Force-correlated activation of the motor cortex during isometric handgrip exercise
R.W. Reid, M.C. Delano, R.W. Wiseman, T.G. Cooper, and R.A.Meyer.
Abstract: Brain areas involved in handgrip force generation were identified by cross-correlation of fMRI voxel intensity vs. handgrip force recorded from a strain-gauge transducer. During 4.5 min isometric contractions in which subjects randomly varied the force, force-correlated activity was confined to the primary motor area. In contrast, during 30 s contractions alternated with 30 s rest periods, the intensity of activation was greater, and extended to both sensory and supplemental motor areas. The results suggest that force output per se is a significant, but minor determinant of the fMRI response during motor tasks.
Introduction: Many blood-oxygen-level-dependent (BOLD)-based fMRI studies have confirmed the location of the motor and sensory homunculi in the cortex of healthy subjects (e.g., 1). However, only a few studies have examined the relationship between the force of voluntary muscle contraction and the magnitude of the corresponding cortical BOLD response. Unfortunately, the results of these studies are not consistent. For example, in a recent study, Dai et al (2) reported a linear relationship between the force of isometric handgrip contractions vs. BOLD response in widespread cortical areas. In contrast, Ludman et al (3) found no force-correlated activation during a dynamic, handgrip weightlifting task. These differing results may arise from differences in the nature and complexity of the task. For example, in the Dai et al study, the subjects were asked to maintain a target force based on visual feedback, whereas no visual feedback is required during weightlifting exercise. Similarly, Thickbroom et al (4) showed that dynamic contractions elicit much greater responses in the sensorimotor areas than isometric contractions, but force pre se had little effect on the intensity of the response to either task. The purpose of this study was to re-examine the force dependence of cortical activity during a very simple motor task, in which no feedback was provided, and the dynamic component of the contractions was minimized.
Methods: The study was approved by the University Committee on Research Involving Human Subjects, and all subjects (n=14, age 19-55 years, 6 female) gave informed, written consent. Anatomical (SPGR, TR/TE=7.6/3.0, 30o pulse, 24 cm FOV, 124 x 1.5 mm slices) and one-shot echo-planar functional images (TR/TE= 3000/45, 24 cm FOV, 64x64 matrix, 24 x 7 mm slices) were acquired via the standard bird-cage head coil on a 1.5 T GE/CVi system. Subjects lay supine in the magnet with their right hand positioned over the lower abdomen. The force of handgrip contractions was measured with a custom-built isometric strain-gauge force transducer, connected via rf-filtered leads to the excitation and amplifer circuits of a Gould transducer amplifer. Each subject's maximum voluntary force (MVC) of handgrip contraction was measured before each protocol. FMRI data was motion corrected and analyzed using the AFNI package (5).
In protocol 1, echo-planar images and force were acquired while subjects contracted continuously for 4.5 min. Before this protocol, the subjects were instructed to "randomly change the force" of contraction every 30 s. In protocol 2, subjects performed four 30 s isometric contractions, flanked by 30 s rest periods, at each of 3 different MVC's (nominally 25%, 50%, and 100% MVC; 90 images/slice/force). Before each run, the subjects were allowed to practice achieving the target force with visual feedback, but feedback was removed during data acquisition. In both protocols, the only cue provided was a timer which alternated between red and green every 30 s. Two subjects participated in both protocols.
Results: Figure 1 shows a sample force record (top) from a subject in protocol 1, and the corresponding intensity changes (bottom) within an ROI identified by cross-correlation of voxel intensity vs. this force record. The ROI was confined to the primary motor area. Comparable results were obtained in 8 subjects, and there was no consistent activation of ROI's in other areas during this protocol.
Figure 2 compares the mean force (top, mean of 8 subjects) during the nominal 50% MVC run of protocol 2, and the mean intensity changes (bottom) in the force-correlated ROI in the motor cortex. Note that without visual feedback, force decreased during each 30 s contraction, and most subjects did not achieve the target force (actual forces were 20±2, 36±3, and 55±5 % MVC, x±SE. n=8). The ROI intensity change in protocol 2 was only weakly dependent on force (Linear Regression: % change in intensity = 0.0215 * (% MVC) + 0.56 %., n= 24, r=0.503).
Discussion: The results suggest that only in the primary motor area is the BOLD response directly correlated with muscle force output per se. During cyclic contractions, a substantial component of the activity observed in this area, and in other sensorimotor areas, must be related to other components of the task, e.g., force initiation, relaxation, dynamic motion, feedback regulation, etc.
References:
1.
Lotze, M., Erb, M., Flor, H., Huelsmann, E., Godde B., Grodd, W., Neuroimage, 11,
2000.
2.
Dai, H.,
Liu, J.Z., Sahgal, V., Brown, R.W.,Yue,
G.H., Exp. Brain Res., 140, 290, 2001.
3.
Ludman, C.N., Cooper,
T.G., Ploutz-Snyder, L.L., Potchen, E.J.,Meyer, R.A., NMR Biomed., 9, 228, 1996.
4.
Thickbroom, G.W.,
Phillips, B.A., Morris,
Cox, R.W.,Hyde, J.S., NMR
Biomed., 10,
171, 1997.
