How Useful is Imaging in Tendinopathy

Peter Malliaras5th of February / blog / tendinopathy-updates / how-useful-is-imaging-in-tendinopathy

Dear all 

Welcome to tendinopathy blog 41 (subscribe here).

Remember to register for Erik Meira’s Hip and Knee course – early bird registration ends very soon.

The blog this week is a foray into the utility of imaging in tendinoathy. You have to be so careful when interpreting imaging findings in tendinopathy. There are two papers that I have selected that take a very different view of the importance of imaging findings. Our quest will be to try and find the truth. Is it in the middle or polarized towards one of these studies?

Value of quantitative MRI parameters in predicting and evaluating clinical outcome in conservatively treated patients with chronic midportion Achilles tendinopathy: a prospective study

Study 1 – what they did: Tsehaie et al. performed a prospective observational study. They followed 20 people with midportion Achilles tendinopathy during the course of 12 weeks conservative treatment and then for a further 12 weeks – 24 weeks in total. They were prescribed a typical Alfredson training program to perform 3x15 repetitions of knee bent and straight calf raises eccentrically 2x/day for 12 weeks. (who ever follows that program to letter…no one!). A blinded MRI trained researcher also collected MRI data on each patient, including tendon volume (Volume), tendon maximum cross-sectional area (CSA), tendon maximum anterior-posterior diameter (AP), and Signal intensity, quantified as the weighted mean T2 relaxation time (SI). They analysed the prognostic value of the MRI scans – ie whether they predicted clinical outcome.

What they found: The intra- and inter-observer reliability of the MRI outvomes was fair to excellent – volume was by far the most reliable, the remainder were fair. Victorian Institute of Sport Outcome (VISA) improved significantly over the 24 weeks – by an average of 12.3 points. None of the MRI measures were associated with change in VISA. Also, changes seen in MRI measures over the the 24 weeks were within minimal detectable changes, so could have been explained by measurement error. A key limitation is the relatively small sample size meaning the study is most likely underpowered.

Refractory patella tendinopathy with failed conservative treatment—shock wave or arthroscopy?

 Study 2 – what they did: Williams et al. performed a single-centre retrospective study. They included 40 patients who had been treated for patellar tendinopathy between 2012 and 2014. All patients had had an MRI scan and were categorized as having pathology that involved the tendon itself or the retropatellar fat pad. All patients underwent ESWT and if they did not respond they had arthroscopic debridement. They had VISA scores for all patients at baseline and then again at 6-months follow-up.


What they found: 80% of patients with tendon involvement responded to ESWT. In contrast, 100% of patients with fat pad involvement failed ESWT treatment, but had reported good or excellent outcomes after surgery. The authors then declare that based on these finding their patients only receive ESWT if they have tendon involvement, and proceed straight to surgery if they have fat pad involvement. Interesting ! 

Clinical interpretation: As promised, 2 studies that consider the use of imaging in tendinopathy for prognostic purposes. Interestingly, they reach opposite conclusions! Granted they are looking at different tendons, but this is, I believe, not the main reason for difference in findings. The big issue with the Williams study is that it is retrospective. It is very likely that bias from the authors regarding their categorization of pathology and what this means for outcome influenced the results. We have no idea what they told patients about the imaging findings. Eg ‘look you have a type of pathology that we don’t think will respond to shockwave, but what the heck, let’s try it anyway’.

Although underpowered, the first study represents much better science, i.e. it is prospective and used blinded MRI assessors. It seems in contrast though, two prior studies by Khan et al. 2003 and Archambault et al. 1998 did find a link between imaging at baseline and outcomes. found less severe pathology on MRI and US respectively, were associated with better outcomes in Achilles tendinopathy.

The key take home is that there is at best conflicting that imaging findings relate to clinical outcome.

So far we have focused on one potential use of imaging and that is predicting outcome or prognosis. There are other uses such as diagnosis and predicting onset of symptoms in people who are asymptomatic. Here is a short infographic providing what I hope is a balanced view of potential uses of imaging in tendinopathy. If I was a teacher writing a report card for my pupil ‘imaging’, it would probably be a bare minimum pass and go like this…‘imaging has their moments but these are rare and overall it has to be said is a poor contributor in my tendinopathy class and it must be said pretty dispensable’.

See you next time


Peter Malliaras
Tendinopathy Rehabilitation