269

REFRACTIONS AND REFLECTIONS

NUMBER 3

NOVEMBERDECEMBER 2001

2001 by Elsevier Science Inc. 0039-6257/01/$see front matter
All rights reserved. PII S0039-6257(01)00260-0

JOSEPH MILLER, EDITOR

Dynamic Retinoscopy: The Missing Data

David G. Hunter, MD, PhD

The Krieger Childrens Eye Center at the Wilmer Institute, The Johns Hopkins University School of Medicine, Baltimore,
Maryland, USA

Abstract.

Dynamic retinoscopy is a well described but often overlooked technique that allows rapid
assessment of accommodative ability. The key to the technique is the neutralization of the retinoscopic
reflex that occurs when the patient accommodates on a target adjacent to the retinoscope. This clinical
tool can provide critical data that can help solve treatment dilemmas, such as when a child presents
with high hyperopia or when a patient presents at any age with possible accommodative insufficiency.
In this review, performance of dynamic retinoscopy will be detailed, with the applicability of the tech-
nique demonstrated with use of case examples. (

Surv Ophthalmol 46:

269274, 2001. 2001 by
Elsevier Science Inc. All rights reserved.)

Key words.

accommodation

dynamic retinoscopy

esotropia

refraction

A 6-month-old boy referred for chronic overflow
tearing has a cycloplegic refractive error of

5.00 D
OU. Should you treat or observe? A 16-month-old
girl has 1.75 D of anisometropia. Is she beginning to
develop amblyopia? A 25-year-old woman complains
of asthenopia and headaches that have persisted de-
spite full recovery from a job-related hyphema 6
months previously. Near acuity is J1

OU. Is she
malingering?
The clinical situations above are often considered
difficult treatment dilemmas. Why so difficult? The
clinician is missing a key piece of information;
namely, how well (and how symmetrically) can the
patient accommodate in real-life situations? That
missing information can be readily available with use
of a well described

3,6

but often overlooked tech-
nique: dynamic retinoscopy. The purpose of this ar-
ticle is to review the performance and application of
this valuable clinical tool.

Retinoscopy

During retinoscopy, the examiner views the red
reflex of the eye through the peephole of the retino-
scope while sweeping a linear streak of divergent
light across several meridians. When the eye is fo-
cused in the plane of the peephole, all light return-
ing from the retina passes through the peephole,
and the red reflex appears to fill with light (neutral-
ization, Fig. 1A). When the subjects eye is focused
beyond the peephole (behind the examiner), only a
portion of returning light passes through the peep-
hole, and the red reflex appears as a band of light
(the shape of the bulb filament) in the same merid-
ian as that of the filament, moving in the same direc-
tion as the light source (with motion, Fig. 1B).
When the eye is focused in front of the peephole,
the band of light moves in the direction opposite the
light source (against motion). For accurate deter-
mination of distance refraction, the patient is en-
270 Surv Ophthalmol 46 (3) NovemberDecember 2001

HUNTER

couraged to maintain a state of fully relaxed accom-
modation. This is accomplished through fixation on
a distant target, or by using cycloplegia in some
adults and most children.
For dynamic retinoscopy, cycloplegia is not used,
and accommodation is not relaxed. The goal is to
ask the patient to use accommodation to move the
point of best focus (the point that is optically conju-
gate to the retina) at distance to the peephole of the
retinoscope at near (Fig. 1A), without the use of ad-
ditional lenses. This provides rapid and accurate as-
sessment of accommodative ability in both eyes, as
the examiner quickly switches back and forth be-
tween eyes. Several variations of the technique, some
requiring plus lenses, have been described.

2,4,7,10

This
report will describe techniques of dynamic retinos-
copy that do not (for the most part) utilize addi-
tional plus lenses.

Dynamic Retinoscopy Technique

A near fixation target must be provided to per-
form the test. This typically consists of a small letter
chart. Hold the target with the free hand, or secure
it as close as possible to the retinoscope peephole
without occluding the peephole or the light source
(Fig. 2A). Seat the patient in the examining chair,
with the room darkened to facilitate visualization of
the retinoscopic reflex. Direct a dim light source
(typically a dim reading lamp attached to the chair
stand) toward the examiner so that the patient can
see the near fixation target in the dim room. The pa-
tient should wear distance correction, except in spe-
cial circumstances (for example, to determine how
well the patient can function without glasses).
Ask the subject to fixate on a distant target. The
retinoscopic reflex is assessed in both eyes, and
with motion should be observed. Check several
meridians in case astigmatism is present. Next ask
the subject to fixate on the near target, with the reti-
noscope and target held approximately at the nor-
mal reading position. The with motion will rapidly
change to neutralization or slight against motion.
If neutralization is incomplete, try moving back
slightly to see if accommodation is borderline, or en-
courage the patient to try harder (looking at finer
print, for example). The examiner, retinoscope, and
target should move together as a single unit. Switch
back and forth quickly between the two eyes during
the assessment. Avoid off-axis viewing, which can
produce misleading responses. Again ask the subject
to fixate on the distant target. The reflex should rap-
idly convert to with movement. Now move closer
(Fig. 2B) and ask the subject to maintain fixation
at near for a longer period of time. This places
more demands on the accommodative system, which
helps estimate whether the accommodative effort is
sustainable.
The results of dynamic retinoscopy, when normal,
can be described as rapid, complete, and steady
OU. Examples of abnormal responses can thus in-
clude incomplete, sluggish, momentary accom-
modation only, accommodative lag, or asymmet-
Fig. 1. Schematic representation of standard retinoscopy. The light source illuminating the eye is not shown, but is as-
sumed to be a linear streak of divergent light produced by the linear filament of the retinoscope. The return path of light
from patient to examiner is shown A: At neutralization, the far point of the eye is at the peephole of the retinoscope; all
returning light passes through and the entire pupil appears to fill with light. B: When the far point of the eye is behind
the examiner, a portion of returning light is blocked by the retinoscope peephole; reflected light thus appears in only
part of the pupil, and with motion is observed as the streak is swept across the eye.
DYNAMIC RETINOSCOPY

271

ric. Dynamic retinoscopy can thus be performed,
and the results recorded, in seconds. If possible, dy-
namic retinoscopy should be performed routinely
on new patients prior to cycloplegia to avoid the
need for a postcycloplegic evaluation.

Alternatives and Limitations

There are potential alternatives to dynamic retin-
oscopy, but each has disadvantages. Near acuity mea-
surement provides some indication of maximal ac-
commodative ability, but it does not provide any
indication of the effort required to read the letters,
and it is subject to errors if the wrong testing dis-
tance is used or an improper add is given. Formal
measurement of accommodative amplitudes is time
consuming and must be performed monocularly,
making it impossible to compare the accommoda-
tive state of the two eyes simultaneously. Measure-
ment of accommodative amplitudes provides no in-
formation about the time required to achieve the
endpoint or of the stability of the endpoint. These
tests are unreliable and time-consuming when per-
formed on older children, and not applicable to pre-
verbal children.
All assays of accommodation, including dynamic
retinoscopy, may generate false responses. Falsely
abnormal results are obtained if a subject is inat-
tentive or uncooperative. This can be avoided by
strongly encouraging the patient to attend to a near
target with fine detail. Falsely normal results can oc-
cur if a patient with accommodative insufficiency is
able to generate an accommodative response for the
duration of the test, but is unable to maintain this
level of accommodation for extended time periods
when reading at home. A false normal result is less
likely to occur if the patient is asked to maintain fo-
cus on the near target for several seconds during
testing; patients who are symptomatic generally are
not able to maintain near focus for long under the
intense demands of dynamic retinoscopy.
Near retinoscopy is used to measure distance re-
fraction, not accommodation. This technique, de-
scribed by Mohindra et al,

9

should not be confused
with dynamic retinoscopy. Near retinoscopy is per-
formed under monocular conditions. The retino-
scope filament serves as the target. Under these cir-
cumstances, there is normally no accommodative
stimulus (but see the section Infants, below). The
distance refraction is estimated empirically by sub-
tracting 1.25 D from the value obtained.

Special Situations

CHILDREN

It may be necessary to ask the patient to read the
letters out loud or to name a detail in a small picture
to stimulate accommodation. It is best to avoid ques-
tions such as what color is the bird, for these ques-
tions will stimulate cogitation rather than accommo-
dation. Instead, ask for observational detail: How
many toes are on the birds feet?

ACCOMMODATIVE LAG

If the conversion of the with retinoscopic reflex
to neutralization at near is equivocal, try separating
the target from the retinoscope, moving it closer to
Fig. 2. Dynamic retinoscopy. A: A fine target, such as a small letter chart, is placed adjacent to, but not occluding, the
retinoscope peephole and light source. B: The examiner attracts attention to the accommodative target and evaluates
the retinoscopic reflex in both eyes. In this case, the examiner has moved quite close to the patient to further stress the
patients accommodative system.
272 Surv Ophthalmol 46 (3) NovemberDecember 2001

HUNTER

the patient while keeping the retinoscope in the
same place. This should generate additional accom-
modative effort and produce unequivocal against
movement. If the target must be moved more than
12 cm in front of the retinoscope to achieve neu-
tralization, the patient may be demonstrating clini-
cally significant accommodative lag, that is, the pa-
tient is able to accommodate, but not able to
generate sufficient accommodation to focus accu-
rately on the near target.

ASTIGMATISM

In all patients, initially assess the reflex in multiple
meridians to identify significant astigmatism. When
uncorrected astigmatism is present, the near reflex
will remain with in some meridians, and be neu-
tralized or even switched to against in other merid-
ians. When astigmatism is observed, it should be
quantified with a standard refraction.

INFANTS

In infants, dynamic retinoscopy is performed by
assessing the response to an appropriate near target
only; there is no need to assess the change in the re-
flex during fixation on a distant target. At this age,
there still may be little interest in the standard pedi-
atric accommodative targets, and little accommoda-
tive ability may be detected before 24 months of
age.

5,11

Larger squeaky toys may help create interest.
Two maneuvers that may go against common clini-
cal judgment are useful when an infant appears to
show no accommodative response to a target. First,
move closer. Recall that babies have a high capacity
for accommodation but little interest in remote tar-
gets. Up to age 4 months, the accommodative re-
sponse may be all or none, with neutralization noted
at a distance of just 0.2 m in some cases. Second, try
using a small fixation light as a target. Although a
white light is not a good accommodative target in
older children under monocular conditions (see
Near Retinoscopy, above), in our experience it is
quite a good accommodative stimulus for infants
when viewed binocularly.

Case Examples

How can we use dynamic retinoscopy to solve a va-
riety of clinical dilemmas? Consider the following
examples.

CASE I

A 6-month-old boy referred for chronic overflow tearing
has a cycloplegic refractive error of

5.00 OU. There is no
strabismus. Should you treat or observe?

Dynamic retinoscopy can be reassuring in these
cases. If this boy can accommodate normally at near
while maintaining centered corneal light reflexes,
frequent observation alone is probably safe. If ac-
commodation is insufficient, repeat the assessment
through partial hyperopic correction. For example,
start with

2.00 lenses; if accommodation is still in-
sufficient, try increasing the amount of plus power
in 1- to 2-D increments until the patient is able to
neutralize the reflex at near. Consider deferring
glasses until similar measurements have been ob-
tained on more than one visit. If the patient cannot
accommodate to a near target (by dynamic retinos-
copy) without the aid of partial hyperopic correction
on consecutive visits, prescribe glasses.
In my experience performing dynamic retinos-
copy routinely on highly hyperopic infants and chil-
dren over the past 10 years, I have found that many
high hyperopes develop a strategy for avoiding eso-
tropia while accommodating 815 D to view near
targets. These children will accommodate fully, but
only briefly, on the near target; presumably they ac-
commodate just long enough to identify key features
of the object of interest. Once the object has been
evaluated, the child will immediately relax accom-
modation. This might be considered an accommo-
dative burst adaptation. The response is recorded
as rapid and complete, but momentary. As they
grow, these patients tend to either 1) require partial
hyperopic correction if the hyperopia does not
change, or 2) become less hyperopic, thereby avoid-
ing the need for glasses.
If dynamic retinoscopy had not been assessed
prior to cycloplegia, the patient will need to return
for a postcycloplegic evaluation before a treatment
decision is made. Consider this a good opportunity
to warn the parents that glasses may be in the childs
future. Parents do not react favorably to the idea
that their normal-appearing child may need glasses,
and spreading the shock out over several visits reli-
ably defuses the reaction.
Similarly, newborn babies may have high hyper-
opia, discovered incidentally. Infants have not yet
learned to accommodate accurately,

5

and dynamic
retinoscopy may therefore be unreliable. In the ab-
sence of strabismus, it is reasonable to defer assess-
ment of accommodation and prescription of glasses
until 46 months of age, depending on severity.

CASE II

A 4-month-old girl presents with an extensive anterior
polar cataract in her right eye. There is no view of the poste-
rior pole prior to dilation, but a good, clear view after dila-
tion. Intraocular surgery can probably be avoided if chronic
dilation is tolerated. How frequently should the dilating
drops be instilled to avoid inducing refractive amblyopia
while still adequately dilating the pupil?

In this case, chronic pupillary dilation is desired to
allow a formed image to reach the retina. A noncy-
cloplegic dilating drop is unlikely to keep the pupil
DYNAMIC RETINOSCOPY

273

dilated for long enough to be useful. However, a
cycloplegic agent will paralyze accommodation,
thereby optically penalizing an eye that is already
amblyopic. Bifocals could compensate for the cyclo-
plegia, but they are difficult to manage effectively in
such a young patient. One treatment approach in
this case would be to administer a dose of the cyclo-
plegic agent, then follow the patient with dynamic
retinoscopy every 12 days for 1 week to determine
1) when the cycloplegic agent had worn off suffi-
ciently to allow good near vision, and 2) when the di-
lating effect had worn off sufficiently to again ob-
scure the entrance pupil. Drops could then be
prescribed on a time interval that maximized the
number of days with dilation but without cyclople-
gia. For example, if one drop of atropine to the right
eye gave 12 days of total cycloplegia, 3 days of dila-
tion with only partial cycloplegia, and 2 days of full
accommodative ability but marginally dilated pupils,
the parents might be advised to instill one drop of
atropine to the right eye every 57 days, combined
with part-time occlusion of the left eye on days 26.

CASE III

A 5-year-old boy with hyperopic anisometropic amblyopia
OD is treated with glasses and atropine penalization of the left
eye. Cycloplegic refraction is OD

3.50 sphere, OS

0.50
sphere. Is a bifocal also necessary to help the amblyopic eye fo-
cus at near?

An amblyopic eye tends to accommodate poorly,
especially when the eye has significant hyperopia.
Thus the question, Is a bifocal necessary? fre-
quently arises when treating these patients. The use
of atropine penalization increases the relevance of
the question, since the amblyopic eye will be respon-
sible for fine near work. Dynamic retinoscopy with
best distance correction in place and a cycloplegic
agent instilled in only the penalized eye can answer
this. If the amblyopic eye does not accommodate re-
liably, a bifocal lens should be prescribed only for
the amblyopic eye until later testing reveals im-
proved accommodation in that eye.

CASE IV

A 16-month-old girl has 1.75 diopters of hyperopic aniso-
metropia. Is she beginning to develop amblyopia?

Although accommodative effort is generally sym-
metric, we do not know in this case whether the
child is always accommodating with the less hyper-
opic eye (and thus probably already amblyopic or at
high risk), or whether she is spontaneously accom-
modating with the more hyperopic eye at times (and
thus at lower risk for amblyopia). Dynamic retinos-
copy can be performed with particular attention
paid to the more hyperopic eye. If the eye cannot ac-
commodate even under monocular conditions, it is
probably amblyopic. If it is consistently hypoaccom-
modating, then the anisometropic correction re-
quired to restore symmetry should be prescribed,
and prophylactic, part-time occlusion or penaliza-
tion should be considered (especially if glasses are
refused). Alternatively, if the eye shows evidence of
full accommodation to a near stimulus when view-
ing under binocular conditions, observation alone is
acceptable.

CASE V

A 6-year-old girl complains of tiring after reading for a
few minutes despite near acuity of J1

i

n each eye, normal
convergence, and a cycloplegic refraction of

1.50 OU. Are
her complaints fictitious?

Measurement of normal J1

near acuity alone in
this patient does not fully characterize her accommo-
dative status. Dynamic retinoscopy might reveal rapid,
complete, but unsteady or discontinuous accommo-
dation in this low hyperope, confirming the diagnosis
of accommodative insufficiency and reinforcing the
decision to treat with glasses. Alternatively, a briskly
normal dynamic retinoscopy response is consistent
with a nonocular etiology to her complaints. If symp-
toms persisted in the latter case, a trial of a low-power
reading add may be indicated in case of a falsely nor-
mal response to dynamic retinoscopy.

CASE VI

A 5-year-old boy presents for follow-up of high hyperopia,
accommodative esotropia, and amblyopia. Three siblings,
aged 4 months, 16 months, and 3 years, accompany him.
Mother wants the other children evaluated formally for am-
blyopia on a future visit, but the family has no insurance
and must pay out-of-pocket for every exam. Which child
should be examined first?

Dynamic retinoscopy can be used as a quick
screening tool for potentially amblyopiagenic refrac-
tive error. By using an accommodative target and
quickly switching back-and-forth to view the reflexes
in both eyes, it is possible to detect refractive asym-
metry or high hyperopia with poor accommodation.
Significant astigmatism can be detected with or with-
out use of an accommodative target. Esotropia in-
duced by accommodation can also be detected. In
this case, all three siblings can be screened in sec-
onds, and any child with abnormal findings should
be examined promptly.

CASE VII

A 7-year-old boy with Down syndrome has a cycloplegic
refraction of

0.75 OU. There is no strabismus, and he is
asymptomatic. Are bifocals indicated

?
Patients with Down syndrome have poor ability to
sustain accommodation.

1,4

Patients with other devel-
opmental delays may be similarly affected.

8

There is
274 Surv Ophthalmol 46 (3) NovemberDecember 2001

HUNTER

a high likelihood that this child will need bifocals,
even though the hyperopia is minimal. Dynamic reti-
noscopy, performed prior to cycloplegia or on a fol-
low-up visit, can help determine whether this patient
will benefit from a bifocal add. We have found that
patients with Down syndrome often accept glasses
much more readily when bifocal adds are included.

CASE VIII

A 25-year-old woman complains of asthenopia and
headaches that have persisted despite full recovery from a
job-related hyphema 6 months previously. Near acuity is
J1

OU. Is she malingering?

Ocular trauma can impair accommodation. Often
but not always heralded by pupillary abnormalities,
the problem can sometimes be confirmed by formal
measurement of accommodative amplitudes, but it
might be missed if accommodation is complete but
slow or unsteady in the traumatized eye. More rapid
and accurate confirmation can be achieved with dy-
namic retinoscopy, with which the trauma-induced
subnormal accommodation of the right eye can be
detected by rapidly comparing the state of accom-
modation in the two eyes.
In this patient, if unilateral accommodative insuf-
ficiency is detected, the patient may benefit from a
monocular bifocal add. If dynamic retinoscopy is
normal, a nonocular etiology to the symptoms
should be sought.

Summary

With dynamic retinoscopy, neutralization of the
retinoscopic reflex can be detected bilaterally when
the patient accommodates on a target adjacent to
the retinoscope. This rapidly performed clinical tool
can provide critical data that can help solve diagnos-
tic or therapeutic dilemmas involving high hyper-
opia or possible accommodative insufficiency.

Method of Literature Search

The MEDLINE ([Link]/entrez/
[Link]) database was searched using the
dates 1966 to the present using the phrase

dynamic
retinoscopy

. No limitations on language were applied.
In March 2001, the search yielded 29 articles. Addi-
tional relevant articles were identified from the ref-
erences cited in the articles found with the MED-
LINE search. The Association for Research in Vision
and Ophthalmology Abstract Search and Program
Planner was also searched. Articles and abstracts
considered of highest relevance to the topic were in-
cluded in this article.

References

1. Cregg M, Woodhouse JM, Pakeman VH, et al: Accommoda-
tion and refractive error in children with Down syndrome:
cross-sectional and longitudinal studies. Invest Ophthalmol
Vis Sci 42:5563, 2001
2. del Pilar Cacho M, Garcia-Munoz A, Garcia-Bernabeu JR, Lo-
pez A: Comparison between MEM and Nott dynamic retinos-
copy. Optom Vis Sci 76:6505, 1999
3. Guyton DL, OConnor GM: Dynamic retinoscopy. Curr Opin
Ophthalmol 2:7880, 1991
4. Haugen OH, Hovding G: Strabismus and binocular function
in children with Down syndrome. A population-based, longi-
tudinal study. Acta Ophthalmol Scand 79:1339, 2001
5. Haynes H, White BL, Held R: Visual accommodation in hu-
man infants. Science 148:52830, 1965
6. Jackson E: Skiascopy and its practical application to the study of
refraction. Philadelphia, Edwards and Docker, 1895, pp. 868
7. Jackson TW, Goss DA: Variation and correlation of clinical
tests of accommodative function in a sample of school-age
children. J Am Optom Assoc 62:85766, 1991
8. Leat SJ: Reduced accommodation in children with cerebral
palsy. Ophthalmic Physiol Opt 16:38590, 1996
9. Mohindra I, Held R, Gwiazda J, Brill J: Astigmatism in in-
fants. Science 202:32931, 1978
10. Rosenfield M, Portello JK, Blustein GH, Jang C: Comparison
of clinical techniques to assess the near accommodative re-
sponse. Optom Vis Sci 73:3828, 1996
11. Tondel G, Horner D, Bradley A: Infant accommodation dy-
namics measured by eccentric videorefraction [Abstract]. In-
vest Ophthalmol Vis Sci 42:2075, 2001

Supported by grants from Research to Prevent Blindness (Lew
R. Wasserman Merit Award), The Roy and Niuta Titus Founda-
tion, and The Helena Rubinstein Foundation. The author has no
proprietary or commercial interest in any product or concept dis-
cussed in this article.
Reprint address: David G. Hunter, MD, PhD, Wilmer 233, The
Johns Hopkins Hospital, 600 N. Wolfe St., Baltimore, MD 21287-
9028.