High prevalence of retinal vascular changes in never-treated essential hypertensives: an inter- and intra-observer reproducibility study with non-mydriatic retinography

Background: The clinical significance of stratifying cardiovascular risk in hypertensive patients on the basis of retinal changes such as arteriolar narrowing or arterio-venous crossing has been criticized. Aim: Objectives of the study were: (i) to compare the prevalence of retinal abnormalities detected by non-mydriatic retinography with that of other quantitative markers of target organ damage (TOD), such as echocardiographically determined left ventricular hypertrophy (LVH), carotid structural abnormalities and microalbuminuria in recently diagnosed and never treated hypertensives; (ii) to assess the inter- and intra-observer reproducibility in evaluating retinal microvascular changes. Methods: One hundred ninety-seven grade 1 (73%) and grade 2 essential hypertensives (119 males; mean age 46.8 ± 12.0 years, duration of hypertension: 2.3 ± 1.8 years) referred for the first time to our outpatient hypertension hospital clinic were subjected to the following procedures: (i) repeated clinic blood pressure (BP) measurements; (ii) electrocardiogram; (iii) routine blood chemistry and urinalysis; (iv) 24-h urine collection for microalbuminuria; (v) 24-h ambulatory BP monitoring; (vi) non-mydriatic retinography; (vii) echocardiogram; (viii) carotid ultrasonography. Retinal changes were evaluated according to a modified Keith, Wagener and Barker (KWB) classification by two physicians, who had no knowledge of the patients' characteristics. These following markers of TOD were considered: (i) left ventricular mass index ≥125 g/m2 in men and ≥110 g/m2 in women; (ii at least one carotid plaque (focal thickening >1.3 mm) or diffuse common carotid thickening (≥0.9 mm); (iii) microalbuminuria (urinary albumin excretion ≥30 and <300 mg/24 h). Results: The prevalence rates of LVH, carotid structural alterations and microalbuminuria were 12.9, 26.0 and 8.6% respectively; while the distribution of patients in the different degrees of hypertensive retinopathy made by two independent readers (1 and 2) was: 0 = 15.2, I = 25.4, II = 58.9, III = 0.5% (1); 0 = 14.7, I = 27.9, II = 56.8, III = 0.5% (2), p = NS. The overall prevalence of retinal changes was 84.3% and 84.7%, respectively, and the inter- and intra-observer reproducibility 89.1, 91.6 (1) and 90.2% (2), respectively. Conclusions: Our data indicate that: (i) the prevalence of initial retinal changes is far higher than that of other prognostically validated quantitative markers of cardiac and extracardiac TOD; (ii) the inter- and intra-observer reproducibility between two skilled readers in detecting these abnormalities with non-mydriatic retinography is excellent; (iii) the high prevalence of retinal changes in untreated subjects with mild hypertension offers a new piece of evidence that they cannot be considered a proof of TOD. © 2004 Taylor & Francis on licence from Blood Pressure.