Doc Ophthalmol (2015) 130:221–229 DOI 10.1007/s10633-015-9483-0
ORIGINAL RESEARCH ARTICLE
Pattern visual evoked potentials for identifying malingering I-Ting Sun • Jong-Jer Lee • Hsiu-Mei Huang Hsi-Kung Kuo
•
Received: 21 October 2014 / Accepted: 21 January 2015 / Published online: 25 January 2015 Ó Springer-Verlag Berlin Heidelberg 2015
Abstract Purpose To investigate the efficacy of pattern visual evoked potentials (VEPs) in evaluating objective visual acuity (VA) and discriminating malingerers. Methods Two hundred and forty-nine eyes of 249 patients aged 20–65 years were included. There were 147 eyes with macular diseases (group 1) and 102 eyes with optic nerve diseases (group 2). Amplitudes and latencies were analyzed and correlated with bestcorrected visual acuity by a regression analysis. We found the best-correlated mode of pattern VEP, determined the relations, and then calculated the pattern VEP-estimated VA (PVEP-VA) of all 249 eyes, another 30 malingering eyes, 13 eyes with macular diseases, and 17 eyes with optic nerve diseases, and used a receiver operating characteristic (ROC) curve to determine a cutoff for acceptable variance between PVEP-VA and subjective VA to discriminate malingerers. Results The best correlation was between the amplitude of 500 checkerboard size (Amp500 ) and VA in every group. Significant correlation was between Amp500 and VA, where p \ 0.0001 in group 1 and p = 0.020 in group 2. A logarithmic curve best fitted
I.-T. Sun J.-J. Lee H.-M. Huang H.-K. Kuo (&) Department of Ophthalmology, Kaohsiung Chang Gung Memorial Hospital and Chang Gung University College of Medicine, 123 Dapi Road, Niao Sung District, Kaohsiung City, Taiwan e-mail: [email protected]
the correlation in the regression analysis, where y = 1.731 - 1.569x (R2 = 0.611, p \ 0.0001) in group 1 and y = 2.413 - 2.169x (R2 = 0.531, p \ 0.0001) in group 2 [x: log(Amp500 ), y: PVEPVA (logMAR)]. By using the relations and ROC curve, we determined a variance value of 0.4041 (logMAR) with 100 % sensitivity and 94.0 % specificity in group 1 and 0.3658 with 70.6 % sensitivity and 50.5 % specificity in group 2 to discriminate malingerers. Conclusions The pattern VEP amplitude of 500 checkerboard size was useful to assess VA and can be helpful in discriminating malingering from real disability. Keywords Pattern visual evoked potentials Visual acuity Objective assessment
Introduction The visual evoked potential (VEP) offers quantitative data on the function of the visual pathways and visual cortex. In recent years, the possibility of using the VEP to objectively assess the vision of patients has been examined. Objective visual acuity estimation by the VEP has been used in infants, children with cerebral visual impairment, and subjects suspected of nonorganic/functional visual loss, including malingering [1–5]. VEP-based visual acuity can be estimated from
123
222 Table 1 Diagnosis and the number of eyes in each group
Doc Ophthalmol (2015) 130:221–229
Diagnosis
Number of eyes
Group 1 Age-related macular disease
147
54.91 ± 10.16
113
59.60 ± 4.32
Macular pucker
13
39.32 ± 5.35
Choroidal neovascularization
11
43.44 ± 6.78
Diabetic maculopathy
10
40.22 ± 6.96
Group 2
102
45.44 ± 10.63
Optic neuritis
32
43.78 ± 9.86
Optic atrophy Traumatic optic neuropathy
22 26
45.30 ± 9.75 39.44 ± 8.64
Anterior inflammatory optic neuropathy
16
59.32 ± 4.89
6
43.79 ± 3.98
Toxic optic neuropathy
the sweep VEPs, steady-state VEPs, or pattern appearance VEPs. There are various ways to analyze the transient recordings; visual acuity can be expressed as the highest spatial frequency to which a response is recorded [6, 7] or by fitting a regression line or polynomial to amplitude response functions [8– 12]. Kurtenbach et al. [13] compared three methods of the VEP recordings to estimate visual acuity and found that the sweep VEP results were less variable and more reproducible than other methods, but require manual waveform detection and analysis. The pattern VEP is clinically attractive because of its ease of analysis and short recording time. The pattern VEP is generally used for the detection of visual disturbances, including optic neuritis, ischemic optic neuropathy, compressive optic neuropathy, demyelinization disease, and maculopathy [14, 15]. It has been used to assess objective visual acuity (VA) in previous studies, where only a limited number of cases were included [16–21]. However, there is little consensus regarding the correlation between the pattern VEP and VA, and the application for discriminating malingering. In this study with large number of cases, we investigated the efficacy of the pattern VEP to predict objective visual acuity and diagnose malingering.
Methods Participants With IRB approval, the medical records of 751 patients who received the pattern VEP examinations
123
Average age
between January 2008 and April 2013 at the same medical center were reviewed retrospectively. Patients who had anterior segment abnormalities and obvious brain lesions, those aged older than 65 or younger than 20 years, those with poor VEP waveforms, and those with glaucoma and normal and amblyopic eyes were excluded. If the patient had similar ocular diseases at both eyes, we selected the right eye. Finally, we selected 249 eyes of 249 patients and divided them into two groups as follows (Table 1): Group 1: 147 eyes with macular diseases, such as age-related macular disease (AMD), macular pucker, choroidal neovascularization (CNV), and diabetic maculopathy; and group 2: 102 eyes with optic nerve diseases, such as optic neuritis, optic atrophy, traumatic optic neuropathy, anterior ischemic optic neuropathy, and toxic optic neuropathy. Procedure All patients underwent biomicroscopic and dilated fundus examinations and underwent the VEP recordings. Additional examinations such as fluorescein angiography (FAG), indocyanine green angiography (ICG), optical coherence tomography (OCT), visual field tests, color vision tests, and magnetic resonance imaging (MRI) were performed if necessary to help diagnose macular and optic nerve diseases. The bestcorrected visual acuities (BCVAs) were recorded in all patients with a Landolt C chart just before VEP examinations. The Landolt C chart VA was converted to logMAR VA for the statistical analysis. The transient pattern VEP was recorded using an LKC UTAS-E 3000 system (LKC Technologies Inc.,
Doc Ophthalmol (2015) 130:221–229
Gaithersburg, MD, USA) in accordance with the International Society for Clinical Electrophysiology of Vision (ISCEV) recommendations. The patients’ refractions were corrected with trial lenses using a Landolt C chart before the VEP was examined. All patients were instructed to maintain fixation at the center of a stimulus located at a distance of 1 m in front of a 20 9 30 cm black-and-white video display monitor (contrast 99 %). The reversal rate was 1 reversal per second. The checkerboard stimulus subtended a visual angle of 5.7° vertically and 8.5° horizontally on either side of the fixation. The P100 amplitude and latency for five different checkerboard sizes (1000 , 500 , 250 , 120 , and 60 min of arc) were recorded. The fixation stability of the eyes was closely monitored by an experienced electrophysiology technician. If the cooperation of the patient was poor, the VEP examination was repeated. All the VEP examinations were performed under the same laboratory conditions. Data analysis We recorded the P100 amplitude and latency for each checkerboard size (1000 , 500 , 250 , 120 , and 60 min of arc) in all patients. P100 amplitude and the latency of each checkerboard size were correlated with BCVA by multiple regression analyses using IBM SPSS Statistics 18 software. We performed a linear regression analysis first to investigate the confounding effect of age on the pattern VEP latency and amplitude parameters among the enrolled patients (aged 20–65 years) and found that there was no significant interaction between the VEP parameters and age groups. Therefore, we did not perform statistical adjustments to age match the results of the enrolled patients. In order to evaluate the diagnostic validity of the test in distinguishing malingering patients, we also Table 2 Diagnosis and number of malingering eyes
223
enrolled 30 malingering eyes of 30 malingerers and compared the pattern VEP-estimated VA (PVEP-VA, logMAR) of all 249 eyes and those 30 malingering eyes. Then, we made a receiver operating characteristic (ROC) curve to determine a cutoff value for acceptable variance between PVEP-VA and subjective visual acuity to discriminate malingerers. The malingerers were all exaggerators who had simply macular disease or optic nerve disease without other clinically detectable combined ophthalmic disorders and whose acuity was worse than expected, where 13 eyes with macular diseases (group 1m) and 17 eyes with optic nerve diseases (group 2m) (Table 2). The patients whose BCVA was incompatible with ocular anatomic status were suspected malingerers. The malingering eyes were diagnosed clinically if a discrepancy in the initially presented BCVA (VAi) and BCVA after the simulation examination (VAase) was greater than 1 line on the Landolt C chart. The VAase was assessed after certain technical procedures and behavior observations. Sensitive clinical tests, including fogging, prism dissociation test, fixation techniques, distance test, grimacing in front of the patient, handshaking test, and pupillary light reflexes, were used for the detection of malingering [22–25]. VAi and VAase were checked by an experienced clinician. Briefly, the stepwise algorithm of our study can be more intuitively understood using the following rule set: 1.
2.
3.
Find whether the P100 amplitude or latency of the pattern VEP was best correlated with subjective visual acuity. Use the best-correlated mode of the pattern VEP results to perform linear analyses to determine a relation. Enroll malingering patients and use the above relation to calculate the PVEP-VA.
Diagnosis
Number of eyes
Group 1m: Macular diseases
13
52.47 ± 10.03
7
59.89 ± 4.98
6
43.81 ± 9.89
Age-related macular disease Diabetic maculopathy Group 2m: Optic nerve diseases Optic neuritis
Average age
17
41.86 ± 12.41
12
41.92 ± 11.96
Toxic optic neuropathy
4
44.65 ± 9.87
Traumatic optic neuropathy
1
30
123
224
4.
5.
Doc Ophthalmol (2015) 130:221–229
Compare the PVEP-VA with the subjective VA to investigate the variance in all patients (malingerers and non-malingerers). Perform a ROC analysis to obtain a cutoff value for acceptable variance, where sensitivity is more relevant than specificity.
where p \ 0.0001 in group 1 and p = 0.020 in group 2 (Table 3). The mean VEP Amp500 parameters for each group were 5.10 ± 1.95 lV in group 1 (macular diseases) and 5.89 ± 2.49 lV in group 2 (optic nerve diseases). Pearson’s correlation coefficient between Amp500 and VA was -0.737 (p \ 0.0001) in group 1 and -0.656 (p \ 0.0001) in group 2. We found that a logarithmic curve fits best the correlation and was better than a straight line or reciprocal curve in the regression analysis (figures of straight line or reciprocal curve were not showed). The relations were y = 1.731 - 1.569x (R2 = 0.611, p \ 0.0001) in group 1 and y = 2.413 - 2.169x (R2 = 0.531, p \ 0.0001) in group 2 [x: log(Amp500 ), y: PVEP-VA (logMAR)] (Figs. 1, 2).
Results The demographic data of all patients are shown in Table 1. The mean ages were as follows: group 1, 54.91 ± 10.16 years; and group 2, 45.44 ± 10.63 years. Significant correlation was between the amplitude of 500 checkerboard size (Amp500 ) and VA in every group,
Table 3 Multiple linear regression models between the visual evoked potential data and visual acuity (logMAR) Model
Unstandardized coefficients
Standardized coefficients
B
Beta
t
Standard error
Significance
95 % Confidence interval for B Lower limit
Upper limit
Group 1 (macular diseases) Constant
.889
.391
2.275
.026
.110
-.003
.003
-.145
-1.048
.298
-.009
.003
Lat500
.000
.004
.020
.126
.900
-.007
.008
Lat250
.004
.003
.175
1.201
.234
-.003
.011
0
Lat1000
Lat12
1.668
.001
.003
.025
.243
.809
-.005
.006
Lat60
.001
.002
.004
.041
.968
-.004
.004
Amp1000
.006
.014
.042
.402
.689
-.022
.033
Amp500
-.117
.018
-.696
-6.361
.000
-.153
-.080
Amp250
-.006
.019
-.038
-.339
.736
-.044
.031
0
.001 .010
.012 .012
.012 .081
.121 .821
.904 .414
-.022 -.014
.025 .034
Amp12 Amp60
Group 2 (optic nerve diseases) Constant
1.055
.599
Lat1000
.001
.007
0
.010
.007
Lat250
-.002
.007
Lat120
.004
.004
-.010 .022
Amp500
-.074
Amp250 Amp120
Lat50
Lat60 Amp1000
Amp60
1.760
.087
-.160
2.269
.206
.838
-.013
.016
.347
1.331
.191
-.005
.025
-.082
-.307
.760
-.017
.012
.143
.869
.391
-.005
.013
.004
-.343
-1.759
.090
-.017
-.003
.028
.151
.796
.431
-.035
.079
.031
-.492
-2.433
.020
-.136
-.012
-.050
.028
-.349
-1.819
.080
-.137
-.022
.044
.026
.334
1.678
.102
-.009
.096
-.015
.016
-.130
-.981
.333
-.047
.016
Dependent variable: visual acuity (logMAR) Lat latency, Amp amplitude
123
.043
Doc Ophthalmol (2015) 130:221–229
225
Fig. 1 Correlation between log(Amp500 ) and visual acuity (logMAR) in group 1. The solid line represents the linear regression line fitting the data (y = 1.731 1.569x, p \ 0.0001)
Fig. 2 Correlation between log(Amp500 ) and visual acuity (logMAR) in group 2. The solid line represents the linear regression line fitting the data (y = 2.413 - 2.169x, p \ 0.0001)
The mean ages of the malingerers were as follows: macular disease malingering patients (group 1m), 52.47 ± 10.03 years, and optic nerve disease malingering patients (group 2m), 41.86 ± 12.41 years. By using the relations to determine the variance between the PVEP-VA and subjective VA in all patients (malingerers and non-malingerers), we determined that a variance value of 0.4041 (logMAR) resulted in 100 % sensitivity and 94.0 % specificity in group 1 and
a variance value of 0.3658 in 70.6 % sensitivity and 50.5 % specificity in group 2 to identify malingering in the ROC curve analysis (Tables 4, 5; Figs. 3, 4).
Discussion There is a long tradition of using the VEPs to assess visual acuity [26–31]. However, until now, there has
123
226 Table 4 Diagnostic validity (sensitivity and specificity) at each cutoff point for variance between the pattern visual evoked potential-estimated visual acuity and subjective visual acuity in patients with macular diseases
The most useful cutoff points were found at 100 % sensitivity and 94.0 % specificity when using a variance cutoff value of 0.4041
Table 5 Diagnostic validity (sensitivity and specificity) at each cutoff point for variance between the pattern visual evoked potential-estimated visual acuity and subjective visual acuity in patients with optic nerve diseases
The most useful cutoff points were found to be at 70.6 % sensitivity and 50.5 % specificity when using a variance cutoff value of 0.3658
Doc Ophthalmol (2015) 130:221–229
Range of variance (logMAR)
Sensitivity
1 - Specificity
Specificity
.3857
1.000
.119
.881
.3861
1.000
.112
.888
.3898
1.000
.097
.903
.3953
1.000
.082
.918
.4001
1.000
.075
.925
.4028
1.000
.067
.933
.4041
1.000
.060
.940
.4082 .4208
.923 .923
.060 .052
.940 .948
.4427
.846
.052
.948
.4591
.846
.045
.955
.4770
.846
.037
.963
.5014
.846
.030
.970
.5232
.769
.030
.970
Range of variance (logMAR)
Sensitivity
1 - Specificity
Specificity
0.2888
0.765
0.615
.385
0.2926
0.765
0.604
.396
0.3072
0.765
0.593
.407
0.3214
0.706
0.571
.429
0.3236
0.706
0.560
.440
0.3303
0.706
0.549
.451
0.3382
0.706
0.538
.462
0.3454 0.3540
0.706 0.706
0.527 0.516
.473 .484
0.3613
0.706
0.505
.495
0.3658
0.706
0.495
.505
0.3687
0.647
0.495
.505
0.3736
0.647
0.484
.516
0.3775
0.647
0.473
.527
0.3821
0.647
0.451
.549
been no firm theoretical link between visual acuity and the VEP recordings. Estimating visual acuity by the VEP examination is difficult because of fluctuations in accommodation and gaze direction, and a pronounced variability between subjects, both with respect to the absolute amplitude and to the shape of the tuning curve [5]. Iyer et al. [32] found that intermediate checkerboard sizes might be able to efficiently and robustly estimate visual acuity. They found that checkerboard size of 60, 32, 23 and 12 min of arc had the highest
123
correlation with extrapolated visual acuity using the sweep VEP and could be conducted without compromising accuracy to estimate visual acuity. In our study, though we used the pattern VEP, we also found that intermediate checkerboard sizes, especially 50 min of arc, had better correlation with visual acuity. The better correlation of VEPs produced by 500 checkerboard sizes and VA might have resulted from the fact that patients being studied had underlying maculopathy and optic neuropathy, whose visual acuity might
Doc Ophthalmol (2015) 130:221–229
Fig. 3 Receiver operating characteristic (ROC) curve to determine the diagnostic validity (sensitivity and specificity) for variance between the pattern visual evoked potentialestimated visual acuity and subjective visual acuity in patients with macular diseases. Area under curve: 0.987, p \ 0.0001
be mediated by portions of the visual field eccentric to the center, leading to variable and small VEPs produced by smaller checkerboard sizes. Jeon et al. [20] assessed the use of the pattern VEP recordings to predict objective VA in 38 eyes from normal or
227
amblyopic subjects and in 19 eyes with optic neuritis. They found that there was a significant correlation between the Amp500 and visual acuity in both normal and amblyopic eyes (r = -0.7649, p \ 0.0001) and in eyes with optic neuritis (r = -0.762, p = 0.0002), but there was no significant correlation between latency period and VA. In our study, we also found that the Amp500 was significantly correlated with VA in every group, rather than the latency period. Jeon et al. [20] performed linear regression analyses and determined relations of y = -0.072x ? 1.22 in normal or amblyopic eyes and y = -0.108x ? 1.55 in eyes with optic neuritis [x: Amp500 (lV), y: VA (logMAR)]. In the literature, Odom et al. [21] compared the subjective VA and the pattern VEP data of adults. Their linear regression analyses also gave good agreement between the subjective VA and the pattern VEP results. However, in our study, where a relatively large number of cases were included, we found a logarithmic curve best fit the correlation between Amp500 and VA in the regression analyses. Patients with visual disturbances may exaggerate the decrease in VA of the injured eye to maximize compensation. Such patients are particularly difficult to approach; therefore, an objective measurement of VA will provide an important contribution to the evaluation of such cases. Electrophysiological testing
Fig. 4 Receiver operating characteristic (ROC) curve to determine the diagnostic validity (sensitivity and specificity) for variance between the pattern visual evoked potential-estimated visual acuity and subjective visual acuity in patients with optic nerve diseases. Area under curve: 0.668, p = 0.029
123
228
can be used to evaluate the level of underlying organic dysfunction in patients with a non-organic overlay superimposed upon real dysfunction [33–36]. The pattern VEP may be a useful tool for determining the discrepancy in objective VA [34]. Rover and Bach successfully used a simultaneous recording of a pattern electroretinogram and the pattern VEP to reveal malingering [37]. Gundogan et al. [35] found that the PVEP-VA was well correlated with VAase (r = 0.670), and the sensitivity and specificity of the pattern VEP to diagnose malingering were 97.2 and 62.5 %, respectively. In our study, if we used a logMAR VA value of 0.4041 variance in the relation (y = 1.731 - 1.569x) in the macular diseases group, the sensitivity and specificity of the pattern VEP to diagnose malingering were 100 and 94.0 %, respectively. This means that if a subject is a malingerer, the pattern VEP is very useful in diagnosing it (100 % sensitivity), but there is a risk of misdiagnosing 6.0 % of non-malingers as malingerers. Similarly, in the optic nerve diseases group, a logMAR VA value of 0.3658 in the relation (y = 2.413 - 2.169x) was useful in diagnosing real malingerers with 70.6 % sensitivity; however, there is a relatively high risk of misdiagnosing 49.5 % of non-malingerers as malingerers. As a result, though the pattern VEP is a useful tool to help discriminate malingering, other adjunctive tests are required for confirmation, especially for patients with optic nerve diseases. The major limitations of our study were that it was a retrospective review of medical records and the variable nature of the VEP results between subjects, though we had to consider confounding factors such as age and the subject’s cooperation during the examination. Secondly, standard Landolt C does not really provide as precise visual acuity as provided by ETDRS, particularly for moderately to severely reduced visual acuity and converting it to logMAR could not moderate such weakness. Correlations with the VA would be better if visual acuity had been measured with ETDRS chart. Additionally, results were specific to the cases being studied and had limited generalizability because of the many exclusion factors in selecting the 249 patients and because of the limited numbers of underlying pathology in the patients with malingerers and the non-inclusion of malingerers with no organic dysfunction. In summary, there was a significant correlation between the pattern VEP amplitude of 500
123
Doc Ophthalmol (2015) 130:221–229
checkerboard size (Amp500 ) and VA. Estimation of VA through the correlation of Amp500 in the pattern VEP should be a useful reference to identify malingering patients. Further studies are required to assess reliability and to perform a statistical verification of the efficacy of our methods in the assessment of VA by the pattern VEP. Acknowledgments The authors have no financial or proprietary interest in any material or method mentioned. No financial support was provided for this work. Conflict of interest
None.
References 1. Katsumi O, Denno S, Arai M, De Faria JL, Hirose T (1997) Comparison of preferential looking acuity and pattern reversal visual evoked response acuity in pediatric patients. Graefes Arch Clin Exp Ophthalmol 235:684–690 2. Norcia A, Tyler C (1885) Spatial frequency sweep VEP: visual acuity during the first year of life. Vis Res 25:1399–1408 3. Ghasia F, Brunstom J, Tychsen L (2009) Visual acuity and visually evoked responses in children with cerebral palsy: gross motor function classification scale. Br J Ophthalmol 93:1068–1072 4. Strasburger H, Remky A, Murray IJ, Hadjizenonos C, Rentschler I (1996) Objective measurement of contrast sensitivity and visual acuity with the steady-state visual evoked potential. Ger J Ophthalmol 5:42–52 5. Bach M, Maurer JP, Wolf ME (2008) Visual evoked potential-based acuity assessment in normal vision, artificially degraded vision, and in patients. Br J Ophthalmol 92:396–403 6. Johansson B, Jakobsson P (2000) Fourier analysis of steadystate visual evoked potentials in subjects with normal and defective stereo vision. Doc Ophthalmol 101:233–246 7. Towle VL, Harter MR (1979) Objective determination of human visual acuity from the visual evoked potential. Percept Psychophys 25:497–500 8. Chan H, Odom JV, Coldren J, Dove C, Chao GM (1986) Acuity estimated by visually evoked potentials is affected by scaling. Doc Ophthalmol 62:107–117 9. Harter MR, Towle VL, Zakrzewski M, Moyer SM (1977) An objective indicant of binocular vision in humans: sizespecific interocular suppression of visual evoked potentials. Electroencephalogr Clin Neurophysiol 43:825–836 10. Heine S, Ruther K, Isensee J, Zrenner E (1999) Clinical significance of objective vision assessment using visually evoked cortical potentials induced by rapid pattern sequences of different spatial frequency. Klin Monbl Augenheilkd 215:175–181 11. Odom JV, Hoyt CS, Marg E (1981) Effect of natural deprivation and unilateral eye patching on visual acuity of infants and children. Evoked potential measurements. Arch Ophthalmol 99:1412–1416
Doc Ophthalmol (2015) 130:221–229 12. Sokol S (1978) Measurement of infant visual acuity from pattern reversal evoked potentials. Vis Res 18:33–39 13. Kurtenbach A, Langrova´ H, Messias A, Zrenner E, Ja¨gle H (2013) A comparison of the performance of three visual evoked potential-based methods to estimate visual acuity. Doc Ophthalmol 126:45–56 14. Sokol S (1976) Visually evoked potentials: theory, techniques and clinical applications. Surv Ophthalmol 21:18–44 15. Weinstein GW (1978) Clinical aspects of the visually evoked potential. Ophthalmic Surg 9:56–65 16. Howe JW, Mitchell KW, Robson C (1981) Electrophysiological assessment of visual acuity. Trans Ophthalmol Soc UK 101:105–108 17. Raniel Y, Pratt H, Neumann E, Schacham SE (1989) Miniature fiber-optic pattern-reversal stimulator for determination of the visual evoked potential threshold; comparison with Snellen acuity. Graefes Arch Clin Exp Ophthalmol 227:212–215 18. Steele M, Seiple WH, Carr RE, Klug R (1989) The clinical utility of visual-evoked potential acuity testing. Am J Ophthalmol 108:572–577 19. Emmerson-Hanover R, Shearer DE, Creel DJ, Dustman RE (1994) Pattern reversal evoked potentials: gender differences and age-related changes in amplitude and latency. Electroencephalogr Clin Neurophysiol 92:93–101 20. Jeon J, Oh S, Kyung S (2012) Assessment of visual disability using visual evoked potentials. BMC Ophthalmol 12:36 21. Odom JV, Maida TM, Dawson WW, Romano PE (1983) Retinal and cortical pattern responses: a comparison of infants and adults. Am J Optom Physiol Opt 60:369–375 22. Kramer KK, La Piana FG, Appleton B (1979) Ocular malingering and hysteria: diagnosis and management. Surv Ophthalmol 24:89–96 23. Slavin ML (1990) The prism dissociation test in detecting unilateral functional visual loss. J Clin Neuroophthalmol 10:127–130 24. Nuzzi R, Chiado Piat L (1994) Outpatient tests for visual acuity evaluation in malingerers: a review and personal experience. Ann Ophthalmol 26:175–182 25. Golnik KC, Lee AG, Eggenberger ER (2004) The monocular vertical prism dissociation test. Am J Ophthalmol 137:135–137
229 26. Cobb WA, Ettlinger G, Morton HB (1968) Cerebral potentials evoked in man by pattern reversal and their suppression in visual rivalry. J Physiol 195:33–34 27. Harter MR, White CT (1970) Evoked cortical responses to checkerboard patterns: effect of check-size as a function of visual acuity. Electroencephalogr Clin Neurophysiol 28:48–54 28. Arden GB, Lewis DR (1973) The pattern visual evoked response in the assessment of visual acuity. Trans Ophthalmol Soc UK 93:39–48 29. Regan D (1977) Rapid methods for refracting the eye and assessing the visual acuity in amblyopia using steady-state visual evoked potentials. In: Desmedt J (ed) Visual evoked potentials in man: new developments. Clarendon Press, Oxford, pp 418–426 30. Regan D (1978) Assessment of visual acuity by evoked potential recording: ambiguity caused by temporal dependence of spatial frequency selectivity. Vis Res 18:439–443 31. Teping C (1981) Determination of visual acuity by the visually evoked cortical potential (author’s transl). Klin Monbl Augenheilkd 179:169–172 32. Iyer KK, Bradley AP, Wilson SJ (2013) Conducting shorter VEP tests to estimate visual acuity via assessment of SNR. Doc Ophthalmol 126:21–28 33. Chen CS, Lee AW, Karagiannis A, Crompton JL, Selva D (2007) Practical clinical approaches to functional visual loss. J Clin Neurosci 14:1–7 34. McBain VA, Robson AG, Hogg CR, Holder GE (2007) Assessment of patients with suspected non-organic visual loss using pattern appearance visual evoked potentials. Graefes Arch Clin Exp Ophthalmol 245:502–510 35. Gundogan FC, Sobaci G, Bayer A (2007) Pattern visual evoked potentials in the assessment of visual acuity in malingering. Ophthalmology 114:2332–2337 36. Nakamura A, Akio T, Matsuda E, Wakami Y (2001) Pattern visual evoked potentials in malingering. J Neuroophthalmol 21:42–45 37. Rover J, Bach M (1987) Pattern electroretinogram plus visual evoked potential: a decisive test in subjects suspected of malingering. Doc Ophthalmol 66:245–251
123
ORIGINAL RESEARCH ARTICLE
Pattern visual evoked potentials for identifying malingering I-Ting Sun • Jong-Jer Lee • Hsiu-Mei Huang Hsi-Kung Kuo
•
Received: 21 October 2014 / Accepted: 21 January 2015 / Published online: 25 January 2015 Ó Springer-Verlag Berlin Heidelberg 2015
Abstract Purpose To investigate the efficacy of pattern visual evoked potentials (VEPs) in evaluating objective visual acuity (VA) and discriminating malingerers. Methods Two hundred and forty-nine eyes of 249 patients aged 20–65 years were included. There were 147 eyes with macular diseases (group 1) and 102 eyes with optic nerve diseases (group 2). Amplitudes and latencies were analyzed and correlated with bestcorrected visual acuity by a regression analysis. We found the best-correlated mode of pattern VEP, determined the relations, and then calculated the pattern VEP-estimated VA (PVEP-VA) of all 249 eyes, another 30 malingering eyes, 13 eyes with macular diseases, and 17 eyes with optic nerve diseases, and used a receiver operating characteristic (ROC) curve to determine a cutoff for acceptable variance between PVEP-VA and subjective VA to discriminate malingerers. Results The best correlation was between the amplitude of 500 checkerboard size (Amp500 ) and VA in every group. Significant correlation was between Amp500 and VA, where p \ 0.0001 in group 1 and p = 0.020 in group 2. A logarithmic curve best fitted
I.-T. Sun J.-J. Lee H.-M. Huang H.-K. Kuo (&) Department of Ophthalmology, Kaohsiung Chang Gung Memorial Hospital and Chang Gung University College of Medicine, 123 Dapi Road, Niao Sung District, Kaohsiung City, Taiwan e-mail: [email protected]
the correlation in the regression analysis, where y = 1.731 - 1.569x (R2 = 0.611, p \ 0.0001) in group 1 and y = 2.413 - 2.169x (R2 = 0.531, p \ 0.0001) in group 2 [x: log(Amp500 ), y: PVEPVA (logMAR)]. By using the relations and ROC curve, we determined a variance value of 0.4041 (logMAR) with 100 % sensitivity and 94.0 % specificity in group 1 and 0.3658 with 70.6 % sensitivity and 50.5 % specificity in group 2 to discriminate malingerers. Conclusions The pattern VEP amplitude of 500 checkerboard size was useful to assess VA and can be helpful in discriminating malingering from real disability. Keywords Pattern visual evoked potentials Visual acuity Objective assessment
Introduction The visual evoked potential (VEP) offers quantitative data on the function of the visual pathways and visual cortex. In recent years, the possibility of using the VEP to objectively assess the vision of patients has been examined. Objective visual acuity estimation by the VEP has been used in infants, children with cerebral visual impairment, and subjects suspected of nonorganic/functional visual loss, including malingering [1–5]. VEP-based visual acuity can be estimated from
123
222 Table 1 Diagnosis and the number of eyes in each group
Doc Ophthalmol (2015) 130:221–229
Diagnosis
Number of eyes
Group 1 Age-related macular disease
147
54.91 ± 10.16
113
59.60 ± 4.32
Macular pucker
13
39.32 ± 5.35
Choroidal neovascularization
11
43.44 ± 6.78
Diabetic maculopathy
10
40.22 ± 6.96
Group 2
102
45.44 ± 10.63
Optic neuritis
32
43.78 ± 9.86
Optic atrophy Traumatic optic neuropathy
22 26
45.30 ± 9.75 39.44 ± 8.64
Anterior inflammatory optic neuropathy
16
59.32 ± 4.89
6
43.79 ± 3.98
Toxic optic neuropathy
the sweep VEPs, steady-state VEPs, or pattern appearance VEPs. There are various ways to analyze the transient recordings; visual acuity can be expressed as the highest spatial frequency to which a response is recorded [6, 7] or by fitting a regression line or polynomial to amplitude response functions [8– 12]. Kurtenbach et al. [13] compared three methods of the VEP recordings to estimate visual acuity and found that the sweep VEP results were less variable and more reproducible than other methods, but require manual waveform detection and analysis. The pattern VEP is clinically attractive because of its ease of analysis and short recording time. The pattern VEP is generally used for the detection of visual disturbances, including optic neuritis, ischemic optic neuropathy, compressive optic neuropathy, demyelinization disease, and maculopathy [14, 15]. It has been used to assess objective visual acuity (VA) in previous studies, where only a limited number of cases were included [16–21]. However, there is little consensus regarding the correlation between the pattern VEP and VA, and the application for discriminating malingering. In this study with large number of cases, we investigated the efficacy of the pattern VEP to predict objective visual acuity and diagnose malingering.
Methods Participants With IRB approval, the medical records of 751 patients who received the pattern VEP examinations
123
Average age
between January 2008 and April 2013 at the same medical center were reviewed retrospectively. Patients who had anterior segment abnormalities and obvious brain lesions, those aged older than 65 or younger than 20 years, those with poor VEP waveforms, and those with glaucoma and normal and amblyopic eyes were excluded. If the patient had similar ocular diseases at both eyes, we selected the right eye. Finally, we selected 249 eyes of 249 patients and divided them into two groups as follows (Table 1): Group 1: 147 eyes with macular diseases, such as age-related macular disease (AMD), macular pucker, choroidal neovascularization (CNV), and diabetic maculopathy; and group 2: 102 eyes with optic nerve diseases, such as optic neuritis, optic atrophy, traumatic optic neuropathy, anterior ischemic optic neuropathy, and toxic optic neuropathy. Procedure All patients underwent biomicroscopic and dilated fundus examinations and underwent the VEP recordings. Additional examinations such as fluorescein angiography (FAG), indocyanine green angiography (ICG), optical coherence tomography (OCT), visual field tests, color vision tests, and magnetic resonance imaging (MRI) were performed if necessary to help diagnose macular and optic nerve diseases. The bestcorrected visual acuities (BCVAs) were recorded in all patients with a Landolt C chart just before VEP examinations. The Landolt C chart VA was converted to logMAR VA for the statistical analysis. The transient pattern VEP was recorded using an LKC UTAS-E 3000 system (LKC Technologies Inc.,
Doc Ophthalmol (2015) 130:221–229
Gaithersburg, MD, USA) in accordance with the International Society for Clinical Electrophysiology of Vision (ISCEV) recommendations. The patients’ refractions were corrected with trial lenses using a Landolt C chart before the VEP was examined. All patients were instructed to maintain fixation at the center of a stimulus located at a distance of 1 m in front of a 20 9 30 cm black-and-white video display monitor (contrast 99 %). The reversal rate was 1 reversal per second. The checkerboard stimulus subtended a visual angle of 5.7° vertically and 8.5° horizontally on either side of the fixation. The P100 amplitude and latency for five different checkerboard sizes (1000 , 500 , 250 , 120 , and 60 min of arc) were recorded. The fixation stability of the eyes was closely monitored by an experienced electrophysiology technician. If the cooperation of the patient was poor, the VEP examination was repeated. All the VEP examinations were performed under the same laboratory conditions. Data analysis We recorded the P100 amplitude and latency for each checkerboard size (1000 , 500 , 250 , 120 , and 60 min of arc) in all patients. P100 amplitude and the latency of each checkerboard size were correlated with BCVA by multiple regression analyses using IBM SPSS Statistics 18 software. We performed a linear regression analysis first to investigate the confounding effect of age on the pattern VEP latency and amplitude parameters among the enrolled patients (aged 20–65 years) and found that there was no significant interaction between the VEP parameters and age groups. Therefore, we did not perform statistical adjustments to age match the results of the enrolled patients. In order to evaluate the diagnostic validity of the test in distinguishing malingering patients, we also Table 2 Diagnosis and number of malingering eyes
223
enrolled 30 malingering eyes of 30 malingerers and compared the pattern VEP-estimated VA (PVEP-VA, logMAR) of all 249 eyes and those 30 malingering eyes. Then, we made a receiver operating characteristic (ROC) curve to determine a cutoff value for acceptable variance between PVEP-VA and subjective visual acuity to discriminate malingerers. The malingerers were all exaggerators who had simply macular disease or optic nerve disease without other clinically detectable combined ophthalmic disorders and whose acuity was worse than expected, where 13 eyes with macular diseases (group 1m) and 17 eyes with optic nerve diseases (group 2m) (Table 2). The patients whose BCVA was incompatible with ocular anatomic status were suspected malingerers. The malingering eyes were diagnosed clinically if a discrepancy in the initially presented BCVA (VAi) and BCVA after the simulation examination (VAase) was greater than 1 line on the Landolt C chart. The VAase was assessed after certain technical procedures and behavior observations. Sensitive clinical tests, including fogging, prism dissociation test, fixation techniques, distance test, grimacing in front of the patient, handshaking test, and pupillary light reflexes, were used for the detection of malingering [22–25]. VAi and VAase were checked by an experienced clinician. Briefly, the stepwise algorithm of our study can be more intuitively understood using the following rule set: 1.
2.
3.
Find whether the P100 amplitude or latency of the pattern VEP was best correlated with subjective visual acuity. Use the best-correlated mode of the pattern VEP results to perform linear analyses to determine a relation. Enroll malingering patients and use the above relation to calculate the PVEP-VA.
Diagnosis
Number of eyes
Group 1m: Macular diseases
13
52.47 ± 10.03
7
59.89 ± 4.98
6
43.81 ± 9.89
Age-related macular disease Diabetic maculopathy Group 2m: Optic nerve diseases Optic neuritis
Average age
17
41.86 ± 12.41
12
41.92 ± 11.96
Toxic optic neuropathy
4
44.65 ± 9.87
Traumatic optic neuropathy
1
30
123
224
4.
5.
Doc Ophthalmol (2015) 130:221–229
Compare the PVEP-VA with the subjective VA to investigate the variance in all patients (malingerers and non-malingerers). Perform a ROC analysis to obtain a cutoff value for acceptable variance, where sensitivity is more relevant than specificity.
where p \ 0.0001 in group 1 and p = 0.020 in group 2 (Table 3). The mean VEP Amp500 parameters for each group were 5.10 ± 1.95 lV in group 1 (macular diseases) and 5.89 ± 2.49 lV in group 2 (optic nerve diseases). Pearson’s correlation coefficient between Amp500 and VA was -0.737 (p \ 0.0001) in group 1 and -0.656 (p \ 0.0001) in group 2. We found that a logarithmic curve fits best the correlation and was better than a straight line or reciprocal curve in the regression analysis (figures of straight line or reciprocal curve were not showed). The relations were y = 1.731 - 1.569x (R2 = 0.611, p \ 0.0001) in group 1 and y = 2.413 - 2.169x (R2 = 0.531, p \ 0.0001) in group 2 [x: log(Amp500 ), y: PVEP-VA (logMAR)] (Figs. 1, 2).
Results The demographic data of all patients are shown in Table 1. The mean ages were as follows: group 1, 54.91 ± 10.16 years; and group 2, 45.44 ± 10.63 years. Significant correlation was between the amplitude of 500 checkerboard size (Amp500 ) and VA in every group,
Table 3 Multiple linear regression models between the visual evoked potential data and visual acuity (logMAR) Model
Unstandardized coefficients
Standardized coefficients
B
Beta
t
Standard error
Significance
95 % Confidence interval for B Lower limit
Upper limit
Group 1 (macular diseases) Constant
.889
.391
2.275
.026
.110
-.003
.003
-.145
-1.048
.298
-.009
.003
Lat500
.000
.004
.020
.126
.900
-.007
.008
Lat250
.004
.003
.175
1.201
.234
-.003
.011
0
Lat1000
Lat12
1.668
.001
.003
.025
.243
.809
-.005
.006
Lat60
.001
.002
.004
.041
.968
-.004
.004
Amp1000
.006
.014
.042
.402
.689
-.022
.033
Amp500
-.117
.018
-.696
-6.361
.000
-.153
-.080
Amp250
-.006
.019
-.038
-.339
.736
-.044
.031
0
.001 .010
.012 .012
.012 .081
.121 .821
.904 .414
-.022 -.014
.025 .034
Amp12 Amp60
Group 2 (optic nerve diseases) Constant
1.055
.599
Lat1000
.001
.007
0
.010
.007
Lat250
-.002
.007
Lat120
.004
.004
-.010 .022
Amp500
-.074
Amp250 Amp120
Lat50
Lat60 Amp1000
Amp60
1.760
.087
-.160
2.269
.206
.838
-.013
.016
.347
1.331
.191
-.005
.025
-.082
-.307
.760
-.017
.012
.143
.869
.391
-.005
.013
.004
-.343
-1.759
.090
-.017
-.003
.028
.151
.796
.431
-.035
.079
.031
-.492
-2.433
.020
-.136
-.012
-.050
.028
-.349
-1.819
.080
-.137
-.022
.044
.026
.334
1.678
.102
-.009
.096
-.015
.016
-.130
-.981
.333
-.047
.016
Dependent variable: visual acuity (logMAR) Lat latency, Amp amplitude
123
.043
Doc Ophthalmol (2015) 130:221–229
225
Fig. 1 Correlation between log(Amp500 ) and visual acuity (logMAR) in group 1. The solid line represents the linear regression line fitting the data (y = 1.731 1.569x, p \ 0.0001)
Fig. 2 Correlation between log(Amp500 ) and visual acuity (logMAR) in group 2. The solid line represents the linear regression line fitting the data (y = 2.413 - 2.169x, p \ 0.0001)
The mean ages of the malingerers were as follows: macular disease malingering patients (group 1m), 52.47 ± 10.03 years, and optic nerve disease malingering patients (group 2m), 41.86 ± 12.41 years. By using the relations to determine the variance between the PVEP-VA and subjective VA in all patients (malingerers and non-malingerers), we determined that a variance value of 0.4041 (logMAR) resulted in 100 % sensitivity and 94.0 % specificity in group 1 and
a variance value of 0.3658 in 70.6 % sensitivity and 50.5 % specificity in group 2 to identify malingering in the ROC curve analysis (Tables 4, 5; Figs. 3, 4).
Discussion There is a long tradition of using the VEPs to assess visual acuity [26–31]. However, until now, there has
123
226 Table 4 Diagnostic validity (sensitivity and specificity) at each cutoff point for variance between the pattern visual evoked potential-estimated visual acuity and subjective visual acuity in patients with macular diseases
The most useful cutoff points were found at 100 % sensitivity and 94.0 % specificity when using a variance cutoff value of 0.4041
Table 5 Diagnostic validity (sensitivity and specificity) at each cutoff point for variance between the pattern visual evoked potential-estimated visual acuity and subjective visual acuity in patients with optic nerve diseases
The most useful cutoff points were found to be at 70.6 % sensitivity and 50.5 % specificity when using a variance cutoff value of 0.3658
Doc Ophthalmol (2015) 130:221–229
Range of variance (logMAR)
Sensitivity
1 - Specificity
Specificity
.3857
1.000
.119
.881
.3861
1.000
.112
.888
.3898
1.000
.097
.903
.3953
1.000
.082
.918
.4001
1.000
.075
.925
.4028
1.000
.067
.933
.4041
1.000
.060
.940
.4082 .4208
.923 .923
.060 .052
.940 .948
.4427
.846
.052
.948
.4591
.846
.045
.955
.4770
.846
.037
.963
.5014
.846
.030
.970
.5232
.769
.030
.970
Range of variance (logMAR)
Sensitivity
1 - Specificity
Specificity
0.2888
0.765
0.615
.385
0.2926
0.765
0.604
.396
0.3072
0.765
0.593
.407
0.3214
0.706
0.571
.429
0.3236
0.706
0.560
.440
0.3303
0.706
0.549
.451
0.3382
0.706
0.538
.462
0.3454 0.3540
0.706 0.706
0.527 0.516
.473 .484
0.3613
0.706
0.505
.495
0.3658
0.706
0.495
.505
0.3687
0.647
0.495
.505
0.3736
0.647
0.484
.516
0.3775
0.647
0.473
.527
0.3821
0.647
0.451
.549
been no firm theoretical link between visual acuity and the VEP recordings. Estimating visual acuity by the VEP examination is difficult because of fluctuations in accommodation and gaze direction, and a pronounced variability between subjects, both with respect to the absolute amplitude and to the shape of the tuning curve [5]. Iyer et al. [32] found that intermediate checkerboard sizes might be able to efficiently and robustly estimate visual acuity. They found that checkerboard size of 60, 32, 23 and 12 min of arc had the highest
123
correlation with extrapolated visual acuity using the sweep VEP and could be conducted without compromising accuracy to estimate visual acuity. In our study, though we used the pattern VEP, we also found that intermediate checkerboard sizes, especially 50 min of arc, had better correlation with visual acuity. The better correlation of VEPs produced by 500 checkerboard sizes and VA might have resulted from the fact that patients being studied had underlying maculopathy and optic neuropathy, whose visual acuity might
Doc Ophthalmol (2015) 130:221–229
Fig. 3 Receiver operating characteristic (ROC) curve to determine the diagnostic validity (sensitivity and specificity) for variance between the pattern visual evoked potentialestimated visual acuity and subjective visual acuity in patients with macular diseases. Area under curve: 0.987, p \ 0.0001
be mediated by portions of the visual field eccentric to the center, leading to variable and small VEPs produced by smaller checkerboard sizes. Jeon et al. [20] assessed the use of the pattern VEP recordings to predict objective VA in 38 eyes from normal or
227
amblyopic subjects and in 19 eyes with optic neuritis. They found that there was a significant correlation between the Amp500 and visual acuity in both normal and amblyopic eyes (r = -0.7649, p \ 0.0001) and in eyes with optic neuritis (r = -0.762, p = 0.0002), but there was no significant correlation between latency period and VA. In our study, we also found that the Amp500 was significantly correlated with VA in every group, rather than the latency period. Jeon et al. [20] performed linear regression analyses and determined relations of y = -0.072x ? 1.22 in normal or amblyopic eyes and y = -0.108x ? 1.55 in eyes with optic neuritis [x: Amp500 (lV), y: VA (logMAR)]. In the literature, Odom et al. [21] compared the subjective VA and the pattern VEP data of adults. Their linear regression analyses also gave good agreement between the subjective VA and the pattern VEP results. However, in our study, where a relatively large number of cases were included, we found a logarithmic curve best fit the correlation between Amp500 and VA in the regression analyses. Patients with visual disturbances may exaggerate the decrease in VA of the injured eye to maximize compensation. Such patients are particularly difficult to approach; therefore, an objective measurement of VA will provide an important contribution to the evaluation of such cases. Electrophysiological testing
Fig. 4 Receiver operating characteristic (ROC) curve to determine the diagnostic validity (sensitivity and specificity) for variance between the pattern visual evoked potential-estimated visual acuity and subjective visual acuity in patients with optic nerve diseases. Area under curve: 0.668, p = 0.029
123
228
can be used to evaluate the level of underlying organic dysfunction in patients with a non-organic overlay superimposed upon real dysfunction [33–36]. The pattern VEP may be a useful tool for determining the discrepancy in objective VA [34]. Rover and Bach successfully used a simultaneous recording of a pattern electroretinogram and the pattern VEP to reveal malingering [37]. Gundogan et al. [35] found that the PVEP-VA was well correlated with VAase (r = 0.670), and the sensitivity and specificity of the pattern VEP to diagnose malingering were 97.2 and 62.5 %, respectively. In our study, if we used a logMAR VA value of 0.4041 variance in the relation (y = 1.731 - 1.569x) in the macular diseases group, the sensitivity and specificity of the pattern VEP to diagnose malingering were 100 and 94.0 %, respectively. This means that if a subject is a malingerer, the pattern VEP is very useful in diagnosing it (100 % sensitivity), but there is a risk of misdiagnosing 6.0 % of non-malingers as malingerers. Similarly, in the optic nerve diseases group, a logMAR VA value of 0.3658 in the relation (y = 2.413 - 2.169x) was useful in diagnosing real malingerers with 70.6 % sensitivity; however, there is a relatively high risk of misdiagnosing 49.5 % of non-malingerers as malingerers. As a result, though the pattern VEP is a useful tool to help discriminate malingering, other adjunctive tests are required for confirmation, especially for patients with optic nerve diseases. The major limitations of our study were that it was a retrospective review of medical records and the variable nature of the VEP results between subjects, though we had to consider confounding factors such as age and the subject’s cooperation during the examination. Secondly, standard Landolt C does not really provide as precise visual acuity as provided by ETDRS, particularly for moderately to severely reduced visual acuity and converting it to logMAR could not moderate such weakness. Correlations with the VA would be better if visual acuity had been measured with ETDRS chart. Additionally, results were specific to the cases being studied and had limited generalizability because of the many exclusion factors in selecting the 249 patients and because of the limited numbers of underlying pathology in the patients with malingerers and the non-inclusion of malingerers with no organic dysfunction. In summary, there was a significant correlation between the pattern VEP amplitude of 500
123
Doc Ophthalmol (2015) 130:221–229
checkerboard size (Amp500 ) and VA. Estimation of VA through the correlation of Amp500 in the pattern VEP should be a useful reference to identify malingering patients. Further studies are required to assess reliability and to perform a statistical verification of the efficacy of our methods in the assessment of VA by the pattern VEP. Acknowledgments The authors have no financial or proprietary interest in any material or method mentioned. No financial support was provided for this work. Conflict of interest
None.
References 1. Katsumi O, Denno S, Arai M, De Faria JL, Hirose T (1997) Comparison of preferential looking acuity and pattern reversal visual evoked response acuity in pediatric patients. Graefes Arch Clin Exp Ophthalmol 235:684–690 2. Norcia A, Tyler C (1885) Spatial frequency sweep VEP: visual acuity during the first year of life. Vis Res 25:1399–1408 3. Ghasia F, Brunstom J, Tychsen L (2009) Visual acuity and visually evoked responses in children with cerebral palsy: gross motor function classification scale. Br J Ophthalmol 93:1068–1072 4. Strasburger H, Remky A, Murray IJ, Hadjizenonos C, Rentschler I (1996) Objective measurement of contrast sensitivity and visual acuity with the steady-state visual evoked potential. Ger J Ophthalmol 5:42–52 5. Bach M, Maurer JP, Wolf ME (2008) Visual evoked potential-based acuity assessment in normal vision, artificially degraded vision, and in patients. Br J Ophthalmol 92:396–403 6. Johansson B, Jakobsson P (2000) Fourier analysis of steadystate visual evoked potentials in subjects with normal and defective stereo vision. Doc Ophthalmol 101:233–246 7. Towle VL, Harter MR (1979) Objective determination of human visual acuity from the visual evoked potential. Percept Psychophys 25:497–500 8. Chan H, Odom JV, Coldren J, Dove C, Chao GM (1986) Acuity estimated by visually evoked potentials is affected by scaling. Doc Ophthalmol 62:107–117 9. Harter MR, Towle VL, Zakrzewski M, Moyer SM (1977) An objective indicant of binocular vision in humans: sizespecific interocular suppression of visual evoked potentials. Electroencephalogr Clin Neurophysiol 43:825–836 10. Heine S, Ruther K, Isensee J, Zrenner E (1999) Clinical significance of objective vision assessment using visually evoked cortical potentials induced by rapid pattern sequences of different spatial frequency. Klin Monbl Augenheilkd 215:175–181 11. Odom JV, Hoyt CS, Marg E (1981) Effect of natural deprivation and unilateral eye patching on visual acuity of infants and children. Evoked potential measurements. Arch Ophthalmol 99:1412–1416
Doc Ophthalmol (2015) 130:221–229 12. Sokol S (1978) Measurement of infant visual acuity from pattern reversal evoked potentials. Vis Res 18:33–39 13. Kurtenbach A, Langrova´ H, Messias A, Zrenner E, Ja¨gle H (2013) A comparison of the performance of three visual evoked potential-based methods to estimate visual acuity. Doc Ophthalmol 126:45–56 14. Sokol S (1976) Visually evoked potentials: theory, techniques and clinical applications. Surv Ophthalmol 21:18–44 15. Weinstein GW (1978) Clinical aspects of the visually evoked potential. Ophthalmic Surg 9:56–65 16. Howe JW, Mitchell KW, Robson C (1981) Electrophysiological assessment of visual acuity. Trans Ophthalmol Soc UK 101:105–108 17. Raniel Y, Pratt H, Neumann E, Schacham SE (1989) Miniature fiber-optic pattern-reversal stimulator for determination of the visual evoked potential threshold; comparison with Snellen acuity. Graefes Arch Clin Exp Ophthalmol 227:212–215 18. Steele M, Seiple WH, Carr RE, Klug R (1989) The clinical utility of visual-evoked potential acuity testing. Am J Ophthalmol 108:572–577 19. Emmerson-Hanover R, Shearer DE, Creel DJ, Dustman RE (1994) Pattern reversal evoked potentials: gender differences and age-related changes in amplitude and latency. Electroencephalogr Clin Neurophysiol 92:93–101 20. Jeon J, Oh S, Kyung S (2012) Assessment of visual disability using visual evoked potentials. BMC Ophthalmol 12:36 21. Odom JV, Maida TM, Dawson WW, Romano PE (1983) Retinal and cortical pattern responses: a comparison of infants and adults. Am J Optom Physiol Opt 60:369–375 22. Kramer KK, La Piana FG, Appleton B (1979) Ocular malingering and hysteria: diagnosis and management. Surv Ophthalmol 24:89–96 23. Slavin ML (1990) The prism dissociation test in detecting unilateral functional visual loss. J Clin Neuroophthalmol 10:127–130 24. Nuzzi R, Chiado Piat L (1994) Outpatient tests for visual acuity evaluation in malingerers: a review and personal experience. Ann Ophthalmol 26:175–182 25. Golnik KC, Lee AG, Eggenberger ER (2004) The monocular vertical prism dissociation test. Am J Ophthalmol 137:135–137
229 26. Cobb WA, Ettlinger G, Morton HB (1968) Cerebral potentials evoked in man by pattern reversal and their suppression in visual rivalry. J Physiol 195:33–34 27. Harter MR, White CT (1970) Evoked cortical responses to checkerboard patterns: effect of check-size as a function of visual acuity. Electroencephalogr Clin Neurophysiol 28:48–54 28. Arden GB, Lewis DR (1973) The pattern visual evoked response in the assessment of visual acuity. Trans Ophthalmol Soc UK 93:39–48 29. Regan D (1977) Rapid methods for refracting the eye and assessing the visual acuity in amblyopia using steady-state visual evoked potentials. In: Desmedt J (ed) Visual evoked potentials in man: new developments. Clarendon Press, Oxford, pp 418–426 30. Regan D (1978) Assessment of visual acuity by evoked potential recording: ambiguity caused by temporal dependence of spatial frequency selectivity. Vis Res 18:439–443 31. Teping C (1981) Determination of visual acuity by the visually evoked cortical potential (author’s transl). Klin Monbl Augenheilkd 179:169–172 32. Iyer KK, Bradley AP, Wilson SJ (2013) Conducting shorter VEP tests to estimate visual acuity via assessment of SNR. Doc Ophthalmol 126:21–28 33. Chen CS, Lee AW, Karagiannis A, Crompton JL, Selva D (2007) Practical clinical approaches to functional visual loss. J Clin Neurosci 14:1–7 34. McBain VA, Robson AG, Hogg CR, Holder GE (2007) Assessment of patients with suspected non-organic visual loss using pattern appearance visual evoked potentials. Graefes Arch Clin Exp Ophthalmol 245:502–510 35. Gundogan FC, Sobaci G, Bayer A (2007) Pattern visual evoked potentials in the assessment of visual acuity in malingering. Ophthalmology 114:2332–2337 36. Nakamura A, Akio T, Matsuda E, Wakami Y (2001) Pattern visual evoked potentials in malingering. J Neuroophthalmol 21:42–45 37. Rover J, Bach M (1987) Pattern electroretinogram plus visual evoked potential: a decisive test in subjects suspected of malingering. Doc Ophthalmol 66:245–251
123
