C L I N I C A L A N D E X P E R I M E N T A L

OPTOMETRY
cxo_600 1..14

REVIEW

To prescribe or not to prescribe? Guidelines for spectacle

prescribing in infants and children

Clin Exp Optom 2011 DOI:10.1111/j.1444-0938.2011.00600.x

Susan J Leat BSc PhD FCOptom FAAO This paper discusses the considerations for prescribing a refractive correction in infants
School of Optometry, University of and children up to and including school age, with reference to the current literature.
Waterloo, Waterloo, Ontario, Canada The focus is on children who do not have other disorders, for example, binocular vision
E-mail: [email protected] anomalies, such as strabismus, significant heterophoria or convergence excess. However,
refractive amblyogenic factors are discussed, as is prescribing for refractive amblyopia.
Based on this discussion, guidelines are proposed, which indicate when to prescribe
spectacles and what amount of refractive error should be corrected. It may be argued that
Submitted: 26 February 2010 these are premature because there are many questions that remain unanswered and we
Revised: 19 January 2011 do not have the quality of evidence that we would like; the clinician, however, make
Accepted for publication: 7 February decisions on whether and what to prescribe when examining a child. These guidelines
2011 are to aid clinicians in their current clinical decision making.

Key words: amblyopia, anisometropia, astigmatism, childrens vision, hyperopia, myopia, refractive error

There are numerous guidelines that have among optometrists and ophthalmolo- evidence-based guidelines for infants and
been published to help optometrists and gists. The Royal College of Ophthalmolo- young children up to the age of three
ophthalmologists when prescribing for gist guidelines6 were developed by a group years, which are similar in many respects
refractive errors in infants and children. of different eye care professionals, includ- to those given by Leat, Shute and Westall.8
The American Academy of Ophthalmol- ing paediatric ophthalmologists, orthop- Marsh-Tootle11 and Ciner12 published
ogy has published guidelines based on tists, an ophthalmic epidemiologist and an quite comprehensive recommendations in
consensus of opinion among an expert optometrist. Several of these guidelines their textbook chapters.
panel,1 while Miller and Harvey2 suggested are only for a single age (see Directorate The purpose of this paper is to review
recommendations based on consensus of Continuing Education and Training the current evidence, to update these
among members of the American As- [DOCET] recommendations by Far- guidelines and to provide more detail, so
sociation for Pediatric Ophthalmology brother7), an unspecified age6 or a wide that the clinician can see how each guide-
and Strabismus (AAPOS). The American range of ages or refractive errors.3 line relates to the current evidence.
Optometric Association provides guide- Some authors have also developed rec- Although there are many research ques-
lines for correction of hyperopia and ommendations. Leat, Shute and Westall8 tions that still need to be answered, the
myopia based on consensus among expert and Leat9 previously published guidelines clinician has to make a management deci-
optometrists,3,4 and Blum, Peters and Bett- on prescribing for infants and children, sion regarding the child who sits in the
man5 suggested guidelines for referral which were based on the best available chair today. The proposed guidelines are
from vision screening, based on consensus evidence at that time. Bobier10 provided to assist such decisions, based on our

2011 The Author Clinical and Experimental Optometry 2011

Clinical and Experimental Optometry 2011 Optometrists Association Australia 1
Management of refractive error in infants and children Leat

current level of knowledge. Of necessity, NATURAL HISTORY OF REFRACTIVE and degree of astigmatism in the first few
these must be reviewed frequently, ERROR FROM BIRTH TO THREE YEARS years. Of the studies with larger samples,
as knowledge in this area is rapidly There is now general agreement that the eight to 30 per cent have 1.00 D or more of
expanding. range of refractive errors is wider at birth astigmatism at one to two years, four to 24
The proposed guidelines concentrate and in the first year of life than in later per cent at three to four years and two to 17
on the management of refractive error. childhood, that most infants are hyper- per cent at six to seven years24 (see Harvey
Prescribing as part of the management of opic and that the average cyclop- and colleagues24 for more detail). The lon-
ocular misalignment (heterotropia, sig- legic refractive error is approximately gitudinal study of Abrahamsson and col-
nificant heterophoria) or convergence +2.00 D13,14 with a standard deviation of leagues25 found that 90 per cent of Swedish
excess is not covered in detail; however, approximately 2.00 D. There is some children with astigmatism 1.00 D or more
refractive amblyogenic factors, however, uncertainty regarding the changes in the over the age of one year experienced a
are discussed, as is prescribing for refrac- first three months, with some studies decrease in their astigmatism.25 Harvey and
tive amblyopia. showing that the average refractive error colleagues24 found a sustained and higher
The format of the paper is as follows. increases during this time and others prevalence of astigmatism in a Native
First, the main considerations for prescrib- suggesting that it remains static or American population. Those studies that
ing from birth to six years of age, followed decreases.14,15 From three months to show a decrease in prevalence are not in
by school-age children, are discussed, 12 months, there is a period of fast emmet- agreement about when this process ends,
together with the best research evidence ropisation as shown by longitudinal1517 that is, at what age does the prevalence of
that exists to guide a decision to prescribe. and clinical cross-sectional studies.14 In a astigmatism stabilise and become adult-
When evidence from research is scarce or predominantly white sample, Mutti and like? Cross-sectional studies by Mayer and
poor, clinical opinion is added. The guide- colleagues16 showed that the average colleagues14 and Atkinson, Braddick and
lines, which result from this discussion, are cycloplegic spherical equivalent decreases French26 showed that the prevalence stabi-
provided in tabular format (Table 2) and from 2.16 D at three months to 1.36 D lises by 1.5 years. Cross sectional data from
this is followed by notes that relate to this at nine months. This is followed by Gwiazda and colleagues27 show a decreas-
table. a period of slower change until two ing prevalence until approximately three
years for hyperopes and four to five years, while their longitudinal data show
years for myopes.13,14,18,19 A more recent, that it does not stabilise until four to five
INFANTS AND CHILDREN FROM population-based, cross-sectional, Multi- years.13,28
BIRTH TO SIX YEARS Ethnic Pediatric Eye Disease (MEPED) As with spherical error, the rate of
study18 has shown differences between decease of astigmatism is generally associ-
When considering prescribing glasses for
ethnic groups. There was a higher preva- ated with the initial level,22,29 with those
a young child (birth to six years), the fol-
lence and mean hyperopia in Hispanic with higher amounts usually decreasing
lowing questions must be considered:
children compared with African Ameri- more rapidly. With regard to the type of
1. Is the refractive error within the
cans. Table 1 shows a summary of the astigmatism, there is a higher prevalence
normal range for the childs age?
means and lower and upper 95% limits of of all types in infancy. Significant with-the-
2. Will this particular childs refractive
cycloplegic spherical refractive error rule (WTR), against-the-rule (ATR) and
error emmetropise?
according to age calculated from 1.96x the oblique astigmatism are all more common
3. Will this level of refractive error disrupt
standard deviation from studies which in young children than adults.14,22,27 Of
normal visual development or func-
provide this information.14,16,18,19 these types, oblique astigmatism is the
tional vision?
A few infants are myopic at birth and least common.14, 22 There is general agree-
4. Will prescribing spectacles improve
most of those who are either myopic or ment that all types of astigmatism
visual function or functional vision?
hyperopic will emmetropise.13,20,21 The decrease, with infants losing approxi-
5. Will prescribing glasses interfere with
rate of emmetropisation is generally pro- mately two-thirds of their astigmatism
the normal process of emmetropisa-
portional to the initial error. Thus, those between nine and 21 months,22 and that
tion?
who start off close to emmetropia or with a most of this loss occurs in the first 1.5 to
The evidence which helps the clinician to
low amount of hyperopia show little 2 years of life.13,14,2628 Some studies show
answer each of these questions is reviewed
change, while those who have higher that WTR decreases more rapidly,22 while
below.
ametropia generally show greater and others show that ATR is lost more rapidly,
faster changes.16,22 even switching to WTR in some cases.27
Is the refractive error within the There is also a higher prevalence of astig- Most studies have shown that ani-
normal range for the childs age? matism at birth, with as many as 69 per cent sometropia is more common in infants
To answer this question we need to know of full-term newborns having astigmatism than adults. Varghese and colleagues23
the natural history of the refractive error 1.00 D or more.23 In most populations and Zonis and Miller30 reported that 30
and the normal range at each age. there is a decrease in both the prevalence and 17 per cent of newborns, respectively,

Clinical and Experimental Optometry 2011 2011 The Author

2 Clinical and Experimental Optometry 2011 Optometrists Association Australia
 Management of refractive error in infants and children Leat

have anisometropia greater than 1.00 D.

Table 1. Means and upper and lower 95% ranges of cycloplegic spherical refractive error according to age. The various data are placed to compare equivalent or
range (D)

sample, longitudinal, predominantly white)

Ingram, Traynar and Walker31 and Abra-
Upper
95%

4.7
3.4

3.1

3.2
hamsson, Fabian and Sjstrand32 found

Ingram and Barr19 (population sample,

that spherical anisometropia remains

longitudinal, predominantly white)

Mutti and colleagues16 (population
more common in children compared with
range (D)
Lower

adults up to at least four years of life,

95%

-0.4
-0.7

-1.2

-1.0
while the more recent, population-based
MEPED study33 found differences between
ethnic groups, the prevalence of ani-
SE (D)
Mean

0.95
2.2
1.4

1.1
sometropia decreasing from the first year
to the second year of life in children of
(months)

Hispanic origin, but not for African-

Age

6
9

12

42
American children. The studies of Abraha-
msson, Fabian and Sjstrand32 and
Ingram, Traynar and Walker31 were longi-
range (D)
Upper

tudinal and reported that while approxi-

95%

4.1

3.9

3.8
4.0
4.0
mately seven to 11 per cent of one to four
year old children have spherical ani-

nearly equivalent ages across studies. MEPED = Multi-Ethnic Pediatric Eye Disease, SE = spherical equivalent
range (D)

sometropia of 1.00 D or more (compared

Lower

-1.18
95%

-1.5

-1.7
-1.4
-1.2 with zero to five per cent of school
Hispanic
MEPED study18 (cross-sectional, population based,

children3436), it is not the same children

who make up this percentage. Some chil-
SE (D)
Mean

1.3

1.0

1.1
1.3
1.4

dren gain anisometropia during this

period, while others lose it.31,32 This led
based on right eye data)

Abrahamsson, Fabian and Sjstrand32 to

(months)

611

1223

2435
3647
4859
Age

postulate that there are different rates of

emmetropisation between the two eyes,
resulting in transient anisometropia. It is
range (D)
Upper

thought that these transient anisometro-

95%

3.4

3.3

3.5
3.9
4.0

pias are of relatively lower level, for

example, 2.00 or 2.50 D or less and may
range (D)

not lead to amblyopia. Higher levels of

African American
Lower
95%

-2.2

-1.2

-1.7
-1.7
-1.7

anisometropia (3.00 D or more) are more

likely to remain.37
SE (D)
Mean

0.6

0.7

0.9
1.1
1.1

NATURAL HISTORY OF REFRACTIVE

ERROR IN THREE- TO SIX-YEAR-OLDS
There is less change occurring during
(months)

611

1223

2435
3647
4859
Age

this period of life. Gwiazda and col-

leagues13 showed that there is still a slow
movement of the refractive error towards
range (D)
Upper

emmetropia during this period. This was

95%

5.5
4.4
5.1
5.2
4.4
3.6

3.2
3.1
2.9
3.1
2.6
2.9
cross-sectional, predominantly white

evidenced by the finding that the smallest

standard deviation of the populations
Mayer and colleagues14 (clinical,

spherical equivalent refraction occurred

range (D)
Lower
95%

-1.1
-0.2
-0.3
-1.2
-0.8
-1.0

0.0
-0.6
-0.5
-0.6
-0.6
-0.6

at six years,13 at which age the mean is

0.70 to 1.00 D.38,39 There is also less
change occurring in astigmatism in this
SE (D)
Mean

age group compared with younger chil-

2.2
2.1
2.4
2.0
1.8
1.3

1.6
1.2
1.2
1.3
1.0
1.1

dren, although the longitudinal studies

population)

of Gwiazda and colleagues13,28 showed

(months)

that there is still some decrease in astig-

Age

1.5
2.5

matism until approximately four to five

12
18
24
30
36
48
1

4
6
9

years.

2011 The Author Clinical and Experimental Optometry 2011

Clinical and Experimental Optometry 2011 Optometrists Association Australia 3
Management of refractive error in infants and children Leat

Hyperopia
When to consider prescribing What to prescribe Comments, rationale and references

1. Outside the 95% range of Prescribe so as to leave the uncorrected hyperopia See Table 1 and Figure 1 for currently available data,
refraction at any age according to somewhat above the mean for the age (so as to give a which give spherical equivalent and 95% confidence
any currently available data. This slightly greater than average stimulus for limits spanning the ages zero to 4 years
guideline could be applied to other emmetropisationsee text)
refractive errors also, for example,
astigmatism.
2. 3 to 6 months if outside the 95% In addition to the level of hyperopia determined by Mutti and colleagues15 do not give a value of what exact
range cycloplegic refraction, factors that would indicate level of Mohindra refraction would be considered as
correction are VA poorer than 6/100 plus non-cycloplegic high, but from their data it appears that approximately
(Mohindra) refraction that is high15 and presence of 3.25 D of spherical equivalent is outside the normal
against-the-rule astigmatism.13 range (they subtract a correction factor of 0.75 D from
Give a partial prescription for both cylinder and sphere. the gross retinoscopy)
Prescribe for sphere as in 1.
3. 3.50 D in one or more meridian Give a partial prescription. This is based on the randomised clinical trials of Atkinson
at 1 year of age upwards Atkinsons protocol, based on the refraction in plus and colleagues46 and the natural history study of Ingram
cylinder format, at this age was: and colleagues106

Sphere: prescribe 1.00 D less than the least hyperopic

meridian
Cylinder: prescribe half of the astigmatism, if >2.50 D
Or use the approach in 1 above to determine the
correction for hyperopia.
4. >2.50 D at 4 years upwards Still give a partial correction for hyperopia, This is based on studies of visual function and functional
undercorrecting by approximately 1.00 to 1.50 D, which vision25,64,65 and Mayer and co-workers upper 95% range
is the mean hyperopia at this age. This undercorrection is which was 2.6 at 3 years and 2.9 at 4 years.14 For the
not because of emmetropisation (which is almost African American and Hispanic populations, a slightly
completed at this age), but because the child does not higher value of >3.50 D would be justifiable (see Table 1).
require full correction of hyperopia for good function.
5. 1.50 D in the school years A full or near full correction may be given at this age, as Studies on visual function show that hyperopia ranging
without symptoms emmetropisation has essentially ended9,11 from 1.00 D to 2.00 D may impact visual function and
Regarding lower amounts with functional vision8386
signs or symptoms, see text.
Astigmatism
6. >2.50 D at 15 months of age Give partial correction up to 3 to 4 years by which time Based on 15 months being the most critical period for the
upwards emmetropisation is largely completed, that is, decrease development of meridional amblyopia55 and population
cylinder by 1.00 D or give 50%10,46 studies, which show that approximately 5 to 10% of the
population have this amount of astigmatism.13,22,2628 This
is also the criterion used in the study by Atkinson and
colleagues46 for correction of astigmatism.
7. 2.00 D at 2 years of age Give partial cylinder up to 3 to 4 years, after which give Based on findings of better VA in children whose
upwards full cylinder46 astigmatism was corrected at this age20 and at 2 years
approximately 5 to 10% have astigmatism 2.00 D13,22,27
8. 1.50 D at 4 years upwards Give full cylinder, although in cases of previously Cowen and Bobier107 found that the 95th percentile for
uncorrected high astigmatism, a reduced prescription astigmatism was 1.25 D in children of mean age 4 years.
may be given initially, to allow the child to adapt Five per cent or less have 2.00 D and 5 to 20% have
between 1.00 and -2.00 D astigmatism at this age.13,2628
Roch-Levecq and colleagues66 reported functional benefits
of correcting 1.50 D astigmatism in 4- to 5-year-olds.
Also see clinical recommendations.11
9. Correct oblique astigmatism My clinical instinct would be to correct approximately 3/4 Oblique astigmatism is a risk factor for amblyopia.25
1.00 D from 1 year onwards to the age of 2 and then correct the full amount Mayer and colleagues14 show that oblique astigmatism of
1.00 D is rare after 12 months.
10 0.75 D at school age without Prescribed as in 8 above Congdon and colleagues96 found that correction of this
symptoms. For lower amounts level improved VA
with signs and symptoms, see
text.

Table 2. Guidelines for prescribing for refractive error in children. Guidelines in italics are those that are based on clinical opinion
rather than a research evidence base. MEPED = Multi-Ethnic Pediatric Eye Disease, VA = visual acuity

Clinical and Experimental Optometry 2011 2011 The Author

4 Clinical and Experimental Optometry 2011 Optometrists Association Australia
 Management of refractive error in infants and children Leat

When to consider prescribing What to prescribe Comments, rationale and references

Anisometropia
11. Anisometropia with amblyopia Correct the full anisometropia and astigmatism but
correct the hyperopia or myopia according to age
12. 3.00 D at 1 year upwards Prescribe the full anisometropia if amblyopia is already This is based on reports by Abrahamsson and
present (see above). If there is no amblyopia, a reduced colleagues37 that 3.00 D of anisometropia is less likely
anisometropic prescription could be considered (for to be transient
example, prescribing 1.00 D less than the full difference
between the eyes) and prescribing for astigmatism and
spherical error according to age. According to
Marsh-Tootle,11 if amblyopia can be demonstrated to be
absent, a prescription is not necessary.
13. 1.00 D but <3.00 D after 1 year Monitor first over 4 to 6 months. If it persists, prescribe This is based on reports of transient anisometropia31,32,36
of age as in 11 above.
14. 1.00 D of spherical hyperopic Prescribe as in 11 above. If amblyopia is absent, may These levels of anisometropia have been found to be
anisometropia, 2.00 D of monitor first. amblyogenic at this age32,60,61
spherical myopic anisometropia or
1.50 D of cylindrical
anisometropia after 3.5 years of
age
Myopia
15. <-5.00 D, during the first year Reduce by 2.00 D. Undercorrect because Clinical opinion and guidelines agree to prescribe when
emmetropisation does occur for myopes.13,21 -5.00 D1,12 but not less than -3.00 D.4 In the MEPED
study,18 less than 1% of children between 6 to 72 months
had <4.00 D of myopia.
16. <-2.00 D myopia from one year or Reduce by 0.50 D or 1.00 D until school age. The MEPED18 study showed that <-1.2 to -1.7 is the
when child is walking Undercorrect because some emmetropisation is still lower end of the 95% range in African Americans and
occurring.13,21 Hispanics in the US. Clinical opinion varies widely,
between correcting -0.75 D to -4.00 D, in infants and
toddlers.1,2,4,7,11,101,108
17. 4 years to early school years <-1.00 D or lower amounts if it improves VA and the child Congdon and colleagues96 found that correction of
appreciates it, that is, correct for function. Can give full 0.75 D improved VA. Clinical opinion suggests
correction at this age. correcting <1.00 D to 1.50 D in preschoolers2,11,101,108
and <-0.50 D to <-2.00 D in school children.4,8,11,100102
18. School age myopia Prescribe full correction. Cases of myopia with near Guideline for bifocal correction based on the Correction of
esophoria and larger lag of accommodation (>0.43 D) or Myopia Evaluation Trial study for 6- to 11-year-olds81,82
with shorter habitual reading distances may be
considered for a +2.00 D addition progressive lens.
Aphakia or pseudophakia
19. In first few months Overcorrect by 2.00 to 3.00 D, because the childs world Intraocular lenses may be implanted at surgery109
is near, reducing to a single vision intermediate add of
1.00 to 1.50 D by 1 year.8,9 Contact lenses are often the
correction of choice.
20. 2 to 3 years onwards Distance correction with bifocals when the child can They will require bifocal/progressive addition lens
adapt to these correction for life109

Table 2. Continued

Will this particular childs cases, at least in the early years of life. studies40 and by some human data.15,22
refractive error emmetropise? Alternatively, those who will not emmetro- Mutti and colleagues15 showed how the
Although the majority of children will pise and who have a high refractive error probability of emmetropisation decreases
emmetropise, this is not true for all. We might benefit from spectacle correction. as hyperopia increases (Figure 1). The
would like to be able to predict those who There is some evidence that children with probability is less than 50 per cent for
will fully emmetropise, as there is likely to very high refractive errors are less likely to three-month-olds, who had a cycloplegic
be no need to prescribe spectacles in these emmetropise. This is suggested by animal spherical equivalent refraction greater

2011 The Author Clinical and Experimental Optometry 2011

Clinical and Experimental Optometry 2011 Optometrists Association Australia 5
Management of refractive error in infants and children Leat

1.0 amblyopia,37 with 30 per cent of those with

Probability of emmetropisation this level of anisometropia showing
0.9
increasing anisometropia over the follow-
0.8
ing nine years, 60 per cent developing
0.7 amblyopia and 90 per cent retaining ani-
0.6 sometropia of 1.00 D or more at the age of
five years. Children with lower amounts of
0.5
anisometropia, for example, up to 2.00 D,
0.4 are more likely to lose it.36,43 To summarise,
0.3 currently the only methods we have to
0.2 determine if a particular case of ani-
sometropia is transient and therefore does
0.1
not need intervention are:
0.0 1. monitor a child over a period of four to
2.00 3.00 4.00 5.00 6.00 7.00
six months
Sph. eq. ref. error at three months (D) 2. consider the visual acuitiesif am-
blyopia is already present it requires
Figure 1. The probability of reaching 2.00 D by 18 months of treatment and indicates that the ani-
age as a function of the level of cycloplegic spherical equiva- sometropia has been present for some
lent at three months of age. From Mutti DO, Mitchell GL, time
Jones LA, Friedman NE, Frane SL, Lin WK, Moeschberger 3. be aware that higher levels of ani-
ML, and Zadnik K. Accommodation, acuity, and their relation- sometropia (for example, 3.00 D or
ship to emmetropization in infants. Optom Vis Sci 2009; 86: more and particularly 5.00 D or
666676. Reproduced with permission. more37) are less likely to be transient.
Lastly, children with low vision are less
likely to fully emmetropise.44 Therefore,
these children can be prescribed, with the
main consideration being to optimise
than 5.00 D. Apart from the cycloplegic Apart from the predictive factors visual function.
refraction, visual acuity (VA) poorer than described above, namely, level of ametro-
6/100 and a higher non-cycloplegic pia, VA and type of astigmatism, which are Will this level of refractive error
Mohindra retinoscopic result also pre- still not as accurate as we would like, disrupt normal visual
dicted which infants would emmetropise. the other main tool that we have to predict development or functional vision?
Gwiazda and colleagues13 found that emmetropisation is to monitor the There is evidence that uncorrected high
hyperopic children with WTR astigmatism refraction. Those that emmetropise lose refractive error (hyperopia, astigmatism
show different patterns of emmetropisa- approximately one-half of their spherical and anisometropia) during the first few
tion compared with those with ATR astig- equivalent refractive error in the first year42 years of life is a risk factor for amblyopia.
matism. Although both groups lost their and approximately one-third between nine The studies of Abrahamsson and col-
astigmatism, the hyperopic children with and 21 months.22 With regards to astigma- leagues25 and Ingram and colleagues45
ATR astigmatism at six months main- tism, approximately two-thirds of the astig- indicate that there is an increased chance
tained their hyperopia (approximately matism is lost between nine and of monocular or binocular amblyopia in
2.00 D on average), while those with WTR 21 months.22 one-year-olds with 3.50 D or more in one
astigmatism lost their hyperopia (ending Currently, there is no way to predict with meridian, in four-year-olds if the most
up at approximately 0.75 D hyperopic at certainty whether a particular childs ani- hypermetropic meridian is 2.00 D or more
age six years). In contrast, Ehrlich and col- sometropia (measured at one point in and if there is increasing or unchanged
leagues22 did not find a similar relation- time) is transient or will remain into adult- refractive error between one and four
ship. Ingram and colleagues41 showed that hood (with the risk of amblyopia). There is years. Atkinson and colleagues20,46 showed
there is an association between lack of no simple relationship between the ani- that partial correction of hyperopia
emmetropisation and the presence of stra- sometropia measured at one time, whether greater than 3.50 D at nine to 11 months
bismus. What is not clear is whether stra- spectacles are prescribed and whether the resulted in improved VA at four years of
bismus interferes with emmetropisation or anisometropia persists or amblyopia devel- age and may reduce the incidence of
whether those who do not emmetropise, ops.32 Nevertheless, if the anisometropia is esotropia. Aurell and Norrsell47 found that
and thus maintain their higher hyperopia, 3.00 D or more at one year, there is a high infants who maintained more than 4.00 D
are more likely to develop strabismus. risk of it remaining and resulting in of hyperopia were more likely to develop

Clinical and Experimental Optometry 2011 2011 The Author

6 Clinical and Experimental Optometry 2011 Optometrists Association Australia
 Management of refractive error in infants and children Leat

esotropia. Clinical retrospective studies of Harvey and colleagues58 indicate that cor- Will prescribing spectacles
children with high bilateral uncorrected recting astigmatism at 4.75 years or later improve visual function or
hyperopia have also shown a connection may be too late to allow development of functional vision?
between poorer acuity and high hypero- optimal visual function, while the results of By visual function, we mean psychophysi-
pia. In children with 5.00 D or more, 25 to Atkinson and colleagues20 indicate that we cal measures of the sensory capability of
43 per cent have acuity of 6/12 or should correct astigmatism of 1.00 D or the visual system, such as VA or contrast
worse48,49 and 87 per cent have acuity more as early as two years to optimise acuity sensitivity, while functional vision is used
worse than 6/6.50 Poor accommodation development. If, however, we were to to refer to how the person as a whole is
and stereopsis have also been associated correct the levels based on Atkinson and able to use vision in performing everyday
with high hyperopia.51 It has been sug- colleagues,20 we might find ourselves pre- tasks, which are dependent on vision.62
gested52 that it is hyperopic children with scribing for up to 20 to 45 per cent of the With regards to visual function, there
poorer accommodation who may develop population, because according to some are few randomised clinical trials that
amblyopia and, consistent with this, studies, 20 to 45 per cent of two-year-olds have studied the effects of a prescription
Schoenleber and Crouch53 found that still have 1.00 D or more of astigma- in the pre-school age group. The two
none of their high hyperopes who were tism.13,14,27 This would not be clinically rea- studies by Atkinson and colleagues20,46
able to co-operate for amplitude testing sonable. Thus, the suggested values for were randomised clinical trials, in which
had sufficient accommodation to main- prescribing for astigmatism in Table 2 are one group of nine- to 11-month-olds with
tain a 50 per cent reserve of accommoda- based on the 95 per cent upper limits of the hyperopia of 3.50 D or more in the most
tion for extended periods of viewing. distribution of astigmatism with respect to hyperopic meridian was given a partial
With regards to astigmatism, researchers age, in addition to the evidence that higher spectacle correction and the other group
have sought associations between meridi- levels of astigmatism are associated with (controls) was not. The prescribing proto-
onal amblyopia and astigmatism with visual function deficits. col can be seen in Table 2, guideline 3.
mixed results, although they have shown We have already seen that anisometro- The incidence of strabismus and amblyo-
that recognition acuity and other measures pia of 3.00 D or more at one year is likely pia was reduced in the children who were
of visual function are decreased in astigma- to cause amblyopia,37 as is persisting ani- prescribed glasses compared with the con-
tism. The age of the child when the astig- sometropia of 1.00 D or more. Donahue59 trols in the first study but the incidence of
matism is present seems to be a factor. The showed that anisometropia after the age of strabismus was not reduced in the second
visual system may not be very sensitive to three years is more likely to cause amblyo- study. They were followed until the age of
uncorrected astigmatism in the first year of pia than before that age. It also appears four years, at which time more children in
life54 but from one year onwards, there is that different types of anisometropia the corrected group obtained a VA of
evidence that uncorrected astigmatism, might be more or less likely to cause better than 6/9 than in the control group.
particularly oblique astigmatism, is associ- amblyopia. In a cross-sectional study of The only other such clinical trial is that by
ated with meridional amblyopia.20,25,55 clinic patients, Weakley60 found that more Ingram and colleagues,63 in which infants
Dobson and colleagues56 found no evi- than 1.00 D of spherical hyperopic ani- aged six months with +4.00 D or more of
dence of meridional amblyopia in six- sometropia was associated with amblyopia hyperopia in one meridian were randomly
month-olds up to three-year-olds with and decreased steroacuity, while spherical assigned to spectacle or no spectacle treat-
astigmatism of 2.00 D or more, although myopic anisometropia had to be greater ment. The protocol was a little unusual as
the acuity for both vertical and horizontal than 2.00 D before amblyopia occurred. cycloplegic retinoscopy was performed at
gratings was decreased in children with Cylindrical anisometropia (either myopic one metre but a 1.75 D correction factor
astigmatism, which may be because most of or hyperopic) had to be greater than was subtracted. The spectacle prescription
the children with astigmatism were also 1.50 D before amblyopia occurred. Weakly appears to have been a dioptre under-
hyperopic. In three- to four-year-olds, also found that the degree of amblyopia corrected in both meridians, that is, the
1.50 D or more of astigmatism is associated increased with the amount of anisometro- full astigmatic correction was given.
with poorer recognition acuity, such that pia (of any kind). Dobson and colleagues61 They found no impact of the spectacle
for every dioptre increase in cylinder, there found somewhat similar results; amblyopia correction on the incidence of strabismus,
was a half-line decrease in VA.57 In older and intraocular differences of VA were even when compliance with wear was
children with 1.00 D or more of astigma- associated with 1.00 D or more of hyper- taken into account. They found a signifi-
tism (first corrected at the age of 4.75 to opic anisometropia and 2.00 to 3.00 D or cant difference in VA between the spec-
13.5 years), a range of visual functions more of cylindrical anisometropia. Stere- tacle and non-spectacle wearers only
(grating acuity, letter acuity, vernier acuity, oacuity seemed to be more sensitive to when compliance was taken into account,
contrast sensitivity and steroacuity) is the presence of anisometropia; 0.50 D or with the compliant spectacle wearers
impaired.58 After optical correction, there more of hyperopic, myopic or cylinder having better VA. Differences between the
was some improvement up to one year but anisometropia was associated with a studies are that the studies of Atkinson
deficits still remained. The results of decrease of stereoacuity. and colleagues20,46 prescribed a smaller

2011 The Author Clinical and Experimental Optometry 2011

Clinical and Experimental Optometry 2011 Optometrists Association Australia 7
Management of refractive error in infants and children Leat

percentage of the refraction and pre- showed that four- to seven-year-old chil- Will prescribing glasses interfere
scribed a little later in life than the dren with more than 2.00 D of uncor- with the normal process of
Ingram and colleagues63 study (nine rected hyperopia had poorer emergent emmetropisation?
months compared with six months). It literacy skills measured on several tests Experimental animal studies clearly show
also appears that in the Ingram and col- than emmetropes, although in this study that refractive correction will influence
leagues63 study, any controls who devel- the children with hyperopia performed the development of refractive error67,68
oped strabismus during the study were equally well on tests of visual motor and and therefore we need to consider this
prescribed treatment involving spectacles, visual perceptual skills. The fact that both possibility in humans also. The human evi-
occlusion and/or surgery. Atkinson and colleagues20,46 and Shankar, dence of whether a prescription for glasses
There is clinical evidence that amblyo- Evans and Bobier64 found poorer perfor- has some effect on emmetropisation is
pia due to high isometropic hyperopia mance on some but not all tests indicates equivocal and there are few randomised
responds to treatment with refractive cor- that the poorer performance of the clinical trials that can give solid evidence
rection,49,50,52,53 although the time-course hyperopic children does not seem to be in humans. In the study by Atkinson and
for improvement varies from one to part of a general developmental delay. colleagues,46 there was no difference in
several years.50,52 Many children in these The results of Rosner and Rosner65 indi- the reduction of hyperopia comparing
studies achieved a final VA of 6/7.5 or cate that prescribing for hyperopia those who were fitted with a partial pre-
better.49,52 On the other hand, the out- greater than 2.50 D before the age of four scription and the controls. Ingram and
comes for other children were not so years may reduce deficits in visual percep- colleagues41 also found no significant
good. The percentage of children whose tual skills later in life. In a recent study, difference overall. However, when they
final VA with spectacle correction was Roch-Levecq and colleagues66 showed re-analysed their intervention group
6/12 or poorer ranged from 11 to 50 per that to three- to five-year-olds with uncor- according to the amount of spectacle lens
cent.4850,53 Surprisingly, three of these rected hyperopia of 4.00 D or more, three wear, they did find a differencethe com-
studies found that the final outcome of VA year olds with 2.00 D or more of astigma- pliant spectacles wearers emmetropised
was not dependent on the age of first spec- tism and four to four- and five-year-olds less than the non-compliant spectacle
tacle prescription,49,50,52 which ranged with 1.50 D or more of astigmatism wearers or the controls. In the study by
from seven months to 12 years. To con- had poorer visuomotor skills and perfor- Ingram and colleagues41 it appears that a
clude, these clinical studies indicate that mance intelligence scores than a control greater percentage of the refractive error
moderate improvements can be obtained group with lower refractive errors. Impor- was corrected in the spectacle prescription
for children who already have bilateral tantly, after the children with these compared with the studies by Atkinson
refractive amblyopia due to hyperopia, but higher ametropias were prescribed and colleagues,20,46 which may have caused
do not indicate whether we can prevent glasses, their visuomotor skills perfor- the different resultsthere would have
amblyopia by even earlier spectacle pre- mance improved to the level of the been a smaller stimulus for emmetropisa-
scription. The best current evidence for control group in only six weeks, although tion. Friedman, Neumann and Abel-
prevention is based on the randomised it must be noted that they were not fol- Peleg48 reported retrospective clinical data
clinical trials of Atkinson and col- lowed longer than six weeks and there- of 39 children with high levels of ametro-
leagues20,46 and Ingram and colleagues63 fore a Hawthorne effect is almost pia, who were treated with spectacle cor-
described above. certainly in operation. These studies do rection at one to 2.5 years (we are not told
With regards to functional vision, there not prove a causal relationship between whether this was a partial or full correc-
are studies that have shown that young hyperopia and these skills, because there tion). Sixty-four per cent of the hyperopic
children with uncorrected hyperopia are likely to be many other influences, eyes, 60 per cent of the astigmatic eyes and
perform more poorly on some tests. such as IQ and family background, which 50 per cent of the myopic eyes showed
Atkinson and colleagues46 followed their interact in a complex fashion. To prove some decrease of the ametropia up to the
corrected and uncorrected hyperopic a causal relationship, the impact of age of seven to 10 years. However, this was
infants to the age of 5.5 years. At the age spectacle correction should be studied not compared with a control group that
of three years, they still had 3.50 to 4.00 D either over a longer period of time (to had no correction. An interesting study
of hyperopia on average. At 5.5 years, avoid a Hawthorne effect) or in a clinical that may have relevance involves adult
they faired more poorly on a range of trial. Thus, we do not have the quality of monovision contact lens wearers,69 which
visuomotor and visuocognitive tests and evidence that we would like regarding showed that a refractive difference devel-
had poorer visual attention than the this question and this is an area that oped between the eyes. If adults are influ-
emmetropic children (although the requires more research. Therefore, the enced by correction, we may anticipate a
authors note that there was no significant guideline is not based on these studies greater effect in young children. There-
difference between the corrected versus alone but also on studies of risk factors fore, with the current information, it
the uncorrected hyperopic children). In a for amblyopia and epidemiological behoves the clinician to be conservative,
small study, Shankar, Evans and Bobier64 studies.14,18,25 that is, we cannot assume that prescribing

Clinical and Experimental Optometry 2011 2011 The Author

8 Clinical and Experimental Optometry 2011 Optometrists Association Australia
 Management of refractive error in infants and children Leat

glasses does not influence refractive intervention resulted in less myopic pro- 2.00 D or more hyperopia without glasses
development. gression. The largest and most ethnically did less close work and reading than con-
diverse study was the Correction of Myopia trols with lower refractions, while the
Evaluation Trial (COMET).81,82 This hyperopic children with glasses reported
CHILDREN IN THE SCHOOL YEARS
found that the group fitted with +2.00 D the same amount.84 Rosner and Rosner85
During the school years, there are slightly addition progressive addition lenses had reported that first to fifth graders with
different considerations. Emmetropisa- less myopic progression compared with 1.50 D or more of hyperopia had poorer
tion is essentially complete by six years13 those with single vision lenses. The differ- school achievement than other children
and the most sensitive part of the critical ence was statistically significant (0.20 D and Williams and colleagues86 found
period is over (although various aspects of over a three-year period) but was not con- similar results, namely, uncorrected
vision may not be adult-like until eight sidered to be clinically significant.81 A sub- hyperopic children with a total of 3.00 D
years or even until the teenage years and analysis, however, showed that myopic hyperopia in the two eyes summed had
there are different critical periods for dif- children with a larger lag of accommoda- poorer performance on standardised
ferent functions9,7073). During these years, tion (greater than 0.43 for a 33 cm target, school tests. Two older reviews of the lit-
the refraction of children with higher which can be measured with dynamic reti- erature concluded that hyperopia (specifi-
hyperopia and with emmetropia remains noscopy) in combination with a near eso- cally hyperopia 1.00 D or more) is
unchanged, while the refraction of chil- phoria gained a clinically significant associated with poor reading skills (non-
dren with moderate hyperopia still shows a benefit from progressive addition lenses specific reading difficulty).87,88 It is pos-
drift towards emmetropia up to nine or (0.64 D less myopic progression over three sible that it is specifically those children,
10 years of age74 and early onset myopia years).82 Similarly, those with the larger lag who fail to accommodate for their moder-
commences. Thus, with age, there is a of accommodation plus a closer working ate hyperopia, who are most likely to
slow movement of the population mean distance or a lower baseline myopia expe- benefit from a hyperopic prescription for
towards emmetropia and then myo- rienced clinically a significant reduction reading,89 but this is an area that needs
pia38,39,75 and a slow increase of the range in myopic progression (0.44 D and 0.48 D, more study.90 Anisometropia is also
of refractive error of the population, as respectively). related to poor reading, although there is
shown by an increase in the standard In school age children compared with no evidence of such a relationship for
deviation.13,38,75 From six years onwards, younger children, there are fewer guide- astigmatism.87 However, when children
when early onset myopia starts,13,74 there is lines on what level of hyperopia should be with an explicit diagnosis of specific
also an increase in the prevalence of corrected in the absence of symptoms and reading disability (dyslexia)91,92 are con-
higher amounts of astigmatism, and in there are very limited current data on sidered, there is little evidence of any
individual children, increases in astigma- which to make this judgement. The follow- relationship to refractive error.93 As
tism occur simultaneously with increases ing studies give some indications of when mentioned above, an association between
in myopia.28 Thus, during these years, to prescribe. Mutti83 presented data from a performance on tests such as reading and
correction is more for function, with a longitudinal study of school children. refractive error does not prove causality.
consideration of symptoms and school Visual acuity was poorer in the children When we consider these studies together
performance. with uncorrected hyperopia (spherical (those on VA, accommodative lag and
In the school years, myopia should be equivalent) of 2.00 D or more compared poorer reading), there are indications
corrected for function with full correc- with those who had a correction. For those that higher levels of uncorrected hypero-
tion. There is no evidence that a partial who wore glasses and had hyperopia of pia may have functional impacts on vision
correction reduces the progression of 1.00 D or more, corrected VA was a line and near work. Taking both the modal
myopia.76 In fact, undercorrection may better than uncorrected VA. In other and median values of hyperopia from
lead to further progression of myopia.77 words, uncorrected hyperopia of 1.00 D or among these studies seems to indicate that
There are numerous randomised clinical more can impact VA. This was for distance 1.50 D or more of hyperopia should be
trials that have examined the impact of VA measured at one point in time. There- considered for correction even in the
progressive lens additions on the progres- fore, it is reasonable to assume that near absence of symptoms. It is clear that more
sion of myopia.7881 Most have shown a acuity and acuity for sustained tasks would studies are required to confidently answer
small but statistically significant differ- be more impacted. In this study, they also the question of what level of hyperopia
ence, although Edwards and colleagues80 measured the lag of accommodation. should be corrected at this age.
found no effect in a group of Hong Kong Uncorrected hyperopia of 1.50 D or more There is little solid evidence for or
children. Leung and Brown79 found an was also associated with 2.00 D or more of against the benefit of correcting lower
effect of the power of the addition, accommodative lag (at a 4.00 D demand), levels of hyperopia. Correcting small
+2.00 D resulting in more myopic control which is a significant defocus for near refractive errors generally (myopia,
than +1.50 D, and in a cross-over study, work. A recent study in Australia of hyperopia, astigmatism and anisometro-
Hasebe and colleagues78 found that earlier 12-year-old children found that those with pia) in school children (for example,

2011 The Author Clinical and Experimental Optometry 2011

Clinical and Experimental Optometry 2011 Optometrists Association Australia 9
Management of refractive error in infants and children Leat

0.50 D to 1.00 D for astigmatism or up to With all these considerations in mind, such as poor co-ordination, slower devel-
1.50 D for hyperopia) is controversial and the guidelines shown in Table 2 have been opment of fine motor skills, reduced
there are no solid studies to give guid- developed. They are based on the very few attention for near tasks, excessive activity
ance. Robaei and colleagues94 considered randomised clinical trials that have been and asthenopia, headaches or learning
the spectacle usage of 12-year-old children undertaken. This is the highest level of difficulties in older children are also indi-
with hyperopia of less than 2.00 D or astig- evidence. When these are not available, cators of the potential benefit from a
matism less than 1.00 D (termed non- the guidelines are based on epidemio- prescription.11,12,99,100 Many authors1012
refractive spectacle wearers in this study) logical studies that give the expected recommend monitoring the refraction
and found that 62.2 per cent used their age-related range of refraction and longi- (hyperopia, myopia or astigmatism) in
spectacles at least sometimes. In an earlier tudinal and cross-sectional studies, includ- infants and toddlers before prescribing.
study of six-year-old children, they found ing clinical studies, which link refractive Frequently unchanging or increasing
that 42.3 per cent of those with these error with outcomes. When none or very refractions are associated with amblyo-
lower refractive errors were symptomatic few of these are available, the guidelines pia.25,32 This is unless factors such as
before but not after wearing spectacles.95 are based on current clinical opinion and demonstrable amblyopia indicate pre-
In one of the few studies to apply different other guidelines (shown as italics in scribing immediately. The other main
cut-off criteria to examine the improve- Table 2). These show when spectacle pre- factor, which will influence ones likeli-
ment with a spectacle prescription, scription would be considered. In the fol- hood of prescribing for hyperopia, is the
Congdon and colleagues96 found that a lowing section, which gives notes on the presence of heterophoria. Correction of
cut-off of -0.75 D or less of myopia, 1.00 D guidelines, other factors that would influ- hyperopia to optimise alignment (with a
or more of hyperopia and 0.75 D or more ence a prescribing decision are discussed. bifocal in cases of convergence excess eso-
of astigmatism was effective in discriminat- There are some instances when phoria) is a consideration.12
ing six- to 19-year-old children, who spectacle correction is essential. This Guideline 1 (Table 2) suggests prescrib-
gained improvement in VA, although would include children with anisome- ing if the refraction is outside the 95%
none of their criteria distinguished tropic amblyopia, very high refractions of limits for a particular age. Guideline 3
between the children who did or did not any kind with reduced VA and children (Table 2) is based on the studies of Atkin-
use their spectacles. who are aphakic or pseudoaphakic. Chil- son and colleagues,20,46 which indicate
On this question of prescribing for low dren with aphakia or pseudophakia functional improvements when children
refractive errors, clinical opinion varies. require glasses or contact lenses to with hyperopia in the least hyperopic
Some clinicians suggest that children with correct any residual hyperopia plus a cor- meridian of 3.50 D or more were given a
smaller refractive errors (down to 0.75 D) rection for near because they have no partial prescription. For the current data
associated with symptoms (asthenopia, dif- accommodation. for white children, these guidelines are
ficulty with focusing, headaches) may fairly similar. This is not the case for
benefit from spectacle prescription.8,11,97,98 African American or Hispanic children
NOTES ON MANAGEMENT
Other factors that would indicate a pre- according to the MEPED study, which
scription for lower levels of hyperopia are In prescribing for higher hyperopes, apart shows the higher 95% limit of the spheri-
reduced uncorrected vision, the presence from the level of hyperopia, factors that cal equivalent normal range to be greater
of esophoria or esotropia (perhaps indi- may give further indication of the need for than 3.50 D. At present, we do not know
cating a bifocal), higher than normal lags intervening with a correction are reduced whether we should follow the guideline
of accommodation, difficulty with close uncorrected vision, reduced corrected VA based on the functional improvements in
work (for example, squinting, blinking or or stereopsis and whether there is reduced English children, which would mean pre-
poor attention span) or reports of sus- or insufficient accommodation. Accom- scribing glasses for more than five per cent
pected or diagnosed reading difficul- modation could be measured with of children in the African American or
ties.11,99 These smaller prescriptions would dynamic retinoscopy or by amplitude Hispanic groups, or whether we should
usually be given for part-time wear. For testing depending on the childs age. The prescribe only for those who fall outside
myopia, most clinical opinions indicate clinician should consider if there is exces- the 95% range for their ethnicity. The
correcting the refractive error once the sive lag of accommodation without latter approach would indicate that in
child reaches -1.00 D,11,100,101 although a correction (in the case of dynamic reti- some way, these ethnic groups are more
some say a prescription can be considered noscopy) or if there is sufficient amplitude immune to the functional impact of
at less than -0.50 D.8 Milder and Rubin102 of accommodation to overcome the higher hyperopia or better able to com-
state that a prescription would usually be hyperopia and accommodate for a near pensate with accommodation.
required at less than 2.00 D.102 Certainly, a task, allowing 50 per cent of the amplitude When prescribing for infants with
prescription can be offered once the child in reserve.53 Clinical observation and hyperopia, there are several approaches
starts to notice difficulty with blackboard opinion, including the authors own expe- that could be adopted to determine how
work.100 rience, indicate that signs and symptoms much hyperopia to correct. We could pre-

Clinical and Experimental Optometry 2011 2011 The Author

10 Clinical and Experimental Optometry 2011 Optometrists Association Australia
 Management of refractive error in infants and children Leat

scribe to bring the uncorrected portion the prescribing appointment. This allows life can be monitored. Emmetropisation is
just within the normal range, for example, time for the spectacles to be ordered and active, the visual world that is important to
to the 95% limit. This would leave a large dispensed and for the child to adapt to babies is close and the visual demands of
stimulus for emmetropisation and there- them. At this follow-up visit, the optom- babies do not include a need for clear
fore potentially encourage a greater etrist should question the parents regard- distance vision. Therefore, it is only the
amount of emmetropisation. Clinical ex- ing any signs of strabismus and should very high refractive errors that should be
perience suggests that children who are carefully check for strabismus and corrected. The clinician should be aware
prescribed in this way may be more at risk changes in phoria, as well as measuring that high myopia at this age is associated
of developing esotropia, although evi- the VA and over-refraction. with prematurity, in particular with retin-
dence from research has not confirmed In the pre-school years, the general rule opathy of prematurity14,104 and ocular or
this. It seems that the childs accommoda- for prescription of glasses is that while neurological conditions unless there is a
tion cannot overcome the very large emmetropisation is active, the refractive family history of degenerative myopia,11 so
uncorrected hyperopia but a correction error is undercorrected, unless other that a referral for an ophthalmological or
that is small enough to bring them just factors such as the need to treat amblyopia neurological examination may be war-
within the normal range allows them to or strabismus or to optimise ocular ranted. Very high levels of myopia are also
accommodate for the remaining hypero- alignment outweigh the need to leave a associated with amblyopia.105 From the age
pia, resulting in esotropia.102 Another stimulus for emmetropisation. Emmetro- of one year, children are starting to
approach is to prescribe to leave the pisation may be active for astigmatism up explore their environment and take an
uncorrected portion equal to the average to four to five years and possibly up to six interest in distance activities and therefore
for the age. This would give the child an years for spherical ametropia,13 and even are likely to benefit from a correction, but
average stimulus for emmetropisation, until nine to 10 years for some moderate they do not have a requirement for fully
which may not be the optimal stimulus to hyperopes.74 Also, while emmetropisation focused distance vision.10 By reducing the
emmetropisation considering their higher is still active, the optometrist should prescription, some stimulus to emmetropi-
than normal level of hyperopia. Thus, the monitor the child frequently and maintain sation is maintained.
approach suggested here is to prescribe to an undercorrection according to these When prescribing for school children,
leave the
Ambliopa y uncorrected portion just above guidelines. It is tempting not to decrease the author finds that the full non-
the mean for the age, leaving a stimulus the prescription, when the child is func- cycloplegic subjective refraction for occa-
estrabismo tienen
for emmetropisation, which is still larger tioning well and visual function is good. sional or full-time wear can be considered.
manejo
thanprioritario
the average. For example, at one year However to prevent any interruption to This means that for children with previ-
the mean according to Mayer and col- emmetropisation it would seem prudent ously uncorrected high hyperopia, the
leagues14 is approximately 1.75 D spheri- to do this. Therefore, the optometrist prescription would be reduced from the
cal equivalent (cycloplegic refraction), so should remember to advise the parent retinoscopic result and that generally most
the clinician might consider prescribing from the outset that the prescription may prescriptions would be reduced compared
to leave approximately 2.00 to 2.25 D have to be changed frequently. If the with any cycloplegic findings to allow for
undercorrected. This is still prescribing to parents understand that the clinician tonus.
leave the uncorrected portion within the hopes to decrease the prescription, they Consider this
normal limits, as suggested by Marsh- are usually happier (parents are always
CONCLUSION
Tootle.11 Alternatively, the clinician could more concerned when a prescription has
apply the Atkinson and colleagues proto- to be increased). This paper has reviewed the evidence that
col,46 which in practice gives a similar In cases of anisometropia with amblyo- is currently available and has attempted to
result. If this approach of prescribing and pia, refractive correction is the usual first bring this together to guide the clinician
leaving a greater than average stimulus management option. Full refractive cor- who works with children. There are rea-
to emmetropisation is used, the child rection alone often results in some sonable data available regarding the
must be monitored very frequently (for improvement of VA, most of which occurs natural history of refractive error develop-
example, every month initially) and the in the first four months, although some ment for the population as a whole and we
parent warned that at the first sign of a improvement may continue to occur up to have some knowledge of the risk factors
strabismus, they should return. If that one year.103 After this four-month period for abnormal visual development; how-
happens, the prescription should be of refractive correction, occlusion therapy ever, we currently lack the ability to
increased to optimise ocular alignment12,98 may not be necessary in some cases and in accurately predict which children will
or to the full hyperopic prescription.102 those that do require occlusion, the emmetropise. There are also very few
In prescribing for any of these young improved VA after a period of spectacle studies on the impact of spectacle pre-
patients, especially when a larger prescrip- wear may make compliance better. scription on the childs visual system and
tion is given, it is imperative to see the With respect to correcting myopia in functional vision. Prescribing spectacles
child approximately four to six weeks after infancy, most myopia in the first year of when a risk factor is present would seem to

2011 The Author Esperar al menos 4 meses Clinical and Experimental Optometry 2011
antes
Clinical and Experimental Optometry 2011 Optometrists deAustralia
Association instaurar tratamiento 11
ortptico
Management of refractive error in infants and children Leat

be logical to avoid the development of 8. Leat SJ, Shute RH, Westall CA. Assessing based sample of astigmatic children. I. Refraction
Childrens Vision: A Handbook. Oxford: and amblyopia. Acta Ophthalmol (Copenh) 1990;
amblyopia, for example, but without more
Butterworth-Heinemann, 1999. 68: 428434.
longitudinal studies and clinical trials we 9. Leat SJ. Pediatric assessment. In: Rosenfield M, 26. Atkinson J, Braddick O, French J. Infant astigma-
cannot be sure whether this is the case. We Logan N, Edwards K, eds. Optometry: Science, tism: Its disappearance with age. Vision Res 1980;
Techniques and Clinical Management. Edin- 20: 891893.
can now identify reasonably well the child
burgh: Butterworth-Heinemann/Elsevier, 2009. 27. Gwiazda J, Scheiman M, Mohindra I, Held R.
who is outside the limits of the normal p 439460. Astigmatism in children: Changes in axis and
distribution, based on the natural history 10. Bobier WR. Evidence-based spectacle prescribing amount from birth to six years. Invest Ophthalmol
for infants and children. J Modern Optics 2007; 54: Vis Sci 1984; 25: 8892.
and clinical data that have been reviewed
13671377. 28. Gwiazda J, Grice K, Held R, McLellan J, Thorn F.
here. It would be useful to have 11. Marsh-Tootle W. Infants, toddlers and children. Astigmatism and the development of myopia in
population-based data published in the In: Benjamin WJ, ed. Borishs Clinical Refraction. children. Vision Res 2000; 40: 10191026.
Philadelphia: WB Saunders, 1998. p 10601118. 29. Saunders KJ, Margaret Woodhouse J, Westall CA.
format of the clinical data of Mayer and
12. Ciner EB. Management of refractive error in Emmetropisation in human infancy: Rate of
colleagues14 for the various components of infants, toddlers and preschool children. Probl change is related to initial refractive error. Vision
refractive error, so that a more exact idea Optom 1990; 2: 394419. Res 1995; 35: 13251328.
13. Gwiazda J, Thorn F, Bauer J, Held R. 30. Zonis S, Miller B. Refraction in the Israeli
could be determined of where a child of a
Emmetropization and the progression of mani- newborn. J Ped Ophthalmol 1974; 11: 7781.
particular age lies with respect to the fest refraction in children followed from infancy 31. Ingram RM, Traynar MJ, Walker C. Screening for
population mean and ranges. We also to puberty. Clin Vis Sci 1993; 8: 337344. refractive errors at age 1 year. Br J Ophthalmol
14. Mayer DL, Hansen RM, Moore BD, Kim S, Fulton 1979; 63: 243250.
need to know at exactly what age and level
AB. Cycloplegic refractions in healthy children 32. Abrahamsson M, Fabian G, Sjstrand J. A longi-
of ametropia we should intervene. This aged 1 through 48 months. Arch Ophthalmol 2001; tudinal study of a population based sample of
seems particularly problematic with astig- 119: 1625. astigmatic children. II. The changeability of ani-
15. Mutti DO, Mitchell GL, Jones LA, Friedman NE, sometropia. Acta Ophthalmol (Copenh) 1990; 68:
matism, in which effects on visual func-
Frane SL, Lin WK, Moeschberger ML et al. 435440.
tion, including meridional amblyopia, Accommodation, acuity and their relationship to 33. Borchert M, Tarczy-Hornoch K, Cotter SA, Liu N,
have been identified at levels of astigma- emmetropization in infants. Optom Vis Sci 2009; Azen SP, Varma R. Anisometropia in Hispanic
86: 666676. and African American infants and young chil-
tism that are quite prevalent in the popu-
16. Mutti DO, Mitchell GL, Jones LA, Friedman NE, dren: The Multi-Ethnic Pediatric Eye Disease
lation and therefore might be considered Frane SL, Lin WK, Moeschberger ML et al. Axial study. Ophthalmology 2010; 117: 148153.e1.
normal for age. It is clear that the field of growth and changes in lenticular and corneal 34. De Vries J. Anisometropia in children: Analysis of
power during emmetropization in infants. Invest a hospital population. Br J Ophthalmol 1985; 69:
refractive development and correction is
Ophthalmol Vis Sci 2005; 46: 30743080. 504507.
in need of further research. 17. Edwards M. The refractive status of Hong Kong 35. Phelps WL, Muir J. Anisometropia and strabis-
Chinese infants. Ophthalmic Physiol Opt 1991; 114: mus. Am Orthopt J 1977; 27: 131133.
297303. 36. Almeder LM, Peck LB, Howland HC. Prevalence
18. Multi-Ethnic Pediatric Eye Disease Study Group. of anisometropia in volunteer laboratory and
REFERENCES Prevalence of myopia and hyperopia in 6- to school screening populations. Invest Ophthalmol
1. American Academy of Ophthalmology Pediatric 72-month-old African American and Hispanic Vis Sci 1990; 31: 24482455.
Ophthalmology/Strabismus Panel. Preferred children: The Multi-Ethnic Pediatric Eye Disease 37. Abrahamsson M, Sjstrand J. Natural history of
Practice Pattern Guidelines. Amblyopia. San study. Ophthalmology 2010; 117: 140147.e3. infantile anisometropia. Br J Ophthalmol 1996; 80:
Francisco, CA: American Academy of Ophthal- 19. Ingram RM, Barr A. Changes in refraction 860863.
mology; 2007. Available from: http://www. between the ages of 1 and 3 1/2 years. Br J Oph- 38. Twelker JD, Mitchell GL, Messer DH, Bhakta R,
aao.org/ppp. Accessed on August 27, 2010. thalmol 1979; 63: 339342. Jones LA, Mutti DO, Cotter SA et al. Childrens
2. Miller JM, Harvey EM. Spectacle prescribing rec- 20. Atkinson J, Braddick O, Robier B, Anker S, ocular components and age, gender and ethnic-
ommendations of AAPOS members. J Pediatr Oph- Ehrlich D, King J, Watson P et al. Two infant ity. Optom Vis Sci 2009; 86: 918935.
thalmol Strab 1998; 35: 5152. vision screening programmes: Prediction and 39. Zadnik K, Mutti DO, Friedman NE, Adams AJ.
3. American Optometric Association. American prevention of strabismus and amblyopia from Initial cross-sectional results from the Orinda
Optometric Association: Care of the patient with photo- and videorefractive screening. Eye 1996; longitudinal study of myopia. Optom Vis Sci 1993;
hyperopia. Available from: http://www.aoa.org/ 10: 189198. 70: 750758.
documents/CPG-16.pdf. Accessed on July 14, 21. Ehrlich DL, Atkinson J, Braddick O, Bobier W, 40. Smith EL 3rd, Hung LF. The role of optical
2010. Durden K. Reduction of infant myopia: A longi- defocus in regulating refractive development
4. American Optometric Association. Optometric tudinal cycloplegic study. Vision Res 1995; 35: in infant monkeys. Vision Res 1999; 39: 1415
Clinical Practice Guideline; Care of the patient 13131324. 1435.
with myopia. Available from: http://www.aoa. 22. Ehrlich DL, Braddick OJ, Atkinson J, Anker S, 41. Ingram RM, Arnold PE, Dally S, Lucas J. Emmet-
org/documents/CPG-16.pdf. Accessed on Weeks F, Hartley T, Wade J et al. Infant ropisation, squint, and reduced visual acuity after
October 9, 2010. emmetropization: Longitudinal changes in treatment. Br J Ophthalmol 1991; 75: 414416.
5. Blum HL, Peters HB, Bettman JW. Vision refraction components from nine to twenty 42. Pennie FC, Wood ICJ, Olsen C, White S,
Screening for Elementary Schools: The Orinda months of age. Optom Vis Sci 1997; 74: 822843. Charman WN. A longitudinal study of the bio-
Study. Berkeley, CA: University of California 23. Varghese RM, Sreenivas V, Puliyel JM, Varughese metric and refractive changes in full-term infants
Press, 1968. S. Refractive status at birth: Its relation to during the first year of life. Vision Res 2001; 41:
6. Royal College of Ophthalmologists. Guidelines newborn physical parameters at birth and gesta- 27992810.
for the management of amblyopia. Avail- tional age. PLoS One 2009; 4: e4469. 43. Birch E, Stager D, Everett ME. The natural
able from: http://www.rcophth.ac.uk/docs/ 24. Harvey EM, Dobson V, Clifford-Donaldson CE, history of infantile anisometropia. Invest Ophthal-
publications/GuidelinesfortheManagementof Green TK, Messer DH, Miller JM. Prevalence of mol Vis Sci 1995; 36: S845.
Amblyopia.pdf. Accessed on February 21, 2010. astigmatism in Native American infants and chil- 44. Du JW, Schmid KL, Bevan JD, Frater KM, Ollett
7. Farbrother JE. Spectacle prescribing in child- dren. Optom Vis Sci 2010; 87: 400405. R, Hein B. Retrospective analysis of refractive
hood: A survey of hospital optometrists. Br J Oph- 25. Abrahamsson M, Fabian G, Andersson AK, errors in children with vision impairment. Optom
thalmol 2008; 92: 392395. Sjstrand J. A longitudinal study of a population Vis Sci 2005; 82: 807816.

Clinical and Experimental Optometry 2011 2011 The Author

12 Clinical and Experimental Optometry 2011 Optometrists Association Australia
 Management of refractive error in infants and children Leat

45. Ingram RM, Arnold PE, Dally S, Lucas J. Results development of young hyperopes and age of first myopia progression and their interaction with
of a randomised trial of treating abnormal hyper- lens correction. Clin Exp Optom 1986; 69: 166 treatment in COMET children. Invest Ophthalmol
metropia from the age of 6 months. Br J Ophthal- 168. Vis Sci 2004; 45: 21432151.
mol 1990; 74: 158159. 66. Roch-Levecq AC, Brody BL, Thomas RG, Brown 83. Mutti DO. To emmetropize or not to
46. Atkinson J, Braddick O, Nardini M, Anker S. SI. Ametropia, preschoolers cognitive abilities, emmetropize? The question for hyperopic devel-
Infant hyperopia: Detection, distribution, and effects of spectacle correction. Arch Ophthal- opment. Optom Vis Sci 2007; 84: 97102.
changes and correlatesOutcomes from the mol 2008; 126: 252258. 84. French AN, Rose KA, Burlutsky G, Mitchell P.
Cambridge infant screening programs. Optom 67. Wallman J, Winawer J. Homeostasis of eye growth Does correction of hyperopia affect the pattern
Vision Sci 2007; 84: 8496. and the question of myopia. Neuron 2004; 43: of childrens activities, and does this differ from
47. Aurell E, Norrsell K. A longitudinal study of chil- 447468. that of emmetropic children? Invest Ophthalmol
dren with a family history of strabismus: Factors 68. Wildsoet CF. Active emmetropizationevidence Vis Sci 2009; 50: E-Abstract 3961.
determining the incidence of strabismus. Br J for its existence and ramifications for clinical 85. Rosner J, Rosner J. The relationship between
Ophthalmol 1990; 74: 589594. practice. Ophthalmic Physiol Opt 1997; 17: 279 moderate hyperopia and academic achievement:
48. Friedman Z, Neumann E, Abel-Peleg B. Outcome 290. How much plus is enough? J Am Optom Assoc
of treatment of marked ametropia without stra- 69. Wick B, Westin E. Change in refractive ani- 1997; 68: 648650.
bismus following screening and diagnosis before sometropia in presbyopic adults wearing monovi- 86. Williams WR, Latif AHA, Hannington L, Watkins
the age of three. J Pediatr Ophthalmol Strab 1985; sion contact lens correction. Optom Vis Sci 1999; DR. Hyperopia and educational attainment in a
22: 5457. 76: 3339. primary school cohort. Arch Dis Child 2005; 90:
49. Edelman PM, Borchert MS. Visual outcome in 70. Gwiazda J, Bauer J, Thorn F, Held R. Develop- 150153.
high hypermetropia. J AAPOS 1997; 1: 147150. ment of spatial contrast sensitivity from infancy to 87. Simons HD, Gassler PA. Vision anomalies and
50. Fern KD. Visual acuity outcome in isometropic adulthood: Psychophysical data. Optom Vis Sci reading skill: A meta-analysis of the literature. Am
hyperopia. Optom Vis Sci 1989; 66: 649658. 1997; 74: 785789. J Optom Physiol Opt 1988; 65: 893904.
51. von Noorden GK, Avilla CW. Accommodative 71. Gordon GE, McCulloch DL. A VEP investigation 88. Grisham JD, Simons HD. Refractive error and the
convergence in hypermetropia. Am J Ophthalmol of parallel visual pathway development in reading process: A literature analysis. J Am Optom
1990; 110: 287292. primary school age children. Doc Ophthalmol Assoc 1986; 57: 4455.
52. Klimek DL, Cruz OA, Scott WE, Davitt BV. 1999; 99: 110. 89. Arnold RW. Pseudo-false positive eye/vision pho-
Isoametropic amblyopia due to high hyperopia 72. Madrid M, Crognale MA. Long-term maturation toscreening due to accommodative insufficiency.
in children. J AAPOS 2004; 8: 310313. of visual pathways. Vis Neurosci 2000; 17: 831837. A serendipitous benefit for poor readers? Binocul
53. Schoenleber DB, Crouch ER Jr. Bilateral hyper- 73. Lewis TL, Maurer D. Multiple sensitive periods in Vis Strabismus Q 2004; 19: 7580.
metropic amblyopia. J Pediatr Ophthalmol Strabis- human visual development: Evidence from visu- 90. Simons K. Hyperopia, accommodative dysfunc-
mus 1987; 24: 7577. ally deprived children. Dev Psychobiol 2005; 46: tion and reading. Binocul Vis Strabismus Q 2004;
54. Gwiazda J, Mohindra I, Brill S, Held R. Infant 163183. 19: 6970.
astigmatism and meridional amblyopia. Vision Res 74. Jones LA, Mitchell GL, Mutti DO, Hayes JR, Moe- 91. Nandakumar K, Leat SJ. Dyslexia: A review of two
1985; 25: 12691276. schberger ML, Zadnik K. Comparison of ocular theories. Clin Exp Optom 2008; 91: 333340.
55. Gwiazda J, Bauer J, Thorn F, Held R. Meridional component growth curves among refractive error 92. American Psychiatric Association. Task Force on
amblyopia does result from astigmatism in early groups in children. Invest Ophthalmol Vis Sci 2005; DSM-IV. DSM-IV: Diagnostic and Statistical Manual
childhood. Clin Vis Sci 1986; 1: 145152. 46: 23172327. of Mental Disorders. Washington, DC: American
56. Dobson V, Harvey EM, Clifford-Donaldson CE, 75. Pointer JS. A 6-year longitudinal optometric Psychiatric Association, 1994.
Green TK, Miller JM. Amblyopia in astigmatic study of the refractive trend in school-aged 93. Evans BJW, Drasdo N, Richards IL. Dyslexia: The
infants and toddlers. Optom Vis Sci 2010; 87: 330 children. Ophthalmic Physiol Opt 2001; 21: 361 link with visual deficits. Ophthalmic Physiol Opt
336. 367. 1996; 16: 310.
57. Dobson V, Miller JM, Harvey EM, Mohan KM. 76. Adler D, Millodot M. The possible effect of 94. Robaei D, Kifley A, Rose KA, Mitchell P. Refrac-
Amblyopia in astigmatic preschool children. undercorrection on myopic progression in chil- tive error and patterns of spectacle use in 12-year-
Vision Res 2003; 3: 10811090. dren. Clin Exp Optom 2006; 89: 315321. old Australian children. Ophthalmology 2006; 113:
58. Harvey EM, Dobson V, Miller JM, Clifford- 77. Chung K, Mohidin N, OLeary DJ. Undercorrec- 15671573.
Donaldson CE. Changes in visual function follow- tion of myopia enhances rather than inhibits 95. Robaei D, Rose K, Kifley A, Mitchell P. Patterns of
ing optical treatment of astigmatism-related myopia progression. Vision Res 2002; 42: 2555 spectacle use in young Australian school chil-
amblyopia. Vision Res 2008; 48: 773787. 2559. dren: Findings from a population-based study. J
59. Donahue SP. Relationship between anisometro- 78. Hasebe S, Ohtsuki H, Nonaka T, Nakatsuka C, AAPOS 2005; 9: 579583.
pia, patient age and the development of amblyo- Miyata M, Hamasaki I, Kimura S. Effect of pro- 96. Congdon NG, Patel N, Esteso P, Chikwembani F,
pia. Am J Ophthalmol 2006; 142: 132140. gressive addition lenses on myopia progression in Webber F, Msithini RB, Ratcliffe A. The associa-
60. Weakley DR. The association between nonstrabis- Japanese children: A prospective, randomized, tion between refractive cutoffs for spectacle pro-
mic anisometropia, amblyopia and subnormal double-masked, crossover trial. Invest Ophthalmol vision and visual improvement among school-
binocularity. Ophthalmology 2001; 108: 163171. Vis Sci 2008; 49: 27812789. aged children in South Africa. Br J Ophthalmol
61. Dobson V, Miller JM, Clifford-Donaldson CE, 79. Leung J, Brown B. Progression of myopia in 2008; 92: 1318.
Harvey EM. Associations between anisometropia, Hong Kong Chinese schoolchildren is slowed by 97. Brookman KE. Low ametropias. In: Brookman
amblyopia, and reduced stereoacuity in a school- wearing progressive lenses. Optom Vis Sci 1999; 76: KE, ed. Refractive Management of Ametropia.
aged population with a high prevalence of astig- 346354. Boston: Butterworth-Heinemann; 1996. p 123
matism. Invest Ophthalmol Vis Sci 2008; 49: 4427 80. Edwards MH, Li RW, Lam CS, Lew JK, Yu BS. The 143.
4436. Hong Kong progressive lens myopia control 98. Carlson NB. Hyperopia. In: Brookman KE, ed.
62. Colenbrander A. Visual functions and functional study: Study design and main findings. Invest Oph- Refractive Management of Ametropia. Boston:
vision. Int Cong Ser 2005; 1282: 482486. thalmol Vis Sci 2002; 43: 28522858. Butterworth-Heinemann, 1996. p 4572.
63. Ingram RM, Arnold PE, Dally S, Lucas J. Results 81. Gwiazda J, Hyman L, Hussein M, Everett D, 99. Cotter SA. Management of childhood hyperopia:
of a randomised trial of treating abnormal hyper- Norton TT, Kurtz D, Leske MC et al. A random- A pediatric optometrists perspective. Optom Vis
metropia from the age of 6 months. Br J Ophthal- ized clinical trial of progressive addition lenses Sci 2007; 84: 103109.
mol 1990; 74: 158159. versus single vision lenses on the progression of 100. Werner DL, Press LJ. Clinical Pearls in Refractive
64. Shankar S, Evans MA, Bobier WR. Hyperopia and myopia in children. Invest Ophthalmol Vis Sci 2003; Care. Boston: Butterworth-Heinemann Medical,
emergent literacy of young children: Pilot study. 44: 14921500. 2002.
Optom Vis Sci 2007; 84: 10311038. 82. Gwiazda JE, Hyman L, Norton TT, Hussein M, 101. Goss DA. Myopia. In: Brookman KE, ed.
65. Rosner J, Rosner J. Some observations of the rela- Marsh-Tootle W, Manny R, Wang Y et al. Accom- Refractive Management of Ametropia. Boston:
tionship between the visual perceptual skills modation and related risk factors associated with Butterworth-Heinemann, 1996. p 1344.

2011 The Author Clinical and Experimental Optometry 2011

Clinical and Experimental Optometry 2011 Optometrists Association Australia 13
Management of refractive error in infants and children Leat

102. Milder B, Rubin ML. The Fine Art of Prescribing

Glasses without Making a Spectacle of Yourself,
3rd ed. Gainsville, Florida: Triad Publishing
Company, 2004.
103. Cotter SA. Treatment of anisometropic amblyo-
pia in children with refractive correction. Oph-
thalmology 2006; 113: 895903.
104. Quinn GE, Dobson V, Kivlin J, Kaufman LM,
Repka MX, Reynolds JD, Gordon RA et al. Preva-
lence of myopia between 3 months and 5
1/2 years in preterm infants with and without
retinopathy of prematurity. Ophthalmology 1998;
105: 12921300.
105. Ciner E. Refractive error in young children. In:
Moore BD, ed. Eye Care for Infants and Young
Children. Boston: Butterworth-Heinemann,
1997. p 4774.
106. Ingram RM, Walker C, Wilson JM, Arnold PE,
Dally S. Prediction of amblyopia and squint by
means of refraction at age 1 year. Br J Ophthalmol
1986; 70: 1215.
107. Cowen L, Bobier WR. The pattern of astigmatism
in a Canadian preschool population. Invest Oph-
thalmol Vis Sci 2003; 44: 45934600.
108. Viner C. Refractive examination. In: Harvey W,
Gilmartin B, eds. Paediatric Optometry. Oxford:
Elsevier Butterworth-Heinemann, 2004. p 2126.
109. Zetterstrom C, Kugelberg M. Paediatric cataract
surgery. Acta Ophthalmol Scand 2007; 85: 698710.

Corresponding author:
Dr Susan J Leat
School of Optometry
University of Waterloo
Waterloo
Ontario N2E 3G1
CANADA
E-mail: [email protected]

Clinical and Experimental Optometry 2011 2011 The Author

14 Clinical and Experimental Optometry 2011 Optometrists Association Australia