This is the first post in my #NoMore series. It has to do with movement beliefs and manual handling theories which need updating based on the most up to date evidence. The main purpose of this series is to encourage a shift from the traditional kinesiopathological model of movement to a movement optimist approach. Another purpose is to educate people in the current evidence in a short and concise(ish) manor. I hope that you learn something and enjoy the read.
First, I would like to acknowledge three blogs which were extremely influential in the creation of this post:
1. Ben Cormack - Core Stability - Does it matter? A Look at the evidence
2. Brendan Moutt - Why we need to stop blaming the Transversus Abdominis for back pain
3. Greg Lehman - Core stability and pain: is it time to stop using the word stability to explain pain?
Next, I should start with a brief overview on some statistics for back pain. Throughout their lifetime, 60-80% of people experience low back pain at least once (Gordan & Bloxham, 2016). The majority of this pain can be labeled as non-specific in nature, meaning not caused from a insidious pathology. Finding a solution to this problem is an extremely difficult and complicated task; one of the proposed ways to do this is through core stabilization exercises to help stabilize the spine.
Brief History of the Core Stability Theory for LBP
First we must start with a bit of history to understand why lack of core stability has earned blame as the cause of low back pain.
The setting is Queensland University in Australia in the 1990's. Two researchers, Dr. Paul Hodges and Dr. Carolyn Richardson, published a series of papers which looked at transverse abdominis (TVA) and multifidus timing while performing various tasks in people with and without low back pain.
The first study, published in November 1996, looked at the difference in TVA timing in subjects (n=15) with low back pain (LBP) and subjects (n=15) in a control group without LBP. The researchers instructed both groups to move their shoulders in various planes of motion and they measured the timing of TVA and multifidus via EMG. They identified that these two core muscles turned on before or within 70-113 msec of the deltoid activation. Therefore the conclusion from their research was that in those with LBP the TVA and multifidus was delayed in activating (by 70-113 msec).
The second study, published in February of 1998, investigated the association between lower limb movement in people with LBP (n=15) and a control group without LBP (n=15). While performing these movements they had an EMG recording timing for TVA, rectus abdominis, abdominis obliquus externus, abdominis obliquus internus, and the erector spinae. With this study they found that the TVA was delayed 57-86 msec in those who were experiencing LBP.
One of the main conclusions from these two studies from Hodges and Richardson was " The mechanism of delay in abdominal muscle activity is also uncertain. Furthermore, it is not known whether the delay in abdominal muscle contraction associated with limb movement proceeds the onset of LBP or is a result of the pain." They hypothesized that a diminishment of stabilization of the lumbar spine is one of the contributors for those experiencing LBP.
Now, we should be asking ourselves (like Hodges & Richardson did) why was there this delay. Was this caused by the LBP or was it in response to the LBP. Whichever it is we must also question the relevance of a ~50 ms delay.
This research lead to the idea of the feed-forward criteria as speculated by Hodges & Richardson (1998) for the core stabilizer muscle timing. The idea is that the brain prepares/plans for the movement prior to and not in response to the movement; this was proposed as a result of their findings that prior to the movement of the deltoid, the TVA and multifidus activated.
And BOOM, this research was extrapolated by chiros, physios, and other manual therapists as we need to focus on spinal stability for rehab because the TVA and other spinal stabilizers are weak and the spine is unstable. This is however not what this study found; all it found was that there was a delay.
Current Research on Core Stability and LBP
When we move to focus on more recent research, which looks at the utilization of core stabilization exercises and outcomes of reduction in LBP, we find the following:
1. Research conducted by Vassejen et al. (2012) looked at the effect of core stabilization exercise on the feed forward mechanism of activation with a similar deltoid movement that Hodges used. They also compared this core stabilization group to a high load sling exercise group, and to a general exercise group. After an intervention period of 8 weeks in all groups they looked to see if there was a difference in onset of the abdominal muscles.
After 8 weeks they found that there was no difference in pain between the three groups. They did find that their was a change from 19 msec delay to 15 msec in the high load sling exercise group, but no changes in the two other groups. From these findings they concluding that the "abdominal muscle onset was largely unaffected by 8 weeks of exercise in chronic LBP patients. There was no association between changes in onset and LBP". Therefore, from this study we find that 8 weeks of stabilization exercise training didn't have an effect on the timing via the feed forward mechanism. This means that if this delay is the cause of pain we may not be able to affect it through stabilization specific exercises.
2. Another RCT published by Carnes, Foster, & Wright (2006) looked at spinal stabilization exercise against "conventional physiotherapy" exercises (general activity exercises) in patients who experienced recurring LBP with a follow up at 6 and 12 months. Using metrics to analyze physical functioning, pain, psychological distress, and quality of life to determine if there was a difference between the two groups and also when compared to a control group.
Their conclusion from this research was that "specific spinal stabilization exercises provide no additional benefit in terms of physical functioning, pain, psychological distress, and quality of life over a package of care consisting of advice and conventional physiotherapy (mainly exercise and manual therapy). Both groups had a clinically meaningful improvement in function and reduction in pain over time, but no statistical difference between groups was shown." Therefore, both groups were able to experience decrease in their scores, showing that general exercise is equally as effective in providing clincially meaningful improvements as spinal stabilization exercises.
3. Next we must look to see if a change in abdominal muscle contraction is associated with improved clincial outcomes, luckily we have Marrion et al. (2012) who carried out exactly this research. They looked at spine stabilization exercises in the treatment of chronic LBP to see if it had a better clinical outcome. They also tried to identify if there was a change in the feed forward activation of TVA, obliquus internus, and obliquus externus during rapid arm movement pre and post stabilization exercise program. They also measured to see if there was a decrease in pain (0-10 scale) and disability (Roland - Morris Questionnaire).
They found that post intervention there was a significant difference in the ability for the subjects to voluntarily activate their TVA (4.5%, P=0.045), however based on their data they concluded that clinical outcomes based on the utilization of spine stabilization exercises didn't depend on the patients ability to recruit TVA during the task. Another way they phrased that was that the TVA muscle function before or after intervention wasn't a statistical predictor of a good clinical outcome. This is a significant result because if the voluntary activation of TVA isn't related to beneficial clinical outcomes, then we have to identify some other cause for the decrease in pain and disability.
4. One last study which I will look at is Marshal et al. (2013), Pilates or cycling for chronic LBP. The participants (n=64) were randomized into two groups; one which completed specific trunk exercise (the Pilates group) and one which cycled on a stationary bike. Both groups went through 8 weeks of intervention. At the start of the study, at 8 weeks, and at a 6 month follow up, they had the participants rate their pain, disability, and catastrophizing.
They found that after 8 weeks of intervention disability was significantly lower in the Pilates group. They also found that pain and catastrophizing were reduced from baseline in both groups. When they moved to the follow up at 6 months they determined that there was no long-term difference between groups on clinically meaningful improvements. It is important to note whether or not having a decrease in pain at 8 weeks was relevant. Since these patients were chronic in nature, the difference in the groups could have been a result of their beliefs on pain and not due to the stabilization exercises. Either way this study is important because it is more evidence to show that general exercise is equally as effective of a treatment for LBP as spinal stabilization exercise in the long term.
When we summarize the studies looked at in this post, there is a clear direction in which the evidence appears to point. However, how can we be sure that this is indeed the way the evidence actually looks and not just the result of my selection bias (and me not wanting to summarize study after study)? Well, we can look at systematic reviews and meta analysis and find out what additional information they have to offer.
There have been numerous systematic reviews and meta-analyses (Saragiotto, et al, 2016; Rackwitz, et al., 2006: Macedo et al., 2009; May & Johnson, 2008; Ferreira et al., 2006; Smith, Littlewood, & May, 2014) published regarding stabilization exercises and their influence on outcomes for LBP. The conclusion of most include a statement which reads similar to this: there is no additional benefit of prescribing motor control or TVA activation exercises over other forms of exercise for long term outcomes.
However, we do find one outlier to this general theme by Bystrom, Rasmussen-Barr, & Grooten, (2013). Their review did find that motor control exercises had better outcomes, however the researchers stated "It is to date not known if the effect of MCE (motor control exercises) on pain and physical impairment in LBP is due to the isolated activation of the local musculature or subsequent stages of the intervention involving loaded postures engaging all trunk muscles." They were not able to identify if the intervention was successful due to its specificity or due to the engagement in activity.
Another review I want to highlight, which had a very strong conclusion, was by Smith, Littlewood, & May (2014). They found "there is strong evidence stabilization exercises are not more effective than any other form of active exercise in the long term. The low levels of heterogeneity and large number of high methodological quality of available studies, at long term follow-up, strengthen our current findings, and further research is unlikely to considerably alter this conclusion."
Fairly conclusively, the evidence points (and not just due to my selection bias) that the theory of the spine being unstable and needing stability does not stand up to the academic process. However, as we know about many other theories out there, it takes a lot of time to reverse false narratives which have been widely regarded for some time. Hopefully this post is a drop in the bucket which moves us closer to better understanding.
5 Key Take Aways
1. LBP is a multifactorial issue and many aspects of the biopsychosocial framework have relevant influence upon pain and disability. Many times a single intervention as a blanket treatment is not the best option.
2. Yes, core stabilization exercises help to reduce LBP. However, it doesn't seem like it is because there is a change in TVA timing or spinal stability, and it doesn't appear to reduce pain any more than other active exercise in the long term (Smith, Littlewood, & May, 2014).
3. A great take away is that we don't have to prescribe these over-complicated stabilization exercises if that isn't what the person in pain wants. We can get people back to doing what they want to do through general graded activity.
4. Focus on individuality, educate patients about pain, treat the whole person in front of you and not only the condition.
5. Empowering someone in pain to be able to choose the activity can have a positive influence on their adherence to the exercise (Middleton, Anton, & Perri, 2013). If they are more active they will also get all of the other secondary benefits that exercise gives.
All the information in this post gives us all the more reason to be/become Movement Optimists
Thank you for reading,