Peripheral input and phantom limb pain: A somatosensory event-related potential study

Following amputation, nearly all amputees report nonpainful phantom phenomena and many of them suffer from chronic phantom limb pain (PLP) and residual limb pain (RLP). The aetiology of PLP remains elusive and there is an ongoing debate on the role of peripheral and central mechanisms. Few studies have examined the entire somatosensory pathway from the truncated nerves to the cortex in amputees with PLP compared to those without PLP. The relationship among afferent input, somatosensory responses and the change in PLP remains unclear.

Transcutaneous electrical nerve stimulation was applied on the truncated median nerve, the skin of the residual limb and the contralateral homologous nerve in 22 traumatic upper-limb amputees (12 with and 10 without PLP). Using somatosensory event-related potentials, the ascending volley was monitored from the brachial plexus, the spinal cord, the brainstem and the thalamus to the primary somatosensory cortex.

Peripheral input could evoke PLP in amputees with chronic PLP (7/12), but not in amputees without a history of PLP (0/10). The amplitudes of the somatosensory components were comparable between amputees with and without PLP. In addition, evoked potentials from the periphery through the spinal, subcortical and cortical segments were not significantly associated with PLP.

Peripheral input can modulate PLP but seems insufficient to cause PLP. These findings suggest the multifactorial complexity of PLP and different mechanisms for PLP and RLP.

Peripheral afferent input plays a role in PLP and has been assumed to be sufficient to generate PLP. In this study we found no significant differences in the electrical potentials generated by peripheral stimulation from the truncated nerve and the skin of the residual limb in amputees with and without PLP. Peripheral input could enhance existing PLP but could not cause it. These findings indicate the multifactorial complexity of PLP and an important role of central processes in PLP.