Chapter 48 The Optic Nerve in Glaucoma DOUGLAS R. ANDERSON Table Of Contents |
THE NORMAL RETINA AND OPTIC NERVE PRESSURE-INDUCED DAMAGE VISUAL EFFECTS FUNDUSCOPIC FINDINGS CLINICAL EXAMINATION OF THE OPTIC NERVE REFERENCES |
THE NORMAL RETINA AND OPTIC NERVE | |
The approximately 1 to 1.5 million axons that form the optic nerve arise
from the retinal ganglion cells and course toward the optic disc in
a well-known pattern (Fig. 1). From the nasal retina the fibers take a straight course toward
the disc. Axons originating temporal to the fovea arc around the macula
to enter the upper and lower poles of the optic nerve head. The macula
fibers pass directly to the temporal quadrant of the disc in the papillomacular
bundle.
The axons maintain an orderly grouping in the retina as bundles partitioned by sheets of Müller's cells.1–3 A sheen reflected from the many bundles taking parallel courses gives the retina a striated appearance4 that is most prominent where the nerve fiber layer is thickest near the disc, especially in the arcuate bundles arriving at the poles of the disc from the temporal arcades.5,6 The attentive observer can see the striae of light reflected from these bundles and recognize their partial or total absence when bundles of axons disappear in glaucomatous or other forms of optic atrophy.4,7–10 The backscatter of light is determined by the cylindric nature and size of the light-scattering structures, possibly the axonal microtubules. They produce reflection with directional, spectral, and polarization (retardation) properties that have been studied and quantified, providing a basis for measuring the integrity of the retinal nerve fiber layer.11–14 The manner of layering of axons within intraretinal bundles2,3,6,15 may not be exactly the same in all primate species. However, a basic orderliness is present in the retina and seems to be maintained in the chiasm and beyond to the lateral geniculate body.16–19 Discrete scotomas and other defects caused by damage to nerve fiber bundles are produced by localized insults in the retinal nerve fiber layer or in the optic nerve head. Diffuse pressure on the posterior optic nerve by a mass lesion typically produces a central scotoma or diffuse visual dysfunction, but occasional instances of more localized damage in this region produce well-localized nerve fiber bundle defects, attesting to the existence of orderly bundles in this region also.12–15 On arriving at the lateral geniculate body, the axons are still organized and synapse in an orderly array. In the retinal nerve fiber layer, axons converge from every direction toward the optic disc and turn to enter the optic nerve through an opening in the outer retina, the choroid, and the sclera. The features and anatomic variation of the normal optic nerve head, or optic disc, are illustrated in Figures 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, and 16. A physiologic excavation (cup) results if the chorioscleral canal is larger than required for the approximately 1 to 1.5 million axons and the supporting glial cells and blood vessels. The size of the excavation depends on how ample is the size of the chorioscleral canal.20,21 In discs where the chorioscleral canal matches the number of axons, the chorioscleral canal itself is typically somewhat taller than it is wide (Figs. 2 and 3). However, because the number of nerve fibers entering the upper and lower poles of the disc is greater than in the temporal and nasal sectors, the boundary of the physiologic cup is more or less circular.22 When the disc is large, it may be nearly circular but the cup likewise nearly circular. However, except in discs of anomalous shape, the width of the rim of neuroretinal tissue is noted in normal, nonglaucomatous optic discs to be greatest in the inferior meridian followed by the superior meridian, and narrowest in the temporal quadrant.23 The shape of the physiologic cup is affected by the obliquity of the wall of the canal through the choroid and sclera. The wall of the cup is steep where the wall of the canal is angled outward.24 Where the canal has a wall perpendicular to the ocular coats the cup has a sloping wall (Fig. 4). The slope of the cup may vary from one sector to another, and there is considerable individual variation in the size and shape of the chorioscleral canal and therefore of the optic cup. In many eyes, the optic nerve exits through a canal the passes nasally and upward into the orbit, so that the upper nasal neuroretinal rim is bounded by a steep wall when viewed from the ophthalmoscopic viewpoint, while the lower temporal wall is of the cup is sloped, so that ophthalmoscopically the cup wall is, in essence, being viewed from within the cup rather than from above. In other discs, particularly those with a large cup, the walls are perpendicular to the scleral wall in all meridians. There is also typically no evident slope of the exit canal in eyes with only a small dimple of a cup, resulting from a near-perfect match between the size of the chorioscleral canal and the amount of neuroretinal tissue (axons and glia) that comprise the optic disc. The normal anatomic variations were noted and classified by Eschnig,25 and the glaucomatous changes to be discussed below are superimposed on the anatomic character present natively. The canal through which the optic nerve exits from the eye traverses several layers of peripapillary tissue: the outer retina, the pigment epithelium, Bruch's membrane, the choriocapillaris, the outer choroid, and the sclera. In nearly all human eyes, in all or some meridians, a flange of sclera extends as far forward as the pigment epithelium, forming something of a border for the canal,25,26 and this can be seen ophthalmoscopically as a scleral rim around the disc (Figs. 5, 6, and 7). With or without such an obvious flange, the several chorioretinal layers may not be perfectly aligned with one another. Perhaps most typically, as the optic nerve exits obliquely in an upper nasal direction, the lower temporal sector of the disc viewed ophthalmoscopically is seen to have a crescent of misaligned edges of the several chorioretinal layers, such that the outer choroid and sclera may be left uncovered by the choriocapillaris and pigment epithelium (Fig. 8). Because of the tilt, the scleral flange is not seen as a narrow band; instead, by ophthalmoscopy its inner surface is viewed obliquely as it merges with the optic nerve sheath. In some myopic eyes, the crescent is most extensive on the temporal side, and in other eyes the crescent may be directly inferior, or for that matter less frequently in almost any position. In yet other variations the crescent encompasses much or all of the entire circumference of the disc, with misalignment or retraction of the various chorioretinal layers. When these eyes have glaucoma and the thin or retracted choroid surrounds the disc entirely, the area of thin choroid and absent retinal pigment epithelium has been termed a glaucomatous halo. The deformed anatomy associated with the tilted exit canal or the peripapillary crescent may mark the sector of the disc that is most susceptible to injury from elevated intraocular pressure27; for example, the lower temporal sector of the disc is most frequently the first sector affected. |
PRESSURE-INDUCED DAMAGE |
In the past, some believed that the damage to the optic disc in glaucoma
occurred as the result of some sort of local sclerosis. Elevated intraocular
pressure was not thought to be the cause, but simply to be an
accompanying manifestation of a sclerotic process, that affected both
the trabecular meshwork and the optic disc. Seeming evidence for this
idea was a dissociation between the degree of cupping and the degree
of pressure elevation,28–30 which is now attributed to individual variation in susceptibility to damage
from intraocular pressure.31 A causative role of intraocular pressure is evident from cases of secondary glaucoma and from the fact that nerve damage is produced by elevated intraocular pressure in glaucoma produced experimentally in animals. Experienced clinicians have long agreed that progressive damage to the disc is halted when satisfactory lowering of intraocular pressure is achieved,32 and several recent clinical trials have now also shown more formally a cause-and-effect relationship between the level of intraocular pressure and damage.33–35 However, considerable individual variation is evident. Some individuals have intraocular pressure well above the normal range (ocular hypertension) for a long time without developing the anatomic and functional changes typical of glaucoma. Others, meanwhile, undergo progressive harm while the intraocular pressure is in the statistically normal range (normal-tension glaucoma), and the rate of harm is reduced with the intraocular pressure is lowered. Something makes the some optic nerves more susceptible to pressure-induced injury, and the gradation of susceptibility extends down into the normal range. Factors that determine the susceptibility of the optic nerve to pressure-induced injury must exist and vary in amount to explain the individual variation in susceptibility to pressure-induced injury. Although the occurrence of damage does correlate with the level of intraocular pressure (and with its correlate, the resistance to aqueous humor outflow), the correlation with age is equally strong.28,29 This indicates that some of the susceptibility factors are age-related and that they are as important as the intraocular pressure in determining the visual loss in glaucoma. That some of the factors are genetic is evident from relationships to family history and race. Genetic makeup may impact not only the tendency to elevated intraocular pressure, but also susceptibility to harm that depends on the level of intraocular pressure. Observed relationships of particular susceptibility to female gender, migraine and other manifestations of vascular dysregulation, and clinical correlates or manifestations of vascular occlusive disease suggest the types of physiologic processes that may be altered by altered by the level of intraocular pressure. Details of the pathophysiologic mechanism by which intraocular pressure produces optic nerve damage, exactly what the participating causative factors are, and how they interact together remain largely unknown.36 However, some clues are emerging. In some cases, inability to regulate blood flow in the optic nerve (dysregulation or inadequate autoregulation, which may be associated with migraine, with vascular overreaction to cold or stress, or with low blood pressure) is postulated by many37,38 to limit the ability of blood vessels in the optic nerve head to maintain sufficient blood flow when the intraocular pressure rises and challenges the circulation by raising venous pressure. It has been shown that individuals vary in their ability to regulate blood flow in the optic nerve,39 and potential mechanisms for local blood flow control by optic nerve capillaries have been studied in the laboratory.40 In other cases, vascular occlusive disease seems to play a role. If ischemia is a major cause for optic nerve damage in glaucoma, the different clinical presentation and appearance of a glaucomatous disc from that of a major acute vascular occlusive event (acute anterior ischemic optic neuropathy) remains a puzzle. The are other events in the pathogenic pathways. Pressure seems to affect the astroglia, which express inducible nitric oxide synthase when the pressure is high,41 which on one hand might enhance vasodilation, but can also cause cytotoxic levels of nitric oxide. Axonal transport is impaired in the region of the lamina cribrosa, perhaps as a result of local ischemia,42 which may be reversed in time.43 With more serious or prolonged ischemia, damage to the axon membrane and entry of calcium may stimulate calcium-dependent proteases with irreversible damage to the axon segment would be expected. Any of these local events may prohibit transport of trophic factors, in particular nerve growth factor, to the retinal ganglion cells. In turn, after a time of inadequate arrival of growth factors, apoptosis is triggered with eventual death of the ganglion cell and its entire axon.44 Once triggered, apoptosis may occur even if the axon ischemia and blockage was reversed. It may be speculated that individual variation in the degree and duration of ischemia, free radical damage, and time of growth factor depravation before apoptosis is triggered might account for individual variation in susceptibility and in the rate of glaucomatous damage. Release of glutamate by dying ganglion cells has been postulated to add to the damage through excitotoxicity of adjacent retinal ganglion cells.45 While an interesting possibility, it remains uncertain whether excitotoxicity plays a major role in the pathogenic process. Especially in cases in which the intraocular pressure is not particularly high, the collapse and stretching of the lamina cribrosa and other connective tissue in the region remains unexplained but is the histologic correlate to the clinically different appearance of the glaucomatous disc from other forms of optic atrophy (simple loss of axons). Recent evidence46 that a thin cornea places an eye at higher risk of damage when carrying an elevated intraocular pressure has suggested to some that the character of connective tissue in these individuals affects vulnerability, but there remains debate about whether the thin cornea simply produces an erroneous pressure measurement that masks the magnitude of the elevation of intraocular pressure. Individuals vary in the rate of damage and in the level of intraocular pressure that is harmful, but perhaps also in the details of which of several root causes might participate or dominate in an individual, and which consequential events in the pathogenic sequence might be influenced by the intraocular pressure. Young individuals with secondary glaucoma and substantially elevated intraocular pressure develop cupping that is in part a diffuse stretching of the supporting tissues, and the elastic stretching may be reversible, while axon loss is not reversible. Older individuals with normal or slightly elevated intraocular pressure may develop shallower cupping with pallor thought to be related to chronic ischemia, which may involve all sectors of the disc or be quite localized. Cases may vary further depending on whether an occlusive disease or regulatory dysfunction participates in the postulated ischemic component, with the level of intraocular pressure possibly being more relevant in nonocclusive cases.47 There is more to learn about the entire pathogenic process, how to identify the dominant participating pathogenic processes in a given individual, and the pathways that are common to all cases that produce the typical clinical features of chronic glaucoma. Such information can be used in part to assess individual risk. Moreover, each of the contributing causes and participating pathophysiologic pathways involved in all cases or in particular identified individuals may be amenable to treatment that will improve the visual outcome. We wait for the day when we will know more about the pathogenic process, know what the various susceptibility factors are, are able to measure them for the purpose of predicting how much pressure the eyes of a given person can tolerate, and can attack the susceptibility factors therapeutically. Until then, we know that a particular person is highly susceptible to future glaucomatous damage mainly when we observe that the pathogenic process is active or that the damage has begun,31 for example noting splinter optic disc hemorrhages or early glaucomatous anatomic or psychophysical (visual field) abnormality. The level of intraocular pressure is a weak predictor, but remains the focus of attention mainly because it is the only risk factor participating directly in the pathogenic mechanism that we know both how to measure and to treat. The susceptibility factors seem in large part to be in the constitution of the person, perhaps the result of systemic influences (e.g., vascular status) or the fact that the two eyes of a person possess similar anatomic and physiologic traits on a genetic basis. Thus, if one eye has suffered damage from elevated intraocular pressure, the other eye is likely soon to follow if it too has or develops a pressure elevation.48 The two eyes may not always be damaged to the same degree. When one eye carries a somewhat higher pressure, it is usually the more damaged eye.49,50 Occasionally, a unilateral carotid artery obstruction, an anatomical difference between the two eyes (e.g., marked anisometropia) or some unidentified factor seems to make one eye more susceptible than the other, even with symmetric intraocular pressures. It appears that the susceptibility factors are absent, or present only to a low degree, in the majority of persons, especially the young. Therefore, many people can carry a moderately elevated intraocular pressure for many years without damage, but there is a limit, and few eyes will tolerate a pressure of 50 or 60 mm Hg for very long. Approximately half of eyes have suffered harm when discovered to have an intraocular pressure of 35 mm Hg.30 It can be presumed that susceptibility factors are stronger in persons who suffer damage from mildly abnormal pressures (in the low and middle 20s) or in the normal range than in the majority of eyes, which typically tolerate such modest pressures easily for a long time. A few persons are exceptionally susceptible to damage and they suffer damage from modest pressure elevations, or in the high teens (low-tension glaucoma, or more accurately, normal-tension glaucoma). Damage while the pressure is in the low teens occurs even more rarely. Although it is sometimes thought surprising that a pressure in the normal range can be damaging, it is not so surprising if it is kept in mind that 18 mm Hg is, as an absolute pressure level, nearly 80% of 23 mm Hg, which is readily accepted as potentially damaging to the susceptible person. If susceptibility factors lower the resistance of the disc to the point that 23 mm Hg is damaging, it is not a much further step to make the disc sensitive to 18 mm Hg. This reasoning suggests that it is arbitrary to consider low-tension glaucoma as being different than ordinary chronic open-angle glaucoma. The distinction that is made in traditional diagnostic classification results from fallaciously equating abnormal (i.e., a pressure level that is infrequent) with unsafe (i.e., a pressure that is harmful). Indeed, there is growing sentiment that the basic nature of the disease is the same when the intraocular pressure is normal as when it is abnormally high, so that primary open angle glaucoma should no longer be distinguished from normal-tension glaucoma. While it is important to clarify our thinking by considering these diseases to have the same underlying mechanisms, there may be cases of glaucomatous damage that occurs without the intraocular pressure playing a part (and these would mainly have normal intraocular pressures), and in any event, cases that occur with normal intraocular pressure must have particularly active participation of the susceptibility factors. Noting that a case if glaucoma has a normal intraocular pressure guides the clinician to recognize that these susceptibility factors are more dominant in this patient than the intraocular pressure and one day may guide that clinician to focus attention on treatment directed at these factors in these cases and on the intraocular pressure in those cases with particularly elevation intraocular pressure. In typical cases of chronic glaucoma, the damage occurs over a prolonged period of time. After the most susceptible axons are already damaged, it may be speculated that damage to other, previously unaffected axons may occur as a result of one or more of the following: (1) age causes a reduction in resistance to damage so that the next most susceptible axons become susceptible to the pressure that is being carried; (2) weakening of the disc by partial cupping increases the susceptibility of the remaining axons; (3) intraocular pressure progressively rises or has higher peaks of pressure as glaucoma continues; (4) damage depends not only on the pressure level but also on the time of exposure to the pressure (thus, the damage occurs as a cumulative effect of the force exacted by the intraocular pressure); or simply (5) somewhat arbitrarily a certain number of axons (or proportion of remaining axons) suffer during numerous small episodes, with cumulative damage. Sometimes, however, in susceptible individuals glaucomatous cupping of the disc can proceed rapidly, within a month or two, even with moderate pressure elevation (25 to 30 mm Hg) if the pressure rose to this new level rather suddenly and exceeds the susceptibility level of all the axons all at once. This situation is encountered with a sudden (but perhaps mild) pressure rise in secondary glaucomas. Many cases may suffer no harm from a sudden pressure rise to 35 mm Hg from trauma or uveitis, for example, but a few eyes will suffer rapid, nearly total cupping in just a week or two with lesser elevation of pressure, say to the mid or upper 20s. Therefore, cases with recent onset of pressure rise must be monitored closely until it is clear whether this eye is being damaged or not. In some cases of chronic glaucoma the cupping and field loss are recognized to progress in a series of small episodes a year or two apart, perhaps as a result of episodic upward swings of intraocular pressure or episodic swings of the constitutional factors that decrease the tolerance level. Even in cases that seem to be a continuous slow loss of axons, there may be smaller more frequent episodes that affect a few axons or bundles every week or two, but are each too small to be perceived as separate events. A single dramatic episode of transient vulnerability is sometimes blamed for nerve damage and field loss in patients who have had a hemodynamic crisis. Patients with such a history are less likely than others to progress51 if they have returned to good health and episodes of ill health are not recurrent. Progression may be accompanied by the occurrence of splinter hemorrhages that appear transiently on the optic disc.52–57 Although observed only occasionally, such hemorrhages may occur between ophthalmoscopic examinations in the majority of untreated patients and in patients whose glaucoma is progressing despite treatment. They are seen most often in cases of normal-tension glaucoma, perhaps because they most likely have an abundance of susceptibility factors with intraocular pressure that has not been lowered sufficiently to halt damage. Often one region of the disc is affected earlier and more severely than another. When axon loss occurs more predominantly in certain bundles, cupping characteristically extends toward the disc rim in the sector that has lost neural tissue. If the preferential loss is in the most typical location (in the upper and lower sectors, most characteristically just temporal to the inferior pole), the cup expands vertically,22 and may form a notch at the disc rim, for example, at the inferior the pole of the optic disc. In such typical cases, the loss of axons and ganglion cells predominates in the corresponding arcuate regions of the temporal retina. In contrast, when axon loss is evenly diffuse, the cup expands concentrically,58,59 and there is an accompanying generalized depression of the field. In the middle of the spectrum, where the majority of cases lie, there may be some widespread diffuse loss of axons but more severely near the pole of the disc. Thus, as the cup extends and deepens vertically, there is also temporal and nasal unfolding of the cup. |
VISUAL EFFECTS |
Loss of axons naturally affects the visual function in the regions of the retina from which the axons arose. Color sense,60–63 contrast sensitivity,63–66 and acuity67,68 are among the functions lost, in addition to the differential intensity visual threshold typically measured in classic visual-field testing.69–75 A certain proportion of nerve fibers must be lost in the affected retinal region before the diminished visual function is recognizably reduced from the normal range, from the previously documented visual status, or from the threshold in the surrounding unaffected (or less affected) region.76 For example, a 50% loss of axons in a bundle serving a given retinal region may produce a 0.5 log-unit (5 dB) reduction in visual sensitivity. If one or several adjacent bundles lose 50% or more of their axons, a scotoma results in which visual sensation in the corresponding region of the field will be recognizably reduced compared to that in the surrounding regions. A localized scotoma may be recognized because of major loss of axons in a confined region while no axons are lost in 95% of the disc area. However, if the axon loss is more diffuse, the resulting mild widespread depression may remain for quite a while within the range of normal visual sensation. Such diffuse axon loss is more difficult to recognize as an acquired reduction of visual sensation,58,63,66,71,77–79 unless comparison is made either with a previously documented baseline in that person or with the opposite, less affected eye. Fortunately, bilaterally equal diffuse loss of nerve fibers with bilaterally equal cupping and loss of visual field is infrequent. Even when axon loss is diffuse, one region is usually more affected than another and one eye more than the other. Therefore, glaucomatous nerve damage and visual-field loss can usually be recognized at an acceptably early stage by fundus examination and automated perimetry by80: (1) characteristic localized cupping and field loss, (2) by asymmetry of field in the upper and lower positions, (3) by asymmetry of cupping or field loss between the two eyes that is in keeping with asymmetry of intraocular pressure, or (4) by a change from a previous baseline disc photograph or baseline threshold perimetry. The conclusion that an acquired or asymmetric generalized depression of visual field is glaucomatous must be in keeping with other findings suggestive of glaucoma and out of keeping with any other causes of visual depression that might be present, such as amblyopia, anisometropia, alteration in media clarity, or indeed simply deviant ocular anatomy or function that may accompany high refractive error. |
FUNDUSCOPIC FINDINGS | |||||||||
When evaluating the optic disc clinically, it is helpful to have a mental
image of the underlying tissue changes. These are twofold: altered
configuration of the lamina cribrosa and other connective tissue, and
loss of neuroretinal tissue. Connective tissue changes consist of elastic stretching caused by the force of the intraocular pressure and of connective tissue disruption with collapse of the layered sheets of the lamina cribrosa. Reversibility of cupping sometimes seen when the intraocular pressure is lowered if there remains an elastic element to the stretching backward bowing of the lamina cribrosa and outward stretching of the scleral border of the optic nerve head, especially in the young. This reversibility should not be construed as axon recovery or regeneration. Much of the connective tissue disruption and collapse is irreversible and produces a permanent component to the excavation, in addition to the loss of axons and perhaps of some astroglia. It seems that the change in the lamina cribrosa and other connective tissue that accounts for the fact that glaucomatous atrophy of the optic disc differs in clinical and histologic appearance from that of other optic atrophies. When there is considerable backward bowing of the lamina cribrosa without much axon loss, the disc may be severely cupped with surprisingly little visual effect detected on clinical examination. Adding to the excavation caused by backward bowing and collapse of the lamina cribrosa is the loss of axons, as well as reduction or displacement of the astroglia and microvasculature within the optic nerve head. This neuroretinal tissue diminishes in the rim that surrounds any physiologic excavation that may have been present. As the supporting tissue recedes and the tissue bulk diminishes, the surface of the disc and the surface of its cup collapse. The surface of the neuroretinal rim recedes backward, but may not be conspicuous because the adjacent inner retinal surface also recedes backward as the nerve fiber layer of the retina also becomes thin. In patients in whom backward movement of the disc surface is the earliest sign recognized, the disc has sometimes been described as “saucerized.”81 Easier to see but difficult to describe and quantify is the excavation, or cup, in the center of the optic disc. It is a three-dimensional cavity that can be appreciated in stereoscopic photographs, binocular views clinically, and more modern imaging methods. Although three-dimensional aspects can be quantified, clinicians struggle to describe, quantify, or communicate the qualities of the cup in two-dimensional terms, such as having a diameter (cup/disc ratio) or area in a two-dimensional monocular ophthalmoscopic or photographic view, or a two-dimensional shape, such as round or oval. The reference plane is often by necessity perpendicular to the line of sight of the photograph or biomicroscopic view, although some make an effort to describe the cup/disc ratio or shape in a plane perpendicular to the axis of the nerve as it exits. Conventionally, the cup diameter is sometimes recorded at the plane of the retinal surface but as more sophisticated technology arose, other useful reference planes have been defined, and there is no universal standard or definition of how the size and shape of a cup should be recorded and communicated. While waiting for standardized ways to quantify the disc configuration, clinicians do make an effort to describe and record the extent of the excavation. There is considerable variation in the appearance of glaucomatous cupping (Figs. 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, and 32), sometimes depending on the physiologic configuration before the onset of glaucomatous disease. The lateral expansion of the cup is easier to appreciate than the backward recession of the surface, and it is sensible to gauge progression of glaucomatous cupping as an increase in the cup diameter. Where the wall of the cup is steep, a cup margin is easy to delineate and consensus is easy to reach about enlargement of the cup when tissue is being lost is such a sector. However, in any sector(s) in which the cup has a sloping surface, it may be difficult or meaningless to define a cup margin with two-dimensional parameters, making it impossible to quantify in a uniform manner what is meant by a cup radius or cup/disc ratio for that meridian.82 One can conceive of the reference plane either as concentric with the ocular coats or as perpendicular to the axis of the exiting optic nerve, perhaps achieving reasonably reproducible assignment of a cup/disc ratio within a group that has reached a pragmatic consensus in its terminology.83
To be clear, the purpose of noting the quantified or qualitative features of the disc in clinical practice is twofold with reference to glaucoma: to recognize whether the disc is damaged and to recognize whether there has been further damage since a previous examination. For the first purpose it is important to be familiar with the variations in appearance of normal optic discs (Figs. 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, and 16), and how each of these may change as damage occurs (Figs. 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, and 32). With experience, presence of damage may be recognized by a single examination. In the next paragraphs the features to note will be described, but there is room for improvement in the ability to recognize abnormality accurately and efficiently in clinical practice. An advantage of automated image analysis is that it is less dependent on the experience and expertise of the individual clinician. For the second purpose (determining change), a means to record the state of the disc on the baseline occasion is needed. Drawings are perhaps used by most clinicians, but depend on the observational method used (direct ophthalmoscopy, binocular indirect ophthalmoscopy, biomicroscopy) as well as the talent of the artist-clinician to represent a three dimensional configuration on a two-dimensional drawing. High-magnification stereoscopic photographs are often a better record of the disc's state, but too often after years pass, progression is still difficult to judge if a different camera is used, exposure, magnification, and degree of stereopsis is not the same, and the development or removal of cataract affects the image. Sometimes, too, in regions of the disc without vessels or surface texture, the tissue surface may not be easily appreciated even in stereoscopic photographs. Images created by other methods (for example scanning ophthalmoscopy) are sometimes subject to the same lack of standardization or reproducibility. Both recognition of the presence of damage and the recognition of change are particularly problematic in one frequent disc configuration, that with a somewhat tilted exit of the nerve upward and nasally (Figs. 8 and 14, 15, and 16). In these discs, the cup surface slopes in the lower temporal quadrant, which is often also devoid of major vessels to help localize the surface. In the ophthalmoscopic view, this surface simply recedes away from the observer and loss of tissue is difficult to identify, as the tissue simply becomes thinner and paler. Only in the more advanced stages might the cupping be recognized by the formation of a distinct notch at the inferior pole at one edge of the lower temporal sector. It is particularly unfortunate that this region of the disc is so typically the earliest and most involved, corresponding to the upper hemifield visual field abnormality. Moreover, in this sector it is difficult to evaluate the thickness of the layer of retinal nerve fibers crossing the disc margin.82 These problematic features are most difficult in the disc of with high axial myopia, because the sclera simply curves backward to merge with vaginal sheaths of the optic nerve, the choroid gradually thins within the β zone as it approaches the nerve exit, and it becomes arbitrary to designate any location as the margin of the disc itself. Fortunately, the difficulties and limitations just emphasized are problematic only in some cases, both in describing and quantifying the features of a normal and abnormal optic disc, but we can nonetheless make useful notations of the characteristics of the optic disc configuration. Efforts to observe and record these features are in most cases very useful for the diagnosis of glaucoma and for monitoring its progress. The area (or diameter) of pallor may be described separately from the size of the cup.84 As a descriptive matter, the region of pallor represents the base of the cup, and its diameter is judged there, while the cup diameter is conventionally judged at or near the inner retinal surface. It is not typical of glaucoma for the tissue of the neuroretinal rim to be pale in glaucoma, but for the tissue to be absent in the affected part of the neuroretinal rim. If there is any rim tissue present with a sector of pallor, nonglaucomatous atrophy must be suspected (except at times as a residual of an acute attack of glaucoma). Glaucomatous loss of axons and thinning of the neuroretinal tissue may be diffuse or localized. If the axons are lost diffusely, the excavation may enlarge concentrically,58,85,86 retaining its round outline, and may be difficult to distinguish from a physiologic configuration.87 This is completely analogous to the difficulty in recognizing diffuse axon loss by visual field testing. The pathologic nature of the enlarged excavation can be recognized if it can be compared to the previous status of the disc58 (hence the importance of careful documentation of the optic disc appearance in all patients with elevated intraocular pressure) or if the degree of enlargement is different in the two eyes59 (acquired enlargement may be asymmetric but the physiologic cups are nearly always the same in the two eyes,88–96 unless there is anisometropia or other congenital asymmetry). Sometimes the experienced observer can pinpoint the physiologic nature of the large excavation by recognizing that the diameter of the scleral canal is unusually large in comparison with the diameter of the emerging central retinal vessels (Figs. 11 to 13). Large, round scleral canals, as opposed to smaller canals that are taller than they are wide, seem particularly common in black patients, and in such discs the branches of the central retinal vessels may dive through the pink neuroretinal rim rather than pass along its surface into the large physiologic cup. In the absence of such clues, and in any event if the intraocular pressure is elevated, an excavation larger than 0.6 cup-to-disc ratio is usually considered suspicious and should observed carefully until time proves that it is not enlarging. Typically, before the cupping proceeds very far, preferential loss of neuroretinal tissue becomes evident, usually near the poles of the disc. The damage can be recognized either by noting that the rim is thinner than it should be compared to other sectors, or by the fact that the cup has become distinctly oval in the vertical direction. When the disc (and cup) is large enough to judge the breadth of the rim, a normal disc is has the broadest rim in the inferior meridian, followed in order by the superior, nasal, and temporal rim23 (Figs. 11 and 12). If the rim is not thickest at the two poles of the disc and thinnest temporally, glaucomatous damage should be strongly suspected (Fig. 13). The neuroretinal tissue becomes thinner in the affected sector such that the cup expands toward the lower temporal or upper temporal margins of the disc, assuming a vertically oval outline.22,97 Recognition of thinning of the neuroretinal rim near the vertical poles of the disc, with corresponding enlargement of the cup in a vertical direction, is the single most effective way to recognize the presence of glaucomatous damage.98,99 Depending on the configuration of the disc, the loss of tissue in the vertical sections may be most easily recognized as thinning of the rim, by a vertical extension of the cup or by a vertical shape to the region of pallor representing the floor of the cup. Except when the disc is obviously anomalous, any vertical tendency of these sorts deserves attention. For example, when the disc itself is small, and prior to the onset of glaucoma was completely full of neuroretinal tissue (essentially no cup except for a central dimple), the onset of glaucomatous tissue loss can be deceivingly small and unapparent.100 Therefore, in a noticeably small disc, any cup is suspect, but particularly if the cup is recognized to be a thin vertical slit with a vertically oriented pale base, it should not be passed over as nonglaucomatous. Such cases often seem to have field loss or retinal nerve fiber loss become evident before the glaucomatous nature of the cupping is recognized. The strong preference for early involvement of the vertical sectors notwithstanding, the localized loss of tissue may sometimes occur in any sector of the disc, even nasally (associated with a temporal wedge field defect), and a cup with extension localized toward any point on the disc circumference is decidedly abnormal. Localized tissue damage is typically associated with a nerve fiber bundle defect in the appropriate portion of the visual field. In myopic eyes with the disc tilted and a crescent over the temporal sector, thinning of the temporal rim can be glaucomatous, and the field defects are more often closer to the point of fixation than in other eyes. Evaluation of discs in high axial myopia is often made even more difficult because of an enlarged anomalous disc with an irregular shape embedded in a staphyloma, making the edge of the disc difficult to define, and loss of nerve fibers nearly impossible to recognize with certainty. These classic appearances of glaucomatous cupping at various stages and in discs of various anatomic configurations has been described.22,85,86,97,101–112 Representative examples are shown in Figures 13 and 19 to 32. Glaucomatous cupping is recognized not only on the basis of the illustrated typical preferential loss of tissue at the poles of the disc but importantly by also looking for asymmetry in the cupping of the optic discs of the two eyes (which are normally the same88–96) or a change from the previous status (hence the importance of documenting carefully the disc status of all glaucoma suspects, preferably with photographs). Especially without the advantage of prior photographs, the degree of damage must take into account the size of the disc,113 and clues about its preglaucomatous configuration, paying more attention to any cupping in small discs100 and recognizing that large cups are expected physiologically in large discs. In nearly all cases of nonglaucomatous optic atrophy, the affected tissue turns pale and the loss of mass is inconspicuous. Rare exceptions occur in moderately large discs with cylindric cups in which glaucoma-like thinning of the rim is noticeable after arteritic anterior ischemic optic neuropathy114,115 (nonarteritic forms tend to occur in small discs) and very rarely in other optic atrophies. Misalignment and irregularity of the retinal pigment epithelium, choriocapillaris, choroid, and sclera at the disc margin (or partial atrophy of some of these tissue layers) around part or all of the disc circumference gives the appearance of a peripapillary crescent or halo.27,116–128 A crescent is often seen in the sector of the disc that shows the most profound cupping.27,124,126 These regions of disrupted or thin peripapillary tissue have been descriptively divided into α regions of clumped or irregular retinal pigment epithelium and a β region, closer to the disc, in which the choroidal and pigment epithelial tissue are thin with a pale appearance. At the very edge of the disc is typically a thin white rim usually perceptible around the entire circumference, representing a flange of sclera that separates the disc from the choroid in humans. Advanced cupping in older people with low grade elevation of intraocular pressure is also seen in the company of a prominent peripapillary halo (surrounding the entire disc instead of only a portion of the circumference) with β-type thinning of choroidal tissue, but the association is not universal. One form of cupping in normal-tension glaucoma is rather shallow cupping with a prominent halo of thin pigment epithelium and pale choroidal tissue. Another characteristic form in normal-tension glaucoma is a narrow localized crescent at the inferior pole associated with a discrete notch in the neuroretinal rim in which a splinter hemorrhage is sometimes seen. It is not clear how many of these crescents and halos represent congenital preexisting misalignment of the layers, how many are age-related changes that may occur before the glaucoma develops, and how many are atrophy of the peripapillary tissue that accompanies the insult to the disc. Certainly examples of each can be found. The congenital variety has been postulated to mark a sector of the disc that later is more vulnerable to glaucomatous damage, accounting for field defects close to fixation in cases of temporal myopic crescents and inferior notches of the disc. In keeping with this, it has been postulated that individuals without any peripapillary crescent may be less susceptible to damaging effects from moderate ocular hypertension.129,130 Enlargement of the β zone has been seen with age in nonglaucomatous patients, but is more frequent in those with glaucoma associated with elevated pressure, often older people with mild elevation of intraocular pressure.121 Younger individuals with more marked elevation of intraocular pressure (often secondary glaucoma), however, typically develop deep glaucomatous cupping with loss of all nerve fibers, but no development or enlargement of the peripapillary tissue disturbance.131,132 Transient splinter hemorrhages at the disc margin52–57 (Figs. 31 and 32) can be seen to occur in most cases of untreated glaucoma if observations are made often enough.56 When they are seen in treated cases, it usually signifies that cupping and field loss are progressing,53,54 and sometimes the hemorrhage is seen at the same time that a new field defect appears. Because hemorrhages are easy to overlook and transient, fundus examinations must be both careful and frequent if these hemorrhages are to be detected. Along with loss of inner retinal nerve fiber layer thickness and disc tissue, the retinal arteries become narrower, presumably because of reduced metabolic need, and the narrowing is often most marked near the disc.133 This narrowing is nonspecific for glaucoma but occurs with all forms of optic atrophy. Given the anatomic variation in physiologic configuration of the optic nerve head, the varying levels of intraocular pressure that may develop, the individual variation in the level of intraocular pressure that is harmful, and the varying age at which glaucoma may appear, there is a variety of appearances of glaucomatous discs. In younger patients, often with secondary glaucoma and considerable pressure elevation, reversible elastic expansion of support tissue produces a symmetric expansion of the physiologic cup (deep round cup surrounded by a donut-like rim of neuroretinal tissue), which may later develop a diffuse loss of the neuroretinal rim until there is total cupping, or a localized loss to produce a notching the rim that later also becomes thinner as well. In such cases, there is no accompanying atrophy of the peripapillary choroid and retinal pigment epithelium. In somewhat older patients with more moderate idiopathic (primary) pressure elevation, the cupping is shallower and less likely to show general reversible enlargement of the cup before localized excavation is evident. A tilted disc with a sloping surface of the inferotemporal neuroretinal rim and an adjacent crescent is a common physiologic configuration (or more directly temporal in somewhat myopic eyes), and these sectors seem most often to suffer the earliest and most severe damage. These varieties represent the most common patients, and hence the most typical groups, as represented in the figures in this chapter. In some, typically older, patients the excavation is shallow with little or no elevation of intraocular pressure. Peripapillary zones of depigmented or absent pigment epithelium (as a crescent or halo) are more common in older patients who develop damage with little or no pressure elevation, as are splinter hemorrhages. Despite these generalizations, there is considerable variation in the combination and overlap of these features, so that it is difficult to separate the varieties of cupping into distinct groups.134 In infantile glaucoma, as the entire globe enlarges, a concentric enlargement of the cup results when the scleral canal is stretched along with the rest of the globe. The lamina cribrosa of the glaucomatous infant eye may also be elastically bulged backwards. This element of the acquired cupping in infants is not the result of axonal loss. Relaxation of the stretched tissue when the intraocular pressure is relieved may account for the reversibility of cupping often observed in these infant eyes104,135–145 (and occasionally observed to a lesser degree in other glaucomatous eyes in later decades of life146). The cupping caused by scleral stretching may not be completely reversible because with remodeling of collagen in stretched tissues still undergoing development and growth, some of the enlargement of the eye (and of the scleral canal) is permanent. Thus, after successful pressure control, the eye in infantile glaucoma may remain myopic with a large cup without any axon loss. Of course the pressure can also damage the axons, often preferentially at the poles of the disc as in adult glaucoma. This element of cupping is not reversible and is accompanied by visual loss. In acute angle-closure glaucoma (Fig. 33), the optic disc may swell,147–149 perhaps as a result of ischemia, but because of corneal edema and preoccupation with the angle status, it is not often seen. Even after an attack lasting several days, there is often no visual loss; the disc may remain normal appearing or may develop some degree of pallor resembling that of nonglaucomatous optic atrophy.150–152 Excavation of the disc does not seem to result from an acute attack but may result from persistent residual pressure elevation after the attack. If excavation is already present at the time that a patient presents with symptomatic angle closure, the cupping is evidence that for some time an asymptomatic, perhaps gradual, rise of intraocular pressure elevation preceded the abrupt onset of symptoms.
Usually all other forms of optic nerve disease (and inner retinal disease) produce simple atropy of the disc without cupping (Figs. 34 to 36), but occasional cases of cupping have been seen with anterior ischemic optic neuropathy, particularly due to giant cell arteritis (Figs. 37 and 38).
|
CLINICAL EXAMINATION OF THE OPTIC NERVE | |
For glaucoma detection, an eye examination needs to include an examination
of the optic disc. This supplements tonometry for glaucoma detection
by revealing cases of low-tension glaucoma and patients with
variable pressure who happen to have a normal pressure at the time of
a routine eye examination. Visual-field examination is appropriate
when the pressure is elevated (even if the disc appears healthy) or
when the disc is suspiciously abnormal (even if the
pressure is normal). On the basis of family history, the intraocular pressure, the disc configuration, or the visual field, a patient may be diagnosed as having glaucoma, may be suspected possibly as having early glaucoma, or may be judged at risk of developing glaucoma. In any of these situations, it is wise to record the status of the discs with careful drawings or preferably (Figs. 18, 19, 39, and 40), if available, stereoscopic disc photographs. In addition, it is wise to record a carefully quantified visual-field examination (kinetic perimetry with several isopters or automated perimetry with a thresholding strategy) as a baseline. Visual acuity and intraocular pressure are measured on every visit thereafter, the optic disc may be examined without necessarily dilating the pupil on every visit, but also with a dilated pupil at regular intervals. Frequent disc examination will increase the chance of observing transient disc hemorrhages in patients in whom glaucoma is progressing.56 If the patient's course seems stable and if the intraocular pressure is not different than usual, the optic disc may be examined through a dilated pupil on approximately every third or fourth visit. A visual-field examination may be conducted during one of the intervening visits, leaving one intervening visit for a refraction repeat gonioscopy, or whatever else might be required for the patient's care. An office visit every 3 months has been traditional for patients with glaucoma and for patient suspicious for glaucoma who are thought to be stable but is not always appropriate.153 In patients whose stable course has been well documented for a year or two, the interval can be lengthened to 4 months, and in selected patients to 6 months or even longer. On the other hand, patients about whom there is reason for concern (e.g., the pressure is not substantially lowered from the pretherapeutic level, a regular visit shows suspicious change in the disc or field, or a new therapy has been instituted) may need not only to be seen more frequently for a while, but may also require disc and field examinations on every visit until it is either evident they are progressing or evident that they are stable. Clinical examination of the optic disc is best done through a dilated pupil with the magnified stereoscopic view of slit-lamp biomicroscopy. The angled slit beam reveals the contour of the disc's surface, but the slit is helpful even when placed directly in line with the biomicroscope. As it intersects the surface, a thin, well-focused beam produces a discrete line, which the examiner views with stereopsis. The Hruby (approximately 55 diopter) lens is convenient for routine use and usually will give an adequate view, but a funduscopic contact lens (approximately 64 diopters) should be used in persons whose discs are not seen clearly with the Hruby lens. Apart from the additional inconvenience, a disadvantage of using the contact funduscopic lens routinely on all patients is that the anesthetic eye drops and gonioscopic solution render the cornea less suitable for obtaining a high-quality fundus photograph. The use of a +78 diopter or +90 diopter indirect lens in conjunction with a slit-lamp biomicroscope is becoming increasingly popular. It must be kept in mind that with less magnification (with a stronger power lens) appreciation of depth is reduced because axial magnification (depth) relates to the square of transverse magnification (diameter of the disc in the image). Stereoscopic color disc photographs (taken best at twice the usual magnification) lack the advantage of having a slit to highlight the surface in areas of homogeneous texture and are thus not a substitute for biomicroscopic examination. However, good stereoscopic fundus photographs are best for recording the status of the disc for the sake of future comparison. The monocular magnified view of direct (or monocular indirect) ophthalmoscopy is the next best choice for viewing the disc and sometimes the only method that can be used with an undilatable pupil. Although the fundus as a whole should always be examined with a binocular indirect ophthalmoscope during a complete eye examination, this instrument is the least suitable for judging glaucomatous excavation of the disc. The strong illumination seems to render a pink color even to an atrophic disc, and with the usual +20 diopter or +30 diopter lens, the stereopsis and magnification without the biomicroscope are not sufficient for the examiner to appreciate saucerized or sometimes even frankly excavated discs. When bundles of nerve fibers drop out, their absence can be noted in the retinal sheen produced by the retinal nerve fiber layer.4,8–10,85 Such defects are easiest to see in the thickest portions of the nerve fiber layer, namely, close to the disc and especially in the arcuate bundles approaching the poles of the disc.5,6 The loss of nerve fibers can be recognized ophthalmoscopically but, as shown in Figure 41, is demonstrated most beautifully in wide-angle fundus photographs taken with blue or green filters.10,154 It is difficult to see if the background is lightly pigmented or if imperfect media produce an imperfect view of the fundus. Some observers are more skilled than others in recognizing nerve-fiber layer disease. Fluorescein angiography has shown areas of slow filling or nonfilling in damaged segments of the disc,155–159 evidence that the microvasculature is lost along with the other elements of the neuroretinal tissue. Fluorescein angiography has not been used for routine clinical evaluation, but in cases of uncertainty about the existence of pressure-induced disc damage, the demonstration of a filling defect could potentially be of value in recognizing neuroretinal tissue loss in discs with very sloped margins, such as in high axial myopia.158 Emerging technologies to record and quantify the optic nerve configuration and the retinal nerve fiber layer are presently used in some offices. The techniques are undergoing improvement. Criteria for abnormality and for progression of damage are being validated. These are described in Chapter 48A. |