Clinical background
Glaucoma is the leading cause of irreversible blindness and is estimated to affect approximately 67 million people worldwide. The pathological correlate of disease is the loss of retinal ganglion cells (RGCs) and their optic nerve axons. Glaucoma is a silent, slowly progressive disease that causes irreversible vision loss. Elevated intraocular pressure (IOP) is a major risk factor. Other risk factors include family history of glaucoma, black race, increasing age, myopia, and abnormal blood pressure.
Key components of the clinical assessment for glaucoma include measurement of IOP, and examination of the optic nerve head. The diagnosis is made based on visualizing characteristic patterns of damage to the optic nerve head. Features of glaucomatous optic neuropathy include evidence of retinal nerve fiber layer loss, excavation and cupping of the nerve head, focal or diffuse loss of the neuroretinal rim, and disc hemorrhage ( Box 26.1 ). These findings correlate with visual field deficits in a retinotopic fashion when vision loss is present. A complete eye assessment that includes the anterior-chamber angle helps to determine whether these changes are secondary to specific etiologies such as angle closure glaucoma, or whether the entity is primary open-angle glaucoma in which no abnormality is detected other than optic nerve head pathology and possibly elevated IOP.
- •
Glaucomatous optic neuropathy shows characteristic patterns of optic nerve damage
- •
Elevated intraocular pressure is a major risk factor for the development of glaucoma
- •
All current therapies for glaucoma aim to lower intraocular pressure
- •
Many patients continue to lose vision despite treatments to lower pressure
All treatments for glaucoma are based on lowering IOP by pharmacological agents in the form of eye drops or surgical methods. Glaucoma may also occur in patients without evidence of elevated IOP, so-called low-tension glaucoma, where the IOP lies within the range observed in the general population. Large randomized prospective clinical trials have demonstrated that reducing IOP helps protect vision loss compared to untreated patients, including those without obvious elevated IOP. Many patients, however, continue to lose sight in spite of adequate IOP-lowering treatment. In this context, factors other than IOP that are implicated in RGC injury and death in glaucoma are under active investigation. Understanding how and why glaucoma progresses will propel the development of novel adjunctive treatments to prevent blindness.
Retinal pathology in glaucoma
The pathologic basis for vision loss in glaucoma appears to be RGC injury and death. The RGC injury in glaucoma is typically slow, partial, and progressive, accounting for specific patterns of vision loss that deepen and expand over time. Some evidence suggests that RGC death is apoptotic in nature, and primary mechanisms leading to programmed cell death have been reviewed elsewhere. Experimental work performed in glaucoma models with elevated IOP has been optimized over the years to study the sequence of pathological events triggered by IOP elevation.
As RGCs die, histological examination of the optic nerve head reveals progressive optic nerve head excavation, with progressive tissue atrophy and gliosis. In the retina, there is reduced density of surviving RGCs and thinning of the inner nerve fiber layer. These changes are likely due to atrophy and/or loss of the RGC cell body. Previous morphological studies demonstrated increased susceptibility of larger optic nerve fibers to IOP and this was interpreted as selective injury to magnocellular neurons. However, it is now accepted that cell atrophy may have accounted for some of these observations, and recent studies show that at least 10 RGC types in nonhuman primates are larger than magnocellular RGCs.
In addition to inner retinal pathology, photoreceptor pathology in glaucoma has been reported but remains controversial. Horizontal cell abnormality was reported previously in two glaucoma eyes with elevated IOP requiring enucleation. A recent report demonstrated abnormal phosphorylation of tau protein involving the horizontal cells of human glaucoma surgical enucleation specimens ( Box 26.2 ).
- •
The major pathologic correlate of vision loss in glaucoma is retinal ganglion cell injury and death
- •
Excavation of the optic nerve head with tissue atrophy and gliosis may also be observed
- •
In addition to retinal ganglion cell death, photoreceptor and horizontal cell pathology have been reported in glaucoma
- •
Multiple mechanisms of cell death are implicated in the pathobiology of glaucomatous optic nerve damage
- •
Elevated intraocular pressure in experimental primate glaucoma closely reproduces optic nerve and visual field changes of human disease
Pathological events associated with RGC death include glial cell alteration at the optic nerve head, disruption of axonal transport mechanisms leading to growth factor deprivation, oxidative stress, glutamate excitotoxicity, immune alterations, and vascular pathology ( Figure 26.1 ).
Etiology
The cause of open-angle glaucoma is not clearly established. The major risk factor, elevated IOP, can be considered as the cause of the glaucomatous injury in many cases. Support for this comes from nonhuman primate studies in which damage from elevated IOP closely mimics human glaucoma. In this model, IOP elevation is caused by laser-induced scarring of trabecular meshwork in the eye. The IOP elevation is typically induced in one eye, and eye pressure is chronically monitored during experiment ( Figure 26.2 ). The experimental primate model of glaucoma is highly relevant to human disease due to similar anatomy and physiology of central visual pathways, characteristic glaucomatous optic nerve changes as observed in human glaucoma, and visual deficits similar to those observed in glaucoma patients. In addition to IOP, optic disc changes and visual electrophysiological changes can be monitored in vivo. After chronic exposure to elevated IOP there is blocked anterograde transport to the lateral geniculate nucleus (LGN) and retrograde transport at the level of the lamina cribrosa, RGC death, and atrophy of surviving cell bodies and dendrites.
Anatomy and pathophysiology
The unmyelinated RGC axon inside the eye becomes myelinated as it leaves the eye beyond the lamina cribrosa. The RGC axon is long and forms the intraorbital, intracanalicular, and intracranial components of the optic nerve, optic chiasm, and optic tract ( Figure 26.3 and Box 26.3 ). In primates most RGC axons are retinogeniculate, and target the LGN directly, while a remaining 10% target other subcortical structures including the superior colliculus, pretectal nuclei, accessory optic system, and suprachiasmatic nucleus.
- •
Most of the retinal ganglion cell axon lies outside the eye, forming intraorbital, intracanalicular, and intracranial components of the optic nerve, optic chiasm, and optic tract
- •
Most retinal ganglion cells terminate in the lateral geniculate nucleus
- •
In the lateral geniculate nucleus, there is anatomic segregation of functionally distinct visual channels, namely the magno-, parvo-, and koniocellular pathways
The LGN conveys visual information received from the retina to the primary visual cortex in humans and nonhuman primates. This structure is composed of neuronal cell bodies arranged into six anatomically segregated layers that carry signals from the three major magno-, parvo-, and koniocellular vision pathways. Each LGN layer receives input from one eye only: layers 2, 3, and 5 receive input from the ipsilateral eye, and layers 1, 4, and 6 from the contralateral eye. In the two most ventral layers, magnocellular neurons convey motion information, and in the four remaining dorsal LGN layers, parvocellular neurons convey red–green color information. Koniocellular neurons are found intercalated between principal layers and convey blue–yellow color information. Eighty percent of neurons in the LGN are relay neurons that comprise the axons of the optic radiations to the primary visual cortex. Approximately 20% of LGN neurons stay within the LGN: these are GABAergic interneurons. Of the total input to the LGN, less than 10% derive from RGCs, with the remaining 90% coming from GABAergic interneurons, cortical, and subcortical synaptic inputs.
Transsynaptic degeneration of the lateral geniculate nucleus in glaucoma
Transsynaptic degeneration is a phenomenon in which injured neurons spread injury to previously uninjured neurons connected by a synapse. Within the central nervous system, injury typically spreads from a population of neurons to other anatomically and functionally connected neurons. This pathological process accounts for the progressive cognitive decline in diseases such as Alzheimer’s disease. Transsynaptic degeneration likely plays a role in the progressive loss of vision in glaucoma. Evidence from independent laboratories confirms that RGC damage leads to injury to target neurons of the LGN. Experimental work in models with elevated IOP has helped us to understand the sequence of pathological events triggered by IOP elevation. Attention to pathology within the length of the RGC axon and also its LGN target has shed new light on the progressive nature of central visual changes in glaucoma.
Evidence of transsynaptic degeneration in glaucoma comes mainly from the monkey model of glaucoma. Elevated IOP causes varying degrees of injury to myelinated optic nerve fibers behind the globe, most of which are destined for the LGN. Using established histomorphometric techniques ( Figure 26.4 ), the degree of damage ranges from no loss of optic nerve fibers to total replacement of axons by glial scar in this model.
Examination of the LGN following elevated IOP reveals metabolic changes detected by altered cytochrome oxidase enzyme activity in LGN layers connected to the experimental eye. Size measurement of LGN neurons connected to the glaucoma eye in this model shows significant atrophy of neurons, and relay neurons in magno- and parvocellular layers. Furthermore, quantitative assessment by three-dimensional morphometric techniques revealed significant loss of neurons in both magno- and parvocellular layers ( Figure 26.5 ). A linear relationship between LGN neuron loss and mean IOP was observed. Surviving neurons also showed increasing atrophy with mean IOP more pronounced in parvocellular layers.
In the koniocellular pathway, a selective marker for these neurons, called alpha subunit of type II calmodulin-dependent protein kinase (CaMK-II alpha), showed reduced immunoreactivity in the LGN. In ocular hypertensive monkeys without evidence of optic nerve fiber loss, decreased LGN immunoreactivity of this major postsynaptic density protein suggests early neurochemical alterations in the blue–yellow pathway in response to elevated IOP. In this group of ocular hypertensive monkeys, marked alterations were noted in the dendrites, with measurable reduced dendrite complexity and field area in the magno- and parvocellular layers of the LGN. These findings suggest that not all LGN changes are attributable to deafferentation within the visual system. Transsynaptic degeneration of the LGN in primate glaucoma appears to affect the three major vision channels, namely the magno, parvo- and koniocellular pathways ( Box 26.4 ). While changes to relay neurons in experimental glaucoma are described above, it is not known whether GABAergic interneurons in the LGN are also altered in glaucoma, as has been observed following enucleation and monocular deprivation.
- •
The spread of disease from injured neurons to connected neurons, known as transsynaptic degeneration, is a feature of neurodegenerative diseases and glaucoma
- •
In glaucoma, degenerative neuron changes appear diffuse, affecting all major vision pathways
- •
Early neurochemical changes have been observed in blue–yellow koniocellular neurons of the lateral geniculate nucleus (LGN) in glaucoma
- •
Major changes in metabolic activity in addition to neurochemical changes have been observed in ocular dominance columns driven by the glaucomatous eye in the primate glaucoma model
- •
Some mechanisms of injury common to neurodegenerative diseases also appear relevant to glaucomatous injury
Glial cells including astrocytes and NG-2 cells appeared altered in experimental glaucoma in the optic tract and LGN. Other studies show the involvement of other cell types such as microglia and vascular cells in the LGN in glaucoma. Mechanisms related to transsynaptic degeneration in glaucoma may be relevant to strategies aimed at preventing the spread of disease to visual centers in the brain and presumably disease progression.
Visual cortex changes in glaucoma
Relay neurons of the LGN project to the primary visual cortex. Here, neurons are arranged into six layers subdivided into sublayers. The M, P, and K geniculate axons terminate in sublayers of layer 4 and superficial layers in eye-specific columns called ocular dominance columns. In monkeys with unilateral glaucoma, a relative decrease in metabolic activity has been detected with cytochrome oxidase activity in ocular dominance columns driven by the glaucomatous eye, compared to those driven by the fellow nonglaucoma eye ( Figure 26.6 ). Neurochemical changes in the visual cortex involving growth cone-associated protein-43 (GAP-43) and inhibitory neurotransmitter receptor γ-aminobutyric acid (GABA) A receptor subtype are additional evidence of neuroplasticity in the primate visual system. Apart from these neurochemical changes, direct evidence of neuron loss in the primary visual cortex in primate glaucoma is lacking. However, relative differences in metabolic activity of ocular dominance columns between the glaucoma and nonglaucoma eye appeared more pronounced with increasing optic nerve fiber loss.
Central visual system changes in human glaucoma
Decreased LGN neuron density was reported in humans, with more pronounced effect in magnocellular layers, a subject of considerable controversy. Recent findings in human glaucoma support observations of central visual pathway neural degeneration in experimental primate glaucoma. In an index human glaucoma case, postmortem analysis of the visual system revealed neuropathology in the prechiasmal intracranial optic nerve, LGN, and visual cortex and correlated with visual field defects in a retinotopic fashion. Thus, in this case with superior bilateral visual field defects, marked inferior optic nerve atrophy and decreased phosphorylated neurofilament, neuron atrophy in the lateroposterior LGN and marked thinning of the inferior bank of the primary visual cortex was noted. In a study of human primary open-angle glaucoma patients with vision loss, atrophy of the LGN by magnetic resonance imaging has been reported. Functional neuroimaging (fMRI) showed decreased blood oxygen level-dependent (BOLD) signal in human primary visual cortex in patients with primary open-angle glaucoma ( Box 26.5 ). Thus, pathology in central vision centers is present in at least some glaucoma patients.
