Association of ST2 levels with cardiac structure and function and mortality in outpatients

Background ST2, an interleukin-1 receptor family member up-regulated in the setting of cardiomyocyte strain, has prognostic value in patients with acute myocardial infarction, chronic severe heart failure, and acute heart failure. The predictive value of ST2 levels in outpatients is unknown. We studied the clinical and echocardiographic correlates of ST2 levels and evaluated their prognostic use in outpatients referred for echocardiograms. Methods ST2 levels were measured in 588 outpatients referred for echocardiogram. Subjects were analyzed by quartile as well as by optimal ST2 cut-point (28.25 ng/mL) derived from receiver operating characteristic curve analysis. All-cause death at 1 year was the primary outcome. Results In this cohort with mean age of 68 +/- 12 years and median ST2 level of 19.8 ng/mL (interquartile range 15.8-23.7), 25 deaths occurred. Heart rate, creatinine clearance, use of diuretics, and the presence of right ventricular hypokinesis were independently associated with ST2 levels. At 6 months, no patients with ST2 below the median had died. Patients with high ST2 levels had an increased risk of death (adjusted hazard ratio [HR] 2.5, P = .02); those with elevated levels of both ST2 and B-type natriuretic peptide were at even higher risk (adjusted HR 4.3, P = .01 vs none elevated). Conclusions ST2 levels reflect right-side heart size and function and are independent predictors of 1-year mortality in outpatients referred for echocardiograms. The optimal cut-point derived in this cohort is comparable with the previously identified prognostic cut-point for sicker patients. ST2 may be an especially strong prognostic marker for short-term mortality risk. (Am Heart J 2010; 160: 721-8.)