Pearl of the Month

Does this patient have bacterial conjunctivitis?


A 54-year-old pharmacist with a history of gout, hypertension, and conjunctivitis presents for evaluation of pink eye in the summer. The morning before coming into the office, he noticed that his right eye was red and inflamed. He self-treated with saline washes and eye drops, but upon awakening the next day, he found his right eye to be crusted shut with surrounding yellow discharge. He has not had any changes to his vision but endorses a somewhat uncomfortable, “gritty” sensation. He reports no recent cough, nasal congestion, or allergies, and he has not been around any sick contacts. His blood pressure is 102/58 mm Hg, pulse is 76 bpm, and body mass index is 27.3 kg/m2. His eye exam reveals unilateral conjunctival injections but no hyperemia of the conjunctiva adjacent to the cornea. Mucopurulent discharge was neither found on the undersurface of the eyelid nor emerging from the eye. Which of the following is the best treatment for this patient’s condition?

A) Erythromycin 5 mg/gram ophthalmic ointment.

B) Ofloxacin 0.3% ophthalmic drops.

C) Antihistamine drops.

D) Eye lubricant drops.

E) No treatment necessary.

This patient is an adult presenting with presumed conjunctivitis. Because he is presenting in the summer without observed purulent discharge, his condition is unlikely to be bacterial. This patient does not need treatment, although eye lubricant drops could reduce his discomfort.

Nearly 1% of primary care office visits1 and 300 million in annual costs2 are spent evaluating and treating “pink eye.” After ruling out serious eye disease, clinicians need to determine which cases of suspected conjunctivitis are most likely to be bacterial to allow for judicious use of antibiotic eye drops. This is an important undertaking as most patients assume that antibiotics are needed.

McKenzie Momany

How do we know which history and clinical exam findings to lean on when attempting to categorize conjunctivitis as bacterial or not? If a patient reports purulent discharge, doesn’t that mean it is bacterial? Surprisingly, a systematic review published in 2016 by Narayana and McGee found that a patient’s self-report of “purulent drainage” is diagnostically unhelpful, but if a clinician finds it on exam, the likelihood of a bacterial etiology increases.3

Narayana and McGee analyzed three studies that enrolled a total of 281 patients with presumed conjunctivitis who underwent bacterial cultures. They then determined which findings increased the probability of positive bacterial culture. From strongest to weakest, the best indicators of a bacterial cause were found to be: complete redness of the conjunctival membrane obscuring tarsal vessels (the vessels visible on the inside of everted upper or lower eyelids) (likelihood ratio, 4.6), observed purulent discharge (LR, 3.9), matting of both eyes in the morning (LR, 3.6), and presence during winter/spring months (LR, 1.9). On the other hand, failure to observe a red eye at 20 feet (LR, 0.2), absence of morning gluing of either eye (LR, 0.3), and presentation during summer months (LR, 0.4) all decreased the probability of a bacterial cause. This review and different study by Stenson et al. unfortunately have conflicting evidence regarding whether the following findings are diagnostically helpful: qualities of eye discomfort (such as burning or itching), preauricular adenopathy, conjunctival follicles, and conjunctival papillae.3,4 Rietveld and colleagues found that a history of conjunctivitis decreased the likelihood of bacterial conjunctivitis.5

Dr. Doug Paauw

Ultimately, if the former indicators are kept in mind, primary care clinicians should be able to decrease the prescribing of topical antimicrobials to patients with non-bacterial conjunctivitis.

Pearl: The best indicators of a bacterial cause in patients with presumed conjunctivitis are complete redness of the conjunctival membrane obscuring tarsal vessels, observed purulent discharge, and matting of both eyes in the morning. Presentation during the summer months and having a history of conjunctivitis decreases the likelihood of bacterial conjunctivitis.

Ms. Momany is a fourth-year medical student at University of Washington, Seattle. Dr. Paauw is professor of medicine in the division of general internal medicine at the University of Washington and serves as third-year medical student clerkship director at that university. Contact Dr. Paauw at [email protected].


1. Azari AA and Barney NP. JAMA. 2013 Oct 23; 310(16):1721-9.

2. Smith AF and Waycaster C. BMC Ophthalmol. 2009 Nov 25. doi: 10.1186/1471-2415-9-13.

3) Narayana S and McGee S. Am J Med. 2015;128(11):1220-4.e1.

4) Stenson S et al. Arch Ophthalmol. 1982;100(8):1275-7.

5) Rietveld RP et al. BMJ. 2004 Jul 24;329(7459):206-10.

Next Article: