Repetitive Transcranial Magnetic Stimulation in Spatial Attention After Stroke

BACKGROUND: A frequent and disabling impairment in persons who suffered acquired brain injury after stroke is left hemispatial neglect (LHSN). LHSN is a spatial attentive syndrome characterized by a reduced ability to attend, perceive and consciously represent the left contralesional space, in the absence of a primary sensory deficit. Commonly, the LHSN is due to a lesion on the right inferior parietal cortex and/or on the right fronto-parietal areas. Furthermore in stroke, LHSN is often associated with a mixture of attention and motor deficits leading to complex cognitive and behavioral pictures, which may even interfere with the administration of standard cognitive treatments for LHSN. However, several studies already demonstrated the efficacy of rehabilitation treatments on patients with LHSN, in terms of reducing disability, using various rehabilitative techniques and assessment tests. The rehabilitation techniques can be distinguished, accordingly to the used tools in: cognitive treatments (i.e. visual scanning training) and non-invasive brain stimulation methods (i.e. r-TMS). Conventional cognitive treatments (CCT) for LHSN involve different types of exercises aimed at reducing attentive bias for the ipsilesional space and/or promoting awareness of contralesional space by a continuous training of the patient's spatial exploration abilities. Instead, the TMS is a non-invasive method to modulate excitability of the cerebral cortex. The frequency of TMS pulse determines the type of modulation on the cerebral cortex (i.e. inhibitory <1 Hz; excitatory> 5 Hz). The most used TMS paradigm for the rehabilitation of LHSN is the r-TMS (repetitive transcranial magnetic stimulation) and it is based on the model of the inter-hemispheric competition. According to the model, the cerebral areas responsible for spatial orientation are contralateral to the attended hemifield (i.e. the left cerebral areas are responsible for the spatial orientation towards the right hemifield) and exert a reciprocal inter-hemispheric inhibition. The left and right fronto-parietal cortices are part of a intra-hemispheric network, which together with a parietal inter-hemispheric pathway, control the spatial orientation. In particular, a lesion on the right parietal cortex causes disinhibiting of the left parietal cortex and therefore a pathological over-activation of the latter.

According to the model, this left over-activation inhibits the right contralateral neural activity, as an increase of the healthy hemisphere inhibition activity on the damaged hemisphere. For these reasons, in the LHSN, a pathological orientation or pathological attention bias is determined towards the right hemifield after right stroke. Often, the r-TMS treatments aimed at rebalancing the inter-hemispheric competition through a stimulation of the parietal-parietal pathway, in particular inhibiting the intact areas of the left parietal cortex. The results showed significant post-treatment improvements in several neuropsychological and behavioral tests. In particular, a recent study showed how low- frequency r-TMS on the intact contralesional parietal areas reduced the clinical symptoms of LHSN after 10 treatment sessions. Improvements were stable 15 days after the treatment.

Despite evidence of efficacy of single treatment on cognitive symptoms of LHSN, many authors agree that the use of different rehabilitative techniques on the same patient are useful for generalizing clinical improvement and have a greater potential in reducing long-term global disability. However, in the literature, only few study tested effects of combined interventions on LHSN. Here, we aim at assessing the efficacy of a novel rehabilitation protocol, based on r-TMS in combination with a CCT, based on visual scanning training AIMS

To compare the efficacy of a combined r-TMS and CCT intervention aiming at reducing cognitive symptoms of LHSN, in a sample of right Stroke patients with LHSN, within the context of an RCT.

To evaluate the long- term impact of the intervention To evaluate the effect of the intervention on other clinical measures (activity of daily living, attentive and motor functions) and on psychophysiological indices.

METHODS: Multicenter, Randomized Controller Trial, with blind assessments on pre-test, post-test and on 3 months follow-up.

POPULATION: Right stroke patients with clinical evidence of LHSN INTERVENTION: Inhibitory low-frequency r-TMS on the intact left parietal cortex combined with a conventional cognitive treatment (visual scanning training) for 2 weeks. Multidisciplinary assessment with clinical and psychophysiological indices.

CONTROL: SHAM group with a placebo r-TMS stimulation combined with visual scanning training for 2 weeks. Multidisciplinary assessment with clinical and psychophysiological indices.

OUTCOMES:

The primary outcome is represented by a specific assessment of cognitive symptoms in LHSN with the Behavioral Inattention Test (BIT)

The secondary outcomes will consider the impact of the intervention on other clinical and psychophysiological indices. In particular we will test:

Activity of daily living and everyday life independence, attentive functions and motor independence.

A psychophysiological index based on Visual-Evoked Potentials

(VEPs). In particular we will focus on the N1 component, a posterior negativity, which is altered in terms of latency and amplitude in LHSN population. In order to study effects of the r-TMS in the inter-hemispheric parietal pathway, we will extract indices of inter-hemispheric transmission time (IHTT) on the N1 latency and of spatial attention bias on N1 amplitude.

EXPECTED RESULTS: In the intervention and in the control groups we expect a significant improvement of the clinical (cognitive, activity of daily living, attentive and motor functions) and psychophysiological indices, because both groups will be treated with an effective rehabilitation protocol (visual scanning training).

However, unlike the standard methods of rehabilitation, the new techniques of magnetic stimulation such as r-TMS allow the execution of a cognitive task through the pre-empowerment of a specific network or neuronal circuit. This effect could facilitate experiential learning during cognitive training, with a richer and more articulated neural environment, as well as selectively stimulated according to the areas most involved in the lesion. In the case of LHSN the mechanism of empowerment concerns the preserved parietal areas and the inter-hemispheric connectivity. As a consequence, r-TMS could increase the "responsiveness" of the peri-lesional areas and of the inter- hemispheric connectivity during cognitive training, increasing its effectiveness with respect to the SHAM condition. Therefore, we expect a larger improvement on the intervention group, because the r-TMS could modulate the responsiveness of the specific spatial attention pathway. At three months, it is expected a stabilization of functional improvements.

Randomization method In order to ensure that each arm will contain an equal number of individuals, eligible subjects will be randomly assigned to one of the two groups with a blocked randomization method based on a block size of 4.

Assessor and patient blindness: obviously, the physician who will administer the r- TMS or the SHAM stimulation will not be blind to the randomization. However, to ensure a double-blind assessment, pre-treatment assessments will be performed prior to randomization. An assessor not aware of the patient randomization group will conduct post-treatment and follow-up assessments. Also the visual scanning protocol will be administered by therapists unaware of the patient randomization and patients themselves will be instructed not to reveal any information on the brain stimulation treatment received.

Case report form and data monitoring Specific case report forms (CRF) for every test in the assessments are already available and will be used. During the treatment the responsible physician will report any adverse event in the treatment-CRF. All information and results will be promptly reported on the electronic database. A researcher will be responsible for the electronic database, data analyses and will draft the single patient record at the end of all procedures.

Maximization of inter-rater reliability of assessment and treatment methods

In order to minimize biases deriving from inter-rater measurement errors, the following interventions will take place during the start-up stage:

1. Collegial assessment of voluntary patients by the various assessors and therapists in order to standardize the administration modalities and resolve discrepancies between scoring and treatment procedures.
2. Subsequent development of an "assessment manual" and of a "treatment manual" containing all information necessary to the administering and scoring procedures.

Cases lost at follow-up In case of lost to follow-up, the information available up to that moment will be considered. If participant is lost at follow-up, it will be analyzed on the basis of the "intention to treat" principle. Finally, the reason for quit the study will be reported for each participant lost at follow-up.

Sample size

The sample size was calculated using the following formula:

https://www.dropbox.com/s/y56umrt7z93en1w/Simple%20size%20formula.docx? dl=0

The sample size resulting from the formula calculated for the BIT is 25.2. Consequently, the minimum sufficient simple to reach the primary aim is, assuming approximately 10% of subjects lost to follow-up, 60 subjects (30 X group) recruited over a three-year period.

Data Analyses In order to evaluate cognitive symptoms of LHSN in the two groups (primary outcome), differences in the BIT will be analyzed, between the pre-treatment (T0), post-treatment (T1) and follow-up (T2) phases for both groups of patients (r-TMS + CCT group and SHAM + CCT group). For this purpose, analyses of covariance will be performed for the BIT scale using a mixed-model ANCOVA with a 2X3 design, where the "between" factor is represented by the randomization group (r-TMS + CCT, SHAM + CCT) and the "within" factor is represented by the assessment time (T0, T1, T2). Whenever necessary, Greenhouse-Geiser correction will be applied and corrected p-values will be reported. In the ANCOVA, an adjustment will be made for the pre-treatment values and other possible confounding factors such as age, gender and education of the participants. Besides p-values, effect sizes will be provided in order to assess the size of treatment effect. Similar analyses will be performed also for the secondary outcomes on clinical and psychophysiological measures. Data analysis will be performed using Matlab (The Mathworks Inc.) and SPSS (version 13).