Accepted Manuscript

Functional and cognitive vision assessment in children with autism spectrum disorder

Sahithya Bhaskaran, MS, Linda Lawrence, MD, Jeyaseeli Flora, MRSc, VijayalakshmiPerumalsamy, MS

PII: S1091-8531(18)30122-8

DOI: 10.1016/j.jaapos.2018.03.010

Reference: YMPA 2866

To appear in: Journal of AAPOS

Received Date: 9 December 2017

Revised Date: 27 March 2018

Accepted Date: 30 March 2018

Please cite this article as: Bhaskaran S, Lawrence L, Flora J, Perumalsamy V, Functional and cognitivevision assessment in children with autism spectrum disorder, Journal of AAPOS (2018), doi: 10.1016/j.jaapos.2018.03.010.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service toour customers we are providing this early version of the manuscript. The manuscript will undergocopyediting, typesetting, and review of the resulting proof before it is published in its final form. Pleasenote that during the production process errors may be discovered which could affect the content, and alllegal disclaimers that apply to the journal pertain.

https://doi.org/10.1016/j.jaapos.2018.03.010

MAN

USCR

IPT

ACCE

PTED

ACCEPTED MANUSCRIPT

Functional and cognitive vision assessment in children with autism spectrum disorder Sahithya Bhaskaran, MS,a Linda Lawrence, MD, Jeyaseeli Flora, MRSc,a and Vijayalakshmi Perumalsamy, MSa Author affiliations: aAravind Eye Care System, Madurai, Tamil Nadu, India Submitted December 9, 2017. Revision accepted April 1, 2018. Correspondence: Dr. Sahithya Bhaskaran, Aravind Eye Care System, No.1, Anna Nagar, Madurai – 625020, Tamil Nadu, India (email: bsahithya@gmail.com). Word count: 2,655 Abstract only: 186

MAN

USCR

IPT

ACCE

PTED

ACCEPTED MANUSCRIPT

Abstract

Purpose

To assess functional vision in children with autism spectrum disorder (ASD) with a cognitive

visual function battery in addition to standard ophthalmic examinations.

Methods

Subjects were recruited from a school for children with ASD. In addition to a comprehensive

ophthalmic examination, all children underwent cognitive vision assessment at a tertiary

ophthalmological care center in India.

Results

A total of 30 children were included. The distribution of the number of children with mild to

moderate versus severe ASD was nearly equal based on CARS autism scores. The majority of

subjects had normal color vision (16/18), contrast (24), shape discrimination (26), and perception

of directionality (28). Most were not able to identify optical illusions or differentiate tests of

emotions. Ocular pursuits, saccades, and recognition of size differences were often abnormal.

Poor visual closure was noted in (11) subjects. The duration of fixation to Heidi face target was

inversely proportional to the severity of ASD. The study further established that cognitive visual

impairment was present in children with ASD irrespective of their severity of ASD.

Conclusions

All subjects had some form of cognitive visual impairment independent of ASD severity.

MAN

USCR

IPT

ACCE

PTED

ACCEPTED MANUSCRIPT

Autism spectrum disorders (ASD), a group of neurobiological disorders occurring in children,

are generally identified by 30 months of age. The Diagnostic and Statistical Manual of Mental

Disorders (DSM-5) of the American Psychiatric Association defines ASD as “persistent

difficulties with social communication and social interaction” and “restricted and repetitive

patterns of behaviors, activities or interests” present since early childhood, “limiting and

impairing everyday functioning.”1 The Centers for Disease Control and Prevention (CDC)

estimates the prevalence of ASD in children as 1:68 (1 in 42 boys; 1 in 189 girls).2 There is no

prevalence data for ASD in the South India. Approximately one-third of children with ASD have

an intellectual disability.3 Visual impairment, especially cognitive visual impairment in children

with ASD, has yet to be studied in depth. The literature on vision-related problems in children

with ASD is sparse,4 and a comprehensive ophthalmological evaluation with appropriate

interventions should be mandatory before the diagnosis is made. Traditional methods of

ophthalmological examination may be challenging if the child is nonverbal or unable to

understand typical communication. Cognitive vision assessments are not typically performed,

despite the fact that these abnormalities can have a profound effect on communication,

education, and social-emotional development of the child. This study aimed to investigate

whether children with ASD also have cognitive visual dysfunction.

Subjects and Methods

The Institutional Ethics Committee of Aravind Medical Research Foundation approved this

study, which followed the tenets of the Declaration of Helsinki. The parents of the children

provided written informed consent before their children were enrolled. The examinations were

conducted at Aravind Eye Care System, Madurai, South India, during January 2017. All 30

students attending a school for children with ASD were included.

MAN

USCR

IPT

ACCE

PTED

ACCEPTED MANUSCRIPT

All children had been diagnosed with ASD by the same pediatric neurologist and clinical

psychologist team using the validated Childhood Autism Rating Scale (CARS), a clinical rating

scale for trained clinicians to score ASD by direct observation of the child. Scores range from 15

to 60, with 30 being the cutoff rate for a diagnosis of autism. A score of 30-37 indicates mild to

moderate autism; of 38-60, severe autism.5

A detailed history was taken for all children, and all children received comprehensive

ophthalmic evaluation performed by a pediatric ophthalmologist with experience in evaluations

of children with autism (SB). A low-vision rehabilitation specialist experienced in assessments

for persons with developmental disabilities and cognitive visual impairment (JF) and the same

ophthalmologist (SB) performed the functional and cognitive visual assessments on a different

day with the refractive correction in place.

Ophthalmologic Evaluation

Depending on age and ability, visual acuity at near and distance was measured by Teller acuity

cards adopting the standard procedure (Teller Acuity Cards, University of Washington Precision

Vision, Woodstock, IL),6 Lea Symbol 15 Line Pediatric Eye Chart (Good-Lite, Elgin, IL), by

verbally identifying or matching, or Snellen eye test chart by copying or verbal response

binocularly and, if able, monocularly.

All patients underwent fundus examination using indirect ophthalmoscopy.

Accommodation was assessed by dynamic retinoscopy, using the method described by Hunter.7

Ocular alignment was assessed using the Hirschberg test and alternate cover-uncover test. In the

presence of strabismus, a complete orthoptic evaluation was performed. Visual fields were

assessed using the confrontation method. Stereopsis was assessed using Lang Stereotest I (Lang-

Stereo-test AG, Switzerland)8; because the stereoscopic clues were familiar no pretest was

MAN

USCR

IPT

ACCE

PTED

ACCEPTED MANUSCRIPT

performed. Pupillary reaction was tested using a flashlight; anterior segment examination, using

a handheld slit-lamp. Cycloplegia was achieved with 2 drops 10 minutes apart of cyclopentolate

1% and phenylephrine 2.5%; refraction was performed 45 minutes later. A refractive error of ≥

−1.00 DS was categorized as myopia; ≥ +1.0 DS, as hyperopia; and cylinder of ≥0.75, as

astigmatism. Glasses were prescribed with full cycloplegic correction if the above criteria were

met.

Cognitive Visual Assessment

Cognitive visual impairment in its broadest sense refers to a condition leading to

misinterpretation of the visual world either with respect to where things are or concerning what

things are.9 Children who were nonverbal were assessed with the help of the caregiver or parent

and a special educator. Each child was asked to perform all tests demonstrated by the observer or

special educator for at least 2 to 3 trials. Even after the trials, if the child could not perform the

test, the result was recorded as “absent.” If the child did not attempt the test, it was categorized

as “not testable.”

The child’s fixation to a 5" Heidi fixation target (Good-Lite, Elgin, IL) at 30 cm was

observed and the duration of fixation recorded in seconds by the same observer, only once for

each child, so that repeated testing did not affect fixation time from loss of interest or attention.

Color vision was assessed by Ishihara color vision test (Kanehara Shuppan Co Ltd, Bexco,

Haryana), with nonverbal children asked to trace the number or the pattern.

The Hiding Heidi low-contrast test was used to assess contrast sensitivity in nonverbal

children. The test picture and control were moved in opposite directions at the same speed, 30

cm from the child: the result was considered positive if the child fixed on or pointed to the face

picture. Children unable to respond verbally were tested using the Pelli-Robson Contrast

MAN

USCR

IPT

ACCE

PTED

ACCEPTED MANUSCRIPT

Sensitivity Chart (Precision Vision, Woodstock, IL) at a distance of 3 m.10 Contrast sensitivity of

≥5% on Hiding Heidi and ≥2% on Pelli-Robson was considered normal.

Saccadic and pursuit eye movements were assessed using the validated NSUCO (Nova

Southeastern University College of Optometry) oculomotor test. The results were scored based

on four factors: ability, accuracy, head movement, and body movement. The scoring was based

on a 5-point scale, with 5 being highest; a score of ≤3 was a failure, and a score >3 was

considered normal.11

The Lea Mailbox Game was used to assess the visual recognition of line directions. The

child was asked to drop the card into the mailbox slot oriented in horizontal, vertical, and oblique

axes at a distance of 0.5 m.12

The Lea Puzzle (Good-Lite, Elgin, IL) was used to assess the concept of same/different.

The child was asked to match the puzzle pieces in three dimensions using color cues and then on

the flip side of the puzzle. If the child was successful, the 3D puzzle pieces were matched to the

2D symbols. As a modification in our study, a final step was included where the child was asked

to match the black-and-white side of the puzzle with a crowded background, designed by

introducing a collage from varied pictures in the 3D black-and-white puzzle background to

assess the effect of crowding.

The Lea Rectangle Game (Good-Lite, Elgin, IL), a modification of Efron’s rectangles,

was used to assess the child’s ability to appreciate differences in size by matching one set of 5

rectangles according to size and length to a similar set of a different color.13 Figure–ground

discrimination was assessed by displaying several familiar objects cluttered together in a tray and

having the child pick out a specific object.14 Visual closure was assessed by displaying a familiar

picture partially hidden in view and the child was asked to name or match the complete object.14

MAN

USCR

IPT

ACCE

PTED

ACCEPTED MANUSCRIPT

To assess the perception of optical illusions, a 2D illusion showing a rabbit and a duck were

shown on a 22̋ TV screen at a distance of 1 m. The child was given 10-15 minutes to find the

hidden images. The ability to identify both the animals meant the child was able to perceive this

illusion.15 To assess the ability to perceive emotions, the child was presented with emojis 6.5ʺ in

diameter, with four different emotions: happy, sad, angry, and fearful on a 22ʺ screen at one

meter and asked to identify. The ability to identify 2 was considered a positive response.16,17

A questionnaire for characteristics of cerebral visual impairment (CVI) was adapted from

the CVI inventory developed by Dutton.18 Parents and special educators were asked 5 screening

questions, individually or together. A score of 3 or more suggested the presence of CVI. The

modified CVI inventory asked whether the child (1) has difficulty in walking down stairs (for

visual reasons), (2) does not see things that move quickly (eg, small animals), (3) does not see

something that is pointed out in the distance (despite requisite visual acuity), (4) has difficulty

locating an item of clothing in a pile, or (5) has difficulty copying words or pictures.

Statistical Analysis

The statistical analysis was performed with STATA version 14.0 (Stata Statistical Software,

release 14 [2015]; Stata Corp, College Station, TX). The Fisher exact test and correlation test

were used for analysis.

Results

Of the 30 children included in the study, 25 were boys. The mean age at testing was 9.5 years

(range, 5-14 years). No children had undergone prior eye examinations. There were no

comorbidities in any subject. Based on the CARS scoring, 16 children (53%) had mild to

moderate ASD; 14 (47%), severe ASD.

Visual acuity was assessed binocularly with Teller acuity cards in 24 children (80%) and

MAN

USCR

IPT

ACCE

PTED

ACCEPTED MANUSCRIPT

monocularly with Lea symbol or Snellen charts in 6 (20%). See Table 1. Dynamic retinoscopy

was performed on 27 children (90%), with hypo-accommodation in 2 (7%). Accommodation

could not be assessed in 3 children because of inadequate fixation. Of the 30 children, 22 (73%)

were orthophoric; 8 (27%), exotropic; and none, esotropic. Visual fields by confrontation was

normal in 30 children (100%). Stereopsis was assessed in 17 children (56%). Gross stereopsis of

≤550 arcsec was detected in 9 children (30%); 8 children (26%) had no detectable stereoacuity

on Lang 1 stereotest, and 13 (44children could not be tested. Anterior segment, pupillary

reactions, and fundus examination were normal in all 30 children. Cycloplegic refraction was

performed on 60 eyes of 30 children. Emmetropia was present in 48 eyes (80%) eyes; myopia, in

6 eyes (10%) of 3 children; and myopic astigmatism, in 6 eyes (10%) of 3 children.

Functional and Cognitive Visual Assessment

The duration of fixation to 5̋ Heidi fixation target was ≤5 seconds all 30 children. On correlation

analysis, the CARS scores and the duration of fixation to the Heidi target correlated negatively

and was statistically significant at 5% probability (r = −0.44).

Color vision testing was completed in 18 children (60%). Of the 25 boys, 2 had a color

vision deficiency; testing to identify the type of color vision deficiency was not performed.

Twenty-four children (80%) had normal contrast sensitivity; 5 (17%), abnormal. Contrast

sensitivity could not be assessed in 1 child. Saccades were normal in 7 children (24%) and

abnormal in 19 (63%). Pursuit movements were normal in 11 children (57%) and abnormal in 17

(36%).

The mailbox test was performed in all three directions by 28 children (93%). Two

children (7%) could not perform in all three directions. The ability to match the Lea Puzzle was

performed by most of the children in 3D and 2D (Table 2). Two children with mild-to-moderate

MAN

USCR

IPT

ACCE

PTED

ACCEPTED MANUSCRIPT

ASD who matched the 2D and 3D puzzle failed the test when it was introduced with a crowded

background (Table 2). With regard to other cognitive functions (Table 3), 10 children (33%)

were able to match the rectangles; 19 (63%) could not; and 1 child was unable to perform the

test. Figure–ground discrimination was present in 14 children (47%) and absent in 15 (50%).

Visual closure was present in 15 children (50%) children and absent in 11 (36%). Four children

(13%) could identify both animals in the optical illusion, and 21 (70%) could identify only one

animal (an abnormal response). Six children (20%) could identify the difference between emojis,

and 21 (70%) could not (Table 3). Eight children (26%) had a score of at least 3 on the modified

CVI inventory, suggesting the possibility of CVI, which was independent of severity of autism

(Table 4).

Discussion

There is no known prevalence for ASD at this time in the Indian population. In this cohort, there

was a gender disparity greater than reported by the CDC2; however; we included children from a

single school. The distribution of the number of children with mild-to-moderate versus severe

ASD was nearly equal based on CARS scores. This study included children with severe ASD,

whereas earlier studies concentrated on children with high-functioning or mild ASD.3,13,15,17 All

the tests used are standard clinical tests. Previous studies in children with ASD have adopted

similar testing methods,19 although normative data for these methods is lacking. There is no

evidence-based protocol for children with special needs. The assessments for functional and

cognitive visual functions suggested in this study are easy to administer, inexpensive, portable,

and well tolerated.

In this study, 20% of children had refractive errors requiring spectacle correction. Ikeda

and colleagues20 reported a refractive error rate of 29% and a 21% incidence of strabismus;

MAN

USCR

IPT

ACCE

PTED

ACCEPTED MANUSCRIPT

Thankappan and colleagues,21 of 50% and 7.5%, respectively. In the general population in this

age group, the incidence of refractive errors is 2.2%22; of strabismus, 3%-5%.23 Abnormal

response to any one of the cognitive visual tests was considered CVI. All children in our study

had CVI, emphasizing the need for a more comprehensive evaluation in children with ASD than

only standard ophthalmic examination alone so that appropriate educational modifications can be

made.

The duration of fixation to the Heidi 5ʺ face target for children with severe ASD showed

reduced duration of fixation compared with children with mild-to-moderate ASD. Further studies

in children with ASD could provide evidence that this simple, portable, and inexpensive test may

be a useful screening tool for suspected ASD. Ocular saccades, pursuits, size perception, and

visual closure affect reading, academic tasks, and daily activities that may affect behavior.

Three children with mild-to-moderate ASD who performed the 3D puzzle could not

perform the same test when it incorporated a crowded background; no such difference was noted

in children with severe ASD. A larger sample is needed to assess the effect of crowding and its

possible effects on educational interventions.

Most children in our study were not able to identify optical illusions, indicating difficulty

in perceiving visual complexity: 70% did not recognize the facial expressions emoji test.

Whether this result is due to CVI or is rather a characteristic of children on the autism spectrum

deserves further study.

Limitations of this study include the small sample size and the fact that the results may

not be applicable to the general population with ASD. Future studies should include cross-

sectional analysis of a larger population. Also, our study lacked age-matched controls and

normative data for comparison with typically developing children.

MAN

USCR

IPT

ACCE

PTED

ACCEPTED MANUSCRIPT

The number of “not testable events” was low, reinforcing that children with ASD can

cooperate for a wide range of visual function and cognitive vision tests. Augmentative

communication through teachers and parents can improve testability. Rigorous studies are

needed to prove the validity and reliability of the methods employed to identify the ability of

abnormal test results to predict life challenges and to help guide intervention strategies.

MAN

USCR

IPT

ACCE

PTED

ACCEPTED MANUSCRIPT

References

1. American Psychiatric Association. Diagnostic and Statistical Manual of Mental

Disorders. 5th ed. Virginia, USA: American Psychiatric Association; 2013.

2. Christensen DL, Baio J, Van Naarden Braun K, et al; Centers for Disease Control and

Prevention (CDC). Prevalence and characteristics of autism spectrum disorder among

children aged 8 years--Autism and Developmental Disabilities Monitoring Network, 11

Sites, United States, 2012. MMWR Surveill Summ 2016;65:1-23.

3. Boraston ZL, Corden B, Miles LK, Skuse DH, Blakemore SJ. Brief report: perception of

genuine and posed smiles by individuals with autism. J Autism Dev Disord 2008;38:574-

80.

4. Simmons DR, Robertson AE, McKay LS, Toal E, McAleer P, Pollick FE. Vision in

autism spectrum disorders. Vision Res 2009;49:2705-39.

5. Childhood Autism Rating Scale. https://www.special-

learning.com/article/childhood_autism_rating_scale. Accessed June 22, 2018.

6. Droste PJ, Archer SM, Helveston EM. Measurement of low vision in children and

infants. Ophthalmology 1991;98:1513-18.

7. Hunter DG. Dynamic retinoscopy: the missing data. Surv Ophthalmol 2001;46:269-74.

8. Ancona C, Stoppani M, Odazio V, La Spina C, Corradetti G, Bandello F. Stereo tests as a

screening tool for strabismus: which is the best choice? Clin Ophthalmol 2014;8:2221-7.

9. Dutton GN. Cognitive vision, its disorders, and differential diagnosis in adults and

children: know where and what things are. Eye (Lond) 2003;3:289-304.

10. Pelli DG, Robson JG, Wilkins AJ. The design of a new letter chart for measuring contrast

sensitivity. Clin Vis Sci 1988;2:187-99.

MAN

USCR

IPT

ACCE

PTED

ACCEPTED MANUSCRIPT

11. Maples WC, Atchley J, Ficklin T. Northeastern State University College of Optometry’s

oculomotor norms. Journal of Behavioral Optometry 1992:3;143-50.

12. Williams C, Gilchrist ID, Fraser S, et al. Normative data for three tests of visuocognitive

function in primary school children: cross-sectional study. Br J Ophthalmol 2015;99:752-

6.

13. Smith D, Ropar D, Allen HA. Visual integration in autism. Front Hum Neurosci

2015;9:387.

14. Jadue Jadue TC, Figueroa LFO. Habilidades visoperceptuales en niños escolarizados de 7

a 12 años con ambliopía refractiva. Ciencia & Tecnología para la Salud Visual y Ocular

2017;15:31.

15. Allen ML, Chambers A. Implicit and explicit understanding of ambiguous figures by

adolescents with autism spectrum disorder. Autism 2011;15:457-72.

16. Churches O, Nicholls M, Thiessen M, Kohler M, Keage H. Emoticons in mind: an event-

related potential study. Soc Neurosci 2014;9:196-202.

17. Rump KM, Giovannelli JL, Minshew NJ, Strauss MS. The development of emotion

recognition in individuals with autism. Child Dev 2009;80:1434-47.

18. Dutton G, Bax M. Visual Impairment in Children Due to Damage to the Brain. Clinics in

Developmental Medicine 186. London: MacKeith Press; 2010.

19. Coulter RA, Bade A, Tea Y, et al. Eye Examination Testability in Children with Autism

and in Typical Peers. Optom Vis Sci 2015;92:31-43.

20. Ikeda J, Davitt BV, Ultmann M, Maxim R, Cruz OA. Brief report: incidence of

ophthalmologic disorders in children with autism. J Autism Dev Disord 2013;43:1447-

51.

MAN

USCR

IPT

ACCE

PTED

ACCEPTED MANUSCRIPT

21. Thankappan B, Sidhan N, Aparna KS. Ocular disorders in children with autism in special

schools. J Med Sci Cin Res 2017:05:07:25199-203.

22. John DD, Paul P, Kujur ES, David S, Jasper S, Muliyil J. Prevalence of refractive errors

and number needed to screen among rural high school children in southern India: a cross-

sectional study. J Clin Diagn Res 2017;11:NC16-NC19.

23. Kothari M. Clinical characteristics of spontaneous late-onset comitant acute

nonaccommodative esotropia in children. Indian J Ophthalmol 2007;55:117-20.

MAN

USCR

IPT

ACCE

PTED

ACCEPTED MANUSCRIPTTable 1. Visual acuity at distance

Visual acuity test Visual acuity, Snellen equivalents

No.

Snellen chart or Lea symbols optotypes

6/6 4 6/12 2

Teller Acuity Cards 6/9.5 6 6/15 12 6/36 4 6/60 1 5/60 1

MAN

USCR

IPT

ACCE

PTED

ACCEPTED MANUSCRIPTTable 2. Lea Puzzle responses

Puzzle Response Autism spectrum disorder, no. (%)

Mild-moderate Severe ASD 3D Color Presenta 16 (100) 12 (86)

Absentb 0 2 (14) Black & white Present 16 (100) 12 (86)

Absent 0 2 (14) 2D background Present 15 (94) 11 (79)

Absent 1 (6) 3 (21) Crowded background Present 13 (72) 12 (86)

Absent 3 (18) 2 (14) aAble to match shapes. bUnable to match shapes.

MAN

USCR

IPT

ACCE

PTED

ACCEPTED MANUSCRIPTTable 3. Cognitive visual function (N = 30 children)

Characteristics Present,a Absent,b Not testable,c

no. (%) no. (%) no. (%) Saccades 7 (24) 19 (63) 4 (13) Pursuits 11 (36) 17 (57) 2 (7) Mailbox 28 (93) 2 (7) 0 Rectangle game 10 (33.3) 19 (63.3) 1 (3) Figure–ground discrimination 14 (47) 15 (50) 1 (3) Visual closure 15 (50) 11 (36) 4 (14) Optical illusion 4 (13) 21 (70) 5 (17) Recognize emotions 6 (20) 21 (70) 3 (10) aAble to perform test. bUnable to perform test or failed test. cDid not attempt test.

MAN

USCR

IPT

ACCE

PTED

ACCEPTED MANUSCRIPTTable 4. Association between Childhood Autism Rating Scale score and cognitive visual abilities

Variables P valuea Modified CVI inventory 0.192 Stereopsis 0.896 Lea puzzle (3D black and white) 0.209 Lea puzzle (2D) 0.315 Lea crowded 0.999 Lea puzzle 3D color 0.209 Optical illusion 0.158 Visual closure 0.064 Saccades 0.427 Pursuits 0.846 Rectangle game 0.94 Mailbox 0.734 Emotions 0.312 Figure–ground discrimination 0.170

CVI, cerebral visual impairment. aFischer exact test.