The neurobiology of chronic pain, including chronic back pain, is unknown. Structural imaging studies of the spine cannot explain all cases of chronic back pain. Functional brain imaging studies indicate that the brain activation patterns are different between chronic pain patients and normal subjects, and the thalamus, and prefrontal and cingulate cortices are involved in some types of chronic pain. Animal models of chronic pain suggest abnormal spinal cord chemistry. Does chronic pain cause brain chemistry changes? We examined brain chemistry changes in patients with chronic back pain using in vivo single- voxel proton magnetic resonance spectroscopy ((1)H-MRS). In vivo (1)H-MRS was used to measure relative concentrations of N-acetyl aspartate, creatine, choline, glutamate, glutamine, gamma-aminobutyric acid, inositol, glucose and lactate in relation to the concentration of creatine. These measurements were performed in six brain regions of nine chronic low back pain patients and 11 normal volunteers. All chronic back pain subjects underwent clinical evaluation and perceptual measures of pain and anxiety. We show that chronic back pain alters the human brain chemistry. Reductions of N-acetyl aspartate and glucose were demonstrated in the dorsolateral prefrontal cortex. Cingulate, sensorimotor, and other brain regions showed no chemical concentration differences. In chronic back pain, the interrelationship between chemicals within and across brain regions was abnormal, and there was a specific relationship between regional chemicals and perceptual measures of pain and anxiety. These findings provide direct evidence of abnormal brain chemistry in chronic back pain, which may be useful in diagnosis and future development of more effective pharmacological treatments.