Rehabilitation of visual field defects after stroke (CVA) using Virtual Reality vision training

Occipital cortical damage after stroke (CVA) leads to half-sided blindness,
with consequential loss of mobility (no permission to drive a car, but also
bumping into objects/people when walking) and limitations in reading, media use
and social interaction.
CVA incidence (ca 41.000 per year. Source: Hartstichting) leads to occipital
damage in approx. 30% of the patients. Prevalence of CVA in The Netherlands is
approx. 452.000 (https://www.volksgezondheidenzorg.info). These numbers are
expected to rise as a consequence of increased lifespan and improving
care/treatments after stroke.
At the moment, there is no validated method for reducing a visual field defect
in stroke patients. We developed a home e-training (duration: 4-6 months) and
tested it in approx. 80 patients with a (partial) recovery in approx. 70% of
the participants. We found significant improvement of the quality of life
related to the size of the visual recovery. Some salient results were regaining
flight authorization after training by 3 pilots, regaining driving license by
professional drivers en re-entry as teacher.
It is therefore our opinion that gradual improvement by reorganisation and
optimalisation of cortical processing (Perceptual Learning) as a result of
visual training can reduce the burdens of the visual field disorder (homonymous
visual field defect: HVFD; a.k.a. hemianopia).
Our method is currently only experimentally (participation in studies) or
against cost price available by the HemianopsieStichting (approx. 20 patient
per year; over 1700 visitors of our website in 2017).

1. In two earlier pilot RCT studies we made use of in-house developed hardware:
the "iBox". We want to enhance comfort and ease of use of the hardware, that
basically consists of a large box containing a computer, screen, mouse and
keyboard. However, training equipment should be portable and easier to use,
both for patient and therapist. We therefore developed a small training module,
which is basically a virtual reality headset: the "visionTrainer". We wish to
establish whether the visionTrainer yields the same or better results as the
iBox.

2. We also want to establish what the effects of training are when parts of the
defect and the intact field are trained simultaneously. More specific, we want
to see whether small negative training effects, which occurred sometimes in the
case that intact training followed defect training, are no longer observed
after training both fields simultaneously. This approach ensures that training
is the same during the training-period; only the targeted visual areas differ.

Every patient constitutes his/her own control subject in a double-training
paradigm. Every patient completes two blocks of training. During the first
block, training is aimed at 1 half of the defect plus its corresponding
diametrically located part in the contralateral visual field. During the second
block of training the other half of the defect plus its corresponding
diametrically located part in the contralateral visual field is trained. The
order in which a patient carries out the training is randomised.

Training of the defect leads to reduction of the defect. Training of the
contralateral intact visual field, however also has very small effects on the
defect. These small effects are positive if intact-training precedes
defect-training, but are negative if defect-training precedes intact-training.
To avoid these small negative effects, training will in this study be directed
simultaneously on a defect part and an intact par of the visual field.

Each training program will be preceded by two Baseline measurements, carried
out with an intermediate period of 8 weeks. This approach serves to support the
assumption that the phase of natural or spontaneous recovery has ended.
Directly after completing the second baseline measurement, the training
equipment is set up for the patient, the training instruction is delivered and
a hard-case with the hardware (visionTrainer headset, laptop) is given for use
at home.

The patient trains at home for 1 hour per day, 5 days per week. The hour per
day can be split into 2x 30 min or 3x 20 min if desired. The patient makes
his/her own choice where to sit or lay down to perform the training.

The measurements that will be carried out (baseline 2x, intermediate and
post-training) are: Humphrey perimetry, Goldmann perimetry, reading test (all 4
visiting days) en Goal Attainment Scaling (visiting days 2, 3 and 4)

The training exists of the presentation of elementary visual stimuli in a wide
border-area between the intact and defect visual field. The stimuli are white
dots, circling clockwise or counter clockwise in a 'patch' of varying size that
is determined by the Cortical Magnification Factor. Because of this, every
stimulus covers an equal part of cortical tissue.
The stimuli are presented alternately on different locations in the defect. The
patient must direct his/her gaze to a central fixationpoint and at the same
time direct visual attention towards the defect. This is the same as 'looking
out of the corner of your eye' to the stimuli that are presented and decide
whether the stimulus can indeed be detected and perhaps even discriminated. The
'looking out of the corner of your eye' is called 'covert attention shift' and
this forms the instruction to the patient. Performing a covert attention shift
is required to reach omprovement by training.

The training is performed at home using the in-house developed trainingmodule
"visionTrainer". The visionTrainer is a virtual reality headset. The headset
can -as the name suggests- be worn on the head like a diving- or skimask. This
has the benefit of a much more comfortable position during training in
comparison with the currently used "iBox", behind which the patient must take a
seat, just like with a perimeter. That way, the training duration will not be
limited by possible physical discomfort.
The visionTrainer is connected with a MacBook Pro laptop by a 5 meter long
USB-cable. The laptop presents the training-software on the visionTrainer. De
patient operates the visionTrainer through a wireless mouse.

In previously conducted studies we have found that virtually all patients are
capable of meeting the requirements of training 1 hour per day, 5 days per
week. In some cases a patient chooses to split up the hour of training into 2x
30 minutes or even 3x 20 minutes. Sometimes a patient appears capable of
training even more than 1 hour per day.
Perimetry and eye tracking are a general routine within ophthalmological
departments and form no risk.