engleski

S2PLIT Score: a simple risk score predicting post-discharge 1-year mortalityin patients with acutely decompensated heart failure

Background: The acutely decompensated heart failure (ADHF) is a condition associated with poor outcomes, especially during the first year after hospitalization. Purpose: To assess the performance of the S2PLIT scoring model in predicting 1-year all- cause mortality in ADHF patients. Methods: Clinical and laboratory data of 340 patients admitted for ADHF at a single- center ICCU were retrospectively examined. Variables that were significantly associated with 1-year mortality in multivariate regression analysis adjusted for age, sex, NYHA class, post- discharge medications, and comorbidities were included in the risk stratification model. Hosmer-Lemeshow test and C-statistic were used to determine the validity and predictive power of the model. Kaplan-Meier survival analysis was used to assess survival among risk groups. Results: The average age of the studied population was 74 ± 9.8 years and 50.6% were women. Seventy-eight patients (22.9%) were NYHA II, 162 (47.8%) were NYHAIII while 100 patients (29.5%) belonged to NYHA IV functional class. The average left-ventricular ejection fraction (LVEF) was 42.2 ± 9.5% while average systolic blood pressure (SBP) was 137.4 ± 27.4 mm Hg. Serum creatinine, uric acid, and sodium levels averaged 142.3 ± 96.2 μmol/L, 489 ± 172.9 μmol/L, and 138.2 ± 4.4 mmol/L, respectively. Significant independent predictors for 1-year all-cause mortality in our sample were LVEF, SBP, prior history of ADHF hospitalization(s) and serum creatinine, uric acid, and sodium levels. According to calculated S2PLIT score (Table1), 153, 78 and 109 patients were stratified into a low-, intermediate- and high- risk groups with the observed mortality rates of 9.8% (15/153), 33.3% (26/78) and 91.7% (100/109), respectively. The obtained area under the curve (AUC) for the proposed score model was 0.900 (95% CI 0.864-0.937, SE 0.019, p < 0.001) (Figure 1A) with clear separation among respective risk groups in terms of cumulative survival (Figure1B). Conclusions: The S2PLIT scoring model performed well and demonstrated high predictive power for all-cause mortality in ADHF patients during the 1-year period following discharge.

acutely decompensated heart failure ; 1-year mortality ; risk stratification ; S2PLIT Score

Presented at the Rapid Fire Oral Abstract Session.