Psychophysical correlates of phantom limb experience.

Phantom limb phenomena were correlated with psychophysiological measures of peripheral sympathetic nervous system activity measured at the amputation stump and contralateral limb. Amputees were assigned to one of three groups depending on whether they reported phantom limb pain, non-painful phantom limb sensations, or no phantom limb at all. Skin conductance and skin temperature were recorded continuously during two 30 minute sessions while subjects continuously monitored and rated the intensity of any phantom limb sensation or pain they experienced. The results from both sessions showed that mean skin temperature was significantly lower at the stump than the contralateral limb in the groups with phantom limb pain and non-painful phantom limb sensations, but not among subjects with no phantom limb at all. In addition, stump skin conductance responses correlated significantly with the intensity of non-painful phantom limb paresthesiae but not other qualities of sensation or pain. Between-limb measures of pressure sensitivity were not significantly different in any group. The results suggest that the presence of a phantom limb, whether painful or painless, is related to the sympathetic-efferent outflow of cutaneous vasoconstrictor fibres in the stump and stump neuromas. The hypothesis of a sympathetic-efferent somatic-afferent mechanism involving both sudomotor and vasoconstrictor fibres is proposed to explain the relationship between stump skin conductance responses and non-painful phantom limb paresthesiae. It is suggested that increases in the intensity of phantom limb paresthesiae follow bursts of sympathetic activity due to neurotransmitter release onto apposing sprouts of large diameter primary afferents located in stump neuromas, and decreases correspond to periods of relative sympathetic inactivity. The results of the study agree with recent suggestions that phantom limb pain is not a unitary syndrome, but a symptom class with each class subserved by different aetiological mechanisms.