Sterile Mucopurulent Conjunctivitis Associated With the Use of Dorzolamide Eyedrops

Dorzolamide hydrochloride, a nonbacteriostatic sulfonamide derivative, is the first topical carbonic anhydrase inhibitor clinically available. It reduces intraocular pressure by inhibiting aqueous humor production in the ocular ciliary processes.¹

The most common adverse effects associated with dorzolamide use are bitter taste and local ocular complaints such as burning, blurred vision, itching, tearing, foreign body sensation, stinging, eyelid discomfort, and nonspecific conjunctival hyperemia.^1,2 As dorzolamide represents a completely new class of topical ocular drugs, its spectrum of side effects is not yet as well defined as for older drugs. With the increased use of dorzolamide, it is important to be aware of possible unreported side effects that will invariably occur. We report the unusual finding of a severe sterile purulent conjunctivitis in 7 patients using dorzolamide drops that resolved immediately after discontinuation of treatment. We believe that this condition is most probably attributed to the use of this drug.

A 73-year-old male patient had been treated bilaterally with 0.5% betaxolol hydrochloride twice a day and 2% pilocarpine 3 times a day during several years for chronic angle-closure glaucoma after an acute attack of angle closure in the right eye. Because of bradycardia of unknown origin, treatment with betaxolol was discontinued and replaced by 2% dorzolamide hydrochloride (Trusopt) drops. Other topical medications included occasional fucidic acid gel for chronic blepharitis and, once to twice a year, additional 0.1% fluorometholone drops for episodes of marginal keratitis. One month after the change of therapy, the patient returned, complaining of red eyes with a bilateral purulent discharge that was first ascribed to recrudescence of blepharitis. There was a diffuse papillary reaction in the inferior palpebral conjunctiva and cul-de-sac that was less prounouced in the superior palpebral conjunctiva as well as hyperemia of the bulbar conjunctiva that was preponderant inferiorly. Treatment with fucidic acid gel was temporarily discontinued to perform conjunctival bacterial cultures, according to standard ophthalmologic microbiologic practice, which did not show any growth of organisms. A 10-day course of a fixed combination of 0.1% dexamethasone phosphate and 0.5% chloramphenicol drops was given in addition to the long-term fucidic acid gel treatment. One month later, the patient consulted us again and complained of the same, persisting symptoms. The purulent discharge was massive and the cul-de-sac was thickened with rugae (Figure 1). A second bacterial culture was again negative. Pilocarpine drops were discontinued and replaced by systemic acetazolamide sodium, 500 mg twice a day, because a cataract operation was planned. Several courses of antibiotic drops, including a neomycin sulfate– polymyxin B sulfate–gramicidin combination, the fluoroquinolone 0.3% lomefloxacin hydrochloride, and again 0.5% chloramphenicol, had absolutely no effect on the purulent conjunctivitis. The planned cataract operation had to be postponed 3 times. It was only after the discontinuation of the dorzolamide drops that rapid resolution of all symptoms within 2 days occurred, with progressive regression of signs. The cataract of the right eye was operated on uneventfully 1 week later. The patient remained without recurrence during a follow-up of 2 years.

A 77-year-old pseudophakic patient was sucessfully treated for glaucoma with 0.5% timolol maleate (Timoptic) for more than 2 years. Six months before his referral, 0.5% timolol maleate drops had been replaced by dorzolamide drops because of "cardiovascular problems." The patient had also been treated for blepharitis and meibomitis in the past. The reason for his referral was a bilateral refractory purulent conjunctivitis present for the last 4 months that had been treated with 4 different antibiotics, including fucidic acid, lomefloxacin, chloramphenicol, and a combination of neomycin and bacitracin. Bacterial cultures had been performed twice after a 48-hour antibiotic washout period. The initial swab revealed no bacterial growth and the second swab showed only saprophytic organisms (Staphylococcus epidermidis and Corynebacterium). On examination, a thickened inferior palpebral conjunctiva with a prominent papillary reaction was noted. Some papillae were very large, having the size of follicles, and there were rugae in the inferior cul-de-sac with prominent purulent secretions. Dorzolamide drops were discontinued and replaced by systemic acetazolamide. Chloramphenicol drops that had been prescribed by his ophthalmologist were continued. The purulent discharge and symptoms resolved almost completely within 36 hours with progressive regression of the other signs and no recurrence during an 11-month follow-up.

An 81-year-old female patient was treated for 2 years with 2% carteolol hydrochloride twice a day in both eyes for ocular hypertension; dorzolamide was added in her left eye because of uncontrolled intraocular pressure. Three months after initiation of dorzolamide treatment, she complained of occasional redness and tearing in the left eye that was attributed to a moderate ectropion of the inner third of the inferior lid; a plasty of the lacrimal punctum was performed with resolution of symptoms. Several months later, she complained of stinging and purulent discharge in her left eye. On examination, there was an inferior bulbar and palpebral hyperemia and a thickened conjunctiva without a frank papillary reaction but the presence of purulent secretions. The cornea was clear and no dermatitis was noted. Conjunctival bacterial cultures were negative. Dorzolamide treatment was interrupted and all symptoms resolved within 24 hours and signs resolved within 3 days.

Four additional bilateral cases that received bilateral dorzolamide therapy, for which no bacterial examinations were performed, were diagnosed soon after we had been aware of the described side effect. The durations of dorzolamide treatment until occurrence of the adverse clinical signs were 12, 21, 21, and 22 months. The duration until dorzolamide discontinuation was 2 months in two patients, 3 months in one patient, and 4 months in another patient. All patients had been treated unsuccessfully by several courses of topical antibiotics and at least 1 course of corticosteroid eye preparation. Resolution of symptoms and disappearance of the mucopurulent discharge occurred in all 4 patients 24 to 72 hours after discontinuation of topical dorzolamide treatment, with progressive regression of signs.

Dorzolamide is a topically active carbonic anhydrase inhibitor used in a 2% ophthalmic solution for the treatment of elevated intraocular pressure in patients with ocular hypertension or glaucoma.¹ It may be used as first-line therapy for patients who are unable to tolerate β-blockers. It can also be used as an add-on therapy when more than one drug is needed. Dorzolamide is generally well tolerated and has few adverse effects. Mild stinging and burning on instillation, resolving within minutes, and a bitter taste following instillation have been reported in up to 27% of patients.^1,2 Allergic reactions, mainly nonspecific conjunctival irritation and lid reactions, have been described in 10% to 15% of patients. White granular deposits on the bottle tip have also been reported.³

We describe herein an unreported adverse effect due to dorzolamide treatment occurring in 7 patients. The well-defined clinical picture of a sterile purulent conjunctivitis developed gradually in all patients after several weeks or months of dorzolamide treatment and all patients were mistakenly treated for a bacterial conjunctivitis for 2 to 4 months. The following elements represent strong arguments for a relation between the clinical picture reported and dorzolamide use: (1) the prompt and spectacular improvement in all 7 patients after the discontinuation of dorzolamide, (2) the unilateral involvement in the patient who was treated unilaterally with dorzolamide, and (3) the repeatedly negative bacterial cultures in the 3 first patients. The presence of the preservative benzalkonium chloride in other drops that were well tolerated by all patients eliminates this substance as a possible factor for the reported side effect. The absence of itching and dermatitis as well as the gradual developement of symptoms over months speaks more for a toxic type of conjunctival involvement rather than a predominantly allergic reaction. Although allergic and toxic conjunctivitis are theoretically different clinical entities,⁴ it is not always easy to sort out the contribution of toxic or immunologic mechanisms in clinical practice. Instead of discussing the mechanism involved, we stress the well-characterized findings that all patients had in common. Dorzolamide is used predominantly in an elderly glaucomatous population in which blepharitis is a common finding. The clinical picture described herein might well be ascribed to an infectious conjunctivitis related to blepharitis, and thus may lead to prolonged, unnecessary antibiotic use as was the case for all our patients. Because of the delayed onset of adverse signs, after prolonged dorzolamide use, the link to dorzolamide is not always obvious. It is therefore important that the clinician be aware of this adverse effect, which should be suspected in cases of recalcitrant and unexplained purulent conjunctivitis in patients taking dorzolamide drops.

Corresponding author: Carl P. Herbort, MD, PD, Department of Ophthalmology, University of Lausanne, 15 Ave de France, CH-1004 Lausanne, Switzerland.