MRI structural brain changes associated with sensory and emotional function in a rat model of long-term neuropathic pain

Abstract

In human conditions, chronic pain is associated with widespread anatomical changes in the brain. Nevertheless, little is known about the time course of these changes or the relationship of anatomical changes to perception and behaviour. In the present study, we use a rat model of neuropathic pain (spared nerve injury, SNI) and 7 T MRI to determine the longitudinal supraspinal changes associated with pain-like and anxiety-like behaviours. SNI rats and sham controls were scanned at seven time points, 1 week before surgery, 2 weeks after, and then once a month for 5 months. At each time point we performed behavioural tests, including thermal and mechanical sensitivity, and tests of locomotion and exploratory behaviour (open field and elevated plus maze). We found that SNI rats had early and sustained thermal and mechanical hyperalgesia, and developed anxiety-like behaviours several months after injury. Compared to sham controls, SNI rats had decreased frontal cortex volumes several months after surgery, coincident with the onset of anxiety-like behaviours. There was also decreased volume in retrosplenial and entorhinal cortices. We also explored areas that correlated with mechanical hyperalgesia and found that increased hyperalgesia was associated with decreased volumes in bilateral S1 hindlimb area, anterior cingulate cortex (ACC, areas 32 and 24), and insula. Overall, our results suggest that long-term neuropathic pain has widespread effects on brain anatomy related to the duration and magnitude of the pain.

Figure 1

Behavioural results. SNI rats showed pronounced immediate and sustained mechanical (A) and thermal (cold) (B) hyperalgesia. There was a significant group x time point interaction for locomotion in the open field, with SNI rats showing relatively less movement during middle, but not late time points (C). SNI rats also began to display anxiety-like behaviours several months after injury, demonstrated here (D) as decreased exits from the closed arms of the elevated plus maze. * p < 0.05 simple effects tests (shown only for C and D).

Figure 2

(A) Results from the linear mixed effects model showing areas of relative decreased volume in SNI rats compared to controls. Decreases were found in several cortical regions, summarized in Table 1. The plots to the right show the mean relative voxel sizes (or the Jacobian determinant mean values, which indicate an expansion (>1) or compression (<1) relative to the reference space) for all the significant (above a threshold of t=3.5) voxels, for each group plotted over time. Plots on the far right show the relative voxel size for each rat over time. (B) The results for one significant cluster from the whole brain analysis (prefrontal cortex, PFC, including secondary motor cortex (M2), anterior cingulate cortex area 24 (ACC (24)), and anterior cingulate cortex area 32 (ACC (32)). There were significant differences between SNI and sham at time points 19 and 24 weeks post-surgery, but not at earlier time points. RS, retrosplenial cortex. Ent, entorhinal cortex. Error bars are standard errors of the means. The vertical line indicates the approximate time the surgeries were performed. * p < 0.05 simple effects tests.

Figure 3

Results from the von Frey (mechanical hyperalgesia) analysis, showing (A) significant clusters where increasing mechanical hypersensitivity predicted decreased volume in rats with SNI. Six significant clusters were identified in the analysis and are shown in the figure. (B) The scatter plot for one region (right S1HL) is shown (mean relative voxel sizes, or the Jacobian determinant mean values, indicate an expansion (>1) or compression (<1) relative to the reference space) and a regression line is added for each rat. The von Frey scores for all time points post-surgery are included. For abbreviations see Table 2. Error bars are standard errors of the means.