Acta Ophthalmologica 2015

Universal ocular screening of 1021 term infants

using wide-eld digital imaging in a single public
hospital in India a pilot study
Anand Vinekar,1 Indu Govindaraj,1 Chaitra Jayadev,1 Anupama Kiran Kumar,1 Praveen Sharma,1
Shwetha Mangalesh,1 Lakshmipathy Simaldi,2 Kavitha Avadhani,1 Bhujang Shetty1 and Noel Bauer3
1

Narayana Nethralaya Postgraduate Institute of Ophthalmology, Bangalore, India

Department of Pediatrics, K.C. General Hospital, Bangalore, India
3
Faculty of Ophthalmology, Maastricht University, Maastricht, the Netherlands
2

ABSTRACT.
Purpose: Universal eye screening of neonates is currently not standard of care.
Early detection of abnormalities could oer prompt management and a reduction in
visual morbidity. We report a pilot study using wide-eld digital imaging to screen
all infants at birth to explore its feasibility as a tool for universal screening.
Methods: Consecutively enrolled 1021 term infants in a public hospital were
imaged within 72 hr of birth using the Retcam shuttle (Clarity MSI, USA).
Anterior and dilated posterior segment images were obtained. Infants with
abnormal images were examined clinically, and medical or surgical treatment
was given when needed, at no cost to the family.
Results: Of the 1021 healthy full-term newborns, 48 babies had abnormal
ndings (4.7%). Retinal haemorrhages were the most common (52.1%)
abnormality of which 24% were macular. A ridge resembling retinopathy of
prematurity in both eyes constituted 18.8% of all abnormalities. Nine infants
(18.8% of abnormalities and 0.9% overall) had conditions meriting medical or
surgical intervention and would have been missed otherwise. Seven of these nine
babies (0.7%) needed medical attention with a systemic work-up for conditions
such as posterior uveitis with linear perivasculitis, salt and pepper retinopathy or
posterior synechia. Two of nine babies (0.2%) required surgical intervention for
unilateral cataract and retinoblastoma.
Conclusion: Universal eye screening of all neonates using wide-eld digital
imaging is possible and safe. Extrapolating our results to the national scenario,
we estimate that 226 950 infants requiring treatment would go undiagnosed
annually. Universal infant eye imaging strategies must receive serious national
consideration.
Key words: infant blindness retcam screening universal imaging wide-eld

Acta Ophthalmol.
2015 Acta Ophthalmologica Scandinavica Foundation. Published by John Wiley & Sons Ltd

doi: 10.1111/aos.12685

Introduction
Universal eye screening of infants is
not standard of care even in developed
countries. In the United Kingdom

(Royal College of Ophthalmologists

& the British Paediatric Association
1994) or the United States of America
(American Academy of Pediatrics, Section on Ophthalmology, American

Association for Pediatric Ophthalmology & Strabismus, American Academy

of Ophthalmology, American Association of Certied Orthoptists 2008), the
recommendation is to perform red
reex testing (RRT) at birth or thereafter. In India, even in the bigger cities,
red reex testing is not universally
practised. The ocular health of infants
is often restricted to screening of retinopathy of prematurity (ROP), and
even this is performed in a handful of
tertiary care centres in the larger cities
(Charan et al. 1995; Gilbert et al.
1999). The reasons are manifold and
include lack of experts, poor access to
health care, reduced awareness and
poverty (Gilbert et al. 1999). Recently,
to address this issue of universal
screening, the Indian Government has
launched a programme, which aims
to detect over 30 conditions in the
016 years population, the Rashtriya
Bal Swasthya Karyakram (RBSK)
(National Child Health Program)
under the National Rural Health Mission (NRHM, Ministry of Health &
Family Welfare 2013). Although infant
eye screening is one of the listed activities, operational guidelines have not
been detailed.
Our institute manages a multicentre
tele-ROP screening programme, Karnataka Internet Assisted Diagnosis of
Retinopathy
of
Prematurity
(KIDROP) which provides ROP
screening in 83 neonatal centres in
three zones of rural regions in the
Southern Indian state of Karnataka in

Acta Ophthalmologica 2015

collaboration with the NRHM (Vinekar 2011; Vinekar et al. 2011a,b; Vinekar et al. 2014a,b; Hungi et al. 2012).
On recently reviewing our images, we
observed that 7.9% of premature
infants being imaged for ROP were
serendipitously detected to have other
diagnoses (C. Jayadev, A. Vinekar, S.
Mangalesh, P. Mahendradas, V. Kemmanu, A. Mallipatna, N. Bauer,
B. Shetty, unpublished). This led us to
explore the feasibility of using the
ongoing ROP programme to implement universal infant eye screening.
The aim of this pilot study was to
highlight the prevalence of ocular
abnormalities in all newborn infants
screened using wide-eld digital imaging in a single public hospital and
explore the feasibility and ecacy of
universal infant eye imaging as a tool
for universal screening in the country.

Materials and Methods

A public general hospital within Bangalore Metropolitan city limits, already
included in KIDROP, was chosen for
universal eye screening. This multispecialty hospital provides free or subsidized care to a large population from a
lower socio-economic background.
Inclusion into this study was not mandatory. The treating neonatologists,
paediatricians and our team created
awareness among the mothers and
families of the babies of the possible
benets of enrolling in the study. The
Institute Research Board and the Ethics Committee approved the consent
form and the protocol of imaging and
screening.
Between 1 September 2012 and 31
March 2013, 1954 babies were born in
the study hospital. Of these, parents of
1021 (52.3%) babies were present on
the days of screening and consented for
enrolment. Our team visited the study
centre three times a week. Infants born
on the day of the visit or between two
visits and still admitted were oered
inclusion. All babies were inborn, and
no deliberate attempt was made to
exclude any neonate, unless the paediatrician advised otherwise. Owing to
the pilot nature of this study, no
attempt was made to call back infants
who were discharged from the hospital
during the interval between each of our
teams visits. Additionally, this service
was oered at no cost to the parents or
to the hospital. All enrolled infants

were born at term and had a birthweight more than 2000 g. Babies born
<2000 g at birth were enrolled for the
ongoing ROP screening programme
which was performed once a week
and at a dierent locations in the
hospital (Vinekar et al. 2014a,b).
Our team of trained technicians is
specialized in infant eye imaging using
the Retcam shuttle (Clarity MSI,
Pleasanton, CA, USA). On an average,
they perform over 1100 imaging sessions in a month in the KIDROP
programme (Vinekar et al. 2014a,b).
The screening was performed in a room
adjacent to the neonatal intensive care
unit (NICU) and was assisted by a
neonatal nurse. Pupillary dilatation
was achieved using a commercially
available combination of cyclopentolate 0.5% and phenylephrine 2.5%
(Auropent Plus, Aurolab, Madurai,
TN, India) administered by the nurse
under supervision. A sterile infant wire
speculum was used in one eye at a time
after instilling topical anaesthetic proparacaine 0.5% solution. An anterior
segment image was obtained without
the use of any externally attached lens,
and retinal images were obtained using
the 130-degree (ROP lens). Images
were captured in video mode, reviewed
and relevant stills saved. There were no
ocular or systemic complications during or after any of the imaging sessions.
The paediatric retina specialist at the
base hospital scheduled babies who
had abnormal ndings for a clinical
exam within 7 days of the session and
follow-up was determined on the merit

of each case. Our institute provided

ophthalmic screening, medical and surgical management for these infants
when required, at no cost to the family
or to the referring hospital.

Results
The total number of babies screened
during the study period was 1021 [559
(54.8%) males and 462 (45.2%)
females]. Strikingly, of the 1021 deliveries, 990 (96.9%) were by Caesarian
section and only 31 (3.1%) were delivered vaginally.
Forty-eight (4.7%) babies had an
ocular pathology in at least one eye.
Retinal haemorrhages were the most
common nding and were present in at
least one eye in 25 babies, accounting
for 52.1% of abnormalities detected
and 2.4% of all babies screened.
According to Egges classication
(Egge et al. 1980), six (0.5%) had grade
I, 13 (1.2%) had grade II and six
(0.5%) had grade III retinal haemorrhages (Fig. 1). ROP-like ridge was
noted in the temporal periphery in nine
babies, which accounted for 18.8% of
abnormalities and 0.9% of all babies
screened (Fig. 2). Over 8 weeks, this
ridge as documented in follow-up sessions resolved in all cases spontaneously without progressing to type 1
ROP. Two babies with retinal dysplasia (4.2% of abnormalities), one baby
with a peripheral scar (2.1%), one with
persistent fetal vasculature syndrome
(2.1%) and one with a naevus (2.1%)
were advised observation.

Fig. 1. Grade 3 (Egges) retinal haemorrhages were seen in 0.5% of infants imaged.

Acta Ophthalmologica 2015

Discussion

Fig. 2. Retinopathy of prematurity like ridge was seen in 0.9% of all infants imaged, which
resolved spontaneously.

Fig. 3. Salt and pepper retinopathy detection resulted in early diagnosis of systemic comorbidities
including cardiac disease.

Medical or surgical intervention of

any form was required in nine babies,
which accounted for 18.8% of abnormalities detected and 0.9% of all babies
screened. Seven of these nine babies
(0.7% of all babies) needed medical
attention with a systemic work-up
prompted after the ophthalmic diagnosis was made. Five of these had posterior
uveitis with linear perivasculitis; none
had any coexisting systemic features to
suspect an ocular pathology. One had
salt and pepper retinopathy (Fig. 3),
which prompted cardiac and auditory
referral and abnormalities were detected

in both systems thereafter. One baby

had posterior synechia and was evaluated for a possible intrauterine infection
with coexistent uveitis. This baby had no
other systemic symptoms (other than
growth retardation) at the time the
ocular diagnosis was made. Two of nine
babies (0.2% of all babies) required
surgical intervention. One baby (50% of
surgical diagnosis) had a unilateral cataract (Fig. 4) for which lensectomy with
anterior vitrectomy was performed. The
other baby was diagnosed to have retinoblastoma (Fig. 5), which was treated
appropriately.

Universal eye screening of newborn

infants is not a common practice even
in several developed countries. In the
United Kingdom (Royal College of
Ophthalmologists and the British Pediatric Association 1994) and the United
States of America (American Academy
of Pediatrics 2008), RRT is recommended at birth with formal visual
acuity assessment commencing around
3 years of age. In Sweden, over 90% of
babies are routinely examined for congenital cataract with positive results.
(Magnusson et al. 2013). However, the
limitations and fallacies of red reex are
known. Khan & Al-Mesfer, in their
study, showed how the red reex failed
miserably in detecting 100% of cases of
retinoblastoma (Khan and Al-Mesfer,
2005). In countries such as India and
other middle-income countries, eye
screening for all babies is not yet part
of the healthcare delivery system (Vinekar et al. 2007). Whereas mass training
of health workers and physicians to
perform red reex testing remains a
possible strategy, our study did not
aim to study or evaluate this.
The inadequacy of systems in detecting every child with an early eye
problem is evidenced by the prevalence
of infant blindness data worldwide
(Gilbert et al. 1999). Clearly, several
conditions are either present or evolve
during infancy and remaining undetected. Alternate early screening strategies need exploration.
In India, and perhaps in most other
developing countries, universality of
screening is a distant reality. The launch
of the ambitious RBSK programme in
India (Ministry of Health & Family
Welfare 2013) promises 27 million
infants born each year, universal screening of 30 conditions. Two of these 30
conditions pertain to ophthalmology
and are retinopathy of prematurity
and vision and eye diseases. However,
there are no operational guidelines yet
for the latter (Ministry of Health &
Family Welfare 2013). We aimed to test
the feasibility of introducing universal
screening piggybacking on our existing
tele-ROP programme. The outcomes
would help in formulating guidelines
for scalability of the nationwide expansion of this programme planned by the
Government.
In summary, this pilot study shows
that 48/1021 or 4.7% had ocular

Acta Ophthalmologica 2015

reported how early detection of fungal

vasculitis detected on routine Retcam
imaging had prompted early systemic
treatment, saving the life of the infant.
(Vinekar et al. 2011a,b).
(3) Highly signicant (4.2%), which
potentially changed the course of
vision or life, as in infants with cataract
and retinoblastoma, respectively.

Fig. 4. Cataract was detected and surgically managed early owing to imaging.

Fig. 5. Retinoblastoma was detected in one patient (2% of abnormalities) and would have gone
undetected without imaging.

abnormalities of which 9/1021 (0.89%)

required some form of intervention. Of
signicance is the fact that all these
infants were healthy and normal at the
time of examination. A universal
screening programme of this nature
will be most eective if it detects
conditions that can inuence vision or
life earlier than the natural history of
these conditions or if prior information
could alter the course of management
in these babies.
As there is no control of what we
nd, it is important to ascertain how
to interpret what we nd, especially in
a scenario where scalability is being
considered. In this regard, our positive

ndings could be categorized into the

following:
(1) Interesting, but without major
clinical signicance, which include
self-resolving retinal haemorrhages,
spontaneously regressing ridges, retinal naevus and peripheral scars.
(2) Important (14.6%), with clinical
value which would include retinal vasculitis (which prompted cross-consultation and early management of other
comorbidities including cardiac and
auditory defects), retinal dysplasia
and inoperable persistent fetal vasculature syndrome (which resulted in early
enrolment into the vision rehabilitation
programme). In the past, we have

A recent publication from the Peoples Republic of China reported the

ndings of perinatal ocular examinations performed in 3573 healthy infants
and found 24.4% to have some abnormality. Of these, 88% had retinal
haemorrhages. Excluding retinal haemorrhages, they found that 2.2% of
babies screened had vision-threatening
ocular abnormalities (Li et al. 2013).
Our study has similarities in some of the
diagnoses, but diers in methodology:
(i) we did not exclude mothers with
known transmitted disease (e.g. STD,
HIV, hepatitis B) (ii) or infants with low
Apgar score or (iii) evidence of systemic
disease (iv) our study was piggybacked
on an existing tele-ROP screening programme (Vinekar et al. 2012, 2014a,b)
and was not a stand-alone project. We
chose all babies because we wanted to
explore the feasibility of universal
screening in a real-world scenario.
The studys limitations are as follows:
(i) rstly, it is a single public hospital
study. Although we had enrolled 52.3%
of babies born (1021/1954) during the
study period, this serves as a pilot study.
(ii) secondly, of the 1021, 990 (96.9%)
infants were born by Caesarian section.
This is an unusually high incidence and
does not mimic the situation in other
centres. It is likely that the prevalence of
haemorrhages is underestimated in our
study because it is higher in vaginally
born infants (Singh et al. 1980; Hughes
et al. 2006; Watts et al. 2013). Of the
933 babies born during the study period
but not enrolled, 893 (95.7%) were born
by Caesarian section. This is comparable with the enrolled cohort and would
potentially eliminate any selection bias
resulting from the mode of delivery (iii)
Maternal nutrition, antenatal risk factors and infections were not evaluated.
When the national scenario is considered, of the 27 million live births
annually (Indias Population 2014), and
an average national infant mortality
rate (IMR) of 58/1000, there would be
25.5 million survivors. Extrapolating
our results, 1 118 500 (4.7%) could

Acta Ophthalmologica 2015

Table 1. Estimating the countrys annual case load of infant eye conditions requiring medical and
surgical management.

Number of subjects
Ocular abnormalities (4.7%)
Medical conditions (0.69%)
Surgical conditions (0.2%)

Our
study

Indian
population

1021
48
7
2

25.5 million (after correction of infant mortality rate)

1 269 000
175 950
51 000

have had abnormal ocular ndings and

226 950 would have required medical
or surgical intervention (Table 1). This
indeed is a disturbing scenario and
warrants a detailed cost-impact study
to evaluate the economic and workforce feasibility of implementation and
possible expansion of this programme
to other centres.
In conclusion, this report establishes
that wide-eld, portable ocular digital
imaging is safe and eective in screening all live births in the Indian
scenario. Our pilot study shows that
approximately 0.9% of so-called normal infants were detected with treatment requiring conditions that resulted
in early and eective management that
altered the course of their ocular
and systemic diseases. To expand the
scope of eye care for infants from
retinopathy of prematurity towards
universal screening which the Government has mandated, we have
launched the FOREVER programme
focus on ROP, eye care, vision, eye
cancer and rehabilitation to address
more comprehensively, the visual and
ocular needs of all live births. This
pilot of universal imaging has provided the initial data in exploring
strategies that may be considered in
operationalizing
this
programme.
However, for increasing the scalability
of universal screening, low-cost cameras made available at all centres will
be necessary together with logistical
consideration of training and accrediting imaging technicians.
With the shortage of paediatric eye
specialists in our country, imaging
performed by accredited non-physicians and interpreted by remote experts
could be a viable strategy of solving the
lack of access to subspeciality eye care
in the Indian scenario. (Vinekar et al.
2014a,b). Hence, imaging at or immediately after birth oers the opportunity for the family to bring their infant
into contact with the healthcare delivery system. Additional requirements

include the availability of large image

reading centres with a facility to create
a national registry (Holmstr
om et al.
2014). The expanded multicentre ongoing study and cost-utility analysis will
further determine the feasibility of its
nationwide implementation.

References
American Academy of Pediatrics, Section on Ophthalmology, American Association for Pediatric
Ophthalmology and Strabismus, American Academy of Ophthalmology, American Association of
Certied Orthoptists (2008): Red reex examination in neonates, infants, and children. Pediatrics
122: 14011404.
Charan R, Dogra MR, Gupta A & Narang A (1995):
The incidence of retinopathy of prematurity in a
neonatal care unit. Indian J Ophthalmol 43: 123
126.
Egge K, Lyng G & Maltau JM (1980): Retinal
haemorrhages in the newborn. Acta Ophthalmol
(Copenh) 58: 231236.
Gilbert CE, Anderton L, Dandona L & Foster A
(1999): Prevalence of visual impairment in children: a review of available data. Ophthalmic
Epidemiol 6: 7382.
Holmstr
om G, Hellstr
om A, Jakobsson P, Lundgren P,
Tornqvist K & Wallin A (2014): Evaluation of new
guidelines for ROP screening in Sweden using
SWEDROP a national quality register. Acta
Ophthalmol [Epub ahead of print].
Hughes LA, May K, Talbot JF & Parsons MA
(2006): Incidence, distribution, and duration of
birth-related retinal hemorrhages: a prospective
study. J AAPOS 10: 102106.
Hungi B, Vinekar A, Datti N, Kariyappa P, Braganza S, Chinnaiah S, Donthi K & Shetty B (2012):
Retinopathy of prematurity in a rural Neonatal
Intensive Care Unit in South India a prospective
study. Indian J Pediatr 79: 911915.
Indias Population 2014. Available online at http://
www.indiaonlinepages.com/population
/india-current-population.html. Accessed on Feb
7th, 2014.
Khan AO & Al-Mesfer S (2005): Lack of ecacy of
dilated screening for retinoblastoma. J Pediatr
Ophthalmol Strabismus 42: 205210.
Li LH, Li N, Zhao JY, Fei P, Zhng GM, Mao JB &
Rychwalski PJ (2013): Findings of perinatal ocular
examination performed on 3573, healthy full-term
newborns. Br
J Ophthalmol 97: 588591.
Magnusson G, Bizjajeva S, Haargaard B, Lundstr
om
M, Nystr
om A & Tornqvist K (2013): Congenital
cataract screening in maternity wards is eective:
evaluation of the Paediatric Cataract Register of
Sweden. Acta Paediatr 102: 263267.

Ministry of Health and Family Welfare (2013): Operational Guidelines for Rashtriya Bal Swasthya
Karyakram. Child health screening and early intervention services under NRHM. http://mdm.nic.in/
Files/School%20Health%20Programme/Nutrition_
Support/Rastriya_Bal_Swaasthya_Karyakram.pdf
Royal College of Ophthalmologists and the British
Paediatric Association (1994): Ophthalmic services
for children. Report of joint working party.
London: Royal College of Ophthalmologists and
the British Paediatric Association.
Singh YP, Gupta SL, Jain IS, Gupta A & Bhakoo
ON (1980): Congenital ocular abnormalities of the
new born. J Pediatr Ophthalmol Strabismus 17:
162165.
Vinekar A (2011): IT-enabled innovation to prevent
infant blindness in rural India: the KIDROP
experience. J Indian Bus Res 3: 98102.
Vinekar A, Dogra MR, Sangtam T, Narang A &
Gupta A (2007): Retinopathy of prematurity in
Asian Indian babies weighing greater than 1250
grams at birth: ten year data from a tertiary care
center in a developing country. Indian J Ophthalmol
55: 331336.
Vinekar A, Avadhani K, Braganza S, Shetty B,
Dogra M & Gilbert C (2011a): Outcomes of a
protocol-based management for zone 1 retinopathy of prematurity: the Indian Twin Cities ROP
Screening Program report number 2. Am J Ophthalmol 152: 712713.
Vinekar A, Avadhani K, Maralusiddappa P, Prabhu
VM, Mahendradas P & Indumathi VA (2011b):
Retinal vasculitis as an early indicator of systemic
candidal abscesses in a premature infant. J
AAPOS 15: 9697.
Vinekar A, Avadhani K, Dogra M, Sharma P,
Gilbert C, Braganza S & Shetty B (2012): A novel,
low-cost method of enrolling infants at risk for
Retinopathy of Prematurity in centers with no
screening program: the REDROP study. Ophthalmic Epidemiol 19: 317321.
Vinekar A, Gilbert C, Dogra M, Kurian M, Shainesh
G, Shetty B & Bauer N (2014a): The KIDROP
model of combining strategies for providing retinopathy of prematurity screening in underserved
areas in India using wide-eld imaging, tele-medicine, non-physician graders and smart phone
reporting. Indian J Ophthalmol 62: 4149.
Vinekar A, Jayadev C & Bauer N (2014b): E-ROP vs
KIDROP: why tele-medicine in retinopathy of
prematurity is necessary in middle-income countries. JAMA Ophthalmol doi:10.1001/jamaophthalmol.2014.4913. [Epub ahead of print].
Watts P, Maguire S, Kwok T, Talabani B, Mann M,
Wiener J, Lawson Z & Kemp A (2013): Newborn
retinal hemorrhages: a systematic review. J
AAPOS 17: 7078.

Received on June 24th, 2014.

Accepted on January 12th, 2015.
Correspondence:
Anand Vinekar, MD, FRCS
Head, Department of Pediatric Retina
Narayana Nethralaya Postgraduate Institute
of Ophthalmology
Bangalore
India
Tel: +91 80 66121400
Fax: +91 80 23377329
Email: anandvinekar@yahoo.com

This study was not externally funded.