Retinal vessel metrics: normative data and their use in systemic hypertension

PURPOSE OF REVIEW In-vivo measurement of retinal vascular calibers may be used as a tool to study the pathophysiology and clinical status of the microvasculature of the retina. The aim of this study was to generate normative data for retinal vessel parameters, and to evaluate the clinical relevance in systemic hypertension. METHODS Fundus photographs from 4309 participants of the Gutenberg Health Study were assessed using the 'retinal vessel analyzer' software (IMEDOS). We generated age and sex-specific nomograms in a disease-free subpopulation of 890 participants for determining the central retinal arteriolar equivalent (CRAE), the central retinal venular equivalent, and the arteriovenous ratio (AVR). RESULTS Women had higher values of CRAE, central retinal venular equivalent, and AVR than men, and the decrease in measures with increasing age was less steep in women than in men. Systemic hypertension was associated with lower values [odds ratio (OR), 95% confidence interval (CI) referring to area below the 5% percentile] of AVR (men: OR 2.41, 95% CI 1.669-3.490, P < 0.001; women: OR 3.01, 95% CI 2.126-4.268, P < 0.001) and CRAE (men: OR 2.60, 95% CI 1.563-4.326, P < 0.001, women: OR 3.00, 95% CI 2.004-4.487, P < 0.001). Both median CRAE and AVR were lower in participants with uncontrolled hypertension (172.28, range 83.05-251.04; and 0.81, range 0.56-1.04) versus those with screening-detected hypertension (175.72, range 101.23-222.09, P < 0.001; and 0.82, range 0.64-1.05, P = 0.001), and versus those with controlled (179.10, range 108.19-221.92, P < 0.001; and 0.84, range 0.60-1.08, P < 0.001) hypertension. CONCLUSION The study provides sex and age-specific normative data for retinal vasculature. Persons with untreated or insufficiently treated hypertension are more likely to have retinal vessel equivalents outside the reference range.