Prediction of Running-Induced Achilles Tendinopathy

Ebonie Rio is a world-renowned tendon expert, and part of the Physio Network team. She and other experts analyse and summarise the latest and most clinically relevant research for you. This blog was originally one of Ebonie’s research reviews, but we decided to make it freely available to all. Enjoy!

I want to discuss the some of the highlights and challenges outlined in ‘Prediction of running-induced Achilles tendinopathy with pain sensitivity – a 1-year prospective study’ (Brund et al 2018) and provide a guide as to how you might use the findings in this paper in your clinical practice.

Background and Objectives

Injury risk prediction is often touted as a panacea for sports medicine. It is tantalising to think we may be able to predict those at greater risk of sustaining an injury, and particularly an injury such as Achilles tendinopathy which is often persistent and affects training volume and performance. This study aimed to do just that – to predict Achilles tendinopathy injury in runners – by following running athletes for one year and examining the local and widespread mechanical sensitivity to stimuli.

The study aim was to investigate whether mechanical pain sensitivity in the Achilles tendon could predict the development of Achilles tendinopathy (the clinical presentation of Achilles pain) in recreational male runners.  The researchers hypothesised that those with the highest pain sensitivity (pre-selected cut off was 441kPa) would be more prone to develop Achilles tendinopathy.

Methods

99 male athletes (aged 18-60 years) were followed for the study and full data was available for 98.  Athletes had to have been running at least twice per week for two years with no injuries in the preceding three months. This study was part of a larger trial so there were some other exclusion criteria around footwear and treadmill experience.

At baseline, body mass index and history of previous injuries were collected. Pressure pain thresholds (mechanical sensitivity) were collected over each Achilles tendon (reported as mid-portion, 2-3cm proximal to insertion) and the infraspinatus muscle (as the proxy for widespread sensitivity, as it is in a location distant to the Achilles).

Although the title and abstract read as though this was observational, it was part of a bigger study so runners were provided footwear (conventional neutral shoe with a 12mm heel to toe drop).  They had to run in these for at least 10km per week for the first 500km of the year with no other guidelines around pace or frequency. It is worth defining a few pain terms, so we are all on the same page.

Pain terms

Mechanical sensitivity is how sensitive you are to a mechanical stimulus (such as pressure), and can be tested using a device termed an algometer (little gadget with a small rubber probe where you apply pressure over the skin).  The measurement is described as a pressure pain threshold, that is the amount of pressure (threshold) that can be applied to the participant before they report pain.

Endogenous pain inhibition may be thought of as your own descending inhibitory process that turns down a nociceptive signal and can occur at multiple spinal and supra-spinal levels. It is a natural phenomenon that appears to be disrupted in some pain states. Remember noxious stimuli or a nociceptive stimulus is an input – it is the brain that decides whether we need protection based on lots of factors. See Explain Pain supercharged for a thorough and completely inspirational review of all things pain; another great resource is https://bodyinmind.org/.

The authors clearly defined injury for the reader – the absence from running for a minimum of one week due to a musculoskeletal complaint in the lower extremity or the spine caused by running.  Runners completed a weekly survey.  Anyone that reported an injury was seen by a sports physician or sports physiotherapist and runners had Achilles imaging at this point if an injury was suspected.  Baseline ultrasounds were not recorded (understandably – imaging adds cost to studies).

Running data were obtained from GPS and recall when required – a fantastic feat in this many people.  GPS data enables accurate tracking of load – this is important to both understand the generalisability to patients in front of you, as well as to track internal loads. I applaud the consideration of pain science in a tendinopathy study, and more novel studies like this are needed to understand how and why the pain of tendinopathy occurs.

Limitations / things to keep in mind

The challenge here is that we don’t know the prevalence of pathology in this cohort prior to injury.  In runners, tendon pathology may be present (strongly linked with loading and running injury history – up to 36% of high-volume runners have asymptomatic tendon pathology (Kevin Lieberthal, SMA conference presentation). This means pathology can be present with or without pain and was likely present prior to any onset of pain. Pathology may affect mechanical sensitivity too. This could inflate the differences between a painful group and a ‘control group’ if asymptomatic pathology was included in the control group (Rio, 2018). Also, one of the biggest risk factors for injury is previous injury.  Previous injuries are reported, but the number of people with previous Achilles tendinopathy isn’t reported.

Furthermore, there is variation in exactly how pressure pain thresholds (PPT) are measured between studies because of the instructions given to participants, so it can be difficult to compare between studies. Some studies ask the participant to tell the researcher when it becomes painful, or reaches a certain level of pain, or when a feeling of pressure turns to discomfort.  The instructions can impact the outcome.

The data obtained from mechanical sensitivity testing (using the pressure algometer) was divided into high and low sensitivity (441kPa) after PPT measurements were completed at baseline. This statistical method of using a cut-point is commonly used in research, however it is not without problems. It was elegantly described at the recent Sports Medicine Australia conference 2018 by ‘Data Dave’ (ok that’s not his real name – it’s David Carey). My understanding of this is that this method risks missing a finding that may be there, and possibly other statistical analysis may be more appropriate. It creates a dichotomous outcome (in this case either high or low) based on a continuous measure such as PPT, i.e. someone with a PPT of 442 is in a different group to someone with a score of 440 but could be in the same group as someone with a much higher number.  Also, the cut off is based on this group – so this cannot be applied to the next group of people as their average may be different – like body temperature there is some normal variability in PPT.

Runners have two feet and this provides huge challenges for Achilles tendon research! The Brund study provides data for each leg so some people had one leg high sensitivity and one leg low. The challenge with splitting into 2 legs is that the 2 legs have the same human being attached to it – meaning the same external factors may be affecting legs in different groups – this becomes tricky for researchers to deal with during analysis.

Results / what did they find?

10 people developed Achilles tendinopathy during the year (of which 2 were bilateral i.e.12 ‘legs’ were injured). 7 legs were in the high sensitivity group while 5 were in the low group. They found no difference between groups when measured at any of the timepoints.

Clinical implications / closing thoughts

Pain is the complex interaction of biology and context (biopsychosocial). Lots of people are doing research in those with pain and looking at sensitivity – this research is coming at the injury from a completely different perspective. Achilles tendinopathy, frequently defined as pain and dysfunction of the Achilles tendon, is commonly seen in relation to a change, such as a change in load. The authors hypothesised that the footwear may have increased the load on the Achilles tendon and plantar fascia (though injury rates were low – incidence of 10/98).  An externally imposed load may be ok for some people but exceed others’ capacity. Conversely if a slow progressive load was applied it is feasible that the injury rate may be same or less than this (because tendons, muscles and bones do love to adapt), so that strategy may not help answer this particular question.  This was a pragmatic study and the authors recommend future studies controlling some of the running variables. Many of you that work with runners will know it can be difficult to control runners loading!

Based on this study, we cannot use baseline mechanical sensitivity to predict who will sustain an Achilles tendon injury in male runners. It is fantastic to see novel approaches to a clinical challenge and I hope you found this review interesting!

Acknowledgments: Sincere thanks to Mick Girdwood and Meg Hawke for their assistance!