Br J Ophthalmol 1998;82:793–796 793

Emmetropisation, axial length, and corneal

topography in teenagers with Down’s syndrome
Stephen J Doyle, John Bullock, Cath Gray, Alison Spencer, CliV Cunningham

Abstract astigmatism is probably related to this

Aim—To study the refractive status and failure of emmetropisation. The preva-
corneal topography in Down’s syndrome. lence of 2% keratoconus in Down’s syn-
Method—A matched cohort subgroup of drome compares with that found by other
50 individuals with Down’s syndrome in authors of between 5.5 and 15%. The 6%
the Manchester area aged 15–22 years was with inferior steepening on topography
studied by refraction, corneal topography, will be followed up over the next few years
A-scan biometry, slit lamp examination, to see if there is any development of clini-
and orthoptic examination. cal keratoconus. Hence we will see if
Results—(1) A linear relation was found corneal topography is useful as a screen-
between axial length and spherical ing tool for preclinical keratoconus in this
equivalent refraction. There was no statis- high risk group.
tical relation between keratometry and the (Br J Ophthalmol 1998;82:793–796)
axial length. (2) 80% of the group had a
hyperopic refraction (mean +2.46 D,
range +0.5 to +7.5 D); 18% were myopic Individuals with Down’s syndrome have been
(mean −2.75 D, range −0.5 to −8.0 D); and reported to have a higher incidence of refrac-
2% were emmetropic (within plus or tive errors1–3 and keratoconus.4–6 All these stud-
minus 0.5 D of zero). The overall mean ies have been on random groups, presenting to
spherical equivalent refraction was +1.43 normal clinical practice or found in various
(SD 2.86) D. 63% of eyes could see 6/12 or institutions. We present some findings from a
study on a representative sample of the Greater
better and 66% of the individuals had a
Manchester Down’s Syndrome Cohort, all of
binocular vision of 6/12 or better. (3) Cor-
whom resided in the family home.
neal topography was generally of a regular
“bow tie” pattern, but there was a high
Manchester Royal Eye incidence of oblique cylinders. Mean cyl- Method
Hospital inder strength was 1.14 (1.15) D. (4) The A cohort consisting of 90% of the total known
S J Doyle prevalence of overt keratoconus was 2%. Down’s syndrome births in Greater Manches-
J Bullock 6% had corneal topography with inferior ter over the 7 year period between August 1973
C Gray
A Spencer steepening which may be a preclinical and August 1980 had been studied since birth
keratoconic process. by the Hester Adrian Research Centre of the
Liverpool John Moores Conclusions—In this cohort of late teen- University of Manchester and comprised 181
University agers with Down’s syndrome, emmetropi- individuals. Their medical, social, and psycho-
C Cunningham sation has failed to occur in most logical histories were well known because of
Correspondence to:
individuals. In a similar aged group of this study.7 By September 1995, 118 of these
Dr Doyle, Manchester Royal non-disabled individuals one would expect were still living locally, the rest having been lost
Eye Hospital, Oxford Road, about 83% emmetropic (plus or minus to the study because of movement out of the
Manchester M13 9WH. area or death. All 118 were contacted by post
0.25 D), 13% myopic, and 4% hyperopic.
Accepted for publication The wide spread of oblique cylinders and and in the period allocated for the study, 57
19 November 1997 the small proportion of with the rule had been studied, of which 50 were selected
and agreed by the original researchers of the
8 cohort to be representative of the total cohort
6 for age, mental ability, general health, and
Spherical equiv refraction (D)

social class. This subgroup (29 male, 21

4 female) of ages 15–22 years (mean 17.4 years)
2 was studied by refraction, orthoptic examin-
ation, corneal topography, A-scan biometry,
0 and slit lamp examination. Refraction was
–2 done in the normal way, without cycloplegia.
Corneal topography was done using a TMS-1
–4
machine. Biometry was done using a Storz
–6 A-scan ultrasound machine. Slit lamp examin-
ation and fundal examination through the
–8
undilated pupil using an indirect 78 D lens was
–10 also done.
20 21 22 23 24 25 26
Axial length (mm)
Results
Figure 1 Plot of spherical equivalent refraction v axial length as measured by A-scan A linear relation was found between axial length
biometry. and spherical equivalent refraction (Fig 1).
794 Doyle, Bullock, Gray, et al

8 Corneal topography was generally of a regu-

SE right eye
lar “bow tie” pattern that fitted the optical
6 refraction, but there was a high incidence of
Spherical equivalent (D) SE left eye
4
oblique cylinders with 38% of eyes having their
axis outside 10° from the orthogonal axes; 22%
2 had cylinders “with the rule” (plus or minus
10° of 90°) and 39% had cylinders “against the
0 rule” (plus or minus 10° of 180°) (Fig 4).
–2
Mean cylinder strength was 1.14 (SD 1.15)
D (Fig 5). Topography was successful in all but
–4 one subject and he had minimal astigmatism.
Hence, it was felt unlikely that he would have
–6 any irregular topography. Average keratometry
–8
for the group was 43.97 (2.17) D. There was
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 no statistical relation between keratometry and
Patients the axial length.
Figure 2 Bar chart of spherical equivalent refraction for each eye. There was one (2%) overt keratoconus (Fig
6). Three (6%) had corneal topography with
1 inferior steepening which may be a preclinical
keratoconic process (Fig 7).
Visual actuity (decimal fraction)

0.9 Right eye

Left eye Two (4%) had childhood squint operations,
0.8
although neither of these had any amblyopia.
0.7 There were three amblyopic eyes in three
0.6 patients, two associated with large cylinders
and small A esotropia patterns. The other
0.5
amblyope was in a V esotrope. There were four
0.4 other A esophorias, giving a total of six (12%)
0.3 with an A esotropic pattern. There were no
exophorias or exotropias. Five (10%) had
0.2
latent nystagmus.
0.1 Fourteen (28%) had minor bilateral lens
0
opacities of no visual significance in any case.
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 Five (10%) had “flake” lens opacities, two
Patients (4%) had sutural changes, and seven (14%)
Figure 3 Bar chart of visual acuity for each eye. had “blue dot” opacities. The posterior poles
were normal in all cases. No optic disc had a
10 cup/disc ratio of greater than 0.3.
In terms of general health, 36% (18) had
8
conductive hearing loss; this was mostly due to
repeated middle ear infections, and nine (18%)
Number of cases

had had grommets, tonsillectomy, and/or

6 adenoidectomy.

4 Discussion
The linear relation between refraction and
axial length (and also vitreous cavity length) in
2 Down’s syndrome is also found in the normal
population.8 The process of emmetropisation
0
in Down’s syndrome in the first 2 years of life
10.00 18.00 25.00 40.00 51.00 60.00 75.00 85.00 95.00 105.00 115.00 135.00 155.00 165.00 175.00 has been studied by Woodhouse’s group at the
Axis of positive cylinder University of Wales, CardiV School of
Figure 4 Bar chart of axes of the refractive cylinders. Optometry.9 They have shown that the distri-
bution of refractive errors among infants with
Forty (80%) of the group had a hyperme- Down’s syndrome is similar to the norm; but,
tropic refraction ( mean +2.46 D, range +0.5 rather than narrowing with age as in the
to +7.5 D); nine (18%) were myopic (mean normal population, the distribution widens,
−2.75 D, range −0.5 to −8.0 D); one (2%) was and the prevalence of refractive errors in-
emmetropic (within plus or minus 0.5 D of creases in children with Down’s syndrome. In
zero). The overall mean spherical equivalent our cohort of teenagers with Down’s syn-
refraction was +1.43 (SD 2.86) D (Fig 2). In drome, it is clear that emmetropisation has
terms of visual acuity, 63% of eyes could see failed to occur in most individuals. In a similar
0.5 (6/12) or better and 66% of the individuals aged group of non-disabled children one
had a binocular vision of 6/12 or better (Fig 3). would expect about 83% emmetropic (plus or
Visual acuity was measured with a Snellen minus 0.25 D), 13% myopic, and 4%
chart at 6 metres in 40 cases, Sheridan- hyperopic.10 Cylinders also tend to decrease
Gardiner cards in six cases, two used reduced with emmetropisation, but this also appears
Snellen, and two used CardiV charts. not to have occurred with this group with a
Anisometropia had a mean of 0.4 D (SD mean cylinder of 1.14 (1.15) D. The wide
0.47 D, max 1.5 D); 58% wore glasses at least spread of oblique cylinders and the small
some of the time. proportion of with the rule astigmatism is again
Emmetropisation, axial length, and corneal topography in teenagers with Down’s syndrome 795

4 This group was part of a cohort that had

Cylinder RE been closely followed since birth by the Hester
3.5 Adrian Research Centre and had appropriate
Cylinder LE
3
intervention at an early stage. Often, Down’s
individuals do not develop as they should
Cylinder (D)

2.5 because of poor hearing and/or vision. Screen-

ing in childhood should hence be especially
2
thorough as they are prone to middle ear
1.5 problems and refractive errors, either of which
can stunt intellectual and social development.
1 This group performed visually surprisingly
well, with two thirds seeing 6/12 or better. We
0.5
think that visual performance has been
0 maximised by early refraction and appropriate
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49
glasses in almost all cases, with good follow
Patients up, and it emphasises the importance of good
Figure 5 Bar chart of magnitude of the refractive cylinders. early ophthalmic care for those with Down’s
syndrome.
It is interesting that the posterior segments
were normal in all cases. All the ocular
pathologies seem to be confined to the
anterior segment. We do not know if there is a
reason for this or whether it is a chance
finding.
The prevalence of keratoconus in Down’s
syndrome has been reported as being between
5.5 and 15% by various authors4–6 compared
with the general population of about 50 per
100 000 (0.05%).12 Many reports in the litera-
ture suggest an abnormality in collagen me-
tabolism in patients with keratoconus. The
gene encoding the á-1 chain of type VI
collagen, a major constituent of the corneal
stroma, is on chromosome 21 at locus
21q22.3. As Down’s has a trisomy 21, it has
been speculated there might be a connection
between this gene and the higher incidence of
Figure 6 Corneal topography of subject with clinical keratoconus.
keratoconus in Down’s syndrome. Indeed, one
team has done linkage analysis to see if a muta-
tion of this collagen VI gene contributed to a
keratoconus occurring in three generations of
one family.13 The study excluded a gene locus
for keratoconus on the most telomeric region
of chromosome 21 in this family group, but the
interesting speculation remains. Keratoconus
is essentially a post-pubertal disease and
mostly presents between the ages of 15 and 25
years (Manchester Royal Eye Hospital Kerato-
conus Survey, 1995, unpublished) The fact
that we found one (2%) overt keratoconic is
not surprising. What is more interesting is to
see whether the three with inferior steepening
on topography become truly keratoconic over
the next few years. With corneal topography
becoming more widely available, the possibility
of screening a higher risk group for kerato-
conus such as individuals with Down’s syn-
drome becomes more feasible. Added risk fac-
Figure 7 Corneal topography of subject with inferior steepening
tors for keratoconus development, as in the
probably related to this failure of emmetropisa- non-Down’s population, are atopy, an eye rub-
tion. It is known that corneal topography bing habit, and myopic shift in refraction.
changes from with the rule to against the rule Individuals with Down’s syndrome showing
with aging in the normal population11 and indi- any of these characteristics should probably
viduals with Down’s syndrome age more have corneal topography in their mid to late
rapidly than normal. However, it is unlikely teens as a screen for possible early keratoconus.
that this is the cause among these teenagers as Our understanding of emmetropisation and
the onset of more rapid aging does not occur the pathogenesis of keratoconus is at best
until much later.7 sketchy and much work remains to be done.
796 Doyle, Bullock, Gray, et al

1 Catalano RA. Down syndrome. Surv Ophthalmol 9 Woodhouse JM, Pakeman VH, Parker M, et al. Refractive
1990;34:385–98. errors in infants and young children with Down’s
2 Shapiro MB, France TD. The ocular features of Down’s syndrome. MAMH Congress “The mentally retarded in
syndrome. Am J Ophthalmol 1985;99:659–63. 2000’s Society”, Rome 1994 and personal communica-
3 Eissler R, Longnecker JP. The common eye findings in tion.
mongolism. Am J Ophthalmol 1962;54:398–406. 10 Teasdale TW, Goldschmidt E. Myopia and its relationship
4 Cullen JF, Butler HG. Mongolism (Down’s syndrome) and to education, intelligence and height. Acta Ophthalmol
keratoconus. Br J Ophthalmol 1963;47:321–30. Suppl 1988;185:41–3.
5 Pierse D, Eustace P. Acute keratoconus in mongols. Br J
Ophthalmol 1971;55:50–4. 11 Hayashi K, Hayashi H, Hayashi F. Topographic analysis of
6 Walsh SZ. Keratoconus and blindness in 469 institutional- the changes in corneal shape due to ageing. Cornea
ised subjects with Down’s syndrome and other causes of 1995;14:527–32.
mental retardation. J Ment Defic Res 1981;25:243–51. 12 Kennedy RH, Bourne WM, Dyer JA. A 48 year clinical and
7 Turner S, Sloper P, Knussen C, et al. Health problems in epidemiological study of keratoconus. Am J Ophthalmol
children with Down’s syndrome. Child, Care, Health and 1986;101:267–73.
Development 1990;16:83–97. 13 Rabinowitz YS, Maumenee IH, Lundergan MK, et al.
8 Hosaka A. The growth of the eye and its components. Acta Molecular genetic analysis in autosomal dominant kerato-
Ophthalmol Suppl 1988;185:65–8. conus. Cornea 1992;11:302–8.