Purpose: To describe retinal vascular caliber and correlates in people with type 2 diabetes.
Design: Population-based study.
Participants: Thirteen hundred seventy persons diagnosed to have diabetes at or after 30 years of age in an 11-county area in south central Wisconsin from 1980 to 1982.
Methods: Retinal photographs of 7 standard fields were taken; light box grading was done to determine retinopathy severity. Computer-assisted grading was done from a digitized image of field 1 to determine the central retinal arteriolar equivalent (CRAE; arteriolar caliber) and central retinal venular equivalent (CRVE; venular caliber).
Main outcome measures: Retinal arteriolar and venular calibers.
Results: In multivariable analyses in persons with panretinal photocoagulation excluded, while controlling for refractive error, CRAE was associated independently with age (per 10 years, beta = -2.0 microm), mean arterial blood pressure (BP) (per 10 mmHg, beta = -2.2 microm), smoking status (current vs. never smoked, beta = 5.6 microm), and intraocular pressure (IOP) (per 1 mmHg, beta = 0.2 microm). The CRVE was associated independently with age (per 10 years, beta = -2.5 microm), mean arterial BP (per 10 mmHg, beta = -2.1 microm), smoking status (current vs. never smoked, beta = 11.6 microm), pack-years smoked (per 10 pack-years, beta = 1.0 microm), body mass index (per kg/m2, beta = 0.3 mm), pulse rate (per 10 beats/minute, beta = 1.5 microm), retinopathy severity (per 1 level, beta = 1.05 microm), and IOP (per 10 mmHg, beta = -0.5 microm). Smaller CRAEs and CRVEs were found in eyes with panretinal photocoagulation treatment than in eyes without such treatment.
Conclusions: In persons with type 2 diabetes, variations in retinal vascular caliber are related to various systemic and ocular factors. Understanding these relationships may provide further insights into early retinal vascular changes in diabetes.