CENTRAL PAIN IN THE ABSENCE OF FUNCTIONAL SENSORY THALAMUS

Since the pioneering publication by Dejerine and Roussy, the thalamus has been commonly implicated in the pathogenesis of central pain. It is well known that cerebral lesions that spare the thalamus as well as both small and large ones within that structure produce similar pain syndromes, but could such pain develop if the thalamus were completely destroyed? CT scans of our patients suggested that it could. Moreover, exhaustive physiological explorations of the thalamus with macro- and microstimulation and microelectrode recordings in 2 more patients in whom CT scans suggested thalamic preservation despite massive suprathalamic infarcts, for the purpose of carrying out deep brain stimulation (DBS), revealed no evidence of ascending or descending diencephalic function. Their sensory examinations were similar to those of patients with congenital hemiatrophy who had undergone hemispherectomy. In 1 case, microstimulation of periaqueductal grey-periventricular grey (PAG-PVG) on the unaffected side of the brain induced the spectrum of responses correctly associated with these areas and during acute stimulation nearly abolished the patient's allodynia and hyperpathia. These observations suggest that processes occurring at a subdiencephalic level (or possibly ipsilaterally) may be involved in the generation of pain. Stroke-induced central pain remains a therapeutic dilemma for the pain surgeon, and the mechanisms underlying this pain state remain a mystery. Since the early publication of Dejerine and Roussy, the thalamus has been thought to play an important role in the pathogenesis of central pain. In fact, in most cases of pain due to brain pathology, the term 'thalamic pain' is loosely applied, even though the lesions may be located in the cortex, subcortical white matter OT brainstem. We have had the opportunity to carry out detailed physiologic thalamic explorations in 2 patients with central pain secondary to large suprathalamic infarcts, in whom we could find no evidence of medial or ventrobasal thalamic function. This indicated that sites other than the ipsilateral cerebral cortex and sensory thalamus mediate certain central pain states.