Ocular therapeutics - Part 12

Anti-glaucoma
The goal of glaucoma treatment was summarised by the European Glaucoma Society as 'preservation of visual function adequate to the individual needs with minimal or no side effects, for the expected lifetime of the patient, without any disruption of his/her normal activities, at a sustainable cost'.1
There is now a wide range of anti-glaucoma agents (Table 1) to allow the prescriber to tailor therapy to the individual, bearing in mind the target pressure to be achieved, the patient's co-morbidity and concurrent medical therapy, tolerability of and concordance with therapy. All drugs used for the treatment of glaucoma lower the intraocular pressure in one (or both) of two ways - by reducing aqueous inflow or increasing aqueous outflow. An increase in outflow may be achieved by increasing outflow through conventional (trabecular) or non-conventional (uveoscleral) pathways (see Part 5 and Figure 1).

Sympathomimetics
The original sympathomimetic drug used in the treatment of ocular hypertension and open-angle glaucoma, adrenaline, has been discontinued but the more lipophilic pro-drug dipivefrine remains available. Dipivefrine is converted into adrenaline by esterases in the cornea and stimulates both <03B1> and <03B2> adrenoceptors. This results in a decrease in aqueous humour production and an increase in aqueous humour outflow through the trabecular meshwork leading to a fall in intraocular pressure of 15-20 per cent. However, its non-selectivity leads to both ocular and systemic side effects and, in clinical use, its place has been taken by more selective agents which reduce intraocular pressure by 20-25 per cent. Apraclonidine and brimonidine are relatively selective drugs, stimulating <03B1>2-adrenoceptors, resulting in a reduction in aqueous production and an increase in uveoscleral outflow. Brimonidine is the preferred agent for long-term use as it is about 30 times more selective than apraclonidine for <03B1>2-adrenoceptors and thus has a reduced incidence of alpha1-mediated side effects such as mydriasis, eyelid retraction and conjunctival blanching. In addition, the effect of apraclonidine reduces with time and so it is only licensed for short-term use. Side effects of sympathomimetics are shown in Table 2.

Parasympathomimetics
Since the discontinuation of IsoptoCarbachol, pilocarpine is the only commercially-available parasympathomimetic used in the treatment of glaucoma. It reduces intraocular pressure by 15-20 per cent by increasing aqueous outflow through the trabecular meshwork. In the treatment of open-angle glaucoma, pilocarpine's short duration of action, necessitating instillation three to four times a day, and its troublesome ocular side effects of induced myopia and ciliary spasm, have led to its replacement by better-tolerated agents with lower instillation frequencies. However, it is still an essential part of the treatment of acute closed-angle glaucoma where it is used to constrict the pupil to free the anterior chamber angle. The stimulation of systemic receptors of the parasympathetic nervous system leads to a wide range of adverse effects including: sweating, salivation, nausea, vomiting, tremor and hypotension.

Beta adrenergic antagonists
The beta adrenergic antagonists or beta-blockers were drugs of first choice for many patients for the treatment of open-angle glaucoma until the prostaglandin analogues became licensed as first line agents. They are effective, lowering the intraocular pressure by 20-25 per cent by reducing aqueous humour production, require once daily or twice daily instillation and have few serious ocular side effects. The original beta-blocker, timolol, is the gold standard treatment of open angle glaucoma against which all other agents are compared. This group of drugs can be classified as selective, ie selectively blocking <03B2>1 receptors such as betaxolol, or non-selective ie blocking both <03B2>1 and <03B2>2 receptors as is the case for the remainder of the class. The potential advantages of using a selective <03B2>-blocker are questionable.2,3 One of the <03B2>-blockers, carteolol, possesses intrinsic sympathomimetic activity (ie it is a partial agonist) which may give the drug some advantages over the other beta-blockers such as less nocturnal bradycardia and less change in the lipid profile.4-6
Although the summary of product characteristics (SPC) for Timoptol states that 'provided that the intraocular pressure is maintained at satisfactory levels, many patients can be placed on once-a-day therapy', a gel-forming solution which provides a longer contact time than the aqueous solution is preferred for once daily instillation.7 Generally, therapy should be started at the lower strength available and only increased if the response is unsatisfactory. However, betaxolol is available in two strengths which have been shown to be equipotent.8 Betoptic 0.25 per cent suspension contains betaxolol molecules ionically bound to amberlite resin. Upon instillation the betaxolol molecules are displaced by ions in the tear film. This displacement process occurs over several minutes and enhances the ocular comfort and efficacy of the formulation.9
Four of the five <03B2>-blockers are available as unit dose, preservative-free preparations. In addition timolol is available in combination with dorzolamide and with latanoprost
(Table 1).
The current preference for prostaglandin analogues as drugs of first choice is mainly due to the systemic side effects of beta adrenergic antagonists (Table 3) and their contraindication in a high proportion of patients. The Committee on Safety of Medicines has advised against the use of all topical <03B2>-blockers, including betaxolol, in patients with asthma or a history of obstructive airways disease.10 These drugs are also contraindicated in patients with bradycardia, heart block and uncontrolled heart failure. They may mask the symptoms of thyrotoxicosis or hypoglycaemia and should be used with caution in patients with labile diabetes.
Interactions with drugs affecting the cardiovascular system are possible and care should be taken when co-prescribing systemic <03B2>-blockers, certain calcium channel blockers and reserpine because of additive effects.

Carbonic anhydrase inhibitors
For many years, the only carbonic anhydrase inhibitor available was acetazolamide which had to be administered systemically. There are now two topical carbonic anhydrase inhibitors available (Table 1), one of which is available in a combination product with timolol.
Carbonic anhydrase inhibitors reduce intraocular pressure by 15-20 per cent by reducing aqueous formation as a result of the inhibition of the enzyme which catalyzes the cellular production of carbonic acid. This acid dissociates to form hydrogen and bicarbonate ions, the latter plays a key role in the formation of aqueous humour.
Being less potent ocular hypotensives as <03B2>-blockers or prostaglandin analogues, topical carbonic anhydrase inhibitors are licensed as second line agents, that is they may be used as adjunctive therapy to first line agents or as monotherapy in patients unresponsive to or intolerant of first line agents. The ocular hypotensive effect of brinzolamide administered twice daily is equivalent to that of dorzolamide administered three times daily and the former causes less ocular discomfort on instillation. The combination product Cosopt has been shown to be more effective than either dorzolamide or timolol alone and equal in efficacy to a loose combination of the two drugs.11,12
Acetazolamide is administered intravenously in cases where the intraocular pressure needs to be reduced rapidly, for example preoperatively or in acute closed-angle glaucoma, and is used orally following surgery or in complex glaucoma cases. Side effects, contraindications and drug interactions of topical carbonic anhydrase inhibitors are shown in Table 4.

Prostaglandin analogues and
prostamides
These drugs (Table 5) are all now licensed as first line agents and are being used as such more frequently because they are potent ocular hypotensives, reducing IOP by 25-30 per cent, and are almost free of systemic side effects. The prostaglandin analogues, latanoprost and travoprost, are analogues of prostaglandin F2<03B1>. The ocular hypotensive effect achieved by an increase in uveoscleral outflow is thought to be as a result of activation of prostanoid FP receptors leading to increased production of matrix metalloproteinases which degrade the extracellular matrix of the ciliary muscle.
Bimatoprost is a synthetic prostamide structurally related to prostaglandin F2<03B1>13 but it does not act through any known prostaglandin receptors. However, a specific prostamide receptor has not yet been structurally identified. Bimatoprost reduces IOP not only by enhancing uveoscleral outflow but also by increasing trabecular outflow.14
Parrish et al compared the efficacy and tolerability of latanoprost, bimatoprost, and travoprost and stated that the drugs were comparable in their ability to reduce IOP in open-angle glaucoma and ocular hypertensive patients but that latanoprost exhibited greater ocular tolerability.14 This is confirmed by a study examining patient persistence with therapy in which latanoprost-treated patients demonstrated significantly greater persistency than did those treated with either bimatoprost or travoprost.15 However, Bournias et al studied patients with uncontrolled glaucoma or ocular hypertension on a variety of treatment regimens including latanoprost and found that when bimatoprost was substituted for latanoprost the prostamide therapy was well-tolerated and helped many more patients reach low target pressures when used as a replacement for latanoprost.16
Patients starting on therapy with one of these drugs should be informed of the possibility of eyelash growth, darkening of the eyelid skin and increased iris pigmentation. Some of these changes may be permanent, and may lead to differences in appearance between the eyes when only one eye is treated. The change in eye colour is due to an increase in the number of melanosomes (pigment granules) in the melanocytes of the iris and not to an increase in the number of melanocytes. The long-term effects on the melanocytes and any consequences thereof are currently unknown. The change in eye colour has predominantly been seen in patients with mixed coloured irides, ie, blue-brown, grey-brown, yellow-brown and green-brown; however, it has also been observed in patients with brown eyes. Typically, the brown pigmentation around the pupil spreads concentrically towards the periphery in affected eyes, but the entire iris or parts of it may become more brownish. The incidence in patients with mixed colour irides ranged from 7 to 85 per cent with latanoprost, with yellow-brown irides having the highest incidence. In patients with homogeneously blue eyes, no change has been observed and in patients with homogeneously grey, green or brown eyes, the change has only rarely been seen. The change in iris pigmentation occurs slowly and may not be noticeable for several months. After discontinuation of therapy, no further increase in brown iris pigment has been observed. It has not been associated with any symptom or pathological changes in clinical trials to date.
Other side effects of the prostaglandin analogues and bimatoprost are shown in Table 6.
Latanoprost is available as a combination product with timolol 0.5 per cent, Xalacom which was launched in 2001 with a frequency of once daily in the morning, shortly before the introduction of travoprost and bimatoprost. At the time of the launch of Xalacom, latanoprost did not have a first line licence and therefore Xalacom is licensed for the reduction of intraocular pressure in patients with open-angle glaucoma and ocular hypertension who are insufficiently responsive to topical beta-blockers. Unlike the combination of dorzolamide and timolol (Cosopt) there was no published data about the efficacy of the preparation compared with the loose combination at the time of launch and prescribers have had to wait until 2004 before this data became available.17 It was of concern that the authors of the paper comparing the efficacy of Xalacom with its component drugs administered at the licensed dosages of timolol 0.5 per cent bd and latanoprost 0.005 per cent once daily in the evening could not prove equal efficacy according to the study design. The authors suggest that the difference in IOP lowering between the unfixed and fixed treatment may be due to the timing of the latanoprost dose (morning cf evening) and the once daily versus twice daily dosing of timolol. They recommend additional studies to determine whether an evening dose of the fixed combination is more effective in controlling IOP.

Ocular lubricants
Ocular lubricants in the form of eye drops, gels and ointments (Table 7) are used to treat the discomfort associated with conditions in which the tear film is reduced or unstable, such as keratoconjunctivitis sicca and xerophthalmia. They are also used to prevent exposure keratitis in patients undergoing surgery, in intensive care, in ectropion or other abnormalities of the eyelids where part of the ocular surface is exposed, as contact solutions in diagnostic procedures, as comfort solutions in contact lens wearers and to lubricate artificial eyes. Lubricant eye ointments are also used in the treatment of recurrent corneal erosions. The compounds used in artificial tear preparations are widely used in ophthalmology to prolong the ocular contact time. Allergan's Liquifilm system (polyvinyl alcohol 1.4 per cent) is used as a base for levobunolol, flurbiprofen and fluorometholone eye drops and Alcon's Isopto system (hypromellose 0.5 per cent) as a base for their atropine eye drop.
The substituted cellulose ethers, hydroxypropylmethylcellulose (HPMC or hypromellose), hydroxyethylcellulose and carboxymethylcellulose (carmellose) are widely used for the treatment of tear deficiency. However, their retention time in the eye is short (less than 30 minutes) and they need to be instilled at frequent intervals. Carmellose has been shown to a useful adjunct in Lasik surgery18 and to have a potential cytoprotective effect on the ocular surface when used before insertion of contact lenses.19
Polyvinyl alcohol (PVA) enhances the stability of the tear film with a 1.4 per cent concentration increasing the tear break up time by a factor of 1.89, with higher concentrations increasing it still further. Solutions of PVA may be thickened or gelled by sodium bicarbonate and borate and sodium, potassium and zinc sulphate. Another polymer, polyvinylpyrrolidone (povidone, PVP) is used at a 5 per cent concentration for the symptomatic treatment of dry eyes.
When aqueous deficiency is associated with threads and filaments of mucus in the tear film, a preparation of acetylcysteine 5 per cent in hypromellose (Ilube) is useful. Acetylcysteine, an N-acetyl derivative of the naturally occurring amino acid L-cysteine, is thought to exert its mucolytic action by opening up disulphide linkages. Acetylcysteine may also have a beneficial effect on corneal wound healing.20
Carbomers (polyacrylic acids) have a much longer retention time than the cellulose ethers and are generally used four times a day. In an in vitro study, carmellose artificial tears were shown to be less toxic than carbomer gel formulations so that questions about the benefits of using high-viscosity gels in the treatment of dry-eye syndrome still remain.21 However, Bron et al22 found polyacrylic acid gel was as safe as and more effective than polyvinyl alcohol in the treatment of patients with dry eyes. Choice of carbomer product may be based on preservative content; multi-dose Viscotears and Liposic contain cetrimide; GelTears and Liquivisc contain benzalkonium chloride; a preservative-free form of Viscotears is also available. Patients may prefer a particular presentation, for example, Liquivisc is presented as a bottle while the other products are packaged in tubes. In terms of cost, GelTears is the least expensive preparation and is the brand cited in the Drug Tariff, which guides pharmacists on preparations they may dispense when filling NHS prescriptions.
Lubricating eye ointments may be applied at night or used to lubricate the eye surface in cases of recurrent corneal epithelial erosion. These preparations contain paraffins which are not miscible with the tear film and may therefore cause temporary visual disturbance.
Sodium hyaluronate has been described as the most promising tear substitute for the treatment of xerophthalmia in Sjgren's syndrome23 and has been compared favourably against Tears Naturale,24 polyvinyl alcohol 1.4 per cent25 and a range of artificial tears.26 It is available in both multi-dose and unit dose forms but, unfortunately, is not prescribable at NHS expense as the commercially available products are classified as devices rather than medicines. Unfortunately, marketing of ocular lubricants as CE marked products rather than medicines is a growing trend. The product Systane, found by Christensen et al 27 to be superior at reducing both the signs and symptoms of dry eye compared to the carboxymethylcellulose-containing control, fits into the same category.
Ocular lubricants are generally well tolerated but transient burning on instillation can occur and the instillation of gels and ointments can lead to a temporary blurring of vision. Allergic reactions to preservatives and other excipients can occur. Contact lenses should not be worn when drops containing preservatives, gels or ointments are instilled.
Patients with Sjgren's syndrome have shown improvement in dry eye and dry mouth with pilocarpine given orally at a dose of 5mg bd - qds. The major adverse effect of this therapy was sweating.28-30

Analgesics
Pain is associated with actual or potential damage to tissues. It is initiated at specialised pain endings in peripheral tissues and is relieved by analgesics which act peripherally or centrally. Salicylates and non-salicylate non-steroidal anti-inflammatory drugs (NSAIDs) act predominantly peripherally while paracetamol and opioids act centrally.

Salicylates
Aspirin or acetylsalicyclic acid is the oldest non-narcotic analgesic. In addition to its analgesic properties, it is anti-inflammatory and antipyretic and in low doses inhibits platelet aggregation. Aspirin inhibits synthesis of the prostaglandin PGE2 by irreversible acetylation and inactivation of the enzyme cyclo-oxygenase. Inhibition of cyclo-oxygenase in the hypothalamus of the brain results in a reduction of the thermoregulator setting, giving aspirin its antipyretic action. The action of aspirin is dose dependent; inhibiting platelet aggregation at low doses (75-150mg daily), being analgesic at 300-600mg per dose and anti-inflammatory at higher doses of 3-4g/day.
Aspirin is used for the relief of headaches, musculo-skeletal pain, dysmenorrhoea and pyrexia. It is contra-indicated in patients less than 16 years, unless specifically indicated, because of an association with Reye's syndrome30 and should be used with caution in asthmatic patients in some of whom it can precipitate bronchospasm. Gastric irritation with aspirin can be minimised by taking the dose after food. Anti-inflammatory doses of aspirin may also cause mild chronic salicylate intoxication characterised by dizziness, tinnitus, and deafness. Aspirin interacts with a large number of other drugs. The most important interactions are with anticoagulants and selective serotonin re-uptake inhibitor (SSRI) antidepressants where there is an increased risk of bleeding and with the antimetabolite methotrexate where reduced excretion of the antimetabolite can lead to methotrexate toxicity.

Paracetamol
Paracetamol is often the first drug used for management of mild to moderate pain but is also used as an adjunct to opioids in the management of more severe pain. It has analgesic and antipyretic efficacy similar to aspirin but is a very weak inhibitor of peripheral cyclo-oxygenase and therefore has little or no anti-inflammatory action. In recommended doses of up to 4g per day, paracetamol rarely causes significant side effects but overdose can lead to serious liver toxicity, an effect that may be seen even at normal doses in patients with pre-existing liver disease. Because it is virtually free of side effects, it should be considered the treatment of choice for mild-to-moderate pain. It has been used safely for years in children and is the mild analgesic of choice for the pregnant patient. Paracetamol is available in a wide range of preparations including combinations of paracetamol with mild opiates such as codeine, dihydrocodeine and dextropropoxyphene. Concerns about use of paracetamol in suicide attempts have lead to restrictions on the pack sizes which can be sold in pharmacy and non-pharmacy outlets.31

Non-steroidal anti-inflammatory drugs (NSAIDs)
These drugs are analgesic, antipyretic and anti-inflammatory working primarily by inhibiting cyclo-oxygenase (COX) in injured or inflamed tissues. They reduce or eliminate production of the sensitisers for peripheral nociceptors and are used in the treatment of the inflammatory arthritides, advanced osteoarthritis, back pain and soft-tissue disorders.
There are two forms of cyclo-oxygenase, COX-1 and COX-2. COX-1 is the enzyme producing the prostaglandins necessary for normal renal function and integrity of the gastric mucosa; therefore its inhibition, while resulting in analgesia and a reduction in inflammation, also leads to gastric ulceration and renal impairment. COX-2 is induced by inflammatory mediators such as interleukin-1 and marked increases in this enzyme are found in inflammation. Older traditional NSAIDs such as ibuprofen and diclofenac inhibit both enzymes to a similar degree, while etodolac and meloxicam inhibit COX-2 up to 50 times more than COX-1. Newer agents, the 'coxibs' celecoxib, etoricoxib and valdecoxib and the recently discontinued rofecoxib are even more COX-2 selective. This selectivity is said to give COX-2 inhibitors a safer side effect profile as the potential for gastric ulceration and renal impairment is reduced, although this has recently been questioned.32 Moreover, COX-2 selective NSAIDs may be safer than classic NSAIDs with respect to platelet function during the perioperative period.33 However, these newer agents are much more costly than traditional NSAIDs and the National Institute for Clinical Excellence (NICE) has issued guidance as to when they should be used.34
There is a large range of NSAIDs available and the choice of drug may depend upon side effect profile, range of dosage forms, legal category and drug selectivity. The Committee on Safety of Medicines has advised that all NSAIDs are associated with serious gastro-intestinal toxicity, the risk being higher in the elderly. Evidence on the relative safety of seven non-selective NSAIDs indicates differences in the risks of serious upper gastro-intestinal side-effects. Of the non-selective NSAIDs, azapropazone is associated with the highest risk, ibuprofen with the lowest and piroxicam, ketoprofen, indometacin, naproxen and diclofenac are associated with intermediate risks (possibly higher in the case of piroxicam). Selective inhibitors of cyclo-oxygenase-2 are associated with a lower risk of serious upper gastro-intestinal side-effects than non-selective NSAIDs with celecoxib having the lowest risk.35
While it is preferable to avoid NSAIDs in patients with active or previous gastro-intestinal ulceration or bleeding, and to withdraw them if gastro-intestinal lesions develop, nevertheless patients with serious rheumatic diseases (eg rheumatoid arthritis) are usually dependent on NSAIDs for effective relief of pain and stiffness. Those at risk of duodenal or gastric ulceration (including the elderly) who need to continue NSAID treatment should receive either a selective inhibitor of cyclo-oxygenase-2 alone, or a non-selective NSAID with gastroprotective treatment such as an H2-receptor antagonist, a proton pump inhibitor or misoprostol.

Opioids
Opioid analgesics mimic the action of endorphins, endogenous neuropeptides.
There are three specific opioid receptors (mu), <03BA> (kappa) and <03B4> (delta). The results of activation of these receptors are shown in Table 8.
Opioids are used for a range of indications including the relief of moderate to severe pain in areas as diverse as osteo-arthritic, obstetric, post herpetic neuralgic, chronic non-malignant and post-operative pain as well as in terminal care. The effects of opioids on the gastro-intestinal tract and respiratory system have, respectively, led to their use as anti-diarrhoeals and cough suppressants.
Opioid analgesics share many side-effects, though qualitative and quantitative differences exist. The most common include nausea, vomiting, constipation, and drowsiness. Larger doses produce respiratory depression and hypotension. Other side-effects include difficulty with micturition, ureteric or biliary spasm, dry mouth, sweating, headache, facial flushing, vertigo, bradycardia, tachycardia, palpitations, postural hypotension, hypothermia, hallucinations, dysphoria, mood changes, dependence, miosis, decreased libido or potency, rashes, urticaria and pruritus.
Opioids should be avoided in acute respiratory depression, acute alcoholism and where there is a risk of paralytic ileus. In addition to interfering with respiration, opioids affect pupillary responses vital for neurological assessment so they should be avoided in raised intracranial pressure or head injury. Due to the risk of a pressor response to histamine release, injection of opioids should be avoided in phaeochromocytoma. These drugs should be used with caution in hypotension, hypothyroidism, asthma and decreased respiratory reserve, prostatic hypertrophy, convulsive disorders, dependence, pregnancy and breast-feeding. They may precipitate coma in hepatic impairment, require dose reduction or avoidance in renal impairment and a reduced dose in elderly and debilitated patients. Use of cough suppressants containing opioid analgesics is not generally recommended in children and should be avoided altogether in those under one year.

References
1 European Glaucoma Society Terminology and Guidelines for Glaucoma 2nd edition 2003.
2 Watson PG, Barnett MF, Parker V. et al. A 7 year prospective comparative study of three topical beta blockers in the management of primary open angle glaucoma. Br J Ophthalmol, 2001; 85:962-8.
3 Kirwan JF, Nightingale JA, Bunce C et al. Do selective topical antagonists for glaucoma have respiratory side effects? Br J Ophthalmol, 2004;88:196-198.
4 Netland PA, Weiss HS, Stewart WC et al. Cardiovascular effects of topical carteolol hydrochloride and timolol maleate in patients with ocular hypertension and primary open-angle glaucoma. Night Study Group. Am J Ophthalmol, 1997;123:465-77.
5 Bartlett JD, Olivier M, Richardson T. Central nervous system and plasma lipid profiles associated with carteolol and timolol in postmenopausal black women. J Glaucoma, 1999;8:388-95.
6 Stewart WC, Dubiner HB, Mundorf TK et al. Effects of carteolol and timolol on plasma lipid profiles in older women with ocular hypertension or primary open-angle glaucoma. Am J Ophthalmol, 1999;127:142-7.
7 Shedden A, Laurence J, Tipping R. Timoptic-XE 0.5 per cent Study Group. Efficacy and tolerability of timolol maleate ophthalmic gel-forming solution versus timolol ophthalmic solution in adults with open-angle glaucoma or ocular hypertension: a six-month, double-masked, multicenter study. Clin Ther, 2001;23:440-50.
8 Weinreb RN, Caldwell DR, Goode SM, Horwitz BL, Laibovitz R, Shrader CE et al. A double-masked three-month comparison between 0.25 per cent betaxolol suspension and 0.5 per cent betaxolol ophthalmic solution. Am J Ophthalmol, 1990;110:189-92.
9 Weinreb RN, Jani R. A novel formulation of an ophthalmic beta-adrenoceptor antagonist. Journ Parenteral Sci Technol, 1992;46:51-3.
10 Committee on Safety of Medicines. Bronchospasm associated with cardioselective and topical <03B2>-blockers. Current problems no.28, May 1990.
11 Strohmaier K, Snyder E, DuBiner H, Adamsons I. The Dorzolamide Study Group. The efficacy and safety of the dorzolamide-timolol combination versus the concomitant administration of its components. Ophthalmol, 1998;105:1936-44.
12 Boyle JE, Ghosh K, Gieser DK, Adamsons IA. The Dorzolamide-Timolol Study Group. A randomized trial comparing the dorzolamide-timolol combination given twice daily to monotherapy with timolol and dorzolamide. ibid 1998;105:1945-51.
13 Woodward DF, Phelps RL, Krauss AH et al. Bimatoprost: a novel antiglaucoma agent. Cardiovasc Drug Rev,2004;22:103-20.
14 Parrish RK, Palmberg P, Sheu WP. XLT Study Group A comparison of latanoprost, bimatoprost, and travoprost in patients with elevated intraocular pressure: a 12-week, randomized, masked-evaluator multicenter study. Am J Ophthalmol, 2003;135:688-703.
15 Reardon G, Schwartz GF, Mozaffari E. Patient persistency with ocular prostaglandin therapy: a population-based, retrospective study. Clin Ther, 2003;25:1172-85.
16 Bournias TE, Lee D, Gross R et al. Ocular hypotensive efficacy of bimatoprost when used as a replacement for latanoprost in the treatment of glaucoma and ocular hypertension. J Ocul Pharmacol Ther, 2003;19:193-203.
17 Diestelhorst M, Larsson LI. European Latanoprost Fixed Combination Study Group A 12 week study comparing the fixed combination of latanoprost and timolol with the concomitant use of the individual components in patients with open angle glaucoma and ocular hypertension. Br J Ophthalmol, 2004;88:199-203.
18 Lenton LM, Albietz JM. Effect of carmellose-based artificial tears on the ocular surface in eyes after laser in situ keratomileusis. J Refract Surg, 1999;15(Suppl):S227-31.
19 Vehige JG, Simmons PA, Anger C et al. Cytoprotective properties of carboxymethyl cellulose (CMC) when used prior to wearing contact lenses treated with cationic disinfecting agents. Eye Contact Lens, 2003;29:177-80.
20 Aldavood SJ, Behyar R, Sarchahi AA et al. Effect of acetylcysteine on experimental corneal wounds in dogs. Ophthalmic Res. 2003;35:319-23.
21 Diebold Y, Herreras JM, Callejo S et al. Carbomer- versus cellulose-based artificial-tear formulations: morphologic and toxicologic effects on a corneal cell line. Cornea, 1998;17:433-40.
22 Bron AJ, Mangat H, Quinlan M et al. Polyacrylic acid gel in patients with dry eyes: a randomised comparison with polyvinyl alcohol. Eur J Ophthalmol, 1998;8:81-9.
23 Baudouin C, Pisella PJ, Brignole F. Current treatments of xerophthalmia in Sjgren's syndrome. Rev Med Interne, 2004 May;25:376-82.
24 Iester M, Orsoni GJ, Gamba G et al. Improvement of the ocular surface using hypotonic 0.4 per cent hyaluronic acid drops in keratoconjunctivitis sicca. Eye. 2000;14:892-8.
25 McDonald CC, Kaye SB, Figueiredo FC et al. A randomised, crossover, multicentre study to compare the performance of 0.1 per cent (w/v) sodium hyaluronate with 1.4 per cent (w/v) polyvinyl alcohol in the alleviation of symptoms associated with dry eye syndrome. Eye, 2002;16:601-7.
26 Solomon A, Merin S. The effect of a new tear substitute containing glycerol and hyaluronate on keratoconjunctivitis sicca. J Ocul Pharmacol Ther, 1998;14:497-504.
27 Christensen MT, Cohen S, Rinehart J, et al. Clinical evaluation of an HP-guar gellable lubricant eye drop for the relief of dryness of the eye. Curr Eye Res. 2004;28:55-62.
28 Vivino FB, Al-Hashimi I, Khan Z et al. Pilocarpine tablets for the treatment of dry mouth and dry eye symptoms in patients with Sjgren syndrome: a randomized, placebo-controlled, fixed-dose, multicenter trial. P92-01 Study Group. Arch Intern Med, 1999;159:174-81.
29 Tsifetaki N, Kitsos G, Paschides CA et al. Oral pilocarpine for the treatment of ocular symptoms in patients with Sjgren's syndrome: a randomised 12-week controlled study. Ann Rheum Dis, 2003;62:1204-7.
30 MCA/CSM Aspirin and Reye's syndrome in children up to and including 15 years of age. Current Problems in Pharmacovigilance, 2002;28:4.
31 Medicines, Ethics and Practice: A Guide for Pharmacists. 28th Edition July 2004 Pharmaceutical Press.
32 Laporte JR, Ibanez L, Vidal X et al. Upper gastrointestinal bleeding associated with the use of NSAIDs: newer versus older agents. Drug Saf, 2004;27:411-20.
33 Blaicher AM, Landsteiner HT, Al-Falaki O et al. Acetylsalicylic acid, diclofenac, and lornoxicam, but not rofecoxib, affect platelet CD 62 expression. Anesth Analg, 2004;98:1082-5.
34 National Institute for Clinical Excellence Guidance on the use of cyclo-oxygenase (Cox) II selective inhibitors celecoxib, rofecoxib, meloxicam and etodolac for osteoarthritis and rheumatoid arthritis. Technology appraisal guidance no 27, July 2001.
35 MCA/CSM Non-steroidal anti-inflammatory drugs (NSAIDs) and gastrointestinal (GI) safety. Current Problems in Pharmacovigilance, 2002;28:5.

<2666> Lucy Titcomb is Directorate Pharmacist, Ophthalmology, Birmingham and Midland Eye Centre, Sandwell and West Birmingham NHS Trust, and Bernard Gilmartin is Professor of Optometry, School of Life and Health Sciences, Aston University