|
|
||||||||
1From the Department of Psychology, Concordia University, the 2Lady Davis Institute for Medical Research, and the 3Department of Ophthalmology, Sir Mortimer B. Davis Jewish General Hospital, Montreal, Quebec, Canada.
| Abstract |
|---|
|
|
|---|
METHOD. Visual acuities of 23 patients with MH (age range, 5282) were tested, using standard ETDRS and Landolt-C charts. Optical coherence tomography was used to confirm the diagnosis of MH.
RESULTS. Acuities correlated strongly, before and after surgery (r = 0.92 and r = 0.95, respectively). However, paired t-tests determined that resolution acuity was significantly more impaired at both time points than was recognition acuity (P < 0.001). Using Bland-Altman plots, the limits of agreement between the two acuity types indicated that resolution acuity differed from recognition acuity by up to five lines before surgery and up to 3 lines after surgery.
CONCLUSIONS. ETDRS and Landolt-C acuities differ in a clinically significant way in patients before and after MH surgery. Measuring recognition acuity by reading letters may lead to an overestimate of visual ability at the retinal level in patients with MH by including compensatory top-down cognitive processes that are unavailable for resolution tasks.
The Snellen chart has become the most frequently used tool for recognition acuity in the applied setting. Detailed evaluation of this measure has revealed several flaws in its design, resulting in the constant development of improved eye charts,1 2 3 4 5 including the ETDRS chart, which has become the scientific measure of choice. Resolution tasks are used when examining children, patients who are unfamiliar with the Latin alphabet, or those who are unable to communicate verbally. Here, the optotypes generally consist of one repeating symbol, where only the orientation of the target varies. The most commonly used resolution acuity charts are the Landolt-C and the Tumbling-E. Generally, recognition and resolution acuities are considered equivalent in their ability to assess visual function in normal observers, as differences were small and deemed clinically insignificant (Raasch TW, et al. IOVS 1984;25:ARVO Abstract 87).1 6 It has been pointed out that the letter C, which is also contained in the ETDRS chart and is the sole symbol of the Landolt-C chart, is more difficult to identify than other letter optotypes.1 5 7 Although the letters T and Z have been considered easier to recognize than a Landolt-C, the letter R has been found to be more difficult.8 Furthermore, letter confusion has been particularly problematic among letters that contain curved components, such as C, S, D, and O.9 A scaling factor of 0.95 for letter optotypes has been suggested to obtain the same acuity as with Landolt rings.8 Whether these two types of acuities are comparable in the presence of visual disease has only been investigated in patients with amblyopia10 and remains to be fully evaluated in other diseases.
One form of disease in which surgical intervention is evaluated with the use of eye charts is macular hole (MH), an age-related condition that affects the central area of the retina (macula). In addition to declining acuity, this condition creates distortions and/or blind spots (scotomas) in the central visual field. Because of the central scotoma created by the presence of an MH, patients are forced to fixate eccentrically. Therefore, a decrease in acuity is expected, as both recognition and resolution acuity have been demonstrated to decline as a function of retinal eccentricity.11 12 13 14 Previous work with a scanning laser ophthalmoscope (SLO) has indicated that patients fixate at or near the edge of the MH before surgery and that fixation returns toward the central area of the macula after successful surgical closure of the defect.15 16 Nakabayashi et al.15 used the amount of fixation shift and its associated improvement of visual acuity to define functionally successful MH closure, in addition to the established standards for successful MH surgery.17
In the present study, we investigated whether differences exist between acuity-chart types in patients with a diagnosis of MH. It was hypothesized that, even though recognition (ETDRS) and resolution (Landolt-C) acuities would be highly correlated, resolution acuity would be more impaired, and this difference would be clinically significant in the presence of MH as well as after MH surgery.
| Methods |
|---|
|
|
|---|
Participants
The participants were 18 women and 5 men with a mean age of 71.0 ± 7.7 years (SD; range, 5282), currently being treated by one retinal surgeon. The sample consisted of 23 eyes of these 23 patients with a diagnosis of idiopathic MH in only one eye, with no concomitant retinal disease and a healthy second eye. MH diameter was measured with optical coherence tomography (OCT 3; Carl Zeiss Meditec Inc., Dublin, CA) and ranged from approximately 200 to 900 µm (
0.83.4° of visual angle). Information on duration of symptoms was not available. Eight eyes had clear intraocular lens implants at the time of diagnosis whereas, of the remaining 15 eyes, 2 (patients 15 and 22) showed clinically significant signs of cataract formation during the study. All patients were scheduled to undergo 25-gauge transconjunctival sutureless vitrectomy with gas tamponade between June 2004 and July 2005. Postsurgically, MH status was evaluated by OCT scan.
Apparatus and Procedure
Participants underwent refraction with an autorefractor (Autorefractor ARK-760A; Nidek, Gamagori, Japan), and then wore appropriate trial lenses for acuity measurement. Acuities were determined in the eye that was scheduled to receive surgery, whereas the other eye was occluded. Standard, retroilluminated ETDRS (Early Treatment Diabetic Retinopathy Study) and Landolt-C charts (Lighthouse International, New York, NY) were presented in that order at a distance of 1 or 2 m, to accommodate the acuity range of the patients. The Landolt-C chart contained rings with four orientations (right, left, up, down). Black optotypes were displayed at full contrast on a white background (luminance 185 cd/m2). The luminances fell within an acceptable range of established optimal parameters for eye-chart illumination.7 Participants were encouraged to identify optotypes one by one in each line while the experimenter scored correct responses. Testing stopped whenever a participant was unable to identify five consecutive optotypes in a line correctly. The same procedure was repeated between 1 and 3 months after MH surgery. For statistical analysis, acuities were expressed in logMAR (logarithm of the minimum angle of resolution) units.
| Results |
|---|
|
|
|---|
2 = 0.69, observed power = 1.0). Analysis of the postoperative data resulted in identical findings (t(22) = 6.66, P < 0.001,
2 = 0.67, observed power = 1.0).
|
|
| Discussion |
|---|
|
|
|---|
|
The results regarding heteroscedasticity of residuals in the present sample further indicate that a relationship exists between mean acuity on both charts and acuity differences between the charts. As acuity declined, differences between recognition and resolution acuities increased, both before surgery and when surgery failed. This discrepancy was reduced in patients with successful closure of the MH.
Resolution acuity was significantly poorer than recognition acuity in the present sample. Theoretically, by simply guessing the orientation of a Landolt-C (chance of one in four or one in eight, depending on the chart), resolution acuity should be better than letter acuity (chance of 1 in 10 or 1 in 26, from the patients perspective). In fact, the opposite is the case. In an attempt to explain why letter and resolution acuities differ so drastically in a clinical population, the characteristics of the optotypes must be considered. It has been established that the type of letter used in a chart has great impact on the legibility and difficulty of the task.1 The use of letters cognitively facilitates the acuity assessment. In addition, the identification of a letter is a more complex task than identifying the orientation of a ring.7 Under conditions of impairment, it may be this complexity that helps an observer to reconstruct the target cognitively. Components of a letter, such as diagonals or curves, facilitate its accurate identification, even if the entire target is not resolved. Conversely, resolving the gap in a ring is not facilitated by such component cues and becomes more difficult, thereby explaining why resolution acuities were poorer.
The difference between the two acuity types, before as well as after surgery may, in part, be explained by a change in fixation. Before surgery, monocular fixation is most likely to occur at or near the MH edge.14 15 As all participants in the present study had a healthy second eye, the affected eye never had to develop a new and stable preferred retinal locus. It has previously been reported that the better eye generally dominates tasks that require fixation,22 which explains part of the poor monocular performance. If surgery fails to close the retinal defect (patients 2, 6, 19, and 20) but flattens and reattaches the MH edges, acuity may still improve but fixation remains peripheral and nonoptimal. In patients in whom MH closure is accomplished, monocular fixation returns toward the center of the fovea, and acuity improves as the scotoma becomes smaller or disappears. This improvement may lead to more stable monocular fixation23 and differences between acuity types may be reduced, approaching values of normal observers. It is possible, however, that acuity may simply improve as patients learn to use eccentric fixation effectively with time.24
Of particular relevance to the present study, is an eye chart that was developed for patients with MH.25 This chart displays multiple identical Landolt-C targets and is similar to an approach previously applied to age-related macular degeneration.26 The rationale underlying this chart design is that, by looking anywhere on the chart, at least one of the targets will fall on the most sensitive area of the retina. We questioned whether this measure reflects normal rather than optimal visual acuity, as patients may not consciously be able to use the part of the retina that provides optimal resolution.16 26 27 Horiguchi et al.25 were able to obtain acuities that were significantly better than those obtained with the standardized Landolt-C chart.25 Their acuities differed by approximately four lines before and two lines after closure of the MH. These differences could not be explained by the optotype used, as both chart types only contained Landolt rings. Here, the differences occurred because patients were not required to fixate on one target at a time but a full field of identical targets was presented. Patients achieved better acuities when fixation was not crucial in the measurement process. This raised the question as to which type of acuity test would be the best indicator of retinal resolution, to predict functional and structural outcome after treatment.28 29 30
Anstis11 reported that letter acuity declines by 0.4 logMAR units per degree of retinal eccentricity. Therefore, a patient with an MH 2° in diameter (1° of eccentric fixation) should exhibit visual acuity of 0.4 logMAR. ETDRS acuities in the present sample are generally poorer than these expected values. This additional decline in performance is consistent with work by Sjaarda et al.,31 32 who described reduced functional ability in the area surrounding an MH. Therefore, visual function in the presence of MH is characterized by the absolute scotoma at the center, surrounded by a relative scotoma in the area immediately surrounding the MH.
Additional factors influencing visual acuity measurement should be considered, such as testretest variability and possible fatigue effects. A certain amount of error is to be expected in any acuity measurement. The ETDRS chart has been shown to have testretest variability of approximately 0.18 logMAR units.33 Because the present data were collected in the context of clinical treatment, testretest data were not available. Nevertheless, if it is assumed that this error is present in both acuity measures simultaneously, the differences of 3 to 5 lines of acuity in patients with MH remain noteworthy. It is possible that Landolt-C acuities may have been affected by fatigue in the present study design as resolution acuity was always measured second, after recognition acuity. Still, the protocol did not require participants to undergo a lengthy series of acuity tests, and this type of systematic error is assumed to be negligible.
Previous work using SLOs in the context of MH research has greatly contributed to the understanding of structurefunction relationships.15 16 27 34 35 This type of technology may shed more light on the question of whether Landolt-C acuity may be an indicator of retinal resolution, whereas ETDRS acuity may additionally reflect the patients cognitive ability to use this resolution. In conclusion, recognition and resolution acuities differ in a clinically relevant way in patients with MH, before as well as after MH surgery. The choice of acuity chart should be carefully considered in the context of clinical research and ophthalmic practice for the comparison of research findings as well as the ability to predict functional outcomes.
| Footnotes |
|---|
Presented at the Canadian Society for Brain, Behaviour and Cognitive Science, Montreal, Canada, 2005.
Submitted for publication October 3, 2005; revised November 23, 2005, and January 6, January 20, and March 13, 2006; accepted May 31, 2006.
Disclosure: W. Wittich, None; O. Overbury, None; M.A. Kapusta, None; D.H. Watanabe, None.
The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked "advertisement" in accordance with 18 U.S.C.
1734 solely to indicate this fact.
Corresponding author: Walter Wittich, Department of Ophthalmology, Pavilion E-008, Sir Mortimer B. Davis Jewish General Hospital, 3755 Cote-Sainte-Catherine, Montreal, Quebec H3T 1E2, Canada; walterwittich{at}sympatico.ca.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
E. G. Gonzalez, L. Tarita-Nistor, S. N. Markowitz, and M. J. Steinbach Computer-Based Test to Measure Optimal Visual Acuity in Age-Related Macular Degeneration Invest. Ophthalmol. Vis. Sci., October 1, 2007; 48(10): 4838 - 4845. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |