Smoking as a risk factor for choroidal neovascularization secondary to presumed ocular histoplasmosis syndrome.

PURPOSE

To investigate the relationship of smoking to choroidal neovascularization (CNV) secondary to presumed ocular histoplasmosis syndrome (POHS).

DESIGN

Retrospective, case-control study.

PARTICIPANTS

A total of 568 patients 18 to 50 years of age, 142 of whom were diagnosed with CNV secondary to POHS in a private retina practice between July 1, 2000, and August 1, 2010. Four hundred twenty-six were controls selected from a private comprehensive ophthalmology practice at the same location.

METHODS

A retrospective medical record review was performed for all participants. Age, gender, zip code, CNV diagnosis date, insurance status, and smoking status at CNV diagnosis date were collected first for the POHS patients. For each of these 142 patients, 3 randomly selected comprehensive clinic patients, whose visit date fell within 3 months of the corresponding POHS patient's CNV diagnosis date, served as controls. Age, gender, zip code, visit date, reason for visit, insurance type, and smoking status were recorded. Descriptive statistics were calculated for cases and controls.

MAIN OUTCOME MEASURES

Logistic regression analyses were performed for both univariate and multivariate models, with CNV secondary to POHS as the main outcome variable and smoking as the main predictor variable, while adjusting for age, gender, insurance type, median household income, and education level.

RESULTS

The mean age of patients with CNV secondary to POHS was 39.0±7.1 years, whereas that of the control patients was 35.7±9.1 years. Of the patients with CNV secondary to POHS, 47.2% were current or former smokers (42.3% current, 4.9% former). Of the control patients, 22.5% were current or former smokers (21.8% current, 0.7% former). Age, insurance type, median income, educational attainment, and smoking status were significant in the univariate models. In the final adjusted logistic regression model, only age (odds ratio [OR], 1.04; 95% confidence interval [CI], 1.02-1.07; P = 0.001), level of educational attainment by zip code (OR, 0.95; 95% CI, 0.92-0.98; P = 0.001) and smoking status (OR, 2.83; 95% CI, 1.86-4.31; P<0.0001) were significant.

CONCLUSIONS

The odds of a smoker having CNV secondary to POHS are almost 3 times that of a nonsmoker. In this study, the odds of having CNV secondary to POHS increased with age and decreased with increasing level of educational attainment.