Visual field defects in nonarteritic anterior ischemic optic neuropathy, plotted with a Goldmann perimeter (using I-2e, I-4e, and V-4e targets), show absolute inferior altitudinal defect with I-2e, I-4e, and V-4e isopters. The visual acuity in the eye was 20/20.
Visual field defects in nonarteritic anterior ischemic optic neuropathy, plotted with a Goldmann perimeter (using I-2e, I-4e, and V-4e targets), show inferior altitudinal defect with I-2e and inferior nasal defect with I-4e and V-4e isopters. The visual acuity in the eye was 20/20.
Hayreh SS, Zimmerman B. Visual Field Abnormalities in Nonarteritic Anterior Ischemic Optic NeuropathyTheir Pattern and Prevalence at Initial Examination. Arch Ophthalmol. 2005;123(11):1554-1562. doi:10.1001/archopht.123.11.1554
Copyright 2005 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.2005
To evaluate the pattern of various types of visual field defects and their prevalence at initial examination of nonarteritic anterior ischemic optic neuropathy (NA-AION).
The data were compiled from 312 consecutive eyes (in 265 patients) that fulfilled our inclusion and exclusion criteria. A comprehensive ophthalmic evaluation was performed, including recording of visual acuity, visual fields with a Goldmann perimeter (using I-2e, I-4e, and V-4e targets regularly), and intraocular pressure; slitlamp examination of the anterior segment; ophthalmoscopy; color fundus photography; and in acute cases, fluorescein fundus angiography. The visual field defects were divided into 2 groups: (1) general field defects and (2) various types of scotoma in the central 30°. The prevalence of various types of visual field defects was estimated for I-2e, I-4e, and V-4e isopters by dividing the total number of eyes with the defect by the total number of eyes that could see that particular target. Exact 95% confidence limits for the prevalence were computed.
Of the 265 patients, 169 (63.7%) were male and the mean ± SD patient age was 55.0 ± 9.1 years. The median interval between the first visual field test and the onset of NA-AION was 2 weeks. Of the 312 eyes, the I-2e target was seen by 75.3%, the I-4e target by 90.7%, and the V-4e target by 100%. Overall prevalence of general visual field defects was 83.4% with I-2e, 78.8% with I-4e, and 68.9% with V-4e, whereas the prevalence of scotoma(s) within the central 30° was 55.3%, 49.5%, and 36.2%, respectively. Central scotoma was seen in 48.5% with I-2e, 43.8% with I-4e, and 29.2% with V-4e. A detailed prevalence of various types of visual field defects is given. Relative inferior altitudinal defect was most common (34.9% with I-2e and 22.3% with I-4e), but the absolute inferior altitudinal defect was seen in only 8.0%. By contrast, absolute inferior nasal sector visual loss was the most common defect detected in NA-AION (22.4%), but it occurred in only 3.4% with I-2e and 11.0% with I-4e. Overall, loss of the nasal part of the visual field was the most common occurrence.
Our study demonstrated that NA-AION eyes may initially show a variety of optic nerve–related visual field defects. Our study also showed that an absolute inferior nasal visual field defect is much more common (22.4%) than an absolute inferior altitudinal visual field defect (8.0%) in NA-AION and could be considered the most characteristic single field defect in NA-AION. We found that a combination of relative inferior altitudinal defect with absolute inferior nasal defect is usually the most common pattern in NA-AION.
In nonarteritic anterior ischemic optic neuropathy (NA-AION), visual field abnormality is an important clinical criterion for diagnosis and determination of the full extent of visual loss and disability. Available information in the literature on the visual field defects in NA-AION is mainly based on manual kinetic perimetry with a Goldmann perimeter1- 12 and, more recently, using automated static threshold perimetry.13- 16Table 1 summarizes the information revealed by manual kinetic perimetry.1- 8,10- 12 Practically all the studies are retrospective rather than planned studies of a large cohort of patients with NA-AION, specifically and systematically evaluating the pattern of various types of visual field defects and their prevalence at the initial examination of patients with NA-AION. Our planned study started in 1973, when automated perimetry did not exist. Therefore, we continued to use manual kinetic perimetry with a Goldmann perimeter throughout the study to ensure consistent data from a large cohort of patients. The objective of the study was to identify and assess the pattern of visual field defects seen at the initial visit, since that constitutes an important clinical criterion in the diagnosis of NA-AION. This study consists of a large cohort of consecutive patients with NA-AION, investigated thoroughly in our clinic at the University of Iowa Hospitals and Clinics from 1973 to 1990, who fulfilled our inclusion and exclusion criteria.
We have systematically investigated patients with NA-AION in a planned study in the Ocular Vascular Clinic at the University of Iowa Hospitals and Clinics since 1973. From this cohort of patients, the initial visual field abnormality data were compiled from 312 consecutive eyes seen up to 1990 that fulfilled our inclusion and exclusion criteria.
To be included in this study, an eye was required to have a definite diagnosis of NA-AION and a satisfactory and reliable visual field test result plotted with a Goldmann perimeter. The criteria required for diagnosis of NA-AION included the following: (1) a history of sudden visual loss, usually discovered in the morning, and an absence of other ocular and neurologic diseases that might influence or explain the patient’s visual symptoms; (2) optic disc edema at onset that was documented in the Ocular Vascular Clinic or by another ophthalmologist; (3) spontaneous resolution of optic disc edema observed usually within 2 to 3 months; (4) optic disc–related visual field defects in the eye; and (5) no neurologic or ocular disorder that could be responsible for optic disc edema and visual impairment.
Patients who had any retinal or optic nerve lesion or any other factor that would have influenced the visual fields were excluded. Patients with NA-AION with only background diabetic retinopathy were included, but those who had active neovascularization, vitreous hemorrhages, traction detachment, or other complications that influenced the visual fields were excluded. Patients with a diagnosis of glaucoma and visual field loss were excluded; however, those with elevated intraocular pressure with a documented normal field before the onset of NA-AION were included. Eyes with unreliable visual field test results were excluded.
A detailed ophthalmic and medical history was obtained at the patient’s first visit to our clinic (by S.S.H.). A comprehensive ophthalmic evaluation was performed at that time (by S.S.H.), and this invariably included recording of visual acuity, visual fields with a Goldmann perimeter, relative afferent pupillary defect, and intraocular pressure; slitlamp examination of the anterior segment, lens, and vitreous; direct and indirect ophthalmoscopy; color fundus photography; and in acute cases, fluorescein fundus angiography. In cases in which giant cell arteritis was suspected based on systemic symptoms, elevated erythrocyte sedimentation rate and/or C-reactive protein level, or suspicion of arteritic AION,17,18 patients had temporal artery biopsy performed to rule out giant cell arteritis.
All patients had visual fields plotted with a Goldmann perimeter, using I-2e, I-4e, and V-4e targets regularly, although occasionally other targets, including I-1e or those between I-4e and V-4e, were used if it was thought that this would provide additional information for evaluation of the visual status. The testing was performed by trained and highly experienced perimetrists, who throughout the years have been tested periodically for interindividual and intraindividual reliability of their visual fields, with excellent results. These patients were mixed with other patients who were concurrently having perimetry for a variety of diseases, and the perimetrists were completely unaware of the ophthalmic diagnosis. The perimetrists frequently made written comments about the visual field with regard to testing difficulties encountered and patient cooperation and reliability. In particular, they carefully evaluated the area of visual loss while visually monitoring the fixation by the patient. Patients with extensive central visual loss and difficulty in maintaining fixation on the fixation hole of the Goldmann perimeter were given a large black cross to view, placed in the perimeter bowl in the center of the visual field. Only eyes whose visual field test results were judged reliable were included in the evaluation. Visual fields plotted during the first visit to our clinic were examined and coded for the pattern of visual field defects.
We divided the visual defects into 2 groups: (1) general visual field defects and (2) various types of scotoma in the central 30°. For initial compilation of the data in the data forms for each eye, we classified the various types of visual field defects that could possibly occur in NA-AION into 85 categories. However, for data analysis and presentation of results, the use of 85 categories was unwieldy; therefore, we condensed those into 22 major categories (15 for the general visual field defects and 7 for scotomas within the central 30°; Table 2) and 58 subcategories to obtain comprehensive information. For each eye, the nature of 1 or more visual field defects present was recorded separately for I-2e, I-4e, and V-4e isopters. Most of the eyes had more than 1 category of visual field defect, and it was impossible to present the data for all of the hundreds of different possible combinations.
Although in this study visual field information was available in 312 eyes with NA-AION, not every eye saw the I-2e target or both the I-2e and I-4e targets, depending on the severity of the visual loss; therefore, the prevalence of the various types of visual field defects was analyzed for only those eyes that could see that particular target and not for all 312 eyes in each isopter. The prevalence of various visual field defects was estimated for I-2e, I-4e, and V-4e isopters by dividing the total number of eyes with the defect by the total number of eyes that could see that particular target. The exact 95% confidence limits for the prevalence were computed. Descriptive statistics for the demographic variables were also obtained.
The visual field data in this study came from 312 eyes (158 right eyes and 154 left eyes) of 265 patients (47 bilateral and 218 unilateral). Of the 265 patients, 169 (63.7%) were male. The mean ± SD patient age was 55.0 ± 9.1 years (range, 22-81 years; 25% were younger than 48 years and 25% were older than 58 years). The median interval between the first visual field test and the onset of NA-AION was 2 weeks (interquartile range, 1-4.1 weeks).
Table 3 gives the overall distribution in the 312 eyes that were tested, giving the number of eyes with the target not seen and the various combinations of general field defects and scotomas. In visual fields plotted with a Goldmann perimeter, defects seen with smaller isopters are more frequent and more extensive than those with larger isopters. Therefore, some of the eyes could see the V-4e target but not the smaller targets (I-2e or I-2e and I-4e), which explains the difference in the number of eyes with I-2e, I-4e, and V-4e isopter information. Similarly, a visual field defect may not be detected with larger targets (V-4e and/or I-4e) but be present with smaller targets (I-4e and/or I-2e). Also, an eye may have general field defects and/or scotoma(s) but not necessarily both all the time.
Table 4 gives the prevalence of the major categories of visual field defects for I-2e, I-4e, and V-4e isopters, and Table 5 gives a detailed list of the prevalence for all subcategories of visual field defects. To obtain a valid prevalence of various types of visual field defects for each isopter, the prevalence was calculated for that isopter only among the eyes that could see that target and not for the entire 312 eyes in the study.
An inferior altitudinal visual field defect is almost invariably described as the classic field defect in NA-AION. A review of the literature on visual fields in NA-AION, in which visual fields were plotted with manual kinetic perimetry, shows that the reported prevalence rate of inferior altitudinal visual field defect varies from 25% to 79% (Table 1). Ellenberger et al,3 however, in the 64 eyes from their study, reported no such defect but recorded seeing inferior (34 eyes) or superior (17 eyes) arcuate field defects or central scotoma (8 eyes). Other less common types of visual field defect reported in NA-AION are given in Table 1. In recent years, the use of automated static threshold perimetry has become common; among these studies, we found only 2 reports13,16 that provided some information on the pattern of visual field defects in NA-AION. Traustason et al,13 using Octopus automated static perimetry (Haag-Streit, Bern, Switzerland) within the central 30° in 47 eyes, reported inferior altitudinal defects in 55.3% and “diffuse” visual field defects in 45%. WuDunn et al,16 using Humphrey perimetry, tangent screen, or both in 31 cases with bilateral NA-AION, reported that the visual field was “severely depressed” or “unable” to record in 17 eyes and overall defects that involved the inferior hemisphere in 35 eyes, superior hemisphere in 18, central in 8, and constricted in 6. A review of the reported studies in the literature on the overall pattern of visual field defects in NA-AION reveals that (1) in almost all of the studies, the visual field information was given simply as part of the general clinical features of NA-AION rather than based on specific, planned, and systematic investigation of visual field patterns seen in NA-AION at initial examination; (2) no mention was made of whether those were relative or absolute defects (see later in this section); and (3) most studies were based on a comparatively small number of eyes. Ours, by contrast, was a systematic planned study of the largest number of consecutive NA-AION eyes (312 eyes) reported so far, with visual field defects categorized into almost all possible types that could be seen in NA-AION and data analyzed not only for all of those types but also for whether they were relative (detected only with I-2e and/or I-4e isopter of the Goldmann perimeter) or absolute (detected with I-2e, I-4e, and V-4e isopters). Therefore, our findings are likely to be different and will, we hope, provide more reliable information on the various patterns of visual field defects seen at the initial visit in patients with NA-AION.
The various types of visual field defects seen in our study (Tables 4 and 5) differ from those reported in the literature (Table 1). There may be several reasons for this.
The first reason is the number of patients in a study: obviously, the larger the series, the more reliable the results are likely to be. Our study was a planned systematic study of 312 eyes.
The second reason is the method used to plot visual fields: visual field information provided by manual kinetic perimetry performed with a Goldmann perimeter, which we used, may be very different from that provided by automated static threshold perimetry (Humphrey 30-2 or 24-2 SITA [Swedish Interactive Testing Algorithm]). We did not use automated perimetry because our study started before automated perimetry was available. Therefore, we continued to use manual kinetic perimetry with a Goldmann perimeter throughout the study to ensure consistent data for a large cohort of patients. Moreover, the changing face of automated perimetry would make such long-term studies difficult; automated perimetry is still evolving. Both types of perimetry have their advantages and disadvantages, and one should be aware of those when interpreting the findings.
The third reason is peripheral vs central visual loss in NA-AION and their role in functional disability. Nonarteritic anterior ischemic optic neuropathy is a common visually disabling disease. To assess that disability, one needs information on both peripheral and central visual field defects. The kind of functional disability produced by the 2 types can be very different.
It is well established that constant tracking provided by the peripheral visual fields is essential for sensory input in our day-to-day activities, for example, driving and “navigating” generally. In view of that, to assess the visual function disability produced by NA-AION, it is important to have complete information about the peripheral visual fields and any impairment in them. Automated perimetry provides information on only up to approximately 24° to 30° in the periphery. Kinetic perimetry, by contrast, provides peripheral visual field information all the way to approximately 80° to 90° temporally, 70° inferiorly, 60° to 70° nasally, and 50° to 60° superiorly. Thus, for evaluating visual functional disability in NA-AION, the visual field plotted with manual kinetic perimetry provides far superior information about peripheral visual field defects.
It is well known that a defect in the central visual field usually results in deterioration of visual acuity.
Our experience in dealing with patients with NA-AION for some 35 years has shown us that loss of the peripheral visual field, particularly in the lower part, can be highly disabling. For example, a person with central scotoma but with a normal peripheral visual field can lead a fairly normal life, although he or she will not able to read, write, and see properly in the center. By contrast, patients with bilateral absolute inferior altitudinal defects or with markedly constricted visual fields but with perfectly normal visual acuity (Figure 1) are markedly disabled, stumbling, and unable to drive or navigate. Obviously, this has enormous practical implications. Disability boards tend to deny eligibility to persons with NA-AION who can see 20/20 in both eyes despite the fact that they have complete loss of the lower half of the visual fields in both eyes (Figure 1); from a practical point of view, they are legally blind.
The fourth reason is the distinction between relative and absolute visual field defects. A review of the literature shows that descriptions of the various types of visual field defects in NA-AION almost never specify whether those were relative or absolute (Table 1). However, a distinction between relative and absolute visual field defects is essential for the following reasons.
The pattern of visual field defects is different in these 2 types: in visual fields plotted with a Goldmann perimeter, defects seen with smaller isopters are more frequent and more extensive than those with larger isopters (Figure 2). The prevalence of pattern of various types of visual field defects therefore depends on whether it is relative or absolute and, even among the relative field defects, whether it is detected with only I-2e or both I-2e and I-4e isopters (Figure 2) (Tables 4 and 5). Of the 312 eyes in our study, the I-2e target was seen by 75.3%, I-4e by 90.7%, and V-4e by 100% (Table 3). The overall prevalence of general visual field defects was 83.4% with I-2e, 78.8% with I-4e, and 68.9% with V-4e, whereas the prevalence of scotoma(s) within the central 30° was 55.3%, 49.5%, and 36.2%, respectively (Table 3). Although the relative inferior altitudinal defect was most common (34.9% with I-2e and 22.3% with I-4e), that was not true of the absolute inferior altitudinal defect (8.0% with V-4e). In sharp contrast, absolute inferior nasal sector visual loss was the most common defect seen in NA-AION (22.4%) but was seen in only 3.4% with I-2e and 11.0% with I-4e. We have found that the most common pattern is a combination of relative inferior altitudinal defect with relative or absolute inferior nasal defect (Figure 2). From the data in Table 5, we also calculated the prevalence rate of any visual field defects present in the V-4e isopter in the superior (in 24.4%), inferior (in 72.1%), nasal (in 50.3%), and temporal (in 11.9%) regions of the visual field, located in only 1 region or extending to adjacent regions. (Most of the eyes had more than 1 location of visual field defect.) This further revealed that the most common combination is inferior and nasal absolute visual field defect in NA-AION and much less frequently involves the superior and temporal regions.
In addition, the extent and severity of visual field defect have functional significance because in the region of the relative field defect, there is still a variable amount of intact visual function: more in the defect seen with the I-2e isopter than that seen with the I-4e isopter. For example, an absolute inferior altitudinal visual field defect (Figure 1) is far more visually disabling than only a relative inferior altitudinal defect, and the latter is less disabling when it is detected with I-2e rather than I-4e (Figure 2).
The fifth reason that the prevalence of various types of visual field defects seen in our study differs is the characterization of visual field defects in NA-AION. Information provided by automated perimetry may result in different characterization of the pattern of the visual field defects compared with that of kinetic perimetry. For example, in automated perimetry, a large scotoma or defect that extends all the way to 30° is almost invariably interpreted as an “altitudinal field defect” when kinetic perimetry may not show this.
Tables 4 and 5 give details of the prevalences of various categories of visual field defects seen in this study. These tables illustrate the following 2 particularly interesting and important features.
Table 4 illustrates that with a V-4e isopter, an absolute inferior nasal sector defect was seen in 22.4%, and if the prevalence of an absolute inferior nasal step (Table 5) is combined with that, then an overall inferior nasal defect was seen in 35.9%; in contrast, absolute inferior altitudinal defects were seen in only 8.0%. Thus, contrary to the prevalent impression, an absolute inferior nasal sector defect is the most common type of visual field defect in NA-AION. Our experience since 1990 has further confirmed this. Other types of visual field defect are much less common.
Tables 4 and 5 also demonstrate that overall, loss of the nasal part of the visual field is the most common occurrence in NA-AION. This is because (1) a nasal field defect (ie, nasal quadrant + nasal step + nasal vertical) was seen in 44.2% compared with the corresponding temporal field defect (ie, temporal quadrant + temporal wedge) seen in only 10.6%; (2) nasal vertical, almost hemianopic, visual field defect was seen in 1.6%, but no temporal vertical defect was seen in this study; and (3) of the 16 eyes with only a residual peripheral island field remaining with the V-4e isopter, none was in the nasal region, but in 14 eyes it was present in the temporal region.
The finding that the most common loss in NA-AION is of the nasal part of the visual field can be explained by the pattern of location of the watershed zone between the posterior ciliary arteries in relation to the optic nerve head.19- 22 The part of the optic nerve head located in the watershed zone is the most vulnerable to ischemia. In our studies, the most common location of the watershed zone is in or adjacent to the temporal part of the optic disc (see figure 3A and B in the article by Hayreh22). Therefore, the temporal part of the optic nerve head is more vulnerable to ischemic disorders than the nasal part, hence the more prevalent type of nasal visual loss in NA-AION. When the watershed zone is located only in the superior temporal part of the optic nerve head (see figure 16 in the article by Hayreh21), it corresponds to the frequently seen inferior nasal visual field defect in NA-AION.
Table 4 also demonstrates that among scotomas within the central 30°, central scotomas of one or another type (including centrocecal scotoma) were seen in 48.5% with I-2e, 43.8% with I-4e, and 29.2% with V-4e. Unfortunately, the frequent occurrence of a central scotoma is not even mentioned in most descriptions of visual field defects in NA-AION. As was expected, the severity of peripheral constriction of visual fields was most marked with I-2e (in 19.6%) and least with V-4e (in 4.8%).
In 6.4% of the eyes in our study, no visual field defect could be detected at the initial visit even with I-2e, despite all other classic findings of NA-AION; in some we could detect a defect only with the I-1e target, whereas others had incipient NA-AION20,23 (and later developed visual field defects).
Table 5 gives detailed information about the prevalence of various subcategories of visual field defects seen in NA-AION in the present study. This information is highly relevant for several reasons, including detecting the severity of visual disability, particularly central visual disability. For example, whether an altitudinal or a vertical field defect spares the central 5°, bisects it (Figures 1 and 2), or includes it determines the level of visual acuity in that eye; this information is lacking in the literature. Similarly, the size of a central scotoma or centrocecal scotoma determines the level of visual acuity: the larger the scotoma, the worse the visual acuity. The same applies to the height of the centrocecal scotoma.
Correspondence: Sohan Singh Hayreh, MD, MS, PhD, DSc, FRCS, FRCOphth, Department of Ophthalmology and Visual Sciences, University Hospitals and Clinics, 200 Hawkins Dr, Iowa City, Iowa 52242-1091 (email@example.com).
Submitted for Publication: June 14, 2004; final revision received November 10, 2004; accepted February 3, 2005.
Financial Disclosure: None.
Funding/Support: This study was supported in part by grants EY-1151 and RR-59 from the National Institutes of Health, Bethesda, Md, and in part by an unrestricted grant from Research to Prevent Blindness, Inc, New York, NY. Dr Hayreh is a Research to Prevent Blindness Senior Scientific Investigator.
Acknowledgment: We are grateful to J. J. S. Barton, MD, and T. J. Martin, MD, for their help with the initial compilation of the data in the data forms and to W. L. M. Alward, MD, R. H. Kardon, MD, PhD, and M. Wall, MD, for their helpful suggestions and critique of the manuscript.