Identifying brain nociceptive information transmission in patients with chronic somatic pain

Abstract

Introduction: Recent advances regarding mechanisms of chronic pain emphasize the role of corticolimbic circuitry in predicting risk for chronic pain, independently from site of injury-related parameters. These results compel revisiting the role of peripheral nociceptive signaling in chronic pain. We address this issue by examining what brain circuitry transmit information regarding the intensity of chronic pain and how this information may be related to a common co-morbidity, depression.

Methods: Resting state functional MRI was used in a large group of chronic pain patients (n=40 chronic back pain, CBP, and n=44 osteoarthritis, OA patients), and in comparison to healthy subjects (n=88). We used a graph theoretical measure, degree count, to investigate voxel-wise information sharing/transmission in the brain. Degree count, a functional connectivity based measure, identifies the number of voxels functionally connected to every given voxel. Subdividing the chronic pain cohort into discovery, replication, and also for overall group we show that only degree counts of diencephalic voxels centered in the ventral lateral thalamus reflected intensity of chronic pain, independently of depression.

Results: Pain intensity was reliably associated with degree count of the thalamus, which was correlated negatively with components of the default mode network and positively with the periaqueductal grey (in contrast to healthy controls). Depression scores were not reliably associated with regional degree count.

Conclusion: Collectively the results suggest that, across two types of chronic pain, nociceptive specific information is relayed through the spinothalamic pathway to the lateral thalamus, potentiated by pro-nociceptive descending modulation, and interrupting cortical cognitive processes.

Figure 1.

Depression and pain are not correlated in our patient group. (A) Distribution of VAS and BDI scores across all patients in the discovery group (CBP N = 20, OA N = 22) (left and middle panels), and their correlation (right panel) (r = 0.167, N = 42). (B) Group-average spatial distribution of connectivity degree (number of functional connections at each node). Blue represents relatively fewer connections; red indicates more connections. CBP, chronic back pain; OA, osteoarthritis.

Figure 2.

Functional connectivity degree in the thalamus predicts subjective pain. (A) Functional connectivity degree in the thalamus correlated most strongly to pain in our discovery group. Maps are thresholded at z > 2.3 and cluster-corrected for multiple comparisons at P < 0.05. (B) Higher thresholding of the z-stat correlation maps (z > 4.5) shown in A demonstrates that the highest correlations between connectivity degree and VAS are located in the thalamus. (C) Posthoc correlation between average degree within the map in A and pain is highly significant. (D) Pain scores in the replication group were predicted based on the relationship between pain and degree in the discovery group (as shown in panel C).

Figure 3.

Functional connectivity of the thalamus is disrupted in patients with pain. (A) The correlation map between degree and pain in the combined group analysis was thresholded at z > 4.5, and the remaining voxels (shown in green) were used as seeds to determine functional connectivity to the thalamus. (B) Results of a 2-group unpaired t test, comparing functional connectivity between patients and healthy controls (z > 2.3, cluster-corrected for multiple comparisons at P < 0.05). Red indicates voxels that had greater positive connectivity to the thalamus in patients, blue indicates greater positive connectivity in healthy controls. (C) The average z-stat values from the map shown in B and SE for each group. CBP, chronic back pain; OA, osteoarthritis.

Figure 4.

Thalamic regions with strongest relationship between pain intensity and functional connectivity (degree) exhibit structural connections to somatosensory, premotor, and primary motor cortical regions. (A) Segmentation (colored regions) of the thalamus is based on probabilistic diffusion tensor imaging (DTI) tractography, available as FSL's thalamic connectivity atlas. The black outlines indicate the thalamic regions with the highest relationship between degree and pain intensity (same as shown in A). (B) The table shows the correspondence of thalamic clusters to histologically defined locations of major nuclei and their probabilistic DTI connectivity to respective cortical regions. The 2 clusters outlined by the black line in A, are thought to include VL, VA, and VP nuclei and project mainly to somatosensory, primary motor, and premotor cortical regions. FSL, FMRIB's Software Library.