CVI Home
Articles and Videos
Research
APH Home
J.E. Jan
M. Groenveld
A.M. Sykanda
C.S. Hoyt
Originally appeared in Developmental Medicine and Child Neurology
Recent reports suggest that transient and permanent cortical visual impairment (CVI) is more common in children than previously thought (Whiting et at. 1985, Hoyt 1986, Roland et al. 1986). In our multidisciplinary diagnostic evaluation of children with ocular and cortical visual loss, we have noted a number of behavioral features which were characteristic of CVI. These have not been studied systematically before in the literature. The aim of this paper is to analyse the common behavioural features of 50 children with permanent CVI and their diagnostic significance.
CVI was suspected when the degree of visual loss was unexplained by the ocular examination alone. It was diagnosed in the presence of severe visual loss, normal or minimal ocular findings and clinical, electrodiagnostic and CT evidence of post-geniculate lesions involving the visual cortex. Children who had incomplete investigations, profound mental retardation, more than minimal ocular findings or transient CVI were not included in this study.
The selection criteria applied to a group of 130 children with CVI who were followed by the Visually Impaired Program at the Children's Hospital, Vancouver, B.C. They were diagnosed between 1983 and 1985 as having visual loss. This number represents slightly less than 10 per cent of the visually impaired children seen in the program. The 50 children selected for this study all had a multidisciplinary evaluation, including neurological, ophthalmological and psychological investigations, as well as physiotherapy and speech, and language assessments.
Measurements of distance visual acuity by formal techniques was impossible and/or inconclusive in most cases because of the children's lack of visual attention and variability of vision. Hence acuity was graded as no apparent vision' (0), 'light perception' (1), 'ability to see a colourful toy measuring five to six inches within three feet' (2), 'within 10 feet' (3), and 'beyond 10 feet' (4). The term 'useful vision' was applied to children who had more than light perception, but this does not imply that light perception is not beneficial. When CVI occurred during the pre- and perinatal period it was termed congenital, and acquired when it occurred later. Visual field testing, which was difficult, was done by moving colourful toys within the visual fields of these children and observing their reactions.
The ages of the children at the time of evaluation ranged from six months to 17 years (mean 5.5 years). 34 children had pre- and perinatal visual impairment; the other 16 acquired their visual loss before the age of three years. The causes of CVI and the results of CT investigations are shown in Tables I and II.
| N | ||
| Pre- and perinatal | ||
| Asphyxia | 24 | |
| Cerebral dysgenesis | 4 | |
| Cerebral haemorrhage | 4 | |
| Infection | 2 | |
| Acquired | ||
| Shunt failure | 7 | |
| Asphyxia | 4 | |
| Injury | 4 | |
| Dehydration | 1 | |
| Total | 50 | |
| N | ||
| Occipital infarcts | ||
| Bilateral | 18 | |
| Unilateral | 4 | |
| Cerebral atrophy | 12 | |
| Periventrieular leukomalacia | 10 | |
| Cerebral dysgenesis | 4 | |
| Normal | 2 | |
| Total | 50 | |
*Only abnormalities causing cortical visual loss are listed.
| N | ||
| Cerebral palsy | 44 | |
| Mental retardation | 38 | |
| Epilepsy | 29 | |
| Hydrocephalus | 16 | |
| Learning difficulties | 4 | |
| Myelomeningocele | 2 | |
*N=50; all children had additional neurological disorders.
Additional handicaps
All the children had associated neurological problems (Table III). Most were multiply handicapped and received complex supportive services. Eight children were in special classes for the handicapped and one child was attending a regular class, but required help. The rest were too young, or not educable. The severity of the handicaps tended to increase with decreasing visual acuity.
Visual acuity
Five children had no apparent vision, 16 had light perception, nine were able to see the colourful toys within three feet, eight others could see them within 10 feet and 12 could see them over 10 feet away. 33 children had no associated ocular pathology, 16 had minimal optic-nerve atrophy and one child had questionable optic-nerve hypoplasia.
Variable visual function
The visual function of all the children with residual sight varied from day to day and even from hour to hour. The visual acuity of only seven children could be tested with Snellen eye charts, but results changed from one examination to the next. Snellen charts are designed to measure the acuity of patients with ocular visual loss, and had severe limitations when used to test children with CVI. After obtaining a history of visual function from the parents, the most effective way of testing vision was simply to play with the children, to observe how far they could see and how well they used their sight.
Tiredness, a noisy environment, preoccupation with other activities, medication, illness or seizures caused major fluctuations in their visual function. All those patients with useful sight could see 'better' in a familiar environment and when it was explained what they should look for, and where. They all exhibited impaired visual attention and lack of visual curiosity. Some children were aware of distant objects but could not identify them, while others had more difficulty in spotting them but could interpret visual information better. Spontaneous visual activity tended to be of short duration, and visual learning appeared to be tiring. Some children closed their eyes while they were listening: with their eyes closed, their balance often improved. Three children who had vision beyond 10 feet actively looked away from people or events. All three had CT evidence of bilateral striate cortex damage.
Most of our patients with useful sight viewed objects at close range. Several children were unable to distinguish or count objects unless they were widely separated and could be viewed individually. They paid closer attention to one rather than to a variety of toys in front of them.
Visual self-stimulation
Only one patient exhibited visual self-stimulation, such as flicking fingers in front of the eyes against a light source. Seven children enjoyed staring into lights.
Use of touch
Thirty-one children regularly used touch to identify objects. Nine of these had light perception, nine were able to see colourful toys within three feet, six saw objects within 10 feet and another seven beyond 10 feet.
Reaching on visual cues
Thirty-three children reached for visual cues. 11 of them, after visually locating the object, consistently looked away to either side, usually with a slight downward gaze, during the act of reaching. They did not close their eyes. The CT scans of these 11 patients revealed occipital infarcts in seven, dilated occipital horns of the lateral ventricles in two, and marked periventricular leukomalacia in one. The scan was normal for one child.
Appearance
None of our patients 'looked blind', in contrast to children with congenital or early acquired ocular visual loss, but their faces were often expressionless and they lacked visual communications skills. Usually their eye-movements were smooth, though often aimless. They did not have sustained nystagmus or deep-set orbits. Only seven children showed occasional beats of nystagmus, all of whom had optic nerve defects acquired under the age of one year.
Colour perception
Nineteen of the 50 children were able to name colours (their ability to match colours was not studied). The other 31 could not be tested because they were too young, too handicapped, had no apparent vision, or had only light perception. Naming colors appeared to be much easier than naming objects or shapes.
| Head control | Useful vision | No useful vision |
| Good | 17 | 3 |
| Poor | 2 | 16 |
*N=38.
Visual field defects
Accurate visual field testing of our multi-handicapped patients was extremely difficult. Nine children appeared to have double, but unequally affected homonymous hemianopias, as their vision seemed to be better on one side. Six patients had severely restricted visual fields, all of whom had hypoxic-ischaemic damage to the striate cortex.
Light sensitivity
Three children were markedly photophobic, but one child's photophobia gradually disappeared several months after the initial insult. All three had cortical visual loss following neonatal asphyxia; their ocular examinations were normal, but they did not have electroretinograms. On CT scans two had markedly dilated occipital horns of the lateral ventricles and the other had bilateral occipital infarctions. Two of these patients had light perception only, while the other was able to see beyond 10 feet.
Head control
Thirty-eight patients were non-ambulant because of cerebral palsy. Among these, a close relationship was noted between good head-control and useful sight (Table IV).
Children who were sitting propped up with their heads hanging forward had very restricted visual environments, which was thought to impede the development of their vision. This in turn might have resulted in diminished motivation to raise the head and lack of facilitation for better head-control. In a pilot project, head-supporting collars were provided for four totally blind children with poor head-control to determine whether visual efficiency would improve. Preliminary data on two of the children suggest a positive benefit.
Orientation and mobility skills
Of the 12 patients who were functionally ambulant, 10 had sufficient sight to avoid obstacles, but only five were able to use their vision for close work. Two children frequently bumped into objects, one of whom consistently sustained bruises and cuts, and occasionally fractures. Another patient, who was almost totally blind for conscious visual analysis of his environment and for identification of motionless objects, could ride a bicycle without hurting himself. His brain damage was restricted to the striate cortex bilaterally. Parents occasionally commented that their children saw moving objects better than stationary ones, and occasionally that they 'saw' better when travelling in a car. The vast majority of our patients appeared to have difficulties with depth perception: their reach was inaccurate and they could not estimate distances. They were unsuited for training with a cane or dog because of their difficulties with spatial interpretation.
It is not clear why visual function, including visual acuity, is so inconsistent in children with CVI. However, variable performance is characteristic of brain-damaged children and appears to be a sign of cortical malfunction.
Children with CVI, even with minimal vision, are able to identify colours much more readily than shapes. Colour perception, in contrast to form perception, is represented bilaterally in the brain, so only extensive cerebral lesions will eliminate it. Furthermore, colour perception requires far fewer neurons than form perception (Wiesel 1982).
Of the children who were ambulant, twice as many were able to avoid obstacles as were able to use their vision for close work. The most plausible reason for this is the presence of an extra-geniculostriate (collicular) visual system (Perenin and Jeannerod 1978, Zihl 1980), which enables them to perceive motion but not to make a conscious visual analysis. Campion et al. (1983) summarised the evidence for and against the existence of extra-geniculostriate visual pathways in man, and their presentation was followed by numerous peer commentaries. On the basis of animal studies and case reports of adult patients, most investigators agree that while the primary visual pathway is for conscious analysis of the environment, there is another subconscious system which is concerned with orientation during travelling. This visual pathway is thought to involve the peripheral retina, the Y ganglion cells, the optic nerves, the superior colliculus, pulvinar and the occipito-parietal regions. This so-called 'collicular system' is connected to the primary visual pathway, but the exact cortical localisation is still unclear.
One-third of our patients who reached on visual cues turned their head to the side while reaching, as if they were using peripheral visual fields. This was also observed by Gordon (1968). Reaching towards visual stimuli can occur in the absence of geniculo-striate pathways. Weiskrantz et al. (1974) came to this conclusion after studying visual capacity in the hemianopic field, following a restricted occipital ablation. Perenin and Jeannerod (1978) have also analysed reaching to visual stimuli in patients who had hemispherectomies and unilaterally lacked the striate cortex. Benton et al. (1980) emphasised the role of the temporal crescent in the perception of movement: this crescent is the peripheral unpaired portion of the visual field, extending between 60° and 100° for fixation in the horizontal meridian. This visual field is represented at the anterior end of the striate cortex (Bosley et al. 1985). Analysis of the CT scans of our patients who turned their heads away during reaching could not differentiate these two mechanisms. It may be argued that while the use of the peripheral reties is important in movement-oriented visual functions, possibly the process of visual perception is so difficult for these children that it is easier to reach on propioceptive feedback generated by the arm movement. However, in that case they could simply close their eyes, but they do not, so this explanation probably can be dismissed.
It is not surprising that visual self-stimulation was rare among our patients, a contrast to children with ocular blindness, since the purpose of these mannerisms is to stimulate the visual cortex (Jan et al. 1983).
Most of our children with useful sight viewed objects at a close range, even though they lacked significant refractive errors. This mechanism was probably used to achieve linear magnification or to reduce complex visual information, i.e., 'less crowding'.
Visual field testing was exceptionally difficult. Nevertheless, moving colourful toys in their visual fields and observing the reactions of our multi-handicapped children gave many clues. Van Hof-van Duin and Mohn (1984), who studied visual field defects of children who had had cerebral hypoxia, related the severely constricted peripheral visual fields specifically to hypoxic-ischaemic brain damage. This was evident in our studies as well.
Three of our patients who developed CVI kept following neonatal asphyxia exhibited photophobia. Denny-Brown and Chambers (1976), after carrying out extensive experiments in Macaque monkeys, concluded that the animals shunned bright light in the absence of cortical areas 17, 18 and 19. It is most unfortunate that these three patients did not have electroretinograms because Nickel and Hoyt (1982) showed that hypoxic events can cause transient but notable electroretinographic changes in children. Furthermore, the retinas of comatose adult patients who had been on a respirator before they died showed selective, characteristic alterations (Foos and Rhodes 1984). Therefore the aetiology of the photophobia could have been cerebral or retinal.
Monkeys and cats with total lack of retinalgeniculo-cortical pathways have been trained from virtual sightlessness to a state of visual competence (Humphrey 1974, Norrsell 1983). All animal studies suggest that the immature brain has a greater potential for recovery than the adult brain. Zihl (1980) suggested that adults with visual-field defects can be visually rehabilitated to a limited degree, but Bailliet et al. (1985) concluded that restitution of visual fields after damage to the striate cortex in humans is not possible with existing methods. The ages of their patients ranged from 56 to 66 years; to our knowledge there are no similar studies on the rehabilitation of children with permanent CVI.
Accepted for publication 8th October 1986.
J. E. Jan, M.D., F.R.C.P.(C)*, Professor, Division of Child Neurology, University of British Columbia.
M. Groenveld, Ph.D.*, Developmental Psychologist, Children's Hospital, Vancouver, B.C.
A.M. Sykanda, B.S.R., M.A.*, Senior Physiotherapist, Children's Hospital,Vancouver, B.C.
C. S. Hoyt, M.D., Professor, Department of Ophthalmology, University of California, San Francisco.
*Visually Impaired Program, Children's Hospital, Vancouver, B.C.
Correspondence to Visually Impaired Program, Children's Hospital, 4480 Oak Street, Vancouver, B.C. V6H 3V4.
The common behavioral features of 50 children with permanent cortical visual impairment (CVI) are described. CVI is frequently associated with specific behavioural characteristics. The majority of these children have residual vision, but they all have variable and inconsistent visual performance, including visual acuity. They see better in familiar environments and when they understand what to look for and where to look for it. They often use touch to Identify objects. Their ability to identify colours is much stronger than their perception of form. Many turn their heads to the side when they are reaching. Nystagmus and visual self-stimulation are exceptionally rare. They appear to have great difficulty with the cognitive evaluation of visual perception in spatial terms. Head elevation is worst in those with least vision, and without head elevation the possibility of visual stimulation is further restricted.
Bailleit,R., Blood,K.M.T., Bach-y-Rita,P. (1985) 'Visual field rehabilitation in the cortically blind?' Journal of Neurology, Neurosurgery and Psychiatry, 48, 1113-1124.
Benton,L.,Levy,I.,Swash,M. (1980) 'Vision in the temporal crescent in occipital infarction.' Brain, 103, 83-97.
Bosley, T.M., Rosenquist, A.C., Kushner, M.,Burke, A.,Stein, A.,Dann, R.,Cobbs, W.,Savino, P.J.,Schatz, N.J.,Alavi, A.,Reivich, M.(1985) 'Ischemic lesions of the occipital cortex and optic radiations: positron emission tomography.' Neurology, 35, 470-484.
Campion,J.,Latlo,R.,Smith,Y.M. (1983) 'Is blindsight an effect of scattered light, spared cortex and near-threshold vision.' Behavioral Brain Science, 6, 423-486.
Denny-Brown,D., Chambers,R.A. (1976) 'Physiological aspects of visual perception. I: Functional aspects of visual cortex.' Archives of Neurology, 33, 219-227.
Foos,R.F., Rhodes,R.H. (1984) 'Respirator retina.' Archives of Ophthalmology, 102, 296-303.
Gordon, N. (1968) 'Visual agnosia in childhood.' Developmental Medicine and Child Neurology, 10, 377-379.
Hoyt,C.S. (1986)'Cortical blindness in infancy.' Pediatric Ophthalmology and Strabismus, 23, 235-243.
Humphrey,N.K. (1974) 'Vision in a monkey without striate cortex: a case study.' Perception, 3, 241-255.
Jan,J.E., Freeman,R.D., McCormick,A.Q., Scott,E.P., Robertson,W.D., Newman,D.E. (1983) 'Eye-pressing by visually impaired children.' Developmental Medicine and Child Neurology, 25, 755-762.
Nickel,B.L., Hoyt,C.S. (1982) 'The hypoxic retinopathy syndrome.' American Journal of Ophthalmology, 93, 589-593.
Norrsell,K. (1983) 'Visually guided behavior of cats in the absence of retino-geniculo-cortical pathways.' Acta Ophthalmologica, Suppl. 160, 1-99.
Perenin,M.T., Jeannerod,M. (1978) 'Visual function within the hemianopic field following early cerebral hemidecortication in man.' Neurophysiologica, 16, 1-13.
Roland,E.H., Jan,J.E., Hill,A., Wong,P.K. (1986) 'Cortical visual impairment following birth asphyxia.' Pediatric Neurology, 2, 133-137.
Van Hof-van Duin,J., Mohn,G. (1984) 'Visual defects in children after cerebral hypoxia.' Behavioral Brain Research, 14, 147-155.
Weiskrantz,L., Warrington,E.K., Sanders,M.D., Marchall,J. (1974) 'Visual capacity in the hemianopic field following a restricted occipital ablation.' Brain, 97, 709-728.
Whiting,S., Jan,J.E., Wong,P.K.H., Flodmark, O., Farrell,K., McCormick,A.Q. (1985) 'Permanent cortical visual impairment in children.' Developmental Medicine and Child Neurology, 27, 730-739.
Wiesel,T.N. (1982) 'The postnatal development of the visual cortex and the influence of environment.' Bioscience Reports, 2, 351-377.
Zihl,J. (1980) '"Blindsight": Improvement of visually guided eye movements by systematic practice in patients with cerebral blindness.' Neurophychologica, 18, 71-77.