Study Population

The Singapore Indian Eye Study is a population-based study of ethnic South Asian Indians in Singapore described elsewhere. In brief, we used age-stratified random sampling to select ethnic Indians 40 to 80 years of age from the entire population of Singapore during the study period. Of the 4497 eligible adult ethnic Indians, 3397 participated in our study (75.5% response rate). The study was conducted from June 2007 through March 2009 at the Singapore Eye Research Institute.

Study Examinations

Details of our study design and methodology have been described. We conducted standardized systemic and ocular examinations, administered interviewer-administered questionnaires, and carried out standard blood investigations for all participants.

Standardized slit-lamp examinations were performed by study ophthalmologists to determine the presence or absence of corneal arcus in our participants. As defined in our study protocol, corneal arcus was defined as absent or present if the arc of the involved cornea was more than 180 degrees.

A detailed interviewer-administered questionnaire was used to collect relevant sociodemographic data and medical history from all participants. Our main outcome measure was CVD, which was collected using an interview-based, standardized questionnaire on this outcome. CVD was defined as a self-reported history of previously recognized myocardial infarction, stable and unstable angina pectoris, stroke, or transient ischemic attack that was diagnosed by a doctor or required treatment, similar to epidemiologic cross-sectional studies. We also collected data such as country and state of birth with ancestral origin (India), marital status, education, occupation, current housing status, participants’ lifestyle factors, current medications, lifestyle risk factors (e.g., smoking and alcohol intake), systemic medical and surgical history, and family history of eye diseases. Being overweight was defined as body mass index of more than 25 kg/m ² . Blood pressure was measured with a digital automatic blood pressure monitor (Dinamap model Pro Series DP110X-RW, 100V2; GE Medical Systems Information Technologies Inc, Milwaukee, Wisconsin, USA) after the participants were seated for at least 5 minutes. Hypertension was defined as systolic blood pressure of 140 mm Hg or more, diastolic blood pressure of 90 mm Hg or more, or use of anti-hypertensive medication.

We also collected nonfasting venous blood samples for biochemistry tests including serum lipids (total cholesterol, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol), glycosylated hemoglobin A1c), creatinine, and random glucose. All cholesterol levels were analyzed in participants on a continuous scale. Where applicable, hypercholesterolemia was defined as total cholesterol of 6.2 mmol/L or more or use of lipid-lowering drugs. A high level of low-density lipoprotein (LDL) cholesterol was defined as LDL cholesterol of 4.1 mmol/L or more and a low level of high-density lipoprotein (HDL) cholesterol was defined as HDL cholesterol of less than 1.0 mmol/L. Diabetes mellitus was defined as random glucose of 11.1 mmol/L or more, use of diabetic medication, or a physician diagnosis of diabetes mellitus. We calculated the Framingham score for all study subjects using recognized criteria. The risk factors used in the Framingham score were age, total cholesterol, HDL cholesterol, blood pressure, and cigarette smoking. In brief, risk status was determined by a 2-step procedure: the number of risk factors was assessed first, which then was used to determine the 10-year risk assessment using Framingham scoring. Persons were divided into 3 groups of low, medium, and high risk, based on these combinations.

Statistical Analysis

Statistical analysis included descriptive statistics, where the mean and standard deviation were calculated for the continuous variables, whereas frequency distribution and percentages were used for categorical variables. Differences between participants with and without corneal arcus initially were assessed using the t test or chi-square test, as appropriate. Multivariate logistic regression analysis was performed to evaluate the association between prevalence of corneal arcus and CVD while adjusting for potential confounders by using a backward selection method, considering all risk factors initially in the logistic regression. The estimate of odds ratio (OR) and its relative 95% confidence interval (CI) were calculated. We estimated the Akaike information criterion (AIC) to determine the likelihood of CVD using the Framingham score, corneal arcus, and a combination of both. AIC is used as a tool for model selection, with the model having the lowest AIC being the best. All reported P values were compared to a significance level of 5%. All analyses were performed using STATA software version 11 (StataCorp LP, College Station, Texas, USA).