increased physical activity or mental health.

Level of evidence: Level III.

Keywords: osteoarthritis; pain; physical activity; physical function; total knee replacement.

Figure 1

Subject selection from OAI database. n, number of participants included in the analysis; OAI, Osteoarthritis Initiative; TKA, total knee arthroplasty.

Figure 2

Demonstrate matching of TKA and non‐TKA groups in the OAI cohort. Participants in the TKA group had their first TKA from the cohort entry (not before) to 48 months, and the baseline for this group was the clinical visit before the TKA was performed. The propensity score matching control group was selected from participants who did not conduct a TKA at or prior to the 48‐month visit. OAI, Osteoarthritis Initiative; TKA, total knee arthroplasty.

Figure 3

Change in (a) the Western Ontario and McMaster University Index (WOMAC) pain and (b) function scores from baseline in the TKA and non‐TKA groups over 60 months. *Positive scores indicate worsening and negative improvement from baseline. Vertical bars indicate 95% CIs for the mean scores. CI, confidence interval; TKA, total knee arthroplasty.

Figure 4

Comparison between the TKA and non‐TKA groups on change in secondary outcome. (a) The change of SF‐12 Mental, Positive scores indicate improvement and negative indicate deterioration from baseline. (b) The change of SF‐12 Physical, Positive scores indicate improvement and negative indicate deterioration from baseline. (c) The change of PASE, Positive scores indicate improvement and negative indicate deterioration from baseline. (d) The change of 20‐m walk speed, Positive scores indicate improvement and negative indicate deterioration from baseline. (e) the change of five times chair stands, timed, Positive scores indicate worsening and negative improvement from baseline. Vertical bars indicate 95% CIs for the mean scores. 95% CI, 95% confidence interval; PASE, Physical Activity Scale for the Elderly; SF‐12, Short Form 12‐Item Health Survey; TKA, total knee arthroplasty.