Prematurity and the Eye

Gerd Holmström

Prematurely born children have an increased risk to develop visual and eye problems
compared to children born at term. In the neonatal period they may develop retinal disease,
while later there is a risk of various visual dysfunctions.
Retinopathy of Prematurity (ROP)
Retinal disease in prematurely born infants was first decscibed by Terry in the beginning of
the 1940ies and was given the name Retrolental fibroplasia (RLF). In the middle of the
1980ies the disease was given a new name, Retinopathy of prematurity (ROP), and a new
classification by an international committee. ROP was divided into five stages, where stage 5
represents a total retinal detachment and a blind eye. Recently, a revised classification has
been presented (ICROP 2005), describing aggressive and rapidly progressing ROP in the
most immature infants. Without adequate treatment this type of ROP most often leads to
progression to stage 5 ROP and blindness. ROP stages 2 and 3 and Aggressive ROP are
illustrated in Figure 1.
Early stages of ROP (ROP 1 and 2) often regress spontaneously, while more severe stages
need treatment to prevent retinal detachment. In the 1970ies, treatment of the peripheral retina
with so-called cryo therapy was introduced, to destroy the anoxic part of the eye that was
stimulating the deleterious production of growth factors and vascular proliferations. In the
1990’s laser treatment has become the method of choice. Laser therapy is also a destructive
method, but more beneficial for the eye. Given at a correct time, it is successful in most cases
7 8
of ROP. If retinal detachment occurs, so-called vitrectomy may be performed. In the most
advanced cases with total retinal detachment (ROP stage 5), the chance of useful visual
function is, however, negligible regardless of surgical method.
A. B.
C. D.
Figure 1. Normal fundus, including retina, optic disc and vessels (A). ROP stage 2, including
ridge at the border between vascularised and non-vascularized retina (B). ROP stage 3,
including vascular proliferations (C). Aggressive posterior ROP (D). (ICROP 2005)
Incidence of ROP
The incidence of ROP is related to the quality of neonatal care. In poor countries almost no
ROP occurs since prematurely-born infants rarely survive. In middle-income countries with
an increasing survival, there is a new epidemic of ROP leading to severe visual impairments
and blindness. In Sweden several population-based studies on the incidence of ROP have
7 9
been performed during the last decades. Twenty years ago we reported an ROP incidence of
40% in infants with a birth weight of 1500 grams or less born in the Stockholm County during
the years of 1988 to 1990 (Holmström et al 1993). A new study in exactly the same
geographical area 10 years later, revealed a similar incidence but a change in the distribution
of ROP. The more ”mature” infants had a reduced incidence of ROP, while the most
immature babies, previously not surviving, had the highest risk of ROP and particularly
severe ROP, see figure 2. (Larsson et al 2002).
Figure 2. Probality of ROP in relation to gestational age at birth (weeks) in two consecutive
studies on ROP (1988 to 1990 and 1998 to 2000, respectively) in the Stockholm area of
Sweden (Larsson et al 2002).
During the last decade, improvements in neonatal care have resulted in a new population of
extremely immature babies with a high risk of various complications. A recent study
enclosing all infants in Sweden born before 27 weeks of gestations from 1 April 2004 to 31st
March 2007, revealed that 70% of the infants survived up to one year of age (EXPRESS
Group 2009). Major morbidities , i.e. intraventricular haemorrhage, broncopulmanry
dysplasia and / or ROP, were found in 45 % of these infants. The ophthalmic data were
separately collected and analysed. We recently reported that 73% of this population developed
8 0
ROP; mild ROP in 38 % and severe ROP in 35 % (Austeng et al 2009). Treatment was
performed in 20% of the infants. The most immature infants had the highest risk to develop
ROP and to reach treatment criteria, and they also developed ROP earlier and were treated
earlier than those born at a slightly higher gestational age (Austeng et al 2009).
Screening for ROP
ROP is one of the few causes of childhood blindness, which in most cases can be prevented.
Screening for ROP with repeated eye examinations in the neonatal period is therefore of
utmost importance for detection of severe ROP and for initiation of treatment at a correct
time. Since the incidence of ROP is related to the neonatal care and socio-economic situation
of the country, guidelines for screening have to be adapted to the country per se.
In Sweden, the first national guidelines were suggested after the population-based study in
1988-1990 (Holmström et al 1993). ROP screening was recommended to include all infants
born before 33 weeks of gestation and to start at 5 to 6 weeks of life and continue with one to
two weeks interval up to term or until regression of ROP. In case of progressing ROP, the
intervals between examinations were reduced and treatment was performed if indicated.
Based on the results of the incidence study in the Stockholm area during 1998 to 2000,
modification of screening recommendations with reduction of inclusion criteria to infants
born before the 32nd week of gestation were proposed (Larsson & Holmström 2002).
Analyses of the natural history of ROP in infants with a gestational age of less than 27 weeks
at birth have revealed new information on the importance of time and retinal location of ROP
at onset of the disease (Austeng et al 2010). These findings will probably lead to
modifications of the screening guidelines in the group of extremely preterm infants.
To further improve the quality of ROP screening, a national web based register for ROP, the
so-called SWEDROP, has been developed ( The register was started in late
2006, has a steering group of six regional responsible ophthalmologists and today has a cover
of around 80% of the prematurely-born infants in the country. In the near future we hope to
have reliable data to further modify and improve the screening guidelines, leading to
improved outcome for the infants.
8 1
Other ophthalmological sequels in prematurity
Retinopathy of prematurity (ROP) is a neonatal complication of the eye. Long-term follow-up
studies in childhood and adolescence, however, have revealed various visual dysfunctions in
prematurely-born children. Population-based studies on children of 10 years age in Denmark,
United Kingdom and New Zealand have reported increased risks to develop refractive errors,
strabismus, visual field defects, contrast sensitivity defects, and visual perceptual deficits.
Children from the population-based study in Stockholm during 1988 to 1990 have been
followed up repeatedly up to 10 years age (Holmström et al 1998, 1999). At that age, an
extensive ophthalmological follow-up was performed together with a similar group of
children born at term in the same geographical area. The visual acuity was found to be
reasonable in a majority of the preterm children and only 2 % had a visual impairment ( 0.3),
while a significant proportion had a subnormal visual acuity (Larsson et al 2005). Contrast
sensitivity was slightly, but significantly, reduced (Larsson et al 2006), strabismus was much
commoner (Holmström et al 2006), visual fields were slightly reduced (Larsson et al 2004)
and refractive errors (Larsson 2003, Holmström et al 2005, Larsson et al 2006) were much
more common in the preterm children compared to those born at term. The various
ophthalmologic sequels were most common in children with previous severe ROP and with
neurological problems, but also the children without such complications had an increased risk
to develop ophthalmic problems. Finally, our studies revealed a significant correlation
between visual dysfunction at 10 years and various school problems (Holmström & Larsson
Ophthalmological follow-up in prematurely-born children
As illustrated above, prematurely-born children indeed have an increased risk to develop
various ophthalmological problems and dysfunctions. Major sequelae are well-known, while
more subtle dysfunctions recently have been reported to affect the quality of life of these
children. Hence, the question arises if these children ought to be offered a long-term followup,
and in that case, which children and when and by whom. There is no consensus in the
literature, but some authors suggest frequent follow-up of all prematurely-born children
during childhood, some suggest follow-up of all children with ROP, while some propose
follow-up of only those with severe ROP and neurological sequealae. The Stockholm children
born in 1988 to 1990 had been followed up repeatedly up to 10 years, giving us the possibility
to evaluate risk factors which could predict visual dysfunction at 10 years. Multiple regression
8 2
analyses revealed that, apart from children who had been treated for ROP and children with
neurological sequelae, also those with significant refractive errors (anisometropia and
astigmatism) at 2.5 years, had an increased risk to develop visual dysfunction, according to
our definition, at 10 years of age (Holmström&Larsson 2008). Based on our findings, we
recommend that children who have been treated for ROP and children with obvious
neurological lesions have a regular ophthalmic follow-up. Regarding the children with
untreated ROP and without ROP, we propose that at least one follow-up is offered at around
2.5 years when the refraction is stabilized and amblyopia, with or without strabismus, is still
treatable. This examination can preferably be performed by an orthoptist with a possibility to
refer to a paediatric ophthalmologist, when indicated.
Prematurely-born infants and children are a risk group for various visual and
ophthalmological problems. Screening programmes to detect and treat these children are
necessary, but will continuously need to be modified parallel to new achieved knowledge. An
upcoming 6.5-year follow-up of the Swedish cohort of 500 extremely preterm infants born
before 27 weeks of gestation will hopefully provide us with new knowledge of the visual
function in this group of children.
Austeng D, Källen K, Ewald U, Jakobsson P & Holmström G (2009). Incidence of
retinopathy of prematurity in infants born before 27 weeks’ gestation in Sweden. Arch
Ophthalmol, 127:1315-9.
Austeng D, Källen K, Ewald U, Wallin A & Holmström G (2009). Treatment of ROP in a
population of extremely preterm infants born before the 27th week of gestation in Sweden. Br
J Ophtalmol, Nov 30, Epub ahead of print.
Austeng D, Källn K, Hellström A, Tornqvist K & Holmström G (2010): Natural history of
ROP in a population of extremely preterm infants born before the 27th week of gestation in
Sweden. Arch Ophthalmol, In press.
EXPRESS group (2009): One-year survival of extremely preterm infants after active perinatal
care in Sweden. JAMA, 301: 2225-2233.
8 3
Holmström G, el Azazi M, Jacobson L & Lennerstrand G (1993): A population based,
prospective study of the development of ROP in the Stockholm area of Sweden. Br J
Ophthalmol, 77: 417-23.
Holmström G, el Azazi M & Kugelberg U (1998): Ophthalmological follow-up of preterm
infants – a population-based study of the refraction and its development. Br J Ophthalmology,
82: 1265-71.
Holmström G, el Azazi M, Kugelberg U. Ophthalmological follow-up of preterm infants – a
population-based study of visual acuity and strabismus (1999). Br J Ophthalmology, 83: 143-
Holmstrom GE & Larsson EK (2005): Development of spherical equivalent refraction in
prematurely born children during the first 10 years of life: a population-based study. Arch
Ophthalmol, 123: 1404-11.
Holmström G, Rydberg A & Larsson E (2006): Strabismus and its development in 10-yearold
prematurely-born children – a population-based study: J Pediatr Ophthalmol Strabismus,
Holmström G & Larsson E (2008): Longterm follow-up of visual functions in prematurelyborn
children – a prospective population-based study up to 10 years of age. JAAPOS, 12:157-
International Commitee for the Classification of Retinopathy of Prematurity (ICROP) (2005).
The International Classification of retinopathy of prematurity Revisited. Arch Ophthalmol,
123: 991-999.
Larsson E, Carle-Petrelius B, Cernerud G, Ots L, Wallin A & Holmström G (2002): Incidence
of ROP in two consecutive population-based studies. Br J Ophthalmology, 86: 1122-26.
Larsson E & Holmström G (2002): Screening for ROP – evaluation and modification of
guidelines. Br J Ophthalmology, 86: 1399-1402.
Larsson E, Rydberg A & Holmström G (2003): A population-based study of the refractive
outcome in 10-year-old preterm and full-term children. Arch Ophthalmol, 121: 1430-1436
8 4
Larsson E, Martin L & Holmström G (2004): Peripheral and central visual fields in 11 year
old children who had been born prematurely and at term. Pediatric Ophthalmol Strab, 41: 39-
Larsson EK, Rydberg AC & Holmstrom GE (2005): A population-based study on the visual
outcome in 10-year-old preterm and full-term children. Arch Ophthalmol, 123: 825-32.
Larsson E, Rydberg A & Holmstrom G (2006): Contrast sensitivity in 10 year old preterm and
full term children: a population-based study. Br J Ophthalmol, 90: 87-90.
Larsson EK & Holmström GE (2006): Development of astigmatism and anisometropia in
preterm children during the first 10 years of life. Arch Ophthalmol, 124: 1608-14.

Ovanstående är ett utdrag ur:

Advances in Pediatric Ophthalmology Research. Gunnar Lennerstrand and Gustaf Öqvist Seimyr, Eds. The Sigvard & Marianne Bernadotte Research Foundation for Children Eye Care. Stockholm, 2010.

Utgiven med anledning av Stiftelsens 20-års jubileum.

Copyright 2023 © All rights Reserved. Hemsida Webbdesign Interwebsite Webbyrå