Peter G.C. Bedford, BVetMed, PhD, FRCVS, DVOphthal, DECVO, ILTM
The many inherent difficulties always encountered in all the domesticated species in the attempted management of glaucoma to retain useful vision range from difficulty in diagnosis to the prevention of retinal ganglion cell death. Clinical experience alone dictates the expected poor prognosis for sight, but recent awareness of the mechanisms almost certainly involved the ganglionopathy clearly indicates that adequate neuroprotection might never be achieved. Not only are possible therapies still conjecture, but the early occurrence of what is probably a self-propagating process of neurodegeneration renders effective therapy particularly difficult in the species we treat. Currently our existing therapies must fall short of the mark and the practical difficulties associated with the assessment of outflow facility, the accurate monitoring of therapy and the complexity of surgical techniques all combine to confound the prognosis. Whilst it is logical that angle-closure-glaucomas can never be treated effectively by carbonic anhydrase inhibition alone, those glaucomas which do lend themselves to this kind of therapeutic approach are often diagnosed when ganglion cell death is already extensive and the loss of sight inevitable. The overriding factor in all glaucoma is the degeneration of the retinal ganglion cell, thus neuroprotection through effective ocular hypotension is the essential requirement of any therapy we utilise. However we are often too late in instituting that therapy and although we may contain associated pain and discomfort, the process of neuroretinal degeneration currently can neither be reversed nor stopped. The most we can achieve through the adequate reduction of intraocular pressure (IOP) is to slow this process down and retain sight for longer periods of time.
What do we understand by the term "glaucoma"? It has been simply defined as the process of ocular tissue destruction caused by a sustained elevation of the IOP above its normal physiological limits. It is the specific effect of that elevated pressure upon the composite parts of the optic nerve which renders glaucoma an emergency. The existence of "normal tension" and "low tension" glaucomas in man has blurred this simple definition for these diagnoses find origin in the clinical similarities of the optic nerve degeneration seen both in association with elevated IOP and other non-pressure related factors such as disc ischaemia or retinal excitotoxicity. It can even be argued that the rise in IOP seen in primary open-angle glaucoma in man is effect rather than cause, with only the effect being assessed and treated by current therapies. Fortunately open-angle glaucoma has limited incidence in the domesticated species for seldom are we in a position to diagnose its early presence and thus inhibit ganglion cell degeneration early in the process. There is evidence to indicate that abnormality in ganglion cell function exists in Beagles with inherited primary open-angle glaucoma before the elevation in IOP occurs, and there is strong temptation to use this evidence to suggest that the IOP changes themselves are purely a secondary feature to another as yet ill-defined disease process.
Only in those glaucomas in which there is demonstrable primary or induced defect in aqueous outflow through the iridocorneal angle can we say that the elevated pressure rise is directly responsible for the ensuing ganglion cell death. Even so, such knowledge does not ensure effective therapeutic control. It remains difficult to define the extent of the ciliary and peripheral anterior synechiae formation caused by an anterior uveitis whilst posterior synechiae formation is usually resistant to therapy. In lens luxation it is the pupillary block achieved by the anterior movement of the lens which causes the collapse of the ciliary cleft and only early lensectomy will restore adequate aqueous outflow. In primary angle-closure glaucoma we describe possible congenital predisposition and physiological pupillary block as the probable exciting factors in the acute cessation of aqueous outflow from the anterior chamber, but such consideration does not exclude other aetiologies. Again lack of aetiological detail renders hypotensive therapy difficult and inherent complications to the surgical techniques usually utilised render prognosis uncertain. However it is likely that all the glaucomas we commonly experience as veterinarians are due to maintenance of a physiologically incompatible rise in IOP and it is the characteristics of that elevated IOP which have been prompted the consultation, whether they be pain, episcleral congestion, corneal oedema, globe enlargement or defective vision. Based on the clinical picture we simply record the elevated IOP, diagnose glaucoma and set about treatment along the traditional hypotensive lines, with the knowledge that effective, long-term reduction in the IOP will approach the best we can achieve. There is a sufficiency of experimental evidence to demonstrate that the process of ganglion cell degeneration, whether it be necrosis or apoptosis, starts within the first few hours of the rise in IOP becoming established and that once triggered this process cannot be stopped. Thus currently the prognosis for sight must always be poor, with the moderating influence of any hypotensive therapy being variably expressed from one patient to another.
MECHANISMS AND TYPES OF GLAUCOMA
There are several classification systems used to describe glaucoma in the domesticated species and as such there is considerable discussion concerning the appropriateness of the terms utilised. "Congenital" dictates a presence at birth and "primary" refers to inherited glaucoma to which there may be congenital predisposition. There is confusion between the terms "narrow angle" and "closed-angle". Both refer to the width of the entrance to the ciliary cleft as assessed by gonioscopy. In primary glaucoma a congenitally narrowed angle may predispose to easier closure, but there has been difficulty in ascertaining if IOP elevates prior to actual closure. It is likely that both terms are simply gradations of the same congenital abnormality. Thus both terms are used and their proponents vigorously justify their usage. It should be noted that the term "goniodysgenesis" is used commonly to mean narrow angle or pectinate ligament dysplasia or both. Its use is limited by our gonioscopic observations, but in essence this term should cover other abnormalities of the ciliary cleft which lie beyond the level of the pectinate ligament.
Glaucoma can complicate other ocular disease processes such as uveitis, lens luxation, neoplasia and cataract and here the term secondary is used. Treatment demands both resolution of the initiating disease and attention to the changes which induce the rise in pressure.
In our patients all glaucomas are characterised by an elevated IOP, although the level of elevation may vary. In those glaucomas in which the elevation is initially low (i.e., open angle glaucoma, melanocytic glaucoma) and some secondary glaucoma, retinal ganglion cell and optic nerve damage are slow to progress. In angle-closure glaucoma the sudden high rise in IOP often renders the eye blind, undoubtedly primarily due to a cessation of axoplasmic flow at the level of the lamina cribrosa.
Retinal ganglion cell degeneration may be necrosis, but the possibility that it is apoptosis triggered by the rise in IOP is plausible, and the respective roles of nitric oxide and glutamate are worthy of discussion. The following observations are part of the current glaucoma debate in terms of pathogenesis and possible therapy.
In human studies it has been widely accepted that tissue ischaemia has a part to play in the initiation or progression of the optic disc damage which occurs in glaucoma. The autoregulation of blood flow within the disc is an essential mechanism in the maintenance of nutrition and an elevation in IOP can interfere with autoregulation.
The hypothesis that nitric oxide (NO) is involved in the degeneration of retinal ganglion cell axons is most appealing for several reasons. The apparent up-regulation and induction of some nitric oxide synthase isoforms (NOS) in astrocytes within the optic nerve head when there is an elevation of IOP has been clearly demonstrated and there is clear evidence of NO toxicity to the axons. NO and endothelin appear to be involved in the regulation of IOP and in the modulation of ocular blood flow, with NO also being involved in apoptosis.
Glutamate levels are elevated in the vitreous of primate, canine and rabbit glaucoma patients and the retinal ganglion cell layer is very susceptible to glutamate toxicity. Excitotoxicity can result in neuronal apoptosis and the mediation of excitotoxicity is by the stimulation of the N-methyl-D aspartate (NMDA) type of glutamate receptor. The overstimulation of the NMDA receptors can lead to increased NO levels. A complex, but potentially vicious circle. However, prevention of NMDA-induced excitotoxicity represents a potential mechanism for neuroprotection.
The possibility that programmed cell death can be triggered by a pressure induced failure of axoplasmic flow has been long hypothesised, and was simply based on the failure of trophic factors to reach the ganglion cell body. However, there are other aspects to apoptosis which lend themselves to its possible consideration in glaucoma, including the roles of NO and glutamate induced excitotoxicity.
Success always demands the use of effective therapy and although several aetiologies are involved in the glaucoma complex the absolute determinant in therapy selection is the amount of primary and/or induced change within the iridocorneal angle.
Medical suppression of an elevated IOP can be attempted using four types of drug: the aqueous formation suppressors; miotics; uveoscleral outflow enhancers; and the hyperosmotic agents. All four are used in the treatment of canine glaucoma, the first three commonly as emergency treatment and in long term control while the hyperosmotic agents are invaluable as emergency and preoperative treatment. A fifth category of drugs, the neuroprotection agents, is beginning to emerge as an important possible addition to medical therapy.
A. Aqueous Formation Suppressors
Carbonic anhydrase inhibitors are used traditionally in the dog and with difficulty in the cat. The alternative use of beta-adrenergic blocking agents is still being evaluated for both species.
i. Carbonic anhydrase inhibitors
Acetazolamide (Diamox; Lederle). An oral dose rate of 50 to 75mg per kg should be used and dosage should be two to three times daily. No ocular side effects are seen, but acute overdosage or long term therapy may produce metabolic acidosis, usually indicated initially by malaise, vomition and diarrhoea.
Dichlorphenamide (Daranide; Merck, Sharpe and Dohme) has provided a useful alternative to acetazolamide in that it is accompanied by less metabolic acidosis. A dose rate of 10 to 12 mg per kg is preferred two or three times daily for the dog. Potassium depletion is prevented by supplementing potassium rich food or by specific medication. 2% dorzolamide HCl (Trusopt; Merck) a topical carbonic anhydrase inhibitor and brinzolamide (Azopt-Alcon) would appear to be as effective and is less irritant.
ii. Beta-adrenergic blocking agents
Timolol maleate (Timoptol; Merck Sharpe and Dohme). Usage in the small animal patient is not indicated because the low concentration of the commercial preparation renders it ineffective in the dog and cat. Concentrations of four per cent plus are required to reduce normal canine IOP by any appreciable degree. Other such agents used in man are betaxolol HCl, carteolol HCl, levobunolol HCl and metipranolol. A combination of timolol and dorzolamide is marketed as Cosopt (Merck, Sharp and Dohme), but experience in the dog and cat is limited.
iii. Alpha2--adrenoreceptor agonists
Two such drugs are currently available. Apraclonidine (Iopidine) reduces aqueous secretion poorly in dogs but brimonidine tartrate (Alphagan; Allergan) seems to be more effective30. It produces less allergic response, probably increases uveoscleral outflow and is also neuroprotective. This drug could prove to be of considerable value to the veterinarian but long term efficacy studies are required to assess its potential use in the dog and cat.
Miotic drugs are either parasympathomimetics, producing direct stimulation (cholinergic) of the iridal musculature (e.g., carbachol and pilocarpine), or anticholinesterase inhibitors producing miosis indirectly by the potentiation of acetylcholine activity (e.g., demecarium bromide). Pilocarpine is perhaps the miotic most often used in the treatment of canine glaucoma. It should be remembered that although the potential to increase the outflow facility exists, the patient must have retained some trabecular meshwork function. Adversely, pilocarpine can sting and it can reactivate and contribute to iritis. Demecarium bromide has been of particular value in maintaining long term miosis in the management of posterior primary lens luxation, but its commercial production has now ceased. Latanoprost (Xalatan--Pharmacia and Upjohn) may prove to be of similar value, although this prostaglandin F2 analogue is used primarily to improve uveoscleral outflow. It also produces long acting miosis and in the absence of a long acting miotic preparation its use in the dog with posterior primary lens luxation could prove invaluable.
C. Uveoscleral Outflow Enhancers
Latanoprost increases the rate of outflow by the uveoscleral route. It is effective against the peptides which are present in the extracellular matrix, rendering the muscle more porous. Brimonidine tartrate also increases uveoscleral outflow but the mechanism for this activity has not yet been defined.
D. Hyperosmotic Agents
A reduction in IOP can be produced effectively and rapidly by increasing the osmolality of the plasma within the ciliary circulation to produce an osmotic pressure gradient across the blood/aqueous barrier within the ciliary epithelium. Hyperosmotic agents are valuable as emergency therapy. Their use preoperatively is an essential adjunct to glaucoma surgery, for the surgical paracentesis effect is less significant when the IOP is low, and the resultant reduction in the total blood volume of the congested globe greatly facilitates the execution of surgery. Mannitol, glycerol and urea are used routinely, all three being effective at 1.0 to 1.5 G per kg body weight.
Neuroprotection and Neuroregeneration
Undoubtedly elevation of the IOP is the most significant trigger factor for glaucomatous optic neuropathy and lowering of the IOP to a normal or subnormal level is the essential factor in treatment. However observation that the NOS and glutamate levels are elevated in glaucoma and that they are involved in retinal ganglion cell necrosis or apoptosis has raised the possibility of neuroprotective therapies and even neuroregeneration. Thus NOS inhibitors, exciting amino acid antagonists, glutamate receptor antagonists, apoptosis inhibitors and calcium channel blockers are all involved potentially in the development of future glaucoma therapies. The calcium channel blockers may reduce the effect of impaired microcirculation to the optic nerve head whilst potentially increasing outflow facility at the level of the trabecular cells.
The difficulty of achieving adequate reduction of the IOP in canine glaucoma by the medical means currently available has prompted the use of several surgical techniques in this species.
A reduction in aqueous production can be achieved by cyclodestruction utilising cryosurgery, heat or laser. The amount of ciliary body damage must be sufficient to ensure that balance is regained between the resultant impaired aqueous production and whatever aqueous drainage is possible. The reopening of a closed ciliary cleft by cyclodialysis involves the breaking down of collapsed cleft tissue and synechiae to separate the ciliary body from the underlying sclera, allowing the anterior chamber to become confluent with the suprachoroidal space. The certain failure to control glaucoma is due to the subsequent closure of the cleft by the rapid formation of postoperative adhesions.
In the dog surgical bypass of the collapsed ciliary cleft is most easily achieved either by iridencleisis or by a corneoscleral (limbal) trephination technique combined with peripheral iridectomy. These techniques allow aqueous to pass directly from the anterior chamber to the subconjunctival tissues where it is absorbed by the vascular and lymphatic elements present. Both may prove successful initially but in the short term fibrin may occlude the sclerostomy wound and long term control may be denied by fibrosis of both the sclerostomy and the subconjunctival tissues.
Shunt (or gonioimplant) surgery offers a realistic approach to the control of IOP for it counteracts the effects of subconjunctival fibrosis to some extent. Several types of shunt exist, basically those with or without valves. Satisfactory results may be obtained using a one-piece Silastic drainage implant consisting of an anterior chamber tube and an attached large surface area strap. The shunt allows aqueous to be diverted from the anterior chamber to the large subconjunctival scar sac which develops around the strap. Further modification of this technique resulting in smaller gonioimplants and even simpler surgery will be possible through the use of fibroblast inhibitor drugs. In the future simple sclerostomy may be all that is necessary to offer the patient effective long term IOP control.
An extended reference list is available on request
1. Nickells, R.W. (1996) Retinal ganglion cell death in glaucoma: the how, the why and the maybe. J. Glaucoma, 6: 123.
2. Offri, R., Samuelson, D.A., Strubbe, D.T., et al (1994) Altered retinal recovery and optic nerve fibre loss in primary open-angle glaucoma in the Beagle. Exp. Eye Res., 58: 245.
3. Haefliger, I.O., Dettmann, E., Liv, R., et al (1999) Potential role of nitric oxide and endothelin in the pathogenesis of glaucoma. Survey of Ophthalmology, 43, S51.
4. Dreyer, E.B., Zurakowski, D., Schumer, R.A., et al (1996) Elevated glutamate in the vitreous body of humans and monkeys with glaucoma. Arch. Ophthalmol., 114, 299.
5. Brooks, D.E., Garcia, G.A., Dreyer, E.B., et al (1997) Vitreous body glutamate concentrations in dogs with glaucoma. Am.J.Vet.Res., 58, 864.
6. Swartz, M., Belkin, M., Yoles, E. et al (1996) Potential treatment modalities for glaucomatous neuropathy:neuroprotection and neuroregeneration. J. Glaucoma 5, 427.
7. Bedford, P.G.C. (1989) A clinical evaluation of a one-piece drainage system in the treatment of canine glaucoma. J. Small. Anim. Pract. 30, 68.
8. Garcia, G.A., Brooks, D.E., Gelatt, K.N., et al (1998) Evaluation of valved and non-valved gonioimplants in 83 eyes of 65 dogs with glaucoma. Anim. Eye. Res. 17, 9.