The correlation between Fischer's ratio and the risk of cardiac dysfunction in heart failure patients

Abstract

Backgrounds: Due to the high mortality and hospitalization rate in chronic heart failure (HF), it is of great significance to study myocardial nutrition conditions. Amino acids (AAs) are essential nutrient metabolites for cell development and survival. This study aims to investigate the associations and prognostic value of plasma branched-chain amino acid/aromatic amino acid ratio (Fischer's ratio, FR) in patients with left ventricular ejection fraction (LVEF) ≤ 50%.

Methods: The value of serum AAs was obtained from 441 cardiovascular patients by liquid chromatography-tandem, including 213 HF patients [all LVEF ≤ 50%, E/e'>14 (n = 101) and E/e'≤14 (n = 112)], and 228 controls without HF [(excluding HF with preserved ejection fraction (EF)]. Two-dimensional echocardiography and Doppler flow imaging determined LVEF and the ratio of early diastolic mitral inflow to mitral annular tissue velocities (E/e'). The abnormal LVEF (≤ 50%) refers to the definition of HF with reduced, mildly reduced, or improved EF. The logistic regression analysis was conducted to measure the FR index and the risk of cardiac dysfunction, and further confirmed by receiver-operating characteristic curves (ROC curve) analysis. In the prospective study, the 188 HF patients were followed up for a mean year (11.74 ± 1.44 months). The event-free HF endpoint or HF readmission was determined by Kaplan-Meier curves, and differences were assessed using log-rank tests, respectively. Cox regression analysis was used to further assess the prognostic value of FR in HF.

Results: FR index decreased gradually along with the control group, HF with E/e'≤14 group, and HF with E/e'>14 group (3.73 ± 1.20 vs. 3.45 ± 0.94 vs. 3.18 ± 0.83, respectively, P < 0.001). Low FR index was associated with HF after full adjustment in all patients [odds ratio (OR): 2.072; 95% confidence interval (CI): 1.546-2.776; P < 0.001]. Meanwhile, low FR index was the independent risk factor for E/e'>14 in HF patients (OR:1.659; 95% CI:1.125-2.446; P = 0.011). The area under the ROC curve for predicting abnormal E/e' was 0.762 (sensitivity 75.51%, specificity 65.05%) by multivariate logistic regression. Furthermore, the decreased FR values indicated poor HF endpoint or HF readmission in HF patients (all Log-rank P < 0.01). The Cox regression analysis showed that low FR tertiles remained correlated to the risk of HF endpoint [hazard ratio (HR): 1.949; 95% CI: 1.006-3.778; P = 0.048].

Conclusions: In all patients, low FR confers an increased risk for predicting HF. Decreased FR levels could also indicate increased left ventricular (LV) filling pressure in HF. In addition, a lower FR value was associated with higher HF endpoint events. Thus, FR can be a valuable indicator of heart function.

Keywords: Cardiac dysfunction; Chronic heart failure; LV filling pressure; Low fischer’s ratio; Prognosis; Risk prediction.

Fig. 1

Flow chart of HF patients’ enrollment analysis. HF, heart failure; HFpEF, heart failure with preserved ejection fraction; E/e’, the ratio of early diastolic mitral inflow to mitral annular tissue velocities; UCG, echocardiography; HFmrEF, HF with mildly reduced ejection fraction; HFrEF, HF with reduced ejection fraction; HFimpEF, HF with improved ejection fraction; LVEF, left ventricular ejection fraction

Fig. 2

The expression of FR in various groups. a Comparison between control (cardiovascular disease without HF) and HF (LVEF≤50%) patients by student’s T-test. b One-way AVONA among the three groups (Control, HF with E/e’≤14, and HF with E/e’ >14), further comparison between the two groups by LSD. LVEF, left ventricular ejection fraction; Other abbreviations are as in Table 1.

Fig. 3

ROC curve for predicting E/e’>14 in HF. Predicted probabilities derived from regression models for detecting E/e’>14. Risk factors include variables of BMI, left ventricular EF, HbA1c, low eGFR, and FR tertiles. ROC, receiver-operating-characteristic curve; Other abbreviations are as in Table 1.

Fig. 4

Correlation between FR and cardiac index in HF

Fig. 5

Follow-up data of HF patients and prognosis analysis. a HF patients who met the HF endpoint, HF readmission, or cardiovascular death had lower FR at baseline. b Kaplan–Meier (KM) curves and log-rank analysis for the HF endpoint according to low FR tertiles. c KM curves and log-rank analysis for the HF readmission according to low FR tertiles. d Cox regression for the HF endpoint according to low FR tertiles. Model 1: Adjusted for gender and age; Model 2: Adjusted for gender, age, BMI, NYHA functional class, smoking, drinking, hypertension, coronary disease, AF, LAD, left ventricular EF, EDD, and ESD, E/e´, log NT-proBNP, HbA1c, TG, HDL-c, LCL-c, ALT, low eGFR, and low HGB. Abbreviations are as in Table 1