Tendency oriented perimetry in children with ocular abnormalities

TENDENCY ORIENTED PERIMETRY IN CHILDREN WITH OCULAR ABNORMALITIES

SANDRA M. BROWN and JOSE MORALES

Department of Ophthalmology and Visual Sciences, Texas Tech University Health Sciences Center, Lubbock, TX, USA

Abstract

Twelve healthy children with congenital or acquired abnormalities of the globe or optic nerve of one or both eyes underwent automated static perimetry testing using the Octopus 1-2-3 perimeter and the Octopus TOP-32 program. Each eye was tested at least twice during one or more sessions. Children with unilateral and/or well-defined anatomic abnormalities displayed good inter-test consistency and good correlation between the predicted and actual visual field defects. Children with bilateral and/or diffuse abnormalities had poor inter-test consistency and poor correlation between the predicted and actual defects. Based on this experience, the authors conclude that the visual fields of pediatric patients with progressive retinal or optic nerve lesions that may cause diffuse depression should be interpreted cautiously, and should be repeated in several sessions to assess consistency. They suggest that therapeutic decisions should not be based solely on the visual field findings unless there is an unequivocal and reproducible shift from normal to abnormal.

Introduction

Tendency oriented perimetry (TOP) decreases testing time by as much as 80% compared to standard full-threshold bracketing strategies. This is primarily due to the fact that significantly fewer stimuli are required compared to full-bracketing strategies.1-3 This makes TOP attractive for use in children, whose performance on full-threshold perimetry is often compromised by fatigue and boredom. In a previous study,4 we assessed the feasibility of TOP in normal children aged six through 12 years using the Octopus TOP-32 program. We found a specificity (normal test in a normal child) of 78%. The purpose of the present study was to assess the utility of TOP in children with discrete, non-progressive defects of the globe or optic nerve.

Address for correspondence: Jose Morales, MD, 3601 Fourth Street STOP 7217, Lubbock, TX 794307217, USA. Email: jose.morales@ttuhsc.edu

Perimetry Update 2002/2003, pp. 51–68

Proceedings of the XVth International Perimetric Society Meeting, Stratford-upon-Avon, England, June 26–29, 2002

edited by David B. Henson and Michael Wall

© 2004 Kugler Publications, The Hague, The Netherlands

Methods

A consecutive series of 12 patients between the ages of six and 12 years was collected from the pediatric ophthalmology practice of one author (SMB). These patients had anatomical abnormalities of the eye(s) or optic nerve(s). In addition to a complete ophthalmological examination, each child underwent automated static perimetry (ASP) testing on the Octopus 1-2-3 perimeter using the TOP-32 program. The details of training and testing have been described previously.4 The testing sequence was not identical for all children and the rationale for the chosen sequence is as follows:

Initially, children with one normal eye were tested normal-abnormal-normal-abnor- mal so that the normal eye served as both training and control. In patient 4, the repeat test on the abnormal eye (test #4) could not be performed due to fatigue. As we learned more about the fatigue effect present by test #4, we modified the sequence to be normal-abnormal-abnormal-normal. In patient 1, this was the planned sequence, but the repeat test on the normal eye (test #4) was deferred due to excellent cooperation and high consistency between the two abnormal eye tests. Children with bilateral, symmetric disease were all tested right-left-right-left. Children with bilateral, asymmetric disease were tested less affected-more affected-less affected-more affected.

Results

Table 1. Summary of patient information and visual field indices

Visual fields are listed in the order performed, e.g., OD3 means that the third test in the sequence was performed on the right eye (representing the second time the right eyes was tested). MD: mean deviation; LV: loss of variance; Bebié: Bebié curve score: N: normal; Bl: borderline; MO: mildly outside of normal limits; Ab: abnormal

For all the figures that follow, the visual fields in the left column correspond to the left eye; the visual fields in the right column correspond to the right eye. The upper row of visual fields was performed first for each eye; the lower row was performed second. Not all patients were tested four times. In order to maintain a consistent rightleft arrangement, photographs of the right eye are placed on the left (the reverse of the customary presentation). Not all patients had photographs of adequate quality for reproduction.

Fig. 1. Patient TM (#1). An 11-year-old boy had blunt trauma OD with resulting iridodialysis and commotio of the macula. The commotio resolved, leaving extensive macular pigmentary irregularity. Best-corrected vision was 20/50. ASP showed an abnormal field OD with generalized depression and ‘migrating’ scotomas.

Fig. 2. Patient KBu (#2). A nine-year-old girl with a new diagnosis of neurofibromatosis was found to have decreased vision OD. Best-corrected vision was 20/200. A +1 right APD was present. The right optic disc showed +2 pallor with temporal excavation and the left optic disc was normal. MRI demonstrated a lesion consistent with a glioma of the right optic nerve. ASP showed significant generalized depression OD but without specific defects.

Fig. 3. Patient HS (#3). A six-year-old girl failed a school screening. Best-corrected vision was 20/25 OU. Both maculas had blunted reflexes and pigmentary graininess. Color vision was normal and there were no iris transillumination defects. ASP suggested a dense superior scotoma OD during the first test that decreased significantly with learning effect by the second examination. Both eyes showed non-specific changes.

Fig 4. Patient DP (#4). An 11-year-old boy had persistent fetal vasculature (persistent hyperplastic primary vitreous) OD. He had a small, eccentric lens opacity where the stalk attached to the posterior lens capsule. This stalk obscured a mildly dysplastic optic nerve. Best-corrected vision was 20/150. ASP showed mild, non-specific changes in both eyes.

Fig 5. Patient DT (#5). A nine-year-old boy had bilateral high myopia with macular hypoplasia OU. Bestcorrected vision was 20/40 OU. ASP (performed with contact lens correction) showed generalized depression and constriction of the field OU.

Fig. 6. Patient KBl (#6). A seven-year-old boy failed a school vision screening. Best-corrected vision was 5 feet/200 OD and 20/20 OS. Color testing with Ishihara plates demonstrated zero of 12 plates OD correct and 12 of 12 plates correct OS. A +1 right APD was present. The right optic nerve was +3 pale and the left optic nerve was +2 pale; the right macula was blunt with pigmentary changes. After work-up, a diagnosis of autosomal dominant optic atrophy was made. ASP was unreliable and demonstrated severe generalized depression of the threshold sensitivity in both eyes without specific patterns.

Fig. 8. Patient CC (#8). A ten-year-old boy had bilateral iris and choroidal colobomas. The right coloboma was inferior to the optic nerve and did not involve the macula; the left coloboma affected the inferior pole of the optic nerve but did not involve the macula. ASP showed dense scotomas corresponding well in location and extent to the colobomas, and were fairly reproducible on repeat testing.

Fig. 10. Patient KF (#10). A ten-year-old girl had encephalitis at the age of six years. Vision was 20/20 and 20/25. The right optic nerve showed diffuse moderate pallor and the left optic nerve showed mild temporal pallor. ASP demonstrated generalized depression in both eyes and high variability between fields.

Fig 12. Patient TB (#12). A seven-year-old boy suffered blunt trauma OD with a hyphema and diffuse macular commotio. He developed a small, full thickness hole in the macula just superior to the fovea. Vision was 2/200 OD and 20/20 OS. ASP demonstrated a discrete, inferior, moderately dense paracentral scotoma in a location consistent with the anatomical defect.

66 S.M. Brown and J. Morales

Discussion

ASP has several advantages over kinetic perimetry: quantitative analysis that allows for detailed longitudinal follow-up, removal of tester bias, and reproducibility of technique. Automated threshold perimetry has not been used in studies reporting on visual field defects in children with well-defined anatomic abnormalities, such as peripheral retinal ablation,5-10 and has anecdotally been considered inaccurate due to the fatigue and boredom that develop during a 16-20 minute testing session (per eye). We have shown that ASP using TOP reduces test time to 2-3.5 minutes (per eye), and allows a 100% test completion rate with a 78% specificity in normal children aged six to 12 years.4 Encouraged by these results, we performed ASP on children with anatomic abnormalities of the eye(s) or optic nerve(s).

Analysis of our series of 12 patients shows that they fall into two broad categories. Children with retinal abnormalities such as colobomas, staphylomas, or choroidal ruptures can map a visual field defect consistent with the anatomical defect, with a good degree of reproducibility and accuracy (i.e., consistent with the predicted visual field defect based on retinal anatomy). These lesions have more defined boundaries between markedly abnormal, and nearlyor fully-normal retinas. Children with more diffuse problems, such as optic nerve atrophy, macular pigmentary disturbance, or macular hypoplasia, will show non-specific generalized depression. In this second group, children with one normal eye can provide their own control to prevent a falsepositive result based on poor test performance. In children who are bilaterally affected, it is not possible to rule out bilateral false-positive tests, or to accept the mean deviation or loss of variance parameters as true measures of retinal sensitivity. Fluctuation in these variables over time should not be over-interpreted as indicating improvement or progression of disease in these patients, unless there is an unequivocal trend that is corroborated by declining visual acuity or objective worsening on examination.

It is important to keep in mind that there are no normal values for interpretation of visual field abnormalities in children because, until recently, reliable ASP was not possible in this population. Therefore, any interpretation that includes the normal values for adults to determine abnormality is inherently mistaken. In adults, the assumption is that the population-based normal values for ASP parameters will deteriorate with increasing age. The other side of this assumption is that normal values are better (e.g., mean sensitivity is higher, mean deviation is lower) with younger age. However, this trend cannot be extrapolated from young adults into the pediatric population, where testability becomes influenced by maturity and attention span. There is a great need for the development of practical, ‘clinical’ normal values for automated perimetry in children. This study is currently underway in our practice.

Table 2 outlines the considerations we found useful for the interpretation of visual fields in children. Performing ASP twice in each eye is possible with TOP perimetry, because the test is much shorter than a standard bracketing strategy and gives us the advantage of identifying the learning and fatigue effects during the same session. Sometimes a further learning effect can be observed during later sessions (patient EB). We have learned that patients with one normal eye are best tested normal-abnormal- abnormal-normal. If the final test on the normal eye is significantly depressed, this indicates that a fatigue effect has developed, which may also affect the accuracy of the second test on the abnormal eye.

Table 2. Guidelines on interpretation of TOP-32 visual fields in children

•test both eyes twice and choose the best (least depressed) field for each eye for interpretation

•false-positive borderline fields will improve with learning; true-positive borderline fields will remain abnormal and show better mapping of defects

•the field indices (MD, LV) and the Bebié curve are depressed substantially with poor performance, and are more resilient when there are specific abnormalities or scotomas

•shifting or ‘dancing’ scotomas are usually not real, but may be seen more often in children with conditions expected to cause generalized central depression, and may be a reflection of erratic fixation

•borderline or moderately abnormal fields that do not correlate with the clinical picture should be repeated during a separate session before interpretation

•the false-positive rate is higher than the false-negative rate; a normal field is most likely truly normal

•some children (and adults) perform poorly on visual field testing and the value of the test will be limited under any circumstances.

•our results and recommendations apply only to neurologically normal children without intellectual delay

We believe that ASP with TOP is a promising method for documenting visual field abnormalities in children aged from six to 12 years. Repeat testing during the same session is possible due to the short duration of each examination, and gives a good indication of consistency and the likelihood of false-positive results. In our cohort, children with non-progressive, discrete retinal defects performed better than children with diffuse macular defects or optic nerve disease, who often displayed non-specific patterns of generalized depression or ‘dancing’ scotomas.

Acknowledgment

The authors and their families have no commercial or proprietary interest in the equipment or software described, and have no financial interest in Interzeag. Dr Brown has never received money from Interzeag. Dr Morales has received speaker honoraria from Interzeag.

References

1.Morales J, Weitzman ML, Gonzalez de la Rosa M: Comparison between tendency oriented perimetry (TOP) and Octopus threshold perimetry. Ophthalmology 107:134-142, 2000

2.Martinez Piñero A, Rodriguez J, Serrano M, González de la Rosa M: Reproducibility and accuracy of tendency oriented perimetry (TOP) (ARVO Abstract 95). Invest Ophthalmol Vis Sci 39:S23, 1998

3.Martinez A, Pareja A, Mantolan C, Sanchez M et al: Results of tendency oriented perimetry (TOP) in a normal population (JERMOV abstract). Vision Res 36:153, 1996

4.Morales J, Brown SM: The feasibility of short automated static perimetry in children. Ophthalmology 108:157-162, 2001

5.Cryotherapy for Retinopathy of Prematurity Cooperative Group: Effect of retinal ablative therapy for threshold retinopathy of prematurity: results of Goldmann perimetry at the age of 10 years. Arch Ophthalmol 119:1120-1125, 2001

6.Quinn GE, Fea AM, Minguini N: Visual fields in 4- to 10-year-old children using Goldmann and double-arc perimeters. J Pediatr Ophthalmol Strabismus 28:314-319, 1991

7.Quinn GE, Dobson V, Hardy RJ, Tung B et al: Visual fields measured with double-arc perimetry in eyes with threshold retinopathy of prematurity from the cryotherapy for retinopathy of prematurity trial. The CRYO-Retinopathy of Prematurity Cooperative Group. Ophthalmology 103:1432-1437, 1996

8.Quinn GE, Miller DL, Evans JA, Tasman WE et al: Measurement of Goldmann visual fields in older children who received cryotherapy as infants for threshold retinopathy of prematurity. Arch Ophthalmol 114:425-428, 1996

9.Fetter WP, Van Hof-Van Duin J, Baerts W, Heersema DJ et al: Visual acuity and visual field development after cryocoagulation in infants with retinopathy of prematurity. Acta Paediatr 81:25-28, 1992

10.Kremer I, Nissenkorn I, Lusky M, Yassur Y: Late visual field changes following cryotherapy for retinopathy of prematurity stage 3. Br J Ophthalmol 79:267-269, 1995