gms | German Medical Science

Objective: Neuro-navigated transcranial magnetic stimulation (TMS) allows for non-invasive mapping of language functions in presurgical planning and for research purposes. We established a semi-standardized neurolinguistic protocol for evaluation of speech and language mapping by TMS.

Method: Twelve healthy left-dominant native speakers of German performed a linguistically controlled picture naming task in random order (drawings of trisyllabic, low-frequency German nouns controlled for number of letters and phonemes as well as for semantic categories). Simultaneously, TMS (5 Pulses at 5 Hz) was applied at 30 predefined target sites covering the pars opercularis and pars triangularis of the left inferior frontal gyrus. Stimulation intensity was defined individually by the threshold for speech-specific symptoms or was set at least at 140 % of resting motor threshold if discomfort prevented higher intensities. The mapping procedure was repeated five times including one sham condition. Audiovisual recordings were analyzed off-line for TMS-induced errors by two speech and language therapists.

Results: At first, trials with non-specific stimulation effects, i.e. disturbed naming in the sham condition already, or with speech and language symptoms indiscernible from reactions to stimulation discomfort and artifacts, were excluded. The remaining conspicuous trials were classified according to the following error categories: no-response error, delay, speech disruption, semantic error, phonematic error, and performance error. Moreover, since overall symptom severity varied considerably between participants, each symptom was rated individually at a three-level severity scale (severe, medium, slight). On average, language-specific disturbances occurred in 34 % of 120 trials under verum stimulation. Our method revealed language mapping results consistent across subjects and separated non-specific from specific TMS-induced symptoms.

Conclusions: We were able to develop a reliable and precise speech and language evaluation protocol for TMS. It allows for error classification comparable across subjects and yielded consistent mapping results despite efficient stimulation intensities and a large inter-individual variability of symptom severity.