Thermal Stimulation Changes Diffusivity of the Spinothalamic Tract

Abstract

Study design: An experimental study.

Objective: This study aimed to investigate task-dependent changes in fractional anisotropy (FA) within the spinal cord during painful stimulation.

Summary of background data: Earlier experiments by Mandl et al (2008, 2013) used non-invasive functional diffusion tensor imaging (fDTI) to detect white matter fibers that were active during functional tasks. In two studies, it was observed that FA of involved white matter tracts exhibited repeatable task-related increases. In this study, we attempted to extend the fDTI work in the spinal cord.

Methods: Twenty-three healthy, right-handed men (mean age 22 yrs, standard deviation [SD] = 4) were invited to participate in this study. Diffusion-weighted images were collected over spinal levels C2 to T4 during a painful thermal stimulus applied to the left thenar eminence. In order to investigate task-related activity, FA values within the contralateral (right) spinothalamic tract were analyzed using a generalized estimating equations (GEE) procedure. As a control, we also examined activity in the ipsilateral and contralateral corticospinal tracts, which are not considered to be involved in nociception.

Results: Significant task-related decreases in FA were observed in the right spinothalamic tract at vertebral levels C2-C5 (Wald X(1) = 17.754, P < 0.001). There was no change in control regions at levels C7-T2 of the same tract, which are located below the level of input from dermatome C6, Wald X(1) = 0.185, P = 0.667. Results in all other regions assessed, that is, the left spinothalamic tract and bilateral corticospinal tract, were also not significant (P > 0.05).

Conclusion: The current findings suggest that task-related changes in FA associated with the transmission of pain signals along the spinal cord can be detected using fDTI. We observed decreased FA values in the contralateral (right) spinothalamic tract following painful stimulation, while no such activity was apparent in control regions.

Level of evidence: 5.

Figure 1

Functional task. Each run was approximately 10 minutes. The run was conducted four times, for a total task time of 40 minutes.

Figure 2

Ten-point time courses in the right spinothalamic tract at vertebral levels C2–C5 (top; corresponding to spinal levels C3–C5 and partial C6) and vertebral levels (bottom; corresponding to spinal levels partial C8) through T2 (bottom). Time courses were created as in Figure 2. Error bars represent 95% confidence intervals.

Figure 3

Ten-point time courses in left and right spinothalamic (red) and corticospinal (blue) tracts from vertebral levels C2–C5 (spinal levels C3–C5 and partial C6). Individual raw scores from 40 time points were smoothed with a 3-point moving average and converted into z-scores using each participant's mean and standard deviation. Ten-point time courses were created for each individual by calculating the median of four measurements for each run. The time courses were then averaged across participants and a cubic spline curve was fitted to the data. Error bars represent 95% confidence intervals.