Unique, common, and interacting cortical correlates of thirst and pain

Abstract

This study used positron-emission tomography to establish the patterns of brain activity involved in the isolated and concurrent experiences of thirst and pain. Ten subjects were scanned while experiencing pain evoked with noxious pressure, while experiencing thirst after the infusion of hypertonic saline, and while experiencing pain when thirsty. After the onset of thirst, noxious pressure evoked more intense sensations of pain. Noxious pressure did not change subjective ratings of thirst. Thirst caused activation in the anterior cingulate (Brodmann area 32) and the insula. Enhanced pain responses were associated with increased activity in cortical regions that are known to correlate with pain intensity, and also with unique activity in the pregenual anterior cingulate and ventral orbitofrontal cortex. These findings suggest a role for limbic and prefrontal cortices in the modulation of pain during the experience of thirst.

Fig. 1.

Subjects were scanned on 13 occasions over ≈2 h. An infusion of hypertonic saline was commenced after the fourth scan and continued for 60 min. Noxious pressure was applied to the thumbnail during scans 2, 4, 9, and 10. Free access to water was given after scan 11. VAS, visual analogue scale; NRS, numerical rating scale.

Fig. 2.

Blood samples were drawn from subjects after scans 1, 5, 8, 11, and 12. Plasma Na and osmolality increased during the infusion of hypertonic saline and were relatively stable after scan 8. Decreases in the proportion of hematocrit reflected increased plasma volume associated with the fluid load. Thirst ratings increased throughout the infusion, diminishing dramatically after free access to water.

Fig. 3.

Activations for pain (blue regions), maximum thirst (green regions), and both conditions (red regions) in the anterior cingulate and insula cortices, shown on an average of the normalized magnetic resonance images for the 10 subjects. The vertical blue shaded regions in the time courses indicate scans during the pain stimuli. (A) Pain-related increases in the mid cingulate (arrow) cortex coincided with increasing intensity of pressure pain. (B) Pain-related signal change in the insula (arrow) did not show an increase at maximum thirst. (C) Pain and thirst activations were observed in medial and lateral mid cingulate cortex, respectively, with overlap at x = 8. (D) Regional CBF in the mid cingulate cortex (arrow) increased steadily in coincidence with the escalation of thirst ratings and decreased after drinking. (E) The rCBF in the posterior insula (arrow) also increased as thirst became more pronounced, but the post-drink reduction in rCBF in the insula was not as marked.

Fig. 4.

Activations associated with thirst when experiencing pain and with increased pain when thirsty demonstrated two distinct time courses. The vertical blue shaded regions in the time courses indicate the scans when noxious mechanical stimuli were applied. Activations having increased pain-related signal intensity during thirst were observed in the mid cingulate cortex (A, arrow) and contralateral S1 (B, arrow). By comparison, activations involving pain-related signal increases only during pain scans at maximum thirst were observed in the pregenual anterior cingulate cortex (C, arrow) and the ventral orbitofrontal cortex (D, arrow).