1040-5488/07/8402-0149/0 VOL. 84, NO. 2, PP.

149–155
OPTOMETRY AND VISION SCIENCE
Copyright © 2007 American Academy of Optometry

ORIGINAL ARTICLE

Association Between Accommodative Accuracy,

Hypermetropia, and Strabismus in Children
with Down’s Syndrome
RUTH E. STEWART, PhD, BSc, MOptom, J. MARGARET WOODHOUSE, PhD, BSc, FSMC,
MARY CREGG, PhD, DipOptom, and VALERIE H. PAKEMAN, BSc, MOptom
School of Optometry & Vision Sciences, Cardiff University, Cardiff, United Kingdom

ABSTRACT
Purpose. A significant proportion of children with Down’s syndrome have been shown to have reduced accommodation.
The purpose of this study was to investigate any association between reduced accommodation and refractive error,
strabismus, visual acuity, and other ocular parameters.
Methods. Subjects were children with Down’s syndrome enrolled in a longitudinal cohort to monitor visual development.
Twenty-seven children with accurate accommodation were age-matched to children with reduced accommodation
based on their most recent assessment for which a full, reliable data set was available. Each child was used only once
for matching. Cross-sectional ocular and visual data were analyzed using ␹2 or Fisher’s exact test, or the Mann–Whitney
U test for (non-normally distributed) quantitative data.
Results. Children with under-accommodation were statistically more likely to have moderate/high hypermetropia
(ⱖ⫹3.00 D) and to be strabismic (most with esotropia). No significant difference between the groups was found for any
other ocular parameters.
Conclusions. This study demonstrates the marked association between under-accommodation, hypermetropia, and
strabismus in children with Down’s syndrome. No causal relation can be demonstrated with these data, but findings
suggest that the link between under-accommodation and hypermetropia (and between accurate accommodation and
emmetropia) is present in early infancy.
(Optom Vis Sci 2007;84:149–155)

Key Words: hypermetropia, Down’s syndrome, accommodation, strabismus

C
hildren with Down’s syndrome are at higher risk than
typical children for a number of ocular disorders, includ-
ing refractive errors,1–3 strabismus,4,5 reduced visual acu-
ity,6–8
and poor contrast sensitivity.9 Under-accommodation is
another common finding in children with Down’s syndrome.1,10 –12
Our group measures accommodation by dynamic retinoscopy,12,13
using a technique that has been validated against autorefraction.14
We conventionally use three viewing distances, 25, 16.7, and 10
cm (accommodative demand of 4.00, 6.00, and 10.00 D, respec-
tively). Defining under-accommodation as an accommodative er-
ror index15 outside the ‘normal’ range of 0 to 2.20 D, we have
reported a prevalence of under-accommodation of 68% among
children aged 0.4 to 7 years.11 Haugen et al.1 described under-
accommodation as ‘accommodative weakness’, which they also
measured with dynamic retinoscopy. Their criterion was based
on a retinoscopic reflex observed from a working distance of 50
cm and was defined as ‘weakness’ if the reflex was ‘with’ for a
target placed at 30 cm from the subject. This is equivalent to a
lag of ⬎1.0 D. Haugen et al. reported a prevalence of “accom-
modation weakness” of 55% among young children with
Down’s syndrome. Conversely, then, approximately 32 to 45%
of children with Down’s syndrome accommodate accurately for
near targets.
Cross-sectional studies have suggested that poorer accommoda-
tive accuracy is associated with hypermetropia and strabismus.11
We have longitudinal data for a cohort of children with Down’s syn-
drome acquired over 12 years, and are therefore in a position to iden-
tify children who have consistently shown accurate accommodation
and compare their ocular and visual characteristics with children from
the same cohort who consistently under-accommodate.

Optometry and Vision Science, Vol. 84, No. 2, February 2007

150 Accommodative Accuracy, Hypermetropia, and Strabismus in Children with Down’s Syndrome—Stewart et al.

METHODS young children was up to 0.75 D under-accommodation or ‘lag’.

Subjects were children with Down’s syndrome (trisomy 21)
living in South and West Wales who participate in our longitudi-
nal studies. Our recruitment protocol (which conforms to the Dec-
laration of Helsinki for research involving human subjects and has
been granted ethical committee approval) has been described pre-
viously.11 Recruits are aged at least 3 month when they enter the
study cohort, many are older and all are seen regularly for ocular
assessment. Parents give informed consent for the data to be used
for research purposes. At the time of the present analysis, there
were 114 children in the cohort.
All the authors were involved in examining the cohort. Their
time involved in data collection ranged from 3 to 12 years. All
authors are experienced in dealing with children with intellectual
disabilities and in the techniques used in the study.
Children were examined at home, at school, or in the University
clinic. Techniques were consistent at each place of examination
and total blackout was available for determination of refractive
error by Mohindra retinoscopy at all venues.16 At homes and
schools, this was achieved by the use of a lightproof playhouse.
Mohindra retinoscopy has been shown to give equivalent results to
cycloplegic refraction both in typically developing children and
children with Down’s syndrome.3,17 For analysis, the refractive
error (equivalent sphere) of only one eye was used; this was the
fixing eye for children with strabismus and the right eye for chil-
dren with no strabismus.
At each visit, accommodation was measured by dynamic reti-
noscopy at three testing distances, 25, 16.7, and 10 cm using a
self-illuminated cube with detailed line drawings mounted on a
rule.11 Children who had spectacles wore them for accommoda-
tion measures. The children viewed the target binocularly, and
accommodation was assessed in the right eye, or the fixing eye of
children with strabismus, along the most minus meridian. Rouse
and Hutter18 found that the ‘normal’ accommodative response in
For the current analysis, consistent accurate accommodation was
defined as ⱕ0.75 D lag of accommodation at ⱖ2 out of three
testing distances on at least two separate occasions. Twenty-seven
children from the cohort of 114 (23.7%) fulfilled these criteria. It
should be noted that Rouse and Hutter tested accommodation at
each child’s habitual working distance (mean 25 cm, range 12.5 to
42.5 cm). Only 61.3% of the children tested used a viewing dis-
tance of equal to or ⬍25 cm. It is, therefore, likely that the criterion
for accurate accommodation employed in the current study under-
estimates the number of subjects who would have ‘normal’ accom-
modation using Rouse and Hutter’s criterion.
Previous studies have shown that ocular measures such as refrac-
tive error and visual acuity change with age.1,3 To control for the
effect of age, each of the children with accurate accommodation
(AA) was age-matched with a child who consistently showed inac-
curate accommodation (IA), using the most recent visit for which
the fullest, reliable data set was available. An IA child was selected
as an age match for an AA child based on being the closest child in
age (to within 9 month) with the fullest ocular data set. An IA child
was used only once for matching. Matching was begun with the
oldest child with AA, as there were relatively few older children
with IA in the cohort for whom data were available. A ‘data set’
consisted of the information shown in Table 1.
Cross-sectional ocular and visual data were analyzed using ␹2 or
Fisher’s exact test (SPSS for Windows Version 11), or the Mann–
Whitney U test for (non-normally distributed) quantitative data.
RESULTS
The mean age (at the latest visit) of the group with accurate
accommodation was 6.3 years (range 0.75 to 13.19 years) and of
the group with inaccurate accommodation was 6.2 years (range
1.08 to 12.43 years). There was no significant difference between

TABLE 1.
Optometric data recorded for each child at each examination (in addition to accommodative accuracy)

Ocular measurement Testing method Notes

Presence or absence of
strabismus
Refractive error
Presence or absence of
nystagmus
Stereopsis
Visual acuity
Presence or absence of
ocular defects such
as lens opacities
Cover test at both distance and near or, when
the child was uncooperative for cover test,
by the Hirschberg test with near fixation
Mohindra retinoscopy
As noted on inspection and during
ophthalmoscopy
Frisby stereo test modified to include a
control plate34
Cardiff Acuity Test, a preferential looking
picture test utilizing vanishing
optotypes35,36
As noted on inspection and by
ophthalmoscopy
Strabismus was recorded if the condition was
present at any distance
Refractive error type was classified as
moderate/high hypermetropia (spherical
equivalent ⱖ⫹3.00 D), low hypermetropia/
emmetropia (⬍⫹3.00 D and ⬎⫺0.75 D) or
myopia (ⱕ⫺0.75 D). This classification is
in line with previous studies3
Stereopsis recorded as present if the first test
plate was successful
Data were used only if appropriate spectacle
correction was worn during testing

Optometry and Vision Science, Vol. 84, No. 2, February 2007

 Accommodative Accuracy, Hypermetropia, and Strabismus in Children with Down’s Syndrome—Stewart et al.
TABLE 2.
Number of children from each accommodation group (n ⫽
27) for whom reliable data were available and statistical
test used in analysis
Parameter tested Test used AA
Presence or absence of ␹2 26a
strabismus
Refractive error type ␹2 26b
Median refractive error Mann–Whitney U 26b
Presence or absence of Fisher’s exact 27
nystagmus
Presence or absence of Fisher’s exact 21
stereopsisa,c
Presence or absence of Fisher’s exact 23
lens opacities
Distance visual acuityd Mann–Whitney U 21
a
One child had strabismus due to pathology and was therefore
excluded from analysis.
b
One child was noncompliant during refraction and was there-
fore excluded from analysis.
c
Five children from the IA group and two from the AA group
were tested but failed the demonstration plate of the stereotest.
Stereotest data were unavailable for two children in the AA group.
d
Only acuity values recorded when the child was wearing
appropriate spectacle correction were included in analysis.
the ages of the two groups (independent t-test: t ⫽ 0.061, df ⫽ 52,
p ⫽ 0.952). Because there were fewer older children than younger
children the discrepancy in age matches was greatest at older ages;
the closest IA age-match to the oldest child in the AA group was
approximately 9 month. The mean absolute difference between
age-matched children was 1.75 month.
For varying reasons, all 27 children from each group (AA and
IA) could not be included in every analysis. The total number of
children for each analysis and the methods of analysis are shown in
Table 2. One child (from the AA group) was noncompliant for
refraction and was excluded from refraction analysis. For statistical
analysis of refractive error data, because there were so few myopes
in either group, children were classified as moderate/high hyper-
metropes (ⱖ⫹3.00 D) or low/nonhypermetropes. The definition
of “significant refractive error” is somewhat arbitrary, especially as
refractive error varies with age. A definition of moderate/high hy-
permetropia ⱖ⫹3.00 D was chosen in order to be consistent with
a previous study by our group3 and is derived from the mean error
⫾1 SD of typical 6 to 8 year olds.19
In the AA group, 4 children had moderate/high hypermetropia
and 22 did not (including 4 myopes). In the IA group, 12 had
moderate/high hypermetropia and 15 did not (including 2
myopes). This difference between the two groups was significant
(␹2 ⫽ 5.307, p ⫽ 0.021). A significant difference between the
(non-normally distributed) refractive error data of the two groups
was also found when median values were analyzed [AA: ⫹1.00 D,
IA: ⫹2.50 D] (Mann–Whitney U: Z ⫽ ⫺3.082, p ⫽ 0.002).
One child (from the AA group) was excluded from analysis of
strabismus as he had pathology (coloboma) of one eye. Among the
remaining members of the AA group, 1 child had strabismus (es-
otropia) and 25 children had no strabismus. Among the IA group,
Optometry and Vision Science, Vol. 84, No. 2, February 2007
151
11 children had strabismus (1 exotropia, 10 esotropia) and 16
children had no strabismus. This difference between the two
groups was significant (␹2 ⫽ 10.294, p ⫽ 0.001). Of the eight
children in the IA group who had strabismus and wore glasses
(including one child with intermittent exotropia, two children
IA
with intermittent esotropia, two children with alternating esotro-
27 pia and three children with constant esotropia), seven had strabis-
mus that persisted even with spectacle correction. Two of these
27 children (including the one with intermittent exotropia) were
27 myopes, four were hypermetropes, and one an astigmatic em-
27 metrope. The one child who was orthophoric with his hyper-
metropic spectacle correction exhibited an alternating esotropia
22
uncorrected. The one child in the AA group who had strabismus
27 unrelated to pathology exhibited an intermittent, alternating es-
otropia with correction for significant astigmatism.
25 There was no significant difference in the number of children
with nystagmus (AA: n ⫽ 3, IA: n ⫽ 5), lens opacities (AA: n ⫽ 2,
IA: n ⫽ 0), or stereopsis (AA: n ⫽ 19, IA: n ⫽ 17) between the two
groups (p ⬎ 0.05). Whenever possible, testing for stereopsis was
attempted on all children with the exception of the child with
coloboma. However, stereopsis data were not available for two
children from the AA group, including the one child with strabis-
mus that was not related to pathology. Three other children in the
AA group were tested but failed the demonstration plate, indicat-
ing that they lacked the cognitive ability to complete the test (all
were orthophoric). Thus the remaining 21 children who were in-
cluded in analysis were able to manage the demonstration plate or
full test. Two children passed the demonstration plate but failed
the Frisby stereotest and were recorded as having no stereopsis;
both were orthophoric. These apparently nonstrabismic children
may have had undetected strabismus or lacked the motivation/skill
to pass the slightly harder task required for stereo positive testing.
In the IA group all children were tested for stereopsis. Five children
failed the demonstration plate (two were orthophoric and three
were constant esotropes). Of those children who were scored pos-
itive on stereo testing, three had intermittent esotropia, one was a
constant esotrope, and the rest were orthophoric. The child with
constant esotropia and apparent stereopsis was noted to have been
orthophoric at a previous visit. This would suggest an intermittent
element to the child’s strabismus.
There was also no significant difference between the (non-
normally distributed) median visual acuity of the two groups [AA:
18.95 cycles per degree (6/9.5), IA: 15.00 cycles per degree (6/12)]
(Mann–Whitney U: Z ⫽ ⫺0.304, p ⫽ 0.761).
Dynamic retinoscopy was performed in the right, or fixing, eye.
Seven children had anisometropia ⫽ ⬎1.00 D (three in the AA
group and four in the IA group), which meant that accommodative
responses could have been driven by the nonexamined eye, and
children misclassified as accurate or inaccurate accommodators.
The above statistical analysis was therefore repeated with anisome-
tropic children excluded, with no change in the findings. There
remained a significant difference in the number of moderate/high
hypermetropes (␹2 ⫽ 4.847, p ⫽ 0.028) and strabismics (␹2 ⫽
8.572, p ⫽ 0.003) with no difference in the number of children
with nystagmus, lens opacities, or stereopsis. The difference in
median refractive error also remained (AA: ⫹1.00 D, IA: ⫹2.50
D) (Mann–Whitney U: Z ⫽ ⫺3.147, p ⫽ 0.002).
152 Accommodative Accuracy, Hypermetropia, and Strabismus in Children with Down’s Syndrome—Stewart et al.

FIGURE 1.
Refractive error for 27 children with Down’s syndrome and accurate accommodation (AA). Refractive error is given by the spherical equivalent of the
fixing eye in children with strabismus, and right eye for children without strabismus. Filled markers represent those visits at which children wore
spectacles, and open markers visits at which children were not wearing spectacles. Dotted lines indicate the range for low hypermetropia/emmetropia
used in the study.

FIGURE 2.
Refractive error for 27 children with Down’s syndrome and inaccurate accommodation (IA). Refractive error is given by the spherical equivalent of the
fixing eye in children with strabismus, and right eye for children without strabismus. Filled markers represent those visits at which children wore
spectacles, and open markers visits at which children were not wearing spectacles. Dotted lines indicate the range for low hypermetropia/emmetropia
used in the study.

Longitudinal refractive error, (represented as spherical equiva-
lent for the right eye of children without strabismus and the fixing
eye of children with strabismus) is shown in Fig. 1 for AA children
and in Fig. 2 for IA children. The visits at which the children wore
a spectacle correction are indicated by filled symbols. The date at
which children were first prescribed spectacles was impossible to
know for children who were under the care of the Hospital Eye
Service (HES) or their own optometrist before joining the study.
Ten children in the AA group and 13 in the IA group had been
prescribed spectacles by the HES, their own optometrist, or the
study team by the date of the visit represented in the cross-sectional
analysis. Two children in the AA group and no children in the IA
group were noncompliant with spectacle wear. The two noncom-
pliant children were both emmetropic astigmats.
From the figures, greater prevalence of emmetropia among chil-
dren with accurate accommodation and of moderate/high hyper-
metropia among children with inaccurate accommodation is clear.
The figures also show the apparent stability of refraction among
the children. Most of the children who fall into the moderate/high
hypermetropic category at later ages were clearly hypermetropic at
their first visit.
Fifteen children were identified as being seen before the age of
12 month and over a time period incorporating the age of 30
month; nine children from the AA group and six children from the
IA group. Two visits were used for analysis of change in refractive
error, the closest visit to 6 month and the closest visit to 30 month.
Table 3 shows the mean refractive error and the change over time
in the two groups. Both groups showed a small increase in hyper-

Optometry and Vision Science, Vol. 84, No. 2, February 2007

 Accommodative Accuracy, Hypermetropia, and Strabismus in Children with Down’s Syndrome—Stewart et al.
TABLE 3.
Change in refractive error between 6 and 30 months
Accommodation group
AA (N ⫽ 9) IA (N ⫽ 6)
Age at younger visit (mo) 6.47 ⫾ 1.56 7.65 ⫾ 2.49
Spherical equivalent (D) ⫹0.58 ⫾ 1.57 ⫹3.08 ⫾ 1.75
Age at older visit (mo) 29.67 ⫾ 1.57 31.96 ⫾ 5.36
Spherical equivalent (D) ⫹1.25 ⫾ 1.29 ⫹3.31 ⫾ 2.47
Change in refractive error (D) ⫹0.67 ⫾ 1.52 ⫹0.23 ⫾ 1.77
metropia. There was no significant difference between the two
groups in the amount of refractive error change over time (t ⫽
0.510, df ⫽ 13, p ⫽ 0.618).
DISCUSSION
When children with Down’s syndrome with accurate accom-
modation were compared with children with inaccurate accommo-
dation (of the same age), there was a significant difference in the
degree of hypermetropia and prevalence of strabismus. Those
children who under-accommodated were more likely to have mod-
erate/high hypermetropia (ⱖ⫹3.00 D) and to be strabismic (the
majority esotropic). Conversely, children who accommodated
accurately were more likely not to have higher amounts of hyper-
metropia or strabismus. In all but one case, the strabismus of chil-
dren who wore glasses persisted both with and without correction.
The two groups of children (those with accurate and inaccurate
accommodation) were matched for age. They were also similar in
the prevalence of spectacle prescription. Compliance was slightly
poorer among the group with accurate accommodation, and this
may reflect the lower refractive error and therefore lack of per-
ceived benefit of spectacles on the part of the children.
The association between refractive error and accommodation in
children with Down’s syndrome reported in this study is in line
with the results of Haugen et al.,1 who showed that “stable, low-
grade hypermetropia (ⱕ⫹2.0 D)” (within the low hypermetropia/
emmetropia range according to our classification) was significantly
correlated with ‘normal’ accommodation, although the authors do
not state whether to define a child as having ‘normal’ accommo-
dation required repeated measures. In a study of emmetropisation
and accommodation in typical children, Ingram et al.20 followed
1119 hypermetropes (of ⫹5.00 D or more in any meridian on
cycloplegic retinoscopy; which would equate to ⱖ⫹3.00 D if a
working distance allowance of 2 D were applied) from the age of 6
month to 3.5 years. They report that children whose first and last
cycloplegic retinoscopies were ⱖ⫹5.50 D (i.e., who remained hy-
permetropic) had higher “apparent accommodative deficits”. They
measured accommodation using a modified version of Mohindra
retinoscopy in which the child binocularly fixated a retinoscope
light held at arms length, viewed in semidarkness. Plus lenses were
placed in front of one eye and their power progressively decreased
until a ‘with’ movement was observed. The end point was taken as
the lowest point at which the retinoscopy reflex was ‘against’. As
this technique for measuring accommodative response did not re-
quire the children to be wearing their distance spectacle correction,
Optometry and Vision Science, Vol. 84, No. 2, February 2007
153
it is difficult to compare the results of their study more directly
with our own.
In a cross-sectional study of children with Down’s syndrome,
Woodhouse et al.11 defining ‘normal’ accommodation by an ac-
commodative error index (AEI) within 2.20 D reported a lower
mean value of hypermetropia among the children within this
group than within the ‘abnormal’ group. Woodhouse et al. also
showed that children with strabismus had a poorer accommodative
response (greater AEI) than children without strabismus, although
strabismus was not significantly associated with a particular refrac-
tive error type.
However, contrary to the above findings and to the current
study, Haugen and Hovding5 found no significant difference in
the frequency of accommodation ‘weakness’ in children with
Down’s syndrome with (17 out of 25) and without (16 out of 35)
strabismus, based on cross-sectional analysis of the children’s last
visit in a longitudinal study. The mean age of the two groups was
not reported. The difference between the results of the two studies
might be explained by the use of different criteria to define accom-
modation ‘weakness’; a lag of ⱖ0.75 D at two or more testing
distances (10, 16.7, and 25 cm) was used in the current study,
whereas a lag of ⬎1.0 D at a testing distance of approximately 30
cm was used by Haugen and colleagues. More children may have
been classified as having accommodation ‘weakness’ if Haugen and
co-workers had assessed accommodation at a closer distance, i.e., at
a greater accommodative demand.
The association between inaccurate accommodation and
moderate/high hypermetropia is not due to the increased accom-
modative demand at near, because all children with significant
refractive error wore their glasses for accommodation measures.
Further, a previous study has shown that under-accommodation
remains similar whether hypermetropia is corrected or not.10
A possible cause for the higher incidence of strabismus in chil-
dren with reduced accommodation might be an abnormal link
between accommodation and convergence. Haugen and Hovding5
suggest that the common occurrence of esotropia in Down’s syn-
drome may be due to increased accommodative effort when having
to compensate for accommodative weakness. However, they com-
ment that other factors must play a role as accommodative weak-
ness is also found in children with Down’s syndrome without
strabismus. Other factors such as weak fusional capacity,21 hypo-
tonia,22 and dysfunction of the accommodation-convergence rela-
tion23 have also been suggested in the etiology of strabismus.
Although the association between strabismus and inaccurate ac-
commodation in the present study is striking, a causal relation
cannot at present be inferred.
In previous reports of under-accommodation in children with
Down’s syndrome, we postulated that premature aging of the lens
(early presbyopia) may be a causative factor.12,13 In light of this
suggestion, it is interesting to note that the two children in the
present study with lens opacities were both from the group with
accurate accommodation. More recently, Cregg et al.10 showed
that children with Down’s syndrome do not behave like presby-
opic adults, i.e., there is no demonstrable physical limitation to the
amount of accommodation they can exert and they found no
difference in accommodation response between corrected and un-
corrected hypermetropes (ⱖ⫹3.00 D). Haugen et al.24 took bio-
metric measurements in a group of young adults with Down’s
154 Accommodative Accuracy, Hypermetropia, and Strabismus in Children with Down’s Syndrome—Stewart et al.
syndrome (mean age 20.0 ⫾ 3.9 years). They found no significant
difference in mean lens thickness or calculated lens power between
the subjects with Down’s syndrome with normal accommodation
and those with weak accommodation. However, they did show that
the calculated lens power in subjects with Down’s syndrome was
significantly weaker than controls. It is therefore unclear to what
extent this influences accommodation. The etiology of the under-
accommodation in Down’s syndrome is yet to be determined.
It is tempting to hypothesize a link between accurate accommo-
dation providing a focused retinal image during near tasks, and
emmetropisation in children with Down’s syndrome. However,
although the association between accurate accommodation and
emmetropia is demonstrated in the present study, this cannot be
taken to indicate a causal relation. As Figs. 1 and 2 show (in line
with previously reported data4), the majority of children with
Down’s syndrome do not show emmetropisation. Rather, the chil-
dren with accurate accommodation appear more likely to be close
to emmetropia in the early months and remain so, while the chil-
dren with inaccurate accommodation appear more likely to have
moderate/high hypermetropia in infancy and to maintain the er-
ror. The role of accommodation in the emmetropisation process
(and conversely in the development of refractive error) remains
unclear for children with Down’s syndrome. In the literature on
typical refractive error development, “it has been claimed that ac-
commodation has a role in the normal postnatal development of
the eye.”,25 although considering the evidence as a whole this the-
ory is not well supported.26 However, there is strong evidence for
the role of excessive near-work in the development of refractive
error (myopia), although this is likely to be a result of factors other
than accommodation.27,28 If the theory of accommodative in-
volvement in myopia development in typically developing chil-
dren were true for children with Down’s syndrome (blur induced
by slight under-accommodation during sustained periods of near
work), it seems surprising that (with the high prevalence of signif-
icant under-accommodation in Down’s syndrome and therefore
retinal blur) more children with Down’s syndrome are not myopic.
In fact, as this study shows, in children with Down’s syndrome the
association is quite the opposite. Perhaps reduced accommodation
does not precede the development of moderate/high hyper-
metropia but rather accompanies it? Here we offer a very small-
scale study of longitudinal change in refractive error over the first
30 month. Among typical children, there is a reduction in hyper-
metropia over the first year and a half of life,29 and a similar
reduction has been shown among children with hypermetropia
between 9 and 36 month, whether corrected or not.30 Among our
children with Down’s syndrome, there is no difference in the ex-
tent to which refractive error changes between the group with
accurate and the group with inaccurate accommodation, and in
fact, both groups show a small increase in hypermetropia, well
within the repeatability error of retinoscopy.17 This suggests that
accurate or inaccurate accommodation is not influencing emmet-
ropisation, but is associated with the refractive error that children
have in very early infancy. Children with Down’s syndrome are
known to have retarded growth that is consistent with the small
eyes of hypermetropia,23 so refractive error development may be
determined in Down’s syndrome by nonoptical factors. Certainly
the development of myopia in Down’s syndrome is associated with
nonoptical factors, namely heart defects.31
Optometry and Vision Science, Vol. 84, No. 2, February 2007
Studies of refractive error and biometric measurements in peo-
ple with Down’s syndrome have shown a correlation between axial
length and refractive error (spherical equivalent)32 and more sig-
nificantly axial length/corneal curvature ratio and spherical equiv-
alent.24 In a study by Haugen et al.,24 the group with Down’s
syndrome were found to have an increased mean keratometry
value, a significant reduction in central corneal thickness, a thinner
lens and higher axial density value, weaker calculated lens power,
and significantly shorter axial length. Despite these significant dif-
ferences compared to controls, the biometric components were still
related in an ordered fashion, similar to controls. Thus, the higher
axial length/corneal radius ratio found in subjects with Down’s
syndrome, was compensated for by a weaker lens. When examined
in a multiple regression analysis, the strongest correlates with
spherical equivalent were axial length and keratometry.
We have recently shown that bifocal spectacles can induce ac-
curate accommodation among school-age children with Down’s
syndrome.33 Prospective studies of early development with both
naturally occurring and bifocal-induced accurate accommodation
may shed light on the precise relation between accommodative
accuracy, refractive error progression, and strabismus in children
with Down’s syndrome.
ACKNOWLEDGMENTS
Financial support for the longitudinal data used in this study has come from
the following sources over the last 12 years; the Medical Research Council, the
Down’s Syndrome Association, Healthcare Foundation, Mencap City Foun-
dation, and the British College of Optometrists.
We thank the children and parents involved in the Cardiff Down’s Syn-
drome Cohort for their continued support of our work.
Received June 6, 2006; accepted November 7, 2006.
REFERENCES
1. Haugen OH, Hovding G, Lundstrom I. Refractive development in
children with Down’s syndrome: a population based, longitudinal
study. Br J Ophthalmol 2001;85:714–19.
2. da Cunha RP, Moreira JB. Ocular findings in Down’s syndrome.
Am J Ophthalmol 1996;122:236–44.
3. Woodhouse JM, Pakeman VH, Cregg M, Saunders KJ, Parker M,
Fraser WI, Sastry P, Lobo S. Refractive errors in young children with
Down syndrome. Optom Vis Sci 1997;74:844–51.
4. Cregg M, Woodhouse JM, Stewart RE, Pakeman VH, Bromham
NR, Gunter HL, Trojanowska L, Parker M, Fraser WI. Development
of refractive error and strabismus in children with Down syndrome.
Invest Ophthalmol Vis Sci 2003;44:1023–30.
5. Haugen OH, Hovding G. Strabismus and binocular function in chil-
dren with Down syndrome. A population-based, longitudinal study.
Acta Ophthalmol Scand 2001;79:133–9.
6. Courage ML, Adams RJ, Reyno S, Kwa PG. Visual acuity in infants
and children with Down syndrome. Dev Med Child Neurol 1994;
36:586–93.
7. Millis E. Visual acuity in children with Down’s syndrome: the influ-
ences of changes in the eye and the effect of environmental and
biological factors. 1988. PhD Thesis. University of London.
8. Woodhouse JM, Pakeman VH, Saunders KJ, Parker M, Fraser WI,
Lobo S, Sastry P. Visual acuity and accommodation in infants and
young children with Down’s syndrome. J Intellect Disabil Res 1996;
40(Pt 1):49–55.
 Accommodative Accuracy, Hypermetropia, and Strabismus in Children with Down’s Syndrome—Stewart et al.
9. Courage ML, Adams RJ, Hall EJ. Contrast sensitivity in infants and
children with Down syndrome. Vision Res 1997;37:1545–55.
10. Cregg M, Woodhouse JM, Pakeman VH, Saunders KJ, Gunter HL,
Parker M, Fraser WI, Sastry P. Accommodation and refractive error
in children with Down syndrome: cross-sectional and longitudinal
studies. Invest Ophthalmol Vis Sci 2001;42:55–63.
11. Woodhouse JM, Cregg M, Gunter HL, Sanders DP, Saunders KJ,
Pakeman VH, Parker M, Fraser WI, Sastry P. The effect of age, size of
target, and cognitive factors on accommodative responses of children
with Down syndrome. Invest Ophthalmol Vis Sci 2000;41:2479–85.
12. Woodhouse JM, Meades JS, Leat SJ, Saunders KJ. Reduced accom-
modation in children with Down syndrome. Invest Ophthalmol Vis
Sci 1993;34:2382–7.
13. Leat SJ, Gargon JL. Accommodative response in children and young
adults using dynamic retinoscopy. Ophthalmic Physiol Opt 1996;16:
375–84.
14. McClelland JF, Saunders KJ. The repeatability and validity of dy-
namic retinoscopy in assessing the accommodative response. Oph-
thalmic Physiol Opt 2003;23:243–50.
15. Chauhan K, Charman WN. Single figure indices for the steady-state
accommodative response. Ophthalmic Physiol Opt 1995;15:
217–21.
16. Mohindra I. A non-cycloplegic refraction technique for infants and
young children. J Am Optom Assoc 1977;48:518–23.
17. Saunders KJ, Westall CA. Comparison between near retinoscopy and
cycloplegic retinoscopy in the refraction of infants and children. Op-
tom Vis Sci 1992;69:615–22.
18. Rouse MW, Hutter RF, Shiftlett R. A normative study of the accom-
modative lag in elementary school children. Am J Optom Physiol
Opt 1984;61:693–7.
19. Kempf GA, Collins SD, Jarman BL. Refractive errors in the eyes of
children as determined by retinoscopic examination with a cyclople-
gic. Results of eye examinations of 1860 white school children in
Washington, DC, by Grover A. Kempf, surgeon, Selwyn D. Collins,
Associate Statistician, Bernard L. Jarman, A. A., Surgeon, United
States Public Health Service. Washington, DC: Government Print-
ing Office; 1928:1–56.
20. Ingram RM, Gill LE, Goldacre MJ. Emmetropisation and accommo-
dation in hypermetropic children before they show signs of squint—a
preliminary analysis. Bull Soc Belge Ophtalmol 1994;253:41–56.
21. Jaeger EA. Ocular findings in Down’s syndrome. Trans Am Ophthal-
mol Soc 1980;78:808–45.
22. Eissler R, Longenecker LP. The common eye findings in mongolism.
Am J Ophthalmol 1962;54:398–406.
Optometry and Vision Science, Vol. 84, No. 2, February 2007
155
23. Fanning GS. Vision in children with Down’s syndrome. Aust J Op-
tom 1971;54:74–82.
24. Haugen OH, Hovding G, Eide GE. Biometric measurements of the
eyes in teenagers and young adults with Down syndrome. Acta Oph-
thalmol Scand 2001;79:616–25.
25. Raviola E, Wiesel TN. An animal model of myopia. N Engl J Med
1985;312:1609–15.
26. Troilo D. Neonatal eye growth and emmetropisation—a literature
review. Eye 1992;6:154–60.
27. McBrien NA, Barnes DA. A review and evaluation of theories of
refractive error development. Ophthalmic Physiol Opt 1984;4:
201–13.
28. McBrien NA, Moghaddam HO, Reeder AP. Atropine reduces exper-
imental myopia and eye enlargement via a nonaccommodative mech-
anism. Invest Ophthalmol Vis Sci 1993;34:205–15.
29. Saunders KJ, Woodhouse JM, Westall CA. Emmetropisation in hu-
man infancy: rate of change is related to initial refractive error. Vision
Res 1995;35:1325–8.
30. Atkinson J, Anker S, Bobier W, Braddick O, Durden K, Nardini M,
Watson P. Normal emmetropization in infants with spectacle correc-
tion for hyperopia. Invest Ophthalmol Vis Sci 2000;41:3726–31.
31. Bromham NR, Woodhouse JM, Cregg M, Webb E, Fraser WI. Heart
defects and ocular anomalies in children with Down’s syndrome. Br J
Ophthalmol 2002;86:1367–8.
32. Doyle SJ, Bullock J, Gray C, Spencer A, Cunningham C. Emmetro-
pisation, axial length, and corneal topography in teenagers with
Down’s syndrome. Br J Ophthalmol 1998;82:793–6.
33. Stewart RE, Margaret Woodhouse J, Trojanowska LD. In focus: the
use of bifocal spectacles with children with Down’s syndrome. Oph-
thalmic Physiol Opt 2005;25:514–22.
34. Saunders KJ, Woodhouse JM, Westall CA. The modified frisby ste-
reotest. J Pediatr Ophthalmol Strabismus 1996;33:323–7.
35. Adoh TO, Woodhouse JM. The Cardiff Acuity Test used for mea-
suring visual acuity development in toddlers. Vision Res 1994;34:
555–60.
36. Woodhouse JM, Adoh TO, Oduwaiye KA, Batchelor BG, Megji S,
Unwin N, Jones N. New acuity test for toddlers. Ophthalmic Physiol
Opt 1992;12:249–51.
J. Margaret Woodhouse
School of Optometry and Vision Sciences
Cardiff University, Redwood Building, King Edward VII Avenue
Cardiff, CF10 3NB, United Kingdom
e-mail: woodhouse@cf.ac.uk