Bacterial Keratitis

Bacterial keratitis is an urgent corneal infection requiring prompt recognition, laboratory evaluation, and treatment.¹ Inflammatory corneal disease — predominantly bacterial — is one of the leading causes of monocular blindness worldwide, with bacterial organisms found in roughly 80% of ulcerative keratitis cases.² Severe visual loss from ulceration and opacity may already be present at diagnosis and can persist despite appropriate treatment, which is why topical fortified antibiotics remain the urgent, empiric mainstay of therapy.³¹

Risk factors and causative organisms differ meaningfully between high-income and low/middle-income countries, so general management guidelines aren't universally transferable.³ Contact lens wear and refractive surgery have grown in importance as risk factors in some studies, while trauma-related cases appear less frequent than historically — contact lens-related and post-refractive-surgery infections are particularly associated with Pseudomonas species and atypical mycobacteria.² Bacterial invasion requires a breach in the corneal epithelial barrier; bacteria then proliferate and trigger both direct microbial damage and an intense host immune response contributing to tissue destruction. Clinical severity ranges widely, from small superficial infiltrates to necrotizing disease, and Pseudomonas aeruginosa classically produces a rapidly progressive, invasive-appearing ulcer.⁴⁵

Patients present acutely with eye pain, photophobia, decreased vision, purulent discharge, and conjunctival injection. A critical, easily-missed point: the idea that bacterial keratitis can only be "confirmed" by a positive culture is misleading — roughly 50% of clinically typical cases are culture-negative, and the etiology of these culture-negative cases remains unknown. A negative culture does not exclude the diagnosis or justify withholding treatment.³

Key differentials: fungal keratitis (often worse outcomes than bacterial ulcers), Acanthamoeba keratitis (diagnosable with in vivo confocal microscopy), viral (especially HSV stromal) keratitis, sterile inflammatory keratitis, autoimmune corneal disease, exposure/neurotrophic keratopathy with secondary infection, and marginal keratitis related to staphylococcal hypersensitivity.⁵

Exam reveals a corneal epithelial defect with underlying stromal infiltrate, stromal suppuration and opacity, anterior chamber reaction with hypopyon in severe cases, conjunctival injection/chemosis, mucopurulent discharge, and surrounding corneal edema. Infiltrate morphology varies by organism; central location, depth, and size have prognostic significance.⁵ Culture remains the prevailing diagnostic standard, and newer molecular techniques — including next-generation sequencing — are expanding what's identifiable, especially for organisms hard to culture conventionally, though they don't yet replace established methods.⁵²

Topical fortified antibiotic drops are first-line, and a recent review found that commonly prescribed topical antibiotics are roughly equally effective against bacterial keratitis, supporting empiric broad-spectrum coverage while culture is pending.¹⁵ Current susceptibility estimation is often based on systemic dosing data rather than the drug concentrations actually achievable at the corneal surface — a real limitation, since better minimum-inhibitory-concentration data specific to corneal tissue would meaningfully improve treatment selection.³ Broad empiric fluoroquinolone use is thought to be contributing to shifts in the microbial spectrum, and rising antimicrobial resistance is a real, if not yet fully quantified, concern.²³

The Steroids for Corneal Ulcers Trial (SCUT) found that adjunctive topical corticosteroids gave no significant overall benefit for bacterial corneal ulcers, but subgroup analysis suggested benefit in ulcers that were central, deep, or large; non-Nocardia; associated with low baseline vision; or when steroids were started early after antibiotics — these are trial subgroup findings, not established universal eligibility criteria, and steroids should still be avoided in Nocardia keratitis specifically.⁵

• Approximately 50% of cases clinically consistent with bacterial keratitis will be culture-negative; negative cultures do not exclude the diagnosis³
• Risk factors and bacterial isolates vary significantly between high-income and low/middle-income countries; regional epidemiology should guide empiric therapy³
• Contact lens-related cases are particularly associated with Pseudomonas species and atypical mycobacteria²
• For appropriately selected cases, early addition of topical steroids after antibiotic initiation may improve outcomes per SCUT subgroup findings⁵
• Common pitfall: delaying treatment pending culture results in an urgent, sight-threatening infection¹
• Common pitfall: interpreting antimicrobial susceptibility data based on systemic concentrations rather than achievable topical corneal concentrations³
• Common pitfall: using topical steroids in all cases, including Nocardia infections, where benefit appears limited or absent⁵