Topical antibiotics and artificial tears associated with reduced infective-conjunctivitis symptoms

Abstract

Question

Among children with infectious conjunctivitis, what is the therapeutic efficacy of topical antibiotics, compared with placebo or no antibiotics, in conjunctival symptom-duration reduction?

Design

Randomized controlled trial (RCT) and meta-analysis (MA).

Setting

RCT: Primary health care in Oulu, Finland.

Participants

RCT: 6 months – 7 years with acute infective conjunctivitis. MA: 1 month – 18 years.

Intervention

RCT: Moxifloxacin eye drops, placebo (carboxymethylcellulose sodium – artificial tears) eye drops, or no intervention.

Outcomes

RCT: Days to clinical cure. MA: Proportion of participants with conjunctival symptoms on days 3 to 6.

Main Results

RCT: 30, 27, and 31 participants were randomized to moxifloxacin, placebo, and no intervention, respectively. Moxifloxacin, compared with no intervention, significantly reduced time to cure, 3.8 vs 5.7 days, difference, −1.9 days (95%CI, −3.7 - −0.1 days). Survival analysis demonstrated that both moxifloxacin and placebo (artificial tears) significantly shortened the time to clinical cure relative to no intervention. MA: Antibiotics reduced the proportion of participants with conjunctival symptoms on days 3 to 6, odds ratio 0.59 (95%CI, 0.39 – 0.91).

Conclusions

Topical antibiotics, and artificial tears, reduced conjunctivitis symptom duration.

Commentary

Spontaneous resolution of acute infectious conjunctivitis is a common clinical observation. Inconsistencies in the literature on the effects of antibiotic therapy on reducing severity or duration of symptoms, however, have sustained the long-term debate whether such therapy is indicated, and if so, its appropriate duration. The current trial sought to address the potential antiseptic effect of placebo when comparing treatment effects of moxifloxacin 0.5% with carboxymethylcellulose sodium (an artificial tear) by including a third group that received no treatment. Results of pair-wise comparisons revealed that placebo containing non-antiseptics may still provide therapeutic benefits by presumably reducing local pathogen loads. As identified in a recent updated Cochrane review,¹ the main sources of heterogeneity exist in the trial design and conduct, contributing to the inconsistent evidence observed. These heterogeneities include a variety of antibiotic classes (fluoroquinolone or non-fluoroquinolone); different timepoints at which outcomes are assessed relative to duration of treatment (“end-of-therapy” or “test-of-cure”); and different analytic approaches (intention-to-treat (ITT) or modified ITT population). Quality head-to-head comparisons are needed to ensure fair assessment of clinical efficacy whether by comparing different classes of antibiotic drugs or different durations of the same treatment.