Brain response to visceral aversive conditioning: a functional magnetic resonance imaging study.

BACKGROUND & AIMS Brain-imaging studies to date have confounded visceral pain perception with anticipation. We used functional magnetic resonance imaging of the human brain to study the neuroanatomic network involved in aversive conditioning of visceral pain and, thus, anticipation. METHODS Eight healthy volunteers (5 male) participated in the study. We used a classic conditioning paradigm in which 3 neutral stimuli (differently colored circles) that acted as conditioned stimuli were paired with painful esophageal distention, air puff to the wrist, or nothing, which acted as unconditioned stimuli. Neural activity was measured during learning, anticipation (pairing only 50% of conditioned stimuli with their unconditioned stimuli), and extinction (unpaired conditioned stimuli) phases. For magnetic resonance imaging, axial slices depicting blood oxygen level-dependent contrast were acquired with a 1.5-T system. RESULTS Neural responses during the learning phase included areas commonly associated with visceral pain (anterior cingulate cortex, insula, and primary and secondary somatosensory cortices) and innocuous somatosensory perception (primary and secondary somatosensory cortices and insula). During the anticipation and extinction phases of aversive stimulation, brain activity resembled that seen during actual painful esophageal stimulation. In contrast, anticipation and extinction of the innocuous somatic stimulus failed to show that effect. CONCLUSIONS We have shown that actual and anticipated visceral pain elicit similar cortical responses. These results have implications for the design and interpretation of brain-imaging studies of visceral pain. They not only contribute to our understanding of the processing of visceral pain, but also have clinical implications for the management of chronic pain states.