Pain, 46 (1991) 9-12 0 1991 Elsevier Science Publishers B.V. 0304-3959/91/$03.50 ADONIS 030439599100146L PAIN 01814 Chronic pain: a PET study of the central effects of percutaneous high cervical cordotomy V. Di Piero ‘,*, A.K.P. Jones ‘, F. Iannotti 3,4, M. Powell 4, D. Perani 1,5, G.L. Lenzi 2 and R.S.J. Frackowiak ’ ’ zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA MRC Cyclotron Unit, Hammersmith Hospital, London W I2 OHS (U.K.), 2 Cattedra di Neuropatologia e Psicopatologia, Dipartimento di Scienze Neurologiche, Urkersitri di Roma ‘La Sapienza, ’ Rome (Italy), 3 Clinica Neurochirurgica, Unicersit; di Napoli, Naples (Italy), 4 Department of Neurosurgery, Middlesex Hospital, London (U.K.), and 5 Istituto di Tecnologie Biomediche Acanzate, Istituto Scientific0 H.S. Raffaele, Milan (Italy) (Received 12 October 1990, revision received 18 December 1990, accepted 18 December 1990) Summary We have studied 5 patients with unilateral, severe chronic pain due to cancer before and after percutaneous, ventrolateral cervical cordotomy to investigate the central effects of the procedure. The aim was to identify the functional anatomical correlates of abolishing unilateral nociceptive input to the brain. Patients were investigated by positron emission tomography using C”02 to evaluate cerebral blood flow. Comparisons were made between the patients with unilateral pain before cordotomy and normal volunteers. These demonstrated signifi- cantly less blood flow in 3 out of 4 of the individual quadrants of the hemithalamus contralateral to the side of pain (P < 0.01-0.05). These differences were abolished by cordotomy. Comparison of the patients before and after cordotomy showed a significant decrease in blood flow in the dorsal anterior quadrant of the thalamus contralateral to the side of pain (P < 0.05) which was normalised after cordotomy. There were no significant changes in the prefrontal or primary somatosensory cortex. We conclude that chronic pain results in a decrease of synaptic activity at thalamic level either from decreased activity in neurones projecting to that region and/or attenuated local neuronal firing. We have demonstrated no secondary remote effects in cortex, indicating the importance of subcortical mechanisms in central responses to chronic pain. Key words: Pain; Cerebral blood flow; Cordotomy; Thalamus; Positron emission tomography (PET) Introduction Antero-lateral cordotomy in the monkey produces degeneration in the ipsilateral nucleus ventralis pos- tero-lateralis (VPL) of the thalamus, in both intralami- nar nuclei and adjacent to the magnocellular part of the medial geniculate body [lo]. More recent quantita- tive studies in monkeys using retrograde transport of horseradish peroxidase have demonstrated extensive projections of the ventral spinothalamic tract to medial and lateral thalamic nuclei. 83% of these terminations were ipsilateral to the side of cordotomy. Cell degener- Correspondence to: Dr. A.K.P. Jones, MRC Cyclotron Unit, Ham- mersmith Hospital, Ducane Road, London W12 OHS, U.K. Tel.: 44 81 740 3172; Fax: 44 81 743 3987. ation of the VPL nucleus following interruption of the ipsilateral spinothalamic tract in man has also been described by Bowsher 131 who examined the brains of 4 patients who underwent open cordotomy for treatment of pain and subsequently died. He reported massive bilateral degeneration of cells in the medial reticular formation of the medulla and to a lesser extent of the pons and mesencephalon. He commented qualitatively on the relatively sparse degeneration seen in thalamus compared to the brain-stem. While these anatomical details are now better understood, how they relate to human pain perception is less clear. In particular, it is not known to what extent and at what level the cere- bral cortex is involved in an integrated response to pain. In the Penfield and Boldrey series of 426 sensory responses in conscious patients evoked by cortical stim- ulation, only 11 (2.5%) were described as painful [ll].