Randomized Controlled Trial on Hemifield Eye Patching and Optokinetic Stimulation in Acute Spatial Neglect
Background and Purpose—Right hemisphere stroke patients frequently experience spatial neglect, a severe lack of awareness for contralesional hemispace. Although neglect counts among the strongest predictors for poor functional outcome after stroke, there is no established therapy, particularly not for the acute stage.
Methods—In a randomized controlled trial, we compared the combined treatment of hemifield eye patching and repetitive optokinetic stimulation in acute stroke patients with neglect to the spontaneous course. Outcome measures were a neuropsychological test battery for neglect as well as scales of functional independence and clinical impairment. Outcomes were assessed at baseline (day 1), post treatment (day 8), and at 1-month follow-up (day 30).
Results—Final analysis included 21 acute right hemisphere stroke patients with neglect (23 enrolled, 2 lost to follow-up) allocated either to the treatment (1 week hemifield eye patching and daily sessions of optokinetic stimulation, n=11) or the control group (no neglect-specific treatment, n=10). At baseline, both groups did not differ in neuropsychological test performance, clinical impairment, or functional disability. At the post treatment session, both groups had improved in all these measures, and results were stable or further improved at follow-up. However, there was no significant difference in this change between the treatment and the control group.
Conclusions—An early intervention of combined hemifield eye patching and optokinetic stimulation in acute stroke patients with spatial neglect has no additive effect to the spontaneous remitting course of the disorder.
Spatial neglect represents a severe cognitive disorder affecting two thirds of patients with acute right hemisphere stroke.1,2 Although there is spontaneous remission in some patients within days, others show persisting neglect signs even after years.2,3 Because of the profound lack of awareness for contralesional hemispace, neglect patients are severely impaired in the activities of daily living.2,3 In fact, spatial neglect constitutes one major predictor for poor functional outcome after stroke.3,4
Despite promising experimental approaches, an established therapy for neglect in clinical practice is still lacking.5 There is a clear need for a neglect treatment that is easily applicable, well tolerated, and effective, inducing long-lasting improvement not only in neuropsychological test performance but also in patients’ functional independence.
We aimed to establish such an intervention for acute stroke patients with neglect by combining 2 bottom-up stimulation techniques: hemifield eye patching (HEP) and repetitive optokinetic stimulation (OKS). HEP is supposed to reduce the salience of hyperattended ipsilesional stimuli and thereby counteracting the interhemispheric attentional imbalance in neglect patients.6,7 HEP may be regarded as a visual type of constraint-induced (forced use) therapy, an approach proven effective in motor rehabilitation.8 OKS, inducing contralesional smooth pursuit eye movements and a visual motion percept, is supposed to work as a correcting input signal for the ipsilesionally shifted subjective egocentric midline in neglect patients.9,10
We a priori hypothesized that HEPOKS in acute stroke patients with neglect induces greater remission of neglect than in the spontaneous course. Beneficial effects, reflected by improvements in neuropsychological tests and functional disability scales, were expected to develop during the treatment period and be persistent at follow-up.
The randomized controlled trial was conducted at the Department of Neurology, University Hospital Schleswig-Holstein, Lübeck, Germany. The study was approved by the local Ethics Committee (#12–064). Eligible were all adults admitted to our stroke unit showing left spatial neglect after acute (<14 days) right hemisphere stroke. Patients were diagnosed as having neglect if they showed pathological performance in at least 2 tests of a neuropsychological test battery11 consisting of the following paper–pencil tests: Line bisection, Star cancellation, Text reading, Bells cancellation, and Ogden figure copying task (see online-only Data Supplement). Exclusion criteria were previous stroke, neurodegenerative disease, or inability to give informed consent.
After obtaining informed consent according to the Declaration of Helsinki, patients were randomized by a third party (Institute of Medical Biometry and Statistics, University of Lübeck) and allocated to either of 2 parallel groups. Patients in the treatment group received HEPOKS in addition to the usual stroke care (physio-, speech, and occupational therapy), whereas patients in the control group had usual care only. HEP was applied by spectacle frames containing noncorrective lenses of which the right half was patched with dark nontranslucent tape. Participants were instructed to wear the glasses all-day for 7 days and only to remove them for the OKS treatment sessions. Investigators, care providers, and patients’ relatives regularly checked on correct use of the glasses. The daily OKS sessions (15 minutes each) were applied at the bedside. Seventy colored geometric objects were coherently moving on an 18.4″ notebook monitor from right to left at varying speed (8–12°/s).
Outcome Measures and Assessments
We prespecified 2 primary outcome measures: (1) mean performance (accuracy) in the neuropsychological test battery and (2) neglect-related functional disability measured by the Catherine Bergego Scale.12
Secondary outcome measures were scores from the individual neuropsychological subtests and further scales of clinical impairment and functional disability (Barthel Index, modified Rankin scale, and National Institutes of Health Stroke Scale).
Participants were assessed at 3 time points (Figure 1): baseline (day 1), post treatment (day 8), and follow-up (day 30). The follow-up assessment took place at 1 of 3 Clinics for Neurological Rehabilitation in Lübeck’s surrounding area.
Statistics were performed using SPSS 22.0 (IBM Corp). The design of the study required 2×3 ANOVAs with repeated measures, with group (treatment versus control) as between-group factor and session (baseline, post treatment, and follow-up) as within-group factors. In case of significant main effects, the ANOVA was followed by post hoc t tests investigating absolute differences (Bonferroni corrected).
Additional details concerning methods, randomization process, and sample size calculation can be found in the online-only Data Supplement.
Recruitment and Baseline Characteristics
From 23 patients randomized (Figure I in the online-only Data Supplement), 2 patients were lost to follow-up, leaving 21 evaluable patients for final analyses (n=11 treatment, n=10 control). At baseline, study groups did not differ significantly in demographic, clinical, and neuropsychological characteristics except for 1 paper–pencil subtest (Table). Lesion overlap analyses are provided in Figure II in the online-only Data Supplement.
Figure 2 shows the group results for the 2 primary outcome measures. Table I in the online-only Data Supplement provides the statistical analyses from the according ANOVAs.
For neuropsychological test accuracy, there was no significant main effect of group or the interaction group*session, but only of session (F(2,18)=32.8, P<0.001). In both groups, neuropsychological test accuracy improved similarly between baseline and post treatment (treatment: d=21±4%, P<0.001; control: d=19±6%, P<0.05). Between post treatment and follow-up, there was additional improvement in the treatment group (d=14±4%, P<0.01) and stable effects in the control group (d=3±4%; P>0.05).
For Catherine Bergego Scale score, there was no significant main effect of group or the interaction group*session, but only of session (F(2,18)=45.2, P<0.001). Within both groups, the Catherine Bergego Scale score decreased equally between post-treatment and follow-up session (treatment: d=−9.0±1.9, P<0.01; control: d=−8.2±1.8, P<0.01).
Primary outcome analyses separately for each participant revealed that there was significant improvement in individual patients, independent of the allocated intervention (Figure III in the online-only Data Supplement). This reflects an interindividual variability in the spontaneous remission of neglect and not a dichotomy of responders and nonresponders to treatment.
Results of the secondary outcome measures are provided in Figure IV and Table I in the online-only Data supplement.
In this randomized controlled trial on acute stroke patients with spatial neglect, we compared an early, 1-week intervention of HEP and OKS (treatment) to the spontaneous course (control). Patients in both groups showed an equal improvement in neuropsychological test performance and neglect-related functional disability over time, independent of the allocated intervention. Data indicate that the HEPOKS intervention had no additive effect to the spontaneous remission of neglect in our acute stroke patients.
Did the study or the intervention fail?
The study fulfilled all requirements of a randomized controlled trial, for example, applying adequate randomization, comparing a treatment to a no-treatment group with similar baseline characteristics and assessing established and relevant outcome measures. By contrast, the positive result of a previous study on HEP in subacute neglect patients was derived from the parallel improvement in 2 experimental groups (HEP versus Visual Scanning Training) without comparison to no-treatment.13 However, spontaneous remission of neglect must be necessarily taken into account before attributing any improvement to an intervention. Other positive studies on HEP did not refer to group differences at baseline which could have explained significant improvements.7,14
But is our study result really a proof for failure of the therapeutic approach in general? There is good evidence for a beneficial impact of HEP and OKS when applied as monotherapies in chronic neglect patients.10,15 That the concurrent application of HEP and OKS caused interferences in our study, is still unlikely, as these interventions act complementary based on their theoretical rationale.6,9
The acute stage of stroke may be better explanation. Our patients were certainly not generally too ill for this bottom-up treatment; tolerance and adherence to therapy was good (no discontinuation). But there was strong spontaneous remission of neglect in individual subjects of the control group, which makes it a strong competitor for any intervention, especially in studies with a moderate sample size. However, neither absolute group values nor individualized analyses pointed to a clinically relevant effect of the HEPOKS intervention missed by our study that would have emerged in a larger study sample.
In this randomized controlled trial, the early intervention of combined HEPOKS had no additive therapeutic effect to the spontaneous remission of spatial neglect in acute stroke patients. We emphasize that this result does not exclude beneficial effects of this therapeutic approach in chronic neglect patients. In the absence of a neglect-specific treatment for the acute stage of stroke, affected patients should receive occupational therapy and physiotherapy as usual care.
Sources of Funding
Dr Machner was supported by a research grant from the Deutsche Forschungsgemeinschaft (MA 5332/1-1).
The online-only Data Supplement is available with this article at http://stroke.ahajournals.org/lookup/suppl/doi:10.1161/STROKEAHA.114.006059/-/DC1.
- Received May 7, 2014.
- Revision received May 7, 2014.
- Accepted May 16, 2014.
- © 2014 American Heart Association, Inc.
- Parton A,
- Malhotra P,
- Husain M
- Karnath HO
- Kerkhoff G,
- Bucher L,
- Brasse M,
- Leonhart E,
- Holzgraefe M,
- Volzke V,
- et al
- Ianes P,
- Varalta V,
- Gandolfi M,
- Picelli A,
- Corno M,
- Di Matteo A,
- et al
- Zeloni G,
- Farnè A,
- Baccini M