A cross-sectional study was conducted in two schools for the blind from the districts of Vellore and Tiruvannamalai in Tamil Nadu (TN), South India after obtaining approval from the Institutional Review Board of Christian Medical College, Vellore (IRB No. 7794). The screening was conducted over 36 months from September 2012 to August 2015, among children less than 18 years who were enrolled in these schools. The first school was a government school and permission was obtained from the local government authorities and school principal. The second school was a private one with three sections for visually challenged children: a primary school, and separate senior schools for boys and girls; screening was conducted after obtaining permission from the principal.

All children 18 years and below whose parents were willing for their child to take part in the screening and gave informed consent were enrolled. Assent was obtained from children above eight years after explaining to them regarding the screening. Children enrolled in these schools had been enrolled either directly to the blind school or had been to a regular school for a few years before their visual impairment was noticed.

Parents were informed regarding the day and time of screening and were present during screening. They contributed to the information collected including onset of visual disturbance and past treatment history, as much as could be recollected.

A team comprising of an ophthalmologist and optometrists from our tertiary care centre visited the schools. Screening was done within the school class rooms. All children underwent distant visual acuity assessment using Snellen’s either number or illiterate E chart based on child’s literacy. The distance of the vision chart was based on child’s vision. Those who could not read the chart at 1 metre were reassessed with counting fingers method followed by projection and perception of light. BCVA and dilated refraction was done for all children. Anterior segment was assessed with flash light examination. Dilated fundus examination was done with indirect ophthalmoscope using a 20 dioptre lens. Documentation was done as per WHO prevention of blindness (WHO/PBL) form [5]. Information was gathered regarding the age at onset of blindness, unaided vision and vision with child’s spectacles, associated general disability and history of previous eye surgery. After ocular examination, the major causes for decrease in vision/ blindness based on anatomical classification were noted. Children who had visual acuity equal to or better than counting fingers at 25 cm were referred to the base hospital for ocular examination and low vision assessment. They underwent slit lamp examination and indirect ophthalmoscopy for fundus assessment; B-scan was done when indicated.

Spectacles were given free of cost when improvement in vision was noted with BCVA. All children who had cataract underwent Lens Matter Aspiration (LMA) along with Primary Posterior Capsulotomy (PPC) and Partial Anterior Vitrectomy (PAV) when indicated. Those with aphakia and good capsular support received secondary Intra-Ocular Lens (IOL) implantation. Low vision assistive devices were dispensed as appropriate.

Descriptive measures such as mean with SD and/or median with IQR was presented for all continuous variables, whereas frequencies and percentages were presented for all categorical variables. All the categorical variables were associated with ocular disability using Chi-square test with continuity correction, and for small sample sizes Fisher’s-exact test was used. SPSS version 16.0 was used for analysis and p <0.05 was considered as statistically significant.

There were 185 students on the rolls of the two schools, and all of them participated in the screening; 121 (65.4%) were males. The mean age at the time of screening was 12.1±3.4 years. The age at onset of blindness was below one year in 108 (58.5%), 1 to six years in 65 (35.2 %) and above six years in 12 (6.3%). Age at the time of screening was <6 years in 15 (8.1%), 7 to 12 years in 54 (29.2%) and >12 years in 116 (62.7%). Positive family history of similar illness was seen in 78 (42%); among those with positive family history, parent/ siblings were affected in 71 (91.5%). History of consanguinity was seen in 118 (63.8%). Those with positive family history were compared with those with and without a history of consanguinity; the difference was not significant (p=0.075).

BCVA in the better eye was 6/24 to 3/60 in 31 (16.8%), <3/60 to no perception of light in 154 (83.2%). The causes for blindness based on anatomical site were retina 77(41.5%), whole globe 59 (32%), cornea 14(7.5%), optic nerve 14(7.5%), lens 10(5.7%), refractive error in 7(3.8%) and uvea 4(2%). Hence 14(7.5%) were preventable and 21(11.4%) were treatable, thus 35(18.9%) were avoidable causes of CHB.

None of the children had associated morbidity as these schools were for children with visual disability alone. Surgical intervention like LMA, PPC and PAV was done in two children and secondary IOL was placed in the sulcus for one child with surgical aphakia.

Population based studies are essential to determine the prevalence of CHB. It is time consuming and needs adequate manpower. It also has the limitation of missing children with blindness/multiple disabilities with visual impairment who could be away in schools for the blind/special schools at the time of screening. Blind school screening can provide details on various causes of blindness among children from the local population. Surveys done in schools for the blind also have their limitations. It cannot estimate prevalence of the causes of blindness. Moreover, children who are blind and do not access blind schools due to lack of awareness, distance and stigma will miss the screening. However, these screening programmes in blind schools can provide data on the various causes of CHB in a community, as well as to note changing trends over the years.

Our cross-sectional study in two blind schools showed retinal dystrophies as the most common cause for CHB, followed by whole globe anomalies. Hereditary forms of disease are common in the Indian population probably due to the culture of consanguineous marriage. In our study, history of consanguinity was seen in 63.8% and family history of similar illness was seen in 42%. Among those with positive family history, parent/ siblings were affected in 91.5%. However, those with positive family history were compared with those with/without a history of consanguinity; the difference was not significant (p=0.075).

Causes of CHB reported from various blind schools in India is shown in [Table/Fig-1] [1,6-12]. Prevalence of CHB and major causes of CHB in India obtained from various population surveys are shown in [Table/Fig-2] [4,13-16]. The causes of CHB from different blind school screening reports from other developing countries are shown in [Table/Fig-3] [17-22].

A changing trend is noticed over the past few years in the major causes of blindness, with a change from corneal blindness to hereditary causes like retinal dystrophies and whole globe anomalies [9]. The decline in corneal causes is probably due to better Vitamin A intake and measles vaccination coverage. In India, until 2015, corneal blindness remained the first or second most common cause of CHB. A similar trend in the causes of CHB was reported from other developing countries where corneal blindness remained the first or second most common cause of CHB until 2015. Congenital anomalies and posterior segment pathologies were the most common causes of CHB in recent years [9].

In our study, more than 50% of children had developed defective vision before one year of age. Ocular morbidity occurring before one year of age will result in dense amblyopia. This emphasises the need for community programmes to screen for ocular morbidity in children. Community and school screening can help in early detection of cataract and refractive errors. Children should also be screened for ocular morbidity in schools, pre-schools and well baby clinics to reduce avoidable blindness in this amblyogenic age group.

A limitation of our study was the lack of availability of documentation regarding child’s previous ocular condition and previous treatment. Parents were often unaware of the cause and sequence of events.

This study was conducted among schools for the blind within a 50 km radius of our tertiary care hospital. Vellore and Tiruvannamalai districts have only one school each, for the blind, and all the children in these two schools were enrolled in our study but the overall numbers were limited.

Retinal dystrophies were the major cause of CHB in our population; one-fifth of children had avoidable blindness. Posterior segment anomalies contributed to more than 50% of the causes for CHB. More than half the children had developed defective vision before one year of age. This emphasises the need for regular community and school screening programmes, health education and to implement regular screening in well baby clinics for early identification of ocular morbidity and prompt management.