Visual Field Development in Infants with Stage 3

580

INVESTIGATIVE OPHTHALMOLOGY & VI5UAL SCIENCE / March 1989

Vol. 30

Investigative Ophthalmology & Visual Science, Vol. 30, No. 3, March 1989 Copyright © Association for Research in Vision and Ophthalmology

Visual Field Development in Infants with Stage 3 Retinopathy of Prematurity Deatrlz Luna.t Velma Dobson,*t Nancy A. Carpenter,* and Albert W.

Binocular visualfielddevelopment was measured in 11 infants who had stage 3 ROP in early infancy and in 11 infants without ROP, matched for birthweight and gestational age. Kinetic perimetry was used to measure visual fields along the 45°, 135°, 225° and 315° half-meridia. Infants were tested at 4, 9, and 18 months from due date. Analyses of variance were used to compare results of the two groups for each age tested. Results at the 4-month test age indicated that both groups had visualfieldswithin the normal range for their age. However, at the 9-month test age the ROP group showed a significantly (P < 0.05) smaller visual field than the control group. At 18 months, the ROP group still showed smaller visual fields than the control group, but the difference was not significant. The results suggest that dysfunction of the peripheral retina associated with ROP may produce a constriction of the visual field or a delay in visual field development. Invest Ophthalmol Vis Sci 30:580-582,1989 Retinopathy of prematurity (ROP) is a disorder characterized by abnormal development of the retinal vasculature of preterm infants. In stage 3 ROP (ICROP),1 the blood vessels in the retinal periphery form a ridge with extension of fibrovascular tissue into the vitreous cavity of the developing eye. Most affected infants will develop normal central vision, but it is reasonable to suspect that peripheral visual function may be affected. Two studies have examined visual fields in children who had ROP. Majima,2 using static and kinetic perimetry, found a significant reduction in peripheral retinal sensitivity in children with cicatricial disease, but not in children whose ROP had resolved without residua. Unfortunately, Majima did not correlate these findings with the active stages of ROP these children had as infants. Tamai and colleagues3 looked at several visual functions, including visual fields of children with ROP who had undergone cryocautery. They found that six out of the 11 infants tested had defects in their visual fields. Again, the active stages of ROP these children had as infants were not reported. Results from these studies suggest that infants who have ROP are at risk for defects in their peripheral vision. It is still not clear, though, how the acute phase of the ROP relates to later visual field development. The purpose of this study was to measure visual field development in a group of infants who had stage 3 ROP during early infancy.

Materials and Methods. Subjects: Infants tested were born at Magee Womens Hospital and The Western Pennsylvania Hospital in Pittsburgh, Pennsylvania between December 1984 and June 1987. All had gestational ages less than 32 weeks. Subjects in the ROP group were five males and six females who had one or more clock hours of stage 3 ROP. Four of these subjects also had plus disease. In all subjects, the ROP had regressed by 5 months corrected (postterm) age. Cicatricial RLF did not occur except in one infant who developed macular heterotopia after data reported in this paper were collected. There were four males and seven females in the comparison group. These controls were matched to the ROP group in birthweight (±200 g) and gestational age (±2 weeks). Secondary disorders related to prematurity were similar in the two groups (see Table 1). Ten infants in the comparison group showed no evidence of ROP and one infant showed stage 1 ROP which had regressed by the time the infant reached term (13 weeks postnatal age). None of the infants in either group had strabismus during the study. Apparatus and procedure: Binocular visual fields were measured along the 45°, 135°, 225° and 315° half-meridia using kinetic perimetry in the manner developed by Mohn and van Hof-van Duin4 (see Fig. 1). Testing was conducted at 4, 9, and 18 months corrected age. All parents or guardians signed informed consents before testing began. Results. For each infant, the medians of the measures taken along each half-meridian were determined. Matched-sample repeated measures analyses of variance were then used to determine whether, for each age tested, visual field size differed between the ROP and the no-ROP groups and whether visual field size varied across directions. The results indicated no effect of direction and no group by direction interaction at any of the test ages. There was no significant difference between groups at the 4-month test age (F(l,5) = 0.549, NS). However, at the 9-month test age the infants with ROP showed significantly smaller visualfieldsthan did the infants without ROP (F(l,8) = 9.41, P < 0.015). By 18 months, the infants with ROP still showed smaller visual fields than infants without ROP but this was not statistically significant (F( 1,4) = 0.294, NS). Figure 2 shows the mean visual field size, averaged

No. 3

across directions, for the ROP and no-ROP groups at the three test ages. The visual field size increases with age in both groups. However, the increase appears to be slower for the ROP group than for the no-ROP group. To investigate further the 9-month results, visual field deficits for individual infants were determined by comparing each infant's median scores along every test direction with normative data of healthy preterm infants.5 A deficit in visual field size was considered to be present if a value fell more than 2 SD below the mean of the healthy preterm group for each direction. Field deficits were found in three out of eight infants in the ROP group and in none of the infants in the no-ROP group. Eye exams obtained during the active phase of ROP were then examined to determine location of ROP in the retina. The eye exam results indicated that the three infants with visual field deficits had had more than 4 clock hours of stage 3 ROP located in the nasal retina. None of the infants whose stage 3 ROP had been located exclusively in the temporal retina showed field deficits. Discussion. The results indicate that infants who have ROP early in life can have a decrease in visual field size as measured in later infancy. It would be of considerable interest to know whether the visual field deficits shown by these infants were due to lasting effects of the ROP on the peripheral retina or to other aspects of these high-risk infants' medical condition. The results of our study allow us to eliminate several possible explanations.

Fig. 1. Black metal arc perimeter with arms extending at45M35°,225 o and315° half-meridia. The infant is positioned in front of the perimeter, 36 cm from the central sphere. An observer stands behind a black felt curtain, prompts the infant tofixateon the central stimulus and monitors the infant's eye movements through a small aperture in the curtain. A second identical sphere is moved by the tester towards the center of the arc along one of the four arms at approximately 2-3 deg/sec. The distance at which the infant first looks at this target is recorded. Each infant is tested with two to three trials per halfmeridian.

581

Reports

Table 1. Complications related to preterm birth

Bronchopulmonary dysplasia Hyaline membrane disease Periventricular leukomalacia Intraventricular hemorrhage Grade III Grade II Grade I Perinatal hypoxia/asphyxia Hyperbilirubinemia

ROP group (n = 11)

No-ROP group (n = 11)

8 1 2

7 2 0

2 1 1 1 10

1 0 0 4 10

First, the reduction in visual field size was not due to a decrease in visual acuity because the spheres used as stimuli were quite large (6° in diameter). In addition, as part of another study6 we measured each infant's visual acuity and found that there was no correlation between visual acuity and visual field size. Likewise, many infants who have ROP also develop strabismus, which can produce an increase in visual field size (exotropia) or a decrease in visual field size (esotropia). However, this cannot explain our results since none of the infants in either group had strabismus at any of the test ages. Central nervous system (CNS) problems are common in this population, as well. Complications include intraventricular hemorrhage (IVH) and periventricular leukomalacia (PVL), which may involve the visual pathways. As shown in Table 1, the incidence of CNS complications was slightly greater in

582

INVESTIGATIVE OPHTHALMOLOGY 6 VISUAL SCIENCE / March 1989

• control group o rop group * p < 0.05

Vol. 30

group was related to the ROP was that the deficit was seen at the 9-month test age, when visual fields are large enough to involve the far periphery of the retina, but not at 4 months, when measured visual field size is only about 50°, even in normal full-term infants.4 We have shown that kinetic perimetry can be used to measure binocular visual fields in infants with stage 3 ROP. Our results revealed a significant reduction in visual field development at 9 months in the ROP group and a tendency towards a delay at 18 months. Comparison of the location of the stage 3 ROP and the direction of the visual field deficit suggested that there was a relationship between the anatomic location of the stage 3 ROP in early infancy and visual field defects in late infancy. We cannot state if these defects will be lasting or if they just represent a delay in visual field development. Additional studies are being conducted on these children to provide an answer to this question. Key words: retinopathy of prematurity, visualfields,infants

corrected age in months

Fig. 2. Mean visual field size, averaged across the 45°, 135°, 225° and 315° half-meridia for the ROP and no-ROP groups, tested at 4, 9, and 18-months corrected age. Bars indicate ± 1 SEM. Visual field size increases with age in both groups. However, the ROP group showed significantly smaller fields at 9 months than did the noROP group.

the ROP group. However, the two infants who had either PVL or PVL plus IVH were not the ones who showed field deficits. The incidence of other complications related to preterm birth was similar in the ROP and no-ROP groups, and thus also could not be responsible for the differences between groups. The most plausible explanation for the visual field deficits seen in this study is that the peripheral retina is affected in some way by the presence of stage 3 ROP. Although we do not have enough data to offer a definitive confirmation of this hypothesis, critical aspects of our results are consistent with this proposition. First, the only infants who showed visual field deficits were those who had stage 3 ROP in the nasal retina in early infancy. Infants who had stage 3 ROP only in the temporal retina showed no visual field deficits, as would be expected, since our visual field testing was done binocularly, which would preclude detection of deficits in the temporal retina (nasal fields). A second aspect of our data consistent with the hypothesis that the reduced field size in the ROP

Acknowledgments. We thank Karen Bonvalot and Jennifer Bossier for their support in collecting the data. From the Departments of t Psychology, 'Psychiatry, and tOphthalmology, University of Pittsburgh, Pittsburgh, Pennsylvania. Presented at the ARVO Annual Meeting, Sarasota, Florida, May, 1988. This research was supported by NIH grant EY-05804, the Magee-Womens Hospital Research Fund, and the Campbell (Charles Talbot) Foundation. Submitted for publication: August 12, 1988; accepted October 12, 1988. Reprint requests: Beatriz Luna, MS, Department of Psychology, Langley Hall, University of Pittsburgh, Pittsburgh, PA 15260.

References 1. The Committee for the Classification of Retinopathy of Prematurity: An international classification of retinopathy of prematurity. Arch Ophthalmol 102:1130, 1984. 2. Majima A: Studies on retinopathy of prematurity: II. Fundus appearance and ocular functions in cicatricial phase of very low birthweight infants. Jpn J Ophthalmol 21:421, 1977. 3. Tamai A, Iyota K, Ueno H, Noda K, and Kishi S: A follow-up study of retinopathy of prematurity, with special reference to the visual functions of the eyes treated by photocoagulation and/or cryocautery. In Acta: XXIV Int Congr Ophthalmol, Vol. 1, Henkind P, editor. Philadelphia, Lippincott, 1983. 4. Mohn G and van Hof-van Duin J: Development of the binocular and monocular visual fields of human infants during the first year of life. Clinical Vision Sciences 1:51, 1986. 5. Luna B, Dobson V, Carpenter N, Bossier J, and Bonvalot K: Development of peripheral vision in high-risk infants. Infant Behav Develop 11 (Special ICIS issue): 196, 1988. 6. Dobson V, Carpenter N, and Biglan A: Visual acuity development in infants with stage III or greater retinopathy of prematurity. ARVO Abstracts. Invest Ophthalmol Vis Sci 29(Suppl):69, 1988.