Language features in the acute phase of poststroke severe aphasia could predict the outcome

Aphasia recovery remains difficult to predict initially in particular for the most severe cases. The features of impaired verbal communication which are the basis for cognitive-linguistic diagnosis and treatment could be part of prediction of recovery from aphasia.

This study investigated whether some components of language screening in the acute phase of stroke are reliable prognostic factors for language recovery in the post-acute phase.

Monocentric prospective study.

University hospital stroke unit.

Eighty-six patients aged between 21 and 92 years (mean=67.4, SD=15.3) were admitted after a first left hemisphere stroke with aphasia and were consecutively included.

Language assessment was performed in the acute phase and 3 months post-stroke with the LAnguage Screening Test (LAST) and the Aphasia Severity Rating Scale (ASRS) of the Boston Diagnostic Aphasia Examination (BDAE). Severe aphasia was defined as ASRS<3. Good recovery was defined as an ASRS≥4. Language scores and other potential predictors of recovery were analysed by comparing groups of patients with good versus poor recovery and as predictors of change with multiple regression approaches.

LAST Total score as well as all the individual items of LAST, NIHSS and ASRS measured in the acute phase significantly differentiated good and poor recovery from aphasia at three months for all aphasic patients and for the most severe cases. In multivariable analyses the repetition score of LAST at the acute phase was significantly associated with the delta of ASRS between the acute phase and 3 months after the stroke reflecting changes in symptom severity.

For patients with initial severe aphasia, word repetition from a language screening task seems to be a more relevant predictor of recovery than initial severity to enrich the prognosis of poststroke aphasia recovery three month after a stroke. These findings show the importance of phonological perception and production as well as speech motor components in the recovery of language. These linguistic aspects of the assessment seem more relevant than severity for prediction in the acute phase.

These findings could improve aphasia management pathway for people with severe aphasia and their families and minimize the evidence-practice gap for speech pathologists.