I read with great interest the article by Alabiad and associates entitled, “Antimicrobial Resistance Profiles of Ocular and Nasal Flora in Patients Undergoing Intravitreal Injections.” I would like to commend the authors for trying to address an important issue. I have some comments about their study design that may explain their conflicting results with the Antibiotic Resistance of Conjunctiva and Nasopharynx Evaluation (ARCANE) Study.
The ARCANE Study was a prospective, randomized, longitudinal study that enrolled patients with unilateral choroidal neovascularization without a prior history of intravitreal injection or recent antibiotic exposure. These inclusion criteria were necessary to minimize bias, to allow for greater clarity in observing trends, and to determine cause and effect. Study participants were randomized to 1 of 4 commercially available antibiotics (ofloxacin 0.3%, azithromycin 1%, gatifloxacin 0.3%, or moxifloxacin 0.5%) and used only their assigned antibiotic after each injection. The fellow untreated eye served as a control. Each participant was mandated to receive 4 consecutive monthly injections, and therefore had a minimum of 5 culture results (baseline plus the 4 after each injection), but some participants had as many as 13 culture results (baseline plus 12) if they received monthly injections throughout the 1-year course of follow-up. Bilateral conjunctival and unilateral nasal cultures on the treatment side were obtained at each visit before the application of topical medications. All study participants received a single application of their assigned antibiotic after each injection and were given a new sample of their antibiotic at each visit, along with written instructions on its use to maximize compliance. The baseline, short-term, and 1-year results of the ARCANE Study are published and demonstrate increased antimicrobial resistance with repeated short-term use of ophthalmic antibiotics.
In contrast, the study by Alabiad and associates seems to be cross-sectional in design, and thus is unable to establish cause and effect. Another major limitation (which the authors acknowledge) is the administration of topical proparacaine hydrochloride 0.5% and other mydriatic medications before culturing, which is problematic because of the bactericidal properties of proparacaine and benzalkonium chloride 0.01%. For this reason, all cultures in the ARCANE Study were obtained before the administration of any topical medication. It is also unclear to what degree patients were compliant with their use of moxifloxacin or the consistency in which they were instructed. Extraordinary measures were taken in the ARCANE Study to maximize compliance, including making replacement samples available. Furthermore, some eyes received several injections, whereas other eyes had not been treated previously, and no detailed information is provided on the time course of previous injection and thus antibiotic exposure. It therefore is difficult to observe definite trends given this lack of uniformity. To avoid this problem, eyes with a history of intravitreal injection were excluded from the ARCANE Study.
In sum, the lack of correlation between use of moxifloxacin and development of antimicrobial resistance observed by Alabiad and associates must be interpreted in the context of the study’s limitations. Nevertheless, the study confirms the increasing prevalence of fluoroquinolone resistance among ocular and nasal flora and re-emphasizes the need for more rational use of ophthalmic antibiotics to reduce the emergence of antimicrobial resistance.