Abstract
Aim
To determine the causes of severe visual impairment and blindness in children in schools for the blind in Maharashtra, India.
Methods
Children aged <16 years with a visual acuity of <6/60 in the better eye, attending 35 schools for the blind were examined between 2002 and 2005, and causes were classified using the World Health Organization's system.
Results
1985 students were examined, 1778 of whom fulfilled the eligibility criteria. The major causes of visual loss were congenital anomalies (microphthalmos or anophthalmos; 735, 41.3%), corneal conditions (mainly scarring; 395, 22.2%), cataract or aphakia (n = 107, 6%), and retinal disorders (mainly dystrophies; n = 199, 11.2%). More than one third of children (34.5%) were blind from conditions which could have been prevented or treated, 139 of whom were referred for surgery. Low vision devices improved near‐acuity in 79 (4.4%) children, and 72 (4%) benefited from refraction. No variation in causes by sex or region was observed.
Conclusions
Congenital anomalies accounted for 41% of blindness, which is higher than in a similar study conducted 10 years ago. Corneal scarring seems to be declining in importance, low vision and optical services need to be improved, and research is needed to determine the aetiology of congenital anomalies.
The control of childhood blindness is a priority of “VISION2020—The Right to Sight”.1 This global initiative aims to eliminate avoidable blindness by the year 2020, and the first phase focuses on the implementation of cost‐effective strategies. Although blindness in children is relatively uncommon, it is a priority of VISION2020, as severe visual loss in early childhood adversely affects development, mobility, education, and social and employment opportunities. The prevalence of blindness in children ranges from approximately 0.3/1000 (of total population) in affluent countries to 1.5/1000 in the poorest2; but owing to demographic differences, the actual number of blind children per million total population ranges from approximately 60 in affluent societies to 600 in the poorest communities.
Epidemiological data on blindness in children in India are incomplete, but population‐based studies have estimated the prevalence as follows: 0.65/1000 (95% confidence interval 0.15 to 1.15/1000) in children aged 0–15 years in urban and rural Andhra Pradesh3; 1.25/1000 in children aged 5–15 years in rural Andhra Pradesh4; and 0.53/1000 in children aged 5–15 years in Delhi.5 These figures are not directly comparable, as different definitions of blindness have been used, and the samples are of different ages. Overall, the prevalence in India is estimated to be 0.81/1000 children,4 ranging from 0.3/1000 in well‐developed states (eg, Kerala) to 1.5/1000 children in the poorest. These figures have been derived using the association between the prevalence of blindness in children and mortality in children aged <5 years (as a proxy for socioeconomic development and healthcare delivery).6
Reliable population‐based data on the causes of blindness in children are difficult to obtain, particularly in developing countries. Examination of children enrolled in special education is one source, but bias is inherent in all facility‐based studies. Advantages of school studies are that a large number of children can be examined quickly by a few examiners. Over the past 10 years, the World Health Organization's classification system7 has been increasingly used, which allows data to be compared.
A study on about 1300 children in nine Indian states undertaken in 1993, showed corneal scaring (mainly from vitamin A deficiency (VAD)) to be the single most common (26.4%) cause of blindness, followed by congenital anomalies (mainly anophthalmos and microphthalmos; 20.7%) and retinal diseases (19.3%).8 The major causes varied from region to region, with corneal scarring being less important in children in schools serving urban areas.8 The study only included 157 children from two schools in Maharashtra.
India is developing rapidly, but improvements are not taking place evenly across the country. Maharashtra is on the western coast and has varying ecology and climate, with a narrow, wet, coastal strip (Konkan, includes the capital Mumbai), a mountainous area, a prosperous, well‐irrigated region (West Maharashtra, with Pune as the largest city), a forested, tribal area (Vidarbha, with Nagpur as the largest city), and dry, arid regions in the north (Khandesh) and centre (Marathwada, with Aurangabad as the largest city). Maharashtra is India's second largest and second most populated state (102 million), with an adult literacy rate of 77%.9 It is among the more prosperous states (per capita income of US $544). The past two decades have seen nearly universal immunisation, increased school attendance, and improved eye care and child care facilities. The infant mortality was 45/1000 live births in 2001 compared with the national average of 64/1000.9
This study was undertaken to determine the causes of blindness in all children attending schools for the blind in Maharashtra.
Methods
Children attending all special schools for blind children in Maharashtra were examined between 2002 and 2005. Schools were contacted through the National Association for the Blind, Mumbai, India, the Poona Blind Men's Association, Pune, India, and other agencies. All the blind schools except four are run by non‐governmental organisations, and none admit children with additional disabilities.
Relevant information was collected from class teachers and children. A brief history of the family, place of residence and whether the parent's marriage was consanguineous were recorded. Information on additional disabilities (eg, mental retardation, physical handicap, epilepsy or deafness) was obtained from children's records. A detailed eye examination was performed by a team of optometrists and ophthalmologists. Distance visual acuity was measured using a Snellen E chart, and near‐vision was assessed using figures equivalent to N18. Simple tests of functional vision were undertaken. Intraocular pressures were not measured, and visual fields were assessed by confrontation. Children were refracted, and assessed for low‐vision devices, if indicated.10 Anterior segments were examined using a torch and loupe or a handheld slit lamp, and posterior segments were examined by direct and indirect ophthalmoscopy after dilating the pupils.
The World Health Organization classification system for children was used to categorise causes using definitions in the coding instructions.7 One major anatomical site and underlying cause were selected for each eye and for each child. The need for optical, surgical or medical interventions was recorded and the visual prognosis assessed. Children requiring further investigations and treatment were referred to the H.V. Desai Eye Hospital, Pune, India. Data were entered and analysed using EPI‐INFO 6 (World Health Organization). A report of the findings and recommendations were given to the principal of each school.
Results
Study population
A total of 35 residential schools were identified and visited (11 in western Maharashtra, 10 in Vidarbha, 10 in Marathwada and 4 in Khandesh, listed in appendix) in which 1985 children were enrolled, 1795 of whom were aged <16 years. About half were boys (1126/1985; 56%), and this sex difference was found in each region (west Maharashtra, 54.6%; Vidarbha, 52.9%; Marathwada, 74.0%; Khandesh, 67.9%). The youngest child was 2 years old.
The vast majority of children were blind (visual acuity of <3/60 in the better eye; 1712/1795; 95.4%), and a further 66 (3.4%) were severely visually impaired (visual acuity <6/60 to 3/60) (table 1). Data presented in this paper are for children aged <16 years who were blind or severely visually impaired (n = 1778). Almost three quarters (73.3%) gave a history of being blind since birth; only 2.3% lost vision during infancy, whereas 13.4% became blind during childhood. Only 18% of children reported another similarly affected family member, and 10% knew their parents were related by birth. We found no difference in rates of consanguinity between regions. More than half of the children (58%) could see well enough to walk about unaided, 25% could recognise faces and 13% could see shapes.
Table 1 World Health Organization categories of visual loss in 1795 children attending 35 schools for the blind in Maharashtra, India.
| Category of visual loss | West Maharashtra | Vidarbha | Marathwada | Khandesh | Whole of Maharashtra |
|---|---|---|---|---|---|
| n (%) | n (%) | n (%) | n (%) | n (%) | |
| Normal vision: ⩾6/18 | 1 (0.2) | 3 (0.6) | 2 (0.4) | 0 (0) | 6 (0.3) |
| Visual impairment: 6/24–6/60 | 3 (0.5) | 4 (0.8) | 2 (0.4) | 2 (0.9) | 11 (0.6) |
| Severe visual impairment: <6/60–3/60 | 25 (4.2) | 19 (3.8) | 10 (2.0) | 12 (5.5) | 66 (3.7) |
| Blind: <3/60 | 561 (95.1) | 473 (94.8) | 474 (97.1) | 204 (93.6) | 1712 (95.4) |
| Total | 590 (100) | 499 (100) | 488 (100) | 218 (100) | 1795 (100) |
Causes of severe visual impairment and blindness
In all 823 (46.3%) children were blind from disorders of the whole globe, microphthalmos being responsible for 30.7% and anophthalmos for 10.4% (fig 1). Corneal causes (including phthisis bulbi) were recorded in 395 (22.2%) children. Disorders of the lens were found in 107 (6.0%) children, 80 of whom had unoperated cataracts. Of these, 16 children had dense amblyopia or had had complicated surgery, and 54 children with pseudophakia/aphakia had visual loss due to coexisting retinal pathology. Retinal factors accounted for 199 (11.2%) cases, the majority being retinal dystrophies. No child was blind from retinopathy of prematurity (ROP). Optic nerve lesions were diagnosed in 81 (4.6%) children and uveal conditions (mainly uveitis and colobomas) in 27 (1.5%) children. We found no major differences in causes between regions: in particular, corneal scarring was not more common in the poorer tribal belts of Khandesh (11.6%) and Vidarbha (14.2%) compared with the prosperous western Maharashtra (12.5%).
Figure 1 Causes of severe visual impairment and blindness.
In more than two thirds (69.2%) of children, an underlying cause could not be determined; most children were blind from prenatal factors (ie, anophthalmos, microphthalmos, cataract and glaucoma). In all, 13.7% of children were blind because of hereditary factors, 11.1% of children had acquired causes of blindness (from trauma, corneal infections or VAD) and 4.6% were blind from intrauterine factors. We found no differences in underlying cause by sex or by region.
Causes in different age groups
Corneal causes seem to have declined in importance over time, as 24.8% of children aged 11–16 years were blind from corneal scarring and phthisis bulbi compared with 19.6% of children aged 7–10 years and 10.7% of children aged <7 years.
Avoidable causes
Preventable causes of blindness, such as corneal scarring and phthisis, mainly attributed to measles, VAD, trauma and harmful traditional medicines, accounted for 22.2% of all causes, whereas treatable causes were found in 219 (12.3%) children. Thus, 614 children (34.5%) had avoidable causes of blindness (table 2). In all, 139 children were referred for evaluation and surgery (eg, cataract extraction, intraocular lens implantation, keratoplasty); the visual acuity could be improved by refraction in 72 children and 79 would benefit from low‐vision devices. A change of school was recommended for 63 children.
Table 2 Causes of avoidable blindness.
| Causes | West Maharashtra | Vidarbha | Marathwada | Khandesh | Whole of Maharashtra |
|---|---|---|---|---|---|
| Preventable | |||||
| Cornea | 73 | 56 | 73 | 30 | 232 |
| Phthisis | 59 | 39 | 47 | 18 | 163 |
| Subtotal | 132 (22.5) | 95 (19.3) | 120 (24.8) | 48 (22.2) | 395 (22.2) |
| Treatable | |||||
| Cataract | 45 | 28 | 22 | 12 | 107 |
| Glaucoma | 17 | 6 | 6 | 3 | 32 |
| Amblyopia | 29 | 18 | 9 | 9 | 65 |
| Uveitis | 5 | 5 | 4 | 1 | 15 |
| Subtotal | 96 (16.4) | 57 (11.6) | 41 (8.4) | 25 (11.5) | 219 (12.3) |
| Total avoidable | 228/586 (38.9) | 152/492 (30.9) | 161/484 (33.2) | 73/216 (33.7) | 614/1778 (34.5) |
Values in parentheses are percentages.
Comparison with the 1993 study
Some differences between the 1993 study and this study are apparent, in that disorders of the lens (cataract and aphakia) seem to be less important now, whereas microphthalmos and anophthalmos have increased in importance from 17.2% to 41.3% (table 3).
Table 3 Changing pattern of childhood blindness between 1993 and 2005.
| 1993 n (%) | 2005 n (%) | |
|---|---|---|
| Cornea | 32 (20.4) | 395 (22.2) |
| Lens | 20 (12.7) | 107 (6.0) |
| Retina | 22 (13.8) | 199 (11.2) |
| Anophthalmos/microphthalmos | 27 (17.2) | 735 (41.3) |
| Other causes | 51 (35.9) | 342 (19.3) |
| Total | 152 (100) | 1778 (100) |
1993 data from Rahi et al.8
Discussion
Studies on children receiving special education have limitations, as children attending special schools may be different from blind children not attending school. In particular, causes associated with additional handicaps, those primarily affecting children from poor remote communities, are likely to be under‐represented, as are causes associated with a high mortality. However, where data from children in special education have been compared with data from community samples the findings have generally been comparable.11
In our study, the observed pattern of causes is intermediate between that seen in industrialised countries and the poorest developing nations. In our study, older children were more likely to be blind from corneal scarring than younger children, which suggests that measles and VAD may be declining in importance. A similar change was documented in Saudi Arabia, where socio‐economic development has led to genetic diseases predominating over infectious causes.12 Corneal blindness was also reported to be less important in younger children than in older children in a recent blind school study in Delhi, India.13 The lower incidence of corneal blindness, particularly in Asia where mortality of children aged <5 years is declining, is almost certainly due to better primary healthcare, higher measles immunisation coverage and child survival initiatives that include control of VAD.
Despite marked differences in ecology, climate and prosperity, we found no differences in causes between different regions within the state. This is in contrast with a study undertaken in schools for blind in Andhra Pradesh, where regional differences were observed.14 In our study there was a preponderance of boys, which is a common finding in facility based studies in developing countries, where cultural factors and economic constraints may lead to sex discrimination.
The most striking finding of our study is the very high proportion of blindness due to microphthalmos and anophthalmos. This has been a common finding in studies on blind chidren in India, including population based‐studies on children in community‐based rehabilitation,8,13,14,15 but the reasons are not known as most causes of blindness are sporadic.16 Known causes include chromosomal abnormalities (<10% of cases in some series)17 and some can be familial, with autosomal dominant, recessive or X‐linked recessive pedigrees being reported. As families with more than one affected individual are uncommon, linkage analysis is usually not possible.18 Several regulatory genes important in ocular development have been implicated in the aetiology, including SHH, PAX6, PAX2 and CHX10,19 as have environmental factors (ie, intrauterine infections, use of pharmaceutical and recreational drugs, alcohol, hyperthermia and maternal hyothyroidism).20,21 Exposure to pesticides has also been postulated,22 which might be relevant in this highly agricultural area where fertilizers and pesticides are used intensively. However, in all series, including a large case series from Andra Pradesh,23 most cases are sporadic and of unknown cause. One hypothesis is that microphthalmos, anophthalmos and coloboma are due to interactions between genes controlling retinoic acid signalling and maternal VAD during early fetal development24—a mechanism similar to folate deficiency and spina bifida—with supportive evidence coming from animal experiments, and epidemiological and laboratory studies. If defects in genes controlling retinoic acid signaling occur at a high rate in the Indian population, where maternal VAD and consanguinity are common, this may explain why anomalies are such an important cause of blindness in Indian children.
Public health approaches to preventing congenital anomalies are limited to health education concerning exposure to known risk factors during pregnancy and rubella immunisation. However, rubella immunisation of infants is not without risk, having the potential to increase the number of babies born with congenital rubella if high coverage is not maintained.25 In India, rubella immunisation is available only privately, with low levels of uptake. Children with microphthalmos can often benefit from refraction and low‐vision services10,26 as has also been shown in this study.
In our study, about one third of children were blind from potentially avoidable causes. No children was blind from ROP, which is reaching epidemic proportions in many middle income countries, and which is being increasingly reported from other cities in India.27,28 There are several explanations: children blind from ROP often have other disabilities, and may not attend school or may have died, and most children in our study were aged >10 years, and ROP may be a cause in younger children. Given the expanding economy in Maharashta, ROP blindness will probably become another avoidable cause unless screening programmes are implemented as services for premature babies expand.
Preventable causes of blindness can be reduced at the primary level of service delivery, whereas treatable causes require specialised, paediatric ophthalmology units, systems for early identification and referral, and increased public awareness. A comprehensive approach is therefore needed, including provision for children with low vision. Maharashtra needs at least five tertiary child eye care centres (one for every 20 million people), but currently has only one (the H.V. Desai Eye Hospital, Pune, India), which was established with support from ORBIS International in 2004.
Acknowledgements
We thank Mr Niranjan Pandya and the principals, staff and students of all the schools for the blind for their cooperation. Dr Rahul Deshpande, Dr Sucheta Kulkarni, Dr Tanaji More, Dr Kishor Jadhav, Dr Seema Jagdale, Dr Sachin Dharwadkar, Dr Ganesh Niras, Dr Amit Vishwe, Dr Geeta Gandhi and Dr Anoopa Bhargav examined the children with the help of the authors and the interns of the Jnana Prabodhini School of Optometry and Bharti Vidyapeeth Medical College School of Optometry, Pune, India. Mrs Anjali Dalvi, a social worker, helped in recording patient history. Mr Santosh Jagtap, Mr Nilesh Khaire and Mrs Sangita Patil helped in data management. Dr Kaumudi Godbole and Dr Kuldeep Dole helped with the write‐up.
Abbreviations
ROP - retinopathy of prematurity
VAD - vitamin A deficiency
Appendix
Names of blind schools, Maharashtra, India
Western Maharashtra:‐
Bhiruratan Damani Andha Shala Nivasi, Solapur
Rajiv Gandhi Memorial School for the Blind, Solapur
Lion's Club Pandharpur Sanchalit Sarkarmanya Andh Vikas Sanstha, Pandharpur, Solapur
Prabodhan Andh Vidyalay, Koregaon, Satara
Shri Sadguru Dhondiraj Nivasi Andh Vidyalay, Palus, Sangli
Dyan Prabodhan Bhavan Sanchalit Andhashala, Miraj‐Tikati, Kolhapur
N.A.B Nivasi Andha Vidyalay, Miraj, Sangali
P.R Lunkad Blind School, Bhosari, Pune
Jagruti School For Blind Girls, Alandi, Pune
Poona School & Home For Blind Boys, Pune
Poona School & Home For Blind Girls, Pune
Vidarbha:‐
Yashwant Andh Vidyalay, Amravati
Apang Kalyan & Punarvasan Sanstha, Buldhana
Kurvey's New Model High School & Junior College, Nagpur
The Blind Boys' Institute, Nagpur
Narendra Bhivapurkar Andha Vidyalay, Amravati
Dyanjyoti Andha Vidyalay, Dhatarkheda, Nagpur
Kanubai Vohra Andha Vidyalay, Akola
NA Vidyalay, Chikhaldara, Amravati
Anand Blind School, Warora, Chandrapur
Shri Govindrao Bijure Andha Vidyalay, Daryapur, Amravati
Marathwada:‐
School for the Blind Bodhani, Kinwat, Nanded
Niwasi Andha Vidyalay, Vasarani, Nanded
Andha Vidyalay, Deglur, Nanded
Shri Guru Ganesh Drishtihin Vidyalay, Jalna
Lt Gangabai Madhavrao Borole Nivasi Andha Vidyalay, Nilanga, Latur
Matoshri Gangadevi Dewda Niwasi Andha Mulanche Vidyalay, Hingoli
M.A.B Andha Vidyalay, Udgir
Taramati Bafna Blind School, Aurangabad
Pradnyachakshu Niwasi Andha Vidyalay, Beed
Shaskiya Andha Vidyalay, Latur
Khandesh:‐
Malegaon Andh Sikshan & Prasikshan Sanstha, Malegaon, Nashik
Girls' School For Blind, Dhule
Andh Mulanchi Shala, Dhule
Rashtriya Prasar Mandal Sanchalit Andh Vidyalaya, Chalisgaon, Jalgaon
Footnotes
Funding: This study was funded by the H.V. Desai Eye Hospital, Pune, India.
Competing interests: None declared.
References
- 1.World Health Organization Global initiative for the elimination of avoidable blindness. WHO Publication Number WHO/PBL/97. 61. Geneva: WHO, 1997
- 2.Rahi J S, Gilbert C E, Foster A.et al Measuring the burden of childhood blindness. Br J Ophthalmol 199983387–388. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Dandona L, Williams J D, Williams B C.et al Population based assessment of childhood blindness in southern India. Arch Ophthalmol 1998116545–546. [PubMed] [Google Scholar]
- 4.Dandona R, Dandona L, Srinivas M.et al Refractive errors in children in a rural population in India. Invest Ophth Visc Sci 200243615–622. [PubMed] [Google Scholar]
- 5.Murthy G V S, Gupta S K, Elloveia L B.et al Refractive errors in children in an urban population in New Delhi. Invest Ophthalmol Vis Sci 200243623–631. [PubMed] [Google Scholar]
- 6.Gilbert C, Rahi J, Quinn G. Visual impairment and blindness in children. In: Johnson G, Minassian D, Weale R, et al eds. Epidemiology of eye disease. 2nd edition. London: Arnold Publishers, 2003
- 7.Gilbert C, Foster A, Negrel A D.et al Childhood blindness: a new form of recording causes of vision loss in children. Bull World Health Organization 199371485–489. [PMC free article] [PubMed] [Google Scholar]
- 8.Rahi J S, Sripathi S, Gilbert C E.et al Childhood blindness in India: causes in 1318 blind school students in nine states. Eye 19959545–550. [DOI] [PubMed] [Google Scholar]
- 9.Registrar General and Census Commissioner Census of India 2001. India: Registrar General and Census Commissioner, 2001
- 10.Silver J, Gilbert L E, Spoerer P.et al Low vision in East African blind school students: need for optical and low vision services. Br J Ophthalmol 199579814–820. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Bulgan T, Gilbert C. Prevalence and causes of severe visual impairment and blindness in children in Mongolia. Ophthalmic Epidemiol 20029271–281. [DOI] [PubMed] [Google Scholar]
- 12.Tabbara K F, Badr I. Changing pattern of childhood blindness in Saudi Arabia. Br J Ophthalmol 198569312–315. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Titiyal J S, Pal N, Murthy G V S.et al Causes and temporal trends of blindness and severe visual impairment in children in schools for the blind in North India. Br J Ophthalmol 200383941–945. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Hornby S J, Adolph S, Gothwal V K.et al Evaluation children in six blind schools in Andhra Pradesh. Ind J Ophthalmol 200048195–200. [PubMed] [Google Scholar]
- 15.Sil A K, Gilbert C E. Childhood blindness in India. J Ind Med Assoc 20019910–15. [PubMed] [Google Scholar]
- 16.Hornby S J, Gilbert C E, Rahi J K.et al Regional variation in children due to microphthalmos, anophthalmos and coloboma. Ophthalmic Epidemiol 20007127–138. [PubMed] [Google Scholar]
- 17.Kallen B, Tornqvist K. The epidemiology of anophthalmia and microphthalmia in Sweden. Eur J Epidemiol 200520345–350. [DOI] [PubMed] [Google Scholar]
- 18.Morrison D, FitzPatrick D, Hanson I.et al National study of microphthalmia, anophthalmia, and coloboma (MAC) in Scotland: investigation of genetic aetiology. J Med Genet 20023916–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Gregory‐Evans C Y, Williams M J, Halford S.et al Ocular coloboma: a reassessment in the age of molecular neuroscience. Med Genet 200441881–891. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Sowden J, Taylor D T. Disorders of the eye as a whole. In: Taylor D, Hoyt CG, eds. Pediatric ophthalmology and strabismus. Philadelphia, USA: Elsevier Saunders, 2005. 206
- 21.Vogt G, Puho E, Czeizel A E. A population‐based case‐control study of isolated ocular coloboma. Ophthalmic Epidemiol 200512191–197. [DOI] [PubMed] [Google Scholar]
- 22.Dolk H, Busby A, Armstrong B G.et al Geographical variation in anophthalmia and microphthalmia in England, 1988–94. BMJ 1998317905–909. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Hornby S J, Ward S J, Gilbert C E.et al Environmental risk factors in congenital malformations of the eye. Ann Trop Paediatr 20022266–77. [DOI] [PubMed] [Google Scholar]
- 24.Hornby S, Ward S J, Gilbert C E. Eye birth defects in humans may be caused by a recessively inherited genetic predisposition to the effects of maternal vitamin A deficiency during pregnancy. Med Sci Monitor 20039HY23–HY26. [PubMed] [Google Scholar]
- 25.Vynnycky E, Gay N J, Cutts F T. The predicted impact of private sector MMR vaccination on the burden of congenital Rubella syndrome. Vaccine 2003212708–2719. [DOI] [PubMed] [Google Scholar]
- 26.Hornby S J, Adolph S, Gothwal V K.et al Requirements for optical services in children with microphthalmos, coloboma and microcornea in southern India. Eye 200014219–222. [DOI] [PubMed] [Google Scholar]
- 27.Gopal L, Sharam T, Shanmugan M P.et al Surgery for stage 5 retinopathy of prematurity: the learning curve and evolving technique. Ind J Ophthalmol 200048101–106. [PubMed] [Google Scholar]
- 28.Gilbert C, Fielder A, Gordillo L.et al Characteristics of babies with severe retinopathy of prematurity in countries with low, moderate and high levels of development: implications for screening programmes. Pediatrics Electronic Pages 2005115518–525. [DOI] [PubMed] [Google Scholar]