Journal of Pediatric Ophthalmology and Strabismus

Preoperative Alternate Occlusion Decreases Motion Processing Abnormalities in Infantile Esotropia

Arthur Jampolsky, MD; Anthony M Norcia, PhD; Russell B Hamer, PhD

Abstract

ABSTRACT

We have examined the effects of preoperative, full-time alternate occlusion on the development of visual motion processing mechanisms. Motion visual evoked potentials (MVEPs) were recorded longitudinally in 14 infantile esotropia patients during the course of standard preoperative occlusion therapy. The MVEP in these patients was initially asymmetric in a fashion consistent with a nasalward/temporalward response bias, with a motion asymmetry significantly higher than that of agematched normals. The magnitude of the developmental motion asymmetry declined significantly after an average of 24 weeks of alternate occlusion. This result implies that the binocular motion-sensitive cells underlying the MVEP retain some degree of plasticity up to at least 1 year of age. Our data suggest further that the persistence of motion asymmetries in untreated infantile esotropia patients is maintained by an active process that can be disrupted by alternate occlusion. Alternate occlusion apparently eliminates a form of abnormal binocular interaction that supports the persistence of the motion asymmetry. We propose that one of the necessary pre-conditions for symmetricization of motion processing in infantile esotropia is the absence of abnormal competitive binocular interactions.

Abstract

ABSTRACT

We have examined the effects of preoperative, full-time alternate occlusion on the development of visual motion processing mechanisms. Motion visual evoked potentials (MVEPs) were recorded longitudinally in 14 infantile esotropia patients during the course of standard preoperative occlusion therapy. The MVEP in these patients was initially asymmetric in a fashion consistent with a nasalward/temporalward response bias, with a motion asymmetry significantly higher than that of agematched normals. The magnitude of the developmental motion asymmetry declined significantly after an average of 24 weeks of alternate occlusion. This result implies that the binocular motion-sensitive cells underlying the MVEP retain some degree of plasticity up to at least 1 year of age. Our data suggest further that the persistence of motion asymmetries in untreated infantile esotropia patients is maintained by an active process that can be disrupted by alternate occlusion. Alternate occlusion apparently eliminates a form of abnormal binocular interaction that supports the persistence of the motion asymmetry. We propose that one of the necessary pre-conditions for symmetricization of motion processing in infantile esotropia is the absence of abnormal competitive binocular interactions.

INTRODUCTION

Full-time alternate occlusion for patients with infantile esotropia has been advocated as a means to prevent abnormal binocular interaction from occurring before the achievement of therapeutic realignment.1-2 It was suggested that full-time alternate occlusion would "keep the binocular slate clean" and, thus, preserve the normal development potential for binocularity. A corollary hypothesis is that alternate occlusion is not harmful. Indirect clinical evidence from enhanced final treatment results3 suggests a beneficial role for preoperative alternate occlusion therapy, but there is no direct evidence that alternate occlusion has either a positive or negative effect, per se, on human visual system development.

This study was designed to examine the effects of full-time alternate-day occlusion by longitudinal monitoring of a newly recognized form of binocularity - sensory motion processing.4 Binocularity was quantified by means of an objective test of the developmental status of binocular motion mechanisms applicable to infants as well as to adults.4·5 This test measures the symmetry of monocular visual evoked responses to nasalward versus temporalward directions of motion.

This study was designed to examine the sensory effects of full-time alternate-day occlusion by longitudinal monitoring of asymmetries in monocular motion processing in infantile esotropia patients. Motion processing was quantified by means of a new objective visual evoked potential test, the monocular motion visual evoked potential (MVEP). We have recently argued that the monocular MVEP can access the responses of a binocular subsystem of motion processing pathways in both infants and adults.4·5 The key feature of the responses is the degree of directionally selective asymmetry, specifically the degree of nasalward/temporalward asymmetry present in the monocular MVEP.

Relationship Between Motion Processing Asymmetries and Binocularity

Directional asymmetries of monocular optokinetic nystagmus (MOKN) and monocular smooth pursuit have long been known to be associated with the lack of binocularity in infantile esotropes. Ciancia6 was the first to report the observation that infantile esotropes had smooth monocular pursuit movements toward adduction, and irregular, deficient, nystagmoid eye movements in monocular abduction pursuit (Fig 1). These monocular tracking asymmetries - good temporal to nasal (adduction), and poor nasal to temporal (abduction) - were later documented with electro-oculograms by Ciancia's group,7·8 and others.911 Oculomotor measures such as MOKN and monocular pursuit movements indicate that the sensory-motor arc shows a strong bias for nasalward motion. The oculomotor nuclei which control optokinetic nystagmus (OKN) and pursuit movements receive sensory information from both cortical and subcortical sources and it is thus difficult to localize the site in the visual pathway which is responsible for producing asymmetric eye movements (Fig 2). The MVEP, on the other hand, reflects sensory processing in visual cortex.

FIGURE 1: (A) Patients with infantile esotropia exhibit smooth adduction tracking (temporal to nasal), and irregular, deficient nystagmoid abduction tracking.6 (B) Normal subjects exhibit smooth tracking in both adduction and abduction.

FIGURE 1: (A) Patients with infantile esotropia exhibit smooth adduction tracking (temporal to nasal), and irregular, deficient nystagmoid abduction tracking.6 (B) Normal subjects exhibit smooth tracking in both adduction and abduction.

FIGURE 2: The sensory-motor arc for optokinetic nystagmus (OKN) and pursuit eye movement responses. Both OKN and pursuit eye movements receive cortical information regarding retinal image motion. The motion visual evoked potential (MVEP) taps the response to retinal image motion at the level of the visual cortex.

FIGURE 2: The sensory-motor arc for optokinetic nystagmus (OKN) and pursuit eye movement responses. Both OKN and pursuit eye movements receive cortical information regarding retinal image motion. The motion visual evoked potential (MVEP) taps the response to retinal image motion at the level of the visual cortex.

Normal infants also show monocular motion asymmetries in both MOKN,1214 and in the MVEP.4 The magnitude of the motion asymmetry declines rapidly during the first 6 months of life. Full maturation of symmetric motion processing may take as long as 2 to 3 years for the most rapid motions and for the motion of fine detailed targets.15,16 Patients whose eyes were aligned for the first time after 2 years show a marked and persistent MVEP asymmetry, which appears to be a lifelong marker of the infantile esotropia.4

Taken together, the data from normal development and esotropia patients imply that early and prolonged disruption of binocularity due to eye misalignment interferes with the normal development of symmetrical motion processing. What is striking about the MVEP and MOKN tests is that these monocular tests can give valuable information about developmental changes in an apparently binocular sensory subsystem. In this study, we show that infants with onset of strabismus before 4 months of age manifest abnormal motion asymmetries before alternate occlusion therapy or surgery and that the magnitude of the developmental motion asymmetry declines during the course of alternate occlusion, even in the absence of binocular input.

METHODS

Patient Selection

The effects of alternate occlusion (using opaque pediatric eye patches) were examined in 14 consecutive infantile esotropia patients each of whom had (1) an onset of constant esotropia before 4 months of age without any significant accommodative component (mean age of onset = 5 weeks), (2) at least 5 weeks of full-time alternate occlusion, and (3) more than one MVEP test during the course of alternate occlusion therapy. MVEP recordings were also made in three additional patients: two patients with no previous history of alternate occlusion, and one patient with an onset of constant esotropia at 8 months of age. Clinical data about each patient are summarized in the Table.

Occlusion Therapy

Alternate occlusion therapy regimens were not equal for all patients (Table). Full-time alternate-occlusion regimens, such as alternate eye occlusion on alternate days for equal vision alternators, were prescribed unless there was a fixation preference, in which case the schedule was changed to not more than 3 or 4 days of occlusion of the better eye, and 1 day occlusion of the poorer eye. The regimens ranged from full-time alternate (RE 1 day: LE 1 day, ie, 1:1) to asymmetric (2:1, 3:1, or 4:1), to alternate occlusion followed by a day of no occlusion (eg, 1:1: off 1 day; 3 patients). Longer periods of continuous occlusion of any one eye, especially in very young infants, may indeed be harmful. Of course, occlusion in the patient with good fusion may disrupt fusion for various reasons, which are not relevant here. The range of occlusion durations at the time of the last MVEP recording was 5 to 68 weeks, with a mean of 31 weeks of occlusion. All but two patients underwent occlusion for 12 weeks or more. The average age at the time of the first MVEP recording was 36 weeks; at the last recording it was 62 weeks. The average age at surgery was 73 weeks.

MVEP Recording

MVEPs were measured monocularly in response to vertical sinusoidal gratings displayed on a video monitor. The gratings were square waves alternated between two positions separated by 90° of spatial phase. The temporal rate of stimulation was 6 Hz (12 changes of direction per sec). The spatial frequency of the gratings was 1 c/deg. The stimulus is illustrated schematically in Figure 3. The gratings were presented for 10 sec and several trials were recorded from each eye. Two bipolar derivations were used: Oz versus Ol and Oz versus 02. The MVEP was subjected to Fourier analysis to extract the amplitude and phase of the evoked response at 6 and 12 Hz using the general methods we have described elsewhere.16

Fourier Method for Detecting Motion Asymmetry

Symmetric and asymmetric MVEPs produce characteristic Fourier spectra. An MVEP in which the response to the two directions of motion is equal yields a response spectrum which is completely composed of even harmonic multiples of the 6 Hz stimulation frequency. An asymmetric MVEP will contain significant additional response components at the odd harmonic multiples of the stimulation frequency. If opposite directions of motion produce larger responses in each eye, as presumed to be the case when a nasalward/temporalward asymmetry is present, the temporal phase of the odd harmonic responses from each eye will differ by 180° of phase in the two eyes. The MVEP asymmetry/symmetry may be quantitatively expressed in terms of an index which reflects the relative asymmetry/symmetry of the MVEP response. This index was obtained by dividing the amplitude of the first harmonic responses by the sum of the first and second harmonic responses. This measure ranges between zero (completely symmetric response) and one (for a completely asymmetric response). Thus, the asymmetry index is that portion of the total MVEP that is asymmetric.

RESULTS

Individual Data

Figure 4 shows the raw MVEP data for patient 16 obtained at two times (at 76 and 85 weeks of age) during a period of full-time alternate occlusion therapy. The MVEP data are shown as vectors in polar coordinates.4 Each MVEP trial generates four vectors (two harmonics recorded for each of two recording channels), although the data from only one recording channel are shown for each MVEP recording session. The length of a vector corresponds to the amplitude of the MVEP response; and the angle within the 360-degree circle indicates the phase of the response (similar to response latency) relative to the stimulus motion. The top pair of polar plots show the data for the Fl (asymmetric) component; the bottom pair shows the F2 (symmetric) data. Data from the right eye are shown as solid vectors. LE data are shown as dashed vectors.

Table

TABLEClinical Demographics of Patient Sample

TABLE

Clinical Demographics of Patient Sample

An asymmetric MVEP is dominated by Fl responses4 with the LE and RE responses being in a 180-degree phase relationship (anti-phase). The anti-phase relationship between LE and RE responses implies that they were evoked by opposite directions of grating motion, implying a nasalward/temporalward asymmetry in the cortical motion responses. This kind of asymmetry appears on the polar plots as a "bowtie" configuration, with the LE and RE vectors pointing to opposite quadrants of the polar plot.

It is evident in Figure 4 that, at 76 weeks, patient 16's MVEP was strongly dominated by the asymmetric component (Fl) and that the symmetric component (F2) was small. By the time of the second recording at 85 weeks, the motion asymmetry index decreased in each eye, although the decrease was more pronounced in the RE. For the RE, the asymmetry index decreased from 0.80 to 0.21 over the 9-week period between the two MVEP measurements. For the LE, the index decreased from 0.63 to 0.43. In addition, the characteristic bowtie configuration in the Fl vectors, which was pronounced at 76 weeks, was absent at 85 weeks.

In general, individual patients exhibited three basic patterns of response to alternate occlusion: (i) a decrease in motion asymmetry in both eyes, (ii) a decrease in only one eye, and (iii) trivial changes in MVEP asymmetry during the course of occlusion therapy. Figure 5 summarizes the changes in asymmetry index for three patients, including patient 16, who showed a decrease in asymmetry in both eyes over the course of alternate occlusion. The vertical axes plot the asymmetry index as described in the Methods. The periods of occlusion are indicated by the thick bars above the age axes. The horizontal solid line indicates the average monocular asymmetry index for 26 normal infants (all 25 weeks of age or older) who had been tested with the same stimulus conditions as the patients. The dashed lines mark the normal average, ± 1 standard deviation. The solid and open squares are the patients' data for the RE and LE respectively. Two of the patients' (3 and 16) asymmetry indices approached normal limits during the period of occlusion therapy. In all three cases, the initially dominant eye (indicated by the asterisk) had the lower asymmetry index at the time of the first MVEP. For patient 10, the dominant eye also ended up with lower asymmetry index. For patient 3, the asymmetry index decreased dramatically and equally in both eyes over the course of occlusion therapy to within the normal range.

FIGURE 3: Motion visual evoked potentials (MVEPs) were measured monocularly in response to vertical sinusoidal gratings displayed on a video monitor. This figure diagrammatically depicts the horizontal back and forth display of the gratings as observed by the patient.

FIGURE 3: Motion visual evoked potentials (MVEPs) were measured monocularly in response to vertical sinusoidal gratings displayed on a video monitor. This figure diagrammatically depicts the horizontal back and forth display of the gratings as observed by the patient.

FIGURE 4: Motion visual evoked potential (MVEP) data from patient 16 at two times during full-time alternate occlusion therapy (76 and 85 weeks of age) The MVEP data from individual trials are shown as vectors in polar coordinates. The filled and open circles show the RE and LE vector averages, respectively. The top pair of polar plots show the data for the Fl (asymmetric) component; the bottom pair shows the F2 (symmetric) data. At 76 weeks, the MVEP was dominated by the asymmetric component (Fl) and the F2 vectors (symmetric response) were small. The motion asymmetry index was 0.80 for the RE and 0.63 for the LE. In addition, the Fl vectors manifest the characteristic "bowtie" configuration indicating a nasalward I temporalward motion asymmetry. At 85 weeks, the motion asymmetry index had decreased in each eye: to 0.21 in the RE and to 0.43 in the LE.

FIGURE 4: Motion visual evoked potential (MVEP) data from patient 16 at two times during full-time alternate occlusion therapy (76 and 85 weeks of age) The MVEP data from individual trials are shown as vectors in polar coordinates. The filled and open circles show the RE and LE vector averages, respectively. The top pair of polar plots show the data for the Fl (asymmetric) component; the bottom pair shows the F2 (symmetric) data. At 76 weeks, the MVEP was dominated by the asymmetric component (Fl) and the F2 vectors (symmetric response) were small. The motion asymmetry index was 0.80 for the RE and 0.63 for the LE. In addition, the Fl vectors manifest the characteristic "bowtie" configuration indicating a nasalward I temporalward motion asymmetry. At 85 weeks, the motion asymmetry index had decreased in each eye: to 0.21 in the RE and to 0.43 in the LE.

Figure 6 shows MVEP asymmetry indices for four patients showing a decrease in asymmetry in one eye and trivial changes in the other during the course of alternate occlusion therapy. Two other patients (4 and 13), who showed an increase in motion asymmetry in the other eye, are also shown. In four of these six patients (1, 4, 13, and 14), the initially dominant eye (indicated by the asterisk) started out having the higher asymmetry index and then decreased to within normal limits by the time of the last MVEP measurement. For two patients (6 and 12), the initially dominant eye had a virtually normal asymmetry index at both MVEP measurements.

Figure 7 shows the asymmetry index data from five patients showing either no change or, in one case (patient 11), an increase in asymmetry in one eye with little change in the other. For patients 9 and 15, the asymmetry index started and remained within the normal limits throughout the period of testing. It is interesting to note that these two patients had received 13 and 14 weeks of occlusion therapy, respectively, before the first MVEP measurement. Patient 7 had only 5 weeks of occlusion therapy which may have been insufficient to produce an effect. One patient (8), who was tested extensively throughout more than 1 year of alternate occlusion therapy, showed no sustained decrease in motion asymmetry.

Group Performance

Figure 8 summarizes the data for all 14 patients. The filled squares at 37 and 61 weeks show the mean asymmetry index ( ± 1 SEM) for the 14 patients at the time of their first and last MVEP measurement during the course of alternate occlusion therapy. The open circle shows the average monocular asymmetry index (0.28, ± 1 SEM) for a group of 26 normal infants 25 weeks of age or older (mean age = 38 weeks). The horizontal dashed lines indicate the 95% confidence limits for normal asymmetry index (normal average ± 1.96 SEM). For the stimulus conditions used in the present study (6 Hz, 1 c/deg), normal infants lose the MVEP asymmetry by about 20 weeks of age.4 In contrast with the normal infants of the same average age, infantile esotropia patients show as large a motion asymmetry index (0.58) at the time of their first MVEP measurement as immature normal infants. (These patients had already had an average of 5.4 weeks of occlusion before the first MVEP). However, after an average of 24 additional weeks of alternate occlusion therapy, the asymmetry index decreased to 0.43. A two-way repeated measures analysis of variance (ANOVA) was performed on the patients' asymmetry index data comparing the first and last MVEP recording sessions during the period of occlusion therapy. Data from the two recording channels were combined and were partitioned according to eye preference noted in the first few examinations. In seven cases, this eye preference was not strong enough to warrant an asymmetrical occlusion regimen (Table). The ANOVA showed that the change in asymmetry index (0.15) during preoperative alternate occlusion was highly statistically significant (P < .012, df = 12).

FIGURE 5: Motion visual evoked potential (MVEP) asymmetry indices for three patients showing a decrease in asymmetry in both eyes during the course of alternate occlusion therapy. The time periods of occlusion are indicated by the thick bars above the age axes. The horizontal solid line indicates the average monocular asymmetry index for 26 normal infants (all 25 weeks of age or older) who had been tested with the same stimulus conditions as the patients. The dashed lines mark the normal average, ± 1 standard deviation. The solid and open squares are the patients' data for the RE and LE respectively. Two of the patients (3 and 16) had data that approached normal limits during the period of occlusion therapy. In all three cases, the initially dominant eye (indicated by the asterisk) had the lower asymmetry index at the time of the first MVEP. For patient 10, the dominant eye also ended up with a lower asymmetry index.

FIGURE 5: Motion visual evoked potential (MVEP) asymmetry indices for three patients showing a decrease in asymmetry in both eyes during the course of alternate occlusion therapy. The time periods of occlusion are indicated by the thick bars above the age axes. The horizontal solid line indicates the average monocular asymmetry index for 26 normal infants (all 25 weeks of age or older) who had been tested with the same stimulus conditions as the patients. The dashed lines mark the normal average, ± 1 standard deviation. The solid and open squares are the patients' data for the RE and LE respectively. Two of the patients (3 and 16) had data that approached normal limits during the period of occlusion therapy. In all three cases, the initially dominant eye (indicated by the asterisk) had the lower asymmetry index at the time of the first MVEP. For patient 10, the dominant eye also ended up with a lower asymmetry index.

FIGURE 6: Motion visual evoked potential (MVEP) asymmetry indices for four patients showing a decrease in asymmetry in one eye and trivial changes in the other during the course of alternate occlusion therapy. Two other patients (4 and 13) who showed an increase in motion asymmetry in the other eye are also shown. In four patients (1, 4, 13, and 14) the initially dominant eye (indicated by the asterisk) started out having the higher asymmetry index and then decreased to within normal limits by the time of the last MVEP measurement. For the other two patients (6 and 12), the initially dominant eye had a virtually normal asymmetry index at both MVEP measurements.

FIGURE 6: Motion visual evoked potential (MVEP) asymmetry indices for four patients showing a decrease in asymmetry in one eye and trivial changes in the other during the course of alternate occlusion therapy. Two other patients (4 and 13) who showed an increase in motion asymmetry in the other eye are also shown. In four patients (1, 4, 13, and 14) the initially dominant eye (indicated by the asterisk) started out having the higher asymmetry index and then decreased to within normal limits by the time of the last MVEP measurement. For the other two patients (6 and 12), the initially dominant eye had a virtually normal asymmetry index at both MVEP measurements.

FIGURE 7: Motion visual evoked potential (MVEP) asymmetry indices from five patients showing either little change or, in one case (11), an increase in asymmetry in one eye with little change in the other. For patients 9 and 15, the asymmetry index started out and remained within the normal range throughout the period of testing. One patient (8), who was tested extensively throughout more than 1 year of alternate occlusion therapy, showed no sustained decrease in motion asymmetry.

FIGURE 7: Motion visual evoked potential (MVEP) asymmetry indices from five patients showing either little change or, in one case (11), an increase in asymmetry in one eye with little change in the other. For patients 9 and 15, the asymmetry index started out and remained within the normal range throughout the period of testing. One patient (8), who was tested extensively throughout more than 1 year of alternate occlusion therapy, showed no sustained decrease in motion asymmetry.

Thirteen out of 14 patients showed a decrease in asymmetry index between their first and last MVEP in their dominant eye, a statistically significant effect (one-tailed sign-test, P < .0017). However, only 8 of the 13 nondominant eyes successfully tested showed a decrease in asymmetry (P > .298, not statistically significant). However, based on the parametric data analysis, there was neither a main effect of eye-dominance, nor a significant Eye versus MVEP-session interaction.

FIGURE 8: Summary of group asymmetry index data for all 14 patients. The filled squares at 37 and 61 weeks show the mean asymmetry index for the 14 patients at the time of their first and last motion visual evoked potential (MVEP) measurement during the period of alternate occlusion therapy. Data from the two recording channels were combined, and were partitioned according to eye preference. An ANOVA showed that the decrease in asymmetry index (from 0.58 to 0.430) was highly statistically significant (P < .012, df = 1). The open circle shows the average monocular asymmetry index (0.28) for a group of 26 normal infants 25 weeks of age or older (mean age = 38 weeks). The horizontal dashed lines indicate the 95% confidence limits for a normal asymmetry index (the normal average, ± 1.96 SEM). The large open square indicates the average asymmetry index (±1 SEM) for the six control patients who had had no alternate occlusion therapy. The asymmetry index of the control patients (0.56), having an average age of 61 weeks, was nearly identical to the index of the younger group of patients. This indicates that infantile esotropia patients will not simply grow out of the motion asymmetry in the absence of alternate occlusion therapy.

FIGURE 8: Summary of group asymmetry index data for all 14 patients. The filled squares at 37 and 61 weeks show the mean asymmetry index for the 14 patients at the time of their first and last motion visual evoked potential (MVEP) measurement during the period of alternate occlusion therapy. Data from the two recording channels were combined, and were partitioned according to eye preference. An ANOVA showed that the decrease in asymmetry index (from 0.58 to 0.430) was highly statistically significant (P < .012, df = 1). The open circle shows the average monocular asymmetry index (0.28) for a group of 26 normal infants 25 weeks of age or older (mean age = 38 weeks). The horizontal dashed lines indicate the 95% confidence limits for a normal asymmetry index (the normal average, ± 1.96 SEM). The large open square indicates the average asymmetry index (±1 SEM) for the six control patients who had had no alternate occlusion therapy. The asymmetry index of the control patients (0.56), having an average age of 61 weeks, was nearly identical to the index of the younger group of patients. This indicates that infantile esotropia patients will not simply grow out of the motion asymmetry in the absence of alternate occlusion therapy.

The natural and important question arises whether the significant decrease in motion asymmetry could be due to factors other than the occlusion therapy. We have addressed this issue by analyzing the data from patients who, at the time of the first MVEP, had experienced no prior alternate occlusion. These data came from the first MVEP measurements from a subgroup (N = 4) of the original 14 infantile esotropia patients, plus two additional patients with no prior alternate occlusion (2 and 5). Only patients older than 20 weeks of age (median age = 41 weeks; mean age = 61 weeks) were included since, below this age, normal infants are still immature (have asymmetrical MVEPs) for the present stimulus conditions. The data from these six patients thus serve as a control to determine the natural asymmetry status of infantile esotropia patients in the absence of any alternate occlusion therapy. Their mean monocular asymmetry index was 0.56 (large open square in Figure 8), virtually identical to the mean index from the initial recording in the larger (and younger) group of patients. Moreover, a linear regression of their asymmetry indexes versus age (weeks) revealed no correlation with age (slope = 0.001, r = 0.18), implying that in the absence of occlusion therapy, the patients will not simply "grow out of it."

DISCUSSION

Infantile esotropia patients who have undergone alternate occlusion therapy show a significant decline in the magnitude of their developmental motion asymmetry (Fig 8). On average, at the time of the first MVEP (37 weeks), the asymmetry index started out at a value appropriate to immature normal infants. By the time of the final MVEP during occlusion therapy (61 weeks), the asymmetry index showed a statistically significant decrease. The group data thus indicate a significant treatment effect of preoperative alternate occlusion: an average of 24 weeks of alternate occlusion therapy applied before surgical eye alignment can significantly decrease an index of abnormal motion processing.

It is also important to note that there are substantial individual differences in response to this treatment (Figs 4-7). Not all patients show a decrease; and asymmetry decreases, when they occur, do not necessarily occur in both eyes. There is a tendency for the preferred eye at the time of initial examination to improve more during the course of occlusion therapy than the nondominant eye.

That the above is the result of alternate occlusion therapy per se is supported by the control MVEP data (Fig 8), which suggest that the significant decline in motion asymmetry would not have occurred in the absence of occlusion treatment. In other words, untreated infantile esotropia patients will not simply grow out of their abnormal motion processing. It should also be noted that patients with infantile esotropia who were aligned after age 2 years maintain high levels of motion asymmetry into adulthood in their MVEPs.4

The fact that the motion asymmetry index can decline during relatively long periods of alternate occlusion has several implications. First, the decrease in MVEP asymmetry index, per se, implies some degree of plasticity in the neural substrate underlying binocular motion processing in infantile esotropia up to at least 1 year of age. Apparently, the abnormalities in at least some of the binocular pathways in infantile esotropia are not "hard-wired." Such plasticity is consonant with our previous finding that early, successful surgical alignment causes a reduction in the asymmetry.17 Second, it suggests that the motion asymmetry present in untreated patients represents either the persistence of, or a pathological modification of, the asymmetric motion processing that is normal in early infancy. Third, the persistence of asymmetry is maintained by an active process that is at least partially disrupted by alternate occlusion. Alternate occlusion apparently eliminates a form of abnormal binocular interaction that supports the persistence of the motion asymmetry.

The present findings indicate that one of the necessary pre-conditions for symmetricization of motion processing is the absence of abnormal binocular competition. This pre-condition is naturally achieved in normal development, where cortical cells are receiving balanced, simultaneous input from the two eyes. In untreated infantile esotropia, the patient experiences continuous, abnormal binocular competition. This pre-condition is also not present in unilateral congenital cataract patients, or in strabismic and anisometropic amblyopes. In these cases, developmental motion asymmetries persist in the form of asymmetrical MOKN responses.9,18 In each of these disorders, cortical binocular cells receive unequal inputs and thus experience abnormal competitive interactions.

The simple absence of any competition, however, is probably not sufficient for the complete maturation of the motion system. First, our motion asymmetry index is higher than normal even after occlusion. Second, MOKN asymmetries persist in patients who had an eye enucleated early in development.39

The clinical outcome of modern treatment of infantile esotropia, including the use of full-time alternate occlusion, is often excellent (A. Jampolsky, clinical observations, personal communication, March 1993). 3,20,21 With early (<24 months) surgery and good (^ 10 prism diopters) ocular alignment, patients can have normal acuities in each eye, with a high likelihood of having binocular fusion, and/or stereopsis ranging from gross (200 to 2000 arcsec) to fine (40 arcsec) in some cases.20,21 Based on these data, and abundant clinical experience, the principal author wishes to emphasize that alternate occlusion therapy is a safe procedure, and can be carried out for many months as long as the occlusion is not applied continuously to one eye for several consecutive days. Our current results are consistent with this overall clinical picture - the developmental motion asymmetry declines during clinical occlusion, but not during clinical strabismus.

Relationship to Animal Deprivation Models

Animal deprivation studies22"27 which have investigated the effects of various experimental manipulations of normal animals, ranging from alternate occlusion28 to experimental ocular (or optical) misalignment.29'31 These studies have found deficits in the visual system, with varying degrees of severity and plasticity, depending on the details of the experimental manipulation (eg, Ud suture versus light occlusion) and its duration relative to the species' sensitive and critical periods.

However, the analogy between these animal deprivation models and clinical application of occlusion therapy is by no means straightforward. The animal literature does not contain examples of studies that are directly analogous to the clinical management of strabismus in humans. We know of no experiments in strabismic (naturally occurring or experimental) animals that reproduced the conditions used in optimal management of human strabismus - namely, effective (and early-for-species) ocular realignment, achieved either with or without judiciously applied occlusion therapy.

Modification of the "Dark-Rearing" Hypothesis

Full-time alternate occlusion, in one schedule form or another, has been routine practice for four decades.1 In 1953, the following hypotheses were stated: "Alternate occlusion keeps the binocular slate clean..., maintains normal innate vision potential... and prevents bad sensorial habits."2 These hypotheses were further elaborated in the context of the experimental animal models of cortical plasticity which emerged in the 1960s and 1970s. Specifically, an analogy was made between the no-stimulus paradigm in dark-reared animals and the no-binocular-stmiulus condition of alternate occlusion.2

The present results suggest a qualitative difference between the effects of full-time alternate occlusion and some of the effects seen in the conventional dark-rearing paradigm. Dark-rearing not only extends the critical period for acuity development, but it also appears to retard the normal developmental progression of visual acuity during the period of deprivation.3234 On the other hand, in our study, under the no-binocular-stimulus condition (full-time alternate occlusion), we find that our index of the development of the sensory motion system proceeded toward normality. It remains an open question as to whether the critical period(s) for plasticity in binocular and motion-processing pathways is also delayed by alternate occlusion in humans. Preliminary data in a monkey model has indicated that the developmental sequence for stereopsis can be delayed by several weeks by alternate occlusion during the early post-natal period.35

In contrast with the prolongation of plasticity by darkrearing are the well-known deleterious effects of rearing when light input is permitted, but pattern information is severely restricted, such as by a diffused image through a sutured lid.36·37 Such pattern deprivation paradigms result in severe loss of responses to pattern information, both at the cortical/cellular level36,37 and behaviorally.38 By analogy, when both eyes are open, the infantile esotropia patient experiences abnormal binocular interactions that are deleterious. Since full-time alternate occlusion eliminates such interactions, we have viewed this treatment as a "darkrearing* paradigm for the binocular visual system.2

Relationship to Other Measures of Binocularity

Stereopsis. It is important to emphasize that present data do not have a direct bearing on the potential for development of stereopsis in infantile esotropia. Although we have inferred that the MVEP appears to be sampling a binocular motion subsystem,4 this system may be partially or entirely distinct from the pathways subserving stereopsis. Physiological studies have identified three classes of binocular cells: cells requiring simultaneous activation from both the left and right eyes (so-called "AND" or "obligate binocular" cells)39,40; cells that receive excitatory input from both the left and right eyes (so-called "OR" cells), and cells that receive excitatory input from one eye and inhibitory input from the other eye.41,42 There are also a small number of truly monocular cells - those cells whose response is in no way influenced by stimuli presented to the other eye. The key feature of cells subserving stereopsis in visual cortex may be the presence of binocular interactions - namely, cells that respond in anything other than a simply additive fashion (BE response = RE response + LE response) to binocular stimulation. The monocular MVEP cannot monitor such binocular interactions by the nature of the monocular testing.

What our data have demonstrated is that alternate occlusion tends to normalize the responses of some classes of binocular cells, implying the presence of plasticity in these cells. Moreover, such plasticity seems to remain in spite of the fact that throughout the entire period of alternate occlusion, all cells other than obligate binocular cells are being constantly stimulated. The fate of cells subserving stereopsis may not be linked to such plasticity since it is well known that the different visual mechanisms have different critical periods.43 The direct analogy with dark-rearing in animal studies is for the binocular AND cells - for these cells full-time alternate occlusion corresponds to no stimulation at all. Td date, the only relevant experimental evidence on this issue comes from the work on infant monkeys cited above.39

Suppression and Anomalous Retinal Correspondence (ARC). Other manifestations of abnormal binocularity, suppression, and ARC, are known to be dependent on an active process that requires continuing non-fusable inputs from the two eyes.44,45 Our data suggest that the maintenance of the developmental motion asymmetry requires similar preconditions and may thus share some underlying mechanisms. Unlike suppression and ARC, the developmental motion asymmetry can be measured under monocular viewing conditions, can be measured without the need for subjective responses, and can be easily quantified.

SUMMARY

1. A new, objective measure of binocular motion processing, the MVEP, has been used to monitor the status of motion processing in 14 infantile esotropia patients during the course of alternate occlusion therapy before surgical eye alignment.

2. The relative normality of the patients' motion processing was assessed by an asymmetry index derived from the raw MVEP data. The index (ranging from 0 to 1) quantifies the degree to which the MVEP is dominated by asymmetric response components: a high index indicates asymmetric (abnormal) motion processing at the level of visual cortex; and a low index value indicates dominance by symmetric (normal) components. For the stimulus condition used in the present study, normal infants develop symmetrical MVEPs by around 20 weeks of age.

3. Compared to age-matched normal infants who have achieved symmetrical MVEPs (mean index of 0.28), the group of infantile esotropia patients had abnormally high asymmetry indices (mean index of 0.58) before, or early in, a prolonged period of alternate occlusion therapy. After an average of 24 weeks of alternate occlusion therapy, the mean asymmetry index of the patients decreased significantly to 0.43.

4. The significant symmetricization of MVEPs in response to prolonged periods of alternate occlusion therapy implies that the binocular motion processing pathways in infantile esotropia patients retain some plasticity up to at least 1 year of age.

5. The present data stand in contrast to the persistence of abnormal (asymmetric) motion processing noted in infantile esotropia patients aligned after age 2 years. Asymmetrical motion processing in such patients may, therefore, be maintained by an active process that is interrupted by alternate occlusion. Hence, early fulltime alternate occlusion may prevent these abnormal binocular interactions, and allow at least some of the sensory substrate to develop toward normality.

6. In spite of the significant effect on the group data, individual differences in the response to occlusion therapy were noted.

7. The MVEP holds promise as a tool with which to monitor the developmental status of a binocular motionprocessing subsystem via an objective monocular measurement. This new tool will undoubtedly help quantify the developmental status of the sensory end of the sensory-motor arc in a patient population too young to monitor using other more subjective measures of binocularity (eg, measures of anomalous retinal correspondence).

REFERENCES

1. Jampolsky A. The physician and the crossed-eyed child. A medical forum. Modern Medicine. 1953;21:144-154.

2. Jampolsky A. Unequal visual inputs and strabismus management. In: Strabismus Symposium, Transactions of New Orleans Academy of Ophthalmology. St Louis, Mo: CV Mosby; 1978:358-592.

3. Foster RS, Paul TO, Jampolsky A. Management of infantile esotropia. Am J Ophthalmol. 1976;82:291-299.

4. Norcia AM, Garcia H, Humphry R, Holmes A, Hamer RD, Orel-Bixler D. Anomalous motion VEPs in infants and infantile esotropia. Invest Ophthalmol Vis Sci. 1991;32:436-439.

5. Norcia AM, Hamer RD, Orel-Bixler D. Temporal tuning of the motion VEP in infants. Invest Ophthalmol Vis Sci. 1990;31<suppl):10.

6. Ciancia AO. La Esotropia Con Limitación Bilateral de la-Abduccion en el Laçante. Archivos de Oftalmologia de Buenos Aires. 1962;36: 207-211.

7. Ciancia AO. Influencia de la oclusión de un ojo sobre los movimientos oculares en la esotropia con limitación las rio dio bilateral de la abducción. In: Souza-Dias C, ed. Anais de V Congresso do Conselho Latino-Americano de Estrabismo. Guarujah, Brasil: Consejo Latino Americano de Estrabismo;1976:37-43.

8. Melek NB, Garcia H, Ciancia AO. La EOG de Seguimiento de Persecución en las Esotropias con Limitación Bilateral de Abducción. Archives de Oftalmologia de Buenos Aires. 1979;54:271-278.

9. Schor CM, Levi DM. Disturbances of small-field horizontal and vertical optokinetic nystagmus in amblyopia. Invest Ophthalmol Vis Sci. 1980;19:668-683.

10. Tychsen LR, Hertig R, Scott WE. Pursuit is impaired but the vestibuloocular reflex is normal in infantile esotropia. Arch Ophthalmol. 1985;103:536-539.

11. Tychsen LR, Lisberger SG. Maldevelopment of visual motion processing in humans who had strabismus with onset in infancy. J Neurosci. 1986;6:2495-2508.

12. Atkinson J. Development of optokinetic nystagmus in the human infant and monkey infant. In: Freeman RD, ed. Development Neurobiology of Vision. New York, NY: Plenum; 1979:277-287.

13. Naegele JR, Held R. The postnatal development of monocular optokinetic nystagmus in infants. Vision Res. 1982;22:341-347.

14. Mohn G. The development of binocular and monocular optokinetic nystagmus in human infants. Invest Ophthalmol Vis Sci. 1989;30(suppl):49.

15. Lewis TL, Maurer D, van Schaik CS. Monocular OKN acuity is asymmetrical in normal 3-month olds. Invest Ophthalmol Vis Sci. 1989;3(Ksuppl):49.

16. Norcia AM, Tyler CW1 Clarke M. Digital filtering and robust regression techniques for estimating sensory thresholds from the evoked potential. IEEE Eng Med Biol. 1985;22:26-32.

17. Norcia AM, Jampolsky A, Hamer RD, Orel-Bixler D. Plasticity of human motion processing following strabismus surgery. Invest Ophthalmol Vis Sci. 1991;32(suppl):1044.

18. Maurer D, Lewis T. Visual outcomes in infant cataracts. In: Simons K, ed. Infant Vision: Basic and Clinical Research. New York, NY: Oxford University Press. In press.

19. Reed MJ, Steinbach MJ, Anstis SM, Gallie B, Smith D, Kraft S. The development of optokinetic nystagmus in strabismic and monocularly enucleated subjects. Behau Brain Res. 1991;46:31-42.

20. Ing MR. Early surgical alignment for congenital esotropia. Transactions of the American Ophthalmological Society. 1981;LXXDC:625-652.

21. Ing MR. Early Surgical Alignment for Congenital Esotropia. San Francisco, Calif: American Academy of Ophthalmology; 1983;90: 132-135.

22. Movshon JA, Van Sluyters RC. Visual neural development. Annual Review of Psychology. 1981;32:477-522.

23. Sherman SM, Spear PD. Organization of visual pathways in normal and visually deprived cats. Physiol Rev. 1982;62:738-855.

24. Fregnac Y, Imbert M. Development of neuronal selectivity in primary visual cortex of cat. Physiol Reu 1984;64:325-434.

25. Hirsch VB, Tieman DG, Tieman SB, Tumosa N. Unequal alternating exposure: effects during and after the classical critical period. In: Rauschecker JP, Marler P, eds. Imprinting and Cortical Plasticity. New York, NY: John Wiley; 1987:287-319.

26. Mitchell DE. The long-term effectiveness of different regimens of occlusion on recovery from early monocular deprivation in kittens. Philosophical Transactions of the Royal Society of London. Series B (Biological Sciences). 1991;333:51-79.

27. Rauschecker JP. Mechanisms of visual plasticity: Hebb synapses, NMDA receptors and beyond. Physiol Rev. 1991;71:587-615.

28. Blake RS, Hirsch HVB. Deficits in binocular depth perception in cats after alternating monocular deprivation. Science. 1975;190:1114-1116.

29. Hubel DH, Wiesel TN. Binocular interaction in striate cortex of kittens reared with artificial squint. J Neurophysiol. 1965;28:1041-1059.

30. Crawford MLJ, von Noorden GK The effects of short-term experimental strabismus on the visual system in Macaca Mulatta. Invest Ophthalmol Vis Sci. 1979;18:496-505.

31. Bennnett MJ, Smith EL III, Harwerth RS, Crawford MLJ. Ocular dominance, eye alignment, and visual acuity in kittens reared with an optically induced squint. Brain Res. 1980;193:33-45.

32. Timney B, Mitchell DE, Cynader M. Behavioral evidence for prolonged sensitivity to effects of monocular deprivation in dark-reared cats. J Neurophysiol. 1980;43:1041-1054.

33. Timney B, Mitchell DE, Giffin F. The development of vision in cats after extended periods of dark-rearing. Exp Brain Res. 1978;31: 547-560.

34. Cynader M, Mitchell DE. Prolonged sensitivity to monocular deprivation in dark-reared cats. J Neurophysiol. 1980;43:1026-1040.

35. Jampolsky A, Brown RJ, Boothe RG, et al. Delay of stereoacuity development in monkeys by full-time alternate occlusion. Invest Ophthalmol Vis Sci. 1993;34(suppl):1188.

36. Wiesel TN, Hubel D. Single-cell responses in striate cortex of kittens deprived of vision in one eye. J Neurophysiol. 1963;26:1003-1017.

37. Ohashi T, Norcia AM, Kasmatsu T, Jampolsky A. Cortical recovery from effects of monocular deprivation caused by diffusion and occlusion. Brain Res. 1991;548:73.

38. Giffin F, Mitchell DE. The rate of recovery of vision after early monocular deprivation in kittens. J Physiol. 1978;274:511-537.

39. Poggio GF, Fischer B. Binocular interaction and depth sensitivity in striate and prestriate cortex of behaving rhesus monkey. J Neurophysiol. 1977;40:1392-1405.

40. Hubel DH, Livingstone MS. Segregation of form, color, and stereopsis in primate area 18. J Neurosci. 1987;7:3378-3415.

41. Ohzawa I, Freeman RD. The binocular organization of simple cells in the cat's visual cortex. J Neurophysiol. 1986;56:221-242.

42. Ohzawa I, Freeman RD. The binocular organization of complex cells in the cat's visual cortex. J Neurophysiol. 1986;56:243-259.

43. Harwerth RS, Smith EL, Duncan GC1 Crawford MLJ, von Noorden GK. Multiple sensitive periods in the development of the primate visual system. Science. 1986;232:235-238.

44. Hallden U. Fusional phenomena in anomalous correspondence. Acta Ophthalmol. 1952;37(suppl):l.

45. Jampolsky A. Characteristics of suppression in strabismus. Arch Ophthalmol. 1955;54:683-696.

TABLE

Clinical Demographics of Patient Sample

10.3928/0191-3913-19940101-04

Sign up to receive

Journal E-contents