Soluble ST2 is a biomarker of cardiomyocyte stretch that is useful in the diagnosis and prognosis of coronary artery disease. Its role in the field of hypertension and hypertensive heart failure (HHF) has not yet been well investigated. We studied the effect of left ventricular remodelling on the concentration of soluble ST2 in a cohort of 210 subjects with hypertension (HT). Left ventricular hypertrophy (LVH) was considered present when echocardiographic left ventricular mass indexed for height in metres (m) was greater than 46.2 g m−1 2.7 in women and 49.2 g m−1 2.7 in men. Subjects were subdivided into three groups: those without LVH (HT, n=83); those with LVH (hypertension with left ventricular hypertrophy (HTLVH), n=50) and those with HHF, n=77). Plasma ST2 and NT-pro BNP were measured using electrochemiluminescence type immunoassay. Subjects with HHF had higher plasma ST2 concentrations compared to HTLVH (134.7±57.3 ng ml−1 versus 23.0±8.3 ng ml−1, P<0.001) and those with HT (134.7±57.3 ng ml−1 versus 14.5±4.9 ng ml−1, P<0.0001). NT-pro BNP levels were similar when HTLVH was compared with HT (P=0.68), but subjects with HHF had significantly higher NT-pro BNP compared to HTLVH (P<0.0002). Soluble ST2 had strong correlation with clinical and echocardiograhic parameters, and correlated well with NT-pro BNP (r=0.41, P<0.0001). Plasma ST2 is a useful biomarker in not only differentiating HHF from HT with or without LVH, but also distinguishes hypertensive LVH from HT without LVH.