Brain · 2014 · 316 citations · 74 references
Health ProfessionsNeurologyNeuroscienceAphasiaClinical NeuroscienceCognitive NeuroscienceAphasia AssessmentAphasia DiagnosisAphasia NeuropathologyAphasia SyndromesNeuroanatomy Of AphasiaPost-stroke AphasiaAphasia PrognosisDiffusion Tractography In NeuroradiologyPerisylvian Cortex DamagePerisylvian Network DysfunctionStroke-induced AphasiaStroke-related AphasiaStroke-related Cognitive ImpairmentWhite Matter Tractography
Stroke-induced aphasia is associated with adverse effects on quality of life and the ability to return to work. For patients and clinicians the possibility of relying on valid predictors of recovery is an important asset in the clinical management of stroke-related impairment. Age, level of education, type and severity of initial symptoms are established predictors of recovery. However, anatomical predictors are still poorly understood. In this prospective longitudinal study, we intended to assess anatomical predictors of recovery derived from diffusion tractography of the perisylvian language networks. Our study focused on the arcuate fasciculus, a language pathway composed of three segments connecting Wernicke's to Broca's region (i.e. long segment), Wernicke's to Geschwind's region (i.e. posterior segment) and Broca's to Geschwind's region (i.e. anterior segment). In our study we were particularly interested in understanding how lateralization of the arcuate fasciculus impacts on severity of symptoms and their recovery. Sixteen patients (10 males; mean age 60 ± 17 years, range 28-87 years) underwent post stroke language assessment with the Revised Western Aphasia Battery and neuroimaging scanning within a fortnight from symptoms onset. Language assessment was repeated at 6 months. Backward elimination analysis identified a subset of predictor variables (age, sex, lesion size) to be introduced to further regression analyses. A hierarchical regression was conducted with the longitudinal aphasia severity as the dependent variable. The first model included the subset of variables as previously defined. The second model additionally introduced the left and right arcuate fasciculus (separate analysis for each segment). Lesion size was identified as the only independent predictor of longitudinal aphasia severity in the left hemisphere [beta = -0.630, t(-3.129), P = 0.011]. For the right hemisphere, age [beta = -0.678, t(-3.087), P = 0.010] and volume of the long segment of the arcuate fasciculus [beta = 0.730, t(2.732), P = 0.020] were predictors of longitudinal aphasia severity. Adding the volume of the right long segment to the first-level model increased the overall predictive power of the model from 28% to 57% [F(1,11) = 7.46, P = 0.02]. These findings suggest that different predictors of recovery are at play in the left and right hemisphere. The right hemisphere language network seems to be important in aphasia recovery after left hemispheric stroke.
74
The assessment and analysis of handedness: The Edinburgh inventory
R. C. Oldfield · Neuropsychologia · 1971 · 36.7K citations
Regression and time series model selection in small samples
Clifford M. Hurvich, Chih‐Ling Tsai · Biometrika · 1989 · 6.4K citations
ANew Look at the Statistical Model Identification
HlROTUGU AKAlKE · 2011 · 4.2K citations
In vivo fiber tractography using DT-MRI data
Peter J. Basser, Sinisa Pajevic, Carlo Pierpaoli et al. · Magnetic Resonance in Medicine · 2000 · 3.1K citations
Medicine, Anatomy, Neuroscience +17