Lesion-aphasia discordance in acute stroke among Bengali-speaking patients: Frequency, pattern, and effect on aphasia recovery

Introduction: Contemporary research papers have highlighted the issue of lesion-aphasia discordance in reference to the classic 'associationist' model provided by Wernicke-Lichtheim. The objective of the present study is to explore frequency, pattern and evolution of lesion-discordant aphasia following first ever acute stroke in Bengali-speaking subjects. Methods: Bengali version of Western Aphasia Battery, a validated scale, was used for language assessment in our study subjects. Lesion localization was done by using Magnetic resonance imaging (MRI) (3T) for ischemic stroke (if not contraindicated) and computed tomography (CT) for hemorrhagic stroke. Among 515 screened cases of first-ever acute stroke, 208 presented aphasia. Language assessment was done between 7 and 14 days in all study subjects and was repeated between 90 and 100 days in patients available for follow-up. Ischemic stroke cases with contraindication for MRI underwent CT imaging. Discordance between lesion and aphasic phenotype was determined only for right-handed subjects with cortical involvement (isolated or in combination with sub-cortical white matter) in the left hemisphere. Appropriate statistical tests were used to analyze the collected data. Results: Lesion-aphasia discordance was found in 20 out of 134 patients with aphasia who were dextral and had cortical involvement in the left hemisphere (14.92%). The pattern of discordance observed were- posterior lesion with Broca's aphasia (4; 20%); posterior lesion with global aphasia (8; 40%); anterior lesion with global aphasia (4; 20%), and posterior lesion with mixed transcortical aphasia (4; 20%). On univariate analysis, the factors significantly associated with lesion-aphasia discordance were hemorrhagic stroke (p = 0.000); posterior perisylvian location (p = 0.002), and higher education (p = 0.048). After adjusting for all other variables, hemorrhagic stroke was found to have strong association with lesion-aphasia discordance (p = 0.001, OR = 11.764, 95% CI, 2.83-50.0). Discordant cases were more likely to recover or change to a milder type compared to concordant cases (p = 0.007, OR = 11.393, 95% CI, 1.960-66.231), after adjusting for all other variables including initial severity of aphasia (p = 0.006, OR = 8.388, 95% CI, 1.816-38.749). Conclusion: Lesion-aphasia discordance following acute stroke is not uncommon among Bengali- speaking subjects. In the discordant group, preponderance towards non-fluent aphasia was observed. Discordance occurred more frequently after hemorrhagic stroke. Subjects with lesion-discordant aphasia presented better recovery during early post-stroke phase.