Functional MRI detection of hemodynamic response of repeated median nerve stimulation

Abstract

Median nerve stimulation is a commonly used technique in the clinical setting to determine areas of neuronal function in the brain. Neuronal activity of repeated median nerve stimulation is well studied. The cerebral hemodynamic response of the stimulation, on the other hand, is not very clear. In this study, we investigate how cerebral hemodynamics behave over time using the same repeated median nerve stimulation. Ten subjects received constant repeated electrical stimulation to the right median nerve. Each subject had functional magnetic resonance imaging scans while receiving said stimulations for seven runs. Our results show that the blood oxygen level-dependent (BOLD) signal significantly decreases across each run. Significant BOLD signal decreases can also be seen within runs. These results are consistent with studies that have studied the hemodynamic habituation effect with other forms of stimulation. However, the results do not completely agree with the findings of studies where evoked potentials were examined. Thus, further inquiry of how evoked potentials and cerebral hemodynamics are coupled when using constant stimulations is needed.

Figure 1

Inter-run habituation effect with median nerve stimulation. Only the very first run showed significant activation. Data was averaged across all ten subjects. A statistical threshold of Z = 3 (p = 0.0013, uncorrected) and a cluster size threshold of 240 mm³ was used to isolate activations. The right hemisphere is on the right and the left hemisphere is on the left.

Figure 2

The average volume of activation for seven runs of data acquisition. The activation volumes were measured in the left M1/S1 area. The activation map was generated using a statistical threshold of Z = 3 and a volume threshold of 240 mm³. The individual activation volumes across all subjects were identified and averaged. The error bars represent one standard error.

Figure 3

The average percent change in a 3×3×3 voxel region of interest in the S1/M1 area across 7 runs of all subjects. The error bars represent one standard error.

Figure 4

Split-half analysis was used to show intra-run habituation. Only the first run was used. The left image represents activation map processed with the first 2.5 cycle data of the first run. The right image is activation map processed with the last 2.5 cycle data of the first run. A statistical threshold of Z = 3 and a volume threshold of 240 mm³ was used. The right hemisphere is on the right and the left hemisphere is on the left.

Figure 5

The average volume of activation clusters found in the first and last 2.5 cycles of scan 1 (figure 4). The activation volumes were measured in the left M1/S1 area. The error bars represent one standard error.