Journal of Pediatric Ophthalmology and Strabismus

Original Article 

Interobserver Variability for Measurement of Grating Acuity in Preverbal and Nonverbal Children Using Lea Grating Paddles

Ajinkya Vivekrao Deshmukh, DNB; Uppal Vipul Gandhi, DNB; Pankti Morakhia, BSc Optom; Ashik Mohamed, MBBS, MTech, PhD; Akshay Badakere, DOMS; Ramesh Kekunnaya, MD

Abstract

This article has been amended to include factual corrections. To read the erratum, click here. The online article and its erratum are considered the version of record.

Purpose:

To study interobserver variability for measurement of grating acuity in preverbal and nonverbal children with developmental delay using Lea Grating Paddles (Lea Hyvärinen).

Methods:

In this prospective study, preverbal patients younger than 3 years and older nonverbal patients (with developmental delay) were included. Those with poor neck holding were excluded. Binocular and monocular visual acuity assessment was done simultaneously using Lea Grating Paddles by two examiners who were blinded from each other's results.

Results:

Thirty-one children were tested (19 [61.29%] males and 12 [38.71%] females). The mean age was 28.5 months (range: 4 to 44 months). A total of 92% of the observations showed interobserver agreement of 0.5 octave or better in binocular testing. Results for monocular testing showed interobserver agreement (0.5 octaves or better) in 95.65% and 95.23% of right and left eyes, respectively. Combined results showed agreement better than 0.5 octaves in 94.20%. The mean binocular visual acuity was 2.07 and 2.14 cycles per degree (cpd) for the first and second observer, respectively (P = .59). Similarly, the mean monocular visual acuity for the right and left eyes was 0.98 and 0.97 cpd for observer 1 and 0.89 and 0.79 cpd for observer 2, respectively (P = .91 and .14, for the right and left eyes, respectively).

Conclusions:

Interobserver variability for grating acuity measurement using Lea Grating Paddles is minimal, signifying it to be a reliable and reproducible alternative to conventional preferential looking testing procedures.

[J Pediatr Ophthalmol Strabismus. 2020;57(5):305–308.]

Abstract

This article has been amended to include factual corrections. To read the erratum, click here. The online article and its erratum are considered the version of record.

Purpose:

To study interobserver variability for measurement of grating acuity in preverbal and nonverbal children with developmental delay using Lea Grating Paddles (Lea Hyvärinen).

Methods:

In this prospective study, preverbal patients younger than 3 years and older nonverbal patients (with developmental delay) were included. Those with poor neck holding were excluded. Binocular and monocular visual acuity assessment was done simultaneously using Lea Grating Paddles by two examiners who were blinded from each other's results.

Results:

Thirty-one children were tested (19 [61.29%] males and 12 [38.71%] females). The mean age was 28.5 months (range: 4 to 44 months). A total of 92% of the observations showed interobserver agreement of 0.5 octave or better in binocular testing. Results for monocular testing showed interobserver agreement (0.5 octaves or better) in 95.65% and 95.23% of right and left eyes, respectively. Combined results showed agreement better than 0.5 octaves in 94.20%. The mean binocular visual acuity was 2.07 and 2.14 cycles per degree (cpd) for the first and second observer, respectively (P = .59). Similarly, the mean monocular visual acuity for the right and left eyes was 0.98 and 0.97 cpd for observer 1 and 0.89 and 0.79 cpd for observer 2, respectively (P = .91 and .14, for the right and left eyes, respectively).

Conclusions:

Interobserver variability for grating acuity measurement using Lea Grating Paddles is minimal, signifying it to be a reliable and reproducible alternative to conventional preferential looking testing procedures.

[J Pediatr Ophthalmol Strabismus. 2020;57(5):305–308.]

Introduction

Assessment of visual acuity is the first and most important clinical measure in ophthalmic practice. Commonly used methods such as Snellen charts or Sloan optotypes are dependent on recognition acuity and not useful in the preverbal pediatric age group.

Visual acuity in these individuals can be assessed and quantified using the grating visual test or the preferential looking acuity test1 and is generally successful.2 The preferential looking test using the Teller Acuity Cards (Vistech) is widely used in clinical practice.3 More recently, the Lea Grating system of preferential looking tests (Lea Hyvärinen) has been developed to assess visual acuity, in which handheld paddles with printed grating lines are used.4,5 Both Teller Acuity Cards and Lea Grating Paddles were found to be efficient and comparable.6 However, Lea Grating Paddles are convenient, quick, smooth, and more engaging, especially for inattentive or fussy patients during monocular testing.6

After an extensive literature search (PubMed, Google Scholar, and Science Direct) we found that, to the best of our knowledge, no study has been conducted to determine interobserver variability of the Lea Grating paddles. Through this study, we aimed to determine the interobserver variability and hence the reliability of Lea Grating Paddles in assessing the grating acuity in preverbal and nonverbal (with developmental delay) children.

Patients and Methods

In this prospective study, preverbal patients younger than 3 years and older nonverbal patients (with developmental delay) were included. Those with poor neck holding were excluded from the study. Thirty-one consecutive patients visiting our outpatient department of the pediatric ophthalmology and strabismus services were recruited. Of the 31 children, 29 fell into the preverbal age group (younger than 36 months). The remaining two children were diagnosed as having delayed speech and were 44 and 43 months old, respectively. Patients included had a wide spectrum of visual and/or ocular pathologies, including various types of refractive errors, amblyopia, comitant and incomitant strabismus, surgical aphakia, pseudophakia, retinopathy of prematurity, and retinal dystrophy.

The study adhered to the tenets of the Declaration of Helsinki and was approved by the institutional review board. Informed consent was obtained from the parent/guardian of each child.

The assessment of visual acuity was performed using the Lea Grating Paddles. The available procedural manual was used as a reference guideline for performing this test. The Lea Grating Paddles consisted of three grating paddles with black and white stripes of varied width on both sides, and one uniform gray color control paddle (Figure 1). The test was performed by holding a striped paddle behind the control paddle and quickly moving them apart, holding them at eye level side by side (Figure 2). The measurement is based on observing the child's eye movements toward the striped area in preference to the gray one, when the grating paddles are presented to the child. A dedicated room with solid color painted walls was chosen to perform the test to make sure there were no evident distractions for the child undergoing the test. The room had standard photopic illumination of 186 lux units.

Lea Grating Paddles (Lea Hyvärinen).

Figure 1.

Lea Grating Paddles (Lea Hyvärinen).

A child sitting comfortably in the lap of his father and the examiner presenting the Lea Grating Paddles (Lea Hyvärinen). The other examiner is standing just behind the first examiner and observing from above the right shoulder of the first examiner at the same eye level to avoid possible parallax error.

Figure 2.

A child sitting comfortably in the lap of his father and the examiner presenting the Lea Grating Paddles (Lea Hyvärinen). The other examiner is standing just behind the first examiner and observing from above the right shoulder of the first examiner at the same eye level to avoid possible parallax error.

The child was seated in the parent's/guardian's lap facing the examiners. One examiner performed the test sitting in front of the patient and the other examiner stood just behind the first examiner and observed from above the right shoulder of the first examiner at the same eye level to avoid possible parallax error. Simultaneous observation by the examiners was chosen over separate sequential testing by both in view of the poor attention span of these children leading to improper response on repeated testing. Both examiners wore neutral or solid color clothing during the test to prevent drifting of the child's attention away from the object of interest.

Keeping in mind the possibility of the child losing interest or becoming irritable on occlusion of one eye during monocular testing, binocular testing was performed first. If the child was cooperative for both monocular and binocular testing, only the monocular testing was considered for the analysis. Pediatric opaque eye patches were used for occluding one eye while performing the test monocularly; however, having the parent/guardian cover the child's eye with his or her hand was opted for when patching was not successful.

Bottles, pacifiers, and a variety of illuminated or unilluminated colorful toys were incorporated to help gain and maintain the child's attention and comfort prior to the actual testing with the grating paddles. The test was performed at a distance of 57 cm. For children who did not respond to presentations at 57 cm, testing was performed at 28 cm. Those who performed best at 57 cm were tested for 84 cm and for better acuity measurements. Accordingly, the acuities were modified as per prescribed guidelines in the testing manual. Both examiners recorded their findings separately and were blinded from each other's test results throughout. Examiners documented their own data sheets of the recorded binocular and monocular visual acuities of all patients.

Statistical analysis was done by a statistician using the SPSS software, version 20 (SPSS, Inc). For calculation of mean and standard deviation of visual acuities, values obtained in cycles per degree (cpd) were converted to logarithm of the minimum angle of resolution (logMAR) units. Interobserver differences for monocular and binocular acuities were calculated to the nearest 0.5 octave (the approximate difference in spatial frequency between adjacent acuity cards). The paired t test was used to analyze the results from the two examiners.

Results

Thirty-one children were tested, of whom 19 (61.29%) were male and 12 (38.71%) were female. The mean age was 28.5 months (range: 4 to 44 months). Five (16.13%) were infants, 11 (35.48%) were between 12 and 24 months, 13 (42.94%) were between 24 and 36 months, and 2 (6.45%) were older than 36 months. Five children (16.13%), including the 2 children older than 36 months, had developmental delay.

Patients comprised various refractive diagnoses. All except 2 children were found to have some form of ocular ailment, the most common being ocular deviation (n = 15). Six children had infantile esotropia, 6 had intermittent exotropia, 2 had refractive accommodative esotropia, and 1 had Duane retraction syndrome type 1. Of the 2 patients with infantile esotropia, 1 had associated dissociated vertical deviation and inferior oblique overaction. Three children were aphakic (following cataract extraction surgery). Five were pseudophakic, of whom 1 had secondary exotropia and 1 had esotropia and glaucoma. Four children were born prematurely and were known to have retinopathy of prematurity in the past. One child was diagnosed as having Leber congenital amaurosis and one had a limbal dermoid.

Monocular testing was possible in 23 children, but in 2 of them, the left eye tests were not reliable and thus they were excluded from the statistical analysis. Therefore, a total of 44 observations were obtained from each observer for the monocular testing. For the remaining 8 children, monocular testing was not reliable for either eye, so only the binocular test observations (8 from each observer) were included in the statistical analysis.

Table 1 compares mean visual acuities for binocular and monocular tests by both observers. The mean binocular visual acuity testing was 2.07 cpd for observer 1 and 2.14 cpd for observer 2. The paired t test showed no significant difference (P = .59) between the two observers. Similarly, the mean for monocular visual acuity testing for the right and left eyes was 0.98 and 0.97 cpd for observer 1 and 0.89 and 0.79 cpd for observer 2, respectively. The P values on the paired t test showed no statistically significant difference (P = .91 and .14 for the right and left eyes, respectively).

Test Results for Binocular and Monocular Testing by Two Observers

Table 1:

Test Results for Binocular and Monocular Testing by Two Observers

For monocular testing, interobserver agreement of 0.5 octave or better was seen in 95.65% and 95.23% for the right eye and left eyes, respectively. For binocular results, 92% of the observations showed agreement of 0.5 octave or better between the two observers. While comparing the overall results for binocular and monocular testing, interobserver agreement of 0.5 octave or better was seen in 94.20%. One hundred percent of the test results (binocular and monocular) showed interobserver agreement of one octave or better.

Discussion

In adults, visual acuity is measured as “recognition acuity,” which uses standard line tests and cannot be used in examining infants and children with developmental delay.5 Visual acuity in these individuals is measured with grating acuity tests. In this grating acuity test, the infant or child detects the presence of parallel lines of decreasing width, a task simpler than recognizing optotypes. When a striped pattern is presented in front of an infant simultaneously with a gray surface of the same size and luminance, the infant is likely to look at the striped pattern because there is more to see than on a gray surface. Conventionally, the Teller Acuity Cards are used to test the preferential looking acuity. These comprise seventeen 25.5 × 55.5 cm cards, each of which has an approximately 4-mm diameter peephole at the center. The more recently developed Lea Grating Paddle system uses handheld paddles to present the gratings instead of the plates. The design and size of the paddles make the test more ergonomic.

A study by Yudcovitch et al6 comparing Lea Grating Paddles and Teller Acuity Cards showed that not only did the testing move quickly and smoothly, but the infants and toddlers were often more attentive to the Lea Grating Paddles than to the Teller Acuity Cards. Moreover, they also found Lea Grating Paddles and Teller Acuity Cards to be comparable and equally efficient in assessment of the visual acuity in this age group. Our study also found Lea Grating Paddles to be convenient, especially when working with inattentive or fussy infants, particularly during monocular testing. In addition, the advantages of lower cost and better portability of the Lea Grating Paddles are desirable features for the clinician. During the course of the study and with more experience, we realized that the use of these paddles can also be extended to get an idea of the child's visual field. Depending on the direction in which the grated paddles are presented to the child from behind the gray paddle and the response obtained, one can estimate whether the child is able to appreciate the target better in one particular direction of field.

Despite the commercial availability of the Lea Grating Paddles for more than a decade, no study has been conducted to report interobserver variability/reliability of this test procedure while assessing the grating visual acuity in the preverbal pediatric age group.

Our study group consisted of five children with developmental delay. Although a formal statistical analysis could not be done due to the smaller sample size of this particular group, this group also had an interobserver agreement of 0.5 octaves or better, which was comparable to the children without any developmental delay but with other ocular conditions. Moreover, all children satisfying the inclusion criteria were included and no exclusions were made depending on the ocular and/or systemic diagnosis. Therefore, these results can be generalized to a larger spectrum of children with varied diagnosis in the preverbal age group.

Results showed greater than 92% of interobserver agreement for both binocular and monocular testing. The mean values obtained by both observers were similar and proved that Lea Grating Paddles have a good interobserver reliability. Interobserver variability for grating acuity measurement using Lea Grating Paddles is minimal, signifying it to be a reliable and reproducible alternative to conventional preferential looking testing procedures such as Teller Acuity Cards.

References

  1. Mayer DL, Dobson V. Visual acuity development in infants and young children, as assessed by operant preferential looking. Vision Res. 1982;22(9):1141–1151. doi:10.1016/0042-6989(82)90079-7 [CrossRef]
  2. Moore BD. Eye Care for Infants and Young Children.Butterworth-Heinemann; 1997:77.
  3. Quinn G, Berlin J, James M. The Teller acuity card procedure: three examiners in a clinical setting. Ophthalmology. 1993;100:488.
  4. Vistech Inc. Teller Acuity Cards Handbook. Revised. Vistech Inc; 1986.
  5. Lea Gratings. Test Instructions for Precision Vision Products. Precision Vision. https://www.precision-vision.com/
  6. Yudcovitch L, Linden ME, Maeda J, Shore N. An evaluation of infant visual acuity using Lea Grating Paddles and Teller Acuity Cards. J Optom Vis Dev. 2004;35(3/4):224–229.

Test Results for Binocular and Monocular Testing by Two Observers

ObserverMean Score (cpd)SD (Octaves)P (Paired t Test)
Binocular observer 12.071.34.59
Binocular observer 22.141.28
Monocular right eye observer 10.981.96.91
Monocular right eye observer 20.971.86
Monocular left eye observer 10.891.61.14
Monocular left eye observer 20.791.68
Authors

From L.V. Prasad Eye Institute, Hyderabad, India.

The authors have no financial or proprietary interest in the materials presented herein.

Correspondence: Uppal Vipul Gandhi, DNB, L.V. Prasad Eye Institute, Kallam Anji Reddy Campus, L.V. Prasad Marg, Banjara Hills, Hyderabad 500034, Telangana, India. Email: dr.uppalg@gmail.com

Received: December 21, 2019
Accepted: April 20, 2020

10.3928/01913913-20200701-02

Sign up to receive

Journal E-contents