TY - JOUR
T1 - Different network functional connectivity characteristics of responders and non-responders to attention training in MS
AU - Prouskas, S. E.
AU - Steenwijk, M. D.
AU - Gehring, K.
AU - Huiskamp, M.
AU - de Jong, B.
AU - Geurts, J. J. G.
AU - Sitskoorn, M. M.
AU - Schoonheim, M. M.
AU - Hulst, H. E.
PY - 2019/6
Y1 - 2019/6
N2 - Background: Cognitive rehabilitation has been suggested as a means to improve cognitive function. However, cognitive rehabilitation is only effective in a subset of MS patients. It is therefore imperative to identify characteristics that might influence the ability to respond to rehabilitation of cognitive impairment. While cognitive impairment in MS is related to changes in functional connectivity (FC) of cognitive brain networks such as the default mode network (DMN), it is unknown how functional network characteristics affect the brain's ability to respond to cognitive training. Aim: To investigate whether baseline resting-state FC within and between the DMN, dorsal attention network (DAN) and ventral attention network (VAN), can distinguish responders from non-responders to an attention training in MS. Methods: Patients were randomized into an attention training (home-based computerized C-Car: 7-week, 45 min/week, N=58, age=48.4±10.2 years, 34 women, RRMS=42, median EDSS=4.0) or a waiting-list control group (CG, N=24, age=48.5±9.4 years, 19 women, RRMS=16, median EDSS=4.0). Neuropsychological assessment at baseline and follow-up included tests of attention, memory, information processing speed, and executive functioning. Based on the CG, a reliable change index (RCI) was calculated, adjusted for practice effects. Responders were defined as patients scoring RCI>1.64 (90% CI) on at least two tests. 3D T1 MRI and resting-state fMRI was obtained at baseline. After preprocessing and denoising using ICA-AROMA, a subject-wise fMRI correlation matrix was computed. Within- and between-network measures of FC of the DMN, DAN, and VAN were calculated using relative correlations and the Brainnetome atlas and compared between responders and non-responders. Results: Responders (N=22) and non-responders (N=36) did not differ significantly in age, sex, education, MS subtype, EDSS, disease duration, or baseline cognition. Responders, compared to non-responders, had lower average FC between DMN and DAN (0.85 vs 0.92 respectively; p=0.04) and DMN and VAN (0.90 vs 0.98 respectively; p=0.04). There were no significant group differences in within-network FC. Conclusion: Lower FC between DMN and attention networks seems an indicator of response to attention training in MS, which might reflect a more intact network functioning at baseline.
AB - Background: Cognitive rehabilitation has been suggested as a means to improve cognitive function. However, cognitive rehabilitation is only effective in a subset of MS patients. It is therefore imperative to identify characteristics that might influence the ability to respond to rehabilitation of cognitive impairment. While cognitive impairment in MS is related to changes in functional connectivity (FC) of cognitive brain networks such as the default mode network (DMN), it is unknown how functional network characteristics affect the brain's ability to respond to cognitive training. Aim: To investigate whether baseline resting-state FC within and between the DMN, dorsal attention network (DAN) and ventral attention network (VAN), can distinguish responders from non-responders to an attention training in MS. Methods: Patients were randomized into an attention training (home-based computerized C-Car: 7-week, 45 min/week, N=58, age=48.4±10.2 years, 34 women, RRMS=42, median EDSS=4.0) or a waiting-list control group (CG, N=24, age=48.5±9.4 years, 19 women, RRMS=16, median EDSS=4.0). Neuropsychological assessment at baseline and follow-up included tests of attention, memory, information processing speed, and executive functioning. Based on the CG, a reliable change index (RCI) was calculated, adjusted for practice effects. Responders were defined as patients scoring RCI>1.64 (90% CI) on at least two tests. 3D T1 MRI and resting-state fMRI was obtained at baseline. After preprocessing and denoising using ICA-AROMA, a subject-wise fMRI correlation matrix was computed. Within- and between-network measures of FC of the DMN, DAN, and VAN were calculated using relative correlations and the Brainnetome atlas and compared between responders and non-responders. Results: Responders (N=22) and non-responders (N=36) did not differ significantly in age, sex, education, MS subtype, EDSS, disease duration, or baseline cognition. Responders, compared to non-responders, had lower average FC between DMN and DAN (0.85 vs 0.92 respectively; p=0.04) and DMN and VAN (0.90 vs 0.98 respectively; p=0.04). There were no significant group differences in within-network FC. Conclusion: Lower FC between DMN and attention networks seems an indicator of response to attention training in MS, which might reflect a more intact network functioning at baseline.
M3 - Meeting Abstract
SN - 1352-4585
JO - MULTIPLE SCLEROSIS JOURNAL
JF - MULTIPLE SCLEROSIS JOURNAL
M1 - P707
ER -