Cross-Cultural Adaptation and Application of the Brazilian Version of the Mini-Linguistic State Examination for Primary Progressive Aphasia Classification

Abstract

<p><span>Primary Progressive Aphasia (PPA) is a neurodegenerative disorder characterized by a gradual and progressive decline in language abilities. According to current criteria (Gorno-Tempini et al. (2011), approximately two-thirds of PPA cases can be classified into three clinical variants: nonfluent PPA (nfvPPA), semantic PPA (svPPA), and logopenic PPA (lvPPA). The remaining one-third of cases present with distinct and/or mixed clinical presentations (mxPPA). The Mini-Linguistic State Examination (MLSE), developed by Patel et al. (2021), is a brief assessment tool designed to assess, classify, and monitor various presentations of PPA. It is a brief assessment tool that utilizes an error classification method to extract more clinical information from a set of speech-language tasks. It assesses five domains of language competence (motor speech, phonology, semantics, syntax, and working memory) using 11 tasks. The objectives of this study were twofold: (1) to introduce the Brazilian version of the MLSE, including the translation and cross-cultural adaptation process, along with the main modifications implemented, and (2) to demonstrate the application of the instrument in classifying PPA variants in a Brazilian sample. The study received ethical approval (</span><span>HCFMUSP, CAAE: 52094121.7.0000.0068)</span><span>, and the sample consisted of 14 PPA patients (2 nfvPPA, 2 svPPA, 6 lvPPA, and 4 mxPPA) and 51 control participants. The PPA diagnosis was made by an interdisciplinary team following current guidelines and using a comprehensive neuropsychological battery. The time from symptom onset ranged from 2 to 3 years for all PPA cases.</span><span> T</span><span>he groups did not show significant differences in age (mean (SD) for PPA and controls were 65.5 (10.2) and 61.8 (11.5), p=.291), education (15.1 (4.4) and 13.8 (4.2), p=.348), and sex (p=.552). The ACE-R and MMSE scores for PPA and controls were 56.4 (17) and 91.2 (4.3), p<0.001, and 20.6 (8) and 28.3 (1.5), p<0.001, respectively. The mean time of MLSE administration was 11.8 min in controls (n=20) and 17.8 min in PPA (n=12). MLSE showed a positive correlation with education in the control group (r=.56, p<0.001). As a group, PPA performed significantly lower than controls in the overall score and all five MLSE domains (p<0.001). The mean scores for controls, nfvPPA, svPPA, lvPPA, and mxPPA in each domain were as follows: overall MLSE: 97 (3.8), 70.5 (10.6), 79.5 (3.5), 75 (13.5), 67.3 (10.3); Motor-speech: 29.6 (1.1), 19 (5.7), 28 (2.8), 29 (1.6), 24 (4.4); Phonology: 29.6 (1.07), 28.5 (1), 28 (1.4), 24 (3.7), 27.5 (2.6); Semantics: 19.2 (1.22), 13.5 (6.3), 9.5 (0.7), 13 (4.3), 11.25 (4); Syntax: 9.3 (1.29), 4 (5.6), 8.5 (0.7), 5.3 (2.7), 2.3 (0.5); Working Memory: 9.3 (1.29), 4.5 (6.3), 5.5 (0.7), 4.2 (3.65), 3.3 (4.2). These preliminary findings suggest that the Brazilian version of the MLSE is suitable for screening for PPA and classifying PPA variants, aligning with results obtained with the original version. Further validation of the MLSE in Brazil is ongoing, including the investigation of psychometric properties and adjusted age and education cut-off scores.</span><span></span></p>