Volkening 2016.
| Study characteristics | ||
| Methods | Double‐blind RCT Setting: clinic, Germany | |
| Participants | 29 first‐ever right hemispheric, right‐handed stroke patients with signs of left‐sided spatial neglect (age 18+) Cathode left = 8, cathode right = 8, sham = 8 Neglect test: according to a cutoff score criterion of ≤ 135 for mild neglect or suspicion of neglect on the Neglect test (NET, adapted German version of the BIT) Age, mean (range): cathode left = 71 (55 to 80), right = 73 (61 to 83), sham = 70 (45 to 82) Sex (men/women): cathode left = 2/6, right = 4/4, sham = 4/4 Days between stroke and treatment, mean (range): cathode left = 1.9 (1.1 to 3.9), right = 1.3 (0.4 to 2.2), sham = 1.0 (0.7 to 1.5) Exclusion: any metal implants; brain tumour, previous epileptic seizure, craniotomy, degenerative or psychiatric disorder, unable to perform the NET | |
| Interventions | As standardised therapy, all patients received smooth pursuit eye movement training (SPT) and visual scanning training (VST). Both training programmes were presented on a 14.1‐inch laptop monitor (60 Hz refresh rate). For SPT, computer‐generated random displays of 350 dots (blue on a white background) moving coherently towards the left hemi‐space (speed: 6.9°/s) were presented. Patients were instructed to look at the displays and to make smooth pursuit eye movements towards the direction of motion and return to the rightward side of the screen whenever they had reached the leftward border of the screen. For VST, different training exercises from the therapy programme Cogpack were used to facilitate exploration of the left hemi‐space. VST programmes and their difficulty levels were adjusted individually depending on each patient’s capabilities. In each session, patients first received 2 to 4 runs of SPT, followed by VST Simultaneously, patients received GVS or sham stimulation. Bilateral bipolar GVS was delivered by a battery‐driven, direct current stimulator (neuroConn Ilmenau, Germany). Two electrodes (anode and cathode) were inserted into natrium‐chloride‐soaked sponges (30 cm² each) and placed over both mastoids. Polarity placements were changed for each of the 3 stimulation conditions: for CL‐G VS, the cathode was placed on the left and the anode on the right mastoid. This electrode setup was reversed for CRGVS. For CL‐ and CR‐GVS, the current was ramped up (in steps of 0.1 mA/s) to 1.5 mA, kept there for 20 minutes, and ramped down again (in steps of 0.1 mA/ s). Conforming to established safety limits, patients were stimulated for only 20 minutes with a current intensity of 1.5 mA. Apart from the intervention, patients received occupational therapy and physiotherapy but no other specific neglect training |
|
| Outcomes | Primary outcome measures
Secondary outcome measures: subjective visual (SVV) and haptic vertical (SHV) were used to assess verticality perception. All taken immediately post intervention with no longer‐term follow‐up |
|
| Notes | ||
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | 17 out of 24 patients were randomly allocated to treatment groups by the research co‐ordinator, who drew cards from a sealed envelope. Since patients with more severe spatial neglect might respond differently to interventions than those with milder impairments, we allocated 7 patients using minimisation (see Scott, McPherson, Ramsay, & Campbell, 2002, for further details). Minimisation was based on NET scores. The NET score range for inclusion was subdivided into 3 strata: 0 to 45, 46 to 90, 91 to 135. Minimisation was performed by a post‐doctoral researcher otherwise not involved in the study |
| Allocation concealment (selection bias) | Low risk | See above |
| Blinding of participants | Low risk | Patients were also blinded to the allocated add‐on intervention (GVS) |
| Blinding of personnel | High risk | Not possible |
| Blinding of outcome assessment (detection bias) All outcomes | Unclear risk | "Outcome measures were assessed by trained neuropsychologists, masked to treatment allocation and not otherwise involved in patients’ treatment." Assumption of unawareness was not tested |
| Incomplete outcome data (attrition bias) All outcomes | High risk | For SVV and SHV, there were more missing data points because some patients were not able to perform tasks at baseline (in total, SVV = 9; SHV = 5) due to impaired comprehension. Baseline measurements were therefore substituted with the group mean. Except for 1 case, all patients were able to perform the tasks after the intervention. Accordingly, LOCF was applied for missing data for the following time points |
| Selective reporting (reporting bias) | Unclear risk | Several primary outcomes |
| Other bias | High risk | Use of LOCF |