Infectious Conjunctivitis

Antibiotics for Infectious Conjunctivitis

What is the problem and what is known about it so far?
Infectious conjunctivitis - also known as "pink eye" - is a common illness with symptoms including red itchy eyes, eye discharge (thick or watery), eyelid swelling and sensitivity to light.  Conjunctivitis may be caused by infection with viruses and/or bacteria.  Regardless of the cause, nearly all cases get better with time.  Still, people with conjunctivitis often seek medical attention for relief of symptoms. This is particularly true of parents whose children are barred from daycare because of pinkeye.

Studies have shown that although the majority of cases are caused by viruses, it is difficult to tell the difference between viral and bacterial conjunctivitis.  Because laboratory evaluation of conjunctivitis takes time and incurs costs, practitioners often prescribe antibiotic eye drops without proof of bacterial infection.  Although there is little evidence that antibiotic therapy helps treat infectious conjunctivitis most hope symptoms will improve more rapidly with antibiotics. This practice has its own costs, including pharmacy costs, the risk of adverse medication effects and antibiotic resistance, and reinforcement of "medicalization" (the cultural phenomenon of turning to the medical profession for more and more of life’s ailments, driving up healthcare costs). 

Why did the researchers do this particular study?
The researchers sought to determine if early or delayed therapy with antibiotics affected the severity or duration of infectious conjunctivitis.  They further sought to determine if education or outcomes would alter parental beliefs about the necessity for antibiotics to treat infectious conjunctivitis. 

Who was studied?
Children over 1 year of age and adults who presented to 30 family medicine offices in England between April 2001 and April 2005 with acute infectious conjunctivitis were recruited to the study.   Slightly more than half of the 307 participants were under the age of 12.  Patients were excluded from the study if they had a concurrent illness requiring oral antibiotics, had received antibiotics in the previous two weeks, had chronic infectious eye disease, had undergone eye surgery within one month, or were allergic to chloramphenicol (the antibiotic eye drop).  These exclusions were chosen because the researchers wanted to focus their study on typical cases of infectious conjunctivitis.

How was the study done?
To address the question of whether treatment of conjunctivitis could influence patient beliefs,  the study was designed to follow both patients’ perceptions of their illness and treatment as well as the effect of different treatment strategies on symptoms.  Study participants were randomly assigned to receive one of three main interventions. The first group began treatment immediately with chloramphenicol (antibiotic) eye drops.  The second group was given a prescription for chloramphenicol eye drops which parents or adult patients could chose to have filled three days after the initial visit.  The third group received neither drops nor a prescription (the control group).  Half of the participants in each main group were given an information leaflet about infectious conjunctivitis; this produced six different study groups.  In each of these six groups, half of the participants had their eyes swabbed for culture; this division resulted in the final 12 study groups.

Participants kept diaries for 14 days, recording data used to evaluate main outcomes: 1) duration of moderately bad symptoms, 2) average severity symptoms over the first three days of the study, and 3) belief in the effectiveness of antibiotics for eye infections.  Data from all groups were compared using a variety of statistical models.

What did the researchers find?
In terms of antibiotic usage, 99% of the immediate antibiotic group, 53% of the delayed antibiotic group, and 30% of the control group used antibiotic drops.  This means that almost one third of patients who were not supposed to get antibiotics did, presumably by obtaining prescriptions from other doctors.

There were no differences in symptom severity between groups over the first 3 days.  Antibiotics did appear to slightly reduce the average duration of moderate symptoms: 3.3 days for the immediate antibiotics group, 3.9 days for the delayed antibiotics group and 4.8 days for the control group.  There was no relationship between having a positive bacterial culture and duration or severity of symptoms. 
Patient education leaflets and eye swab cultures did not affect the primary study outcomes.  Leaflets did improve patient satisfaction in terms of information received and degree to which concerns were addressed.  Obtaining an eye swab for culture appeared to increase patient worries.

Both the immediate antibiotic group and the control group indicated they were more likely to seek medical attention for recurrent conjunctivitis compared with the delayed antibiotic group.  The immediate antibiotic group was more likely than the control group (but not the delayed antibiotic group) to believe antibiotics were effective.  Taken together, these observations are consistent with the idea that patients projected their experiences of treatment toward their understanding of what the ideal treatment would be. 

What were the limitations of the study?
The main limitation of the study involves the possibility of selection bias. The general practitioners who invited participants only did so with one of every three patients they saw. It could be that they opted to provide standard antibiotic treatment for patients whom they thought were more likely to have bacterial conjunctivitis; this would leave fewer bacterial conjunctivitis cases in the study, making antibiotic treatment look less effective.  There were imperfect response rates for diary completion, raising the possibility of bias.  A specific limitation for practitioners in the United States is that chloramphenicol is rarely used in this country to treat conjunctivitis. 

What are the implications of the study?
Consistent with other studies, the authors found that antibiotic therapy did not affect the severity of infectious conjunctivitis.  From a societal point of view, the authors conclude that delayed antibiotic appears to be the best strategy for most cases of infectious conjunctivitis.  Compared with immediate treatment with antibiotics, symptom control was similar, and there was a near 50% reduction in antibiotic use.  Furthermore, with the delayed antibiotic strategy, there was a decreased rate of seeking medical attention for subsequent conjunctivitis symptoms.  The authors interpret this finding as empowering patients against medicalization (giving them the knowledge and skills to care for themselves).

Summarized by Paul Rosenau, MD, Pediatrics Residency, University of Vermont.
Summarized from "A Randomised Controlled Trial of Management Strategies for Acute Infective Conjunctivitis in General Practice." Everitt H.A. et al. British Medical Journal, August 12, 2006, Vol. 333, page 321.