Comparison of postdischarge physiotherapy versus usual care following primary total knee arthroplasty for osteoarthritis: an exploratory pilot randomized clinical trial

Abstract

Objective: To evaluate a pilot trial of a postdischarge physiotherapy intervention to improve patient function versus usual physiotherapy in patients undergoing total knee arthroplasty aiming to assess: recruitment rate, feasibility and acceptability of the intervention and control, suitability of outcomes, retention and adverse events and to inform sample size calculation for a definitive trial.

Design: Exploratory pilot randomized controlled trial using independent assessment.

Setting: Mixed urban and rural, UK.

Participants: Patients undergoing primary, elective unilateral knee arthroplasty for osteoarthritis.

Intervention: Two additional home physiotherapy visits of functional weight-bearing exercises, functional task-specific training versus treatment as usual.

Main outcome: Oxford Knee Score at 12 months. Secondary outcomes: completion rates, adverse events, Knee Injury and Osteoarthritis Outcome Score, leg extensor power, timed 10-m walk, timed sit-to-stand, resource use diaries. Assessments completed at baseline (pre-operatively), 3, 6 and 12 months.

Results: Of 181 eligible participants 107 (59.1%) were randomized over 13 months, one participant withdrew, no adverse events. Intervention group n = 56 (mean age 67.8), control group n = 51 (mean age 70.8). The difference in mean change of Oxford Knee Scores between groups (intervention – control) at 12 months was 0.2 (95% confidence interval (CI) –3.8, 4.2), P = 0.94. Patient diaries revealed non-trial additional physiotherapy requires improved measurement.

Conclusions: Successful recruitment and retention rates were achieved. The intervention appeared feasible and acceptable but may be suboptimal in intensity given recent research. A sample size of 1271 participants would be required for a fully powered randomized controlled trial using the main outcome. However new outcomes, potentially of greater validity and responsiveness, require consideration.