to the editor: I am a new practicing physician in northern Wisconsin and would like some guidance on the correct treatment of conjunctivitis. During my training in family medicine, both visiting with ophthalmologists and pediatricians, it became apparent that all cases of conjunctivitis were treated with topical antibiotics regardless of whether the suspected cause was viral or bacterial. I have seen many children in my office who present with the classic symptoms for viral “pink eye”; the parents of these patients have been told by the day care providers that their child needs an antibiotic in order to return to the day care setting. I am not sure if this is a local or a national problem with the day care providers. Physicians also receive pressure from parents because their children have previously been placed on eye drops for pink eye. When did day care providers begin requiring a child to be on a topical antibiotic for what appears to be viral pink eye before allowing the child to return?

A recent Cochrane review in American Family Physician¹ suggests that we should prescribe topical antibiotics for all affected patients because of the low risk from treatment. However, what should be done about the more common viral pink eye? It is frustrating as a physician to convince parents and day care staff that an antibiotic is not helpful in these cases and should not be used. I would welcome your advice and direction.

in reply: I would like to thank Dr. Bennett for his letter regarding the conflict between the pressure to treat all cases of conjunctivitis (viral and bacterial) with topical antibiotics. As a family physician who has treated many cases of conjunctivitis in infants and toddlers, I can understand your dilemma. You are not alone in your approach to the treatment of viral conjunctivitis. The extant evidence does not warrant the use of topical antibiotics for cases of low-risk viral conjunctivitis, such as cases that are not caused by herpes simplex virus or herpes zoster; however, a recent survey¹ of British general practitioners found that 95 percent of them prescribed topical antibiotics despite their belief that at least one half of the cases they saw were of a viral origin.

Unfortunately, evidence-based guidelines on the treatment of viral conjunctivitis either provide no recommendation or recommend supportive treatment for viral conjunctivitis. The American Optometric Society recommends the use of cold compresses, lubricants, and ocular decongestants.² However, the guidelines of the American Academy of Ophthalmology³ and the British Public Health Laboratory Service⁴ do not offer specific recommendations for treatment of presumptive nonherpetic viral conjunctivitis. Considering the practical realities of parent expectations of antibiotic treatment for all cases of conjunctivitis among children in day care, the low risk of treatment with topical antibiotics, and unclear policy guidelines, it would seem reasonable to pursue one of the following options. One option would be to provide parents with information about the self-limited nature of most cases of conjunctivitis without treatment and the lack of evidence for benefit from topical antibiotics in cases of viral conjunctivitis, and then allow the parents to choose. This is a reasonable option for practical reasons and because a large proportion of primary care physicians are uncertain about whether low-risk cases are viral or bacterial in origin.¹

A second option would be to offer a delayed prescription approach by prescribing antibiotics only for those cases that have not resolved by day 5.⁵ In one qualitative study⁶ of adult patients with conjunctivitis, most patients preferred to wait a few days for treatment once they were made aware of the self-limiting nature of conjunctivitis. However, it is unclear whether parents would take the same attitude toward their children in day care who could not return to day care without having begun antibiotics.

A third approach would be to combine clinical judgment with your knowledge of the parents' expectations and make individual recommendations given your level of certainty of the diagnosis and implications on the physician-patient relationship of your treatment decisions.