Why is treating tendinopathy so tricky?


Tendinopathy is an exceedingly common musculoskeletal condition among active individuals.  Tendinopathy has been an increasingly studied topic of sports medicine and rehab with research actually creating more divergence in the pathophysiological mechanisms with the term “tendinopathy” itself leading insight into this history of equivocation.  Painful tendon conditions were termed “tendonitis” pre-1990s then shifting to “tendinopathy” and even “tendinosis” in light of histological studies demonstrating lack of inflammatory cells in chronic, pathological tendons.  However, advances in histological techniques have recently shown that inflammation likely does play some role in the early stages of tendinopathy while it remains generally accepted that later stage, chronic, degenerative tendinopathies are devoid of these inflammatory markers, but this is a growing topic of debate.  In light of the pathology explanation fluctuations, it is without wonder that rehabbing tendinopathy remains difficult.  Treatment options for those treating painful tendons exist with varying degrees of evidence.

Many models and theories of tendon pathophysiology exist.  The most prevalent and accepted theories are those of mechanical overload resulting in a condition of eventual disrepair and disorganization of the tendon.  These could entail possible minor fibril rupture which leads to acute fibroblast proliferation and release of factors surrounding an initial attempt of healing.  This is proposed to lead to neovascularization, the in-growth of blood vessels and nerves, when this healing process does not, for some reason, complete.  We can see these individual steps, such as the common observation of neovascular in-growth, however the actual initial step of partial rupture has not been observed with no evidence demonstrating they actually occur.  Furthermore, in most cases tendinopathies arise symptomatically in a sort of gradual manner without a clear injurious incident, which would be expected with partial rupture.  Biochemical and histological studies have also shown major differences between complete rupture and healing tendons compared to those with tendinopathy with ruptured tendons lacking the presence of common tendinopathic findings suggesting that tendon rupture/failure may not be linked to pre-existing tendinopathy.

There does appear to an acute stimulus triggering tendinopathy with new research now accumulating suggesting inflammation does, in fact, play a role in the early stages of tendinopathy and possibly in the continual progression of the condition.  Some researchers, such as Jill Cook’s group at La Trobe University, assert that this is likely more of a cellular activation than an actually inflammatory response.  This cell activation idea has lead to attempts at staging tendinopathy with the acute and painful tendon being described as the reactive stage whereby mechanical overload triggers cellular activation leading to increases in proteoglycan production.  Increased hydration of these proteoglycans are then said to increase tendon swelling prompting the disruption of collagen alignment and creating areas corresponding to hypoechoic areas prone to neovacularization.  Tendinopathy is proposed by Cook to occur in stages of reactive tendinopathy, tendon dysrepair, and finally tendon degeneration.

Another hypothesis is a sort of pain-first mechanism with the onset of tendon pain representing a change in nociception via primary ingrowth of nerves and vessels.  We do know that eventual nerve and vessel in-growth is present in late stage tendinopathy but it is difficult to determine when this occurs and the directional relationship towards tendinopathy-does tendinopathy result from this neovascular ingrowth or because of it.  It’s plausible that increases in tendon nociception could lead to mechanical offloading both with movement offloading and offloading at the tissue level.  Offloading at the level of tissue itself is an idea termed tendon stress-shielding which is another hypothesis of tendinopathy supported by evidence that chronically unloaded tendons undergo degenerative processes.

Another plausible hypothesis is that mechanical overload could exceed a sort of tendon homeostasis whereby cells responsible for tissue support and regeneration to continual tissue turnover become stressed beyond their capacity to heal the tendon.  This then may even lead to cellular apoptosis at the tendon which has been observed in large increases of load. It has also been suggested that perhaps there is an oxidative stress component in areas of the tendon which become hypoxic under inadequate oxygen supply and have help explain the mechanism of neovascularization observed in chronic tendinopathy.   Clearly other variables could be involved such as genetics and individual factors with such things as high cholesterol and adiposity being correlated with incidence of tendinopathy.

Compression of the tendon at areas of contact with bone also likely play a role especially with insertional tendinopathies whereby muscle contraction causes increased tendon compression into bone perpendicular to the length of the tendon.  A clear example is of the rotator cuff tendons which must bend or wrap around the head of the humerus at their distal attachments.  Muscle activation would cause the tendon to be firmly pressed into the bone as a sort of bow-stringing effect.  In fact, most tendinopathies are likely insertional with the exception being mid-substance achilles tendinopathy.

Adding to the nearly enigmatic pathophysiology  there also is not clear consensus regarding clinical diagnostic criteria for tendinopathy.  Typically,.tendinopathy is generally diagnosed clinically via local pain related to isometric load and tenderness to palpation.  These criteria are said by Mascia to be anecdotal and lacking specificity Furthermore, imagining modalities are typically unavailable to the rehab specialist with imaging results not consistently correlating to clinical presentation (Masci, 2016.)  Pain does not appear to correlate well to the degree of neovacularization and it is known that asymptomatic athletes can undergo pathological degeneration in the absence of pain.  As such it can be difficult to determine the degree of progression of the condition when a patient presents with tendon pain – is this apparent reactive tendon occurring in a recently otherwise healthy tendon, or was there significant degeneration that has now become symptomatic?   So while it is difficult to diagnose and define tendinopathies it does appear that tendons can undergo normalization over a extended periods of time to improvements in tissue structure as observed with ultrasonography.  So how do we, as clinicians, go about treating and rehabbing painful tendon conditions?

There are several proposed therapies for tendinopathy including plasma-rich platelet injection, high volume injection, extra-corporeal shockwave therapy, stem cell therapy, and of course loading therapies.  The best current evidence in treating tendinopathy is that of tendon loading through exercise.  This is an interesting paradox as it is typically accepted that tendon overload and lack of adaptation to load demands to be a primary driver of tendinopathy.  However, research in the area of loading initiated by Hakan Alfredson and others have shown clear benefit to progressive loading.  An unloaded tendon will regress in it’s load capacity and become prone to increased pathology.  This unloading as noted by possible tissue shielding and/or changes in movement patterns are ways a person may instinctively avoid pain by unloading, but could contribute to continued tendinopathic processes as a result.  Tendon loading provides benefits through the mechanisms of mechanotransduction whereby cells respond to physical load with biophysical and biochemical adaptation.  These adaptations occur as results of tenocyte response with changing cellular structure and composition via alteration in gene expression and protein synthesis.  It also appears that load by muscle contraction is better than stretching for maintaining and promoting load capacity in tendons, although some research exists suggesting static stretching may promote improved tendon efficiency/energy preservation as noted with decreased hysteresis (energy loss) when tendons release their stored energy from load after a stretching program (Kubo, 2002.)  In terms of muscle contraction there is more evidence, at present, for eccentric contraction providing benefit, however, evidence of benefit from concentric contraction exist as well just not in as robust of number as those supporting eccentrics.

The paradox of treating a tendinopathic tendon caused by overload with load makes clinical decisions challenging and there does not appear to be any clear consensus advice on the topic.  I tend to favor Jill Cook’s perspective of the acute pain of tendinopathy as a sort of reactive state which should be allowed to subside before controlled loading is initiated.  My general method for treating tendinopathy (which is ever-evolving) is to reverse titrate any clear offending stimulus or activity with some preservation of of tendon loading with pain-free exercise such as isometrics in positions favoring as little tendon compression as possible.  As an example, isometrics for insertional achilles tendonopathy would be avoided initially  in dorsiflexed positioning as it has been proposed this position could promote insertional compression.  When the pain has shown evidence of regression, typically after about a week, I introduce a gradual loading program.  Research at this point indicates that pain is likely to occur but should not be debilitating, such as to the point of making ambulation difficult; pain is expected, but must be endured to an extent during exercise.

Medical intervention show only tenuous benfefit with anti-inflammatory strategies, such as glucocorticoids and NSAIDs, showing inconsistent benefits but mostly in the short, initial time-frame.  Furthermore, longer duration use, or use after the initial reactive stage, may be harmful as we know that longer durations of glucocorticoids decrease tendon quality and NSAIDs impede healing of muscle and bone injury with both carrying other health risks.  PRP injections so far have not shown much benefit in larger, high quality studies, however, some argue that this has come from variance in the quality of PRP used.  Time will tell the extent of benefit of PRP injection though it does seem that more evidence is emerging dissuading their use.

The British Journal of Sports Medicine released a good guide to treating tendinopathy on their blog that you can access HERE.


1.  Kubo, K., Kanehisa, H. and Fukunaga, T. 2002. Effects of transient muscle contractions and stretching on the tendon structures in vivo. Acta Physiological Scandinavica, 175: 157-164.

2.  Masci, Lorenzo. “Is Tendinopathy Research At A Crossroads?”. British Journal of Sports Medicine 49.16 (2015): 1030-1031. Web. 2 July 2016.

3.  Rees, Jonathan D, Matthew Stride, and Alex Scott. “Tendons – Time To Revisit Inflammation”. British Journal of Sports Medicine 48.21 (2013): 1553-1557. Web. 2 July 2016.

4.  Scott, Alex, Ludvig J. Backman, and Cathy Speed. “Tendinopathy: Update On Pathophysiology”. J Orthop Sports Phys Ther 45.11 (2015): 833-841. Web. 2 July 2016.

5.  Tilley, Benjamin J et al. “Is Higher Serum Cholesterol Associated With Altered Tendon Structure Or Tendon Pain? A Systematic Review”. British Journal of Sports Medicine 49.23 (2015): 1504-1509. Web. 2 July 2016.

6.  Wang QW,Chen ZL,Piao YJ Mesenchymal stem cells differentiate into tenocytes by bone morphogenetic protein (BMP) 12 gene transfer. J Biosci Bioeng2005;100:41822.



Author: Landon Booker, PT, DPT, CSCS

I am a doctor of physical therapy and strength and conditioning specialist practicing in an orthopedic and sports medicine physical therapy clinic in Omaha, Nebraska.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s