Doc Ophthalmol DOI 10.1007/s10633-013-9423-9
ORIGINAL RESEARCH ARTICLE
Contributing factors to VEP grating acuity deficit and inter-ocular acuity difference in children with cerebral visual impairment Nı´vea Nunes Cavascan • Solange Rios Saloma˜o • Paula Yuri Sacai • Josenilson Martins Pereira • Daniel Martins Rocha • Adriana Berezovsky
Received: 15 July 2013 / Accepted: 9 December 2013 Ó Springer-Verlag Berlin Heidelberg 2013
Abstract Purpose To investigate contributing factors to visual evoked potential (VEP) grating acuity deficit (GAD) and inter-ocular acuity difference (IAD) measured by sweep-VEPs in children with cerebral visual impairment (CVI). Methods VEP GAD was calculated for the better acuity eye by subtracting acuity thresholds from mean normal VEP grating acuity according to norms from our own laboratory. Deficits were categorized as mild (0.17 B deficit \ 0.40 log units), moderate (0.40 B deficit \ 0.70 log units) or severe (deficit C0.70 log units). Maximum acceptable IAD was 0.10 log units.
N. N. Cavascan (&) S. R. Saloma˜o P. Y. Sacai J. M. Pereira D. M. Rocha A. Berezovsky Departamento de Oftalmologia, Escola Paulista de Medicina, Universidade Federal de Sa˜o Paulo, Rua Botucatu 822, Sa˜o Paulo, SP 04023-062, Brazil e-mail: [email protected] S. R. Saloma˜o e-mail: [email protected]
Results A group of 115 children (66 males—57 %) with ages ranging from 1.2 to 166.5 months (median = 17.7) was examined. VEP GAD ranged from 0.17 to 1.28 log units (mean = 0.68 ± 0.27; median = 0.71), and it was mild in 23 (20 %) children, moderate in 32 (28 %) and severe in 60 (52 %). Severe deficit was significantly associated with older age and anti-seizure drug therapy. IAD ranged from 0 to 0.49 log units (mean = 0.06 ± 0.08; median = 0.04) and was acceptable in 96 (83 %) children. Children with strabismus and nystagmus had IAD significantly larger compared to children with orthoposition. Conclusion In a large cohort of children with CVI, variable severity of VEP GAD was found, with more than half of the children with severe deficits. Older children and those under anti-seizure therapy were at higher risk for larger deficits. Strabismus and nystagmus provided larger IADs. These results should be taken into account on the clinical management of children with this leading cause of bilateral visual impairment. Keywords Cerebral visual impairment Grating acuity Sweep-VEP Child
P. Y. Sacai e-mail: [email protected] J. M. Pereira e-mail: [email protected] D. M. Rocha e-mail: [email protected] A. Berezovsky e-mail: [email protected]
Introduction Cerebral visual impairment (CVI) is a pediatric neurological disorder caused by post-chiasmatic brain lesion that results in bilateral visual loss [1, 2]. Clinically, it
123
Doc Ophthalmol
manifests as a deficit in visual acuity with normal ocular structures in both eyes [1, 2]. In general, contrast sensitivity is reduced and visual field is constricted, hemianopic or ‘‘Swiss cheese’’ [3–6]. Neurobehavioral signs include preference for brightly colored and moving objects instead of stationary or sharp black and white contrasts [1, 3, 7]. Visual response to complex visual scene used to be reduced, delayed or suppressed, especially in the presence of competing auditory or tactile stimuli [7]. Visual performance apparently is better in familiar settings, and it varies according to the moment, changes in the environment, illumination conditions and visual attention [1, 8]. Ocular movement abnormalities such as strabismus and nystagmus are frequent in patients CVI, and they can find it difficult to make the appropriate eye, head and neck movement to look at an object of visual interest, contributing to the poor visual behavior [2–4]. CVI is the leading cause of childhood blindness in developed countries, and its prevalence has been increasing in middle-income countries due to improvement in medical care and diagnostic equipment. Technological advances of neonatal units allow many premature infants to survive, who are an at-risk group for the occurrence of diseases related to prematurity and its complications [2, 3, 9]. The most common cause of CVI is the hypoxic/ ischemic insult whose damage location depends on the vascular anatomy and the brain maturity [3, 10]. Around 60 % of children with hypoxic–ischemic encephalopathy have CVI [3, 10]. Other etiologies include head injury, hydrocephalus, ventriculo-peritoneal shunt failure, epilepsy, infections, intrauterine exposure to drugs, neurodegenerative or metabolic diseases, tumor and malformation of brain structures [1–4, 8, 10, 11]. Cerebral damages represent a widespread impact, affecting not only the visual system, but also several brain areas, leading to other neurological problems, such as delay in neuropsychomotor development, cerebral palsy, seizures, hearing loss, microcephaly, meningomyelocele and progressive degeneration of central nervous system [1, 8, 9, 11]. Diagnosis of CVI is based on the triple assessment of neurological condition, ocular structure and visual function. Quantitative information about visual acuity deficit supports ophthalmological intervention, establishment of educational strategies and habilitation programs, and moreover, it facilitates social inclusion [5, 12].
123
Sweep visual evoked potential (sweep-VEP) is an electrophysiological technique that provides feasible and reliable grating acuity measurement as showed previously [13, 14]. This objective method does not require verbal, motor or behavioral response from the patient [2]. Then, it is a useful tool for vision evaluation in patients with CVI, which often have limited cognitive and verbal abilities [6, 15]. Reports of children with CVI often present binocularly measured acuities, but seldom show acuity deficit from the normal population and inter-ocular acuity differences (IADs). The purpose of this study was to determine grating acuity deficit (GAD) magnitude measured by sweep-VEP and IAD in children with CVI.
Materials and methods Participants Participants were 115 children (66 males—57 %) with CVI referred for VEP grating acuity measurement to the Clinical Electrophysiology of Vision Laboratory of the Department of Ophthalmology of Federal University of Sa˜o Paulo (UNIFESP), Brazil, in the last 10 years. This retrospective cross-sectional observational study was approved by UNIFESP Committee of Ethics in Research and followed the tenets of the Declaration of Helsinki. Children with CVI were diagnosed clinically on the basis of bilaterally reduced visual acuity, with evidence of post-chiasmatic brain lesion, poor visual behavior and normal ocular structure. Participants with no ophthalmic examination and uncorrected refractive errors were excluded. A total of 1,634 charts were revised, and 245 potential CVI patients were selected. Out of them, 130 were excluded because they presented only binocular VEP grating acuity measurement and/or optic disk pallor. Sweep-VEP measurements were taken from children with CVI aging from 1.2 to 166.5 months (median = 17.7). Thirty-four children were tested in the first year of life (1.2–11.8), 39 in second (12.6–23.3) and 42 from the second year of life on (24.1–166.5). Main etiology was prematurity (n = 28; 24 %), followed by seizures (n = 18; 16 %), perinatal hypoxia (n = 9; 8 %), brain malformations (n = 8; 7 %), acquired hypoxia (n = 7; 6 %) and hydrocephalus (n = 5; 4 %). In 22 (19 %) children, CVI was
Doc Ophthalmol
caused by a combination of two or more etiologies. Other causes (n = 18; 16 %) included encephalitis, meningitis, traumatic brain injury and genetic, infectious or metabolic disorders. Prematurity was considered for children with gestational age of 37 weeks or less according to the American Academy of Pediatrics [16], the Royal College of Ophthalmologists and the British Association of Perinatal Medicine [17]. Regarding ocular motility most patients were exotropic (n = 58; 66 %), typically with variable angle. Out of 88 cases of strabismus, 69 (78 %) were manifest and 19 (22 %) intermittent, as well as 46 (52 %) were unilateral and 42 (48 %) alternating. Thirty-two (28 %) children presented nystagmus that was manifest in 19 (59 %), latent in 7 (22 %) and occasional in 6 of them (19 %). Horizontal nystagmus was found in vast majority of cases (n = 27; 84 %) and vertical nystagmus in 4 (13 %) and only 1 (3 %) child had torsional nystagmus. Eighty-one (70 %) patients were in use of oral antiseizure drugs, and 40 (49 %) children received exclusive monotherapy (17 phenobarbital, 10 valproic acid, 5 benzodiazepine, 4 carbamazepine, 2 oxcarbamazepine, 2 topiramate). Sodium valproate was the main anti-seizure drug prescribed (n = 33; 41 %), mostly in association with one or more of the following substances: benzodiazepine, nitrazepam, phenobarbital and vigabatrin. Four parents did not know the medication in usage. All children presented neuropsychomotor development delay, and 75 % of them (n = 86) were included in habilitation program and/or visual stimulation. Out of 115 children, 35 (30 %) used glasses. Spherical equivalent in right eye ranged from -7.50 to ?6.00 diopters (mean = ?0.26 ± 1.42; median = 0.00) and in left eye from -7.50 to ?5.50 diopters (mean =?0.23 ± 1.37; median = 0.00). Sweep-VEP procedure Monocular grating acuity testing was performed using the PowerDiva (digital infant vision assessment) sweep-VEP system (the Smith-Kettlewell Eye Research Institute, San Francisco, USA) developed by Norcia [18]. This system utilizes two interfacing Macintosh G3 computers: the ‘‘host’’ computer selects the stimulus, establishes stimulus and trial parameters and analyzes the evoked response; the ‘‘video’’ computer controls the display monitor and shows the
Table 1 Testing distances and spatial frequency sweep Testing distance (cm)
Spatial frequency at sweep start (cycles/ degree)
30
0.10
50
0.50
9.46
80
1.00
15.14
100
1.00
18.92
150
2.00
28.38
Spatial frequency at sweep end (cycles/ degree) 3.08
stimuli to the subject. The stimuli were phase-reversal sine-wave gratings presented on a 17.5 in. (29 9 38 cm) high-resolution monochromatic video monitor (M20DCD4RE-Richardson ElectronicsÒ). Stimuli orientation was set vertical for participants without nystagmus and for those with either vertical or torsional nystagmus. Horizontal orientation was used for children with horizontal nystagmus [19]. Mean luminance (142.35 cd/m2) as well as contrast (80 %) was kept constant throughout the session. Patterns were temporally alternated in counterphase with a temporal modulation of 6.6 Hz. During the test, children were positioned seated in their parent’s lap or seated in their wheelchair in front of a display monitor. Their attention was attracted with small toys dangling over the center of the display. Sweep-VEP was recorded only when the subject was alert and fixating the stimuli. Testing distance was determined according to the patient’s age and visual behavior. The ranges of spatial frequencies and testing distances are shown in Table 1. Test stimulation field was 52° 9 65° (for 30 cm distance) to 11° 9 14° (for 150 cm distance), calculated for both vertical and horizontal monitor lengths. Ten linearly spaced spatial frequencies were presented, 1/s, starting at a low spatial frequency. The procedure was performed in a dark room, and the eye of better visual behavior was tested first. Sweep-VEP test, including setup and rest breaks, typically lasted 30–60 min, depending on the subject’s age and cooperation. Responses were obtained from electroencephalogram (EEG) electrodes (Grass Gold Disc Electrodes— E6GH) attached to the scalp with electrode cream and cotton pads. A headband (3M Coban Self-adherent Wrap 1581) was used to keep the electrodes in place. According to the normative protocol of our own laboratory [20], EEG was recorded from two bipolar placements (O1 and O2), 2–3 cm to the left and right
123
Doc Ophthalmol
Fig. 1 Sweep-VEP response; record from right eye of a participant; sweep sets from 1.00 to 18.92 cycles/degree (linear scale); two thresholds were obtained, channel one (O1–Cz) and channel two (O2–Cz); Dots represent noise registered for each one of 10 linearly spaced spatial frequencies presented; grating acuity was estimated by linear fit and extrapolation to zero amplitude, i.e., the value at which the regression line touches the axis of spatial frequencies; the final acuity score was calculated
in logMAR using the results of the channel two with better threshold (8.69 cycles/degree) with the highest signal-to-noise ratio (19.17). Ch channel; 2F1s harmonic of the stimulus frequency; lV microvolts; Spat Freq spatial frequency; Thrsh grating acuity threshold (cycles/degree); SNR signal-to-noise ratio; Sc SNR maximum SNR within the cursors (dotted lines) that define the data used to estimate threshold; Pk SNR maximum SNR at peak mean amplitude in the record
of a ground electrode placed 1 cm above the inion on the midline (Oz), following the international 10–10 electrode placement system [21]. A reference electrode was placed at the vertex (Cz). Visual evoked responses were isolated in real time (sampling rate = 397 Hz) from the EEG by adaptive filter (bandpass). The potential differences were amplified (Neurodata Acquisition System P15, Grass Instrument Co; gain = 10,000; -3 db cutoff at 1 and 100 Hz). Three to 15 repetitions of the sweep were obtained and vector averaged. Amplitude and phase of the first (6 Hz) and second harmonics (12 Hz) of the stimulus frequency were calculated for each channel by discrete Fourier transform. Grating acuity was estimated with an automated algorithm which performed a linear fit and extrapolation to zero amplitude for the final descending limb of the function relating VEP second harmonic amplitude to linear spatial frequency. A signal-to-noise ratio (SNR), calculated as the ratio of power at stimulus frequency to mean power at frequencies ±2 Hz, at peak mean amplitude of 3:1, was required [22]. A SNR of 3:1 corresponds to a false alarm rate of 0.4 %, and it provides an adequate protection level when combined with the phase consistency criteria. In all cases, two thresholds (one for each channel, O1 and O2) were obtained (Fig. 1). The final acuity score was calculated in logMAR (logarithm of the minimum angle of resolution) using the results in cycles per degree of visual angle of the
better threshold channel with the highest SNR, according to the formula:
123
log MAR acuity score ¼ logð30= acuity threshold in cycles per degreeÞ VEP grating acuity threshold and IAD were compared . normative data from our own laboratory [20]. Maxito mum acceptable IAD was 0.10 log units. Since the normal limits of visual acuity vary with age in the age range targeted in this study, all measurements were converted to visual acuity deficits (i.e., the number of log units relative to published age-corrected mean normal values). Then, VEP GAD was calculated for the better acuity eye (log units) by subtracting acuity thresholds from mean normal visual acuity according to age norms [20]. Deficits were categorized as mild (0.17 B GAD \ 0.40 log units), moderate (0.40 B GAD \ 0.70 log units) or severe (GAD C 0.70 log units).
Statistical analysis Statistical analyses were carried out using Stata Statistical Software: Release 12.0. p values equal to or less than 0.05 were considered statistically significant with two-tailed rejection region. On the first-level analysis, GAD and IAD were analyzed by gender, age group (\12, 12–24 and C24 months), main etiology, ocular motility
Doc Ophthalmol Fig. 2 VEP grating acuity threshold; Circles indicate individual values of VEP grating acuity from better vision eye of 115 children with cortical visual impairment; Dashed and dotted lines represent normal mean and normal lower limit, respectively, from our own laboratory (Saloma˜o et al. [20])
condition, use of anti-seizure medication and inclusion in habilitation and/or early visual stimulation program. Statistical models included Student’s t test and oneway analysis of variance (ANOVA), followed by multiple comparison procedure (Tukey’s test), when applicable. The skewness/kurtosis test validated the parametric data analysis methods. If normality test failed, non-parametric tests were used: Mann–Whitney rank sum test and Kruskal–Wallis ANOVA on ranks, followed by multiple comparison procedure (Dunnet’s test), when applicable. On the second-level analysis, multiple logistic regression model was used to investigate the association between severe deficit and age group (\12 and C12 months), gender, presence of prematurity, seizures and ocular motility disorders, use of anti-seizure medication and inclusion in habilitation and/or visual
stimulation program. Odds ratios (OR) with 95 % confidence intervals (CI) were estimated.
Results Inter-ocular acuity difference VEP grating acuity thresholds ranged from 0.23 to 1.41 logMAR (mean = 0.93 ± 0.26; median = 0.95) for the better vision eye (Fig. 2) and 0.26–1.70 logMAR (mean = 1.00 ± 0.27; median = 1.02) for the worse vision eye. IAD ranged from 0 to 0.49 log units (mean = 0.06 ± 0.08; median = 0.04). Normal IAD (B 0.10 log units) was found in 96 (83 %) children. Children with strabismus and nystagmus had IAD significantly larger when compared to children
Table 2 Mean inter-ocular acuity difference, their respective standard deviations and median of 115 participants distributed by ocular motility condition n
%
IAD (log units) Mean
SD
Strabismus
65
56
0.06
0.05
0.04
Strabismus ? nystagmus
23
20
0.13*
0.13
0.08
9
8
0.03
0.03
0.03
18
16
0.03
0.03
0.02
Orthoposition
Statistical model
0.0157*
Kruskal–Wallis
Median
Ocular motility
Nystagmus
p value
IAD inter-ocular acuity difference * p B 0.05
123
Doc Ophthalmol Table 3 Mean VEP grating acuity deficit, their respective standard deviations and median of 115 participants distributed by age group, gender, main etiology, ocular motility condition, anti-seizure medication usage and inclusion in habilitation and/or visual stimulation program n
%
GAD (log units) Mean
SD
Median
p value
Statistical model
0.0002*
ANOVA
Age group (months) \12
34
30
0.53*
0.21
0.57
12–24
39
34
0.77
0.22
0.76
C24
42
36
0.73
0.30
0.76
66 49
57 43
0.71 0.66
0.25 0.28
0.74 0.69
0.3384
t test
Prematurity
28
24
0.62
0.25
0.65
0.4608
ANOVA
Seizures
18
16
0.69
0.28
0.70
0.1309
ANOVA
0.0008*
t test
0.1988
Mann–Whitney
Gender Male Female Main etiology
Perinatal hypoxia
9
8
0.73
0.31
0.73
Brain malformations
8
7
0.61
0.18
0.63
Acquired hypoxia
7
6
0.69
0.28
0.75 0.46
Hydrocephalus
5
4
0.54
0.32
Others
18
16
0.75
0.31
0.83
Miscellaneous
22
19
0.76
0.24
0.75
Ocular motility Strabismus
65
56
0.64
0.27
0.63
Strabismus ? nystagmus
23
20
0.80
0.23
0.78
Nystagmus Orthoposition Anti-seizure medication No
9
8
0.70
0.23
0.69
18
16
0.68
0.28
0.74
34
30
0.56
0.24
0.55
81
70
0.74*
0.26
0.76
No
29
25
0.63
0.25
0.68
Yes
86
75
0.70
0.27
0.74
Yes Habilitation program
GAD VEP grating acuity deficit; ANOVA one-way analysis of variance * p B 0.05
with orthoposition (Kruskal–Wallis; H = 10.397; p = 0.0157) as shown in Table 2. Additionally, IAD was significantly larger (mean = 0.09 ± 0.10; median = 0.07) in patients with deviation with preference for fixation by the fellow eye when (z = 1.934; p = 0.0531) compared to those with alternating deviation (mean = 0.06 ± 0.06; median = 0.04). IAD was comparable either for manifest (mean = 0.08 ± 0.09; median = 0.04) or for intermittent (mean = 0.07 ± 0.06; median = 0.05) deviation. Out of 19 children with IAD above the maximum acceptable ([0.10 log units), all but 1 had
123
strabismus. Twelve presented fixation preference by the fellow eye, and 6 children had alternating strabismus. Exotropia (n = 13) was found in the majority of patients. IAD was comparable for age group, gender, main etiology, using anti-seizure medication and inclusion in habilitation program. VEP grating acuity deficit GAD ranged from 0.17 to 1.28 log units (mean = 0.68 ± 0.27; median = 0.71). Mild GAD was found in 23 (20 %) children, moderate in 32
Doc Ophthalmol Table 4 Odds ratios and 95 % confidence intervals in multivariate analysis of factors related to severe VEP grating acuity deficit Variable
OR (95 % CI)
p value
Age (months) \12
Reference
C12
2.56 (1.04–6.34)
0.041*
Gender Male
Reference
Female
1.20 (0.52–2.75)
0.667
Prematurity No
Reference
Yes
0.76 (0.31–1.83)
0.537
Reference 0.65 (0.20–2.06)
0.463
Seizures No Yes
Ocular motility disorder No
Reference
Yes
0.86 (0.27–2.73)
0.805
Anti-seizure medication No
Reference
Yes
5.30 (2.09–13.42)
\0.001*
Habilitation program No
Reference
Yes
1.18 (0.44–3.14)
0.739
OR odds ratio; CI confidence interval * p B 0.05
(28 %) and severe in 60 (52 %). GAD was significantly smaller for children tested in the first year of life (ANOVA, Tukey’s test; F = 9.45; p = 0.0002), and it was significantly larger in anti-seizure medication users (t = -3.4387; p = 0.0008). GAD was comparable for gender, main etiology, ocular motility condition and inclusion in habilitation programs as shown in Table 3. In multiple logistic regression modeling (Table 4), severe GAD was significantly associated with older age [OR 2.58 (95 % CI 1.04–6.38)] and anti-seizure drug therapy [OR 5.58 (95 % CI 2.17–14.37)]. No interactions were observed with gender, presence of prematurity, seizures and ocular motility disorders, as well as inclusion in therapeutic program.
Discussion Visual acuity evaluation can be challenging in children with CVI. Sweep-VEP is a useful method to
measure grating acuity in this population for many reasons [2]. First, the luminescent stimulus source of the monitor tends to be attractive, since patients with CVI usually show a predilection for bright stimuli. Besides, the child is positioned directly in front of the monitor, so that the VEP stimulus subtends a wide angle of the subject’s visual field, facilitating the recording in cases of motor deficits and visual field defects. In addition, the equipment allows the calculation of the average responses recorded and choice for individual responses, and the examiner can control (pause and restore) the stimulus presentation. Most previous investigations demonstrate reduced VEP grating acuity and slower visual development measured by sweep-VEP in children with CVI [5–7, 12, 15]. Our study included deficit and IAD analysis in this group. We recorded a median VEP grating acuity of 0.93 logMAR (in the better vision eye), which was worse than values from previous studies (0.37–0.88 logMAR) [5, 6, 12, 15]. In these studies, acuity data were obtained under binocular viewing, condition that certainly gets more overall cooperation and attention from the children and can reduce noise of the evoked response [20, 23, 24]. But data of monocular measure as well as IADs contribute to precise and reliable clinical diagnosis of CVI, implying bilateral visual loss. Our results showed acceptable IAD (B0.10 log units) in the vast majority of cases, speaking on behalf of the absence of ocular abnormalities and evidence of post-chiasmatic brain lesion. Additionally, IAD was larger in children with strabismus and nystagmus as expected, whereas these conditions are recognizably predisposing factors to disruption of binocular interaction and simultaneous visual maturation of both eyes [25, 26]. In our group, mean GAD was 0.68 log units, which is more severe when compared to 0.40 log units reported by Watson et al. [15]. Considering that our sample was younger than this previous study (29.97 months vs 12.10 years), roughly it could corroborate with previous findings that early neurological events are associated with more severe visual condition [10], but we cannot affirm the timing of neurological event from Watson’s study. According to magnitude, we found moderate or severe GAD in the vast majority of the cases (80 %, n = 92), finding in agreement with Birch and Bane [27], although obtained by different methods of measurement (electrophysiological vs psychophysical). Analysis by age group showed larger GAD from the second year of life on, suggesting small
123
Doc Ophthalmol
improvement in VEP grating acuity with increasing age and consequently slower visual maturation than normal, as reported previously [5, 7]. Related to the use of anti-seizure medication, GAD was significantly larger in users when compared with non-users. Nowadays, it is known that the cognitive evolution of epileptic patients is significantly linked to the seizure control, which provides a better maintenance of the bioelectrical activities of both cortical and subcortical structures. However, some of the antiseizure drugs widely used such as vigabatrin [28, 29], lamotrigin and topiramate [30, 31] present both neurological and retinal toxicities as side effects that limit their prolonged use. In our study, 21 % (n = 17) of our children received these types of drug, but none of the patients of our sample has been submitted to electroretinogram evaluation. We also did not correlate GAD with the time of drug treatment. Therefore, we cannot rule out its possible action on the retina and its effect on brain electrical conduction. Although antiseizure drugs may lead to attention deficit and irregular electrical activity, sweep-VEP reliability is ensured by speed of execution, discrete Fourier analysis that improves signal-to-noise ratio and extrapolation method that is amplitude-insensitive and does not depend critically on factors such as thickness of skull and small variations in electrode localization [13]. As expected, there was no difference in GAD analyzed by gender, since the visual development is comparable in boys and girls [32, 33]. In our study, hypoxic–ischemic etiologies (prematurity, perinatal hypoxia and acquired hypoxia) put together were main causes of CVI, according to previous reports [1, 6, 8, 10, 11, 27]. GAD also was comparable for etiologic factors, in agreement with previous reports (2007, 2010) [5, 12]. Ocular motility disorders are usually associated with CVI. Out of all our children, only 16 % (n = 18) had orthoposition of the visual axis. Previous investigations showed frequency of strabismus and nystagmus, respectively, of 29–90 and 10–56 % in child population with CVI [7, 10, 11, 27]. Nevertheless, in the current study, there was no difference in GAD when analyzed by ocular motility conditions. Exotropia was found in 66 % of cases, in agreement with literature that reports constant exotropia in childhood is often associated with neurological dysfunction [34]. Surprisingly, children included in habilitation program and/or early visual stimulation (75 %, n = 86)
123
showed comparable GAD to children who did not receive in accompanying therapeutic. However, sweep-VEP depends on the integrity of the maculooccipital pathway (Brodmann area 17), and it reflects central visual function. But patient’s visual behavior depends on this pathway and further visual association cortex (Brodmann areas 18–21), mechanisms of attention and motor responses [35]. Habilitation program and/or early visual stimulation provides to the brain the opportunity to form appropriate connections in order to compensate delayed neuropsychomotor development by the use of residual vision, but not necessarily visual acuity recovery [9, 36]. In multivariable analysis, age and use of antiseizure medication remained significantly associated with severe deficit. Children tested from the second year of life on were 2.6 times as likely to present severe deficit compared with children tested in the first year after adjusting for remaining variables. As well as drug users were 5.3 times as likely to present severe deficit compared with non-users. We conclude that users of anti-seizure medication and children tested from the second year of life on presented significantly higher odds for severe deficit. Further longitudinal investigation is helpful to confirm risk factors associated with visual deficit severity. Besides, to exclude toxic effects of the anti-seizure medication on the retina would be necessary retinal function evaluation by full-field electroretinography, which is impracticable in vast majority of children with CVI without sedation. As a limitation of this study, our participants did not undergo electroretinography procedure. We also call attention that VEP grating acuity measurement in CVI is not only useful clinically, but also extremely valuable in enabling improvement in visual behavior, motor skills and social interaction through habilitation programs. Acknowledgments This study was supported by the Conselho Nacional de Desenvolvimento Cientı´fico e Tecnolo´gico-CNPq (Brası´lia, DF, Brazil) research fellowship (A. Berezovsy and S. R. Saloma˜o). We thank Dr. Flavio E. Hirai for statistical assistance. Conflict of interests
None.
References 1. Jan JE, Groenveld M, Sykanda AM, Hoyt CS (1987) Behavioral characteristics of children with permanent cortical visual impairment. Dev Med Child Neurol 29:571–576
Doc Ophthalmol 2. Good WV, Jan JE, Burden SK, Skoczenski A, Candy R (2001) Recent advances in cortical visual impairment. Dev Med Child Neurol 43:56–60 3. Good WV, Jan JE, deSa L, Barkovich AJ, Groenveld M, Hoyt CS (1994) Cortical visual impairment in children: a major review. Surv Ophthalmol 38:351–364 4. Dutton GN, Jacobson LK (2001) Cerebral visual impairment in children. Semin Neonatol 6:477–485 5. Watson T, Orel-Bixler D, Haegerstrom-Portnoy G (2007) Longitudinal quantitative assessment of vision function in children with cortical visual impairment. Optom Vis Sci 87:80–86 6. Good WV, Huo C, Norcia AM (2012) Spatial contrast sensitivity vision loss in children with cortical visual impairment. Invest Ophthalmol Vis Sci 53:7730–7734 7. Lim M, Soul JS, Hansen RM, Mayer DL, Moskowitz A, Fulton AB (2005) Development of visual acuity in children with cerebral visual impairment. Arch Ophthalmol 123:1215–1220 8. Whiting S, Jan JE, Wong PKH, Flodmark O, Farrell K, McCormick AQ (1985) Permanent cortical visual impairment in children. Dev Med Child Neurol 27:730–739 9. Groenveld M, Jan JE, Leader P (1990) Observations on the habilitation of children with cortical visual impairment. J Vis Impair Blind 84:11–15 10. Hoyt CS (2003) Visual function in the brain-damaged child. Eye 17:369–384 11. Huo R, Burden SK, Hoyt CS, Good WV (1999) Chronic cortical visual impairment in children: aetiology, prognosis, and associated neurological deficits. Br J Ophthalmol 83:670–675 12. Watson T, Orel-Bixler D, Haegerstrom-Portnoy G (2010) Early visual-evoked potential acuity and future behavioral acuity in cortical visual impairment. Optom Vis Sci 84:471–480 13. Tyler CW, Apkarian PA, Levi DM, Nakayama K (1979) Rapid assessment of visual function: an electronic sweep technique for the pattern visual evoked potential. Invest Ophthalmol Vis Sci 18:703–712 14. Good WV (2001) Development of a quantitative method to measure vision in children with cortical visual impairment. Tr Am Ophth Soc 99:253–269 15. Watson T, Orel-Bixler D, Haegerstrom-Portnoy G (2009) VEP vernier, VEP grating, and behavioral grating acuity in patients with cortical visual impairment. Optom Vis Sci 86:774–780 16. American Academy of Pediatrics, Section on Ophthalmology (2001) Screening examination of premature infants for retinopathy of prematurity. Pediatrics 108:809–811 17. The report of a Joint Working Party of The Royal College of Ophthalmologists and the British Association of Perinatal Medicine (1996) Retinopathy of prematurity: guidelines for screening and treatment. Early Hum Dev 46:239–258 18. Norcia AM (1999) PowerDiva manual. Version 1.1.0. The Smith-Kettlewell Eye Research Institute, San Francisco 19. Meiusi RS, Lavoie JD, Summers CG (1993) The effect of grating orientation on resolution acuity in patients with nystagmus. J Pediatr Ophthalmol Strabismus 30:259–261
20. Saloma˜o SR, Ejzenbaum F, Berezovsky A, Sacai PY, Pereira JM (2008) Age norms for monocular grating acuity measured by sweep-VEP in the first three years of age. Arq Bras Oftalmol 71:475–479 21. Nuwer MR, Comi G, Emerson R, Fuglsang-Frederiksend A, Gue0 rite J, Hinrichsf H, Ikedag A, Luccash FJC, Rappelsburger P (1998) IFCN standards for digital recording of clinical EEG. Int Fed Clin Neurophysiol Electroencephalogr Clin Neurophysiol 106:259–261 22. Norcia AM, Tyler CW (1985) Spatial frequency sweepVEP: visual acuity during the first year of life. Vis Res 25:1399–1408 23. Hamer RD, Norcia AM, Tyler CW (1989) The development of monocular and binocular VEP acuity. Vis Res 29:397–408 24. Tyler CW, Apkarian PA (1985) Effects of contrast, orientation and binocularity in the pattern evoked potential. Vis Res 25:755–766 25. Agrawal R, Conner IP, Odom JV, Schwartz TL, Mendola JD (2006) Relating binocular and monocular vision in strabismic and anisometropic amblyopia. Arch Ophthalmol 124:844–850 26. Fu VLN, Bilonik RA, Felius J, Hertle RW, Birch EE (2011) Visual acuity development of children with infantile nystagmus syndrome. Invest Ophthalmol Vis Sci 52:1404–1411 27. Birch EE, Bane MC (1991) Forced-choice preferential looking acuity of children with cortical visual impairment. Dev Med Child Neurol 33:722–729 28. Durbin S, Mirabella G, Buncic JR, Westall CA (2009) Reduced grating acuity associated with retinal toxicity in children with infantile spasms on vigabatrin therapy. Invest Ophthalmol Vis Sci 50:4011–4016 29. Moraes MH, Montenegro MA, Franzon RC, Avila JO, Guerreiro MM (2005) Efficacy and tolerability of vigabatrin in West syndrome. Arq Neuropsiquiatr 63:469–473 30. Abtahi MA, Abtahi SH, Fazel F, Roomizadeh P, Etemadifar M, Jenab K, Akbari M (2012) Topiramate and the vision: a systematic review. Clin Ophthalmol 6:117–131 31. Al-Baradie RS, Elseed MA (2011) West syndrome, can topiramate be on top? Neurosciences (Riyadh) 16:53–56 32. Mayer DL, Beiser AS, Warner AF, Pratt EM, Raye KN, Lang JM (1995) Monocular acuity norms for the Teller Acuity Cards between ages one month and four years. Invest Ophthalmol Vis Sci 36:671–685 33. Saloma˜o SR, Ventura DF (1995) Large sample population age norms for visual acuities obtained with Vistech/Teller Acuity Cards. Invest Ophthalmol Vis Sci 36:657–670 34. Biglan AW, Davis JS, Cheng KP, Pettapiece MC (1996) Infantile exotropia. J Pediatr Ophthalmol Strabismus 33:79–84 35. Wygnanski-Jaffe T, Panton CM, Buncic JR, Westall CA (2009) Paradoxical robust visual evoked potentials in young patients with cortical blindness. Doc Ophthalmol 119:101–107 36. Malkowicz DE, Myers G, Leisman G (2006) Rehabilitation of cortical visual impairment in children. Int J Neurosci 116:1015–1033
123
ORIGINAL RESEARCH ARTICLE
Contributing factors to VEP grating acuity deficit and inter-ocular acuity difference in children with cerebral visual impairment Nı´vea Nunes Cavascan • Solange Rios Saloma˜o • Paula Yuri Sacai • Josenilson Martins Pereira • Daniel Martins Rocha • Adriana Berezovsky
Received: 15 July 2013 / Accepted: 9 December 2013 Ó Springer-Verlag Berlin Heidelberg 2013
Abstract Purpose To investigate contributing factors to visual evoked potential (VEP) grating acuity deficit (GAD) and inter-ocular acuity difference (IAD) measured by sweep-VEPs in children with cerebral visual impairment (CVI). Methods VEP GAD was calculated for the better acuity eye by subtracting acuity thresholds from mean normal VEP grating acuity according to norms from our own laboratory. Deficits were categorized as mild (0.17 B deficit \ 0.40 log units), moderate (0.40 B deficit \ 0.70 log units) or severe (deficit C0.70 log units). Maximum acceptable IAD was 0.10 log units.
N. N. Cavascan (&) S. R. Saloma˜o P. Y. Sacai J. M. Pereira D. M. Rocha A. Berezovsky Departamento de Oftalmologia, Escola Paulista de Medicina, Universidade Federal de Sa˜o Paulo, Rua Botucatu 822, Sa˜o Paulo, SP 04023-062, Brazil e-mail: [email protected] S. R. Saloma˜o e-mail: [email protected]
Results A group of 115 children (66 males—57 %) with ages ranging from 1.2 to 166.5 months (median = 17.7) was examined. VEP GAD ranged from 0.17 to 1.28 log units (mean = 0.68 ± 0.27; median = 0.71), and it was mild in 23 (20 %) children, moderate in 32 (28 %) and severe in 60 (52 %). Severe deficit was significantly associated with older age and anti-seizure drug therapy. IAD ranged from 0 to 0.49 log units (mean = 0.06 ± 0.08; median = 0.04) and was acceptable in 96 (83 %) children. Children with strabismus and nystagmus had IAD significantly larger compared to children with orthoposition. Conclusion In a large cohort of children with CVI, variable severity of VEP GAD was found, with more than half of the children with severe deficits. Older children and those under anti-seizure therapy were at higher risk for larger deficits. Strabismus and nystagmus provided larger IADs. These results should be taken into account on the clinical management of children with this leading cause of bilateral visual impairment. Keywords Cerebral visual impairment Grating acuity Sweep-VEP Child
P. Y. Sacai e-mail: [email protected] J. M. Pereira e-mail: [email protected] D. M. Rocha e-mail: [email protected] A. Berezovsky e-mail: [email protected]
Introduction Cerebral visual impairment (CVI) is a pediatric neurological disorder caused by post-chiasmatic brain lesion that results in bilateral visual loss [1, 2]. Clinically, it
123
Doc Ophthalmol
manifests as a deficit in visual acuity with normal ocular structures in both eyes [1, 2]. In general, contrast sensitivity is reduced and visual field is constricted, hemianopic or ‘‘Swiss cheese’’ [3–6]. Neurobehavioral signs include preference for brightly colored and moving objects instead of stationary or sharp black and white contrasts [1, 3, 7]. Visual response to complex visual scene used to be reduced, delayed or suppressed, especially in the presence of competing auditory or tactile stimuli [7]. Visual performance apparently is better in familiar settings, and it varies according to the moment, changes in the environment, illumination conditions and visual attention [1, 8]. Ocular movement abnormalities such as strabismus and nystagmus are frequent in patients CVI, and they can find it difficult to make the appropriate eye, head and neck movement to look at an object of visual interest, contributing to the poor visual behavior [2–4]. CVI is the leading cause of childhood blindness in developed countries, and its prevalence has been increasing in middle-income countries due to improvement in medical care and diagnostic equipment. Technological advances of neonatal units allow many premature infants to survive, who are an at-risk group for the occurrence of diseases related to prematurity and its complications [2, 3, 9]. The most common cause of CVI is the hypoxic/ ischemic insult whose damage location depends on the vascular anatomy and the brain maturity [3, 10]. Around 60 % of children with hypoxic–ischemic encephalopathy have CVI [3, 10]. Other etiologies include head injury, hydrocephalus, ventriculo-peritoneal shunt failure, epilepsy, infections, intrauterine exposure to drugs, neurodegenerative or metabolic diseases, tumor and malformation of brain structures [1–4, 8, 10, 11]. Cerebral damages represent a widespread impact, affecting not only the visual system, but also several brain areas, leading to other neurological problems, such as delay in neuropsychomotor development, cerebral palsy, seizures, hearing loss, microcephaly, meningomyelocele and progressive degeneration of central nervous system [1, 8, 9, 11]. Diagnosis of CVI is based on the triple assessment of neurological condition, ocular structure and visual function. Quantitative information about visual acuity deficit supports ophthalmological intervention, establishment of educational strategies and habilitation programs, and moreover, it facilitates social inclusion [5, 12].
123
Sweep visual evoked potential (sweep-VEP) is an electrophysiological technique that provides feasible and reliable grating acuity measurement as showed previously [13, 14]. This objective method does not require verbal, motor or behavioral response from the patient [2]. Then, it is a useful tool for vision evaluation in patients with CVI, which often have limited cognitive and verbal abilities [6, 15]. Reports of children with CVI often present binocularly measured acuities, but seldom show acuity deficit from the normal population and inter-ocular acuity differences (IADs). The purpose of this study was to determine grating acuity deficit (GAD) magnitude measured by sweep-VEP and IAD in children with CVI.
Materials and methods Participants Participants were 115 children (66 males—57 %) with CVI referred for VEP grating acuity measurement to the Clinical Electrophysiology of Vision Laboratory of the Department of Ophthalmology of Federal University of Sa˜o Paulo (UNIFESP), Brazil, in the last 10 years. This retrospective cross-sectional observational study was approved by UNIFESP Committee of Ethics in Research and followed the tenets of the Declaration of Helsinki. Children with CVI were diagnosed clinically on the basis of bilaterally reduced visual acuity, with evidence of post-chiasmatic brain lesion, poor visual behavior and normal ocular structure. Participants with no ophthalmic examination and uncorrected refractive errors were excluded. A total of 1,634 charts were revised, and 245 potential CVI patients were selected. Out of them, 130 were excluded because they presented only binocular VEP grating acuity measurement and/or optic disk pallor. Sweep-VEP measurements were taken from children with CVI aging from 1.2 to 166.5 months (median = 17.7). Thirty-four children were tested in the first year of life (1.2–11.8), 39 in second (12.6–23.3) and 42 from the second year of life on (24.1–166.5). Main etiology was prematurity (n = 28; 24 %), followed by seizures (n = 18; 16 %), perinatal hypoxia (n = 9; 8 %), brain malformations (n = 8; 7 %), acquired hypoxia (n = 7; 6 %) and hydrocephalus (n = 5; 4 %). In 22 (19 %) children, CVI was
Doc Ophthalmol
caused by a combination of two or more etiologies. Other causes (n = 18; 16 %) included encephalitis, meningitis, traumatic brain injury and genetic, infectious or metabolic disorders. Prematurity was considered for children with gestational age of 37 weeks or less according to the American Academy of Pediatrics [16], the Royal College of Ophthalmologists and the British Association of Perinatal Medicine [17]. Regarding ocular motility most patients were exotropic (n = 58; 66 %), typically with variable angle. Out of 88 cases of strabismus, 69 (78 %) were manifest and 19 (22 %) intermittent, as well as 46 (52 %) were unilateral and 42 (48 %) alternating. Thirty-two (28 %) children presented nystagmus that was manifest in 19 (59 %), latent in 7 (22 %) and occasional in 6 of them (19 %). Horizontal nystagmus was found in vast majority of cases (n = 27; 84 %) and vertical nystagmus in 4 (13 %) and only 1 (3 %) child had torsional nystagmus. Eighty-one (70 %) patients were in use of oral antiseizure drugs, and 40 (49 %) children received exclusive monotherapy (17 phenobarbital, 10 valproic acid, 5 benzodiazepine, 4 carbamazepine, 2 oxcarbamazepine, 2 topiramate). Sodium valproate was the main anti-seizure drug prescribed (n = 33; 41 %), mostly in association with one or more of the following substances: benzodiazepine, nitrazepam, phenobarbital and vigabatrin. Four parents did not know the medication in usage. All children presented neuropsychomotor development delay, and 75 % of them (n = 86) were included in habilitation program and/or visual stimulation. Out of 115 children, 35 (30 %) used glasses. Spherical equivalent in right eye ranged from -7.50 to ?6.00 diopters (mean = ?0.26 ± 1.42; median = 0.00) and in left eye from -7.50 to ?5.50 diopters (mean =?0.23 ± 1.37; median = 0.00). Sweep-VEP procedure Monocular grating acuity testing was performed using the PowerDiva (digital infant vision assessment) sweep-VEP system (the Smith-Kettlewell Eye Research Institute, San Francisco, USA) developed by Norcia [18]. This system utilizes two interfacing Macintosh G3 computers: the ‘‘host’’ computer selects the stimulus, establishes stimulus and trial parameters and analyzes the evoked response; the ‘‘video’’ computer controls the display monitor and shows the
Table 1 Testing distances and spatial frequency sweep Testing distance (cm)
Spatial frequency at sweep start (cycles/ degree)
30
0.10
50
0.50
9.46
80
1.00
15.14
100
1.00
18.92
150
2.00
28.38
Spatial frequency at sweep end (cycles/ degree) 3.08
stimuli to the subject. The stimuli were phase-reversal sine-wave gratings presented on a 17.5 in. (29 9 38 cm) high-resolution monochromatic video monitor (M20DCD4RE-Richardson ElectronicsÒ). Stimuli orientation was set vertical for participants without nystagmus and for those with either vertical or torsional nystagmus. Horizontal orientation was used for children with horizontal nystagmus [19]. Mean luminance (142.35 cd/m2) as well as contrast (80 %) was kept constant throughout the session. Patterns were temporally alternated in counterphase with a temporal modulation of 6.6 Hz. During the test, children were positioned seated in their parent’s lap or seated in their wheelchair in front of a display monitor. Their attention was attracted with small toys dangling over the center of the display. Sweep-VEP was recorded only when the subject was alert and fixating the stimuli. Testing distance was determined according to the patient’s age and visual behavior. The ranges of spatial frequencies and testing distances are shown in Table 1. Test stimulation field was 52° 9 65° (for 30 cm distance) to 11° 9 14° (for 150 cm distance), calculated for both vertical and horizontal monitor lengths. Ten linearly spaced spatial frequencies were presented, 1/s, starting at a low spatial frequency. The procedure was performed in a dark room, and the eye of better visual behavior was tested first. Sweep-VEP test, including setup and rest breaks, typically lasted 30–60 min, depending on the subject’s age and cooperation. Responses were obtained from electroencephalogram (EEG) electrodes (Grass Gold Disc Electrodes— E6GH) attached to the scalp with electrode cream and cotton pads. A headband (3M Coban Self-adherent Wrap 1581) was used to keep the electrodes in place. According to the normative protocol of our own laboratory [20], EEG was recorded from two bipolar placements (O1 and O2), 2–3 cm to the left and right
123
Doc Ophthalmol
Fig. 1 Sweep-VEP response; record from right eye of a participant; sweep sets from 1.00 to 18.92 cycles/degree (linear scale); two thresholds were obtained, channel one (O1–Cz) and channel two (O2–Cz); Dots represent noise registered for each one of 10 linearly spaced spatial frequencies presented; grating acuity was estimated by linear fit and extrapolation to zero amplitude, i.e., the value at which the regression line touches the axis of spatial frequencies; the final acuity score was calculated
in logMAR using the results of the channel two with better threshold (8.69 cycles/degree) with the highest signal-to-noise ratio (19.17). Ch channel; 2F1s harmonic of the stimulus frequency; lV microvolts; Spat Freq spatial frequency; Thrsh grating acuity threshold (cycles/degree); SNR signal-to-noise ratio; Sc SNR maximum SNR within the cursors (dotted lines) that define the data used to estimate threshold; Pk SNR maximum SNR at peak mean amplitude in the record
of a ground electrode placed 1 cm above the inion on the midline (Oz), following the international 10–10 electrode placement system [21]. A reference electrode was placed at the vertex (Cz). Visual evoked responses were isolated in real time (sampling rate = 397 Hz) from the EEG by adaptive filter (bandpass). The potential differences were amplified (Neurodata Acquisition System P15, Grass Instrument Co; gain = 10,000; -3 db cutoff at 1 and 100 Hz). Three to 15 repetitions of the sweep were obtained and vector averaged. Amplitude and phase of the first (6 Hz) and second harmonics (12 Hz) of the stimulus frequency were calculated for each channel by discrete Fourier transform. Grating acuity was estimated with an automated algorithm which performed a linear fit and extrapolation to zero amplitude for the final descending limb of the function relating VEP second harmonic amplitude to linear spatial frequency. A signal-to-noise ratio (SNR), calculated as the ratio of power at stimulus frequency to mean power at frequencies ±2 Hz, at peak mean amplitude of 3:1, was required [22]. A SNR of 3:1 corresponds to a false alarm rate of 0.4 %, and it provides an adequate protection level when combined with the phase consistency criteria. In all cases, two thresholds (one for each channel, O1 and O2) were obtained (Fig. 1). The final acuity score was calculated in logMAR (logarithm of the minimum angle of resolution) using the results in cycles per degree of visual angle of the
better threshold channel with the highest SNR, according to the formula:
123
log MAR acuity score ¼ logð30= acuity threshold in cycles per degreeÞ VEP grating acuity threshold and IAD were compared . normative data from our own laboratory [20]. Maxito mum acceptable IAD was 0.10 log units. Since the normal limits of visual acuity vary with age in the age range targeted in this study, all measurements were converted to visual acuity deficits (i.e., the number of log units relative to published age-corrected mean normal values). Then, VEP GAD was calculated for the better acuity eye (log units) by subtracting acuity thresholds from mean normal visual acuity according to age norms [20]. Deficits were categorized as mild (0.17 B GAD \ 0.40 log units), moderate (0.40 B GAD \ 0.70 log units) or severe (GAD C 0.70 log units).
Statistical analysis Statistical analyses were carried out using Stata Statistical Software: Release 12.0. p values equal to or less than 0.05 were considered statistically significant with two-tailed rejection region. On the first-level analysis, GAD and IAD were analyzed by gender, age group (\12, 12–24 and C24 months), main etiology, ocular motility
Doc Ophthalmol Fig. 2 VEP grating acuity threshold; Circles indicate individual values of VEP grating acuity from better vision eye of 115 children with cortical visual impairment; Dashed and dotted lines represent normal mean and normal lower limit, respectively, from our own laboratory (Saloma˜o et al. [20])
condition, use of anti-seizure medication and inclusion in habilitation and/or early visual stimulation program. Statistical models included Student’s t test and oneway analysis of variance (ANOVA), followed by multiple comparison procedure (Tukey’s test), when applicable. The skewness/kurtosis test validated the parametric data analysis methods. If normality test failed, non-parametric tests were used: Mann–Whitney rank sum test and Kruskal–Wallis ANOVA on ranks, followed by multiple comparison procedure (Dunnet’s test), when applicable. On the second-level analysis, multiple logistic regression model was used to investigate the association between severe deficit and age group (\12 and C12 months), gender, presence of prematurity, seizures and ocular motility disorders, use of anti-seizure medication and inclusion in habilitation and/or visual
stimulation program. Odds ratios (OR) with 95 % confidence intervals (CI) were estimated.
Results Inter-ocular acuity difference VEP grating acuity thresholds ranged from 0.23 to 1.41 logMAR (mean = 0.93 ± 0.26; median = 0.95) for the better vision eye (Fig. 2) and 0.26–1.70 logMAR (mean = 1.00 ± 0.27; median = 1.02) for the worse vision eye. IAD ranged from 0 to 0.49 log units (mean = 0.06 ± 0.08; median = 0.04). Normal IAD (B 0.10 log units) was found in 96 (83 %) children. Children with strabismus and nystagmus had IAD significantly larger when compared to children
Table 2 Mean inter-ocular acuity difference, their respective standard deviations and median of 115 participants distributed by ocular motility condition n
%
IAD (log units) Mean
SD
Strabismus
65
56
0.06
0.05
0.04
Strabismus ? nystagmus
23
20
0.13*
0.13
0.08
9
8
0.03
0.03
0.03
18
16
0.03
0.03
0.02
Orthoposition
Statistical model
0.0157*
Kruskal–Wallis
Median
Ocular motility
Nystagmus
p value
IAD inter-ocular acuity difference * p B 0.05
123
Doc Ophthalmol Table 3 Mean VEP grating acuity deficit, their respective standard deviations and median of 115 participants distributed by age group, gender, main etiology, ocular motility condition, anti-seizure medication usage and inclusion in habilitation and/or visual stimulation program n
%
GAD (log units) Mean
SD
Median
p value
Statistical model
0.0002*
ANOVA
Age group (months) \12
34
30
0.53*
0.21
0.57
12–24
39
34
0.77
0.22
0.76
C24
42
36
0.73
0.30
0.76
66 49
57 43
0.71 0.66
0.25 0.28
0.74 0.69
0.3384
t test
Prematurity
28
24
0.62
0.25
0.65
0.4608
ANOVA
Seizures
18
16
0.69
0.28
0.70
0.1309
ANOVA
0.0008*
t test
0.1988
Mann–Whitney
Gender Male Female Main etiology
Perinatal hypoxia
9
8
0.73
0.31
0.73
Brain malformations
8
7
0.61
0.18
0.63
Acquired hypoxia
7
6
0.69
0.28
0.75 0.46
Hydrocephalus
5
4
0.54
0.32
Others
18
16
0.75
0.31
0.83
Miscellaneous
22
19
0.76
0.24
0.75
Ocular motility Strabismus
65
56
0.64
0.27
0.63
Strabismus ? nystagmus
23
20
0.80
0.23
0.78
Nystagmus Orthoposition Anti-seizure medication No
9
8
0.70
0.23
0.69
18
16
0.68
0.28
0.74
34
30
0.56
0.24
0.55
81
70
0.74*
0.26
0.76
No
29
25
0.63
0.25
0.68
Yes
86
75
0.70
0.27
0.74
Yes Habilitation program
GAD VEP grating acuity deficit; ANOVA one-way analysis of variance * p B 0.05
with orthoposition (Kruskal–Wallis; H = 10.397; p = 0.0157) as shown in Table 2. Additionally, IAD was significantly larger (mean = 0.09 ± 0.10; median = 0.07) in patients with deviation with preference for fixation by the fellow eye when (z = 1.934; p = 0.0531) compared to those with alternating deviation (mean = 0.06 ± 0.06; median = 0.04). IAD was comparable either for manifest (mean = 0.08 ± 0.09; median = 0.04) or for intermittent (mean = 0.07 ± 0.06; median = 0.05) deviation. Out of 19 children with IAD above the maximum acceptable ([0.10 log units), all but 1 had
123
strabismus. Twelve presented fixation preference by the fellow eye, and 6 children had alternating strabismus. Exotropia (n = 13) was found in the majority of patients. IAD was comparable for age group, gender, main etiology, using anti-seizure medication and inclusion in habilitation program. VEP grating acuity deficit GAD ranged from 0.17 to 1.28 log units (mean = 0.68 ± 0.27; median = 0.71). Mild GAD was found in 23 (20 %) children, moderate in 32
Doc Ophthalmol Table 4 Odds ratios and 95 % confidence intervals in multivariate analysis of factors related to severe VEP grating acuity deficit Variable
OR (95 % CI)
p value
Age (months) \12
Reference
C12
2.56 (1.04–6.34)
0.041*
Gender Male
Reference
Female
1.20 (0.52–2.75)
0.667
Prematurity No
Reference
Yes
0.76 (0.31–1.83)
0.537
Reference 0.65 (0.20–2.06)
0.463
Seizures No Yes
Ocular motility disorder No
Reference
Yes
0.86 (0.27–2.73)
0.805
Anti-seizure medication No
Reference
Yes
5.30 (2.09–13.42)
\0.001*
Habilitation program No
Reference
Yes
1.18 (0.44–3.14)
0.739
OR odds ratio; CI confidence interval * p B 0.05
(28 %) and severe in 60 (52 %). GAD was significantly smaller for children tested in the first year of life (ANOVA, Tukey’s test; F = 9.45; p = 0.0002), and it was significantly larger in anti-seizure medication users (t = -3.4387; p = 0.0008). GAD was comparable for gender, main etiology, ocular motility condition and inclusion in habilitation programs as shown in Table 3. In multiple logistic regression modeling (Table 4), severe GAD was significantly associated with older age [OR 2.58 (95 % CI 1.04–6.38)] and anti-seizure drug therapy [OR 5.58 (95 % CI 2.17–14.37)]. No interactions were observed with gender, presence of prematurity, seizures and ocular motility disorders, as well as inclusion in therapeutic program.
Discussion Visual acuity evaluation can be challenging in children with CVI. Sweep-VEP is a useful method to
measure grating acuity in this population for many reasons [2]. First, the luminescent stimulus source of the monitor tends to be attractive, since patients with CVI usually show a predilection for bright stimuli. Besides, the child is positioned directly in front of the monitor, so that the VEP stimulus subtends a wide angle of the subject’s visual field, facilitating the recording in cases of motor deficits and visual field defects. In addition, the equipment allows the calculation of the average responses recorded and choice for individual responses, and the examiner can control (pause and restore) the stimulus presentation. Most previous investigations demonstrate reduced VEP grating acuity and slower visual development measured by sweep-VEP in children with CVI [5–7, 12, 15]. Our study included deficit and IAD analysis in this group. We recorded a median VEP grating acuity of 0.93 logMAR (in the better vision eye), which was worse than values from previous studies (0.37–0.88 logMAR) [5, 6, 12, 15]. In these studies, acuity data were obtained under binocular viewing, condition that certainly gets more overall cooperation and attention from the children and can reduce noise of the evoked response [20, 23, 24]. But data of monocular measure as well as IADs contribute to precise and reliable clinical diagnosis of CVI, implying bilateral visual loss. Our results showed acceptable IAD (B0.10 log units) in the vast majority of cases, speaking on behalf of the absence of ocular abnormalities and evidence of post-chiasmatic brain lesion. Additionally, IAD was larger in children with strabismus and nystagmus as expected, whereas these conditions are recognizably predisposing factors to disruption of binocular interaction and simultaneous visual maturation of both eyes [25, 26]. In our group, mean GAD was 0.68 log units, which is more severe when compared to 0.40 log units reported by Watson et al. [15]. Considering that our sample was younger than this previous study (29.97 months vs 12.10 years), roughly it could corroborate with previous findings that early neurological events are associated with more severe visual condition [10], but we cannot affirm the timing of neurological event from Watson’s study. According to magnitude, we found moderate or severe GAD in the vast majority of the cases (80 %, n = 92), finding in agreement with Birch and Bane [27], although obtained by different methods of measurement (electrophysiological vs psychophysical). Analysis by age group showed larger GAD from the second year of life on, suggesting small
123
Doc Ophthalmol
improvement in VEP grating acuity with increasing age and consequently slower visual maturation than normal, as reported previously [5, 7]. Related to the use of anti-seizure medication, GAD was significantly larger in users when compared with non-users. Nowadays, it is known that the cognitive evolution of epileptic patients is significantly linked to the seizure control, which provides a better maintenance of the bioelectrical activities of both cortical and subcortical structures. However, some of the antiseizure drugs widely used such as vigabatrin [28, 29], lamotrigin and topiramate [30, 31] present both neurological and retinal toxicities as side effects that limit their prolonged use. In our study, 21 % (n = 17) of our children received these types of drug, but none of the patients of our sample has been submitted to electroretinogram evaluation. We also did not correlate GAD with the time of drug treatment. Therefore, we cannot rule out its possible action on the retina and its effect on brain electrical conduction. Although antiseizure drugs may lead to attention deficit and irregular electrical activity, sweep-VEP reliability is ensured by speed of execution, discrete Fourier analysis that improves signal-to-noise ratio and extrapolation method that is amplitude-insensitive and does not depend critically on factors such as thickness of skull and small variations in electrode localization [13]. As expected, there was no difference in GAD analyzed by gender, since the visual development is comparable in boys and girls [32, 33]. In our study, hypoxic–ischemic etiologies (prematurity, perinatal hypoxia and acquired hypoxia) put together were main causes of CVI, according to previous reports [1, 6, 8, 10, 11, 27]. GAD also was comparable for etiologic factors, in agreement with previous reports (2007, 2010) [5, 12]. Ocular motility disorders are usually associated with CVI. Out of all our children, only 16 % (n = 18) had orthoposition of the visual axis. Previous investigations showed frequency of strabismus and nystagmus, respectively, of 29–90 and 10–56 % in child population with CVI [7, 10, 11, 27]. Nevertheless, in the current study, there was no difference in GAD when analyzed by ocular motility conditions. Exotropia was found in 66 % of cases, in agreement with literature that reports constant exotropia in childhood is often associated with neurological dysfunction [34]. Surprisingly, children included in habilitation program and/or early visual stimulation (75 %, n = 86)
123
showed comparable GAD to children who did not receive in accompanying therapeutic. However, sweep-VEP depends on the integrity of the maculooccipital pathway (Brodmann area 17), and it reflects central visual function. But patient’s visual behavior depends on this pathway and further visual association cortex (Brodmann areas 18–21), mechanisms of attention and motor responses [35]. Habilitation program and/or early visual stimulation provides to the brain the opportunity to form appropriate connections in order to compensate delayed neuropsychomotor development by the use of residual vision, but not necessarily visual acuity recovery [9, 36]. In multivariable analysis, age and use of antiseizure medication remained significantly associated with severe deficit. Children tested from the second year of life on were 2.6 times as likely to present severe deficit compared with children tested in the first year after adjusting for remaining variables. As well as drug users were 5.3 times as likely to present severe deficit compared with non-users. We conclude that users of anti-seizure medication and children tested from the second year of life on presented significantly higher odds for severe deficit. Further longitudinal investigation is helpful to confirm risk factors associated with visual deficit severity. Besides, to exclude toxic effects of the anti-seizure medication on the retina would be necessary retinal function evaluation by full-field electroretinography, which is impracticable in vast majority of children with CVI without sedation. As a limitation of this study, our participants did not undergo electroretinography procedure. We also call attention that VEP grating acuity measurement in CVI is not only useful clinically, but also extremely valuable in enabling improvement in visual behavior, motor skills and social interaction through habilitation programs. Acknowledgments This study was supported by the Conselho Nacional de Desenvolvimento Cientı´fico e Tecnolo´gico-CNPq (Brası´lia, DF, Brazil) research fellowship (A. Berezovsy and S. R. Saloma˜o). We thank Dr. Flavio E. Hirai for statistical assistance. Conflict of interests
None.
References 1. Jan JE, Groenveld M, Sykanda AM, Hoyt CS (1987) Behavioral characteristics of children with permanent cortical visual impairment. Dev Med Child Neurol 29:571–576
Doc Ophthalmol 2. Good WV, Jan JE, Burden SK, Skoczenski A, Candy R (2001) Recent advances in cortical visual impairment. Dev Med Child Neurol 43:56–60 3. Good WV, Jan JE, deSa L, Barkovich AJ, Groenveld M, Hoyt CS (1994) Cortical visual impairment in children: a major review. Surv Ophthalmol 38:351–364 4. Dutton GN, Jacobson LK (2001) Cerebral visual impairment in children. Semin Neonatol 6:477–485 5. Watson T, Orel-Bixler D, Haegerstrom-Portnoy G (2007) Longitudinal quantitative assessment of vision function in children with cortical visual impairment. Optom Vis Sci 87:80–86 6. Good WV, Huo C, Norcia AM (2012) Spatial contrast sensitivity vision loss in children with cortical visual impairment. Invest Ophthalmol Vis Sci 53:7730–7734 7. Lim M, Soul JS, Hansen RM, Mayer DL, Moskowitz A, Fulton AB (2005) Development of visual acuity in children with cerebral visual impairment. Arch Ophthalmol 123:1215–1220 8. Whiting S, Jan JE, Wong PKH, Flodmark O, Farrell K, McCormick AQ (1985) Permanent cortical visual impairment in children. Dev Med Child Neurol 27:730–739 9. Groenveld M, Jan JE, Leader P (1990) Observations on the habilitation of children with cortical visual impairment. J Vis Impair Blind 84:11–15 10. Hoyt CS (2003) Visual function in the brain-damaged child. Eye 17:369–384 11. Huo R, Burden SK, Hoyt CS, Good WV (1999) Chronic cortical visual impairment in children: aetiology, prognosis, and associated neurological deficits. Br J Ophthalmol 83:670–675 12. Watson T, Orel-Bixler D, Haegerstrom-Portnoy G (2010) Early visual-evoked potential acuity and future behavioral acuity in cortical visual impairment. Optom Vis Sci 84:471–480 13. Tyler CW, Apkarian PA, Levi DM, Nakayama K (1979) Rapid assessment of visual function: an electronic sweep technique for the pattern visual evoked potential. Invest Ophthalmol Vis Sci 18:703–712 14. Good WV (2001) Development of a quantitative method to measure vision in children with cortical visual impairment. Tr Am Ophth Soc 99:253–269 15. Watson T, Orel-Bixler D, Haegerstrom-Portnoy G (2009) VEP vernier, VEP grating, and behavioral grating acuity in patients with cortical visual impairment. Optom Vis Sci 86:774–780 16. American Academy of Pediatrics, Section on Ophthalmology (2001) Screening examination of premature infants for retinopathy of prematurity. Pediatrics 108:809–811 17. The report of a Joint Working Party of The Royal College of Ophthalmologists and the British Association of Perinatal Medicine (1996) Retinopathy of prematurity: guidelines for screening and treatment. Early Hum Dev 46:239–258 18. Norcia AM (1999) PowerDiva manual. Version 1.1.0. The Smith-Kettlewell Eye Research Institute, San Francisco 19. Meiusi RS, Lavoie JD, Summers CG (1993) The effect of grating orientation on resolution acuity in patients with nystagmus. J Pediatr Ophthalmol Strabismus 30:259–261
20. Saloma˜o SR, Ejzenbaum F, Berezovsky A, Sacai PY, Pereira JM (2008) Age norms for monocular grating acuity measured by sweep-VEP in the first three years of age. Arq Bras Oftalmol 71:475–479 21. Nuwer MR, Comi G, Emerson R, Fuglsang-Frederiksend A, Gue0 rite J, Hinrichsf H, Ikedag A, Luccash FJC, Rappelsburger P (1998) IFCN standards for digital recording of clinical EEG. Int Fed Clin Neurophysiol Electroencephalogr Clin Neurophysiol 106:259–261 22. Norcia AM, Tyler CW (1985) Spatial frequency sweepVEP: visual acuity during the first year of life. Vis Res 25:1399–1408 23. Hamer RD, Norcia AM, Tyler CW (1989) The development of monocular and binocular VEP acuity. Vis Res 29:397–408 24. Tyler CW, Apkarian PA (1985) Effects of contrast, orientation and binocularity in the pattern evoked potential. Vis Res 25:755–766 25. Agrawal R, Conner IP, Odom JV, Schwartz TL, Mendola JD (2006) Relating binocular and monocular vision in strabismic and anisometropic amblyopia. Arch Ophthalmol 124:844–850 26. Fu VLN, Bilonik RA, Felius J, Hertle RW, Birch EE (2011) Visual acuity development of children with infantile nystagmus syndrome. Invest Ophthalmol Vis Sci 52:1404–1411 27. Birch EE, Bane MC (1991) Forced-choice preferential looking acuity of children with cortical visual impairment. Dev Med Child Neurol 33:722–729 28. Durbin S, Mirabella G, Buncic JR, Westall CA (2009) Reduced grating acuity associated with retinal toxicity in children with infantile spasms on vigabatrin therapy. Invest Ophthalmol Vis Sci 50:4011–4016 29. Moraes MH, Montenegro MA, Franzon RC, Avila JO, Guerreiro MM (2005) Efficacy and tolerability of vigabatrin in West syndrome. Arq Neuropsiquiatr 63:469–473 30. Abtahi MA, Abtahi SH, Fazel F, Roomizadeh P, Etemadifar M, Jenab K, Akbari M (2012) Topiramate and the vision: a systematic review. Clin Ophthalmol 6:117–131 31. Al-Baradie RS, Elseed MA (2011) West syndrome, can topiramate be on top? Neurosciences (Riyadh) 16:53–56 32. Mayer DL, Beiser AS, Warner AF, Pratt EM, Raye KN, Lang JM (1995) Monocular acuity norms for the Teller Acuity Cards between ages one month and four years. Invest Ophthalmol Vis Sci 36:671–685 33. Saloma˜o SR, Ventura DF (1995) Large sample population age norms for visual acuities obtained with Vistech/Teller Acuity Cards. Invest Ophthalmol Vis Sci 36:657–670 34. Biglan AW, Davis JS, Cheng KP, Pettapiece MC (1996) Infantile exotropia. J Pediatr Ophthalmol Strabismus 33:79–84 35. Wygnanski-Jaffe T, Panton CM, Buncic JR, Westall CA (2009) Paradoxical robust visual evoked potentials in young patients with cortical blindness. Doc Ophthalmol 119:101–107 36. Malkowicz DE, Myers G, Leisman G (2006) Rehabilitation of cortical visual impairment in children. Int J Neurosci 116:1015–1033
123
