Shifting Distribution of Chicago-Area Acanthamoeba Keratitis Cases

Acanthamoeba keratitis (AK) is a rare contact lens–related infection with significant morbidity. Acanthamoebae are ubiquitous, free living, and believed to infect the cornea through contaminated water. Risk factors include poor hygiene, swimming while wearing contact lenses, tap water contamination, and use of certain contact lens solutions.¹ A statistically significant increase in cases began in June 2003 in Chicago, Illinois.² Self-reported use of COMPLETE MoisturePLUS multipurpose solution (Advanced Medical Optics, Inc, Santa Ana, California) was strongly associated with AK in 2 studies,^3,4 leading to a product recall on May 25, 2007. Nevertheless, only 55% to 60% of patients with AK reported exposure to COMPLETE MoisturePLUS, while approximately 40% did not.^3,4 Although Acanthamoebae are resistant to multipurpose solutions,⁵ generalized outbreaks have not occurred since introduction. In Chicago, cases continue despite the recall (11 cases were diagnosed from June 1, 2003, to May 31, 2004; 21 cases were diagnosed from June 1, 2004, to May 31, 2005; 15 cases were diagnosed from June 1, 2005, to May 31, 2006; 25 cases were diagnosed from June 1, 2006, to May 31, 2007 [COMPLETE MoisturePLUS was recalled on May 25, 2007, the eve of the Memorial Day holiday weekend; thus, AK cases diagnosed as a result of the recall began June 1, 2007]; 27 cases were diagnosed from June 1, 2007, to May 31, 2008; and 16 cases were diagnosed from June 1, 2008 to November 30, 2009 [which represents only half a year of diagnosed cases]), and the number of cases in which the patients have been using solutions other than COMPLETE MoisturePLUS is increasing.⁶ Surveillance data by the Centers for Disease Control and Prevention presented in January 2009 at the Microbial Testing for Contact Lens Care Products workshop,⁷ sponsored by the US Food and Drug Administration and others, suggest that cases continue elsewhere also. The purpose of this article is to evaluate whether the spatial distribution of AK cases is changing with time.

The University of Illinois at Chicago Institutional Review Board reviewed and approved this research. Patients evaluated at the University of Illinois at Chicago with atypical keratitis were defined as having AK as previously described^2,3 using a validated disease definition.⁸ Geographic information systems were used with exploratory time series analyses to evaluate case distribution in Chicago, where case residence was plotted to overlay 2003 census population density by quartile. Residence at diagnosis was geocoded using the ArcInfo version 9.1 geographic information system (ESRI, Redlands, California) and was limited to the 5-county Chicago metropolitan area; analyses were stratified a priori into 2-year periods (June 1, 2003, to May 31, 2005, and June 1, 2005, to May 31, 2007). To evaluate whether the disease pattern was changing and moving centrally toward the city, the mean distance between the residence of patients with AK and the University of Illinois at Chicago (convenience selection centrally located in Chicago) was compared using a 2-sample t test; euclidean distances were estimated using Hawth's Analysis Tools version 3.27 (http://www.spatialecology.com/index.php).

Comparison of map symbols demonstrates that the cases from June 1, 2003, to May 31, 2005 (n = 22) were distributed to the west, south, and southwest and further from Lake Michigan compared with population density, while the cases from June 1, 2005, to May 31, 2007 (n = 31) continued in the west, south, and southwest but also occurred centrally (Figure). The mean distance between the patients' residence and the University of Illinois at Chicago was statistically farther for cases diagnosed from June 1, 2003, to May 31, 2005, than for those diagnosed from June 1, 2005, to May 31, 2007 (22.90 vs 17.81 miles, respectively; P = .04).

Exploratory time series analysis demonstrates a shifting case distribution, which is incompatible with identified risk factors including hygiene, swimming while wearing contact lenses, ineffective solutions,^1,3,4 or even demographic variables such as socioeconomic status, age, race, or behavior. Although demographic variability may confound AK risk, such as the prevalence of contact lens use by socioeconomic status or race, no contact lens–related risk factors or demographic variables can account for the changing distribution in a 2-year period because few can change at a population level in 2 years. Disease continues, which indicates that prevention will require further risk factor identification and/or a sizeable modification to current hygiene practices and contact lens solutions. Additionally, the changing distribution directs research to factors that can vary by geography over time, requiring study designs where controls are not matched on geography. Results also suggest the need for further exploration of alternative environmental hypotheses, including whether changes in disinfection practices to meet US Environmental Protection Agency disinfection byproduct regulations may be inadvertently shifting the microbial balance of the domestic water supply.²

Correspondence: Dr Joslin, Department of Ophthalmology and Visual Sciences, University of Illinois at Chicago, 1855 W Taylor St, Ste 3.164, Chicago, IL 60612 (charjosl@uic.edu).

Author Contributions: Dr Joslin had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Financial Disclosure: No authors have any financial interest or receive payment as a consultant, reviewer, or evaluator in relation to the contact lens solution company discussed within this article. Dr Joslin has served as a consultant to Bausch & Lomb Inc.

Funding/Support: This work was supported by grant 15689 from the National Institutes of Health, by grants from Prevent Blindness America, Midwest Eye-Banks, and the Campus Research Board of the University of Illinois at Chicago, by the William C. Ezell Fellowship from the American Optometric Foundation/American Academy of Optometry, by a grant from the Karl Cless Foundation, and by grant 09073 from the National Institutes of Health (Dr Shoff).

Role of the Sponsor: The funding organizations provided only support for this research and were not involved in the design and conduct of the study; collection, management, analysis, and interpretation of the data; and preparation, review, or approval of the manuscript.