A pain in the butt! Biomechanics and running retraining for proximal hamstring tendinopathy

It is my great pleasure to put forward this piece from our guest blogger Tom Goom. I was fortunate to meet Tom at my recent running biomechanics course in London where some of the pretty pictures below of his running action come from. I can also vouch for his ridiculously tight hip flexors you can see below – I assessed this for myself! After a couple of beers with Tom after the course I realised not only is he an excellent physiotherapist but his research knowledge and how this applies to clinical practice is exceptional. I look forward to attending one of his running courses in the future – I just have to time my next trip to London. Follow Tom on twitter via @tomgoom

Over to you Tom …..

Proximal Hamstring Tendinopathy (PHT) is a common cause of deep buttock pain amongst distance runners. It can be complex and challenging to treat and, to date, there have been no randomised controlled trials investigating its rehab. Running is often a significant aggravating factor in PHT, finding ways to modify running to reduce symptom aggravation could prove a very valuable treatment.

In recent years running gait retraining has gained momentum as a treatment option for a number of running related injuries. Studies evaluating biomechanical changes have shown that even subtle changes in step rate, stride length or step width can change kinematics and modify loading. While this appears to have potential to help a variety of common running injuries, studies demonstrating changes in pain and function in symptomatic individuals are limited. Promising results have been seen in treatment of patellofemoral pain (Willy et al. 2012, Noehren et al. 2011) but to date there are no studies of gait retraining in tendinopathy.

Before we can suggest gait changes for PHT we have to consider what loading characteristics are likely to aggravate it. Tendon pain is usually aggravated by activities that require the tendon to store and release energy, such as the Stretch-Shortening-Cycle (SSC) in running. Insertional tendinopathy, like PHT, involves localized tendon pathology at the enthesis and compressive load is considered a key factor. In PHT, activities where the hip is flexed or the hamstring is stretched are thought to cause the tendon to be compressed against the ischial tuberosity. A combination of compressive and tensile load on a tendon, as may occur during uphill running when the hip is more flexed, are thought to be especially detrimental to the tendon (Docking et al. 2013). Our first aim then is to identify running kinematics that might be associated with this provocative loading.

Stride length, knee extension at initial contact, trunk position and pelvic tilt are all likely to contribute to hamstring load and will interact considerably. At the end of the swing phase the combination of hip flexion and knee extension induces a substantial elongation stress on the hamstrings (Guex and Millet 2013). It is at terminal swing where hamstring activity is at its highest during running. Due to their origin, insertions and biarticular nature, elongation stress can also be increased by trunk flexion and/ or anterior pelvic tilt. A significant increase in elongation stress is likely to place both compressive and tensile load on the tendon origin, especially if the hip is flexed.

It’s important to consider how the hamstring functions during running. It displays two peaks in activity; terminal swing where it slows knee extension in preparation for foot strike, and shortly after foot strike during the loading response. The initial peak is highest with semimebranonsus peak muscle force being equal to approximately 1.5 time body weight (Lenhart et al. 2014). The second peak is thought to represent the hamstring and quadriceps co-contraction during loading where the quadriceps take the majority of the load.

So, theory aside, what does this mean for gait analysis and retraining? Firstly it’s important to assess stride length and look for evidence of over-striding. Secondly look at hip and knee position at terminal swing and initial contact. Is there a combination of hip flexion and knee extension which will place significant load on the hamstrings? Thirdly assess trunk and pelvic position. Addition of trunk flexion and anterior pelvic tilt to hip flexion and knee extension will potentially increase compressive and tensile load.

I have a history of right sided PHT. It’s resolved now but you can see how my running gait may have contributed. I recently attended Christian’s excellent Running Biomechanics Course and he recorded my gait;

The treadmill obscures the view but you can see initial contact occurs a long way from my centre of mass – I have a significant over-stride. I combine this with trunk flexion and anterior pelvic tilt. Note too my lack of hip extension on the left. This is an important point because it highlights the importance of considering the effects of strength, movement control and flexbility on gait characteristics. In my case my hip flexors prevent any true hip extension.






The impact of this lack of hip extension is that as my left hip reaches its limit of extension my tight hip flexors then anteriorly tilt my pelvis. This coincides with initial contact of the right foot and contributes to hamstring load. It’s a good example of how swing phase on one leg can interact with stance phase on the other – often worth considering this in gait analysis.

To reduce provocative load on my hamstring tendon the first priority would be to reduce the over-stride. A simple cue for this is “take shorter, faster strides” or step rate can be manipulated using a metronome (a higher step rate at the same speed reduces stride length). This is likely to have several potentially beneficial effects. Initial contact will occur closer to the centre of mass, with a more flexed knee and reduced load on the hip and knee joints (Heiderscheit et al. 2011). Activity of gluteus medius and maximus will increase at terminal swing (Chumanov et al. 2012). It may be that the peak force on the hamstring will actually increase (Lenhart et al. 2014) but the hamstring is likely to be active in a range where there is less tendon compression due to altered hip and knee position.

In addition I could try to extend my trunk into a less flexed position or tighten my glutes to achieve a more neutral pelvis. In combination with stride length changes this is likely to reduce provocative load on the hamstring tendon, although it should be noted this has not been demonstrated in any research.

The result (left) is a more upright gait with reduced stride length and a more neutral pelvis. Although I have a history of PHT I no longer have symptoms. It would have been interesting to see my symptom response had I made these changes when pain was more of a feature.

While gait retraining sounds promising it shouldn’t be the only intervention for PHT. Initial treatment may need to focus on reducing pain with isometric loading in non-provocative positions (i.e. with minimal hip flexion). It’s important to modify loading habits that lead to tendon compression, such as sitting, or leaning forward for prolonged periods with the legs straight (an activity I do a lot as a physio working over a treatment couch all day).

When symptoms have settled isotonic loading can be added (with careful monitoring of symptom response) to restore hamstring strength and improve load capacity. Initially this should be performed in positions with minimal hip flexion then progressed into hip flexion range if tolerated. In addition the strength, control and flexibility throughout the kinetic chain should be addressed. Finally power, plyometric and sport specific exercise can be added.

While research in PHT rehab is sparse we are currently examining clinical management of PHT and are hoping to publish a paper on it in the near future. I’m very fortunate to be working with some excellent researchers on this, including Peter Malliaras from Complete Sports Care.

In summary: Multiple gait changes may contribute to loading and symptoms in PHT although we have limited guidance from the research on how to change this. Available evidence from biomechanical studies suggest that stride length and trunk and pelvic position may be key. Gait changes should always be based on individual assessment and other factors may also be implicated. In addition consider the patient’s strength, control, flexibility and biomechanics and how they impact gait, pain and pathology.