Abstract
Purpose: :
To examine longitudinal associations between baseline retinal vascular calibre and incidence of glaucoma in the Blue Mountains Eye Study (BMES).
Methods: :
Of the 3654 baseline BMES participants (1992-94), there were 2461 participants with data on baseline retinal vessel calibre and glaucoma at either 5 or 10 years follow up. After excluding 44 subjects with primary open angle glaucoma (POAG) at baseline, 2417 participants at risk of POAG were included. Central retinal artery and venular equivalents (CRAE and CRVE), the average diameters of these vessels, were measured from retinal photographs using a computer-assisted image program. Associations between baseline retinal vessel calibre and risk of POAG were assessed using generalized estimating equations (GEE) models with multiple logistic regression to incorporate correlation between eyes, adjusting for age, gender, baseline systolic blood pressure and intra-ocular pressure.
Results: :
Of the 82 incident glaucoma cases (104 eyes), 23 were diagnosed at the 5-year, and 59 at the 10-year follow-up visit. Mean baseline CRAE and CRVE were 156.1µm (standard deviation [SD] 15.1) and 233.4µm (SD 22.6), respectively in persons who developed POAG, compared with 160.6 (14.9) µm and 240.3 (22.4) µm, respectively, in those who did not. Each standard deviation decrease in baseline CRAE (-14.9 µm) and CRVE (-22.4µm) was associated with an increased risk of incident POAG (adjusted odds ratio, OR, 1.88, 95% confidence interval, CI, 1.19-2.98 for CRAE; OR 1.68, CI 1.07-2.65 for CRVE). Persons with CRAE in the narrowest quartile of the population were 4 times more likely to develop POAG (OR 4.33, CI 1.29, 14.6) than persons with CRAE in the widest quartile. Similarly, persons with CRVE in the narrowest vs widest quartile of CRVE were 3 times more likely to develop POAG, although this was not statistically significant (OR 3.17, CI 0.88, 11.3). When both CRAE and CRVE were adjusted for simultaneously, associations were no longer significant.
Conclusions: :
This study documents a longitudinal association between narrower retinal vessel calibre at baseline and greater long-term risk of developing clinical POAG, in an older Australian cohort. Our findings suggest involvement of microvascular changes in the pathogenesis of POAG reflecting alterations in the retinal circulation.
Keywords: clinical (human) or epidemiologic studies: risk factor assessment • clinical (human) or epidemiologic studies: prevalence/incidence • imaging/image analysis: clinical