Approximately one in eight children has an episode of acute conjunctivitis annually. The condition is a common cause of telephone calls or visits to primary care physicians, and conventional therapy using topical antibiotics incurs expense and may contribute to antibiotic resistance. In addition, antibiotic therapy has been shown to have limited benefit in many common childhood infections, including otitis media and pharyngitis, raising questions about its use in conjunctivitis. Rose and colleagues conducted a blinded trial of chloramphenicol (Chloromycetin) versus placebo in children with acute conjunctivitis.

Family physicians in 12 participating practices recruited children six months to 12 years of age based on clinical presentation of acute infective conjunctivitis. The only exclusions were allergy to chloramphenicol; use of an antibiotic within the previous 48 hours; immunocompromised state; or evidence of severe infection, particularly periorbital cellulitis.

At baseline, a research nurse assessed the severity of the infection using standardized photographs and obtained conjunctival swabs of the affected eye. Parents were then given eye drops and instructions to give their child one drop every two hours for the first 24 hours when the child was awake, then four times daily until 48 hours after the infection had resolved. The children were randomly assigned to treatment with 0.5 percent chloramphenicol or distilled water containing boric acid and borax. Parents recorded symptoms with every treatment until the condition resolved. Participants were reassessed at seven days, and new conjunctival swabs were obtained. After six weeks, any additional visit to the family physician was recorded, and parents were telephoned to identify any further eye problems. The primary outcome was clinical cure as determined by parents at seven days. The study also documented time to resolution of symptoms and microbiologic cure.

The 163 children treated with antibiotics did not differ in any significant respect from the 163 who were assigned to placebo. Pathogens were identified in 78 percent of the antibiotic group and 77 percent of the placebo group. In 250 children, Haemophilus influenzae accounted for 60 percent of pathogens and Streptococcus pneumoniae accounted for 20 percent. Based on intention-to-treat analysis, clinical cure was achieved at seven days in 86 percent of children treated with antibiotics and 79 percent of those assigned to placebo. Excluding nine children who were lost to follow-up, clinical cure rates at seven days were 86 percent in the antibiotic group and 83 percent in the placebo group. The number needed to treat to achieve one additional clinical cure by day 7 was estimated as 14 to 25 children. The groups did not differ significantly in clinical or microbiologic cure rates or median time to cure, even when the analysis was restricted to children with demonstrated pathogens. Over the six weeks of observation, only one adverse event was recorded. Further episodes of conjunctivitis within six weeks occurred in seven children (4 percent) in the antibiotic group and in five children (3 percent) in the placebo group. Six percent of children from each group consulted their family physicians for other minor problems during the follow-up period.

The authors conclude that symptoms of acute infective conjunctivitis resolve without topical antibiotic therapy in most children. Although in this trial children treated with chloramphenicol had an advantage at day 7 in symptoms and bacteriologic eradication, these results were not clinically or statistically significant. The authors recommend that most cases of conjunctivitis be managed by parents through good hygiene and symptom control, thus avoiding use of medical services and antibiotics.