Date Presented 03/26/20
In this study, fMRI was evaluated as a predictor of individual response to a rehabilitation program. Initial functional connectivity (FC) and acute change in FC after the first therapy session were associated with change in upper-extremity motor function from baseline to one-month follow-up after six-weeks of therapy. This result encourages a large study to examine if fMRI FC is predictive of patients’ responsiveness to an intervention, thus promoting individualized treatment.
Primary Author and Speaker: Amanda Vatinno
Contributing Authors: Andria Farrens, Na Jin Seo, Fabrizio Sergi
PURPOSE: Given the limited time and resources allotted for therapy, post-stroke treatment must be optimized to maximize recovery. To infer individual treatment response and promote personalized rehabilitation, this study sought to determine whether motor improvement after completion of a therapy program may be associated with either baseline neural function (Burke-Quinlan et al., 2015) or with acute change in neural function measured after the first therapy session (Sergi et al., 2011), measured via functional magnetic resonance imaging (fMRI).
DESIGN: This was a pilot study involving a single-arm intervention group of four chronic stroke survivors (>6 months post-stroke) with moderate upper extremity impairment.
METHOD: Participants received 6-week (18-session) goal-directed task-practice therapy. They also received imperceptible vibratory stimulation to the paretic wrist (Seo et al. 2019) during the therapy sessions. Clinical outcome measures including the Wolf Motor Function Test (WMFT) were obtained at baseline, post-therapy, and one-month follow-up. Resting state fMRI was acquired immediately before and immediately after the first therapy session using a 3T MRI scanner. Functional connectivity (FC) was calculated between the primary motor cortex of the lesioned hemisphere and contralateral cerebellum regions 1-4 due to the cerebellum’s role in motor learning. The correlation between FC and change in the WMFT scores (∆WMFT) from baseline to follow-up were examined.
RESULTS: Initial FC positively correlated with ∆WMFT time (r=0.98). Following the first therapy session, FC significantly increased (p<0.05). Change in FC after the first session negatively correlated with ∆WMFT time (r=-0.98). These results suggest that individuals with initially lower FC improved more (i.e., reduced WMFT time more) from the intervention. FC increased after the first intervention session, and individuals who increased FC to a greater extent after the first intervention session improved more upon completion of the 6-week intervention.
CONCLUSION: We show preliminary results that resting state fMRI FC may explain variability in patients’ motor improvement at the end of the intervention program, and may hold value in predicting individual patients’ responsiveness to a given intervention. Due to the small sample size, a larger trial is needed to confirm these results.
IMPACT STATEMENT: This work suggests a potential for occupational therapists to infer patients’ individual responses to an intervention based on examination of neural function, thus promoting personalized rehabilitation.
References
Burke-Quinlan, E., Dodakian, L., See, J., McKenzie, A., Le, V., Wojnowicz, M., ... & Cramer, S. C. (2015). Neural function, injury, and stroke subtype predict treatment gains after stroke. Annals of neurology, 77(1), 132-145. doi: 10.1002/ana.24309
Sergi, F., Krebs, H. I., Groissier, B., Rykman, A., Guglielmelli, E., Volpe, B. T., & Schaechter, J. D. (2011, August). Predicting efficacy of robot-aided rehabilitation in chronic stroke patients using an MRI-compatible robotic device. In Engineering in Medicine and Biology Society, EMBC, 2011 Annual International Conference of the IEEE (pp. 7470-7473). IEEE. doi: 10.1109/IEMBS.2011.6091843
Seo, N. J., Lakshminarayanan, K., Lauer, A. W., Ramakrishnan, V., Schmit, B. D., Hanlon, C. A., ... & Nagy, T. (2019). Use of imperceptible wrist vibration to modulate sensorimotor cortical activity. Experimental brain research, 237(3), 805-816.
