Minimal change in physical activity after lower limb joint arthroplasty, but the outcome measure may be contributing to the problem

a systematic review and meta-analysis

Kathryn Mills*, Brooke Falchi, Caitlin Duckett, Justine Naylor

*Corresponding author for this work

Research output: Contribution to journalArticle

6 Citations (Scopus)


Background: The literature pertaining to changes in physical activity (PA) in people who have undergone lower limb arthroplasty is controversial, but it is unknown whether this is due to participant characteristics or how physical activity is measured. Objectives: To determine whether PA changes after total knee or hip arthroplasty and what explains contradictory results between different published studies. Data sources: Five online databases were searched for keywords and MeSH headings. Reference lists were also hand-searched. Study selection: Cohort studies and the control groups of clinical trials that examined PA levels prior to total arthroplasty as well as 6- and/or 12-month post-operative included. PA could be measured using accelerometery, pedometery or patient reported outcomes. Data extraction: PA outcomes were categorised into frequency, intensity, duration and type. Meta-analysis was performed when possible. Results: Eleven studies, examining 277 individuals with THA and 406 people with TKA were included. Studies differed in the outcome measure reported, device used and placement of accelerometers. When measured as steps per day, pooled data revealed a small increase in the frequency of PA at 12 months post-TKA (SMD 0.44 [0.2, 0.67] I2 = 0%) and moderate increase in individuals post-THA (SMD 0.65 [0.32, 0.97] I2 = 0%). Pooled evidence indicated no change in PA duration when measured as time spent active (SMD 0.05 [−0.42, 0.52] I2 = 46%) or proportion of time spent active (SMD 0.5 [−0.17, 1.16] I2 = 75%) 6-months post-TKA. Data examining PA duration, intensity and type at 12-months could not be pooled. Conclusion: The true nature of changes in PA post-surgery remains largely unclear. While frequency of PA may increase, the balance of evidence indicates that PA does not substantially increase post total joint arthroplasty. Improvements in the consistency of device type, placement and outcome measure would substantially assist in improving knowledge in this area. Trial registration: CRD42015029686.

Original languageEnglish
Pages (from-to)35-45
Number of pages11
JournalPhysiotherapy (United Kingdom)
Issue number1
Publication statusPublished - Mar 2019


  • Accelerometry
  • Exercise
  • Hip
  • Knee
  • Patient reported outcome measures
  • Replacement

Fingerprint Dive into the research topics of 'Minimal change in physical activity after lower limb joint arthroplasty, but the outcome measure may be contributing to the problem: a systematic review and meta-analysis'. Together they form a unique fingerprint.

Cite this