Association between left ventricular outflow tract velocity time integral, ejection fraction, clinical characteristics, and mortality in an undifferentiated cohort of hospitalised patients

Background Left ventricular ejection fraction (LVEF) is commonly used to predict mortality in heart failure patients. However, LVEF is dependent on factors including preload, afterload, contractility, heart rate and measurement accuracy. Recently, left ventricular outflow tract velocity time integral (LVOT VTI) has been shown to confer additional prognostic value in both patients with advanced heart failure and patients hospitalised with heart failure with preserved ejection fraction. Purpose We evaluated the association between LVOT VTI, LVEF, clinical characteristics and mortality in an undifferentiated cohort of hospitalised patients. Methods All hospitalised patients who underwent echocardiogram between January – March 2021 were identified from a single-centre tertiary hospital database. Baseline characteristics, echocardiographic parameters, biomarkers, and mortality at 24 months were collected. Patients were categorised into LVOT VTI tertiles (> 19cm, 13.1 – 18.9cm, < 13cm) and LVEF tertiles (> 50%, 40-50%, < 40%) for analysis; p<0.05 was considered significant. Results Altogether, 653 patients comprising 370 (57%) with LVOT VTI > 19cm, 228 (35%) with 13.1 - 18.9cm, and 55 (8%) with < 13cm were included. There were 183 (28%) deaths. Mortality was associated with increased age (75.6 + 13.5 years vs 64.9 + 16.8 years, p<.0001), high-risk echocardiographic features including increased left atrial size (45.4 + 22.7ml/m² vs 35.8 + 14.9 ml/m², p<.001), E/e’ (14.8 + 8.7 vs 10.4 + 4.3, p<.001), reduced LVEF (51.7 + 15% vs 55.9 + 12%, p=.005) and tricuspid annular plane systolic excursion (TAPSE; 2 + 0.5cm vs 2.1 + 0.7cm, p=.001), and biomarkers including haemoglobin (99 + 45g/L vs 117 + 46g/L, p<.001), albumin 28 + 10g/L vs 32 + 10g/L, p<.001), elevated N-terminal pro brain natriuretic peptide (5126 + 5997ng/L vs 2035 + 3569ng/L, p<.0001), troponin (191 + 1301ng/L vs 102 + 306ng/L, p=.008) and creatinine (152 + 165µmol vs 86 + 99µmol, p<.001) levels on admission. Whilst there was no significant difference between LVOT VTI and mortality (z=-1.73, p=0.083) there was a significant positive relationship between LVOT VTI and LVEF (r [651] =0.49, p<.001). Further, there was a significant association between death and both LVEF tertiles (X2 [2, N=653] = 19.63, p<.0001), and LVOT VTI tertiles (X2 [2, N=653] = 9.05, p=.011). However, small subset analyses revealed similar mortality rates for the lowest LVEF LVOT VTI tertile (46%, n=55), lowest LVEF tertile (46%, n=87), and when combined (50%, n=32). Conclusion In an undifferentiated cohort of hospitalised patients, reduced LVOT VTI and LVEF are both associated with increased mortality. However, LVOT VTI does not appear to outperform LVEF in being associated with mortality.