Cerebral visual impairment (CVI) is a visual disorder caused by damage to the retrogeniculate pathway beyond the lateral geniculate nucleus. It is characterized by reduced visual acuity beyond what would be expected from abnormalities in the eye structure 1). It is the leading cause of visual impairment in children in developed countries and is also increasing in developing countries 1).
The prevalence has risen from 36 per 100,000 in the late 1980s to 161 in 2003. With improved survival rates of preterm infants and advances in perinatal care, the frequency of CVI may continue to increase.
The definition of CVI includes the following five key elements:
Spectrum of visual impairment: Caused by brain abnormalities affecting visual processing pathways
Impairment beyond ocular findings: Presents visual dysfunction greater than predicted from ocular pathology
Low- and high-order visual deficits: Appear as either or both, leading to disease-specific behaviors
Co-occurrence with neurodevelopmental disorders: Coexists with other disorders, but CVI itself is not a disorder of language, learning, or social communication
Delayed recognition: Neurological damage may not be recognized until growth occurs
The terms “cortical visual impairment” or “cerebral visual impairment” are used rather than “cortical blindness.” Since subcortical lesions such as periventricular leukomalacia are also included, the term “cerebral” is considered more accurate1). Recently, Costa et al. proposed “Central Visual Impairment” as a superordinate concept, classifying it into CoVI (cortical) and CeVI (cerebral).
QCan cerebral visual impairment be cured?
A
CVI is a permanent condition but not unchanging. In some patients, improvements in fixation, saccades, pursuit eye movements, as well as visual acuity, contrast sensitivity, and visual field have been observed. Improvement rates are reported to be 46–83%. For details, see the “Standard Treatment” section.
Visual impairment in CVI ranges from no light perception to normal visual acuity. Representative subjective symptoms are shown below.
Visual acuity impairment: Vernier acuity is more affected than grating acuity. In some cases, visual acuity improves under low illumination conditions.
Visual field constriction: constricted visual fields, Swiss cheese-like (scattered scotomas), often hemianopic defects
Simultanagnosia: inability to perceive multiple objects at once
Visual function is easily affected by environmental and medical factors. Seizures or illness can temporarily reduce visual function, and visually complex or unfamiliar environments increase visual difficulties1).
The crowding ratio (single optotype visual acuity ÷ linear visual acuity) was reported to be 2.0 or higher in 41% of the CVI group and 4% of the non-CVI group1).
Perinatal and postnatal hypoxic-ischemic encephalopathy (HIE) in term or preterm infants is the most common cause 1). About half of children with CVI are diagnosed with cerebral palsy, and it tends to be more common in boys.
Hypoxia-Ischemia
Term HIE: Infarction in the watershed areas (fronto-parieto-occipital). Caused by loss of autoregulation of vascular blood flow.
Preterm HIE: Periventricular leukomalacia (PVL). Commonly occurs between 24–34 weeks of gestation. Immature oligodendrocytes and subplate neurons are vulnerable to ischemia. In low birth weight infants, refractive errors such as high myopia and strabismus are frequent, and they often also have lower limb or quadriplegia and spatial cognitive impairment due to periventricular leukomalacia.
Intraventricular hemorrhage (IVH): High risk in preterm infants.
Infection/Inflammation
Meningitis: Accounts for 11.8–15% of CVI cases. Haemophilus influenzae tends to damage the occipital cortex and is the most common causative bacterium.
Infection mechanism: Due to thrombophlebitis, arterial occlusion, hypoxic-ischemic injury, venous sinus thrombosis, and hydrocephalus.
Other Causes
Hydrocephalus: Chronic stretching of the posterior cortex is a frequent mechanism. Shunt malfunction can also be a cause.
Trauma: Accounts for about 4% of cases. Shaken baby syndrome is a typical example.
Epilepsy: Infantile spasms (West syndrome) can cause CVI.
Congenital brain malformations: lissencephaly, schizencephaly, holoprosencephaly, etc. Metabolic diseases and hypoglycemia can also be causes.
In children with evidence of low vision despite normal structural eye examinations, CVI should be actively considered. It is important to suspect CVI from the neonatal period when risk factors can be identified.
Note that in children with cerebral palsy, eye movements themselves may be impaired, leading to the mistaken conclusion that they cannot see even though they can. In children with physical disabilities, visual responses may vary greatly depending on body position; when trunk stability is poor, visual responses decrease. Therefore, it is important to evaluate visual responses in a relaxed and stable state (e.g., while seated in a wheelchair or stroller) as much as possible.
MRI is the most important test for diagnosing CVI. The extent and location of damage help predict prognosis. Lesion patterns on MRI are broadly classified into three types:
Lesion pattern
Prognosis
PVL / brain cyst / brain atrophy
Difficult to improve visual function
Minor damage
Good prognosis expected
Diffuse brain atrophy
Limited improvement
MRI is always recommended for children with low Apgar scores.
VEP was once considered important for diagnosing CVI. However, due to mediation by the extrastriate visual system, normal flash VEP can be recorded even in CVI patients. Therefore, a normal flash VEP does not rule out CVI 1).
EEG was previously considered a valuable diagnostic tool, but with the widespread use of high-resolution imaging, its role in diagnosing CVI has diminished.
Machine learning-based eye tracking is expected as an objective evaluation method for visual processing in CVI. Using indicators such as fixation and saccade latency and frequency, it has been shown to distinguish children with CVI from controls with high accuracy (AUC ≥ 0.90). Combined with AI-generated saliency maps called SegCLIP, it enables quantification of gaze patterns for low-level and high-level visual features.
Currently, there is no established evidence-based treatment. The core of management is prevention, treatment of comorbid eye diseases, rehabilitation, environmental adjustments, and multidisciplinary collaboration.
Ophthalmic Management
Refractive correction: Prescribe glasses for comorbid refractive errors.
Strabismus surgery: Candidates are those with stable visual recovery and controlled neurological complications 1). Large manifest esotropia is an indication for early surgery. Intermittent strabismus is assessed after repeated evaluations. Typically, 15–20% undercorrection is planned (to prevent overcorrection leading to consecutive exotropia).
Rehabilitation
Visual stimulation therapy: Light reflex stimulation (shine a flashlight into each eye in a dark room, 1 minute × 30 times/day), shape recognition exercises.
Environmental adjustments: Minimize patterns, use a simplified environment with high-contrast colors. Double-spaced reading materials. Use of backlit devices.
Utilization of near vision: Near visual acuity is often better than distance visual acuity.
Low illumination environment: Some cases show improved visual acuity when ambient light levels are lowered.
Most children with CVI show some visual recovery, but improvement progresses slowly over several months. Improvement rates are reported to be 46–83%. However, 90% have residual visual impairment and qualify for rehabilitation services.
QCan rehabilitation improve vision?
A
Most children with CVI show some visual improvement over time, but 90% have residual visual impairment. Although visual stimulation therapy is recommended, no studies have shown it to be more effective than natural improvement. Environmental adjustments (high contrast, simplified environment) and multidisciplinary rehabilitation are recommended.
6. Pathophysiology and Detailed Mechanisms of Onset
In term infants, the border zones between the anterior and middle cerebral arteries and between the middle and posterior cerebral arteries are most vulnerable. Loss of autoregulation of cerebral blood flow due to hypoxia leads to hypoperfusion in these watershed areas, causing infarction in the frontal and parieto-occipital regions. The striate cortex, occipital visual areas, temporal lobe, and parietal cortex are also commonly involved.
Hypoxic-ischemic encephalopathy in preterm infants
In preterm infants, the deep periventricular white matter is primarily affected. Injury is most likely to occur between 24 and 34 weeks of gestation. A transient vulnerable watershed zone exists in the periventricular white matter, where long penetrating branches from the middle cerebral artery terminate from the pial surface to the deep periventricular white matter. Capillaries in this region are prone to hemorrhage due to hypoxia-ischemia.
Glial cells and neurons are produced from the germinal matrix and migrate to the cerebrum. Immature oligodendrocytes and subplate neurons located in the periventricular region are more vulnerable to ischemia than mature forms. These contribute to the characteristic pattern of injury in periventricular leukomalacia (PVL).
In CVI, dysfunction of the dorsal stream (where/how pathway) is more common than that of the ventral stream (what pathway)1). Dorsal stream dysfunction manifests as motion perception abnormalities (impaired detection of optic flow and biological motion) and visuomotor integration deficits (optic ataxia)1).
Damage to the developing thalamus contributes to CVI. Significant volume reduction has been observed in the entire thalamus, particularly in the lateral, anterior, and ventral thalamic regions.
A hypothesis for visual improvement suggests that early damage does not cause cell death but rather disrupts normal protein synthesis in neurons, leading to delays in myelination, dendrite formation, and synaptogenesis. Visual improvement in CVI patients may actually represent a form of delayed visual development.
A new method combining eye tracking with AI-generated saliency maps (SegCLIP) has been developed. It can quantify how children with CVI direct their gaze toward low-level and high-level visual features. This method has been validated against functional vision scores and may serve as a non-invasive, quantitative tool for monitoring and evaluating visual processing deficits in CVI.
Unsupervised data-driven clustering analysis identified three distinct CVI subgroups with different 1-year visual outcomes. In one of these groups, visual acuity significantly improved after 1 year. Stratification of patient populations using such methods may be useful for developing individualized intervention plans.
In the only published randomized controlled trial, intraventricular administration of fetal-derived neural stem/progenitor cells resulted in visual improvement of at least one Huo scale grade in 60% of the stem cell treatment group, compared to 33% in the control group 1). However, masking of participants and examiners was not performed, and adverse events including fever, cerebrospinal fluid leakage, and intracranial hemorrhage were observed. Further research is needed to establish optimal stem cell sources and delivery methods.
Chang MY, Borchert MS. Advances in the evaluation and management of cortical/cerebral visual impairment in children. Surv Ophthalmol. 2020;65(6):708-724.
Bauer CM, Merabet LB. Perspectives on Cerebral Visual Impairment. Semin Pediatr Neurol. 2019;31:1-2. PMID: 31548018.
Bauer CM, van Sorge AJ, Bowman R, Boonstra FN. Editorial: Cerebral visual impairment, visual development, diagnosis, and rehabilitation. Front Hum Neurosci. 2022;16:1057401. PMID: 36457755.
Copy the article text and paste it into your preferred AI assistant.
Article copied to clipboard
Open an AI assistant below and paste the copied text into the chat box.
