The Pharmaceutical Journal Vol 264 No 7088 p441-445
March 18, 2000 Continuing education

Eye disorders

Topical ocular antibiotics: part 2

By Lucy C. Titcomb, MCPP, MRPharmS

Ciprofloxacin Ciprofloxacin eye-drops are indicated for the treatment of corneal ulcers and superficial infections of the eye and adnexa (accessory parts or appendages) caused by susceptible organisms. A maximum duration of therapy of 21 days is recommended. The safety and efficacy of ciprofloxacin eye-drops in children under the age of one year has not been established.
Ciprofloxacin has been compared with chloramphenicol and tobramycin in the treatment of conjunctivitis and blepharitis. Power et al investigated the efficacy of ciprofloxacin and chloramphenicol in 57 culture-positive patients and reported that the difference between the groups was not significant.⁹ Safety was also similar, with only one patient from each treatment group suffering an adverse event. Liebowitz¹⁰ compared ciprofloxacin with tobramycin in a placebo controlled trial of 288 culture-positive patients with bacterial conjunctivitis. Both antibiotics were highly effective and significantly superior to placebo.
In a prospective, multicentre clinical study, the same author compared the clinical and antibacterial efficacy of ciprofloxacin 0.3 per cent with that of standard dual therapy in 148 culture proven cases of bacterial keratitis (ie, bacteria had been cultured).¹¹ Most patients in the standard dual therapy group received cefazolin 3.3 per cent with gentamicin or tobramycin 1.4 per cent. Success rates were similar at 92 per cent for ciprofloxacin and 88 per cent for the dual therapy. The most frequently noted untoward ocular event associated with ciprofloxacin therapy was a white crystalline precipitate, commonly located in the superficial portion of the corneal defect (ie, the area of inflammation). This problem was encountered in 35 (16.6 per cent) of patients. A group of ophthalmologists involved in the multicentre study noted the appearance of the white precipitate as the only potentially adverse effect in the ciprofloxacin group.¹² Although the precipitate resolved in all patients and did not appear to cause any scarring, they saw it as a disadvantage because it caused a temporary decrease in vision and prevented adequate evaluation of the corneal infiltrate. Precipitation of ciprofloxacin occurs as a result of a change in the pH of the eye-drop as it mixes with the tear film.¹³

Lomefloxacin Lomefloxacin is indicated for the treatment of acute bacterial conjunctivitis. Following a loading dose of one drop in the affected eye every five minutes for 20 minutes, the drug is instilled twice daily. Lomefloxacin's twice daily dosage is advantageous compared with the more frequent instillation of ciprofloxacin and ofloxacin. A seven to nine day course of treatment is recommended and there are no age restrictions to the use of lomefloxacin.
Several clinical trials have demonstrated the efficacy of lomefloxacin in acute bacterial conjunctivitis. Malminiemi et al¹⁴ compared lomefloxacin and fusidic acid eye-drops in the treatment of 45 patients with presumed acute bacterial conjunctivitis, 31 of whom were culture-positive. Following a loading dose, both antibiotics were instilled twice daily. A significant decrease in clinical symptoms was achieved by both treatments, with a gradual improvement over the treatment period of seven to nine days. There was no significant difference between the general success of therapy in the two groups; however, bacteriological recovery was superior in the lomefloxacin group. An adverse reaction of burning was described in significantly more patients in the fusidic acid group.
In a trial involving 191 subjects with clinically diagnosed bacterial conjunctivitis, Agius-Fernandez et al compared lomefloxacin administered twice daily with chloramphenicol instilled five times a day.¹⁵ Both treatments significantly reduced the conjunctival bacterial colony count score with no difference between the two treatment groups in the 96 patients with bacteriological confirmation of conjunctivitis. The two drugs were equally well tolerated, with no serious systemic or local adverse drug reactions reported. The authors concluded that lomefloxacin 0.3 per cent eye drops instilled twice daily were as effective and well tolerated as chloramphenicol 0.5 per cent eye drops instilled five times daily.
Other trials have compared gentamicin 0.3 per cent, tobramycin 0.3 per cent and norfloxacin 0.3 per cent, administered four times a day with lomefloxacin administered twice daily. Jauch et al have carried out a meta-analysis of all these trials.¹⁶ The authors concluded that lomefloxacin eye-drops, used with a loading dose followed by a twice daily regimen, proved as effective, safe and well tolerated as five established standard treatments used twice, four times or five times times daily. In addition, lomefloxacin caused less discomfort than the other antibiotics upon instillation.
Pooled results comparing in vitro resistance rates for 710 conjunctival isolates to between six and 10 antibiotics was also reported in the meta-analysis. Lomefloxacin showed the lowest rate of resistance at 5.4 per cent, followed by chloramphenicol (9.2 per cent) and norfloxacin (9.7 per cent). The resistance rate for other antibiotics rose from 18.3 per cent with gentamicin to 73.1 per cent with polymyxin (an antibiotic that is not used as a single agent). Fusidic acid, the other antibiotic licensed for twice daily instillation in the treatment of acute bacterial conjunctivitis, was reported as having a resistance rate of 29 per cent. As the authors reported only one partial loss of sensitivity in the lomefloxacin group among surviving bacteria, compared with seven losses (one partial, six complete) in the combined control group, they concluded that lomefloxacin showed a lower risk of generating or selecting new resistant strains.

Ofloxacin Ofloxacin is indicated for the topical treatment in adults and children of external ocular infections, such as conjunctivitis and keratoconjunctivitis, caused by sensitive organisms. Although the summary of product characteristics does not exclude use in neonates, it states that safety and efficacy in the treatment of ophthalmia neonatorum has not been established. A maximum length of treatment of 10 days is recommended.
Ofloxacin has been shown to be as effective as chloramphenicol¹⁷ and gentamicin¹⁸ in 153 and 191 patients, respectively, diagnosed as having external ocular infection. Adverse reactions were reported to be less with ofloxacin in both studies.
Ofloxacin has also been compared with dual antibiotic therapy in the treatment of microbial keratitis. O'Brien et al studied 140 patients with a positive bacterial corneal culture (without fungal or viral infection) to determine the comparative efficacy of ofloxacin eye-drops as a single agent and a combination of cefazolin 10 per cent and gentamicin 1.5 per cent.¹⁹ Results indicated equivalent efficacy between the two groups; however, patients receiving fortified antibiotics reported more severe burning and/or stinging with application of the drops. In a similar study of 118 patients where the comparator was a combination of gentamicin 1.5 per cent and cefuroxime 5 per cent,²⁰ again the authors found equivalence between the two groups with less toxicity in the ofloxacin group.
There have been several reports in Japanese journals of the successful use of ofloxacin 0.3 per cent ointment in the treatment of chlamydial infection.^21-25 An in vitro study showed that the action of ofloxacin was completely lethal when chlamydia-infected cell monolayers were exposed to a concentration of 1g/ml throughout one chlamydial developmental cycle; in contrast, tetracycline inhibited inclusion development at 0.3mg/ml but was only lethal at 2.4mg/ml.²⁶
Despite the publication of several trials reporting the efficacy of fluoroquinolones in the treatment of microbial keratitis, a debate continues as to whether monotherapy with a fluoroquinolone is appropriate.
Bower et al,²⁷ having determined the in vitro antibiotic susceptibilities for 153 isolates, predicted in vitro susceptibility of 98.7 per cent to a fluoroquinolone-cefazolin combination, 97.4 per cent to a fluoroquinolone-gentamicin combination, 88.2 per cent to ofloxacin and 82.3 per cent to ciprofloxacin. On the basis of these in vitro results, they recommend that single-agent therapy is not advisable for vision-threatening bacterial corneal ulcers.
Allan and Dart²⁸ observed that monotherapy might be most appropriate in the developed world where contact-lens related Pseudomonas keratitis was increasingly predominant and pseudomonal resistance to aminoglycosides has become a significant problem. They suggested that a form of combined therapy (eg, fortified penicillin with ofloxacin) might be preferable in developing countries where trauma-related streptococcal keratitis predominated, despite reports of successful treatment of streptococcal infection with fluoroquinolones.^11,19
Goldstein et al²⁹ also listed the gaps in fluoroquinolone coverage for Streptococcus and coagulase-negative Staphylococcus species as a concern over the use of monotherapy in treating bacterial keratitis. In addition, they found a significant increase in the rate of in vitro resistance of Staphylococcus aureus to ciprofloxacin and ofloxacin, from less than 6 per cent in 1993 to 35 per cent in 1997.
The relevance of in vitro resistance to clinical response has been questioned because, as Neu² points out, fluoroquinolone ophthalmic solutions containing 3,000mg/ml are instilled into an eye infected with organisms for which the MICs are commonly of the order of 0.5mg/ml. However, a recently published study from India has shown a link between in vitro resistance and clinical failure to respond in patients with keratitis caused by Pseudomonas isolates. Garg et al³⁰ reported that of 141 culture-proven cases of Pseudomonas keratitis, 22 cases were caused by isolates resistant to ciprofloxacin (mean MIC 43mg/ml). Of the 19 (of 22) cases treated initially with ciprofloxacin, 15 (76.7 per cent) worsened or showed no clinical improvement after three days of intensive therapy and required modification of antibiotic therapy, corneal grafting or evisceration (surgical removal of the contents of the eyeball).
Studies directly comparing the efficacy of the quinolones are sparse. Both ciprofloxacin and ofloxacin were as effective as cefazolin 5 per cent (a standard treatment) in experimental keratitis in the rabbit produced by intrastromal injection of Streptococcus pneumoniae (an organism against which the efficacy of the fluoroquinolones has been questioned).³¹ Lu et al reported a randomised trial involving 217 patients with culture-positive bacterial keratitis in India. They concluded that the efficacy and toxicity of ofloxacin solution in treating bacterial keratitis was equivalent to that of ciprofloxacin solution.³²
Uchida³ reported a multicentre, double-blind trial of lomefloxacin vs ofloxacin in the treatment of bacterial infections of the external eye. He concluded that the results suggest both preparations are outstanding in terms of safety and utility and that neither could be ranked below the other. There was no significant difference in the efficacy of lomefloxacin and ofloxacin measured in terms of clinical signs and symptoms and bacterial eradication in a double-blind study of 100 patients with bacterial conjunctivitis reported by Huguet.³³
New developments in the field of topical ocular antibiotic therapy are currently concentrated on the fluoroquinolone group. In 1996, Herrin¹³ stated that it was likely that the indications for ofloxacin would be extended to include the treatment of corneal ulcers. However, although this has happened in the United States, this indication is still not included in the product licence for the UK market. Herrin also mentioned the imminent introduction of a new formulation of ciprofloxacin, an eye ointment. Markets on both sides of the Atlantic are awaiting this innovation. Newer fluoroquinolones, such as enoxacin, fleroxacin, levofloxacin, perfloxacin and sparfloxacin, are currently being developed as ocular agents.³⁴

Combination products containing polymyxin

Polymyxin B sulphate is used with trimethoprim in Polytrim eye-drops and ointment, with neomycin and gramicidin in Neosporin eye-drops and with bacitracin in Polyfax eye ointment to provide a very broad antibacterial spectrum.
Polymyxin B is active only against Gram-negative aerobic bacteria and is used rarely as a systemic agent because of toxicity.
Trimethoprim has a broad range of antibacterial activity including staphylococci, streptococci, most Enterobacteriaceae and H influenzae. Ps aeruginosa is resistant due to its permeability barrier. The drug is in widespread clinical use as an oral preparation, predominantly for the treatment of urinary tract infection.
Gramicidin and bacitracin are effective against Gram-positive bacteria and the latter is also effective against Neisseria.
Synergistic or additive interaction has been consistently demonstrated between polymyxin and trimethoprim against Enterobacter spp, Klebsiella spp, Serratia spp and Proteus spp, but additive and antagonistic effects were observed with equal frequency against Ps aeruginosa.³⁵
Tirado et al tested Polytrim eye-drops and eye ointment, chloramphenicol eye-drops and eye ointment and fusidic acid eye-drops against 35 bacterial strains including 24 ocular isolates. As eye-drops or ointment, trimethoprim plus polymyxin B had a wider spectrum of activity against clinical isolates than chloramphenicol or fusidic acid. It was the only preparation with anti-pseudomonal activity.³⁶
Gibson reported the success rate of combinations of polymyxin and other antibiotics in treating 230 patients with bacterial conjunctivitis in a double-blind, randomised trial.³⁷ Trimethoprim plus polymyxin B was similar to that of a combination of polymyxin, gramicidin and neomycin (Neosporin) and significantly better than chloramphenicol in treating the conjunctivitis. In this trial, only one patient had a pre-treatment isolate of Pseudomonas spp. Behrens-Baumann et al³⁸ reported a smaller trial comparing the efficacy of trimethoprim-polymyxin B and chloramphenicol ophthalmic ointments in the treatment of bacterial conjunctivitis. They found no significant differences between them with regard to eradication of organisms or clinical improvement.
Neosporin eye ointment applied four times a day was compared with fusidic acid 1 per cent eye-drops twice a day in a trial involving 68 patients with external eye infections.³⁹ Clinical cure rate was equal in both groups at 88-89 per cent, but four patients (16 per cent) in the Neosporin group reported discomfort, while good tolerability was reported in all patients in the fusidic acid group.
Ophthalmic preparations containing polymyxin appear to be well tolerated. Stinging encountered with Polytrim led to the withdrawal of four patients (3.3 per cent) from Gibson's trial.³⁷ Three patients (7 per cent) in the Polytrim group withdrew from Behrens-Baumann's trial because of stinging, increased transient grittiness, conjunctival hyperaemia and periorbital oedema.³⁸ Adverse reactions to Neosporin were similarly small, with only one patient in 52 (1.9 per cent) treated withdrawing from Gibson's trial due to the development of bilateral periorbital oedema.
There are some differences in the antibacterial spectra of the three combinations (see Table 1). Polytrim has the broadest spectrum of activity but is ineffective against Neisseria and Corynebacterium species.
Combination eye-drops containing polymyxin are preserved with thiomersal, allowing use in patients allergic to the more commonly used preservative benzalkonium chloride.

Other antibiotics

Despite the range of commercially available topical ocular antibiotics available, ophthalmologists frequently request topical formulations of other antibiotics. A survey conducted in 1999 by the author of 37 ophthalmic centres in the United Kingdom showed that the majority used antibiotics that are not commercially available. Those used are shown in Figure 1.

Figure 1: Use of non-commercially available topical antibiotics. Key: a = amikacin, b = benzylpenicillin, c = cefuroxime, d = ceftazidime, e = cefazolin, f = ciprofloxacin (preservative free), g = erythromycin, h = gentamicin (forte), i = teicoplanin, j = vancomycin

Fortified gentamicin (1.4 or 1.5 per cent) is frequently combined with cefuroxime in the treatment of bacterial keratitis, while benzylpenicillin is used in the treatment of infections caused by Neisseria gonorrhoea and Streptococcus pneumoniae. Vancomycin and teicoplanin are useful agents against Gram-positive bacteria while ceftazidime is used as an anti-pseudomonal. Erythromycin, used in the form of an eye ointment, is an alternative to ocular tetracycline in the treatment of infections caused by Gram-positive organisms. Ocular erythromycin is also used in combination with systemic tetracycline therapy for chlamydia. Amikacin has been shown to be a useful antibiotic when Pseudomonas with multiple drug resistance is isolated.³⁰ A request for a preservative-free presentation is another reason for in-house preparation.
Unfortunately, it is unlikely that these agents will become commercially available because usage is low, many of these antibiotics are unstable in solution, and preservative-free formulations require the use of expensive single dose presentations.

Mrs Titcomb is directorate pharmacist, ophthalmology, Birmingham and Midland eye centre, City hospital NHS Trust, Birmingham