Cortical visual impairment (CVI) is bilateral visual impairment caused by damage to the posterior visual pathway, the visual cortex, or both.1–5 Clinically, it manifests as a bilateral reduction in visual acuity with normal ocular structures, normal pupil responses, and an absence of nystagmus.1–5 The most common etiology for CVI is hypoxic brain injury.1,2 Other causes include head injury, infection, hydrocephalus, and metabolic disorders.3–5 In children, hypoxic brain damage is often secondary to perinatal hypoxia, a decrease below normal of fetal oxygen levels before delivery from the 28th week of gestation through the first 7 days after delivery.1,2 One study found that at least 60% of children with hypoxic-ischemic encephalopathy (HIE) have CVI.6 The watershed areas of cerebral circulation (extreme peripheral vascular areas with marginal blood flow) are the regions most affected by an ischemic insult (local anemia due to mechanical obstruction of the blood supply).7 Because the watershed regions differ in premature and full-term infants, ischemic brain injury will involve different areas of the brain in a child who is born prematurely as compared with a full-term infant.7 Specifically, the periventricular area is a transient watershed zone in premature newborns. Therefore, periventricular leukomalacia (PVL), defined as damage or softening of the white matter around the ventricles, is a characteristic finding in premature children.7 In children with PVL, the vision loss is primarily due to damage to the optic radiations. In full-term infants, however, the watershed zone is between the anterior and posterior cerebral arteries.7 Therefore, ischemic injury usually involves the frontal or parieto-occipital regions in full-term infants.7 In these children, the visual cortex is particularly vulnerable to hypoxia and hypotension.7 Because both cortical and subcortical damage can diminish vision, some prefer the name cerebral visual impairment.7–9
Because CVI is frequently associated with premature birth and nearly 12% of children in the US are born prematurely,10 it is easy to understand why CVI has become a common cause of bilateral vision impairment in children. Furthermore, the incidence of CVI is increasing, and it is now the leading cause of bilateral vision impairment in children in industrialized nations.2,8 This finding reflects medical advances in pre- and perinatal care, which have increased the survival rate of premature infants and children with birth complications.5,8 Perinatal hypoxia and the other etiologies of CVI historically resulted in a higher infant mortality rate.5,8 The increased survival rate of premature infants with a history of birth complications has increased the infant morbidity rate, specifically brain damage and consequent CVI.
The severity of vision loss in children with CVI ranges from a mild deficit to severe loss. Children with CVI are rarely completely blind and retain residual vision in various forms; e.g., preservation of some visual acuity, visual field, or color perception. In addition to central vision loss, it has been suggested that other aspects of higher-order spatial vision11 and visual perception are likely to be affected as well.12 Interestingly, some children with CVI improve over time whereas others show no change in vision function.13,14 The consequence of brain injury to the developing brain of an infant or child may be substantially different from that in an adult.7 Children appear to be more likely to recover from CVI than do adults.15 It is thought that most children will not ever regain normal vision, but they will often show some degree of improvement.16 However, very little is known about the mechanism of recovery or the specific prognosis of CVI. Current literature reports great uncertainty and variability concerning the natural history of CVI.3,11 It is currently thought that the visual prognosis for children with CVI may depend on the etiology, age of onset, severity, and type of brain damage.2 In general, children with CVI who have extensive neurological damage have a less favorable prognosis for recovery of vision. Additionally, PVL and subsequent damage to the optic radiations may be an indicator of poorer prognosis.3
Lim et al.17 recently reported visual evoked potential (VEP) acuities for children with CVI caused by HIE that they followed (median duration: 2 years, range: 6 month to 6 years). They found that VEP acuity at the last visit was, on average, 1 octave (a factor of 2) better than that at the first visit, with a rate of improvement lower than normal. However, the range of change between visits was −0.22 octaves (small decrease in visual acuity) to 3.09 octaves (large improvement in visual acuity). It is unclear how many patients improved and how many remained stable.
Hypoxic brain injury is rarely isolated to a single area and function of the brain. Consequently, many children with CVI also have cerebral palsy, seizure disorders, and communication disorders. This makes assessment of their remaining vision function very difficult. Measures of visual acuity that require a behavioral response from the child may be affected both by the child's visual impairment and by their performance ability. Therefore, assessment of visual acuity in children who are unable to communicate may be limited to objective, electrophysiological techniques such as the VEP measure of grating acuity (i.e., a determination of the finest stripe size that can be resolved by the visual cortex).18 The VEP provides general information about geniculocalcarine function and occipital responses to visual stimuli. Because the VEP represents a neural response, it is possible to obtain a visual acuity measure on a child who is not verbal and/or unable to generate a motor response to communicate what they see.
The current literature reports great variability in the prognosis of CVI and few quantitative evaluations of vision function are available. The purpose of this study was to evaluate change in vision function in children with CVI over time using a quantitative assessment of vision function. A longitudinal evaluation with a quantitative measure of vision, such as the VEP, will provide valuable information regarding the natural history of CVI, and may be beneficial in determining the prognosis for visual recovery in children with CVI. In addition, by characterizing the way CVI affects vision function over time, we hope to help parents and educators better understand their child's visual abilities and disabilities.
Threshold visual acuity and contrast sensitivity of children with CVI were assessed over time using the sweep VEP.18 Thirty-nine children participated in the visual acuity assessment and 34 of the 39 children participated in the contrast threshold assessment. The experimental protocol was approved by the Committee of the Protection of Human Subjects at the University of California, at Berkeley, and informed consent was obtained from the parents or legal guardians. Each subject received a complete eye exam before data collection to determine eligibility and refractive error. Significant refractive error was corrected before acuity testing. Additionally, children with ocular pathology or whose reduction in vision may be attributed to a condition other than CVI were excluded.
A summary of the study participants and etiology of CVI are listed in Table 1. The presumed causes of CVI are as follows: HIE (n = 14), PVL (n = 7), hydrocephalus (n = 4), hypoxic event in childhood (n = 4), infection (n = 3), trauma (n = 2), static encephalopathy (n = 1), abnormal white matter (n = 1), hemorrhage at birth (n = 1), and two cases with an unknown origin. Six of the seven children who were diagnosed with PVL were born preterm. The average gestational age of the preterm children was 31 weeks (range: 27 to 34 weeks). Thirty-one of the remaining 33 children were born full-term; the gestational age at birth was unknown for two.
At the time of the first VEP, the children ranged in age from 1 to 16 years (mean: 5.0 years; SD: 3.2 years). Each subject was tested at least twice. For patients who were seen multiple times, only the first and last measure was included in the data analysis. The time between measures ranged from 0.6 to 13.7 years (mean: 6.5 years; SD: 3.5 years). For the 34 patients who participated in the contrast threshold assessment, both measures (visual acuity and contrast threshold) were measured during the same testing session. Improvement was defined as a 0.2 log unit change in log MAR (logarithm of the minimum angle of resolution) for grating visual acuity or 0.2 log unit change in contrast threshold. This 0.2 log unit change cutoff was chosen based on the average test-retest reliability of the sweep VEP reported in the literature. Norcia and Tyler18 reported the reliability of VEP acuity measures to be 0.38 octave or 0.11 log MAR. Prager et al.19 reported 0.68 octave (0.20 log MAR), and Lauritzen et al.20 reported 0.13 octave (0.04 log MAR). Therefore, the average test-retest reliability of the VEP acuity measure reported in the literature is 0.40 octave or 0.12 log MAR (approximately one line on a standard chart). The 0.2 log unit cutoff chosen in this study conservatively requires a change of approximately 2 standard deviations of the VEP reliability to be considered a change in visual acuity. Lauritzen et al.20 reported the test-retest reliability of the VEP measure of contrast threshold to be 0.56 octave or 0.17 log MAR. Even though the repeatability of the VEP measure of contrast threshold might be slightly worse than the VEP measure of acuity, the 0.2 log unit cutoff was used for the analysis of both measures of vision function.
The sweep VEP was chosen to provide a quantitative measurement of vision because it offers several advantages for the CVI study population. The majority of the study participants was nonverbal and had significantly compromised motor function. The lack of verbal and motor ability makes a reliable behavioral measure of vision in these children extremely difficult, if not impossible. The VEP may be the most accurate method of testing this population because it provides a quantitative measure of vision and does not require a verbal or motor response from the patient.
Visual Acuity Measurement.
The children were presented with 80% Michelson contrast sinusoidal luminance gratings on a video monitor with an average luminance of 80 cd/m2. Over a period of 10 s, the spatial frequency of the reversing grating was incremented in 19 steps, spanning a 10 to 1 range of spatial frequencies expressed in cycles/degree. Visual acuity was initially swept linearly. But, because of the severely reduced acuity values in the study population, spatial frequency often had to be swept logarithmically. By using a log sweep, more time is spent testing at lower spatial frequencies. The viewing distance was initially 50 cm to present a range of spatial frequencies from 0.6 to 5 cycles/degree. If significant VEP responses remained at the end of this sweep range, the viewing distance was increased to 100 cm (for a 1 to 10 cycles/degree sweep range) or 200 cm (for a 2 to 20 cycles/degree sweep range). The gratings were modulated at a rate of 12 contrast reversals per second. At this rate, the evoked response is nearly sinusoidal in form and therefore only two parameters, amplitude and phase, are needed to describe the resulting waveform. The waveform was analyzed by a discrete Fourier transform at the 12 reversal per second stimulation frequency and compared with an adjacent frequency band at 14 Hz which did not contain any visual evoked activity. The adjacent frequency measurements were used to estimate the noise background during the trial and reject portions of the record with muscle spikes or movement artifacts. Multiple sweeps were recorded and the best one was selected.18 Acuity was determined by linear extrapolation of the final portion of the VEP amplitude to zero amplitude.18 VEP grating acuity (in cycles/degree) was converted into log MAR for data analysis, using the conversion that 1 cycles/degree is mathematically equivalent to 20/600 (log MAR = 1.48) and 30 cycles/degree is mathematically equivalent to 20/20 (log MAR = 0). The authors do not imply that grating acuity is equivalent to Snellen acuity since it is well known that grating acuity underestimates vision loss compared with Snellen equivalent in persons with visual impairment.21
Contrast Threshold Measurement.
The sweep VEP was used to measure contrast threshold for a single, low-spatial frequency grating. The children were presented with a 0.6 cycles/degree vertical grating that was swept logarithmically from 1 to 80% Michelson contrast. Over a period of 10 s, the contrast of the reversing grating was incremented in 19 steps. Again, multiple sweeps were recorded and the best one was selected.18 The contrast threshold was derived by linear extrapolation of the final portion of the VEP amplitude to zero amplitude.
The study participants sat on their parent's lap or in their own wheelchair, during testing. The experimenter initiated each trial and monitored the child's fixation. The experimenter encouraged the participant's fixation by singing and/or dangling a small noisy toy in front of the video monitor. The experimenter was able to pause and resume the trial with a computer mouse device if the child looked away.
As previously stated, a primary goal of this study was to assess the role of various factors on the improvement of vision function in children with CVI. Factors that may contribute to the improvement of vision are the degree of initial impairment, the age of the child, the duration between measures, and the etiology of CVI. The association between the improvement in vision function and the individual initial value (either acuity or contrast), individual age of the child, and the individual duration between measures was analyzed with a two-tailed t-test. ANOVA was used to analyze the relationship between the etiologies (HIE, PVL, hydrocephalus, trauma, infection, etc.) and the visual outcome. Finally, the correlation between acuity improvement and contrast improvement was assessed.
All the subjects had substantial visual impairment. The mean initial acuity for the entire study population was 0.88 log MAR (20/152 grating visual acuity). The initial and final visual acuities for the group as a whole were analyzed using ANOVA and showed a significant improvement (F = 3.97; df = 38; p = 0.004). The group was then split into those who improved and those who did not based on the criterion of a change of 0.2 log units or more. Nineteen of the 39 children (49%) showed improvement, 19 (49%) remained stable, and one (2%) showed a reduction in visual acuity. In those who improved, the average amount of improvement was sizeable (0.43 log unit: 20/205 to 20/76 grating acuity). The rate of change for those who improved was 0.07 log unit per year. The average change in acuity for those who did not improve was −0.03 log unit: 20/112 to 20/121 grating acuity. The rate of change for those who did not improve was less than −0.01 log unit per year.
Figure 1(a, b) graphically displays the change in visual acuity over time for the two groups of children. Figure 1(a) shows the grating acuity threshold as a function of age in those who improved. The y-axis is log MAR. A higher point on the vertical axis indicates a better visual acuity. Therefore, the upward trend (positive slope) shows the improvement in visual acuity over time. The dark black line is the mean change in visual acuity for all the children who improved. Figure 1(b) shows the grating acuity threshold as a function of age in those who did not improve. The line with a negative slope is the graphical representation of one child who showed a reduction of visual acuity. This child had a period of severe and uncontrolled seizure activity (at 5.5 years of age) in the time between the two measures (first measure at 1.8 years and final measure at 10.83 years). He may have suffered further brain damage secondary to this seizure activity that caused further loss of visual acuity. Other than this child, it is striking how stable the children's visual acuities are over time. In those who did improve, the improvement was substantial and in those who did not improve, their vision was strikingly stable.
Even though only 49% of the children improved, the improvement was sizeable in those who did improve. Therefore, it would be extremely helpful if we could better understand the differences between these two groups and predict which children would improve over time. In an effort to better understand the differences between these two groups, the effect of the child's age at the first test, the time between VEP measures, and the initial visual acuity and the etiology of CVI were evaluated as possible factors affecting the prognosis.
The initial visual acuity was significantly worse in the group that improved compared with the group that did not improve (t = 2.8; df = 37; p = 0.007). The mean initial acuity in the group that improved was 1.01 log MAR or 20/205 (95% CI: 0.87 to 1.14 log MAR) and the mean initial visual acuity in those who did not improve was 0.75 log MAR or 20/112 (95% CI: 0.62 to 0.88 log MAR).
The final visual acuity measure was better for those who improved (0.58 log MAR or 20/76 grating acuity) compared with those who did not improve (0.78 log MAR or 20/121 grating acuity) (t = −2.5; df = 36; p = 0.016). So even though the children who improved had a worse initial visual acuity, their visual acuity improved to surpass the initial and final acuity of the children showing no improvement.
Statistical analysis revealed no relationship between the age of the child at the first VEP and the change in visual acuity (t = −0.9; df = 36; p = 0.371). The mean age for the group who improved was 4.5 years (95% CI: 2.9 to 6.1 years) and the mean age for the group that did not improve was 5.4 years (95% CI: 4.0 to 6.8 years).
There was no relationship between the duration between the two measures and the change in visual acuity (t = 1.4; df = 37; p = 0.166). Children who improved had 7.31 years on average between the two measures (95% CI: 4.37 to 7.67 years) and children who did not improve had 6.01 years on average between the two measures (95% CI: 5.67 to 8.95 years).
Most surprisingly, there was no relationship between the etiology of CVI and the child's improvement in visual function (F = 0.28; df = 38; p = 0.60). Of the 14 patients with HIE, six improved and eight did not improve. Of the seven patients with PVL, two improved and four did not improve. Of the four patients with a hypoxic event in childhood, two improved and two did not improve. Of the two patients with CVI due to an unknown etiology, one improved and one did not improve. The following etiology categories had a consistent prognosis, but a statistically significant relationship could not be confirmed because of the small sample size. Both of the patients with CVI caused by trauma did not improve. All four patients with hydrocephalus improved. All three patients with infection improved. The patient with static encephalopathy improved. The patient with abnormal white matter did not improve. The patient with a hemorrhage at birth did not improve.
This is a retrospective study and contrast sensitivity results were not available for five of the 39 children. Therefore, the following contrast sensitivity data is based on a sample size of 34 patients. The mean initial Michelson contrast threshold for the whole group was 7.03%, which represents substantial visual impairment. The initial and final contrast thresholds for the group as a whole were analyzed using ANOVA and showed a significant effect (F = 3.99; df = 33; p = 0.011). The group was then split into those who improved and those who did not based on the criterion of a change of 0.2 log units or more. Sixteen of the 34 children (47%) showed improvement in contrast threshold, 13 children (38%) showed no change in contrast threshold, and five children (15%) showed a reduction in contrast threshold. Again, in those who improved, the average amount of improvement was substantial: 0.57 log unit (10 to 2.6% Michelson). The rate of change for those who improved was 0.09 log unit per year. The average change in contrast threshold for those who did not improve was −0.06 log unit (4 to 4% Michelson). Similar to acuity, the rate of change for those who did not improve was less than −0.01 log unit per year. Figures 2(a, b) graphically display the change in contrast threshold over time. Figure 2(a) shows the contrast threshold as a function of age in those who improved, and Figure 2(b) shows the contrast threshold as a function of age in those who did not improve. A higher point on the vertical axis indicates a better contrast threshold, and an upward trend (positive slope) indicates improvement in contrast threshold over time. The dark black line is the mean change for all the children represented in the graph.
Similar to visual acuity, the amount of improvement observed was associated with the initial contrast threshold (t = 3.6; df = 32; p = 0.001). The initial contrast threshold was significantly worse in those who improved compared with those who did not improve. The mean initial contrast threshold in the group who improved was 10% Michelson (95% CI: 7.25 to 28.21) and the mean initial contrast threshold in the group who did not improve was 4% Michelson (95% CI: 5.03 to 12.94).
The final contrast threshold measure was numerically better for those who improved (2.6% Michelson) compared with those who did not improve (4% Michelson). However, the difference was not statistically significant (t = −1.63; df = 32; p = 0.114). So even though the children who improved had a worse contrast threshold to start with, they improved to meet or barely surpass the children who did not improve.
Unlike visual acuity, the change in contrast threshold was significantly related to the age of the child at the first test (t = −2.26; df = 21; p = 0.035). Those whose contrast threshold improved were, on average, 2.28 years younger at the first test.
Similar to visual acuity, the duration between VEP measures was not significantly different between the two groups (t = 0.005; df = 32; p = 0.996). Children who improved had 6.89 years on average between the initial and final measures (SD: 4.05 years) and children who did not improve had 6.91 years on average between the two measures (SD: 3.74 years).
Figure 3 summarizes the change in visual acuity and the change in contrast threshold. Each data point represents one child. The data points within the box represent children who remained within a 0.20 log unit change and are therefore considered stable or within measurement error. Of the 16 who improved in contrast threshold, 14 also improved in visual acuity (two did not improve). Of the 18 who did not improve in contrast threshold, 13 also did not improve in visual acuity (five did improve). There is a significant correlation between acuity improvement and contrast improvement (r = 0.63; p < 0.001). However, there is large variation making it difficult to predict acuity improvement on the basis of contrast improvement and vice versa on an individual basis.
Longitudinal, quantitative assessment of visual acuity and contrast sensitivity in children with CVI is important to understand the natural history of vision function in CVI. About one-half of the children with CVI in this study showed significant improvement of visual acuity and contrast threshold over time and the improvements in vision function were considerable. However, visual acuity and contrast thresholds were remarkably constant in those who did not improve. As previously stated, the 0.2 log unit change cutoff was chosen somewhat arbitrarily and based on the average test-retest reliability of the sweep VEP reported in the literature. The 0.2 log unit cutoff conservatively requires a change of approximately 2 standard deviations of the VEP reliability to be considered a change in visual acuity. If improvement was defined as a 0.3 log unit change in log MAR for grating visual acuity, five fewer children would be classified as improved and the total percentage of children who improved would be 36%. If the cutoff for contrast threshold improvement was set at 0.3 log unit change, four fewer children would be classified as improved and the total percentage of children who improved would be 35%. However, regardless of the cutoff chosen, some children improved considerably over time and some remain strikingly stable. We evaluated differing factors between the two groups to better understand the prognosis for recovery of CVI.
The degree of initial visual impairment is associated with the amount of improvement observed in these patients both for visual acuity and contrast threshold. Statistical analysis on this relatively small group of patients did not support the involvement of etiology or the timing of the insult in the improvement of visual acuity. This is surprising because a different pattern of brain injury results from the various etiologies of CVI. The pattern of brain injury that results from a hypoxic insult is determined by the age of the child at the time of the hypoxic event. A full-term infant will show a different injury pattern than a premature infant after a similar hypoxic insult. Specifically, full-term infants will sustain injury primarily at the watershed zones in the cerebral cortex with minimal damage to the periventricular white matter, while premature infants who suffer from a comparable hypoxic event will show significant periventricular injury.7 The primary reason for this marked difference in injury pattern is attributed to an age-related change in the location of the intervascular boundary zones that occurs at approximately 34 weeks gestation.7,22 Because there are multiple etiologies of CVI, and because there are distinctively different areas of the brain damaged as a result of each of these etiologies, it would be reasonable to expect that a child's visual outcome would likely be dependent on the etiology of the CVI.
A few studies have discussed differences in visual outcomes in children with different patterns of brain injury.6,23 Cioni et al.6 evaluated patients with CVI with Teller acuity cards and magnetic resonance imaging (MRI) scans. They evaluated 42 infants with optic radiation damage and 19 infants with visual cortex damage. Despite finding significant variation within each study group, they concluded that damage to the optic radiations was predictive of poorer visual function than injury to the visual cortex. However, the patients were only evaluated once in early infancy and the question whether visual improvement occurred over time was not studied.
The expectation is that children with PVL (damage to the optic radiations) would have a poorer prognosis for recovery than children with only cortical damage. Lambert et al.3 reviewed the clinical courses and computed tomographic and MRI scans of 30 children with CVI caused by a hypoxic insult. The degree of injury to the optic radiations, striate cortex, and parastriate cortex were graded on a scale from 0 to 4 by a neuroradiologist. They found that the visual recovery was significantly poorer in the group with computed tomographic and MRI abnormalities in the area of radiations. Furthermore, visual recovery differed significantly with respect to the age at which the hypoxic insult occurred. A statistically significant correlation was established between an early age at the time of hypoxic insult and a poor visual outcome. This finding was initially surprising because most studies have suggested that infant brains have a greater regenerative potential than adult brains after injury.24 However, the reported inverse relationship between timing of hypoxic insult and the visual outcome is thought to be due to the lack of plasticity of the nerve fibers in the optic radiation as compared with the neurons in the visual cortex.3 Additionally, trophic factors are known to be vital at certain points in the development of the visual pathway. An insult in the critical development period of a premature infant may prevent crucial tropic factors from being released and may result in more extensive visual pathway damage.3
Hoyt25 reported a retrospective review of 7200 children with CVI seen between 1979 and 1994. They devised a functional evaluation of vision that included six levels of visual function. Level 1: light perception; level 2: occasional fixation on large objects, faces or movement; level 3: occasional fixation on small objects; level 4: reliable fixation on small objects; level 5: reliable visual acuity; and level 6: completely normal vision. They classified each child by level of visual function over time. This study agreed with the Lambert et al.3 study, showing poorer visual outcomes in children with PVL as compared with those with primary visual cortex damage. At the initial examination, the visual function of children with PVL was slightly worse than those with striate cortex damage. However, the two groups were significantly different when assessed over time. Children with striate cortex damage typically showed some improvement, and in some cases, the improvement was dramatic. However, the children studied with PVL were much less likely to improve and if they did improve, it was rarely more than one level on their functional scale. This study has the advantage of a large study population. However, a significant weakness is the lack of an objective, quantifiable measure of visual acuity at all levels. Levels 1 through 4 may actually be measuring the child's ability to discern brightness differences (i.e., contrast sensitivity) rather than measuring visual acuity.
Matsuba and Jan13 recently reported behavioral vision function longitudinally in a large sample of children with CVI. Vision function was classified into seven different levels and improvement was defined as one level change. Forty-six percent of the children improved one level or more over a period of several years. This result is similar to the current study. Children were more likely to improve if they had fewer comorbid conditions.
We expected that the results of the present study would agree with the Hoyt25 study, but would improve the validity of the conclusion with an objective, quantifiable measure of vision as the outcome measure. It is surprising that we found no association between the etiology and the timing of the insult on the improvement of visual acuity. One explanation is the relatively small number of subjects. Because there are so many different etiologies of CVI, 39 total subjects did not allow for enough subjects per subgroup based on etiology. Additionally, the present study is limited by being retrospective. The subject records included detailed neurological reports and conclusions regarding the etiology and timing of insult could be made. However, we did not have the original imaging scans to specifically evaluate and classify the neurological damage. We feel that such a neuroimaging evaluation by the same neuroradiologist is required to make a statement about the correlation of anatomy and prognosis. Furthermore, the severity and/or extent of brain damage were not evaluated and may be significantly different between the group of children whose visual acuity improved and the group whose visual acuity did not improve.
It is interesting that the age of the child is related to improvement in contrast threshold, but not to the improvement in visual acuity. Those who improved in contrast threshold were younger at the first test. However, there was no difference in the average age between the group who improved in visual acuity and the group who did not improve in visual acuity. In our experience, contrast thresholds appear to improve before changes in visual acuity in these children. This time course of recovery would be in agreement with the time course of normal development. Contrast sensitivity reaches adult levels sooner than visual acuity reaches adult levels.26 A prospective study with matched, controlled follow-up would be beneficial to better understand the time course of visual recovery in these infants and children with CVI.
For both visual acuity and contrast threshold, the mean initial performance was significantly worse in the group that improved as compared with the group which did not improve. One explanation is that the former group had “more room for improvement” which contributed to their greater change in vision function. However, given that their final vision function surpassed that of the no improvement group, we do not feel this is a major contributing factor. Furthermore, even though the initial visual performance for the group who did not change was better than the group who improved, it was still poor and allowed substantial room for improvement that did not occur.
The children in this study are likely to represent more severe cases of CVI. As previously stated, the mean initial visual acuity for the study population was 20/152. It is important to note that this is a VEP grating measurement and it is expected that a behavioral or optotype measure would be much worse.27,28 It is plausible that the prognosis and likelihood of visual recovery may be different for a group of children with less-severe brain damage and milder CVI.
CVI is now the leading cause of bilateral visual impairment in industrialized nations. Despite this, to date, there have been very few longitudinal evaluations of vision function in children with CVI. Therefore, research quantifying and evaluating the various types of vision loss in children with CVI is tremendously important. Indisputably, there is a great need for additional research concerning this common and complex disorder. Quantitative information about a child's condition will not only be useful clinically, but will also be enormously helpful to the child's educators and family, which will ultimately benefit the child.
This work was supported by NEI grant T32 EY07043.
School of Optometry
University of California
Berkeley, CA 94720-2020
1. Fulton AB, Hansen RM, Mayer L, Afshari NA, Afshari MA. The visually inattentive infant. In: Albert DM, Jakobiec FA, eds. Principles and Practice of Ophthalmology. The Harvard System, 2nd ed. Philadelphia: WB Saunders; 2000:4483–90.
2. Good WV, Jan JE, DeSa L, Barkovich AJ, Groenveld M, Hoyt CS. Cortical visual impairment in children. Surv Ophthalmol 1994;38:351–64.
3. Lambert SR, Hoyt CS, Jan JE, Barkovich J, Flodmark O. Visual recovery from hypoxic cortical blindness during childhood. Computed tomographic and magnetic resonance imaging predictors. Arch Ophthalmol 1987;105:1371–7.
4. Chen TC, Weinberg MH, Catalano RA, Simon JW, Wagle WA. Development of object vision in infants with permanent cortical visual impairment. Am J Ophthalmol 1992;114:575–8.
5. Afshari MA, Afshari NA, Fulton AB. Cortical visual impairment in infants and children. Int Ophthalmol Clin 2001;41:159–69.
6. Cioni G, Fazzi B, Ipata AE, Canapicchi R, van Hof-van Duin J. Correlation between cerebral visual impairment and magnetic resonance imaging in children with neonatal encephalopathy. Dev Med Child Neurol 1996;38:120–32.
7. Jacobson LK, Dutton GN. Periventricular leukomalacia: an important cause of visual and ocular motility dysfunction in children. Surv Ophthalmol 2000;45:1–13.
8. Good WV, Jan JE, Burden SK, Skoczenski A, Candy R. Recent advances in cortical visual impairment. Dev Med Child Neurol 2001;43:56–60.
9. Dutton GN, Jacobson LK. Cerebral visual impairment in children. Semin Neonatol 2001;6:477–85.
10. Martin JA, Hamilton BE, Sutton PD, Ventura SJ, Menacker F, Munson ML. Births: final data for 2002. Natl Vital Stat Rep 2003;52:1–113.
11. Good WV. Development of a quantitative method to measure vision in children with chronic cortical visual impairment. Trans Am Ophthalmol Soc 2001;99:253–69.
12. Cannao M, Morretti G, Ricci A. Complex visual problems in childhood. Neuro-ophthalmology 1986;6:321–7.
13. Matsuba CA, Jan JE. Long-term outcome of children with cortical visual impairment. Dev Med Child Neurol 2006;48:508–12.
14. Watson TA, Orel-Bixler D, Haegerstrom-Portnoy G. VEP acuity predicts future behavioral acuity in children with cortical visual impairment. Invest Ophthalmol Vis Sci 2006;47:E-abstract 3477.
15. Miller NR, Newman NJ. Walsh & Hoyt's Clinical Neuro-Ophthalmology, 5th ed. Baltimore: Williams & Wilkins; 1998.
16. Groenendall F, van Hof-van J. Visual deficits and improvements in children after perinatal hypoxia. J Visual Impair Blin 1992;86:215–8.
17. Lim M, Soul JS, Hansen RM, Mayer DL, Moskowitz A, Fulton AB. Development of visual acuity in children with cerebral visual impairment. Arch Ophthalmol 2005;123:1215–20.
18. Norcia AM, Tyler CW. Infant VEP acuity measurements: analysis of individual differences and measurement error. Electroencephalogr Clin Neurophysiol 1985;61:359–69.
19. Prager TC, Zou YL, Jensen CL, Fraley JK, Anderson RE, Heird WC. Evaluation of methods for assessing visual function of infants. J AAPOS 1999;3:275–82.
20. Lauritzen L, Jorgensen MH, Michaelsen KF. Test-retest reliability of swept visual evoked potential measurements of infant visual acuity and contrast sensitivity. Pediatr Res 2004;55:701–8.
21. Dobson V, Quinn GE, Siatkowski RM, Baker JD, Hardy RJ, Reynolds JD, Trese MT, Tung B. Cryotherapy for Retinopathy of Prematurity Cooperative Group. Agreement between grating acuity at age 1 year and Snellen acuity at age 5.5 years in the preterm child. Invest Ophthalmol Vis Sci 1999;40:496–503.
22. Takashima S, Tanaka K. Development of cerebrovascular architecture and its relationship to periventricular leukomalacia. Arch Neurol 1978;35:11–6.
23. Roland EH, Jan JE, Hill A, Wong PK. Cortical visual impairment following birth asphyxia. Pediatr Neurol 1986;2:133–7.
24. Stewart DL, Riesen AH. Adult versus infant brain damage: behavioral and electrophysiological effects of striatectomy in adult and neonatal rabbits. Adv Psychobiol 1972;1:171–211.
25. Hoyt CS. Visual function in the brain-damaged child. Eye 2003;17:369–84.
26. Norcia AM, Tyler CW, Hamer RD. Development of contrast sensitivity in the human infant. Vision Res 1990;30:1475–86.
27. Dobson V, Teller DY. Visual acuity in human infants: a review and comparison of behavioral and electrophysiological studies. Vision Res 1978;18:1469–83.
28. Orel-Bixler D, Haegerstrom-Portnoy G, Hall A. Visual assessment of the multiply handicapped patient. Optom Vis Sci 1989;66:530–6.