pain The Wellcome Trust


  Cognitive behavioural approaches to chronic pain

Complementary medicine for pain

Phantom limb pain

About this site
Contact us
Phantom limb pain

Jonathan Cole

Phantom limb pain – pain appearing to come from where an amputated limb used to be – is often excruciating and almost impossible to treat. New approaches, based on a better understanding of the brain's role in pain, may be opening the way to new treatments.

After amputation of a limb, an amputee continues to have an awareness of it and to experience sensations from it. These phantom limb sensations are also present in children born without a limb, suggesting that perception of our limbs is 'hard-wired' into our brain and that sensations from the limbs become mapped onto these brain networks as we develop.

If phantom limb sensations are normal then so too, alas, is phantom limb pain. This occurs in a majority of those who lose their limbs. (1) In fact, limbs do not need to be lost; it also occurs in conditions in which the brain is disconnected from the body, such as peripheral nerve injuries and after spinal cord injury, when an area becomes insentient (and usually paralysed).

The pain is described in various ways: burning, aching, 'as if the hand is being crushed in a vice,' etc. Such words, however, cannot fully encompass the experience of living with such a pain. In those with chronic pain after spinal cord injury it is frequently the pain rather than the paralysis that interferes with work and social life. One woman has said that paralysis does not stop life, but pain may. (2)

There may be many mechanisms underlying phantom limb pain. Damage to nerve endings is often important: subsequent erroneous regrowth can lead to abnormal and painful discharge of neurons in the stump, and may change the way that nerves from the amputated limb connect to neurons within the spinal cord. There is also evidence for altered nervous activity within the brain as a result of the loss of sensory input from the amputated limb.

Unfortunately, phantom limb pain is generable intractable and chronic; once it develops it persists and is rarely improved by present medical treatments. Destructive surgical procedures are also of limited use. They can be effective for a few months, but pain always returns, frequently worse, and so surgery is only performed in patients with terminal illness.

New treatments
Recently, some potentially valuable treatments have arisen, based on new ways of perceiving the origin of the pain itself.

Flor's group has shown that the development of phantom limb pain is correlated with changes in the way peripheral areas of the body are represented in the sensory cortex. Although is not clear why this should lead to pain, they devised experiments to reverse this cortical plasticity to see whether pain sensations were also altered.

They found that use of an electrical prosthetic limb moved by signals from the patient's muscle reduced the pain if used for several hours per day. Brain imaging revealed that this effect was dependent on a reversion of the sensory cortex to its original state. (3) A task involving repeated touching of the skin over the stump, to improve sensory discrimination there, also reduced phantom limb pain, possibly by replacing some of the sensory input to the brain lost following amputation. (4)

Visual tricks
In his last book Patrick Wall suggested that pain might be considered a 'need state', like thirst, rather than simply a sensation. If so then the 'need' might involve movement to avoid or reduce pain. (5)

Evidence that stimulation of the motor cortex (the area that controls movement) can reduce phantom limb pain has been around for some time. (6) Perhaps more surprising was a trial by Ramachandran and Rogers-Ramachandran in 1996. (7) They asked people with amputations of the arm and phantom limb pain to place their arms inside a mirror box so that they saw their remaining arm mirror-reversed to look like their amputated one. When they moved their remaining arm in the box they were 'fooled' into thinking they were moving their amputated one, and their pain was reduced. Although this has proved less effective in some subsequent trials, it did suggest that phantom limb pain might reflect a loss of motor control to the limb, as well as loss of sensory input from it.

More recently the mirror box has been used with some success in pain that is not due to sensory loss. (8) In fact, a box may not be required. In phantom limb pain due to a peripheral nerve injury (brachial plexopathy), Giraux and Sirigu have shown that merely training patients to imagine their paralysed arms moving in relation to a moving arm on a screen in front of them can relieve phantom limb pain. (9)

They suggest that these attempts to link the visual and motor systems might be helping patients recreate a coherent body image, and so reduce pain as a result of reduced and disordered input. If this approach is successful, it may be that relatively simple treatments, such as patients imagining that they are swinging a golf club with their amputated limb, could have significant pain-relieving benefits.

Finally, in experiments still being developed, we are constructing an arm in virtual reality which subjects with phantom limb pain will move themselves using motion capture techniques. Movement of their stump will be captured by a movement-tracking device, and used to project the movement of the reconstituted limb in virtual reality. We anticipate that this will lead to a sense of re-embodiment in the virtual arm and hence to a reduction of the pain.

These new approaches are all based on a shift in
in phantom limb pain away from the site of damage – the stump – to the centre of pain processing: the brain. It appears that disordered inputs from the limb's sensory systems, combined with disrupted motor signal back to the limb, generate a mismatch between the brain's built-in map of the physical body and what is actually perceived. For some reason, this mismatch results in pain.

Whichever of these new techniques proves effective – and simple enough to be used – the prospects for relief from pain are probably brighter than at any time since Weir Mitchell first coined the term phantom limb pain in 1872.

Professor Jonathan Cole is a Consultant in Clinical Neurophysiology at Poole Hospital a professor at the University of Bournemouth, and Senior Lecturer in Clinical Neurosciences at the University of Southampton, UK. E-mail:

(1) Flor H (2002) Phantom limb pain: characteristics, causes and treatment. Lancet, 1, 182-189.

(2) Cole J (1994) Still Lives; narrative of spinal cord injury. Massachusetts: MIT Press.

(3) Lotze M, Flor H, Grodd W, Larbig W and Birmbaumer N (2001) Phantom movements and pain: an fMRI study in upper limb amputees. Brain, 124, 2268-2277.

(4) Flor H, Denke C, Schaefer M and Grusser M (2001) Sensory discrimination training alters both cortical reorganisation and phantom limb pain. Lancet, 357, 1763-1764.

(5) Wall P D (1999) Pain, the science of suffering. Weidenfeld and Nicolson.

(6) Garcia-Larrea R, Peyron R, Mertens P, Gregoire M C, Lavenne D, Le Bars P, Convers F, Mauguiere F, Sindou M and Laurent B (1999) Electrical stimulation of motor cortex for pain control: a combined PET scan and electrophysiological study. Pain, 83, 259-273.

(7) Ramachandran V S and Rogers-Ramachandran D (1996) Synaesthesia in phantom limbs induced with mirrors. Proc R Soc Lond B Biol Sci, 263, 377-286.

(8) McCabe C S, Haigh R C, Ring E F J, Halligan P W, Wall P D and Blake D R (2003) A controlled pilot study of the utility of mirror visual feedback in the treatment of complex regional pain syndrome (type 1). Rheumatology, 42, 97-101.

(9) Giraux P and Sirigu A (2003) Illusory movements of the paralysed limb restore motor cortex activity. Neuroimage, 20, S107-111.