BackgroundHypertension is a multi-factorial disease of increasing prevalence and a major risk factor for cardiovascular mortality even in the presence of adequate treatment. Progression of cardiovascular disease (CVD) occurs frequently during chronic renin-angiotensin-system (RAS) suppression, and albuminuria is a marker of CV risk. High prevalence of albuminuria in treated hypertensive patients has been demonstrated, but there are no available markers able to predict evolution. The aim of this study was the identification of novel indicators of albuminuria progression measurable in urine of diabetic and non-diabetic patients.Methods1143 hypertensive patients under chronic treatment were followed for a minimum period of 3 years. Among them, 105 diabetic and non-diabetic patients were selected and classified in three groups according to albuminuria development during follow-up: (a) patients with persistent normoalbuminuria; (b) patients developing de novo albuminuria; (c) patients with maintained albuminuria. Differential urine analysis was performed by 2D gel electrophoresis (2D-DIGE) and further confirmed by liquid chromatography-mass spectrometry. Non-parametric statistical tests were applied.ResultsCD59 glycoprotein and alpha-1 antitrypsin (AAT) resulted already altered in patients developing albuminuria de novo, with a similar response in those with maintained albuminuria. A prospective study in a sub-group of normoalbuminuric patients who were clinically followed up for at least 1 year from urine sampling, revealed CD59 and AAT proteins significantly varied in the urine collected from normoalbuminurics who will negatively progress, serving as predictors of future albuminuria development.ConclusionsCD59 and AAT proteins are significantly altered in hypertensive patients developing albuminuria. Interestingly, CD59 and AAT are able to predict, in normoalbuminuric individuals, who will develop albuminuria in the future, being potential predictors of vascular damage and CV risk. These findings contribute to early identify patients at risk of developing albuminuria even when this classical predictor is still in the normal range, constituting a novel strategy towards a prompt and more efficient therapeutic intervention with better outcome.Electronic supplementary materialThe online version of this article (doi:10.1186/s12933-016-0331-7) contains supplementary material, which is available to authorized users.