Tan Charmaine

Background: Acanthamoeba keratitis (AK) is a rare, vision-threatening corneal infection caused by free-living Acanthamoeba species found in environmental sources such as water and soil^1,2. It predominantly affects contact lens wearers, particularly those using soft lenses^1,2. Following infection, organisms form drug-resistant cysts, contributing to prolonged and difficult treatment courses^1,2,5. Early clinical presentation often mimics bacterial or herpetic keratitis, resulting in frequent misdiagnosis and delayed initiation of appropriate therapy^2,5. While high-risk contact lens behaviors, including water exposure, overnight wear, and poor hygiene, are well established, their association with disease severity and treatment response remains incompletely defined^2-5. Delayed diagnosis and advanced disease at presentation are consistently associated with worse visual outcomes and increased need for surgical intervention^2,5. Recent advances in deep learning–assisted analysis of in vivo confocal microscopy (IVCM) offer promising approaches for earlier, noninvasive diagnosis^6,7.

Methods: A PubMed search was conducted from 2010 onward using the terms “Acanthamoeba keratitis,” “contact lens risk factors,” “disease severity,” “treatment outcomes,” “artificial intelligence,” and “confocal microscopy.” Reference lists of relevant articles were also reviewed to identify additional studies.

Results: High-risk contact lens behaviors, particularly water exposure, overnight wear, and lens reuse, are associated with AK, while daily disposable lenses may be protective^1,2. A case–control study (83 cases, 122 controls) confirmed these as independent risk factors³, and a cross-sectional study (n=245) similarly identified water exposure and poor hygiene as significant risks for microbial keratitis⁴. In a 110-case series, AK was frequently misdiagnosed; delayed diagnosis and advanced-stage presentation were associated with worse visual outcomes and increased surgical intervention rates⁵. Biguanide-based therapy achieved high cure rates but required prolonged treatment and was less effective in advanced disease^2,5. Deep learning models using IVCM demonstrated high diagnostic accuracy, with saliency mapping highlighting cystic and inflammatory features; AI-based systems also enabled automated corneal layer localization and robust disease detection, supporting clinical integration^6,7.

Conclusion: Modifiable contact lens behaviors, particularly water exposure while wearing lenses, overnight wear, and poor hygiene, increase the risk of AK and are associated with more advanced disease at presentation, when medical therapy alone is less likely to preserve vision. Diagnostic delay, often due to early misdiagnosis, contributes to worse outcomes, including deeper stromal involvement, prolonged treatment, and higher surgical rates. Deep learning–assisted IVCM is a promising tool to improve early recognition and may help shift AK toward earlier stages and better visual outcomes when combined with behavior‑focused prevention.

Work Cited

1. Gomes de Lacerda, Ana, and Madalena Lira. “Acanthamoeba keratitis: a review of biology, pathophysiology and epidemiology.” Ophthalmic and Physiological Optics, vol. 41, no. 1, 29 Oct. 2020, pp. 116–135, https://doi.org/10.1111/opo.12752.

2. Garg, Diksha, and Sachin Daigavane. “A comprehensive review on Acanthamoeba keratitis: An overview of epidemiology, risk factors, and therapeutic strategies.” Cureus, 26 Aug. 2024, https://doi.org/10.7759/cureus.67803.

3. Carnt, Nicole, et al. “Acanthamoeba keratitis risk factors for daily wear contact lens users.” Ophthalmology, vol. 130, no. 1, Jan. 2023, pp. 48–55, https://doi.org/10.1016/j.ophtha.2022.08.002.

4. Sakr, Suzan Ibrahim, et al. “Impact of contact lens hygiene risk factors on the prevalence of contact lens-related keratitis in Alexandria-Egypt.” Journal of Ophthalmic Inflammation and Infection, vol. 14, no. 1, 20 Aug. 2024, https://doi.org/10.1186/s12348-024-00421-1.

5. Scruggs, Brittni A., et al. “Risk factors, management, and outcomes of Acanthamoeba keratitis: A retrospective analysis of 110 cases.” American Journal of Ophthalmology Case Reports, vol. 25, Mar. 2022, p. 101372, https://doi.org/10.1016/j.ajoc.2022.101372.

6. Essalat, Mahmoud, et al. “Interpretable deep learning for diagnosis of fungal and acanthamoeba keratitis using in vivo confocal microscopy images.” Scientific Reports, vol. 13, no. 1, 2 June 2023, https://doi.org/10.1038/s41598-023-35085-9.

7. Lincke, Alisa, et al. “AI-based decision-support system for diagnosing acanthamoeba keratitis using in vivo confocal microscopy images.” Translational Vision Science & Technology, vol. 12, no. 11, 27 Nov. 2023, p. 29, https://doi.org/10.1167/tvst.12.11.29.