Gluteus Medius – evidence based rehab

This is in part advice for runners and in part something else….

…you’ve heard of a “dance off” and probably a “sing off” this my friends is a geek off!

It’s come about after I’ve heard lots of people being advised not to bother with sidelying exercises for glutes because they “aren’t functional”. Several well-meaning websites even describe them as harmful and more likely to cause “regression” than rehab. I say in my home page for this site that I appreciate everyone’s opinion and I do. I also agree that sidelying work isn’t functional, but I feel it still has a valuable place in glutes rehab and it shouldn’t be treated as ineffective quackery! My point here isn’t that sidelying work is the best, but simply that it has it’s value in rehabbing glutes and should be used as part of a comprehensive programme including weight bearing exercises.

So let’s examine some glutes exercises you can do and some research behind them. All pictures are reproduced from research articles freely available on the web and will be referenced and linked.

Sidelying abduction – reproduced from Distefano et al. 2009 (below). There is also a video although it doesn’t work on the iPad. This page contains links to videos for all the exercises in this article (links may take you to a “mobile page” if using iPad or iPhone. Use the search box and you can locate the full article without signing in).

Distefano et al. 2009 used an EMG study to determine which exercises produced most activation of Gluteus Medius. EMG (Electromyography) uses electrodes to examine muscle activation. They compared a number of exercises including hip clam, single limb squat, single limb deadlift, lateral band walks, multiplanar lunges and multiplanar hops. They concluded that,

“The best exercise for Gluteus Medius was side-lying abduction”

They found single limb squat to produce the second most EMG activity and lateral band walk the third.

Now you may be wondering does this translate to changes that I can feel and see? Will it improve my symptoms? It’s fair to question this, EMG change is all good in theory but what about in practice?

Fredericson, who’s written quite widely on ITB problems and their treatments did a study on runners in 2000. They found that,

“Long distance runners with ITBS have weaker hip abduction strength in the affected leg compared with their unaffected leg and unaffected long-distance runners….symptom improvement with a successful return to the pre injury training program parallels improvement in hip abduction strength.”

So hip abduction, one of the roles of Gluteus Medius, is weaker on the side runners have ITBS. No great surprise there. What exercises did they use to rehab it and return people to running?…..

… guessed it side lying abduction and pelvic drops.

Update 3/6/12:

Thanks to Stuart Palma (a nice bloke despite being a Liverpool fan!) we also have another article to add to the mix. McBeth et al. 2012 compared 3 sidelying exercises sidelying abduction, the clam, and sidelying abduction with external rotation in healthy runners. Unlike previous articles they used a leg weight and a “biofeedback unit” to monitor trunk position. They compared muscle activation in Gluteus Medius, Gluteus Maximus, anterior hip flexors and Tensor Fascia Latae with each exercise. They concluded that,

The sidelying hip abduction exercise was the best for activating Gluteus Medius with little activation of Tensor Fascia Latae and anterior hip flexors”

A few key points here then;

  1. It gives us an indication of specificity – sidelying abduction appears to be able to activate the muscle we are targeting without working the muscles we are not targeting.
  2. Gluteus Medius activation was similar to Distefano’s earlier work at 79.1% of MVIC.
  3. Important to note there is no comparison with weightbearing exercise

Now I could stop there. I’ve made my point – sidelying abduction clearly has a role, but part of this is not about cherry picking a couple of articles to prove a point, it’s about looking at the bigger picture even if that includes research that goes against your theories….

Side plank abduction

This is reproduced from Boren et al. 2011 who did an excellent study and also compared their results with earlier work. Their top 3 exercises for Glute Medius were side plank abduction with dominant leg down, side plank abduction with dominant leg up and single leg squat (in that order). Notice again that these positions, despite being “non-functional” do create a lot of activity in Glute Medius and again more so than weight bearing positions such as single leg squat. Of note too is that they found less activity with side-lying abduction than the previous studies. This raises a good point with research and rehab. Nothing is concrete. You simply cannot say “this exercise has no role” as you will find evidence to support your claim and evidence to refute it. Also they used a slightly different technique which might account for the difference.

The Clam

Again reproduced from Boren et al. 2011 but I’ve modified them with some instructions to clarify (thanks to Debra for the suggestion). The link above also includes detailed descriptions of this exercise and its progressions (it’s all in the appendix at the end of the article). Most people think of the clam as the exercise described in Clam 1 above. In this study 3 progressions of the exercise were included. In many exercises they can be progressed by increasing resistance, this one is progressed by a change in position. The article demonstrated an increase in glutes activity from Clam 1 being lowest to Clam 4 being the highest. Compared to other exercises in the study Clam 4 had a high level of Gluteus Medius activity (77% of MVIC – Maximal Volitional Isometric Contraction) only slightly lower than Single Limb Squat (81% of MVIC). Distefano et al 2009 showed lower levels of activity with the clam but didn’t include the same progressions.

So we’ve seen 3 exercises, all in sidelying, that produce high levels of Gluteus Medius activation, at least comparable to, and in some cases higher than weight bearing exercise. It’s no great surprise then that the English Insitute of Sport uses a selection of sidelying exercises in its “Glutes Circuit”. We were given this by one of their team at a Strength and Conditioning lecture but sadly can’t reproduce it online.

Next we look at weight bearing exercise

Single Limb Squat

Once again from Boren et al. 2011. Single Limb Squat as above showed good Gluteus Medius and Gluteus Maximus contraction with 81% of MVIC for Gluteus Medius and 71% for Gluteus Maximus. Both muscles are hugely useful for runners and so this exercise clearly has its benefits. As a result it’s one I use regularly. The only issue here is the risk of aggravating pain. The deep dip position places greater stress on the ITB and patellofemoral joint. As a result I often start with a shallow knee dip or use this exercise after first rehabbing with sidelying exercises. There is also an issue of control, some patients struggle to even balance on one leg let alone perform a squat.

Distefano et al. 2009 used the slightly different technique shown above, they also reported good activation of both Glute Med and Max. Lateral Band Walk shown below, also showed good Gluteus Medius activation (although less than side lying abduction). This doesn’t appear to have been assessed by Boren et al. 2011.

Wall Press – reproduced from O’Sullivan, Smith and Sainsbury 2010. In the picture below the right Gluteus Medius is being exercised by pushing the left knee, hip and ankle against the wall and maintaining a contraction for 5 seconds.

Wall press was compared to Pelvic Drop and Wall Squat and achieved the highest MVIC of 76%.

Pelvic Drop – picture below reproduced from Bolgla 2005 who found reasonable activation of Glute Medius on pelvic drop of 57% of MVIC and was similar to Boren et al 2011 (58%). In Bolgla’s study it scored the highest of 6 exercises which also included sidelying abduction.

For pelvic drop the standing leg (right in this case) stays straight and you lower your other leg by lowering the pelvis on that side.

Krause et al. 2009 (abstract only) studied the effect of doing exercises on a balance cushion and found an increase in Gluteus Medius activity (compared to balancing on normal floor) although it should be noted this difference wasn’t thought to be statistically significant.

Lubahn et al. 2011 looked at the effect of using resistance band to pull the knee more medially (towards the other knee) during weightbearing exercises. The idea behind this is that the medial pull of the band should increase activation of Gluteus Medius. They found it didn’t increase activation during single limb squat or step up and may result in poorer limb alignment during the exercise.


Acknowledging Limitations and closing thoughts

An important part of any theory is acknowledging limitations. Like I’ve mentioned above nothing is concrete. I’ve barely scratched the surface of research in this area and I acknowledge there is more research out there, undoubtedly with different findings. I also realise that the main measure involved in these studies is surface EMG recording of muscle activity in healthy individuals. This measure is only related to Maximal Volitional Isometric Contraction presented in a percentage and has not included people with injuries. Endurance activity such as running rarely needs maximal voluntary contraction. So we can’t conclude that because sidelying abduction has a higher % of MVIC than single limb squat that it is a better exercise for rehab in return to running. Neither can we conclude that certain exercises are better for rehabbing certain conditions as the research above (with the exception of Fredericson et al. 2000) doesn’t examine the effect of exercise on injury. That said I think the research is a useful indication for strength work – Boren et al suggested that an MVIC of greater than 70% was needed for strength work while ealier research suggested a range of at least 40-60%. Sidelying and weightbearing exercises have both achieved greater than this range and should be capable of producing strength changes.

The aim of this blog was not to prove sidelying exercises are more effective but only to show that they create good levels of glutes activation and have a role in rehab. This idea that exercises must be functional is a slightly limited one. The very fact that sidelying work isn’t functional may be its advantage – it is a task that we don’t do as part of our day, a task that can potentially isolate a muscle to gain good activation. If functional tasks were so good at glutes rehab, we’d all have great glutes just from walking around, climbing stairs and running!

The other advantage of sidelying work is that it can often be done without aggravating symptoms if done correctly. For people with painful ITBS or irritable patellofemoral pain a range of weightbearing exercises can make symptoms worse.

Selecting exercises for yourself or a client is very individual and should be based on addressing specific weaknesses. Take sidelying abduction, some people can do 40-50 with good form and minimal glutes fatigue. There is little point asking them to do this as an exercise as it probably won’t overload the glutes to achieve strength changes. Some people get to 9 or 10 reps and start to fatigue or lose control and they are more likely to benefit from it. For some the issue actually isn’t strength at all, some studies show poor correlation with Glute strength and pelvic position, this is because control of movement can be poor even with good strength. In that situation it’s control and form that needs to be addressed and this is where functional weightbearing movements are more important. I think if your control of pelvic position is poor, but Glute strength is good it’s unlikely sidelying work will be of great benefit.


So what to do for your Glutes? The answer is what works for you. If you find sidelying exercises are getting you results without causing symptoms then great. If not try some of the others above and see which ones seem to work your glutes and get results.


What about reps and sets? Assuming we are working endurance roughly speaking people usually start at around 10-15 reps 3 sets in a session with a rest of 1-2 minutes between each set. Then progress up towards 25 reps, you’re aiming to fatigue the muscle so there will be lots of individual variation in the reps needed to do that. That said if you can do more than 30 reps without fatigue maybe you need a harder exercise? These recommendations are based on those by the American College of Sports Medicine research (abstract only) which I’ve summarised in a table below (click to expand). Remember too that form and control are very important and you should feel it in your glutes not the side of your knee or front of the thigh.

Closing thought, from the research I’ve read and patients I’ve seen, a combination of both functional weight bearing and less functional (sidelying) exercises is most likely to be effective in glutes rehab.


Please feel free to comment, I welcome other opinions even if very different from my own. Where possible back up any claims with research evidence.




6 thoughts on “Gluteus Medius – evidence based rehab

  1. Pingback: Femoro-acetabular Impingement (FAI) | RunningPhysio

  2. Pingback: Patellofemoral Pain Syndrome – PFPS – Part 1 | RunningPhysio

  3. Pingback: Patellofemoral Pain Syndrome – PFPS – Part 2 | RunningPhysio

  4. Ciao, scrivo dall’Italia. Soffro di ITB da circa 4 anni. Ho fatto riposo, tecarterapia ma non riesco a risolvere il problema. Ogni volta che riprendo a correre dopo 2-3 Km mi viene il dolore.
    Ho letto molto su internet e su siti stranieri per cercare di risolvere il problema. Di recente dopo la corsa con dolore mi sono fermato 1 settimana e poi, quando non avevo più dolori, ho provato a seguire il protocollo stretching e potenziamento ITB e glutei ma sento indolenzimento e poi dolore alla ITB. Dall’ecografia fatta un anno fa emerge solo un modesto ispessimento della bandelletta ileo tibiale, il resto è tutto OK! Cosa posso fare?
    Ciao, grazie

  5. Pingback: Resistance training and running | RunningPhysio

  6. Pingback: Resistance training and running | RunningPhysio

Leave a Reply

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

You are commenting using your 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