Predicting Language Outcome After Left Hemispherotomy: A Systematic Literature Review. Issue 2 (April 2021)

• Record Type: Journal Article
• Authors: Lidzba, Karen; Bürki, Sarah E.; Staudt, Martin
• Is Part Of: Neurology. Volume 11:Issue 2(2021)
• Publication Date: 2021-04
• Subjects: Nervous system -- Diseases -- Periodicals; Neurology -- Periodicals; Neurology -- United States -- Periodicals; 616.8
• DOI: 10.1212/CPJ.0000000000000852
• Languages: English
• ISSNs: 2163-0402

Abstract

Abstract : Objective: Hemidecortication is a therapeutic option in patients with drug-resistant structural epilepsy. If surgery is performed early enough in left-hemispheric pathology, the plasticity of the developing brain may enable the right hemisphere to take over language—if this has not occurred before surgery. A systematic overview of potential predictors of language outcome after left hemidecortication in children is warranted. Methods: In a systematic literature review, we analyzed 58 studies on language lateralization after congenital or postneonatally acquired left-hemispheric pathology, and on language outcome after left-sided hemidisconnection, such as hemispherotomy. Single-subject data were pooled to determine the distribution of lateralization across etiologies in congenital lesions and across age groups in acute postneonatal lesions. A hierarchical linear regression assessed the influence of age at surgery, lesion type, age at seizure onset, and presurgery language function on language outcome after left hemidecortication. Results: In acute postneonatal lesions, younger age at injury was significantly associated with right-sided language lateralization (Cramér V = 0.458; p = 0.039). In patients with hemidecortication, age at surgery was not significantly associated with language outcome (Cramér V = −0.056; p = 0.584). Presurgical language function was the most powerful predictor for postsurgical language outcome ( F 4, 47 = 7.35, p < 0.0001), with good presurgical language bearing the risk of postsurgical deterioration. In congenital pathology, right-sided language lateralization was most frequent in pre-/perinatal stroke (Cramér V = 0.357; p < 0.0001). Conclusions: We propose a presurgical decision algorithm with age, presurgical language function, language lateralization, and left-hemispheric structural pathology as decision points regarding surgery.