THE RELATIONSHIP BETWEEN PERIMETRIC AND METABOLIC DEFECTS CAUSED BY EXPERIMENTAL GLAUCOMA

RONALD. S. HARWERTH1 and MORRIS L.J. CRAWFORD2

1College of Optometry, University of Houston; 2Department of Ophthalmology and Visual Science University of Texas–Houston; Houston, TX, USA

Abstract

Purpose: Glaucomatous optic neuropathy causes a loss of visual sensitivity and a reduction in the metabolism of neurons in the afferent visual pathway. The purpose of the present investigation was to correlate the visual and metabolic alterations caused by experimental glaucoma. Methods: The metabolic activities of neurons in the magnocellular and parvocellular pathways were analyzed in tissue from 16 monkeys with visual field defects caused by laser-induced, elevated intraocular pressures. Visual fields were assessed behaviorally by standard clinical perimetry. The effects on the metabolism of neurons that were topographically related to perimetry defects were determined by cytochrome oxidase histochemistry. Results: There was general agreement between the diffuse loss of visual sensitivity and the percentage reduction in cytochrome oxidase reactivity, with a higher correlation for the parvocellular (r = 0.74) than magnocellular pathway (r = 0.52). The percentage reduction in metabolic activities in the subdivisions of the afferent pathway were correlated (r = 0.86) with a slope of 0.6%/%, indicating a proportionally greater loss in the magnocellular than parvocellular pathway. Conclusions: The results do not support the premise that alternative perimetry stimuli based on magno-parvo distinctions will improve the accuracy of clinical perimetry for the early detection of glaucoma. The principal findings are in agreement with structurefunction models that small ensembles of afferent neurons with the highest sensitivity for the stimulus determine visual thresholds, while the density of cytochrome oxidase reactivity is determined by the combined activity of all of the neurons in the sampled site.

Introduction

It is well known that glaucoma is an optic neuropathy caused by the death of retinal ganglion cells, but recent studies have shown that there are also signs of neuropathy in higher brain centers, especially the lateral geniculate nucleus (LGN). For example, studies of Macaque monkeys with experimental glaucoma have shown LGN cell losses, shrinkage of the cell bodies, and reductions in metabolic activity.1-7 Such alterations of neurons suggest that their responses should also be affected, and that part of the loss in visual sensitivity from glaucoma could occur from neural defects that are proximal to the retina. However, although electrophysiological investigations have

Address for correspondence: Ronald S. Harwerth, 505 J. Davis Armistead Building, College of Optometry, University of Houston, Houston, TX 77204-2020, USA. Email: rharwerth@uh.edu

Perimetry Update 2002/2003, pp. 175–186

Proceedings of the XVth International Perimetric Society Meeting, Stratford-upon-Avon, England, June 26–29, 2002

edited by David B. Henson and Michael Wall

© 2004 Kugler Publications, The Hague, The Netherlands

176 R.S. Harwerth and M.L.J. Crawford

demonstrated reduced encounter rates for LGN cells driven by eyes with experimental glaucoma, the response characteristics of the cells innervated by the two eyes were indistinguishable.8 Consequently, the cytological alterations of the LGN could be either an important part of the process of vision loss caused by glaucoma, or simply a reflection of the reduction of ganglion cell innervation. This relationship has additional ramifications from the studies of neurotrophin deprivation as a factor in initiating ganglion cell death from glaucoma, which suggests that atrophy of LGN neurons could precede measurable losses in visual sensitivity.9,10 Therefore, the primary purpose of the present study was to determine the functional significance of neuropathy in higher brain centers. Specifically, the investigation was undertaken to study the relationship between the reduction in neural metabolic activity in the LGN and the functional changes in visual sensitivity, as measured by clinical perimetry.

A second purpose of the study was to determine whether glaucoma causes differential effects in the major anatomical/ functional subdivisions of the retinal-geniculate pathway. The possibility of selective effects is important because, if all the functional classes of ganglion cells are not killed at the same time, or at the same rate, then in the early stages of glaucoma, the loss of visual sensitivity may be caused by a selective loss of a specific population of ganglion cells.11-14 This premise of selective cell death has provided a foundation for the development of psychophysical tests for the diagnosis and assessment of progression of the disease on the basis of the distinctive functional properties of specific populations of retinal ganglion cells.15 As the current example, frequency-doubling perimetry was designed on a stimulus configuration that was specific to a homogeneous, sparse population of retinal ganglion cells that were presumed to be affected during the early stages of glaucoma.16,17 However, the evidence for selective death of retinal ganglion cells is not uncontested. On the one hand, it has been shown that glaucoma causes a selective loss of the larger retinal ganglion cells, while on the other, the majority of studies of morphological and metabolic effects in the LGN have shown that the magnocellular and parvocellular channels are equally affected by glaucoma.1-7 One of the difficulties in reconciling these conflicting results is that the morphological studies have lacked a vision function assessment for determining whether differential effects on the magnocellular and parvocellular neurons depend on the stage of visual field loss and, thereby, on the underlying degree of ganglion cell loss. Thus, as a second purpose, the present study was planned to investigate the effects of the stage of experimental glaucoma on the metabolic activities of neurons in the magnocellular and parvocellular lamina of the lateral geniculate nucleus.

Methods

The subjects of the studies were 16 rhesus monkeys (Macaca mulatta) with unilateral experimental glaucoma induced by laser photocoagulation of the trabecular meshwork of their right eyes.18,19 The experimental and animal care procedures were reviewed and approved by the Institutional Animal Care and Use Committees of the University of Houston and the University of Texas–Houston. The use of animals for these experiments adhered to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research.

Glaucomatous visual field defects were tested by the same methodology and by statistical analyses of perimetry data that have been developed for human patients, and which were applied to experimental glaucoma through behavioral training and testing of the monkey subjects.20,21 For these measurements, a standard clinical instrument, the Humphrey Field Analyzer (HFA I) was attached to a primate-testing cubicle, and the monkeys were trained to fixate and perform the same sort of detection task used for clinical perimetry. After training was completed, standard automated perimetry using the 24-2 pattern of test field locations and the full-threshold test strategy with a size III, white test stimulus, was used to assess the onset and progression of visual field defects caused by experimental glaucoma.

Tissue for histochemical analysis was collected from animals with visual field defects, which represented clinical stages that varied from mild-early to end-stage glaucoma. In order to obtain the neural tissue, the monkeys were overdosed with pentobarbital sodium, and their brains were perfused in situ with paraformaldehyde fixatives and then removed. The lateral geniculate nuclei were dissected, dehydrated, sectioned, and stained for the histochemical assay.5,6,22

The effects of experimental glaucoma on the metabolic activity of neurons in the afferent pathway were assessed by cytochrome oxidase reactivity. Cytochrome oxidase is an essential enzyme in neural energy metabolism, and its concentration within the mitrochondrial membrane is tightly correlated with the level of neural activity.22 This correlation in neural-metabolism has been demonstrated by the intravitreal injection of tetrodotoxin (TTX) to block neural impulses from the retinal ganglion cells, which causes a rapid reduction in the cytochrome oxidase concentration of these neurons, as well as the neurons in other afferent pathway sites (i.e., lateral geniculate nucleus and visual cortex).23-25 In glaucoma, alterations of neural-metabolism in the afferent pathway sites should be retinotopically correlated to the depth of visual field defects and, if they are selectively affected by glaucoma, to neurons in the magnoand parvocellular subdivisions of the parallel afferent pathways.5,6 Therefore, relative losses of cytochrome oxidase activity and visual field defects at concordant locations in lamina innervated by the experimental and control eyes of monkeys were compared in order to establish the relationship between topographically related perimetric and metabolic defects caused by experimental glaucoma.

Results

The relationships between visual field defects and the loss of cytochrome oxidase activity for each of the monkeys were determined at 12 visual field locations. The sample locations are illustrated in Figure 1A for the experimental, right eye by the star-symbols in the nasal field of the perimetry threshold plot. Comparable locations in the temporal visual field of the control eye were also sampled. The visual field locations were selected to lie between the actual perimetry test field locations, so that the means of surrounding measurements could be used as the representative sensitivity, and thus allow for small inaccuracies in the topographical projection of visual field onto the LGN. For each of the visual field locations, a visual sensitivity was derived from the mean of the perimetry sensitivity values surrounding the nominal field site. For example, for the location indicated by the circle in the superior-nasal

visual field of the right eye, the sensitivity was taken as the mean of the four sensitivity values bounded by the circle, which, in this case, is in a deep scotoma (Fig. 1B) and included three zero sensitivity values (mean = 2 dB). In comparison, the mean sensitivity for the corresponding location in the superior-temporal visual field of the normal, left eye was 28.5 dB.

The relative levels of neuronal metabolic activity were determined from histological sections of the monkeys’ right LGNs (i.e., ipsilateral to the experimentally treated eye) that had been processed by established protocols for visualization of cytochrome oxidase reaction product.5,6,22-25 With this procedure, the relative levels of metabolic activity are proportional to the density of cytochrome oxidase staining, with darker staining indicative of higher levels of neural energy metabolism. The example of the grayscale image of a stained histological section of an LGN, presented in Figure 1C, illustrates the effects of glaucomatous neuropathy. The tissue is from a monkey which had developed moderate visual field defects in the right eye (Figs. 1A and B), which are correlated with the obvious differences in staining across the LGN lamina. The more darkly stained lamina (i.e., parvocellular layers 6 and 4, and magnocellular layer 1) had recipient innervation from retinal ganglion cells in the right hemi-retina of the left, untreated control eye, which corresponds to the normal sensitivity of the temporal visual field of the left eye. Conversely, the pale laminae (i.e., parvocellular layers 5 and 3, and magnocellular layer 2), received input from the temporal hemi-retina of the right, treated eye, which is the area of the retina with reduced sensitivity that produced the nasal visual field defects seen in the perimetry data. The relative reduction in staining density of all the lamina with recipient innervation from the right eye, compared to the lamina innervated by the left eye, indicates a general reduction in the neural-driven metabolic activity as a result of neural damage from experimental glaucoma, but does not determine whether the amounts of reduced metabolism and neural damage are related.

In order to relate more directly the depth of visual field defects and the reduction of cytochrome oxidase activity, the relative densities of the cytological stains in adjacent lamina were determined for areas of the LGN that were topographically related to the visual field stimuli. The topographical projections of the visual hemifield locations onto the LGN were based on projection maps of Malpeli and Baker,26 which showed that the visual hemifield representation is divided into upper and lower quadrants by a horizontal meridian projection line (represented by line HM in Fig. 1C), which runs in the anterior-posterior plane along the crest of the nucleus. The horizontal meridian projection intersects neurons within each laminae that represent a constant location in the visual field, and across lamina the midline projection intersects

→

Fig. 1. Examples of the perimetry data and an anatomical section of the right LGN that was stained to visualize cytochrome oxidase. The data are from a monkey with moderate visual field defects from experimental glaucoma. A. The visual thresholds for the right and left eyes. The star-symbols superimposed on the plot for the right eye represent the visual field locations for the correlation of sensitivity and histochemical alterations; comparable locations sampled in the temporal visual field of the left eye. The circles on the left and right visual field data show the locations for threshold measurements that were averaged to compare to the metabolic activity in the areas of the LGN that receive topographically related input from the retina of each eye. B. The grayscale plots of the visual field data to illustrate the stage of glaucoma that caused the reductions of cytochrome oxidase activity in the LGN lamina innervated by the

180 R.S. Harwerth and M.L.J. Crawford

corresponding locations in the visual fields of each eye. The area above the horizontal meridian projection represents inferior visual hemifield locations and the area below the projection line represents the superior visual hemifield. The visual field locations that are circled perimetry data in Figures 1A and B project to the circled areas of the LGN that are marked by a ‘P’ in the parvocellular layers and by an ‘M’ in the magnocellular layers (Fig. 1C). Thus, the glaucoma-induced change in cytochrome oxidase activity was derived from optical density measurements of the levels of staining in corresponding retinal projections from the control and experimental eyes, both for the parvocellular layers (i.e., ratio of layers 5/6) and the magnocellular layers (i.e., ratio of layers 2/1).

A general relationship between perimetric and metabolic defects was developed from the comparisons of the experimental to control eyes, with respect to the losses of visual field sensitivity and reductions in cytochrome oxidase activity, which were derived from the means of the 12 sampled locations for both effects. As would be expected, the sensitivity loss calculated by the average difference in control and experimental eye sensitivities was approximately equal to the mean deviation (MD) global index, which is one of the standard perimetry indices derived from comparisons of a given patient’s perimetry data to the normal expected threshold values. Thus, for this part of the study, the mean sensitivity losses reflect the diffuse defects in visual sensitivity.

The data for the 16 monkeys, expressing the mean sensitivity loss as a function of the mean metabolic loss, for the parvocellular or magnocellular lamina are presented in Figures 2A and 2B, respectively. In both cases, the relationship appears to be linear, but with a tighter correlation for the parvocellular (r = 0.74) than the magnocellular (r = 0.52) lamina. The lower correlation for the sensitivity versus metabolism function for the magnocellular lamina is caused by the larger range of metabolic defects across animals and, as a result, the slope of the function (0.60 dB/%) is approximately onehalf the slope for the function for the parvocellular lamina (1.33 dB/%). A similar effect from the larger range of reduced neural metabolism in the magnocellular neurons is seen in the correlation between metabolic defects in the two neural channels, presented in Figure 2C. The comparison between neural pathways demonstrates that the parvocellular and magnocellular deficits are well correlated (r = 0.86), with slope of 0.60 %/%. This result is in agreement with the sensitivity-metabolic loss functions (Figs. 2A and B) in indicating that, as the stage of visual field loss from experimental glaucoma advances, there is an increasingly greater effect on the metabolism of the magnocellular pathway neurons, compared to the parvocellular pathway neurons. However, these magno-parvo differences in metabolic effects appear to be less significant during the initial stages of glaucoma, because the function intersects both axes at near-zero values.

Two individual examples of the point-by-point relationships between the experimental and control eyes’ sensitivity losses and reductions in cytochrome oxidase are presented in Figure 3. The data in Figure 3A, for an animal (OHT-25) with relatively mild visual field defects, demonstrate systematic changes in sensitivity as a function of reduced metabolic activity. The data for both subdivisions of the retinal-geniculate pathway are well described by linear functions on the coordinates for log sensitivity loss versus percentage cytochrome oxidase reduction. Notably, in agreement with the averaged data in Figure 2, the slope of the function for magnocellular neurons is shallower than the function for parvocellular

Fig. 2. The effects of experimental glaucoma on the energy metabolism of LGN neurons. A. The relationship between sensitivity loss and reduction of cytochrome oxidase in neurons in the parvocellular lamina. Sensitivity losses represent the mean (±1 SD) for the differences between visual sensitivities of the experimental and control eyes for each of the16 monkeys with varying degrees of visual field defects. The cytochrome oxidase reductions represent the average difference (±1 SD) in the optical densities of the lamina innervated by the experimental and control eyes at the 12 sampled locations for each laminae. B. Sensitivity loss versus reduction in cytochrome oxidase for the magnocellular lamina. Other details are the same as for panel A. C. The relationship between the reduction in cytochrome oxidase in the parvocellular and magnocellular lamina. Data are presented for 16 monkeys with varying degrees of visual field defects from experimental glaucoma.

neurons, indicating for a given sensitivity loss measured by perimetry, that there was a greater reduction for neural metabolism for the M than for the P pathway. It is also important to note that, for this animal, the y-intercept values for both functions are negative; implying that the metabolic defects associated with mild visual field defects preceded the sensitivity losses. However, this effect was not consistent across animals.

In comparison, the relationships are not clear-cut with more advanced stages of glaucoma, as is illustrated by the data in Figure 3B. For this animal (OHT-18), the levels of sensitivity loss and reduction in metabolic activity are greater than for OHT-25 (Fig. 3A), but the relationships are not well correlated, especially for the P-path neurons. It is also evident that the large regional variations in visual field sensitivities with advanced glaucoma create considerable uncertainty in the sensitivity loss associated with a specific area of the lateral geniculate nucleus. For that reason, the global relationship between

Fig. 4. The point-by-point relationship between sensitivity loss and reduction in cytochrome oxidase reactivity for each of the 12 sampled locations from each of the 16 monkeys. Data are presented for the parvocellular lamina in A. and the magnocellular lamina in B.

sensitivity and metabolic effects, as presented in Figure 2, is a more appropriate description of the effects of glaucoma than a point-by-point regional relationship. This point is made more clearly by the data in Figure 4, which presents the point-by-point relationships for each of the samples from all of the monkeys. It is obvious that the substantial scatter for sensitivity losses as a function of the cytochrome oxidase reduction in both the parvocellular (Fig. 4A) and magnocellular (Fig. 4B) pathways prohibits the derivation of useful relationships between these two aspects of neural function, at this level of point-by-point assessment.

Discussion

This study has confirmed the results of prior investigations in showing that glaucoma causes significant alterations in the visual pathway at sites that are proximal to the retinal ganglion cells.1-8 In this study, the relative reduction of metabolic activity was determined for the neurons in the LGN lamina that were innervated by the monkeys’ experimental eyes, compared to their control eyes. The level of metabolic activity was assessed by the concentration of cytochrome oxidase, an essential enzyme in neural energy metabolism.22 The principal finding was that, across animals with varying degrees of visual field defects, average neural energy metabolism was reduced in proportion to the diffuse loss of visual field sensitivity (Figs. 2A and B). The average decibel losses in visual sensitivity were correlated with the percentage reductions in cytochrome oxidase activity for both the parvoand magnocellular divisions of the retinal-geniculate pathway, but the correlation was higher and the slope of the function higher for the parvocellular, compared to the magnocellular, pathway. Therefore, a given level of visual field defect was indicative of a larger reduction in metabolism of neurons in the magnocellular pathway than the parvocellular pathway. Nevertheless, because the best-fit functions intersected the axes of the X-Y coordinates close to their origins, these data do not provide evidence of an alteration in LGN physiology preceding the visual field defects. Therefore, although a substantial number of ganglion cells may have died before visual field defects became correlated

with the loss of ganglion cells,12,27,28 the continuity of functions over the full range of defects suggests that the early defects are concurrent at the two levels, and that the LGN changes are a reflection of reduced afferent connections from retinal ganglion cells.

The experimental data also revealed some interesting differences in the effects of experimental glaucoma on the retinal ganglion cells that comprise the parvocellular and magnocellular divisions of the afferent visual pathway. The mean changes in neural energy metabolism in the parvocellular and magnocellular pathways were well correlated (r = 0.86), with the slope of the function relating the percentage of change in each subdivision (0.6 %/%) indicating that the proportional reductions in the magnocellular pathway were greater than for the parvocellular neurons (Fig. 2C). The other parameter of the linear regression, the y-intercept, was essentially at the origin of the axes, which is consistent with the early, initial defects occurring simultaneously in both subdivisions of the visual pathway. These findings, together with the poorer correlation between the sensitivity and metabolic effects for the magnocellular than parvocellular channels, cause difficulties in the application of these results for the development of perimetry stimuli based on the functional characteristics on neurons in each channel.29,30 For example, the origin of the function does not show an offset which would indicate that there is a period during which neurons in one channel are affected prior to the other and, thus, provide an optimal method for the early detection of visual field defects. In addition, although the data suggest that the use of perimetry stimuli with characteristics that are specific to the properties of the magnocellular neurons might reveal larger visual deficits than those designed for the response properties of parvocellular neurons, the difference should increase with the degree of visual deficit, rather than being more efficient for early diagnosis. It is also noteworthy that the relative correlations between sensitivity and metabolism imply that visual field measurements will have less specificity if responses are restricted to the magnocellular neurons (cf. Figs. 2A and B). Therefore, these data do not suggest that alternative perimetry stimuli based on distinctions in the response properties of magnocellular or parvocellular neurons is apt to improve the accuracy of clinical perimetry. Rather, optimization of the structure-function relationship may be accomplished by stimuli that are specific to a homogeneous population of retinal ganglion cells, so that visual sensitivity can be analyzed by basic principles of probability summation.31

The overall results of these investigations have shown qualitative agreement between the average defects in visual and metabolic functions, but not at a more local point-by- point assessment. This level of agreement is not surprising because both aspects are related to glaucomatous optic neuropathy and, therefore, both effects should be proportional to the levels of neural activation from the retinal ganglion cells to the LGN. However, while these diverse methods of investigation have produced compatible results, it is important to recognize that metabolic structure-function relationships are based on fundamentally different principles from the psychophysical structure-function relationships. The density of cytochrome oxidase reactivity at a given anatomical site is determined by the sum of the neural activation of cytochrome oxidase at that site, while psychophysical thresholds at corresponding visual field locations will be determined by a small ensemble of neurons that have the highest sensitivities. It is most likely that the same process of ganglion cell death underlies both the behavioral and cellular deficits, but the correlation of visual losses and the extent of metabolic changes will not necessarily be precise.

Acknowledgments

This study was supported in part by a research grant from Alcon Laboratories, Inc., Fort Worth, TX, and National Institutes of Health/National Eye Institute grants RO1 EY01139, and P30 EY0551 to the University of Houston and grants RO1 EY07751, RO1 EY11545, and P30 EY10608 to the University of Texas–Houston. Additional support was provided by the University of Houston from a John and Rebecca Moores Professorship, and by the University of Texas Health Science Center–Houston from the Herman Eye Fund, The Vale-Asche Foundation, and Research to Prevent Blindness.

References

1.Vickers JC, Hof PR, Schumer RA, Wang RF, Podos SN, Morrison JH: Magnocellular and parvocellular visual pathways are both affected in a macaque model of glaucoma. Aust NZ J Ophthalmol 25:239243, 1997

2.Weber AJ, Chen H, Hubbard WC, Kaufman PL: Experimental glaucoma and cell size, density, and number in the primate lateral geniculate nucleus. Invest Ophthalmol Vis Sci 41:1370-1379, 2000

3.Yucel YH, Zhang Q, Gupta N, Kaufman PL, Weinreb RN: Loss of neurons in magnocellular and parvocellular layers of the lateral geniculate nucleus in glaucoma. Arch Ophthalmol 118:378-384, 2000

4.Morgan JE, Uchida H, Caprioli J: Retinal ganglion cell death in experimental glaucoma. Br J Ophthalmol 84:303-310, 2000

5.Crawford MLJ, Harwerth RS, Smith EL, Shen F, Carter-Dawson L: Glaucoma in primates: cytochrome oxidase reactivity in parvoand magno-cellular pathways. Invest Ophthalmol Vis Sci 41:17911802, 2000

6.Crawford MLJ, Harwerth RS, Smith EL, Mills S, Ewing B: Experimental glaucoma in primates: changes in cytochrome oxidase blobs in V1 cortex. Invest Ophthalmol Vis Sci 42:358-364, 2001

7.Yucel YH, Zhang Q, Weinreb RN, Kaufman PL, Gupta N: Atrophy of relay neurons in magnoand parvocellular layers in the lateral geniculate nucleus in experimental glaucoma. Invest Ophthalmol Vis Sci 42:3216-3222, 2001

8.Smith EL, Chino YM, Harwerth RS, Ridder WH, Crawford MLJ, DeSantis L: Retinal inputs to the monkey’s lateral geniculate nucleus in experimental glaucoma. Clin Vis Sci 8:113-139, 1993

9.Quigley HA, Nickells RW, Kerrigan LA, Pease ME, Thibault DJ, Zack DJ: Retinal ganglion cell death in experimental glaucoma and after axotomy by apotosis. Invest Ophthalmol Vis Sci 36:774786, 1995

10.Quigley HA: Neuronal death in glaucoma. Prog Retinal Eye Res 18:39-57, 1998

11.Quigley HA, Sanchez RM, Dunkelberger GR, L’Hernault NL, Baginski TA: Chronic glaucoma selectively damages large optic nerve fibers. Invest Ophthalmol Vis Sci 28:913-920, 1987

12.Quigley HA, Dunkelberger GR, Green WR: Retinal ganglion cell atrophy correlated with automated perimetry in human eyes with glaucoma. Am J Ophthalmol 107:453-464, 1989

13.Glovinsky Y, Quigley HA, Dunkelburger GR: Retinal ganglion cell loss is size dependent in experimental glaucoma. Invest Ophthalmol Vis Sci 32:484-491, 1991

14.Johnson CA: Selective versus nonselective losses in glaucoma. J Glaucoma (Suppl) 3:S32-S44, 1994

15.Sample PA, Bosworth CF, Blumenthal EZ, Girkin C, Weinreb RN: Visual function-specific perimetry for indirect comparison of different ganglion cell populations in glaucoma. Invest Ophthalmol Vis Sci 41:1783-1790, 2000

16.Maddess T, Henry GH: Performance of nonlinear visual units in ocular hypertension and glaucoma. Clin Vis Sci 7:371-383, 1992

17.Johnson CA, Samuels SJ: Screening for glaucomatous visual field loss with frequency-doubling perimetry. Invest Ophthalmol Vis Sci 38:413-425, 1997

18.Quigley HA, Holman RM: Laser energy levels for trabecular meshwork damage in the primate eye. Invest Ophthalmol Vis Sci 24:1305-1307, 1987

19.Harwerth RS, Smith EL, DeSantis L: Experimental glaucoma: perimetric field defects and intraocular pressure. J Glaucoma 6:390-401, 1997

20.Harwerth RS, Smith EL, DeSantis L: Behavioral perimetry in monkeys. Invest Ophthalmol Vis Sci 34:31-40, 1993

21.Harwerth RS, Smith EL, DeSantis L: Mechanisms mediating visual detection in static perimetry. Invest Ophthalmol Vis Sci 34:3011-3023, 1993

22.Wong-Riley M: Cytochrome oxidase: an endogenous metabolic marker for neuronal activity. Trends Neurosci 2:94-101, 1989

23.Wong-Riley M, Carroll EW: The effect of retinal impulse blockage on cytochrome oxidase activity in the monkey visual system. Nature 307:262-264, 1989

24.Wong-Riley MTT, Tripathi SC, Trusk TC, Hoppe DA: Effect of retinal impulse blockage on cytochrome oxidase-rich zones in the macaque striate cortex. I. Quantitative EM analysis of neurons. Vis Neurosci 2:483-497, 1989

25.Wong-Riley MTT, Tripathi SC, Trusk TC, Hoppe DA: Effect of retinal impulse blockage on cytochrome oxidase-rich zones in the macaque striate cortex. II. Quantitative EM analysis of neuropil. Vis Neurosci 2:499-514, 1989

26.Malpeli JG, Baker FH: The representation of the visual field in the lateral geniculate nucleus of Macaca mulatta. J Comp Neurol 161:569-594, 1975

27.Harwerth RS, Carter-Dawson L, Shen F, Smith EL, Crawford MLJ: Ganglion cell losses underlying visual field defects from glaucoma. Invest Ophthalmol Vis Sci 40:2242-2250, 1999

28.Kerrigan-Baumrind LA, Quigley HA, Pease ME, Kerrigan DF, Mitchell RS: Number of ganglion cells in glaucoma eyes compared with threshold visual tests in the same persons. Invest Ophthalmol Vis Sci 41:741-748, 2000

29.Dacy DM: Primate retina: cell types, circuits, and color opponency. Prog Retinal Eye Res 18:737763, 1999

30.Frishman LJ: Basic visual processes. In: Goldstein EB (ed) Blackwell’s Handbook of Perception, pp 53-91, Oxford: Blackwell Publishers 2001

31.Harwerth RS, Crawford MLJ, Frishman LJ, Viswanathan S, Smith EL, Carter-Dawson L: Visual field defects and neural losses from experimental glaucoma. Prog Retinal Eye Res 21:91-125, 2002