METHODS: We prospectively recruited a cohort of 179 ischemic and 107 non-ischemic heart failure patients. This study mainly focused on ischemic heart failure patients. Non-ischemic heart failure patients were included for the purpose of validation of the risk score in various heart failure groups. Patients were stratified in high risk, moderate risk and low risk groups according to R-hf risk score.
RESULTS: A total of 179 participants with ischemic heart failure were included. Based on R-hf risk score, 82 had high risk, 50 had moderate risk and 47 had low risk heart failure scores. More than half of the patients having R-hf score of <5 had renal failure (n = 91, 50.8%) and anemia (n = 99, 55.3%). Notably, HFrEF was more prevalent in patients with high risk score (74, 90.2%). Patients with high risk score had significantly higher creatinine (2.63 ± 1.96, p
PURPOSE: To examine the association between the modified R-hf risk score and all-cause mortality in patients with HFrEF.
METHODS: Retrospective cohort study included adults hospitalized with HFrEF, as defined by clinical symptoms of HF with biplane EF less than 40% on transthoracic echocardiography, at a tertiary centre in Dalian, China, between 1 November 2015, and 31 October 2019. All patients were followed up until 31 October 2020. A modified R-hf risk score was calculated by substituting brain natriuretic peptide (BNP) for N-terminal prohormone of BNP (NT-proBNP) using EF× estimated glomerular filtration rate (eGFR)× haemoglobin (Hb))/BNP. The patients were stratified into tertiles according to the R-hf risk score. The measured outcome was all-cause mortality. The score performance was assessed using C-statistics.
RESULTS: A total of 840 patients were analyzed (70.2% males; mean age, 64±14 years; median (interquartile range) follow-up 37.0 (27.8) months). A lower modified R-hf risk score predicted a higher risk of all-cause mortality, independent of sex and age [1st tertile vs. 3rd tertile: adjusted hazard ratio (aHR), 3.46; 95% CI: 2.11-5.67; P<0.001]. Multivariate Cox regression analysis indicated that a lower modified R-hf risk score was associated with increased cumulative all-cause mortality [univariate: (1st tertile vs. 3rd tertile: aHR, 3.45; 95% CI: 2.11-5.65; P<0.001) and multivariate: (1st tertile vs. 3rd tertile: aHR 2.21, 95% CI: 1.29-3.79; P=0.004)]. The performance of the model, as reported by C-statistic was 0.67 (95% CI: 0.62-0.72).
CONCLUSION: The modified R-hf risk score predicted all-cause mortality in patients hospitalized with HFrEF. Further validation of the modified R-hf risk score in other cohorts of patients with HFrEF is needed before clinical application.