Functional versus Nonfunctional Rehabilitation in Chronic Ischemic Stroke: Evidences from a Randomized Functional MRI Study

Abstract

Motor rehabilitation of stroke survivors may include functional and/or nonfunctional strategy. The present study aimed to compare the effect of these two rehabilitation strategies by means of clinical scales and functional Magnetic Resonance Imaging (fMRI). Twelve hemiparetic chronic stroke patients were selected. Patients were randomly assigned a nonfunctional (NFS) or functional (FS) rehabilitation scheme. Clinical scales (Fugl-Meyer, ARA test, and modified Barthel) and fMRI were applied at four moments: before rehabilitation (P1) and immediately after (P2), 1 month after (P3), and three months after (P4) the end of rehabilitation. The NFS group improved significantly and exclusively their Fugl-Meyer scores at P2, P3, and P4, when compared to P1. On the other hand, the FS group increased significantly in Fugl-Meyer at P2, when compared to P1, and also in their ARA and Barthel scores. fMRI inspection at the individual level revealed that both rehabilitation schemes most often led to decreased activation sparseness, decreased activity of contralesional M1, increased asymmetry of M1 activity to the ipsilesional side, decreased perilesional activity, and decreased SMA activity. Increased M1 asymmetry with rehabilitation was also confirmed by Lateralization Indexes. Our clinical analysis revealed subtle differences between FS and NFS.

Figure 1

Results of all clinical scales (Fugl-Meyer, ARAT, and Barthel index), for the NFS and FS groups. Values are represented as median and interquartile interval. ^∗ p < 0.05  ^∗∗ p < 0.01 based on Tukey post hoc test. (a) shows the results for the clinical scales in the NFS group, before rehabilitation (P1), immediately after rehabilitation (P2), 1 month after the end of rehabilitation (P3), and three months after the end of rehabilitation (P4). (b) shows the results of clinical scales in the FS group. (c) shows the between-group comparison (NFS × FS) at the different periods of evaluation (P1, P2, P3, and P4).

Figure 2

fMRI of a representative patient (#12) in all periods of evaluation (P1, P2, P3, and P4). Cross-lines are centered over M1 of the ipsilesional hemisphere. At P1, bilateral M1 activity, asymmetrical to the ipsilesional hemisphere. The maps are very sparse, particularly at P1. At P2, sparseness is reduced, and the activity is more confined to M1 and SMA. One month without rehabilitation (P3), the patterns become somehow similar to what they were before treatment onset, which is maintained three months after the end of treatment (P4).

Figure 3

fMRI of patient #3 showing changes from contralesional M1 (at P1) to ipsilesional M1 (at P2). Before rehabilitation (P1), there is an increased activity of M1 in the contralesional hemisphere (ipsilateral to the moving hand) and of SMA. Right after rehabilitation, the activities of both areas are reduced.

Figure 4

Lateralization Index (LI) of ipsilesional and contralesional M1. Values are represented as median and interquartile interval. LI of the control group is presented. For all patients, LI were evaluated at P1, P2, P3, and P4. ^∗ p < 0.05; ^∗∗ p < 0.01.

Figure 5

fMRI of patient #5 showing reduced perilesional activity at P2 with respect to P1. fMRI maps were obtained for the hand movements of the paretic hand. Images show decreased perilesional activity after rehabilitation.