Lien vers les tableaux de données probantes et la liste des références (en Anglais)
The most common type of visual perception disorder following stroke is visual neglect or inattention. Estimates obtained from a systematic review indicated that visual neglect was reported on average in 32% of patients following stroke. The range was wide with the lowest number coming from a prospective study, where assessments were conducted within 3 weeks following stroke in patients with suspected visual deficits, to 82%, when assessed within 3 days of stroke in an unselected sample of general stroke patients (Hepworth et al. 2015). Unilateral spatial neglect (USN) is being more commonly associated with lesions in the right hemisphere (affecting the left side of the body) compared to a left-sided lesion. The presence of neglect has been associated with longer lengths of hospital stay and slower recovery during inpatient rehabilitation (Gillen et al. 2005).
Therapeutic approaches to treat neglect include remedial approaches (e.g., visual scanning, feedback or cueing, virtual reality and mental practice), and compensatory approaches (e.g.prisms, half-field, eye-patching, limb activation). Azouvi et al. (2017) included the results of 37 RCTs in a narrative review assessing rehabilitation techniques for post-stroke spatial neglect, including both top down and bottom up approaches. The authors concluded that one rehabilitation approach cannot be recommended over another, and a combination of several methods may be most effective than a single method. They further concluded that the evidence levels associated with these interventions remain low due to small sample sizes, methodological bias, and contradictory results. Bowen et al. (2013) included the results of 23 RCTs evaluating a variety of cognitive rehabilitation programs compared with an active or inactive control in persons with neglect following stroke. While cognitive rehabilitation approaches were associated with significant improvements in measures of neglect, when measured immediately after the intervention (SMD=0.35, 95% CI 0.09 to 0.62, p=0.0092), they were no longer when measured at follow-up (SMD=0.28, 95% CI -0.03 to 0.59, p=0.079). These techniques were not associated with significant improvements in ADL performance, when measured immediately after the intervention (SMD=0.23, 95% CI -0.02 to 0.48, p=0.068), or at follow-up (SMD=0.31, 95% CI -0.10 to 0.72, p=0.14). In another systematic review, Lisa et al. (2013) reported that among almost all of the 15 included RCTs, there were improvements reported in both the experimental and control groups, but in only 7 trials were there statistically significant between group differences, in favor of the experimental group. Large effect sizes (d > 0.80). were found in only four studies of virtual reality vs. visual scanning training (VST) (d=0.90), somatosensory electrical stimulation + VST vs. sham +VST (d=1.63), TENS vs. control (d=0.87) and optokinetic stimulation vs. control (d=1.59), and individual and group mirror therapy vs. sham (d=2.84 and d= 1.25).
Other forms of treatment for spatial neglect and visual field deficits include the use of noninvasive brain stimulation. Kem et al. (2013) randomized 27 patients admitted for inpatient rehabilitation, with visuospatial neglect to receive repetitive transcranial magnetic stimulation (rTMS). Patients were randomized to receive 10, 20-minute sessions over 2 weeks of 1) low-frequency (1Hz) rTMS over the non-lesioned posterior parietal cortex (PPC), 2) high-frequency (10Hz) rTMS over the lesioned PPC, or 3) sham stimulation. Although there were no significant differences between groups in mean changes in Motor-Free Visual Perception Test, Star Cancellation Test or Catherine Bergego Scale, there was a significant difference among groups in Line Bisection Test change scores (p=0.049). Post-hoc analysis indicated the improvement was significantly greater in the high-frequency rTMS group compared to sham-stimulation group (-36.9 vs. 8.3, p=0.03). Additionally, improvements in mean Korean-Modified Barthel Index scores in both the high and low frequency groups were significantly greater compared to those in the sham stimulation group (p<0.01 and p=0.02, respectively). Yang et al. (2017) reported improvements in mean Behavioural Inattention Test (BIT)-Conventional, following treatment with rTMS, when treatment was combined with a sensory cueing device worn on the left wrist.