Abstract
Transplantation of a donor hand has been successful as a surgical treatment following amputation, but little is known regarding the brain mechanisms contributing to the recovery of motor function. We report functional magnetic resonance imaging (fMRI) findings for neural activation related to actual and imagined movement, for a 54-year old male patient, who had received a donor hand transplant 50 years following amputation. Two assessments, conducted 3 months and 6 months post-operatively, demonstrate engagement of motor-control related brain regions for the transplanted hand, during both actual and imagined movement of the fingers. The intact hand exhibited a more intense and focused pattern of activation for actual movement relative to imagined movement, whereas activation for the transplanted hand was more widely distributed and did not clearly differentiate actual and imagined movement. However, the spatial overlap of actual-movement and imagined-movement voxels, for the transplanted hand, did increase over time to a level comparable to that of the intact hand. At these relatively early post-operative assessments, brain regions outside of the canonical motor-control networks appear to be supporting movement of the transplanted hand.
Introduction
After the first successful long-term outcome for a human hand transplant in 1998, continued surgical innovations have made transplantation of a donor limb a feasible treatment option following hand amputation (MacKay, Nacke, & Posner, 2014; Shores, Brandacher, Schneeberger, Gorantla, & Lee, 2010; Shores, Malek, Lee, & Brandacher, 2017). Little is known, however, regarding the neural mechanisms supporting the recovery of motor function following transplantation. Research with functional magnetic resonance imaging (fMRI) has shown that the neural representation of an amputated limb differs from that of an intact limb, as a result of cortical reorganization following a period of sensory deafferentation (Borsook et al., 1998; Cruz, Nunes, Reis, & Pereira, 2003; Kew et al., 1994; Makin et al., 2015). The neural representation of an amputated limb is typically more spatially diffuse that that of an intact limb and related functionally to non-motor regions. Following transplantation, the neural representation of a transplanted limb may gradually acquire the canonical neural signature of the intact limb (Brenneis et al., 2005; Giraux, Sirigu, Schneider, & Dubernard, 2001; Hernandez-Castillo, Aguilar-Castaneda, Iglesias, & Fernandez-Ruiz, 2016; Neugroschl et al., 2005; Valyear, Mattos, Philip, Kaufman, & Frey, 2019; Vargas et al., 2009). Post-operative cortical reorganization, however, will depend on many variables, including the age of the patient at injury, the time between amputation and transplantation, whether transplantation is unilateral or bilateral, and whether the surgery involves the transplantation of a donor limb or replantation of the patient’s own hand.
The activation of neural networks for imagined movement may be a useful biomarker of recovery of motor function and may even contribute to rehabilitation (Brenneis et al., 2005; Confalonieri et al., 2012; Neugroschl et al., 2005). Data from both neuroimaging and behavioral testing of healthy individuals, and from patients with peripheral and central nervous system damage, suggest that motor imagery engages a network of cortical, subcortical, and cerebellar regions that largely overlaps with the network for motor execution (Hardwick, Caspers, Eickhoff, & Swinnen, 2018; Jeannerod, 2001). That is, actual movement and imagined movement share a substantial degree of neural representation. Neugroschl et al. (2005) reported movement-related fMRI activation for a 27-year old man, both before and after unilateral hand transplantation (performed 2 years following amputation). Before transplantation, activation was present in left-hemisphere primary motor cortex during imagined movement for the amputated (right) hand, though it was less intense than the right motor cortex activation for actual movement of the intact (left) hand. Following transplantation, Neugroschl et al. tested activation related to actual movement, and found that contralateral motor cortex activation was present for the transplanted hand, at 2 months and 8 months post-operatively, though activation for imagined movement was not tested.
We report neural activation for actual and imagined finger movement, at two post-operative time points (3 months and 6 months) in a patient who received a hand transplant 50 years after amputation. To our knowledge, this is the longest time between amputation and hand transplantation reported in the scientific literature, and the first time that actual and imagined movement have been compared post-operatively. Our goal in this preliminary report is to characterize the pattern of neural activation associated with actual and imagined movement, assuming that the neural signature for actual movement would be involvement of primary motor cortex in the hemisphere contralateral to the moved hand, along with activation of ipsilateral cerebellum. Although the present case report is descriptive and qualitative, we hypothesized that for both the intact and transplanted hands, some activation of primary motor cortex during both actual and imagined movement would be detected. In addition, we expected that activation for primary motor cortex would be higher for actual movement than for imagined movement, and that this difference would be more clearly evident for the intact hand than for the transplanted hand.
Methods
Patient
The research reported here was conducted in accordance with the Code of Ethics of the World Medical Association (Declaration of Helsinki) for experiments involving humans. After informed consent and extensive preoperative education and evaluation, the patient was included in an institutional review board approved protocol (IRB # 00056079, clinicaltrials.gov # ).
At the time of hand transplantation, the recipient was a 54-year-old male who had sustained a traumatic amputation of his left hand in a meat grinder at the age of 4 years. No attempt at reconstruction was made and the patient underwent a below the elbow amputation (Figure 1A), with significant occupational and psychosocial impact. The patient reported not using a prosthesis previously; he had no medical problems, aside from occasional gastric reflux. Electrodiagnostic studies showed intact left median, ulnar, and radial motor innervation. The patient has not experienced phantom pain or phantom sensation.
Figure 1.
Patient left limb pre-transplantation (Panel A) and post-transplantation (Panel B).
Hand Transplantation
The donor and recipient were matched for blood type, sex, skin pigmentation, and size. The procurement of the limb was performed at the level of the elbow and transplanted at the mid-forearm. The transplant (Figure 1B) was performed following the standard of care for major limb replantation (Meyer, 1991). The patient received immunosuppression following the approved protocol (Cendales et al., 2018). The patient was discharged from the hospital on postoperative day (POD) 9 and started on an occupational therapy program on POD 11. The patient returned to regular work at month 4 post-transplant. At one year post-transplant, the patient had a moving two-point discrimination of >15mm in all digits. The patient’s Tinel’s sign (percussion over a nerve eliciting a sensation of tingling in the distribution of the nerve) was to the fingertips within the first year post-transplant. The Carroll test (Carroll, 1965), a quantitative evaluation of the upper extremity that includes the measurement of range of motion, pinch, grasp, and grip, improved from 9/99 pre-transplant to 32/99 at 18 months post-transplant. The patient reported performing activities such as fishing, riding a motorcycle, incorporating his left hand, and stated that the transplant had a major positive impact in his psychosocial health. At 18 months post-transplant, the patient reported normal sensation at the proximal aspect of the transplanted forearm and diminished sensation at the fingertips.
MRI Data Acquisition
Structural and functional imaging data were obtained at two time points: 3 months (Session 1) and 6 months (Session 2) post-transplant. Imaging was also conducted at a third time point, 1 year post-transplant, but due to technical difficulties, fewer functional imaging runs were obtained at this last time point, and thus only the data from Sessions 1 and 2 are reported here.
The imaging data were obtained on a 3.0 T GE MR750 whole-body MRI scanner (GE Healthcare, Waukesha, WI) equipped with a 60 cm bore, 50 mT/m gradients, and a 200 T/m/s slew rate. An eight-channel head coil was used for radio frequency (RF) reception. The patient wore earplugs to attenuate scanner noise and foam pads surrounded the head to reduce head motion. We first acquired 3-plane (straight axial/coronal/sagittal) localizer fast spin echo (FSE) images that defined a volume for data collection, using a semi-automated high-order shimming program to ensure global field homogeneity. This was followed by one run of T1-weighted anatomical imaging, one run of resting-state T2*-weighted (functional) imaging, four runs of T2*-weighted imaging, and 1 run of diffusion-weighted imaging (DWI).
For the T1-weighted anatomical images, 162 straight axial slices were acquired with a 3D fast inverse-recovery-prepared spoiled gradient recalled (SPGR) sequence, with TR = 8.21 ms, echo time (TE) = 3.22 ms, inversion recovery time (TI) = 450 ms, field of view (FOV) = 240 mm, flip angle = 12°, voxel size = 1 × 1 × 1 mm, 256 × 256 matrix, and a sensitivity encoding (SENSE) factor of 2, using the array spatial sensitivity encoding technique and extended dynamic range.
T2*-weighted echo-planar (EPI) functional imaging, sensitive to the blood-oxygen-level dependent (BOLD) signal, comprised 36 contiguous slices, acquired at an axial oblique orientation, parallel to the AC-PC plane; TR = 2000 ms, TE = 28 ms, FOV = 240 mm, flip angle = 90°, voxel size = 3.75 × 3.75 × 3.8 mm, 64 × 64 matrix, and a SENSE factor of 1. For each functional run, 200 brain volumes were collected, the first four volumes were not recorded to ensure steady state was reached.
Actual and Imagined Finger Movement Task
Four functional task runs were performed during each session: two actual movement runs, and two imagined movement runs, administered alternately. During each actual movement run, the order of the tasks was as follows: look at the fixation cross (rest condition), then rhythmically flex the fingers in the right hand, rest condition, then rhythmically flex the fingers of the left hand. The patient performed each of these tasks for 20 s, and repeated this sequence five times within each run, for a total of 10 task blocks per hand, and 10 rest blocks, across the two runs. The imagined movement runs were similar, except that the patient was instructed to imagine flexing his fingers, without the actual movement. We viewed the patient during finger movement to assess task compliance, but the pace of the movement was self-determined, and the quality of motor imagery was not assessed. The same tasks were administered for both sessions, with the order of the movement and imagine movement conditions counterbalanced between sessions.
fMRI Preprocessing and Modeling
The MRI images were inspected visually for artifacts and blurring. Initial preprocessing included assessment of data quality using an in-house tool that quantifies several metrics, including signal-to-noise ratio (SNR), signal-fluctuation-to-noise ratio (SFNR), participant motion, and measures of voxel-wise standard deviation (Friedman & Glover, 2006; Glover et al., 2012). Motion correction and high pass temporal filtering (cut off = 100.0 s) were conducted in FSL 5.0.1 (Smith et al., 2004; http://www.fmrib.ox.ac.uk/fsl) and FEAT version 6.0. Structural brain images were skull-stripped using the FSL brain extraction tool (Smith, 2002). Functional image processing included correction for slice-timing and head motion using six rigid-body transformations using FSL MCFLIRT (Jenkinson, Bannister, Brady, & Smith, 2002). The patient’s functional images were coregistered to the structural images in native space and then normalized to the MNI152 T1 template (Montreal Neurological Institute, Montreal, Canada) using a combination of affine and non-linear registrations (Greve & Fischl, 2009; Jenkinson et al., 2002; Jenkinson & Smith, 2001). The functional images were then spatially smoothed with a 5 mm Gaussian kernel.
Voxelwise analyses were conducted within FSL. The 20 s task periods (blocks) of actual movement, imagined movement, and rest, were modeled at the first level using a boxcar function convolved with a double gamma function. The design matrix included two independent events (right hand and left hand), as well as the six nuisance motion regressors. Within each session, activation for actual movement, for each hand, was modeled at the second level from the actual movement > rest contrast. Similarly, within each combination of session and hand, imagined-movement activation was modeled at the second level from the imagined movement > rest contrast. These two contrasts were combined at the second level to assess activation for the actual movement > imagined movement contrast. Note that in this latter contrast, the component effects of actual and imagined movement were each defined relative to rest.
The effects of interest (actual movement and imagined movement, each relative to rest, and actual movement > imagined movement) were analyzed with one sample t-tests using FMRIB Local Analysis of Mixed Effects (FLAME 1 & 2)(Beckmann, Jenkinson, & Smith, 2003; Woolrich, Behrens, Beckmann, Jenkinson, & Smith, 2004) for each session. Each of the contrasts was cluster thresholded at z > 3.0, GRF-corrected at p < .05. Clusters > 100 voxels were retained. The imagined movement > actual movement contrast was also tested but did not yield any significant activation.
Results
Actual Movement versus Rest
Activation for actual movement, relative to rest, is presented in Figure 2; additional details of the activation are presented in Table 1. The pattern of movement-related activation was consistent across the two sessions. In both sessions, actual movement of the right (intact) hand exhibited extensive activation of regions related to motor control. In Session 1 (Figure 2, Panel A), right-hand movement was associated with a large (8807 voxels) cluster of activation with a local maximum in the ipsilateral cerebellum. Two additional clusters (6735 and 3680 voxels) in the dorsal regions of the left hemisphere, with local maxima in the middle temporal gyrus and superior parietal lobule, included primary motor cortex. Ten additional clusters, each less than 500 voxels, with seven local maxima in the left hemisphere, were active during actual movement in Session 1. The activation for actual movement of the right hand in Session 2 (Figure 2, Panel B) exhibited a similar pattern. The local maximum for the largest cluster (5010 voxels) was in the superior parietal lobule of the left hemisphere, but activation extended into left motor cortex. Other large clusters include ipsilateral cerebellum (4108 and 792 voxels), although a cluster of 1799 voxels in the left cerebellum was also present. The additional nine clusters, all less than 1000 voxels each, exhibited six local maxima in the left hemisphere.
Figure 2.
Activation for actual movement, relative to rest, at Session 1 (Panel A) and Session 2 (Panel B). The left hand is the transplanted hand. Local maxima are presented in Table 1. Activation is overlaid on the Montreal Neurological Institute (MNI) template, with axial slices from −30 mm to +72 mm in the z plane. Images are presented in neurological orientation, with left = left.
Table 1.
Activation for Actual Movement
| Voxels | max z | x | y | z | Hem | BA | Location |
|---|---|---|---|---|---|---|---|
| Session 1; Right Hand Actual Movement > Rest | |||||||
| 8807 | 6.5 | 18 | −50 | −28 | R | Cerebellum | |
| 6735 | 7.4 | −28 | −56 | 68 | L | 7 | Superior Parietal Lobule |
| 3680 | 6.1 | −66 | −22 | −10 | L | 21 | Middle Temporal Gyrus |
| 435 | 4.56 | −24 | 12 | 58 | L | 6 | Superior Frontal Gyrus |
| 284 | 4.8 | −46 | 28 | 16 | L | 46 | Middle Frontal Gyrus |
| 237 | 4.37 | 44 | −24 | 26 | R | 13 | Insula |
| 216 | 4.44 | 34 | 2 | 60 | R | 6 | Middle Frontal Gyrus |
| 171 | 4.28 | −28 | 50 | 2 | L | 10 | Middle Frontal Gyrus |
| 166 | 4.63 | −32 | −44 | −14 | L | 37 | Fusiform Gyrus |
| 146 | 3.89 | −40 | 10 | 18 | L | 13 | Insula |
| 120 | 4.76 | −52 | −66 | −32 | L | Cerebellum | |
| 119 | 3.83 | 34 | −48 | 34 | R | 39 | Angular Gyrus |
| 119 | 3.99 | −4 | −28 | 30 | L | 23 | Cingulate Gyrus |
| Session 1; Left Hand Actual Movement > Rest | |||||||
| 2134 | 5.46 | 42 | 42 | 10 | R | 10 | Middle Frontal Gyrus |
| 2081 | 5.85 | 58 | −22 | −26 | R | 20 | Fusiform Gyrus |
| 1130 | 5.47 | 14 | −22 | 72 | R | 6 | Medial Frontal Gyrus |
| 1054 | 5.54 | −30 | −40 | −36 | L | Cerebellum | |
| 749 | 5.13 | −38 | −54 | 46 | L | 40 | Inferior Parietal Lobule |
| 741 | 6.03 | −60 | 30 | −6 | L | 45 | Inferior Frontal Gyrus |
| 735 | 5.58 | −46 | 14 | 24 | L | 9 | Inferior Frontal Gyrus |
| 504 | 5.56 | 18 | −100 | −18 | R | Cerebellum | |
| 451 | 4.75 | 62 | −14 | 10 | R | 42 | Transverse Temporal Gyrus |
| 424 | 6.77 | −2 | 40 | 36 | L | 8 | Medial Frontal Gyrus |
| 330 | 4.91 | −46 | −52 | 20 | L | 22 | Superior Temporal Gyrus |
| 313 | 4.38 | 44 | −60 | 46 | R | 7 | Inferior Parietal Lobule |
| 306 | 4.06 | 60 | −44 | 36 | R | 40 | Supramarginal Gyrus |
| 143 | 4.67 | 40 | −68 | −46 | R | Cerebellum | |
| 125 | 3.82 | −2 | −18 | 8 | L | Thalamus | |
| 106 | 5.1 | 60 | 44 | −18 | R | 47 | Middle Frontal Gyrus |
| 104 | 4.07 | 18 | 52 | 32 | R | 9 | Superior Frontal Gyrus |
| Session 2; Right Hand Actual Movement > Rest | |||||||
| 5010 | 7.87 | −30 | −58 | 64 | L | 7 | Superior Parietal Lobule |
| 4108 | 7.05 | 34 | −38 | −36 | R | Cerebellum | |
| 1799 | 5.44 | −42 | −40 | −24 | L | Cerebellum | |
| 904 | 4.78 | 4 | −80 | 4 | R | 18 | Lingual Gyrus |
| 792 | 6.29 | 12 | −62 | −50 | R | Cerebellum | |
| 764 | 5.74 | −2 | −102 | −10 | L | 18 | Lingual Gyrus |
| 436 | 5.35 | −14 | −24 | 0 | L | Thalamus | |
| 357 | 5.29 | −70 | −18 | −10 | L | 21 | Middle Temporal Gyrus |
| 253 | 5.6 | −36 | −84 | 20 | L | 19 | Middle Occipital Gyrus |
| 214 | 4.18 | −26 | −48 | −34 | L | Cerebellum | |
| 156 | 4.88 | 0 | −14 | 50 | C | 31 | Paracentral Lobule |
| 109 | 4.96 | −38 | −36 | 4 | L | 41 | Superior Temporal Gyrus |
| 105 | 4.34 | 50 | −24 | 38 | R | 2 | Postcentral Gyrus |
| Session 2; Left Hand Actual Movement > Rest | |||||||
| 3042 | 6.37 | −2 | 40 | 36 | L | 8 | Medial Frontal Gyrus |
| 993 | 6.22 | −48 | −58 | 28 | L | 39 | Middle Temporal Gyrus |
| 864 | 4.82 | −16 | −76 | −34 | L | Cerebellum | |
| 776 | 5.78 | −26 | −38 | −34 | L | Cerebellum | |
| 661 | 5.83 | 20 | −26 | 16 | R | Thalamus | |
| 661 | 4.81 | 34 | −24 | 56 | R | 4 | Precentral Gyrus |
| 507 | 6.24 | 46 | 30 | 26 | R | 9 | Middle Frontal Gyrus |
| 178 | 5.01 | −68 | −54 | −14 | L | 20 | Inferior Temporal Gyrus |
| 149 | 3.78 | −40 | −78 | −22 | L | Cerebellum | |
Note. The left hand is the transplanted hand. Voxels = number of voxels in cluster; max z = highest z value within each cluster; Hem = hemisphere; R = right; L = left; C = center; x, y, z = coordinates within Montreal Neurological Institute (MNI) normalized space; BA = Brodmann area. Activation is thresholded at Z = 3.0, Gaussian random field (GRF)-corrected at p < 0.05.
Actual movement of the left (transplanted) hand was associated with activation of right-hemisphere regions related to motor control, as well as ipsilateral cerebellum, though the activation was less extensive spatially than that of the intact hand. In Session 1 (Figure 2, Panel A), actual movement of the left hand was associated with activation of the ipsilateral cerebellum (1054 voxels), and three large clusters in the right hemisphere (1130-2134 voxels). One of these clusters, with a local maximum in the medial frontal gyrus, included right motor cortex, but the local maxima for the other two of these clusters were the right middle frontal gyrus and right fusiform gyrus. Additional activation for left hand actual movement in Session 1 comprised 13 clusters, each less than 1000 voxels, with seven of these in the right hemisphere. In Session 2 (Figure 2, Panel B), movement-related activation for the left hand did exhibit the expected activation of motor-control regions, in the form of three clusters in ipsilateral cerebellum (149-864 voxels), and a cluster in the right motor cortex (precentral gyrus; 661 voxels). However, the most extensive movement-related activation for the left hand in Session 2 was a large cluster (3042 voxels) in left prefrontal cortex, with a local maximum in the left medial frontal gyrus. Four additional clusters were present, with two of the local maxima in the right hemisphere.
Imagined Movement versus Rest
Activation for imagined movement, relative to rest, is presented in Figure 3, with additional details in Table 2. For the right hand, activation in Session 1 (Figure 3, Panel A) included a left-hemisphere cluster in primary motor cortex (685 voxels) and ipsilateral cerebellum (1280 voxels). However, extensive activation was also present bilaterally, for occipital (3112 voxels), parietal (1653 voxels), and superior temporal (1265 voxels) regions, with additional activation for eight smaller (each < 1000 voxels) clusters. In Session 2 (Figure 3, Panel B), the imagined-movement related activation for the right hand was relatively focused on the left-hemisphere motor-control regions, with clusters centered on the precentral gyrus (1057 voxels) and ipsilateral cerebellum (763 voxels). The only additional activation is a small cluster of 106 voxels with a local maximum in the left middle frontal gyrus.
Figure 3.
Activation for imagined movement, relative to rest, at Session 1 (Panel A) and Session 2 (Panel B). The left hand is the transplanted hand. Local maxima are presented in Table 2. Activation is overlaid on the Montreal Neurological Institute (MNI) template, with axial slices from −30 mm to +72 mm in the z plane. Images are presented in neurological orientation, with left = left.
Table 2.
Activation for Imagined Movement
| Voxels | max z | x | y | z | Hem | BA | Location |
|---|---|---|---|---|---|---|---|
| Session 1; Right Hand Imagined Movement > Rest | |||||||
| 3112 | 5.93 | 6 | −86 | 20 | R | 18 | Cuneus |
| 1653 | 5.69 | 36 | −52 | 58 | R | 7 | Superior Parietal Lobule |
| 1382 | 5.82 | −38 | −44 | 54 | L | 40 | Inferior Parietal Lobule |
| 1280 | 4.84 | −40 | −94 | −18 | L | Cerebellum | |
| 1265 | 5.1 | 60 | 0 | −4 | R | 22 | Superior Temporal Gyrus |
| 786 | 6.34 | −44 | −54 | −40 | L | Cerebellum | |
| 685 | 4.58 | 32 | −12 | 46 | R | 4 | Precentral Gyrus |
| 674 | 5.57 | 36 | 54 | 20 | R | 10 | Middle Frontal Gyrus |
| 397 | 4.11 | 58 | −18 | −24 | R | 21 | Middle Temporal Gyrus |
| 174 | 3.74 | −6 | −44 | 70 | L | 5 | Postcentral Gyrus |
| 162 | 4.83 | 24 | −52 | −52 | R | Cerebellum | |
| 154 | 4.31 | −20 | −86 | 30 | L | 18 | Cuneus |
| 129 | 4.44 | −54 | 6 | 0 | L | 22 | Superior Temporal Gyrus |
| 127 | 5.13 | −50 | −20 | 18 | L | 13 | Insula |
| Session 1; Left Hand Imagined Movement > Rest | |||||||
| 228 | 5.3 | −30 | −40 | −36 | L | Cerebellum | |
| 171 | 4.02 | 34 | −26 | 48 | R | 4 | Precentral Gyrus |
| 105 | 4.88 | 20 | −24 | 74 | R | 4 | Precentral Gyrus |
| Session 2; Right Hand Imagined Movement > Rest | |||||||
| 1057 | 5.89 | −18 | −12 | 74 | L | 6 | Precentral Gyrus |
| 763 | 5.9 | 20 | −46 | −32 | R | Cerebellum | |
| 106 | 4.4 | −28 | 66 | 2 | L | 10 | Middle Frontal Gyrus |
| Session 2; Left Hand Imagined Movement > Rest | |||||||
| 1830 | 6.85 | 30 | −18 | 68 | R | 4 | Precentral Gyrus |
| 526 | 7.35 | −22 | −40 | −32 | L | Cerebellum | |
| 324 | 4.38 | −48 | 16 | 34 | L | 9 | Middle Frontal Gyrus |
| 173 | 4.53 | 18 | −12 | 22 | R | Caudate | |
| 156 | 4.64 | −18 | −18 | 22 | L | Caudate | |
| 145 | 5.15 | 4 | −20 | 48 | R | 31 | Paracentral Lobule |
Note. The left hand is the transplanted hand. Voxels = number of voxels in cluster; max z = highest z value within each cluster; Hem = hemisphere; R = right; L = left; C = center; x, y, z = coordinates within Montreal Neurological Institute (MNI) normalized space; BA = Brodmann area. Activation is thresholded at Z = 3.0, Gaussian random field (GRF)-corrected at p < 0.05.
For the left hand, imagined-movement related activation was also evident in Session 1 (Figure 3, Panel A). This activation was limited to three small clusters (105-228 voxels) in the right precentral gyrus and ipsilateral cerebellum. In Session 2 (Figure 3, Panel B), activation for imagined movement of the left hand included a large cluster (1830 voxels) in the right precentral gyrus and a smaller cluster (526 voxels) in the ipsilateral cerebellum. Four additional clusters (145-324 voxels) reflected activation of the caudate bilaterally, left middle frontal gyrus and right paracentral lobule.
Actual Movement versus Imagined Movement
The degree to which activation related to actual movement exceeds that related to imagined movement is presented in Figure 4, with the additional details of the activation in Table 3. For the actual movement > imagined movement contrast, the component effects for actual and imagined movement were each defined relative to rest. Across both sessions, for the right hand, activation of left hemisphere motor-control regions was more prominent for actual movement than for imagined movement. In Session 1 (Figure 4, Panel A), this contrast yielded eight, relatively small clusters of activation, each less than 1000 voxels. Individual clusters, however, were located in the pre- and postcentral gyri of the left hemisphere, and the cerebellum bilaterally. In Session 2 (Figure 4, Panel B), the actual > imagined contrast for the right hand yielded two large clusters. The local maximum for the first cluster (3579 voxels) was located in the right fusiform gyrus but extended inferiorly to include the right cerebellum. The second cluster (3157 voxels) was centered on the left superior parietal lobule but extended to include primary motor cortex. The additional eight activations for this contrast were each less than 250 voxels and represented thalamic and posterior frontal, temporal, and parietal regions.
Figure 4.
Activation for actual movement > imagined movement at Session 1 (Panel A) and Session 2 (Panel B). The left hand is the transplanted hand. Local maxima are presented in Table 3. For the actual movement > imagined movement contrast, the component effects for actual and imagined movement were each defined relative to rest. Activation is overlaid on the Montreal Neurological Institute (MNI) template, with axial slices from −30 mm to +72 mm in the z plane. Images are presented in neurological orientation, with left = left.
Table 3.
Activation for Actual Movement > Imagined Movement
| Voxels | max z | x | y | z | Hem | BA | Location |
|---|---|---|---|---|---|---|---|
| Session 1; Right Hand Actual Movement > Imagined Movement | |||||||
| 805 | 5.19 | −46 | −40 | 60 | L | 1 | Postcentral Gyrus |
| 390 | 4.12 | −10 | −42 | −4 | L | Cerebellum | |
| 210 | 4.68 | −28 | 48 | 2 | L | 10 | Superior Frontal Gyrus |
| 202 | 4.23 | 18 | −50 | −26 | R | Cerebellum | |
| 172 | 4.57 | −4 | −30 | 30 | L | 23 | Cingulate Gyrus |
| 159 | 4.63 | −38 | −20 | 68 | L | 6 | Precentral Gyrus |
| 159 | 4.32 | 42 | −52 | −42 | R | Cerebellum | |
| 109 | 3.67 | 30 | −96 | −16 | R | 18 | Fusiform Gyrus |
| Session 1; Left Hand Actual Movement > Imagined Movement | |||||||
| 1807 | 5.48 | 54 | −12 | −34 | R | 20 | Inferior Temporal Gyrus |
| 298 | 4.43 | −52 | 48 | −18 | L | 11 | Middle Frontal Gyrus |
| 269 | 4.55 | 62 | −14 | 8 | R | 41 | Superior Temporal Gyrus |
| 137 | 4.34 | 62 | −52 | −16 | R | 37 | Fusiform Gyrus |
| 128 | 4.79 | −60 | −8 | −20 | L | 21 | Middle Temporal Gyrus |
| Session 2; Right Hand Actual Movement > Imagined Movement | |||||||
| 3579 | 6.04 | 60 | −56 | −4 | R | 37 | Fusiform Gyrus |
| 3157 | 6.8 | −30 | −60 | 60 | L | 7 | Superior Parietal Lobule |
| 243 | 5.65 | −22 | −2 | −18 | L | 53 | Amygdala |
| 191 | 4.39 | −20 | −26 | 4 | L | 50 | Thalamus |
| 170 | 4.03 | −2 | −74 | 20 | L | 18 | Cuneus |
| 143 | 3.88 | −34 | −82 | 28 | L | 19 | Orbital Gyrus |
| 134 | 4.61 | −60 | −58 | 10 | L | 39 | Middle Temporal Gyrus |
| 114 | 4.11 | 36 | −74 | 22 | R | 19 | Orbital Gyrus |
| 111 | 3.61 | 10 | −50 | 38 | R | 31 | Precuneus |
| 110 | 4.5 | 46 | −46 | −60 | R | Cerebellum | |
| Session 2; Left Hand Actual Movement > Imagined Movement | |||||||
| 481 | 5.14 | −36 | 10 | 56 | L | 6 | Middle Frontal Gyrus |
| 261 | 4.24 | −36 | 46 | 2 | L | 46 | Middle Frontal Gyrus |
| 154 | 4.17 | −10 | −90 | −24 | L | Cerebellum | |
| 115 | 4.25 | −10 | −50 | −52 | L | Cerebellum | |
| 104 | 3.86 | 14 | −78 | 48 | R | 7 | Precuneus |
| 101 | 3.89 | 0 | −34 | −4 | C | Cerebellum | |
| 100 | 4.1 | −64 | −48 | −6 | L | 37 | Fusiform Gyrus |
Note. The left hand is the transplanted hand. For the actual movement > imagined movement contrast, the component effects for actual and imagined movement were each defined relative to rest. Voxels = number of voxels in cluster; max z = highest z value within each cluster; Hem = hemisphere; R = right; L = left; C = center; x, y, z = coordinates within Montreal Neurological Institute (MNI) normalized space; BA = Brodmann area. Activation is thresholded at Z = 3.0, Gaussian random field (GRF)-corrected at p < 0.05.
For the left hand, activation was detected for actual movement relative to imagined movement, but the associated regions were not consistent across the sessions and did not include right-hemisphere primary motor cortex. In Session 1 (Figure 4, Panel A), five clusters were significant for this contrast, but the largest one (1807 voxels) was centered on the inferior temporal gyrus in the right hemisphere. The activation for this contrast in Session 2 (Figure 4, Panel B) yielded seven clusters, each less than 500 voxels, but two of them included the left cerebellum.
To examine the shared activation between the actual and imagined movement conditions, we calculated the percentage of actual-movement voxels that were shared with imagined-movement voxels. For each combination of hand, task condition, and session, we summed the voxels, regardless of location, above our threshold of z > 3.0, relative to rest (Table 4). We then identified those voxels that were common to the actual movement > rest and imagined movement > rest contrasts. These common or overlap voxels are expressed as a percentage of the actual-movement activated voxels in Figure 5. For the right hand, the percentage of actual-movement voxels shared with imagined-movement voxels remained relatively constant across the two sessions at 12-14%. For the left hand, the percentage of overlap voxels the increased from 4.22% in Session 1 to 12.36% in Session 2. Thus, although the left hand was not associated with consistently higher levels of activation for actual relative to imagined movement (Figure 4), the brain regions associated with these two task conditions exhibited increased spatial overlap from Session 1 to Session 2 (Figure 5).
Table 4.
Overlap of Activation for Actual and Imagined Movement
| Session 1 | Session 2 | |
|---|---|---|
| Right (intact) Hand | ||
| Actual Movement | 21655 | 13417 |
| Imagined Movement | 12162 | 2101 |
| Overlap | 2760 | 1857 |
| Left (transplanted) Hand | ||
| Actual Movement | 12208 | 14470 |
| Imagined Movement | 872 | 3431 |
| Overlap | 515 | 1788 |
Note. Values are the number of activated voxels, for all brain regions, relative to rest.
Figure 5.
For the actual movement > rest contrast, the percentage of activated voxels (at z = 3.0 threshold), that are shared with those activated in the imagined movement > rest contrast.
Discussion
These fMRI data demonstrate neural activation related to motor control, within contralateral primary motor cortex and ipsilateral cerebellum, for a transplanted hand, at 3 months and 6 months post-operatively. These findings confirm previous reports that representation of the transplanted hand is integrated into the neural networks of motor control, and that the canonical neural signature of hand and finger movement are re-established (Brenneis et al., 2005; Giraux et al., 2001; Hernandez-Castillo et al., 2016; Neugroschl et al., 2005; Valyear et al., 2019; Vargas et al., 2009). The present findings show, for the first time, that movement-related neural activation within motor control regions can be detected, for a transplanted hand, even when the transplantation follows an amputation conducted 50 years previously.
Although movement-related activation was evident for the transplanted hand, the pattern of activation differed qualitatively from that of the intact hand. When the patient moved the fingers of his right (intact) hand, the neural activation included contralateral primary motor cortex and the ipsilateral cerebellum, as expected. Although additional activation occurred for movement of the intact hand, the most prominent activations were located in motor-control regions and were relatively constant across sessions. Critically, movement-related activation for the left (transplanted) hand also occurred in both assessments, with the expected neural signature, but the activation was less intense and les extensive spatially than that of the intact hand (Figure 2). Further, imagined movement of the intact hand included motor-control regions, and the activation appeared to center more selectively on these regions in Session 2 than in Session 1 (Figure 3). Neural activation for imagined movement of the transplanted hand also included motor-control regions at both sessions. The difference, however, was that for the intact hand, the magnitude of the activation for movement-relevant regions was higher for actual movement than for imagined movement, particularly in Session 2, whereas for the transplanted hand the magnitude of the neural activation patterns for actual and imagined movement were similar. As a result, the actual > imagined contrast yielded higher movement-related activation for the intact hand than for the transplanted hand (Figure 4). However, the spatial overlap of actual-movement and imagined-movement voxels, for the transplanted hand, did increase over time to a level comparable to that of the intact hand (Figure 5). Valyear et al. (2019) obtained a similar pattern for grasp-related activation following unilateral hand transplantation. These authors found that, over the course of 41 months post-transplant, the spatial overlap between the patient’s grasp-related activation and that of healthy control participants increased, reflecting the increased normalization of the neural control of the transplanted hand.
Our assessments at 3 and 6 months post-transplantation are still relatively early in the recovery and rehabilitation process, and the fMRI activation measures reflect the combined influences of developing motor control and sensory feedback, which cannot be distinguished in this movement task. In a study of sensory functioning in a unilateral hand transplant patient, Frey et al. (2008) reported that palmar tactile stimulation of the transplanted hand, at 4 months post-transplant, was associated with neural activation that was indistinguishable in location and amplitude from that of healthy controls. This sensory-related activation occurred in the presence of limited sensitivity of the transplanted hand, suggesting that recovery of the original brain regions for hand sensation may precede the emergence of sensitivity in the transplanted hand. Little is known regarding the time course that is required for the development of normal sensory and motor functioning of a transplanted hand. Valyear et al. (2019) noted that reach-to-grasp kinematics required 41 months following unilateral transplantation before the associated fMRI activation pattern resembled that of healthy individuals. Following bilateral transplantation, cortical reorganization may not be complete at 51 months (Vargas et al., 2009).
Overall, the data confirmed our initial hypotheses, based on previously reported findings (Neugroschl et al., 2005), that the activation would be greater for actual movement than for imagined movement, and that the actual > imagined movement contrast would be more prominent for the intact hand than for the transplanted hand. These results suggest that although the contralateral primary motor cortex and ipsilateral cerebellum are establishing functional connections to the new left hand, other brain regions are supporting left-hand movement. These regions are distributed widely throughout the brain and, for this patient, are also involved in imagined movement. We predict that with additional time and rehabilitation, the level of actual-movement activation for the transplanted hand will more closely approximate that of the intact hand, thus leading to a measurable difference between actual and imagined movement for the transplanted hand.
The present study has several limitations, beyond those associated inherently with a case study. Our fMRI assessments for this patient were only initiated post-operatively, and ideally the post-operative fMRI data would be compared to a pre-operative baseline. In addition, the finger movement was self-paced. Thus, differences in movement-related activation between the hands likely reflect some difference in the pacing of the movement. Similarly, although we did not observe movement of the patient’s fingers during the imagine condition, it is possible that the activation of movement-related regions during this condition reflects subtle movement of the fingers. We do not have a measure of the patient’s ability to imagine movement. Finally, the re-establishment of the neural signature of movement-related activation, following hand transplantation, is ultimately dependent on changes in functional, and possibly structural, connectivity among brain regions (Hernandez-Castillo et al., 2016; Makin et al., 2015; Melton et al., 2016). The present findings, which characterize the location, magnitude, and spatial extent of movement-related activation should be followed by investigation of changes in structural and functional brain connectivity, over time, for this patient.
Conclusions
This fMRI study demonstrated that the neural signature for finger and hand movement can be detected 3 months post-operatively, for hand transplantation occurring 50 years following amputation. In this patient, imagined movement also activated motor-control related regions. The pattern of activation, however, differed markedly between the intact and transplanted hands. The intact hand exhibited substantially greater activation in motor-control regions for actual movement relative to imagined movement. For the transplanted hand, activation of motor-control regions during actual movement was less extensive, and activation in several additional cortical and deep gray matter regions occurred in both the actual and imagined movement conditions. As recovery and rehabilitation continues for this patient, we expect the pattern of movement-related brain activation for the transplanted hand to develop increasing similarity to that of the intact hand, as well as increasing differentiation of actual and imagined movement.
Acknowledgments
The authors wish to acknowledge the Duke VCA Program team for their technical assistance.
Funding
This work was supported by National Institutes of Health (NIH) research grants R01 AG039684 (DJM), R56 AG052576 (DJM), R01 AG042599 (JNB), R01 HL130443 (JNB), a grant from the Department of Defense W81XWH-12-2-0058 (LCC), and the Duke Health Scholar Award (LCC).
Footnotes
Disclosure statement
The authors report no conflicts of interest.
References
- Beckmann CF, Jenkinson M, & Smith SM (2003). General multilevel linear modeling for group analysis in FMRI. Neuroimage, 20(2), 1052–1063. doi: 10.1016/S1053-8119(03)00435-X [DOI] [PubMed] [Google Scholar]
- Borsook D, Becerra L, Fishman S, Edwards A, Jennings CL, Stojanovic M, Papinoclas L, Ramachandran VS, Gonzalez RG, & Breiter H (1998). Acute plasticity in the human somatosensory cortex following amputation. Neuroreport, 9(6), 1013–1017. doi: 10.1097/00001756-199804200-00011 [DOI] [PubMed] [Google Scholar]
- Brenneis C, Loscher WN, Egger KE, Benke T, Schocke M, Gabl MF, Wechselberger G, Felber S, Pechlaner S, Margreiter R, Piza-Katzer H, & Poewe W (2005). Cortical motor activation patterns following hand transplantation and replantation. The Journal of Hand Surgery: British & European Volume, 30(5), 530–533. doi: 10.1016/j.jhsb.2005.05.012 [DOI] [PubMed] [Google Scholar]
- Carroll D (1965). A quantitative test of upper extremity function. Journal of Chronic Diseases, 18(5), 479–491. doi: 10.1016/0021-9681(65)90030-5 [DOI] [PubMed] [Google Scholar]
- Cendales LC, Ruch DS, Cardones AR, Potter G, Dooley J, Dore D, Orr J, Ruskin G, Song M, Chen DF, Selim MA, & Kirk AD (2018). De novo belatacept in clinical vascularized composite allotransplantation. American Journal of Transplantation, 18(7), 1804–1809. doi: 10.1111/ajt.14910 [DOI] [PubMed] [Google Scholar]
- Confalonieri L, Pagnoni G, Barsalou LW, Rajendra J, Eickhoff SB, & Butler AJ (2012). Brain activation in primary motor and somatosensory cortices during motor imagery correlates with motor imagery ability in stroke patients. International Scholarly Research Notices: Neurology, 2012, 613595–613595. doi: 10.5402/2012/613595 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Cruz VT, Nunes B, Reis AM, & Pereira JR (2003). Cortical remapping in amputees and dysmelic patients: a functional MRI study. NeuroRehabilitation, 18(4), 299–305. [PubMed] [Google Scholar]
- Frey SH, Bogdanov S, Smith JC, Watrous S, & Breidenbach WC (2008). Chronically deafferented sensory cortex recovers a grossly typical organization after allogenic hand transplantation. Current Biology, 18(19), 1530–1534. doi: 10.1016/j.cub.2008.08.051 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Friedman L, & Glover GH (2006). Report on a multicenter fMRI quality assurance protocol. Journal of Magnetic Resonance Imaging, 23(6), 827–839. doi: 10.1002/jmri.20583 [DOI] [PubMed] [Google Scholar]
- Giraux P, Sirigu A, Schneider F, & Dubernard JM (2001). Cortical reorganization in motor cortex after graft of both hands. Nature Neuroscience, 4(7), 691–692. doi: 10.1038/89472 [DOI] [PubMed] [Google Scholar]
- Glover GH, Mueller BA, Turner JA, van Erp TG, Liu TT, Greve DN, Voyvodic JT, Rasmussen J, Brown GG, Keator DB, Calhoun VD, Lee HJ, Ford JM, Mathalon DH, Diaz M, O'Leary DS, Gadde S, Preda A, Lim KO, Wible CG, Stern HS, Belger A, McCarthy G, Ozyurt B, & Potkin SG (2012). Function biomedical informatics research network recommendations for prospective multicenter functional MRI studies. Journal of Magnetic Resonance Imaging, 36(1), 39–54. doi: 10.1002/jmri.23572 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Greve DN, & Fischl B (2009). Accurate and robust brain image alignment using boundary-based registration. Neuroimage, 48(1), 63–72. doi: 10.1016/j.neuroimage.2009.06.060 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hardwick RM, Caspers S, Eickhoff SB, & Swinnen SP (2018). Neural correlates of action: Comparing meta-analyses of imagery, observation, and execution. Neuroscience and Biobehavioral Reviews, 94, 31–44. doi: 10.1016/j.neubiorev.2018.08.003 [DOI] [PubMed] [Google Scholar]
- Hernandez-Castillo CR, Aguilar-Castaneda E, Iglesias M, & Fernandez-Ruiz J (2016). Motor and sensory cortical reorganization after bilateral forearm transplantation: Four-year follow-up fMRI case study. Magnetic Resonance Imaging, 34(4), 541–544. doi: 10.1016/j.mri.2015.12.025 [DOI] [PubMed] [Google Scholar]
- Jeannerod M (2001). Neural simulation of action: a unifying mechanism for motor cognition. Neuroimage, 14(1 Pt 2), S103–109. doi: 10.1006/nimg.2001.0832 [DOI] [PubMed] [Google Scholar]
- Jenkinson M, Bannister P, Brady M, & Smith S (2002). Improved optimization for the robust and accurate linear registration and motion correction of brain images. Neuroimage, 17(2), 825–841. doi: 10.1006/nimg.2002.1132 [DOI] [PubMed] [Google Scholar]
- Jenkinson M, & Smith S (2001). A global optimisation method for robust affine registration of brain images. Medical Image Analysis, 5(2), 143–156. doi: 10.1016/S1361-8415(01)00036-6 [DOI] [PubMed] [Google Scholar]
- Kew JJ, Ridding MC, Rothwell JC, Passingham RE, Leigh PN, Sooriakumaran S, Frackowiak RS, & Brooks DJ (1994). Reorganization of cortical blood flow and transcranial magnetic stimulation maps in human subjects after upper limb amputation. Journal of Neurophysiology, 72(5), 2517–2524. doi: 10.1152/jn.1994.72.5.2517 [DOI] [PubMed] [Google Scholar]
- MacKay BJ, Nacke E, & Posner M (2014). Hand transplantation: A review. Bulletin of the Hospital for Joint Diseases, 72(1), 76–88. [PubMed] [Google Scholar]
- Makin TR, Filippini N, Duff EP, Henderson Slater D, Tracey I, & Johansen-Berg H (2015). Network-level reorganisation of functional connectivity following arm amputation. Neuroimage, 114, 217–225. doi: 10.1016/j.neuroimage.2015.02.067 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Melton MS, Browndyke JN, Harshbarger TB, Madden DJ, Nielsen KC, & Klein SM (2016). Changes in brain resting-state functional connectivity associated with peripheral nerve block: A pilot study. Anesthesiology, 125(2), 368–377. doi: 10.1097/ALN.0000000000001198 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Meyer VE (1991). Major limb replantation revascularization In Meyer VE & Black MJM (Eds.), Microsurgical procedures (pp. 36–68). Edinburgh: Churchill Livingstone. [Google Scholar]
- Neugroschl C, Denolin V, Schuind F, Van Holder C, David P, Baleriaux D, & Metens T (2005). Functional MRI activation of somatosensory and motor cortices in a hand-grafted patient with early clinical sensorimotor recovery. European Radiology, 15(9), 1806–1814. doi: 10.1007/s00330-005-2763-4 [DOI] [PubMed] [Google Scholar]
- Shores JT, Brandacher G, Schneeberger S, Gorantla VS, & Lee WP (2010). Composite tissue allotransplantation: hand transplantation and beyond. Journal of the American Academy of Orthopaedic Surgeons, 18(3), 127–131. [DOI] [PubMed] [Google Scholar]
- Shores JT, Malek V, Lee WPA, & Brandacher G (2017). Outcomes after hand and upper extremity transplantation. Journal of Materials Science: Materials in Medicine, 28(5), 72. doi: 10.1007/s10856-017-5880-0 [DOI] [PubMed] [Google Scholar]
- Smith SM (2002). Fast robust automated brain extraction. Human Brain Mapping, 17(3), 143–155. doi: 10.1002/hbm.10062 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Smith SM, Jenkinson M, Woolrich MW, Beckmann CF, Behrens TE, Johansen-Berg H, Bannister PR, De Luca M, Drobnjak I, Fitney DE, Niazy RK, Saunders J, Vickers J, Zhang Y, De Stefano N, Brady JM, & Matthews PM (2004). Advances in functional and structural MR image analysis and implementation as FSL. Neuroimage, 23 Suppl 1, S208–219. doi: 10.1016/j.neuroimage.2004.07.051 [DOI] [PubMed] [Google Scholar]
- Valyear KF, Mattos D, Philip BA, Kaufman C, & Frey SH (2019). Grasping with a new hand: Improved performance and normalized grasp-selective brain responses despite persistent functional changes in primary motor cortex and low-level sensory and motor impairments. Neuroimage, 190, 275–288. doi: 10.1016/j.neuroimage.2017.09.052 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Vargas CD, Aballea A, Rodrigues EC, Reilly KT, Mercier C, Petruzzo P, Dubernard JM, & Sirigu A (2009). Re-emergence of hand-muscle representations in human motor cortex after hand allograft. Proceedings of the National Academy of Sciences of the United States of America, 106(17), 7197–7202. doi: 10.1073/pnas.0809614106 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Woolrich MW, Behrens TE, Beckmann CF, Jenkinson M, & Smith SM (2004). Multilevel linear modelling for FMRI group analysis using Bayesian inference. Neuroimage, 21(4), 1732–1747. doi: 10.1016/j.neuroimage.2003.12.023 [DOI] [PubMed] [Google Scholar]