Article Text


Exercise does not ‘wear down my knee’: systematic reviews and meta-analyses
  1. Alessio Bricca
  1. Correspondence to Alessio Bricca, Research Unit for Musculoskeletal Function and Physiotherapy, Department of Sports Science and Clinical Biomechanics, University of Southern Denmark, Odense 5230, Denmark; alessio.bricca{at}

Statistics from

What did I do?

I investigated the impact of exercise on knee joint articular cartilage. This aim was investigated in three different systematic reviews (SRs) of randomised controlled trials (RCTs) in healthy animals and in humans at risk of or with knee osteoarthritis (OA).

Why did I do it?

Exercise is the first-line treatment for knee OA.1 Yet, many people still believe that exercise may ‘wear down my knee’ creating a barrier to exercise.

Articular cartilage is the hallmark of OA and the structure I studied in my PhD. It is a connective tissue that covers bone ends in the joints and provides lubrication of the meeting surfaces, allowing the transmission of loads with a low frictional coefficient. Aggrecan, collagen and molecular biomarkers are the molecules responsible for maintaining cartilage integrity, function and metabolism, and the loss of some of these components, as occur when the OA disease progresses, jeopardises cartilage health.2

Mechanical loading is essential for articular cartilage thanks to the ability of articular cartilage to convert mechanical loading into cellular response that ultimately may lead to aggrecan and collagen synthesis. However, loading magnitudes and frequencies above (ie, overload) the loading range cartilage can tolerate or below it (ie, underloading) may be detrimental or insufficient to promote cartilage health.3 Collectively, this knowledge suggests that theoretically the load generated by exercise to the knee joint, if not too high or too low, may actually be beneficial for cartilage.

How did I do it?

I performed three SRs of RCTs spanning from healthy knees to knees with OA. The first SR included healthy animals where articular cartilage was assessed via histology, while the second and third SR included people at risk of, or with, knee OA with cartilage assessed via imaging biomarkers (ie, MRI) or molecular biomarkers from joint fluids (ie, blood, urine and synovial fluid), respectively.

I followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. The literature searches were performed in at least five major databases, and the overall quality of evidence was assessed using the Grading of Recommendations Assessment, Development and Evaluation approach. The available evidence was summarised with a narrative synthesis or meta-analysis according to the Cochrane handbook.

What did I find?

Exercise is not harmful for articular cartilage in people at risk of, or with, knee OA as it did not alter articular cartilage either when assessed via MRI4 or molecular biomarkers from joint fluids.5 This suggests that the load generated on cartilage by the exercise interventions in humans was within the loading range tolerated by articular cartilage.

Similarly, a moderate dose of exercise in healthy animals (eg, dogs running 4 km daily) is not harmful, while a too high (eg, dogs running a marathon every day for a year) or too low (eg, rodents allowed only short bouts of exercise) may have a detrimental effect.6

The overall quality of evidence was downgraded to low, suggesting that future high-quality RCTs are needed to improve our confidence in these results.

What is the most important clinical impact?

Patients can be reassured that the exercise commonly prescribed to prevent or treat symptomatic knee OA seems to be safe for articular cartilage health.

Figure 1

Alessio Bricca has been playing football in Italy since he was six years old. His passion for sports and exercise has grown during his football career, first as a player and later as a physical trainer. In April 2015, he moved to Denmark as part of the KNEEMO network to investigate the effect of exercise on knee joint health, within a PhD project at the University of Southern Denmark.


The author would like to thank Professor Ewa Roos and Dr Carsten Juhl for their supervision throughout his PhD. Additional thanks go to all other coauthors including the FoF research group at University of Southern Denmark.


View Abstract


  • Funding This study is funded by European Union Seventh Framework Programme (FP7-PEOPLE-2013-ITN; KNEEMO) (607510).

  • Competing interests None declared.

  • Patient consent Not required.

  • Provenance and peer review Not commissioned; internally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.