Studio con risonanza magnetica del giro temporale superiore e delle sue sottoaree nella schizofrenia: rilevanza per le dimensioni neurolinguistiche

Background: Language disturbances, such as impoverishment, disorganization and dysregulation, are prominent features of schizophrenia. These deficits have been described in patients both in term of thought disorders and schizophasia (dysphasia-like impairments such as neologism, clanging, unintelligible utterance). Moreover, phonetics, access to the lexicon and sometimes syntax are clearly impaired, along with pragmatic deficits (impaired affective/linguistic prosody, theory of mind, non literal expressions comprehension). In the last decades, researches carried out with neuroimaging techniques have suggested the superior temporal gyrus (STG) and its sub-areas, Heschl’s gyrus (HG) and planum temporale (PT), as essential components of the fronto-temporal network involved in speech perception and production. Also, they have highlighted structural abnormalities in HG and STG (especially the left side) and an inversion of PT asymmetry in patients with schizophrenia compared to healthy controls. Such anatomical findings could be linked to language disturbances described in schizophrenia. Aim of the study: To compare STG, HG and PT volumes, PT lateralization index and linguistic/pragmatic dimensions in a sample of patients with a DSM-IV diagnosis of schizophrenia and healthy controls; to investigate correlations between volumes and linguistic/pragmatic indices. Methods: Fifty-eight patients with schizophrenia (mean age±sd: 40.60±11.58; duration of illness: 13.92±10.79; 40 males, 18 females; all right-handed; all Caucasian) and 59 healthy controls (mean age±sd: 40.69±11.10; 32 males, 27 females; all right-handed; all Caucasian) underwent 1.5 T structural magnetic resonance imaging (sMRI). Superior temporal gyrus, HG and PT were manually traced and volumes were calculated using the BRAINS2 software. Narrative, conversational and syntactic comprehension tasks were used to evaluate linguistic skills, while a pragmatic comprehension test was used to assess the ability to understand non literal aspects of language (metaphors and idioms). A partial correlation was used to investigate the link between volumes and linguistic/pragmatic indices. Results: As regards STG, HG and PT volumes, any differences was found between patients and healthy controls. When groups were stratified by sex, a left PT volume loss and a reduced right to left asymmetry was found only in the female patients compared to females controls (PT volume: F=4.54, p= 0.04; PT asymmetry index: F=4.69, p=0.04; ANOVA, age and intracranial volume as covariates). As regards language, patients with schizophrenia showed a widespread impairment of linguistic and pragmatic indices, especially regarding speech fluency, syntactic complexity and diversity, lexical diversity, syntactic rules understanding and non-literal expression comprehension (p<0.01, MANOVA with age, sex and educational level as covariates). Furthermore, distinct language dimensions differently correlated with STG-HG-PT volumes in patients with schizophrenia and controls, particularly with regard to speech fluency, syntactic measures and metaphor/idiom comprehension (p<0.01, partial correlation with age and duration of illness as covariates). Conclusion: This study has highlighted structural anomalies in the PT in females patients with schizophrenia compared to healthy females. Also, several linguistic and pragmatic deficits have been found in patients with schizophrenia compared to controls. Furthermore, distinct correlations between volumes and linguistic indices in patients with schizophrenia and healthy controls suggest a complex neuroanatomical substrate for language dimensions in healthy humans and in schizophrenia patients. In addition, such results suggest the importance to plan studies separately for males and females (and for left-handers/right-handers) in order to reduce the role of confounding variables. Further studies with larger samples should clarify whether these anatomical sex differences in language networks could result in different linguistic abilities.