The effect of neutral density filters on testing in patients with

strabismic amblyopia
Syed Yaser Habeeb, MD, Brian W. Arthur, MD, Martin W. ten Hove, MD, MEng
ABSTRACT RSUM
Objective: To confirm if using a neutral density filter (NDF) affects eyes with strabismic amblyopia differently compared to fellow
non-amblyopic eyes, and to determine if a similar effect could be observed when using a NDF during peripheral visual field
testing.
Design: Prospective controlled case series.
Participants: 19 subjects with strabismic amblyopia with visual acuities between 20/400 and 20/40 in their affected eyes were recruited
to the study. Fellow non-amblyopic eyes served as the control group.
Methods: Visual acuity in both eyes was assessed using a projected Snellen eye chart with two NDFs (0.4 and 3.0 densities). Visual fields
were assessed using a Humphrey perimeter using one NDF (0.4 density). Best corrected visual acuity and visual fields were also recorded.
Results: When using a 3.0 NDF, visual acuity was reduced in all eyes (p 0.0001). When using a 0.4 NDF, visual acuity was
significantly improved in eyes with strabismic amblyopia compared to unfiltered conditions (p 0.0011). There was no significant
effect by NDFs on visual field testing in eyes with strabismic amblyopia or fellow non-amblyopic eyes.
Conclusions: Neutral density filters affect eyes with strabismic amblyopia differently than they do non-amblyopic eyes. A significant
improvement in visual acuity of eyes with strabismic amblyopia was observed when using a 0.4 NDF compared to non-amblyopic
eyes. Visual acuity was reduced in amblyopic as well as non-amblyopic eyes when viewing through a 3.0 NDF. No significant change
in visual fields was observed when using a 0.4 NDF in amblyopic or non-amblyopic eyes.
Objet : Confirmer si lutilisation dun filtre densit neutre (FDN) affecte diffremment les yeux atteints dune amblyopie due au
strabisme comparativement lautre il non amblyope, et dterminer si lon peut observer un effet similaire en utilisant un FDN pour
le test du champ visuel priphrique.
Nature : Srie de cas prospectifs contrls.
Participants : 19 sujets atteints damblyopie due au strabisme, ayant des acuits visuelles variant entre 20/400 et 20/40 dans leurs yeux
affects, ont t retenus pour ltude. Les autres yeux non amblyopes ont servi de groupe tmoin.
Mthodes : Lacuit visuelle des deux yeux a t value par projection de la charte oculaire Snellen avec deux FDN (densits de 0,4
et 3,0). Les champs visuels ont t valus avec un primtre Humphrey utilisant un FDN (densit 0,4). La meilleure acuit visuelle
corrige et les champs visuels ont aussi t nots.
Rsultats : Avec le FDN 3,0, lacuit visuelle tait rduite dans tous les yeux (p 0,0001). Avec le FDN 0,4, lacuit visuelle tait
amliore significativement dans les yeux atteints damblyopie due au strabisme comparativement aux tats non filtrs (p 0,0011).
Les FDN navaient pas deffet significatif sur les tests de champ visuel dans les yeux ayant une amblyopie due au strabisme ou les
autres yeux non amblyopes.
Conclusion : Les filtres densit neutre affectent les yeux atteints damblyopie due au strabisme diffremment des yeux non
amblyopes. Une amlioration significative de lacuit visuelle des yeux atteints damblyopie due au strabisme a t observe lors de
lutilisation du FDN 0,4, comparativement aux yeux non amblyopes. Lacuit visuelle a t rduite chez les yeux amblyopes et non
amblyopes lorsquon regardait travers un FDN 3,0. Aucun changement significatif des champs visuels na t observ avec un FDN
0,4 dans les yeux amblyopes ou non amblyopes.

In 1959, von Noorden and Burian demonstrated that

when tested with a neutral density filter (NDF), visual acuities
in nonamblyopic eyes of individuals with strabismic amblyopia decreased, while visual acuities in the amblyopic eyes
stayed the same or even improved.1 However, when the NDF
was placed before an eye that was amblyopic due to an organic
lesion (such as macular degeneration, optic nerve atrophy, or
chorioretinitis), there was an almost grotesque drop in visual acuity. A potential application of this observation was
that one could use an NDF to differentiate between strabismic and organic amblyopia as a quick screening tool before
proceeding with occlusion therapy, especially when amblyopia is of uncertain origin.2 Von Noorden and Burian posited
that their results confirmed an amblyopic eye is not at its best

under photopic conditions, but shows a relative improvement of its functions under mesopic conditions.1
To our knowledge, the 1959 study has never been replicated, even though the findings are now well accepted in
most textbooks concerning pediatric and neuro-ophthalmology. Moreover, it was unknown whether using an
NDF would exert a similar effect during visual field testing
under mesopic conditions. Thus, the objectives of the present study were twofold: to confirm the effect of an NDF on
the central visual acuity of an eye with strabismic amblyopia and to determine whether a similar effect (improvement as assessed by the mean defect) could be observed
when measuring peripheral visual fields of eyes with strabismic amblyopia viewed through an NDF.

Preliminary results of this study were presented at the Canadian Ophthalmological Society annual meeting in Qubec City, QC, June 2010.
From the Department of Ophthalmology, Queens University, Kingston,
Ont.

Correspondence to Martin W. ten Hove, Department of Ophthalmology,

166 Brock St., Kingston ON K7L 5G2; tenhove@queensu.ca

Originally received Jan. 26, 2012. Final revision Jan. 26, 2012. Accepted
Mar. 26, 2012

348

CAN J OPHTHALMOLVOL. 47, NO. 4, AUGUST 2012

Can J Ophthalmol 2012;47:348 350

0008-4182/11/$-see front matter 2012 Canadian Ophthalmological Society.

Published by Elsevier Inc. All rights reserved.
dx.doi.org/10.1016/j.jcjo.2012.03.039

Effect of neutral density filters on testingHabeeb et al.

METHODS
We recruited, from a pediatric ophthalmology practice,
19 subjects with strabismic amblyopia whose visual acuities were between 20/300 and 20/40 in their affected eyes.
Strabismic amblyopia was defined as reduced vision in 1
eye that had no identified cause other than ocular misalignment. The subjects nonamblyopic eyes served as the control group. The youngest subject was 9 years old and the
oldest was 74 years old. Patients with organic lesions or
media opacities in either eye, neurologic diseases, ocular or
neurologic pathologies affecting the vision of the nonamblyopic eye, or attention deficit disorders were not eligible
for the study.
Best-corrected Snellen visual acuity in both eyes was first
assessed without any filters, then with a 0.4 NDF, and
finally with a 3.0 NDF. Kodak (Rochester, NY) Wratten
No. 96 gelatin filter (the same make and strength [3.0
NDF] as used in the original 1959 study). An NDF decreases transmission of light (luminance) through it
equally at all wavelengths, without affecting color perception. The optical density of a filter is quantified in log
units, which indicate how much light is transmitted
through the filter (transmission 10OD 100). Thus, a
0.4 NDF has a light transmittance of about 40%, while a
3.0 NDF has a 0.1% light transmittance.
Study subjects were allowed 2 minutes initially to adapt
to the reduced illumination of the darkened room. In addition, as per the protocol for the Neutral Density Filter
Test outlined in von Noordens Atlas of Strabismus, the
NDF was held for 1 minute in front of each eye before
visual acuity was determined.2 Two different Snellen
charts were used with each subject to avoid memorization
bias. Visual acuity was converted to logMAR scores and
recorded as the primary outcome. For the second part of
the study, visual fields were assessed in both eyes using a
Humphrey field analyzer (Carl Zeiss Meditec, Jena, Germany), perimeter (visual field assessment 24-2), first without any filter then, following a 5-minute break, with a 0.4
NDF. Here the mean defect was recorded as the primary
outcome.
The data were analyzed to answer 2 questions: (i) Is there a
significant difference in the effect of an NDF on visual acuity

in nonamblyopic eyes versus eyes with strabismic amblyopia?;

and (ii) Is there a significant difference in the effect of an NDF
on the mean defect of visual fields in nonamblyopic eyes
versus eyes with strabismic amblyopia?
Statistical analysis was conducted using SAS software
(SAS Institute, Cary, NC) using the mixed models procedure, assuming compound symmetry structure in the covariance matrix. This approach is similar to a repeated
measures regression but with the advantage of allowing for
pairwise comparisons at the same time. A p value of less
than 0.05 was considered statistically significant.
All data were reviewed by 2 board-certified ophthalmologists and a statistician. The Queens University Health
Sciences and Affiliated Teaching Hospitals Research Ethics Board approved the study for ethical compliance.

RESULTS
A repeated measures ANOVA test showed that the interaction effect between the NDFs and eyes with strabismic amblyopia versus the NDFs and nonamblyopic eyes
was significantly different (p 0.0001). There was no
significant change in visual acuity of nonamblyopic eyes
using a 0.4 NDF versus unfiltered conditions (p
0.3570), but a significant decrease in visual acuity of
nonamblyopic eyes was noted when comparing a 3.0 NDF
with unfiltered conditions (p 0.0001).
There was also a significant improvement in the visual
acuity of eyes with strabismic amblyopia when comparing
a 0.4 NDF to unfiltered conditions (p 0.0011), but a
significant decrease in the visual acuity of eyes with strabismic amblyopia when comparing a 3.0 filter to unfiltered
conditions (p 0.0001). The study thus confirmed that
eyes with strabismic amblyopia and nonamblyopic eyes are
affected differently by NDF filters when measured by central visual acuity. Figure 1 and Table 1 summarize these
results.
A repeated measures ANOVA test showed that there
was no significant effect on the mean defect of visual fields
of either amblyopic or nonamblyopic eyes when comparing a 0.4 NDF to unfiltered conditions (p 0.7409).

1.2
1
0.8
0.6

Fellow eyes

0.4

Amblyopic eyes

0.2
0
No Filter

NDF 0.4

NDF 3.0

Fig. 1Comparing least squares means of amblyopic and fellow nonamblyopic eyes with and without neutral density filters (NDF).
An increase in logMAR represents a decrease in visual acuity.
CAN J OPHTHALMOLVOL. 47, NO. 4, AUGUST 2012

349

Effect of neutral density filters on testingHabeeb et al.

Table 1Pairwise comparisons of least square means for filter
effects broken down by amblyopic and fellow
nonamblyopic eyes (n 19)
Pairwise Comparisons

Change in Mean
logMAR

Effect
on VA

p Value

0.0302
0.6134

2
2

0.3570
0.0001

0.1110
0.2345

1
2

0.0011
0.0001

Fellow nonamblyopic eyes

Unfiltered vs 0.4 NDF
Unfiltered vs 3.0 NDF
Amblyopic eyes
Unfiltered vs 0.4 NDF
Unfiltered vs 3.0 NDF
NDF, neutral density filter; VA, visual acuity.

DISCUSSION
The original study by von Noorden and Burian showed
an improvement in visual acuity of eyes with strabismic
amblyopia when using a 3.0 NDF. In contrast, none of the
subjects in the present study reported any improvement in
visual acuity of the eye with strabismic amblyopia when
using a 3.0 NDF. In fact, using a 3.0 NDF resulted in a
significant decrease in visual acuity of nonamblyopic eyes
as well as amblyopic eyes (p 0.0001). One can only
speculate that we were unable to replicate the results of the
1959 study because of an undisclosed variation in experimental protocol or materials used by von Noorden and
Burian. However, note there was a larger decrease in visual
acuity in nonamblyopic eyes with a 3.0 NDF than in amblyopic eyes with a 3.0 NDF.
In the Atlas of Strabismus, von Noorden suggests using a
0.5 NDF or a 2.0 NDF to distinguish between strabismic
and organic amblyopic.2 It may be that these optical densities are more amenable to obtaining improvements in
visual acuity of amblyopic eyes than is a 3.0 NDF. Differences in optical densities notwithstanding, statistical analysis
showed that the interaction effect between NDFs and eyes
with strabismic amblyopia versus NDFs and nonamblyopic
eyes was significantly different. Moreover, a significant improvement in visual acuity of amblyopic eyes (p 0.05) was
noted when using a 0.4 NDF. This finding is consistent with
several studies, which show a decrease in luminance may improve visual acuity of eyes with strabismic amblyopia.3-7
The function of the amblyopic eye closely resembles
that of a nonamblyopic eye under mesopic conditions, because of the reduced role of cones due to central suppression.3 Eyes with strabismic amblyopia have also been
shown to have improved fixation patterns under conditions of reduced illumination.3,8 In addition, monocular
vision through the amblyopic eye results in improved visual acuity, given that stimulation of the nonamblyopic eye
decreases visual acuity in the amblyopic eye.9 A combination of these 3 factors may thus contribute to the observed
improvement in visual acuity of eyes with strabismic amblyopia when using an NDF.
We did not find any significant effect by NDFs on visual
fields, as quantified by the mean defect, in eyes with strabismic amblyopia. Our results are consistent with the

350

CAN J OPHTHALMOLVOL. 47, NO. 4, AUGUST 2012

study by Mansouri et al., which demonstrated that although the maps of the central visual fields in amblyopic
subjects are affected, this finding does not correlate with
measured visual acuity.10 Thomas showed that contrast sensitivity functions obtained from the foveal region of strabismic
amblyopes resembled those obtained from the peripheral retina of a nonamblyopic eye.11 Since the 24-2 mean defect
measures peripheral retinal function more than central retinal
function, it is not surprising that no difference was found in
the effect of an NDF on mean defect.
In summary, the present study confirms von Noorden
and Burians observation that neutral density filters affect
eyes with strabismic amblyopia differently than they do
fellow nonamblyopic eyes. A significant improvement in
visual acuity of eyes with strabismic amblyopia was observed when using a 0.4 NDF compared to nonamblyopic
eyes. In contrast to the 1959 study by von Noorden and
Burian, visual acuity was significantly decreased in nonamblyopic eyes, as well as in eyes with strabismic amblyopia,
when viewing through a 3.0 NDF. No significant change
in visual fields was observed when using a 0.4 NDF in
nonamblyopic or amblyopic eyes. Future studies should
focus on determining contrast sensitivities using NDFs of
multiple optical densities under varied spatial frequencies.
This would help determine thresholds, if any, at which
eyes with strabismic amblyopia begin to and cease to respond to NDFs of varying intensities. Results can then be
analyzed to determine the ideal NDF resulting in optimal
improvement of visual acuity.

Disclosure: The authors have no proprietary or commercial interest

in any materials discussed in this article.
REFERENCES
1. Von Noorden GK, Burian HM. Visual acuity in normal and amblyopic
patients under reduced illumination. I. Behavior of visual acuity with
and without neutral density filter. Arch Ophthalmol. 1959;61:533-5.
2. Von Noorden GK. Atlas of Strabismus, 4th ed. St. Louis: CV Mosby,
1983.
3. Von Noorden GK, Burian HM. Visual acuity in normal and amblyopic
patients under reduced illumination. II. The visual acuity at various
levels of illumination. Arch Ophthalmol. 1959;62:396-9.
4. Lawill T, Burian HM. Luminance, contrast function and visual acuity in
functional amblyopia. Am J Ophth. 1966;62:511-20.
5. Sloan LL. Variations in acuity with luminance in ocular disease and
anomalies. Doc Ophthalmol. 1969;26:384-93.
6. Hess RF, Howell ER. The luminance-dependent nature of the visual
abnormality in strabismic amblyopia. Vision Res. 1978;18:931-6.
7. Hess RF, Campbell FW, Zimmern R. Differences in the neural basis of
human amblyopias: The effect of mean luminance. Vision Res. 1980;20:
295-305.
8. Von Noorden GK, Burian HM. An Electro-ophthalmographic study of
the behavior of the fixation of amblyopic eyes in light- and dark-adapted
state: A preliminary report. Am. J. Ophthalmol. 1958;46:68-77.
9. Von Noorden GK, Leffler MB. Visual acuity in strabismic amblyopia
under monocular and binocular conditions. Arch Ophthalmol. 1966;
76:172-7.
10. Mansouri B, Hansen BC, Hess RF. Disrupted retinotopic maps in amblyopia. Invest Ophthalmol Vis Sci. 2009;50:3218-25.
11. Thomas J. Normal and amblyopic contrast sensitivity functions in central and peripheral retinas. Invest Ophthalmol Vis Sci. 1978;17:746-53.