Clinical outcomes and prognostic factors associated with acanthamoeba keratitis

Abstract

Purpose: To describe the clinical characteristics, time of presentation, risk factors, treatment, outcomes, and prognostic factors on a recent series of Acanthamoeba keratitis (AK) treated at our institution.

Methods: Retrospective case series of 59 patients diagnosed with AK from January 1, 2004 to December 31, 2008. Of these 59 patients, 51 had complete follow-up data and were analyzed using univariate and multivariate logistic regression analyses performed with "failure" defined as requiring a penetrating keratoplasty (PKP) and/or having (1) best-corrected visual acuity (BCVA) < 20/100 or (2) BCVA < 20/25 at the last follow-up. A single multivariate model incorporating age, sex, steroid use before diagnosis, time to diagnosis, initial visual acuity (VA), stromal involvement, and diagnostic method was performed.

Results: Symptom onset was greatest in the summer and lowest in the winter. With failure defined as requiring PKP and/or final BCVA < 20/100, univariate analysis suggests that age > 50 years, female sex, initial VA < 20/50, stromal involvement, and patients with a confirmed tissue diagnosis had a significant risk for failure; however, none of these variables were significant using multivariate analysis. Univariate analysis, with failure defined as requiring PKP and/or final BCVA < 20/25, showed stromal involvement and initial VA < 20/50 were significant for failure-only initial VA < 20/50 was significant using multivariate analysis.

Conclusions: Symptom onset for AK is greatest in the summer. Patients with confirmed tissue diagnosis and female patients may have a higher risk for failure, but a larger prospective population-based study is required to confirm this. Failure is likely associated with patients who present with stromal involvement and patients presenting with an initial BCVA worse than 20/50.