Transcranial Direct Current Stimulation Does Not Improve Language Outcome in Subacute Poststroke Aphasia
Background and Purpose—The aim of the present study is to investigate the effect of transcranial direct current stimulation on word-finding treatment outcome in subacute poststroke aphasia.
Methods—In this multi-center, double-blind, randomized controlled trial with 6-month follow-up, we included 58 patients with subacute aphasia (<3 months poststroke), who were enrolled in a stroke rehabilitation program. Patients participated in 2 separate intervention weeks. Each intervention week included 5 daily sessions of 45-minute word-finding therapy combined with either anodal transcranial direct current stimulation (1 mA, 20 minutes; experimental group) or sham transcranial direct current stimulation (control group) over the left inferior frontal gyrus. The primary outcome measure was the Boston Naming Test. Secondary outcome measures included naming performance for trained/untrained picture items and verbal communication.
Results—Both the experimental (n=26) and the control group (n=32) improved on the Boston Naming Test over the intervention period and 6-month follow-up; however, there were no significant differences between groups. Also for the secondary outcome measures, no significant differences were found.
Conclusions—The results of the present study do not support an effect of transcranial direct current stimulation as an adjuvant treatment in subacute poststroke aphasia.
Clinical Trial Registration—URL: http://www.trialregister.nl/trialreg/admin/rctview.asp. Unique identifier: NTR4364.
Transcranial direct current stimulation (tDCS) is reported to enhance the effect of speech and language therapy in chronic poststroke aphasia.1 With tDCS, a weak current of 1 to 2 mA is applied; the anodal electrode enhances cortical excitability, whereas the cathodal electrode decreases cortical excitability.2 It is assumed that tDCS enhances learning through long-term potentiation, that is, long-lasting synaptic plasticity.3 For these long-term processes to take place, ongoing synaptic activation is necessary, and therefore, tDCS is usually combined with training/treatment (ie, online tDCS).
At present, studies investigating the effect of online tDCS in subacute aphasia are scarce. Because most recovery is observed in the first 3 months4 and most treatment is provided in this period, studying the potential effect of tDCS in the early stage is clinically of interest. One subacute study reported no effect of tDCS on naming; however, they used offline tDCS (without simultaneous treatment).5 Offline tDCS was also ineffective in chronic aphasia.6 Therefore, the present study aims to investigate the effect of online tDCS in subacute poststroke aphasia.
The data that support the findings of this study are available from the corresponding author on reasonable request.
In this multi-center, double-blind, randomized controlled trial with 2 parallel groups and 6-month follow-up, we included people with aphasia between 3 weeks and 3 months poststroke. Participants were all enrolled in a stroke rehabilitation program including speech and language therapy. For 2 intervention weeks, separated by 2 weeks, regular speech and language therapy was replaced by word-finding therapy combined with either active tDCS (20 minutes, 1 mA; experimental group) or sham-tDCS (control group) over the left inferior frontal gyrus.
On the basis of the power calculation,7 a total number of 58 patients was needed. From March 2014 to October 2016, patients were screened at 4 rehabilitation centers in the Netherlands. Participants were right-handed adults and were native speakers of Dutch. Table I in the online-only Data Supplement presents the inclusion and exclusion criteria. All participants signed informed consent. This study was approved by the Medical Ethics Committee of the Erasmus MC, University Medical Center Rotterdam.
Before intervention, participants were randomized to the experimental group (active tDCS) or to the control group (sham-tDCS). Participants, speech and language therapists, and researchers were all blinded for tDCS condition. Intervention weeks comprised 5 daily 45-minute sessions of word-finding therapy. Before each intervention week, treatment material was selected from the European Data Bank for oral picture naming8; from the first 68 picture items that the participant could not name orally within 20s, we made 2 sets, each of 34 items, matched for word length and word frequency (ie, frequency with which a word occurs): 1 set to be trained (ie, trained items) and 1 control set (ie, untrained items). The online-only Data Supplement includes more information on randomization, blinding, tDCS device, electrode placement, and the word-finding therapy protocol.
The primary outcome measure was word finding, assessed with the Boston Naming Test9 (a 60-item picture-naming test), before and after each intervention week (T1, T2, T3, and T4) and at 6-month follow-up (T5).
Secondary language measures included naming performance on trained and untrained items, which was scored as the percentage correctly named items. Performance was assessed after each intervention week (T2 and T4). In addition, 2 verbal communication measures were administered before the first intervention week (T1), after the second intervention week (T4), and at 6-month follow-up (T5). The Aphasia Severity Rating Scale10 is a 6-point rating scale to judge communicative ability in a semistructured interview. The Amsterdam Nijmegen Everyday Language Test11 is a test with 10 possible everyday scenarios in which the participants’ response is scored on a 5-point scale. To register potential adverse effects, participants rated their discomfort immediately after each therapy session on the Wong-Baker Faces 5-point pain rating scale. The researcher registered any adverse events, as reported by participants, participating speech and language therapists, and physicians.
Baseline differences between groups were tested with independent t tests, χ2 tests, and Mann–Whitney U tests, as appropriate. Outcome measures at each time point were analyzed as dependent variables with generalized estimation equation models on an intention-to-treat basis. Fixed factors in the model were group (experimental and control), time (T1, T2, T3, T4, and T5), and the interaction between these factors (group×time). In generalized estimation equation models, missing values are taken into account by estimating a correlation matrix that represents the within-subject dependencies. P<0.05 was considered statistically significant.
We included 58 participants: 26 were allocated to the experimental group and 32 to the control group. Fifty-six participants completed the 2 intervention weeks (96.6%) and 54 (93.1%) completed follow-up; the online-only Data Supplement presents a flowchart. At baseline, there were no significant group differences in demographic characteristics (Table) and language outcomes measures (online-only Data Supplement).
Participants tolerated treatment well. Wong-Baker Faces 5-point pain rating scale scores ranged from 0 to 2, with a median of 0 (interquartile range, 0–0). Sixty-nine percent reported no pain (score=0); the rest reported very little pain/little pain (score=1–2), 56% of these participants received sham-tDCS. There was no significant difference in pain rating between groups (P=0.725). Reported side effects were headache and skin irritability; no adverse events were reported during treatment.
Both groups improved significantly on the Boston Naming Test (Figure) over the intervention period and at follow-up (P<0.001). No significant effect was found on active tDCS compared with sham-tDCS (P=0.994). Adjustment for aphasia severity (P=0.990) or baseline Boston Naming Test scores (P=0.996) did not alter these results.
For the secondary outcome measures, also no significant effect was found on active tDCS compared with sham-tDCS over the intervention period/follow-up, not for trained items (P=0.616), not for untrained items (P=0.404), and not for verbal communication (Aphasia Severity Rating Scale: P=0.828; Amsterdam Nijmegen Everyday Language Test: P=0.983).
In this multi-center randomized controlled trial, we do not find support for using online tDCS as an adjuvant treatment during word-finding therapy in poststroke subacute aphasia. Our results are in line with 1 earlier study conducted in subacute aphasia,5 which failed to show significant effects of tDCS in an offline treatment paradigm. The authors hypothesized that an effect of tDCS might be difficult to achieve in the subacute phase, as spontaneous recovery is rather high in this phase compared with the relatively stable chronic phase. Our study has some limitations. First of all, we did not include all types of aphasia. As the word-finding treatment with providing cues is an inadequate treatment approach for severe Wernicke’s aphasia and global aphasia, we decided to exclude these patients from our trial. Another limitation of the present study is the relatively short intervention period; however, the intensity and treatment duration as used in our study is in line with studies showing an effect of tDCS in chronic poststroke aphasia.1 For future studies, it will be interesting to investigate whether a higher intensity and longer duration of tDCS treatment may be effective in subacute aphasia. Further, although chronic tDCS studies have reported positive (though small) effects, the impact of several parameters is currently unknown. Parameters such as aphasia type or lesion size/site probably play an important role in tDCS treatment response.1 Although most studies use 1 predefined electrode configuration for all participants, the optimal configuration may well vary across subjects depending on, for example, lesion site. In the present study, conducted in a clinical setting, we had no lesion information of all participants. For future studies, it will be interesting to investigate an individualized application of tDCS.
We thank all participating patients and all participating rehabilitation centers (Rijndam, Libra, Revant and De Hoogstraat). We thank Lisa Snakkers, Stefanie van Vugt, and Marieke Blom-Smink for performing measurements.
Sources of Funding
This work was funded by Erasmus MC Cost-Effectiveness Research and the Dutch Brain Foundation.
The online-only Data Supplement is available with this article at http://stroke.ahajournals.org/lookup/suppl/doi:10.1161/STROKEAHA.117.020197/-/DC1.
- Received September 30, 2017.
- Revision received January 21, 2018.
- Accepted January 29, 2018.
- © 2018 American Heart Association, Inc.
- Shah-Basak PP,
- Wurzman R,
- Purcell JB,
- Gervits F,
- Hamilton R
- Demeurisse G,
- Demol O,
- Derouck M,
- de Beuckelaer R,
- Coekaerts MJ,
- Capon A
- Polanowska KE,
- Leśniak MM,
- Seniów JB,
- Czepiel W,
- Członkowska A
- Volpato C,
- Cavinato M,
- Piccione F,
- Garzon M,
- Meneghello F,
- Birbaumer N
- Spielmann K,
- van de Sandt-Koenderman WM,
- Heijenbrok-Kal MH,
- Ribbers GM
- Kaplan E,
- Goodglass H,
- Weintraub S
- Goodglass H,
- Kaplan E
- Blomert L