Human neurophysiology to diagnose, prevent and treat pain

Abstract

Neurophysiological techniques are useful to diagnose, predict and treat chronic pain, in particular of neuropathic origin. In this talk we shall succinctly review these different approaches, as well as the possibility they open to «back translation» of results gathered in humans to animal models. Definite diagnosis of neuropathic pain (NP) needs confirmation of a lesion or a disease of somatosensory systems, and neurophysiology is crucial to obtain such information. Lesions responsible for NP relate essentially to small fibres A-delta and C in the periphery, and the spino-thalamo-cortical system of pain/temperature at central levels. Among the methods to record responses in pain/temperature pathways, evoked potentials to laser (LEPs) or contact thermodes (CHEPs) are simple and reliable. Their diagnostic efficacy increases when they are coupled to vegetative reactions (SSR), psychophysical measures (reaction times, quantitative sensory tests), and concomitant exploration of non-nociceptive systems (large fibres, dorsal columns). The diagnostic workout should ideally respond to: (a) is there objective somatosensory impairment in painful regions? (b) which is the respective alteration of different sub-systems? (c) are the results concordant with imaging? and (d) how do results aid to patient management? Identification/prevention of neural lesions during surgery can minimise the development of postoperative pain. Joint analysis of nociceptive cortical responses and MRI allowed to classify ∼90% of patients with central pain post-thalamic stroke, and time-frequency LEP analysis may distinguish patients with and without pain following brainstem lesions. All these elements suggest that early neurophysiological studies of patients at risk may help to detect those with highest probability of developing NP, and hence open the way to anticipatory therapy. Recording of sympathetic skin or pupillary responses allows assessing the vegetative arousal provoked by a painful stimulus, indicating that the stimulus was actually perceived. Association and dissociation between vegetative and cortical responses are especially useful in the objective evaluation of allodynic phenomena and clinically dubious perception anomalies. Current trends aim at developing neurophysiological indexes of nociception in non-communicating patients (coma, general anaesthesia). Non-invasive cortical stimulation using magnetic repetitive impulses (rTMS) or scalp-applied currents (tDCS, tACS) are used not only to predict efficacy of subsequent neurosurgical stimulation, but also, increasingly, as stand-alone methods of pain relief. Recent developments allow certain of these techniques to be applied by the patients themselves at home, via internet-based interfacing and control from the hospital.