Biomarker-assist score for reverse remodeling prediction in heart failure: The ST2-R2 score

- Biomarker use for LV reverse remodeling (R2) prediction is unexplored.
- HF etiology and duration, LBBB, LVEF and β-blockers influence R2.
- ST2 was the only biomarker independently associated with LV R2.
- The ST2-R2 score, including 5 clinical parameters and ST2 predicts LV R2.

BackgroundLimited data exists regarding biomarker use to predict left ventricular (LV) reverse remodeling (R2). Our aim was to examine the value of soluble ST2 (ST2), N-terminal-pro-B-type natriuretic peptide (NT-proBNP), high-sensitivity cardiac troponin T (hs-cTnT), and galectin-3 relative to LV-R2 in systolic heart failure (HF), and to develop a clinical score for LV-R2 prediction.MethodsR2 was defined as a) LV ejection fraction (LVEF) increase ≥ 15%, or b) LVEF increase ≥ 10% plus reduction of LV end-systolic diameter index ≥ 20% or LV end-systolic volume ≥ 40%, for 12 months.ResultsWe studied 304 patients (79.6% men, mean age 66.1 ± 12.3 years) with baseline LVEF < 40%. R2 was observed in 104 patients (34.2%). In univariable logistic regression, factors associated with R2 were age (p = 0.02), non-ischemic etiology of HF (p < 0.001), NYHA functional class (p = 0.02), baseline LVEF (p = 0.005), absence of left bundle branch block (LBBB; p = 0.002), ST2 (p = 0.004), NT-proBNP (p = 0.005), and hs-cTnT (p < 0.001); HF duration achieved borderline significance (p = 0.08). In multivariable analysis, ST2 remained the only biomarker associated with LV-R2. We developed the ST2-R2 score for use in clinical practice for predicting R2; variables included were ST2 < 48 ng/mL, non-ischemic etiology, absence of LBBB, HF duration < 12 months, baseline LVEF < 24%, and β-blocker treatment. The score had an area under the curve of 0.79 in the derivation cohort and 0.73 in a separate validation cohort.ConclusionsThe ST2-R2 score, which includes the novel biomarker ST2 and five clinical variables, reasonably predicts LV-R2 in systolic HF patients. ST2 was the only studied biomarker that was independently associated with R2.