Transcranial direct current stimulation over left inferior frontal cortex improves speech fluency in adults who stutter

Abstract

See Crinion (doi:10.1093/brain/awy075) for a scientific commentary on this article.Stuttering is a neurodevelopmental condition affecting 5% of children, and persisting in 1% of adults. Promoting lasting fluency improvement in adults who stutter is a particular challenge. Novel interventions to improve outcomes are of value, therefore. Previous work in patients with acquired motor and language disorders reported enhanced benefits of behavioural therapies when paired with transcranial direct current stimulation. Here, we report the results of the first trial investigating whether transcranial direct current stimulation can improve speech fluency in adults who stutter. We predicted that applying anodal stimulation to the left inferior frontal cortex during speech production with temporary fluency inducers would result in longer-lasting fluency improvements. Thirty male adults who stutter completed a randomized, double-blind, controlled trial of anodal transcranial direct current stimulation over left inferior frontal cortex. Fifteen participants received 20 min of 1-mA stimulation on five consecutive days while speech fluency was temporarily induced using choral and metronome-timed speech. The other 15 participants received the same speech fluency intervention with sham stimulation. Speech fluency during reading and conversation was assessed at baseline, before and after the stimulation on each day of the 5-day intervention, and at 1 and 6 weeks after the end of the intervention. Anodal stimulation combined with speech fluency training significantly reduced the percentage of disfluent speech measured 1 week after the intervention compared with fluency intervention alone. At 6 weeks after the intervention, this improvement was maintained during reading but not during conversation. Outcome scores at both post-intervention time points on a clinical assessment tool (the Stuttering Severity Instrument, version 4) also showed significant improvement in the group receiving transcranial direct current stimulation compared with the sham group, in whom fluency was unchanged from baseline. We conclude that transcranial direct current stimulation combined with behavioural fluency intervention can improve fluency in adults who stutter. Transcranial direct current stimulation thereby offers a potentially useful adjunct to future speech therapy interventions for this population, for whom fluency therapy outcomes are currently limited.

Figures

Figure 1

TDCS montage and behavioural tasks used in intervention. (A) Electrode placement montage used to apply tDCS. Anode (pink) was placed over left inferior frontal cortex, centred on position FC5 of the 10-10 EEG electrode placement system. Cathode (blue) was placed over the right supra-orbital ridge. (B) Choral speech (live voice and recorded voice) and metronome-timed speech (video narrative and conversation) tasks used in each daily intervention session. TDCS (1 mA) was applied concurrently with these tasks for 20 min in 15 male adults who stutter. Another 15 male adults who stutter received sham stimulation for the same period. Both researcher and participant were blind to the stimulation condition.

Figure 2

Trial design. Light grey boxes show baseline and outcome assessment time points. Measures used for the primary outcome (% ds = percentage of dysfluent syllables) and secondary outcomes (SSI-4, OASES) are shown in bold text. Additional measures used for matching group and monitoring adverse effects are shown in regular text. Exploratory measures taken pre-and post-intervention are shown in white boxes for Day 1 of the intervention only. These were repeated on each intervention day. BAI = Beck Anxiety Inventory.

Figure 3

Effect of tDCS on the primary outcome measure: change from baseline in speech disfluency. Bars indicate mean change from baseline in % disfluent syllables (% ds) measured at 1- and 6-weeks post-intervention averaged across speech samples obtained during reading and conversation. Red = tDCS group; grey = Sham group. Error bars indicate standard error of the mean (SEM). Asterisks mark the significant main effect of tDCS.

Figure 4

Effect of tDCS on reading and conversation tasks separately. Bars indicate mean change from baseline in % disfluent syllables (% ds) measured at 1- and 6-weeks post-intervention for the two speaking tasks in the (A) tDCS and (B) Sham groups. Unfilled bars = reading task (Read), striped bars = conversation task (Conv). Error bars indicate SEM. There was a significant interaction between time point and task for the tDCS group only (the significant task difference at 6 weeks post-intervention is marked with an asterisk).

Figure 5

Effects of tDCS on secondary outcomes: change from baseline in SSI and OASES scores. Bars indicate mean change from baseline in (A) SSI-4 scores at 1- and 6-weeks post-intervention, and (B) OASES scores at 6-weeks post-intervention, for the tDCS (red) and sham (grey) groups. Error bars indicate SEM. Asterisks mark the significant main effect of tDCS on the SSI-4 change scores. There was no significant difference between groups for the OASES.

Figure 6

Effects of tDCS on speech disfluency during the 5-day intervention. Bars indicate mean change in % disfluent syllables (% ds) from baseline in speech sample during (A) reading (open bars) and (B) conversation (striped bars) tasks on Days 1 to 5 during the intervention for the tDCS (red) and sham (grey) groups. Error bars indicate SEM. The significant main effect of stimulation is marked with asterisks.