Documenta Ophthalmologica 81: 181-188, 1992. 9 1992 Kluwer Academic Publishers. Printed in the Netherlands.

Comparison of preoperative lO-Hz visual evoked potentials to contrast sensitivity and visual acuity after cataract extraction S H E R I A . C A V E N D E R , R O B E R T R. H O B S O N , G U N G - M E I G E O R G E W. W E I N S T E I N & J. V E R N O N O D O M Department of Ophthalmology, West Virginia University Health Sciences Center, Morgantown, WV 26506, USA

CHAO,

Accepted 3 March 1992 Key words: Cataract, contrast sensitivity, media opacity, visual acuity, visual evoked potential,

visual function Abstract. Cataract patients whose surgical outcomes were in question were referred for testing

by visual evoked potentials, elicited through closed eyelids by a luminance stimulus (flash) that appeared 10 times per second. Visual evoked potentials were rated as normal (predicted acuity of 20/50 or better) or abnormal (predicted acuity of 20/60 or worse). Postoperative Arden and Optronix contrast sensitivities and visual acuities were determined in 37 patients who had no intraoperative or early postoperative complications. Arden grating scores of less than 100 were rated as normal. The optimal and cutoff spatial frequency values were determined for the Optronix scores. Optimal and cutoff values of greater or equal to 1 c/deg and 12c/deg, respectively, were rated as normal. Visual acuities were considered normal at 20/50 or better. Preoperative visual evoked potentials were quantitatively compared to the postoperative contrast sensitivities and visual acuities by 2 • 2 contingency tables. The accuracy of prediction was 79% for the visual acuities, 62% for the Optronix optimal values, 70% for the Optronix cutoff values and 62% for the Arden gratings.

Introduction

C o n t r a s t sensitivity is a m o r e c o m p l e t e m e a s u r e o f visual f u n c t i o n t h a n S n e l l e n visual a c u i t y [1] a n d , as such, is o f t e n r e p o r t e d to b e a m o r e s e n s i t i v e i n d i c a t o r o f visual i m p a i r m e n t t h a n visual acuity [2]. C e r t a i n l y , c o n t r a s t sensitivity a n d r e c o g n i t i o n visual acuity, as m e a s u r e d with S n e l l e n l e t t e r s , m e a s u r e s e p a r a b l e visual functions. P a t i e n t s with n o r m a l S n e l l e n v i s u a l a c u i t y d o n o t always possess n o r m a l c o n t r a s t sensitivity, a n d p a t i e n t s w i t h n o r m a l c o n t r a s t sensitivity m a y h a v e i m p a i r e d S n e l l e n visual acuity [3]. A l m o s t a n y c h a n g e in t h e visual system, f r o m c o r n e a to o p t i c n e r v e , will r e s u l t in a c h a n g e in c o n t r a s t sensitivity. V i s u a l e v o k e d p o t e n t i a l s ( V E P s ) h a v e p r o v e d a useful t o o l in p r e d i c t i n g visual a c u i t y a f t e r c a t a r a c t e x t r a c t i o n . In o u r l a b o r a t o r y , t h e p r e d i c t i o n a c c u r a c y has r a n g e d f r o m 7 0 % to 8 0 % [4-9]. I n t h e p r e s e n t s t u d y w e s o u g h t t o a n s w e r two q u e s t i o n s : If V E P s p r e d i c t p o s t o p e r a t i v e visual a c u i t y accu-

182 rately, can they also predict contrast sensitivity? If so, would they predict visual acuity or contrast sensitivity more accurately?

Subjects and methods Subjects included in the study were referred to the West Virginia University Retinal Function Laboratory, Morgantown, during the period from 1981 through 1983 because of a question of postoperative visual acuity outcome. All subjects were patients with cataracts whose retinas could not be examined because of dense cataracts or who had a retinal disease (or suspected retinal disease) in the affected or fellow eye. Thirty-eight patients participated in the study. They were aged 57 through 91 years, with a mean age of 75.3 years. Only patients who had no operative or postoperative complications were included. Preoperative VEPs were performed, and preoperative and postoperative best corrected visual acuities were measured. Thirty-six patients completed postoperative contrast sensitivity testing on both the Arden plates and the Optronix Series 200; one patient completed only the Arden plates and one only the Optronix test. Although VEP, acuity and contrast sensitivity data were collected on both eyes, only the information obtained from the eye operated on are presented in this study. The VEPs were elicited by a 10-Hz flash produced by a Grass PS-22 photic stimulator at an intensity setting of 1-2. Patients were placed 0.5 m from the flash source in a brightly lit room. The active electrode was placed 1 cm above the inion, and earclip reference and ground electrodes were used. Testing was monocular, while the nontested eye was covered by an opaque black patch. The right eye was always tested first, and the patients kept their eyes closed during testing. A Nicolet 1170 signal averager was used to record the cortical signals. The signal was amplified 10 000 times, with a bandwidth of 0.1 to 100 Hz. Four responses were recorded during recording epochs of 400 msec. Responses were a summation of 128 sweeps. Waveforms were assessed as normal or abnormal on the basis of criteria proposed by Weinstein [4]. A response had to contain both a primary (smaller) and secondary (larger) response for it to be considered normal. An example of a normal 10-Hz VEP and an abnormal VEP is shown in Fig. 1. The normal VEP (top waveform) shows the secondary (larger) response of at least 14 txV in amplitude, the presence of a primary (smaller) response and a normal latency. An abnormal response is pictured below the normal one. Note the reduced amplitude of the response when compared to the normal VEP and the absence of the primary (smaller) response. If an abnormal response was obtained at flash intensity 2 (I-2), the intensity was increased to determine if the dense media was contributing to the abnormality. Intensities of 1-4 and 1-8 were employed to evaluate an abnormal response further. If a normal response was obtained at 1-4 or 1-8, the waveform was considered normal.

183

3.12 W

A

^

h

A

I

Fig. 1. Two VEPs elicited by 10-Hz flicker. The upper VEP is normal, from a subject whose

postoperative visual acuity was 20/20. Note that there is a smaller primary component preceded by a larger, secondary component. The lower VEP is abnormal, from a patient whose postoperative visual acuity was 20/100. Note that there is no small, primary component; also, the large secondary component is reduced in amplitude.

A r d e n contrast sensitivities were performed with the six standard A r d e n test plates. The matte plates are a series of sine gratings, with each plate having an appearance of uniform gray at the top with the sine wave increasing in contrast toward the bottom. Plate 2 has the lowest spatial frequency, 0.2 c/deg, with each plate having a frequency two times that of the previous plate. Plate 7 has the highest spatial frequency, 6 . 4 c / d e g . Patients were positioned 57 cm from the plates, which resulted in a field size of 27.4 ~ in a well-illuminated r o o m of approximately 100 foot-candles. Testing was monocular with optical correction that permitted the patient's best near postoperative visual acuity. The right eye was tested first, and the nontested eye was covered with a white patch so as not to dark adapt that eye. T h e test was explained to the patient, and a demonstration plate was presented. The plates were slowly withdrawn from a pocket and scored at the point when the patient was first able to detect the sine wave. Each plate was scored on the standard scale of 1-20, and the scores from the six plates were s u m m e d to gain the final value. A value of less than 100 was d e t e r m i n e d as normal. We choose 100 as our criterion value, rather than the cutoff value of 82 used by Arden, on the basis of findings of an increase in test scores across all spatial frequencies with increasing age [3, 10]. Optronix contrast sensitivity was acquired with the Series 200 vision tester. The tester consists of a control console, a display console and an observer's box. The standard von Bekesy tracking test consisting of six separate trials, each of which presents a single static sinusoidal grating, was used. T w o demonstration gratings were presented to the patient before the start of the test. Sine wave gratings of 0.5, 1, 3, 6, 11.4 and 22.8 c / d e g were

184

presented on separate trials. The gratings started at zero contrast and increased until the patient was just able to detect the image. The image was then decreased in contrast until the patient was no longer able to see it. Four responses were averaged for each grating. Again, testing was monocular, starting with the right eye. Gratings were presented in a darkened room with the patient 3 m away from the video display screen, with the patient's best corrected visual acuity. The field size for the Optronix measurements was 4.34 ~ A mean and standard deviation were acquired for all contrast thresholds. The optimal and cutoff spatial frequencies were chosen as two values that could easily describe the contrast sensitivity function (Patrick Mulvanny, Optronix, Inc., personal communication). The criterion values of these two measures were selected on the basis of preliminary data that indicated that these values would optimally reflect VEP categories. The criterion optimal spatial frequency value was i c/deg or greater, and the criterion cutoff value was 12 c/deg or greater. The Optronix peak contrast sensitivity (1/threshold contrast) was determined for each patient, irrespective of the spatial frequency at which it occurred.

Results We compared the preoperative VEPs in 37 eyes to the postoperative Optronix and Arden contrast sensitivities and visual acuities. As shown in Table 1, separate 2 x 2 contingency tables were constructed to compare Table 1. Preoperative VEPs in relation to postoperative

findings Postoperative finding

Preoperative VEP (No.) Abnormal

Normal

Total

10 5 15

3 20 23

13 25 38

Arden contrast sensitivity Abnormal 13 Normal 1 Total 14

13 10 23

26 11 37

Optronix optimal spatial frequency Abnormal 3 Normal 12 Total 15

2 20 22

5 32 37

Optronix cutoff spatial frequency Abnormal 9 Normal 6 Total 15

5 17 22

14 23 37

Visual acuity Abnormal Normal Total

185 normal versus abnormal VEPs to normal versus abnormal Arden contrast sensitivity, Optronix cutoff and optimal spatial frequencies and visual acuity. Accuracy, prevalence, sensitivity, specificity, positive predictive values and negative predictive values were determined for each subgroup shown in Table 2. Table 2.

Findings from 2 x 2 contingency tables

VEP and visual acuity VEP and Arden contrast sensitivity VEP and Optronix optimal spatial frequency VEP and Optronix cutoff spatial frequency

n

Sensit- Spec- Positive Negative Prevalence Accuracy ivity ificity predictive predictive value value

38

0.77

0.80

0.67

0.86

0.67

0.79

37

0.50

0.91

0.93

0.43

0.70

0.62

37

0.60

0.66

0.20

0.91

0.13

0.62

37

0.64

0.74

0.60

0.77

0.38

0.70

Preoperative VEPs compared most closely to postoperative visual acuities, with an accuracy of 79% (X 2 = 9.34, p = 0.002). Normal postoperative visual acuity was 20/50 or better; abnormal was 20/60 or worse. The positive predictive value, which indicates the probability of a patient's having p o o r postoperative visual acuity if the VEP is abnormal, was 67%. The negative predictive value, which indicates the probability of a patient's having good postoperative visual acuity if the VEP is normal, was 86%. The single cumulative value for all spatial frequencies obtained from the A r d e n plates had an accuracy of 62% (X 2-- 3.9, p = 0.046). The positive predictive value was 93%, while the negative predictive value was 43%. The accuracy of the postoperative Optronix cutoff spatial frequency was 70% (X 2 = 3.8, p = 0.049). Positive predictive value was 60%; negative predictive value was 77%. Accuracy for the Optronix optimal spatial frequency in comparison to that of the VEP was 62% (X 2 = 0.215, p value was nonsignificant), with a positive predictive value of 20% and a negative predictive value of 91%. The V E P categories were correlated with peak contrast sensitivity for the Optronix and Arden plates to determine if VEPs could predict contrast sensitivity. The V E P categories were not correlated statistically significantly with either Optronix or Arden peak contrast sensitivity ( p > 0.10).

186 Discussion

We observed that 10-Hz flicker VEPs predicted postoperative visual acuity with an accuracy of 79%. Although the accuracy of VEPs in predicting the Optronix high-spatial-frequency cutoff frequency was relatively high (70%), the accuracy of VEPs in predicting Optronix optimal spatial frequency and Arden contrast sensitivity was much poorer (62% for each). Several studies have reported that 10-Hz VEPs are useful in assessing postoperative visual acuity in cases of cataract [4, 5], vitreal opacities [6, 7] and corneal opacities [8]. In general, these studies have reported an accuracy of predicting postoperative visual acuity that ranges between 70% and 80%. Our result, 79%, is within this general range. Our study showed that 10-Hz VEPs were not as good a predictor of postoperative contrast sensitivity as postoperative visual acuity. The best accuracy rate was obtained with the Optronix cutoff spatial frequency data, which has the highest spatial frequency (most closely related to visual acuity). Why were 10-Hz VEPs not able to predict postoperative contrast sensitivity as well as visual acuity? One possibility is that the technique used to evaluate the VEPS was originally designed by Weinstein [4] to correlate with visual acuity. Because visual acuity and contrast sensitivity measure different functions, the VEP criteria predict contrast sensitivity less well. The procedures used to measure contrast sensitivity are of additional concern. Visual acuity as measured clinically is a forced-choice measure; however, the Optronix and Arden gratings did not make use of a forcedchoice method for testing contrast sensitivity. The patients tested by Optronix were asked to signal by pressing a button when they could "just barely see" the gratings and again to press a button when they could no longer detect a pattern. Each patient may use a different criterion for deciding when a contrast pattern is visible, even if the instructions are delivered to each patient the same way every time a test is performed. An individual's perception of when the target is visible will vary, as will their rate of response [11]. The forced-choice method has been shown to be superior in test-retest stability and in reducing test variability [12]. Our decision to compare VEPs to composite contrast sensitivity measures may also have affected the results. Arden scores offered one cumulative value as a comparison of normal or abnormal contrast sensitivity; however, Optronix had six separate scores from which a contrast sensitivity function was obtained. The Optronix high-spatial-frequency cutoff proved to be the most accurate in the contrast sensitivity subgroups. The accuracy was the same between the Optronix optimal spatial frequency (criterion, i>1 c/deg) and the Arden cumulative score. In addition to the Optronix optimal and cutoff values and the Arden cumulative value used for comparison to the VEPs, we also looked at how peak Optronix and Arden contrast sensitivity compared to the VEPs. We found that peak contrast sensitivity, for either the Optronix or Arden plates, was not statistically significantly related to

187 VEP category. We believed this was because the comparisons of peak sensitivity occurred at widely separated spatial frequencies in different individuals. If specific spatial frequencies were used as a comparison, the VEPs might have predicted the sensitivities of one or more spatial frequencies with an accuracy that matched or even surpassed that of visual acuity. However, it is not clear how such a result would have been interpreted. It is known that the VEPs primarily reflect activity of the central 6~ to 12~ of the retina. A field size comparable to that of VEPs was used with the Optronix vision tester, while a field size larger than the area tested by the VEPs was obtained when the Arden gratings were used. This difference could make the comparison of VEPs and contrast sensitivity incompatible, since it is thought that contrast sensitivity measures more than macular function [13]. Many question how the flash VEPs can provide information about macular function. Because the macula is generously represented on a superficial aspect of the occipital cortex, the VEP primarily reflects visual function of the central retina. Since they are elicited by luminance changes, rather than pattern changes, flash VEPs are only indirectly related to visual acuity in that pattern and luminance changes rely on the same neural substrate of the macular region [9]. Flash VEPs are a useful predictor of visual acuity after removal of media opacities. This is especially true in cases of dense opacities [8]. They do not provide as much information on postoperative contrast sensitivity as they do on Snellen visual acuity. Their usefulness as a predictor of contrast sensitivity merits future investigation with a modification of testing procedures to include a forced-choice method of patient response and a strict control of the field size of targets.

References 1. Mannis MJ. Making sense of contrast sensitivity testing. Has its time come? Arch Ophthalmol 1987; 105: 627-9. 2. Bodis-Wollner I. Detection of visual defects using the contrast sensitivity function. Int. Ophthalmol Clin 1980; 20: 135-53. 3. Skalka HW. Comparison of Snellen acuity, VER acuity, and Arden grating scores in macular and optic nerve disease. Br J Ophthalmol 1980; 64: 24-9. 4. Weinstein GW. Clinical aspects of the visually evoked potentials. Trans Am Ophthalmol Soc 1977; 85: 627-73. 5. Odom JV, Hobson R, Coldren JT, Chao GM, Weinstein GW. 10-Hz flash visual evoked potentials predict post-cataract extraction visual acuity. Doc Ophthalmol 1987; 66: 291-9. 6. Vadrevu V. Predicting final visual acuity in diabetic eyes with vitreous hemorrhage. 10 Hz flicker VEPs. Doc Ophthalmol; 1992, accepted. 7. Farber ME, Odom JV, Hobson R. Visual function behind vitreal opacities: VEP assessment [abstract]. Invest Ophthalmol Vis Sci 1985; 26(ARVO suppl): 32. 8. Macsai M, Cavender SA, Michael M, Odom JV. Prediction of visual outcome in pseudo-

188

9. 10.

11. 12.

13.

phakic bullous keratopathy: A comparison of the PAM and VEP [Abstract]. Invest Ophthalmol Vis Sci 1991; 32(ARVO suppl): 1235. Weinstein GW, Odom JV, Hobson RR. Visual acuity and cataract surgery. Ophthalmol Annu 1987; 3: 25-43. Sokol S, Domar A, Moskowitz A. Utility of the Arden grating test in glaucoma screening: High false-positive rate in normals over 50 years of age. Invest Ophthalmol Vis Sci 1980; 19: 1529-33. Weatherhead GR. Use of the Arden grating test for screening. Br J Ophthalmol 1980; 64: 591-6. Higgins KE, Jaffe MJ, Coletta NJ, Caruso RC, de Monasterio FM. Spatial contrast sensitivity, importance of controlling the patient's visibility criterion. Arch Ophthalmol 1984; 102: 1035-41. Comerford JP. Vision evaluation using contrast sensitivity functions. Am J Optom Physiol Opt 1982; 60: 394-8.

Address for correspondence: Sheri A. Cavender, Department of Ophthalmology, WVU Health Sciences Center, Morgantown, WV 26506, USA Phone: (304) 293 3757; Fax: (304) 598 4892

Comparison of preoperative 10-Hz visual evoked potentials to contrast sensitivity and visual acuity after cataract extraction.

Cataract patients whose surgical outcomes were in question were referred for testing by visual evoked potentials, elicited through closed eyelids by a...
415KB Sizes 0 Downloads 0 Views