An organism-based algorithmic approach to Gram-negative bacterial keratitis: a narrative review of characteristics and antibiotic susceptibilities

INTRODUCTION Gram-negative bacterial keratitis is responsible for a significant portion of unilateral blindness around the globe. While vancomycin exists to cover Gram-positive bacteria with little resistance, emerging Gram-negative antibiotic resistance patterns challenge whether typical Gram-negative coverage options such as aminoglycosides or fluoroquinolones are sufficient for empiric therapy in all cases.AREAS COVERED The most common, emerging, and rare Gram-negative organisms involved in bacterial keratitis are identified in this review. While several distinct risk factor associations emerged with certain bacteria, such as contact lens-use, nosocomial exposures, pediatric populations, systemic disease associations, and strong ocular disease associations, clinical characteristics were rarely helpful for reliable Gram-negative organism based on exam alone. Several organisms displayed multi-drug resistance in isolated cases, but emerging Gram-negative bacteria such as Stenotrophomonas and Achromobacter demonstrated high resistance rates to commonly used empiric therapies. Finally, a general algorithm is introduced, with which, by organism, clinicians may guide empiric treatment.EXPERT OPINION While many Gram-negative bacterial keratitis cases typically respond to common empiric aminoglycosides or fluoroquinolones, treating ophthalmologists should be aware of several organisms that may be better treated with other antibiotic categories. An algorithmic approach to Gram-negative bacterial keratitis while considering local geographic patterns may help clinicians optimize outcomes.