Evaluation Of A New Direct-Comparison Aniseikonia
Test
View Shopping Cart
(Analysis based on Version 2 of the
Aniseikonia Software)
Gerard C de Wit PhD
The following technical paper to
appear in as follows: Evaluation of a new direct-comparison Aniseikonia test, Bin. Vis. & Strabismus Q. 18: 87-94 (2003)
Note: Tables and Figures may come
up slowly due to file sizes
ABSTRACT
Background & Purpose: Aniseikonia
is a condition in which the two eyes perceive images of different size or shape,
causing a variety of
visual symptoms including asthenopia. Besides anisometropes (with a prevalence of 5-10% in the
population above age 20 years), also pseudophakes and refractive surgery patents are at risk. For example,
40% of the pseudophakes seem to suffer from Aniseikonia. Reliable measurement and management of
Aniseikonia is therefore important. The "Aniseikonia Inspector" is a new, commercially available,
software product to measure an manage Aniseikonia. The purpose of this study is to evaluate this
Aniseikonia test of the Aniseikonia Inspector.
Methods: Aniseikonia
was induced in four subjects, with normal vision, by means of afocal size
lenses. Using the Aniseikonia Inspector, the
resulting Aniseikonia was measured in vertical, horizontal and diagonal directions.
Results: The
average ratio between the measured Aniseikonia and the induced Aniseikonia was
0.98, 0.89 and 0.93 respectively for the
vertical, horizontal and diagonal directions. For two consecutive measurements. Of the same
aniseikonic state, the difference in measurement value was 97%, 75% and 94% of the time within one
resolution step size (0.5% horizontally and vertically, and 0.7% diagonally).
Conclusion: Aniseikonia
was measured accurately. Measurements in the vertical direction were more accurate that in
the diagonal or
horizontal directions, which is probably due to fixation disparities. The Aniseikonia Inspector is a very
useful new tool in treating the growing number of Aniseikonia patients.
Key Words: Aniseikonia
/ anisometropia / binocular vision / eikonometry / fixation disparity / test
* The author and publisher have a
commercial interest in The Aniseikonia Inspector used in this study.
INTRODUCTION
Incidence of Aniseikonia
After an initial burst of interest
in Aniseikonia in the 1950's, the condition seems to be
out of vogue today (1,2). In contradiction, the number
of patients suffering from Aniseikonia have increased
considerably. Not only anisometropes (with a
prevalence of 5-10% of the population above the age of 20 years
(3)) and other people with inherent optical or
anatomical differences between the two eyes are at risk,
but now also people who have had refractive surgery
(e.g. LASIK, PRK, etc.) and cataract surgery are at
risk (4,5). For example, Kramer et al. (5) found
that 40% of all pseudophakes have ophthalmic
complaints referable to Aniseikonia. Nowadays, in the United
States alone, more than 1.5 million cataract
operations are performed each year, which means
that Aniseikonia can be considered a significant public
health issue (6,7).
Aniseikonia Management
The question remains why eye care
professionals pay relatively little attention to Aniseikonia today. This will be discussed after considering the
basic steps of Aniseikonia management
(Link to Figure 1).
Measure Aniseikonia
Verify if patient is helped
by correcting Aniseikonia
Determine an Aniseikonia corrected
prescription
The first step "Measure Aniseikonia" may seem trivial, but because of a deficiency of
accurate and good instrumentation, measuring Aniseikonia has been difficult. Because of this lack,
very few eye care professionals are able to even
diagnose or measure Aniseikonia at all. Others have
recommended one use certain "rules of thumb",
for estimating Aniseikonia. Research, however, shows that these
rules of thumb (including Knapp's law) have a
large error rate (8-10).
For certain groups of patients such
as pseudophakes and strabismus patients the error
rate of rules of thumb was as high as 60-70% (8). So there
actually was no test or method available to measure Aniseikonia, while it is important to use a reliable
measurement method to manage Aniseikonia.
Verifying if a patient is helped by
correcting the Aniseikonia (subjective eikonometry),
often will be the second step in managing Aniseikonia.
It gives the opportunity to refine the correction
to maximal patient satisfaction. The tolerance level to Aniseikonia also differs a lot from patient to
patient. Some patients can be really grateful when correcting
1% of Aniseikonia, while another might not be bothered
by as much as 3% Aniseikonia. A common way of
verification is to have the patient look through a focal
size lenses with similar magnification as the Aniseikonia found in the measurements. Another way is to
simulate and manipulate the Aniseikonia correction by haploscopic image projection on, for example, a
computer screen.
The last step of Aniseikonia management is the calculation of an aniseikonic
corrected prescription. By changing the shape factor (base
curve and lens thickness), refractive index, and/or
vertex distance of spectacle lenses, the difference in
magnification between the lenses can be modified
to correct for the Aniseikonia. To do this calculation
by hand can be quite a challenge; this is yet
another reason why eye care professionals can be hesitant
to deal with Aniseikonia. However, with the
advent of computers this problem can be easily solved.
A major reason, certainly, why,
currently, eye care professionals pay relatively little
attention to Aniseikonia is that they cannot
perform one (or more) of these basic steps of Aniseikonia management or they consider it is too time consuming or
unprofitable to undertake.
The Aniseikonia Inspector
Recently, a new Aniseikonia product
(Richmond Products p/n 4544) has become commercially available. This
software product performs and manages all three basic
steps of Aniseikonia management (Fig. 1). Therefore, this
product intends to make it possible
and practical to give Aniseikonia the attention it
deserves.
The Aniseikonia test of the
Aniseikonia Inspector is based on the simple technique of
direct comparison eikonometry (11). Although the
technique seems remarkably uncomplicated, it can
have a sensitivity of less than 0.5%. However, a good
layout of the test is important to avoid underestimation
of Aniseikonia. Ogle and Ames (11) mention that
fusion stimuli from objects or contours surrounding the
comparison targets should be eliminated. McCormack et
al. (7) found that another direct comparison Aniseikonia test (the New Aniseikonia Test) significantly
underestimated Aniseikonia. They speculated that
this underestimation is due to a sensory fusion response
resulting in a rescaling of the perceived image.
McCormack et al. also demonstrated a direct
comparison test on a computer screen that did not have a significant underestimation. The
Aniseikonia test of the Aniseikonia Inspector is such a test
on a computer screen and this paper evaluates the
accuracy of this test.
Subjects and Methods
Subjects
Four subjects, aged 28-34 years,
with normal visual acuity and a negative ocular and
vision history, participated in this study to
evaluate the Aniseikonia test of the Aniseikonia Inspector.
Aniseikonia was induced in these
subjects by positioning afocal size lenses (MultiLens,
Sweden) in front of the right eye in random
order. The thickness of the size lenses could have given
the subject a slight clue about the magnification of the
lens; however, the Aniseikonia test did not reveal what Aniseikonia setting was being made, so it is unlikely
that this would have given a bias.
Table
1 shows the curvatures (obtained from the manufacturer) and thicknesses
(measured) of these 4 lenses. Each lens was used both to
magnify and to minify the OD image (by flipping the
lens). The magnification values of each lens was determined by means of optical ray-tracing using
the testing dimensions during the measurements
in this study: 75 cm for the image distance, 16 cm
image size, and 25 mm for the vertex distance. Both the
magnification related to static Aniseikonia (classical Aniseikonia, i.e. difference in retinal image size
with a fixed gaze direction) was calculated as well as
the magnification related to dynamic Aniseikonia (optical anisophoria, i.e. unequal phorias in the various directions of gaze also
due to anisometropic spectacles)
(12-14). This last type of Aniseikonia is probably more appropriate for the Aniseikonia
test of the
Aniseikonia Inspector, because the subjects' gaze
direction was not fixed during the test.
Aniseikonia Test
Figure
2 shows a (gray-scaled) image of the Aniseikonia test of the Aniseikonia
Inspector. The patient looks at this screen through
red-green glasses to separate and isolate binocular
vision. In this study, we used red in front of OD and the
dimensions were the same as those stated for the
determination of the magnification values of each size
lens. An Aniseikonia setting was made by changing the
size of one of the two half circles with keys on the
keyboard or with the mouse until the two half circles
appear equal in size. The Aniseikonia is measured in the
vertical, the horizontal, and the diagonal
direction. In each direction, the measurement is done
twice, once starting with –25% preset Aniseikonia and once starting with +25% preset
Aniseikonia. The average
of these two measurements is taken as the Aniseikonia value. To verify if the Aniseikonia
Inspector measured Aniseikonia correctly, the measured Aniseikonia are plotted against the (dynamic)
magnifications induced by the size lenses. The ideal test
result would be one where a magnification (M) results in
a measured Aniseikonia (A) of: Formula
Around M=0% the aniseikonic function A(M) is predominantly a linear function with
a slope close to –
1. Therefore, a linear regression
analysis is performed on the measured data and the slope
is compared to –1.0 (or –0.995: the slope obtained
when doing a linear regression analysis on Eq. (1) with,
as data points, the same magnifications as in the
experiments). In other words, the ideal Aniseikonia test
should measure –1% of Aniseikonia for each percent of
induced magnification.
RESULTS
Figure
3 shows the measured Aniseikonia as a function of induced magnification for the
vertical, the horizontal, and the diagonal
directions. The slope and R2 value
of the linear regression lines of the data of all subjects is shown in Table
2 .
In the vertical direction the
average slope of the best fit linear regression lines is –0.98,
very close to –1.0.
This suggests that the induced Aniseikonia is measured correctly (i.e. accurately). In the horizontal direction there
may be a minor underestimation (average slope =
-0.89). All of the subjects reported that inequality in
this direction was more difficult to assess than the
vertical direction, because the images were less stable.
This also shows in the R2 values. In the diagonal direction the slope is closer to –1.0, but still there
seems to be a slight underestimation of the Aniseikonia.
Table
3 shows the Y-intercept values of the linear regression lines for the different
subjects, which should represent the inherent Aniseikonia of the subjects. Note that the ideal
Aniseikonia curve,
given by Equation 1, is not a linear function and a
regression line through the ideal data points would have given a
bias of 0.25%. This bias has been subtracted from
the actual Y-intercept, giving the presented 'corrected
Y-intercept'.
The actual Aniseikonia measurements
(i.e. no size lens present) of the subjects is also
shown. This data shows that the subjects did not have a
clinically significant amount (i.e. >1%) (15) of
inherent Aniseikonia. To evaluate if the starting
Aniseikonia value of a measurement had any effect on the
final Aniseikonia setting, Figure
4 shows the number of occurrences of the difference between the first and
second measurement of all 108 trials
(4 subjects × 9 induced Aniseikonia values × 3
directions). The first measurement (M1) of each trial
started at an Aniseikonia value of –25%, while
the second measurement (M2) started at +25%. If
the starting Aniseikonia value would have given a
bias to the measurement, the distribution plots
in Figure 4 would not have been symmetric around M1-M2
= 0%. However, on average the difference
between the two measurements is –0.08%, 0.01%, and
–0.18% respectively for the vertical, horizontal, and diagonal direction. These values are all
close to 0% and small compared to the resolution step size
of 0.5% in the vertical and horizontal direction
and 0.7% in the diagonal direction.
Based on the spread in values, Figure
4 also shows that the repeatability for two
consecutive measurements (even thought the starting value was
different) was very good. The percentage of trials
in which two consecutive measurements were within
one resolution step size from each other is 97%,
75%, and 94% for the vertical, horizontal, and diagonal direction.
DISCUSSION
The Aniseikonia measured with the
Aniseikonia Inspector showed a very good
correspondence with the induced Aniseikonia in the subjects.
In the vertical direction, the slope of the linear
regression line for the 4 subjects ranged from –0.95 to
–1.02, so all very close to the ideal –1.0. In comparison, McCormack et al. (7)
found an average slope of -0.32 for the
New Aniseikonia Test, indicating a significant
underestimation. They speculated that the reason for the underestimation was the visibility of fusion stimuli
from objects, texture, or contours surrounding the comparison
targets. One inherent advantage of a direct
comparison test on a computer screen is that the test
images are themselves often brighter than the
surroundings, while a test on paper requires lighting that will
also illuminate and cast shadows on the surroundings.
In the horizontal direction the
slope of the linear regression line for the Aniseikonia
Inspector data varied more (-0.81 to –1.02) than
in the vertical direction. Also the repeatability
was worse than in the vertical direction. A likely reason
is that fixation disparities and phorias cause the
images to be less stable (11). With the Aniseikonia
Inspector it is possible to correct for these
effects by displacing the two half circles relative to each
other. This way the two half circles can be positioned
symmetrically on top of each other for a more accurate
size comparison. However, some remaining vergence
interference may still cause instability in the
images resulting in less reliable measurements than in the
vertical direction. Ogle et al. (16) reported that the
reliability of the horizontal direction depends on the
viewing distance. At distant vision it is more
reliable than at near vision. Ogle (11) also reported that
comparison targets in the oblique meridians have proven
unreliable because of the disturbing effect of fixation
disparity. However, the subjects in this investigation
all found the diagonal direction clearly easier to do than
the horizontal direction. This is also visible in
the accuracy and repeatability of the measurement
data. The explanation for this incongruity might be that
with the Aniseikonia Inspector it is possibility to move
the half circles relative to each other to compensate
for fixation disparity.
Since, in some patients, the
measurement in the vertical direction might be more accurate
than the measurement in the other directions, the
Aniseikonia Inspector also provides an aniseikonic ellipse
(i.e. complete description of the Aniseikonia (2,11)) based only on the measurement in the vertical
direction plus the prescription of the spectacle glasses used during the test. The idea behind this is that,
in practice, the meridional effects are nearly always
produced by spectacle cylinders (12). If this is
assumed, only one measurement direction is needed to
calculate the overall Aniseikonia part of the
aniseikonic ellipse, which is most important part of the Aniseikonia for clinical purposes (12).
CONCLUSION
The data in this paper indicates
that the Aniseikonia test of the Aniseikonia Inspector
measures Aniseikonia correctly and accurately, especially
in the vertical direction. In the horizontal and
diagonal direction, there may be a minor underestimation
of the Aniseikonia. Because the Aniseikonia
Inspector provides the means not only to
easily measure Aniseikonia, but also to carry out
the two other basic steps of Aniseikonia management
(verification and prescription calculation), it
presents and provides a very useful new tool in treating the
growing number of Aniseikonia patients.
REFERENCES
-
Romano PE. Aniseikonia? YECH! Binocul
Vis Strabismus Q
1999; 14: 173-176.
-
Bennett AG, Rabbetts RB. Clinical
Visual Optics. 3rd ed.
-
Butterworth-Heinemann Ltd, Oxford
1998; 273
-
Weale RA. On the age-related
prevalence of anisometropia.
Ophthalmic Res. 2002;
34:389-392
-
Lubkin V. Aniseikonia at the
Millenium. Binocul Vis, Strabismus Q 1999;
14: 179-182
-
Kramer PW, Lubkin V, Pavlica M,
Covin R. Symptomatic Aniseikonia in Unilateral and
Bilateral Pseudophakia. A Projection Space Eikonometer Study.
Binocul
Vis Strabismus Q 1999;
14: 183-190
-
Foster A. Vision 2020: The
cataract challenge. J. of Community Eye Health 2000;
34: 1
(http://www.jceh.co.uk/journal/34_1.asp)
-
McCormack G, Peli E, Stone P.
Differences in Tests of Aniseikonia. Invest
Ophthalmol Vis Sci 1992; 33:2063-2067
-
Lubkin V, Shippman S, Bennett G.
et. al. Aniseikonia Quantification: Error Rate of Rule
of Thumb Estimation.
Binocul Vis Strabismus Q 1999;
14: 191-196
-
Kramer P. Shippman S. Bennett G.
et. al. A study of Aniseikonia and Knapp's Law Using
a Projection Space Eikonometer. Binocul
Vis Strabismus Q 1999; 14: 197-201
-
Romano PE, Von Noorden GK. Knapp's
Law and Unilateral Axial High Myopia. Binocul
Vis Strab. Q 1999; 14: 215-222
-
Ogle KN. Researches
in Binocular Vision. W.B. Saunders Company, Philadelphia 1950
-
Remole A, Robertson KM. Aniseikonia
and Anisophoria: current concepts and clinical
applications. Runestone Publishing, Waterloo, Canada, 1996
-
Friedenwald JS. Diagnosis and
treatment of anisophoria. Arch. Ophth. 1936;
15: 283-307
-
Scheiman M, Wick B. Clinical
Management of Binocular Vision: Heterophoric, Accommodative,
and Eye Movement Disorders,
JB Lippencott Co. Philadelphia, 1994
-
Michaels DD, Visual
Optics and Refraction: A Clinical Approach, M
Mosby, St. Louis, 1985
-
Ogle KN, Imus HA, Madigan LF,
et. al. Repeatability of Ophthalmo-eikonometer
Measurements. Arch. Ophth. 1940; 24: 1179-1189
Aniseikonia Evaluation Software
Aniseikonia And Cyclodeviation Index
All Products In Alphabetical Order
Richmond Products Home Page View Shopping Cart