Patterns of language improvement in adults with non‐chronic non‐fluent aphasia after specific therapies

Abstract

Background: Methods for functional and linguistic analysis of discourse have been used for describing recovery from aphasia and examining relationships between patterns of recovery and specific therapeutic programmes. This approach, however, has mainly concerned therapeutic programmes for chronic aphasic symptoms (e.g., therapy for chronic agrammatism in non‐fluent aphasic subjects).

Aims: The first aim of this study was to examine whether functional and linguistic analyses of discourse are suitable to describe aspects of language improvement in subjects recovering from non‐fluent aphasia in the first months post‐onset. A second objective was to assess the effectiveness of two therapy programmes for chronic aphasia in increasing informativeness and/or morpho‐syntactic organisation of connected speech. This was made by examining in‐depth the correspondence between each of the two therapy programmes and the results from functional and linguistic analysis of discourse at pre‐ and post‐therapy evaluation.

Methods & Procedures: Three subjects with non‐fluent aphasia (12, 18, and 22 weeks post‐onset, respectively), whose speech was characterised by reduced information content and poor morpho‐syntactic organisation, received two consecutive therapy programmes, each consisting of 35 one‐hour sessions in seven weeks. The first programme consisted of stimulus–response exercises for producing well‐formed sentences (HELPSS, Helm‐Estabrooks, Fitzpatrick, & Barresi, 1981). This was followed by a functional treatment programme in PACE format (Carlomagno, Losanno, Emanuelli, & Razzano, 1991) intended to increase informativeness of communicative (verbal and non‐verbal) behaviour. At the three assessments (before and after HELPSS and after modified PACE) the three participants were asked to describe two cartoon stories and two single pictures. These connected speech samples underwent functional analysis (CIUs, Nicholas & Brookshire, 1993; Main Concept Analysis, Nicholas & Brookshire, 1995) and linguistic analysis (Marini, Caltagirone, Carlomagno, & Nocentini, 2005a; Marini, Boewe, Caltagirone, & Carlomagno, 2005b) in order to examine the pattern of language recovery. The three participants also received functional evaluation with the EFCP (Wirz, Skinner, & Dean, 1990) for assessing changes in language performance in communicative interaction. At the beginning and end of the therapy period, further evaluation was performed by means of standardised aphasia tests (AAT, Italian version, Luzzatti, Willems, & DeBleser, 1991; CADL, Italian version, Pizzamiglio et al., 1984).

Outcomes & Results: Following the therapy programmes, a few changes were observed on standard aphasia tests. However, the informativeness of the speech samples by the three subjects increased. This corresponded to better rating of their language in the EFCP interview. In two subjects, the linguistic analysis of connected speech samples failed to show consistent syntactic organisation at the post‐therapy evaluation and differences between effects of the two programmes were marginal. In the third, the pattern of recovery did not correspond closely to that predicted by the type of treatment, i.e., better syntactic organisation of speech output became evident only after the second therapy programme.

Conclusions: It is suggested that discourse analysis methods are useful for studying functional and linguistic aspects of recovery in subjects with non‐fluent aphasia in the early post‐onset period. Furthermore, in this period, specific therapy for chronic agrammatic symptoms may not reduce them. Nonetheless, this therapeutic approach plays a role in improving language informativeness particularly when combined with a functional approach.

This study has been supported by a grant of the IRCCS Santa Lucia in Rome. The authors are indebted to Carmelina Razzano, Ilaria De Mercurio and Giuseppe Mancini for careful assessment of the participants with standard aphasia tests and Cookie Theft Picture Description task before and after the therapy. They wish to thank Anke Boewe for linguistic revision of the paper. Finally, the authors are grateful to the two anonymous reviewers for their insightful comments on the first version of the paper.

Notes

1. Spontaneous language improvement has been well documented in aphasic subjects in the first months post‐stroke (Basso, Capitani, & Vignolo, 1979; Kertesz & McCabe, 1977; Mazzoni et al., 1992). This improvement may continue for 3 months, 6 months, or even longer (Holland, 1990; Kertesz & McCabe, 1977). It is not our purpose to discuss the temporal aspect or the mechanisms of spontaneous recovery. Rather, the focus here is on whether functional and linguistic analyses of discourse are suitable to evaluate recovery and the pattern of improvement in subjects with aphasia who receive specific therapy programmes in the first months post‐onset, i.e., in subjects where some spontaneous recovery likely occurs.

2. It is not aim of the present study to describe in‐depth gestures of the three subjects and their changes following therapy, as our main interest was to study changes in language performance. However, before and after the therapy period, the three subjects were also given the referential communication task in PACE format described by Carlomagno, Santoro, Menditti, Pandolfi, and Marini (2005). This task allows evaluation of lexical and gestural components of messages used by brain‐damaged patients for conveying discriminating information about 25 referents in face‐to‐face communicative interaction with the examiner. During the study period, production of “informative” gestures increased from 10 to 15 in FS and from 8 to 13 in AS (normal controls  =  1, 3, SD  =  1, 9) while in MR it remained 5. It should be noted that, in this test, coverbal gestures are scored as “informative” when they show semantic relationships with the speech content or they are substitute for expected information that is not in the speech (Carlomagno et al., 2005).

3. A verb retrieval deficit may undermine production of well‐formed sentences in aphasic patients (Berndt, Mitchum, & Wayland, 1997) and may constrain response to specific therapies for agrammatic disturbances like the Mapping Therapy (Schwartz et al., 1994, patient G.R.). The hypothesis of selective verb impairment in agrammatism (Miceli, Silveri, Villa, & Caramazza, 1984) locates the agrammatic disturbances of non‐fluent aphasic patients in a grammatically organised lexical component of the language system. Poor response to the HELPSS therapy in AF and FS might correspond to poor recovery of this component with respect to the noun component. In the case of MR, however, recovery of verb retrieval was only delayed with respect to recovery of the noun component.