Related Experiment Video
Updated: Aug 6, 2026

Intravitreal Injection and Quantitation of Infection Parameters in a Mouse Model of Bacterial Endophthalmitis
Published on: February 6, 2021
Long-term Infectious Keratitis Microbiology and Resistance Patterns With Forecasts to 2030: A 23-Year Analysis
Guillermo Raul Vera-Duarte1, Luis Haro-Morlett, Thelma Cortés-Moreno
1Instituto de Oftalmologia Fundacion Conde de Valenciana IAP, Mexico City, Mexico.
Purpose:
To analyze 23-year trends in microbiology and antibiotic resistance among infectious keratitis (IK) isolates at a tertiary center in Mexico City and to generate validated forecasts through 2030 to inform empiric therapy and stewardship.
Methods:
Corneal scrapings were processed using standardized sampling, culture, and identification protocols; susceptibilities were interpreted per CLSI (VITEK and Kirby-Bauer). Annual Chow tests (2004-2022; ≥3 data points per side) screened for structural breaks in organism prevalence and resistance, and linear regression summarized long-term trends (2002-2024). For forecasting, each microorganism and antibiotic series was modeled as a univariate time series using Historical Mean, linear and quadratic OLS, ETS, and ARIMA (fit on logit-transformed percentages with drift). ARIMA orders were chosen by rolling-origin cross-validation (training 2002-2018). Final model selection was based on lowest MAE on the validation period (2019-2024), and 2025 to 2030 forecasts with 95% CIs were produced from the selected model.
Results:
Among 4748 cultures, 2918 (61.46%) were positive: bacteria 2624 (89.92%) and fungi 294 (10.08%). Bacterial isolates were 81.44% Gram+ and 18.56% Gram-. The leading organisms were Staphylococcus epidermidis, other Gram+, other coagulase-negative staphylococci, Pseudomonas aeruginosa, and Staphylococcus aureus; Fusarium dominated fungi (52.38%). Structural breaks clustered in 2005 to 2012; positive-culture rates rose from ∼35 to 50% pre-2010 to ∼80%, with few new structural breaks after 2020. Gram+ resistance declined in β-lactams/50S agents but increased for fluoroquinolones, vancomycin, and gentamicin. Gram- resistance rose in cephalosporins/carbapenems, while nitrofurantoin decreased; Gram- showed no consistent long-term trend. Historical Mean often minimized MAE; ARIMA/regression/ETS performed best with persistence, curvature, or fluctuations. Forecasts to 2030 indicate broad stabilization, with watchpoints in Gram+ vancomycin/30S agents and Gram-cephalosporin/carbapenem classes; positive rates remain high and negatives decline.
Conclusions:
Infectious keratitis microbiology underwent a late-2000s reconfiguration followed by durable stabilization. Forecasts project continued stability with targeted resistance signals, supporting ongoing local surveillance, antimicrobial stewardship, and optimized diagnostic practices.
Related Concept Videos
Microbiome of the Eye
Clinical Significance of Antibiotic Resistance
Antimicrobial Effectiveness
Mechanism of Antibiotic Resistance in MRSA
