City Research Online

Abstract

Background and Purpose

Several tests of visual functions have been developed and validated for typically developing children but very few have been validated in children with special educational needs (SEN). Cerebral visual impairment (CVI) which is a rising cause of paediatric vision impairment globally is a neurological condition categorized under the umbrella of SEN. Children with CVI are likely to have additional developmental delays in areas such as motor, speech, communication and cognition which makes assessment of visual functions challenging. In addition, other factors may also contribute to the variability of visual functions including location/extent of brain damage, overall development, seizures and medications used by these children. Measuring visual functions is therefore important in this population to understand how well a child performs visually and to understand the benefits of visual rehabilitation. Equally important is availability of visual function tools that are repeatable and can easily be carried out on this population. Children with CVI can present anywhere across low to high visual functioning and therefore validating the clinical tools to measure their visual functions is useful. The current study focused on validating clinical tools for the most commonly measured visual function, i.e., visual acuity (VA) and the parameter that relates closely with functional vision, i.e., contrast sensitivity (CS) in children with CVI by comparing the limits of agreement (LoA) between different tests of VA and CS and determining their repeatability indices. The association of visual functions with other factors such as: brain imaging findings, developmental quotient/age, seizure history/activity and functional vision score were also studied.

Methodology

Children aged 6 months to 7 years with a confirmed diagnosis of CVI by a paediatric neurologist were recruited primarily from a paediatric neurology clinic and some from a vision rehabilitation unit. Demographic and clinical information were elicited from the parents/caregivers and/or extracted from the medical records. Visual acuity was assessed using Teller acuity cards-II (TAC-II) and Peekaboo Vision application (PV app) and CS was assessed using the Hiding Heidi low contrast face test (HH cards) and Ohio contrast cards (OCC). Seizure history as reported by the parents/caregivers and the activity using the electroencephalography findings was noted. Developmental quotient was assessed using the Denver Developmental Screening Test-II (DDST-II) by a clinical psychologist. The brain imaging findings were scored by the neuro-radiologist based on the magnetic resonance imaging (MRI) scanned films and the children were classified as having mild, moderate, and severe damage. The functional vision assessment was measured using the CVI range instrument and the children were categorized into phase I, II and III indicating low, moderate and high functioning CVI respectively. Intra-observer repeatability was carried out within a test-retest duration of 1 month. Chronologically age-similar typically developing children were recruited as the control group.

Results

Demographic and clinical characteristics of children with CVI
A total of 111 children with CVI with a mean age of 3.00±1.85 years (7 months to 7 years, 70.2% males) were recruited in the study. Neonatal hypoglycaemic brain injury was noted to be the most common aetiology of CVI (47.6%). The brain imaging findings revealed that a majority of the children were categorised as having severe damage (66.6%). The most common parent-reported visual concerns included difficulty in recognizing faces (45.4%) and maintaining eye contact (34.5%) in children ≤3 years. While maintaining eye contact (25%) remained as a concern for older children >3 years followed by missing objects in the lower/side field (17.5%).

Visual functions in children with CVI
The testability rates were found to be highest for TAC-II (95.4%) and HH cards (91.8%) in the VA and CS tests respectively. The testing times were noted to be comparable between the VA tests: TAC-II and PV app (p=0.80) and in CS, HH cards was found to be faster compared to OCC (p<0.01). The mean difference between PV app and TAC-II was -0.25±0.40 logMAR, 95% LoA was -1.03 to 0.53 logMAR and this was noted to be significantly different (p<0.01). The PV app over-estimated VA when compared to TAC-II by 0.25 logMAR. The mean difference between HH cards and OCC was 0.06±0.22 logCS, 95% LoA was -0.37 to 0.49 logCS and this was noted to be significantly different (p<0.01). The OCC cards under-estimated CS when compared to HH by 0.06 logCS. The intra-observer repeatability of tests carried out in 21 children with CVI revealed that TAC-II had better repeatability (coefficient of repeatability, CR=0.47) compared to PV app (CR=0.99). In CS tests, OCC had better repeatability (CR=0.24) when compared to HH cards (CR=0.55).

Relationship of visual functions with associative factors in children with CVI
The relationship between visual functions (i.e., VA and CS) and associative factors (such as: seizure frequency, developmental age, functional vision) was analyzed. Children having a seizure episode within the last 3 months had significantly poorer VA and CS measured using TAC-II (p=0.03) and OCC (p=0.02) respectively when compared to the children who had their last seizure episode greater than 3 months ago. The VA, CS and developmental ages were significantly different across the 3 phases of CVI when the chronological age was adjusted, indicating that children with poorer VA, CS and developmental age belonged to low visual functioning group of CVI that was determined based on the functional vision score. Functional vision score was marginally strongly correlated with CS (r=0.86, r2=0.73, p<0.001) when compared to VA (r=-0.83, r2= 0.68, p<0.001). Functional vision score was also noted to have strong correlation with developmental age (r=0.71, r2=0.41, p<0.001). Whereas, developmental age and visual functions were noted to have moderate correlation, i.e., VA: r=-0.54, r2=0.43, p<0.001 and CS: r=0.59, r2=0.66, p<0.001.

Controls

A total of 50 typically developing children were recruited as controls with a mean age of 3.39±1.87 years (6 months to 6.83 years, 38% males). The testability rates were found to be 100% for all the tests (TAC-II, PV app, HH cards and OCC). The testing times were noted to be shorter with the PV app when compared to TAC-II (p=0.04) and in CS, HH cards was found to have shorter testing time when compared to OCC (p<0.01). The mean difference between TAC-II and PV app was -0.14±0.30 logMAR, 95% LoA was -0.72 to 0.44 logMAR and this was noted to be significantly different (p<0.01). The PV app over-estimated VA when compared to TAC-II by 0.14 logMAR. The mean difference between HH cards and OCC was 0.27±0.11 logCS, 95% LoA was 0.06 to 0.49 logCS and this was found to be significantly different (p<0.01). Ohio contrast cards under-estimated CS when compared to HH by 0.27 logCS. The intra-observer repeatability of tests carried out in 16 children revealed that TAC-II had better repeatability (CR=0.27) compared to PV app (CR=0.41). In CS tests, OCC had better repeatability (CR=0.08) when compared to HH cards (CR=0.27).

Discussion and conclusions

The VA and CS estimates using the clinical tools were noted to be significantly different in both children with CVI and controls and the LoA was found to be narrower for the controls when compared to children with CVI. The PV app over-estimated VA when compared to TAC-II and OCC under-estimated CS when compared to HH cards in both groups. In addition to the vision impairment and delayed overall development in children with CVI, the different nature of the tests and step size of the values being measured (i.e., VA and CS) could be the main reasons for the difference in the values between the tests of visual functions. The intra-observer repeatability indices revealed that TAC-II had better repeatability when compared to the PV app. Therefore, indicating that it is not suggestible to use the tests of VA interchangeably. Ohio contrast cards was noted to have better repeatability indices when compared to HH cards and thereby also suggesting not to use the tests of CS interchangeably. This is important in order to avoid incorrect interpretation of the VA and CS values particularly, when used as an outcome measure. It is also important to interpret the change in the VA and CS based on the test being used as each of them have different repeatability indices. In the current study, there was about 2 cards repeatability difference with TAC-II (VA test) and within 1 card repeatability difference with OCC (CS test). In addition, there was a strong correlation between CS and functional vision score. These findings indicate that CS is an essential parameter to be captured in children with CVI to understand their visual concerns better and for planning rehabilitative strategies. Developmental age and parent-reported visual concerns can serve as a referral parameter for paediatricians, paediatric neurologists and developmental psychologists. The visual functions and functional vision are likely to be influenced by the seizures and the overall development of the child and hence it is important to account for these parameters for planning medical interventions and rehabilitative strategies.

Publication Type:	Thesis (Doctoral)
Subjects:	R Medicine > RE Ophthalmology R Medicine > RJ Pediatrics > RJ101 Child Health. Child health services
Departments:	School of Health & Psychological Sciences > Optometry & Visual Sciences School of Health & Psychological Sciences > School of Health & Psychological Sciences Doctoral Theses Doctoral Theses

Export

Downloads