Imagine you had to have a limb amputated. Perhaps you sustained this injury as a hero or heroine soldier, or perhaps you were just caught in a conflict or natural disaster (such as the earthquake in Haiti). Or perhaps the amputation was due to some other accident or disease, such as an amputation to stop a tumor from spreading.
Quite often after amputation, individuals experience phantom limb-the feeling that the arm or leg is still there. Why does this occur? It turns out that while the limb has been amputated and is gone, the brain area that corresponded to the limb remains, and often the rest of the brain interprets the signals from this brain area (e.g., the hand or leg area) as the limb still being present.
Now, imagine that in addition to missing a limb after amputation, you not only have a phantom limb sensation but immense pain from this phantom limb, which is an unusual painful position or the feeling of having a "muscle" spasm.
To visualize what a person experiences, bend your elbow and clench your fist as tightly as you can and keep it clenched. This is very painful. You can relieve the pain simply by opening your fist and relaxing the clench.
But what if this was a phantom hand causing pain? This is not an uncommon problem in amputees. What can the patient do now to relieve the pain?
Birth of a Theory
For millennia, this problem seemed impossible to solve. Often, the phantom clenched fist would eventually relax on its own. However, patients can sometimes have some spasms lasting minutes or longer, or have 10 or more spasms a day. All variety of medicines have been tried, typically with little benefit. I have known of patients on such high doses of opiates that they were too sedated to function well, yet the pain still continued.
Even more striking, following the theory that the pain was coming from the stump of the residual limb, some patients even had an amputation at a higher level in a vain attempt at relief. Others had electrical stimulators placed surgically. As we now know, the pain is coming from the brain, so these therapies are typically not effective.
In 1995, my teacher from medical school-V.S. Ramachandran, MD, PhD, of the department of psychology at the University of California in San Diego-discovered a brilliantly simple and effective treatment.1
Take for example a patient with a right arm above-elbow amputation with severe pain from a phantom hand that often goes into spasm of a clenched fist. Dr. Ramachandran had the patient put one hand on either side of a parasaggitally aligned mirror, positioning the intact arm and hand in the same clenched position as the painful phantom. He instructed the patient to look at the reflection of the intact arm in the mirror, to open the fist of the intact arm, and also try to open the clenched phantom, all the while watching the reflection of the intact arm in the mirror.
Lo and behold, the patient smiled and gasped in astonishment. The clenched phantom had "opened" and his pain resolved. Dr. Ramachadran found in a larger, later study that in most cases, the mirror was most helpful in relieving pain from spasms of a phantom, or the pain of a phantom being in an uncomfortable position.2
What should a patient with pain from an elbow in a flexion spasm or an ankle or great toe in dorsiflexion spasm do? Similar to the phantom fist in clenched spasm, the patient should place one arm or leg on either side of the mirror, place the intact arm in the same position as the painful phantom, and watch the reflection of the intact arm. Then, move the intact limb-i.e., extend the elbow or plantar-flex the ankle or flex the toe-while trying to move the phantom.
Even though this mirror visual feedback (MVF) technique may temporarily relieve the pain from a spasm, does the patient have to constantly use a mirror throughout the day? We have found that in many cases, after using the mirror for a number of times in a day or for a few days, the number of spasms themselves becomes lower, and the severity of the spasms decreases. Often, sooner rather than later, the spasms become minimal or tolerable for the patient without even using the mirror.
Research Support Builds
Subsequent to Dr. Ramachandran's work, others have found the mirror to be helpful for the pain of poorly mobile or spasming phantoms.3-7 In particular, Chan et al completed a randomized controlled study of MVF for phantom limb pain in lower-limb amputees. The study had three arms: MVF; a control in which patients moved the intact leg and tried to move their phantom leg while watching a mirror that was covered (so it did not reflect); and mental visualization-patients tried to visualize moving their phantom leg.
Pain fell by about 90 percent in the patients on MVF, but did not improve (and even rose a bit) for patients in the other two groups. Patients in the covered mirror or mental visualization groups were subsequently crossed over to MVF, and these patients too experienced an 80 to 90 percent drop in pain.
It is absolutely essential to emphasize here that like anything in medicine, MVF does not work for all patients, and even if it works for one patient, it may work to a greater or lesser extent in any individual patient.
In particular it's crucial to appreciate that there can be various types of phantom pain. We have discussed pain from a phantom in spasm or a phantom in an unusual position, and how MVF helps the patent move the phantom limb and relieve the pain. MVF works best for pain requiring movement to relieve the pain. Patients also can alternatively or additionally have a constant burning pain, and MVF is not helpful for this.
We've seen patients with an itch in their phantom-such as on the bottom of their phantom foot. In this case, we've found that by watching the reflection of the intact foot in the mirror and having someone scratch the bottom of the intact foot, some patients can have the itch in their phantom foot "scratched."1,2,8
How might MVF work? We think that it's due to an interaction between vision of the intact limb-which looks like the amputated limb "resurrected"-and proprioception in higher brain centers for the amputated limb. When one watches the reflection of the intact limb, it interacts with the central position sense for the amputated limb. This then feeds into central motor commands that would be sent to the limb, were it still intact.
Even people with all their limbs can experience the MVF effect. Put one hand on each side of a mirror, watching the reflection of one hand. Open and close your hands and move your fingers on both hands, but move the hands and fingers differently. You will experience a strange or jarring feeling.
You are experiencing the brain's reaction to the discrepancy between the vision of the reflection of one hand-which looks like the other hand-and the position sense you have of the other hand. Charles Spence and colleagues from Oxford University have performed a lot of very careful experiments in normal subjects validating this interaction of vision and proprioception as the basis for MVF.9-12
Mirror Therapy for Stroke and CRPS
If MVF works to help move a poorly mobile phantom arm, can it be used in other rehabilitation situations? With Dr. Ramachandran, I was the first to show that MVF may have some benefit in patients with hemiparesis following stroke.13 Subsequently there have been a number of studies with similar findings.14-19
It's crucial to point out that stroke patients are a very heterogenous group. MVF may work for well for some patients, yet only a little or not at all for others. Many further studies are needed.
Typically, patients in this category are told to practice using the mirror about 15 minutes, twice per day, five to six days per week. Patients put one hand on either side of the mirror, watch the reflection of the "good" hand, and move both hands-the affected hand as best as possible.
What exercises should you prescribe? Say a patient is having trouble with hand coordination. Have the patient work on opposing the thumb to each finger in turn. If the patient has trouble with forearm pronation/supination, work on that.
Another condition in which MVF may find wide application is complex regional pain syndrome/reflex sympathetic dystrophy (CRPS/RSD). Dr. Ramachandran suggested some time ago that that this could be a potential application of MVF, and quite a number of studies have shown utility of MVF in CRPS/RSD.20-24
In patients with this condition, MVF may be most helpful in getting patients to move the affected limb. Also, the theory holds that subjects could put one limb on each side of the mirror, watch the healthy limb, and have someone stroke the healthy and the affected limb at the same time.
Perhaps by watching the reflection of the healthy limb-which looks like a normal limb-while being stroked, patients may experience a desensitizing of the hyperesthesia of CRPS/RSD. It would be wonderful if early implementation of MVF could help make moderate or severe CRPS/RSD extinct, but studies will be needed to see whether this is possible.
Finally, I was the first to show that MVF could be helpful for an orthopedic (musculoskeletal) injury.8,25 Again, larger studies are needed to determine whether this is a useful general application.
In susmmary, MVF appears to work by visual feedback from the reflection of the healthy limb acting as a kind of "active assist" to move an affected limb-be it a poorly mobile or spasming phantom limb or a limb affected by stroke, CRPS/RSD or orthopedic injury. Every patient is different, and more and larger studies are needed, but MVF may well be a most useful tool for therapists and patients.
References are available at www.advanceweb.com/pt under the Resources tab.
Eric Altschuler is associate professor in the department of physical medicine and rehabilitation at the University of Medicine & Dentistry of New Jersey.