BACKGROUND: Bacterial keratitis is an important cause of preventable vision loss worldwide. In this study, we aimed to synthesise the evidence for regional and temporal differences in antimicrobial resistance that could influence the choice of empirical therapy. METHODS: For this systematic review and meta-analysis, we searched MEDLINE, Embase, Cochrane Library, Web of Science, clinicaltrials.gov, and the WHO International Clinical Trials Registry Platform from their date of database inception to Oct 3, 2025, for studies that reported the in-vitro susceptibility of bacterial isolates from eyes with keratitis, with no restrictions on age, publication date, or language. We excluded animal studies, single case reports, reviews, conference proceedings, editorials, grey literature, unpublished studies, and studies with an incomplete description of the anatomical source of the samples or samples collected from the conjunctiva. We excluded studies that recruited exclusively from settings such as intensive care or only included individuals with specific syndromes or other disease groups. Except for meticillin-resistant Staphylococcus aureus, studies that only reported resistant or multiresistant bacterial isolates were excluded. We contacted study authors for further information, as required. We pooled data using a random-effects meta-analysis of proportions and examined susceptibility by Global Burden of Disease super region and changes over time. The main outcome was the proportion of susceptible isolates for each drug-pathogen combination. This study was registered with PROSPERO (CRD42023331126). FINDINGS: We extracted data from 340 reports of samples collected between 1975 and 2024 from 47 countries, with eligible data from 129 388 cases of culture-positive keratitis. Most reports were from high-income countries (158 47% of 340), south Asia (85 25%), or east Asia and the Pacific (40 12%), with other regions being deficient in data in comparison. The susceptibility of coagulase-positive Staphylococcus spp to ciprofloxacin in reports from south Asia was 70·8% (95% CI 60·1-79·6), which was lower than that in reports from high-income countries (86·9%, 81·5-90·9). The susceptibility of Pseudomonas spp to ciprofloxacin in reports from south Asia was also lower (80·2%, 95% CI 72·9-85·9), than that in reports from east Asia and the Pacific (96·8%, 94·5-98·1), high-income countries (98·3%, 97·4-98·9), and north Africa and the Middle East (93·8%, 87·4-97·1). We found no conclusive evidence for reduced susceptibility over time for any drug-pathogen combination. INTERPRETATION: Regional differences exist in antimicrobial resistance for some clinically relevant drug-pathogen combinations. A standardised reporting template would reduce heterogeneity and facilitate future research. FUNDING: The National Institute for Health Research Moorfields Biomedical Research Centre, Wellcome Trust.
Tuft et al. (Mon,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: