Effectiveness of language training and non-invasive brain stimulation on oral and written naming performance in Primary Progressive Aphasia: A meta-analysis and systematic review

Highlights

• Language training in patients with PPA.

• Long-lasting effects of language training.

• Effects of language training combined with tDCS.

Abstract

The aim of this systematic review and meta-analysis is to examine the efficacy of language training, alone or in combination with non-invasive brain stimulation techniques, designed to improve oral and written naming abilities in Primary Progressive Aphasia (PPA), and to investigate whether gains can be maintained over time and generalize to untrained items. An electronic database search was conducted up to 31^st of May 2019. Forty-three articles on language training alone and seven articles on the combined treatment fitted the inclusion criteria for the systematic review. Results indicated that language training, alone or in combination with transcranial direct current stimulation (tDCS), improves oral naming accuracy for trained items in patients with PPA, with a long-term maintenance of the gain over time. Only language training combined with tDCS improves oral naming accuracy for untrained items. Considering written naming abilities, language training combined with tDCS improves performance for both trained and untrained items, immediately after training and at follow-up.

Introduction

The diagnostic label of Primary progressive aphasia (PPA) encompasses patients characterized by a gradual, insidious and progressive loss of language abilities (Cerami et al., 2012; Mesulam, 1982, 2001, 2003).

The diagnosis of PPA requires the following three core criteria: i) the patient should have an aphasic disorder as manifested by distortions of word production or comprehension that cannot be attributed to motor or perceptual deficits; ii) language impairment should constitute the most prominent deficit at the initial stages of the illness and the most salient impediment to the pursuit of daily living activities; iii) the underlying disease should be neurodegenerative and, therefore, progressive (Mesulam and Weintraub, 2008; Mesulam, 2001, 2003).

PPA refers to anatomically and pathologically heterogeneous neurodegenerative disorders (Agosta et al., 2013; Caffarra et al., 2013; Gorno-Tempini et al., 2004, 2011; Grossman, 2010, 2012), characterized by a selective deterioration of language (Rohrer et al., 2008), which are currently differentiated in the three main variants, based on the most frequently observed symptoms. The non fluent/agrammatic variant of PPA (Nf/avPPA) is characterized by a progressive effortful, non-fluent speech with grammatical errors and omissions (Gorno-Tempini et al., 2004; Grossman, 2010; Mesulam, 2001) and by naming impairment, with greater difficulty in naming actions than objects (Cotelli et al., 2006; Silveri and Ciccarelli, 2007). Comprehension is spared, with the exception of grammatically complex sentences (Gorno-Tempini et al., 2011). Imaging evidence of atrophy in the left inferior frontal gyrus, insula and supplementary motor areas is necessary for the diagnosis of Nf/avPPA (Cerami et al., 2017; Gorno-Tempini et al., 2011; Iaccarino et al., 2015). Finally, individuals with Nf/avPPA can also show variable degrees of dysarthria, phonological disintegration, labored pronunciation and apraxia of speech (AOS) (Duffy et al., 2017; Josephs et al., 2006, 2012; Mesulam et al., 2012).

The semantic variant of PPA (svPPA) is characterized by fluent speech and difficulties in semantic memory which lead to defective confrontation naming and single word comprehension. Anterior temporal lobe atrophy is the main imaging feature in these patients (Agosta et al., 2010; Cerami et al., 2017; Iaccarino et al., 2015). The logopenic/phonological variant of PPA (L/phvPPA) is characterized by slow speech with long-word finding pauses and anomia, by defective repetition and sentence comprehension linked to left posterior temporal cortex and left parietal lobule atrophy (Cerami et al., 2017; Henry and Gorno-Tempini, 2010; Iaccarino et al., 2015).

Impairment in oral naming have a great clinical relevance in PPA, since word-finding difficulties are often among the initial complaints of these individuals, thus allowing an early detection (Budd et al., 2010; Hillis et al., 2004; Migliaccio et al., 2016). Anomia is a common symptom in neurological and neurodegenerative diseases and several interventions have also been developed to treat naming difficulties (Maher and Raymer, 2004; Rising, 2014; Tippett et al., 2015). Word-finding difficulties can result from different underlying functional impairments, inducing different qualitative errors. In case of impairment of the semantic system, such as in svPPA, word-retrieval deficits reflect lost or incomplete semantic representations, resulting in semantic errors and anomia (Hodges et al., 1995; Howard and Orchard-Lisle, 1984). Anomia can also result from a post-semantic deficit at the level of the phonological (or orthographic) output lexicon itself (for example in Nf/avPPA). In this case, semantic representations are preserved, whereas access to the output lexicon may be impaired (Gainotti et al., 1986; Howard et al., 1995; Howard and Gatehouse, 2006; Howard and Orchard-Lisle, 1984; Lambon Ralph et al., 2000). Naming abilities are moderately affected also in L/phvPPA and are characterized by phonemic paraphasias and word-finding difficulties, compatible with the presence of a lexical-phonological deficit (Gorno-Tempini et al., 2008).

There is a general consensus that aphasia rehabilitation is effective in stroke (Berthier and Pulvermuller, 2011; Cappa et al., 2005; Galletta et al., 2011). In line with the different loci of functional impairment in the naming process, both semantic and phonological approaches have been used for the treatment of anomia in aphasia due to stroke (Hillis, 1998; Howard et al., 1985; Nickels and Best, 1996). Aphasia after stroke is the result of acute, focal damage of components of the brain’s language network, whereas PPA can be conceptualized as a gradual progressive degradation of the efficiency and the activity of the language network. Evidence on the efficacy of speech and language interventions for PPA is limited (Carthery-Goulart et al., 2013; Croot et al., 2009; Henry et al., 2018; Jokel et al., 2014; Kortte and Rogalski, 2013; Lefaucheur et al., 2017; Rising, 2014; Tippett et al., 2015). A growing body of evidence indicates that language training can have beneficial effects also in the treatment of aphasia caused by a neurodegenerative disease (Tippett et al., 2015). The literature reports different therapeutic approaches attempting to tailor the language treatment to the PPA patient's clinical profile (Carthery-Goulart et al., 2013; Croot et al., 2009; Hameister et al., 2017; Jokel et al., 2009, 2014; Jokel et al., 2016; Khayum et al., 2012; Louis et al., 2001; Marcotte and Ansaldo, 2010; Meyer et al., 2016, 2018b; Rising, 2014), with induced long-term benefits (e.g. Henry et al., 2018).

The first aim of this systematic review and meta-analysis is to evaluate the effectiveness of language training, alone or combined with brain stimulation techniques (rTMS and tDCS), to improve oral and written naming abilities in PPA. This focus is motivated by the central role of word-finding difficulties in all PPA variants. The second aim of this study is to examine whether language training (alone and associated with non-invasive brain stimulation techniques) leads to gains that can be maintained over time and if the effects generalize to untrained items.