Is joint instability in knee osteoarthritis mediated by proprioception and/or somatosensation?

Perceived knee instability is a common problem in knee osteoarthritis (OA), affecting approx. 70% of patients.[1] Since knee instability is associated with greater pain, poor physical functioning and life quality a better understanding of the multifactorial nature of joint instability in knee (OA) is desirable.

Several factors have been suggested to contribute to knee instability, including pain, laxity, chronicity, muscle strength, joint proprioception, and somatosensation (hyperalgesia, hypoesthesia). While hyperalgesia (ie increased sensitivity to a painful stimulus) has been linked to poor joint function in patients with knee (OA),[2] the influence of hypoesthesia (ie decreased sensitivity to mechanical stimuli and vibration) on joint function remains unexplained.

Kavchak et al[3] recently examined the relationship between somatosensation, perceived pain and instability during a step-up-and-over task in participants with and without severe knee (OA). Perceived pain and instability were assessed using the visual analogue scale (VAS) and through categorical rating to the query “To what degree does giving way, buckling, or shifting of the knee affect your ability to perform the task?”. Thresholds for mechanical detection (MDT), vibration detection (VDT) and pressure pain (PPP) were assessed before the task.

Compared to controls, participants with knee (OA) showed concurrent signs of hyperalgesia (decreased PPT at most painful and remote site) as well hypoesthesia (increased MDT, VDT at the knee). Hyperalgesia (PPT) was not associated with instability. However, the main finding of this study is that the strongest relationship was found between the perception of instability during the task and increased hypoesthesia to vibration, i.e. the greater the numbness, the greater the instability.

How does hypoesthesia influence knee stability? The hypothesis suggested by the authors is that hypoesthesia reflects impaired afferent input from joint and cutaneous mechanoreceptors which, in turn, might affect knee stability. The impaired processing of proprioceptive information is suggested to be a result of joint pain, evidenced by the lower sensitivity to pressure pain. This hypothesis is, according to Kavchak et al, partly supported by the lack of association between PPT and instability, reflecting that mechanical pain (PPT) is predominantly mediated by deep tissue nociceptors which do not rely proprioceptive information.

However, as the authors acknowledge, proprioception (eg joint position sense, kinesthesia) was not assessed. Furthermore, quadriceps strength, which was estimated using manual muscle testing only, might have contributed to knee instability as well. Indeed, the results of the Amsterdam osteoarthritis cohort show that lower muscle strength was significantly associated with self-reported knee instability, whereas proprioceptive accuracy (motion sense) was not.[1]

Another question that remains concerns the role of vibration sense in knee proprioception. In patients with Ehlers-Danlos syndrome, a connective tissue disorder that makes patients more vulnerable to musculoskeletal problems (particularly to joint instability), no significant correlation was found between proprioception (joint position sense) and VDT at the knee joint.[4] However, the relationship between proprioception and VDT has not yet been investigated in knee osteoarthristis. Future studies need to investigate the influence of somatosensation, propriception (eg joint position sense, kinesthesia) and objective measures of lower extremity muscle strength (eg isokinetic dynamometrie) on knee stability.

About Markus Hübscher

Markus is a postdoctoral fellow in the Arthritis & Musculoskeletal Research Group, University of Sydney. Before coming to Australia in 2011 he worked for several years as a lecturer and researcher in the Department of Sports Medicine, University of Frankfurt, Germany, where he received his Master’s and Ph.D. degrees in Sports Science.

The main focus of Markus’ research is the prevention and management of musculoskeletal injuries and disorders. His postdoc research at Sydney University focuses primarily on exploring the role of sensory function, psychological distress and physical activity in the development of chronic pain and disability in patients with spinal pain. With the team at NeuRA, he is involved in a study that aims at investigating whether exercise-induced muscle pain and movement restriction cause sensory and perceptual changes. Besides his academic work, he has gained extensive practical experience in the field of exercise therapy, having worked for 10 years with musculoskeletal patients from many different populations in outpatient rehabilitation centres.


[1] Knoop J, van der Leeden M, van der Esch M, Thorstensson CA, Gerritsen M, Voorneman RE, Lems WF, Roorda LD, Dekker J, & Steultjens MP (2012). Association of lower muscle strength with self-reported knee instability in osteoarthritis of the knee: results from the Amsterdam Osteoarthritis cohort. Arthritis Care Res, 64 (1), 38-45 PMID: 22213723

[2] Imamura M, Imamura ST, Kaziyama HH, Targino RA, Hsing WT, de Souza LP, Cutait MM, Fregni F, & Camanho GL (2008). Impact of nervous system hyperalgesia on pain, disability, and quality of life in patients with knee osteoarthritis: a controlled analysis. Arthritis Rheum. 59 (10), 1424-31 PMID: 18821657

[3] Kavchak AJ, Fernández-de-Las-Peñas C, Rubin LH, Arendt-Nielsen L, Chmell SJ, Durr RK, & Courtney CA (2012). Association between altered somatosensation, pain, and knee stability in patients with severe knee osteoarthrosis. Clin J Pain, 28 (7), 589-94 PMID: 22146110

[4] Rombaut L, De Paepe A, Malfait F, Cools A, & Calders P (2010). Joint position sense and vibratory perception sense in patients with Ehlers-Danlos syndrome type III (hypermobility type). Clin Rheumatol, 29 (3), 289-95 PMID: 19937459


  1. Great research and discussion. The strong correlation with hypoesthesia (to vibration) and instability was fascinating. Proprioception wasn’t investigated – that would be really helpful. I had a couple questions that I’m hoping someone can answer. What is the role of neurogenic inflammation in OA and consequent wasting of local stabilizers (ie VMO with knee) ? How do you infer that ‘edema’ is from overuse/aggravation or it’s a product of C fibre mediated substance P and CGRP release ? I’m always amazed at how quickly VMO wasting occurs and it usually takes faith (graded aerobic exercise with graded strengthening), customized light compression (to assist lymphatic system) and tactile stimulation (taping too) to get it back. Hope to hear back. Thanks.

  2. Excellent article Markus, thanks! Another consideration in knee instability due to OA would be abberant arthrokinematics that may change the relationship of glide:roll during motion. Any studies in this and/or the effect of manual therapy and exercise on arthrokinematics?

  3. Markus, good article thanks! On the line ‘Thresholds for mechanical detection (MDT), vibration detection (VDT) and pressure pain (PPP) were assessed before the task.’ should the PPP be PPT instead?

    Markus Reply:

    Hi Mark, man thanks for the hint. You are right, PPP should be PPT (pressure pain threshold). Markus

  4. Peter Thomas says:

    This study brings up some points to consider about how to rehabilitate patients in this population. I’ve had good results with taping of knees, feet and hips for knee OA patients. Assuming that the joint cannot give the correct afferent input, joint stabilization and control would be compromised, it is operating on inaccurate information. Decreased reaction time to perturbation and position change in the knee would compromise stability. Gaining additional input through the skin by taping or a tight fitting sleeve could help the body control the knee by supplying position sense information which the joint can no longer provide, allowing the active/dynamic restraints to better hold whatever is left of the joint together. The taping i do for it gives no real mechanical support, so im assuming it just helps improve the ability of whatever strength they have to stabilize, which can be reflected by subjective improvement in stability before resistance exercise has begun.

    Has anyone seen studies on what happens to mechanoceptors in degenerated joints? I’ve always been curious about what happens to them as OA sets in, and if using additional input from the skin could improve position sense and stability during rehabilitation

    Markus Reply:

    Hi Peter, thanks for your comment and interest. There are some animal studies that show functional changes of primary afferent nerves in OA. Wu & Henry (Mol Pain 2010,6:37) for instance performed electrophysiological recordings in a rat model of OA pain and found impaired action potential genesis in mechanoreceptors (especially muscle spindle neurons). The question whether sensory loss resulting from mechanoreceptor damage can be compensated by additional input from the skin remains open. Furthermore, the possible mechanisms of taping need further investigation. However, a recent pilot study concluded that patellar taping modulates brain activity in several areas of the brain during a proprioception knee movement task (Callaghan et al., Phys Ther. 2012;92(6):821-30.)