Graded motor imagery protocol for CRPS
In 2004, Lorimer Moseley published a RCT which assessed the use of a graded motor imagery protocol for the treatment of Complex Regional Pain Syndrome, Type 1 (CRPS-1). I find this protocol to be quite helpful in my treatment of CRPS-1 (when occuring in a patients hand or foot, unilaterally). Each step is performed for 2 weeks.
1. Recognition of laterality in the hand or foot: Moseley states that “recognizing a pictured hand to be a left or right hand activates brain areas involved in higher-order aspects of motor output, the so-called pre-motor cortices, whereas explicitly imagined movements also activate the primary motor cortex.” Perform this step by making flash-cards of pictures of hands or feet in different positions. The patient is then asked to identify these as left or right as quickly as possible and as the examiner, you calculate the % identified correctly. This is to be performed as quickly as possible. Software has also been developed to perform this exercise/test and is found here.
2. Imagined movements: Use the flash-cards of the hands and feet in exercise. The patient will imagine moving the affected limb to the position demonstrated on the card. They will then imagine returning to neutral and repeat this 3+ times.
3. Mirror-box therapy: Create a box to place the affected limb into with a mirror on the side. Place the affected limb inside of the box and using the unaffected limb, assume positions as illustrated on the flash-cards used in the first 2 exercises. While moving the unaffected limb, look into the mirror so that the patient has the illusion that the affected limb is performing the activity. Eventually progress this so that the affected limb is moving inside the box with the unaffected limb, but still paying attention to the mirror.
Read more at www.forwardthinkingpt.com
Update: For a limited time, the NoiGroup has offered the Recognise Application for iphones for free.
Moseley GL. Graded motor imagery is effective for long-standing complex regional pain syndrome: a randomized controlled trial. Pain 2004: 108; 192-198.






Comments
Hi Joe,
How many patients have you used this protocol on? What outcome measure do you use and what scores have you got? I only get a few CRPS a year so I would be interested to generalise. Does it work for central post stroke pain, phantom limb or other condition where a central reorganisation needs to be facilitated? I have tried mirrors on these groups with very variable success.
Kind thoughts,
Hey Steve,
I have used this on probably 10-12 individuals over the past year with rather good success. I have only used it in individuals with CRPS but I would suspect anyone who is in a process of central sensitization to have some success. I stay very strict to this protocol in conjunction with pain education. I do not include any other activities (or modalities—thermal stimuli actually can make the condition worse) until the 6 weeks have been completed and limit their physical activities during this time. You do not want to just do mirror-box interventions without first attempting to have an effect on the pre-motor cortices through the planning of painfree movements through laterality and imagined movements.
I measure all patients using the recognise app for iphone to measure laterality (i used the online version prior to the app from the noigroup), use a tampa scale for kinesiophobia, pain catastrophizing scale, and VAS. I also ask their perceived expectations that their condition will improve pre- and post- pain education. Expectation for improvement is key. If they do not have expectations, help set them through education as well as possible mentorship w/individuals who have been successfully treated for this condition.
Cheers Joe,
Great advice and it is well recieved. I particularly like the choice you make to measure expectation for improvement. I agree and suspect that this expectation represents a state of mind that indicates that the brain is willing. Willing to risk the threat of a pain experience. To revisit high risk areas. And following the protocol as stated would be a zero added pain experience which would build on their willingness. Nice.
I suspect that expectation would be proportional to belief (in the neurophysiological model you teach) and trust (in you).
Kind thoughts.
Steve
Hi,
thanks for your post. I, too have practiced this protocol w/ CRPS clients. However, I haven’t had the outcomes you describe.
1) Are the CRPS patients you treat fairly acute or in the chronic stage ?
2) Are you only progressing the patient if they make improvements on their laterality recognition ?
3) what if they don’t have any improvement with their laterality ? do you still progress them to the 2nd phase ?
Thanks,
Renee, PT
Thanks for your suggested protocol. I was wondering how you might approach graded motor imagery in a patient with bilateral symptoms, particularly when it comes to using the mirror box. Do you simply put their most affected/worst hand inside the box? Or is this form of therapy inappropriate? Looking forward to your thoughts.
Katrina
Hey guys,
I would say by the time they are in my office they would be classified as “chronic” but I hate using the terms acute/chronic. I prefer to classify patients in pain as nociceptive, peripheral neuropathic or central sens.
I have seen patients improve in their laterality measures (using recognise) but like I posted before, this is all they do for 2 weeks. Laterality exercises become their life. If there were no improvements, I would continue to attempt to improve laterality before progressing.
In terms of patient with bilateral symptoms or symptoms that are spreading, I still do laterality and graded motor imagery but I do not do mirror-box. The success of these individuals is much lower, from my antedoctal experience, but I would be interested to see if they would take ownership or tolerate treatment of a realistic, prosthetic limb. There has been some literature to support healthy individual can gain sensation in prosthetic limbs, but more research needs to be done to look at applicabiltiy in individuals with CRPS.