Pugliese, NR

Abstract Funding Acknowledgements Type of funding sources: None. Background. Recent evidence shows epicardial adipose tissue (EAT) acts as a paracrine organ and may directly alter myocardial function by exerting mechanical compression. Purpose. We evaluated EAT thickness through transthoracic echocardiography and investigated its relationship with cardiopulmonary fitness and cardiovascular haemodynamics at rest and during exercise in a population of heart failure (HF) patients with reduced (HFrEF) and preserved (HFpEF) ejection fraction. Methods. We prospectively enrolled 393 consecutive HF outpatients (205 HFrEF, 188 HFpEF) who had been referred to our hospital due to dyspnoea and/or effort intolerance, together with 44 healthy controls. We performed a resting state-of-the-art echocardiographic evaluation, followed by combined cardiopulmonary-echocardiography exercise stress. Results. EAT thickness was higher in HFpEF (median 8 mm, interquartile range [IQR] 4–12 mm) and progressively reduced in controls (median 5 mm, IQR 3–7 mm; p < 0.0001) and HFrEF (median 3 mm, IQR 2–6 mm). In HFpEF, EAT thickness was inversely correlated with peak oxygen consumption (VO2) and peripheral oxygen extraction (AVO2diff), while a direct association was observed for the same parameters in HFrEF (Figure 1). Furthermore, EAT independently predicted peak VO2 and AVO2diff in HFrEF and HFpEF regardless of body mass index and waist circumference. These relationships were direct in HFrEF (standard regression coefficient [SRC] for peak VO2: 0.18, p = 0.02; SRC for peak AVO2diff: 0.17, p = 0.03) and indirect in HFpEF (SRC for peak VO2: -0.33, p < 0.0001; SRC for peak AVO2diff: -0.25, p < 0.0001). As there are no definite cut-off values to define increased EAT, patients were divided based on the median EAT value (5 mm) of the overall population. Thinner EAT (≤5 mm) was associated with worse LV systolic dysfunction (peak average S’) and remodeling (3D LV mass index) in HFrEF. In HFpEF, on the other hand, increased EAT (>5 mm) was related to worse right ventricular systolic dysfunction (3D right ventricular ejection fraction) and more severe left atrio-ventricular (left atrium reservoir strain/left ventricle global longitudinal strain) and right ventriculo-arterial (tricuspid annular plane systolic excursion/systolic pulmonary artery pressure) coupling (Figure 2). Conclusion. In HFpEF, increased EAT thickness is associated with a worse hemodynamic profile and functional capacity. Conversely, in HFrEF, EAT thinning portends more advanced LV dysfunction and impaired cardiopulmonary fitness. Abstract Figure 1 Abstract Figure 2

Abstract Funding Acknowledgements Type of funding sources: None. Background. Recent evidence shows that increased epicardial adipose tissue (EAT) thickness is associated with metabolic syndrome, microvascular dysfunction and enhanced pericardial restraint. Purpose. We measured echocardiography-derived EAT thickness in a population of heart failure (HF) patients with reduced (HFrEF) and preserved (HFpEF) ejection fraction to examine the relationship between EAT and prognosis at clinical follow-up. Methods. We prospectively enrolled 393 consecutive HF outpatients (205 HFrEF, 188 HFpEF) who had been referred to our hospital due to dyspnoea and/or effort intolerance. We performed a resting clinical and biohumoral evaluation, followed by combined cardiopulmonary-echocardiography exercise stress. We considered a composite endpoint of cardiovascular death and HF-related hospitalization during follow-up. Results. Patients with HFpEF displayed greater EAT thickness (median 8 mm, interquartile range [IQR] 4–12 mm) than HFrEF (median 3 mm, IQR 2–6 mm; p < 0.0001). During a median follow-up of 20.9 months (IQR 15-25 months), 34 cardiovascular deaths and 146 HF hospitalizations were reported, with no significant differences between the two HF phenotypes. EAT was shown to predict adverse events independently from body mass index, waist circumference and other well-established prognostic markers in HF (such as NT-proBNP and peak oxygen consumption). The risk of adverse events increased with increasing EAT thickness in HFpEF and with EAT thinning in HFrEF. Kaplan-Meier analyses for the composite endpoint showed that in HFpEF, the survival probability was significantly lower in patients with thicker EAT than those with thinner EAT. In HFrEF, conversely, patients with increased EAT thickness had a higher survival probability than those with reduced EAT thickness (Figure 1). Conclusion. EAT accumulation is increased in HFpEF compared to HFrEF and carries different prognostic meanings in the two subsets. In HFpEF, EAT thickening portends adverse outcomes, which may be due to the secretion of pro-inflammatory and pro-atherogenic adipokines and increased mechanical restraint. In HFrEF, EAT thinning is associated with a worse prognosis, probably reflecting a more advanced catabolic state (e.g., cardiac cachexia). Larger studies are needed to determine whether or not EAT has a causal role in influencing progression and survival in the different HF phenotypes. Abstract Figure 1

Abstract Funding Acknowledgements Type of funding sources: None. Aims. Combined cardiopulmonary exercise test (CPET) and exercise stress echocardiography (ESE) provides a non-invasive tool to study cardiopulmonary pathophysiology. We analyzed how cardiac functional reserve during exercise relates to peak oxygen consumption (VO2). Methods and Results. We performed a symptom-limited graded ramp bicycle CPET-ESE in 30 healthy controls and 357 patients: 113 at risk of developing heart failure (American College of Cardiology/American Heart Association HF Stages A-B) and 244 in HF Stage C with preserved (HFpEF, n = 101) or reduced ejection fraction (HFrEF, n = 143). Peak VO2 significantly decreased from controls to Stage A-B and Stage C (Table 1). A multivariable regression model to predict peak VO2 revealed peak left ventricular systolic annulus tissue velocity (S"), peak TAPSE/PAPs (tricuspid annular plane systolic excursion/systolic pulmonary artery pressure) and low-load left atrial reservoir strain/E/e’ were independent predictors, in addition to peak heart rate, stroke volume and workload (adjusted R²=0.76, p < 0.0001). The model was successfully tested in subjects with atrial fibrillation (n = 49), and with (n = 224) and without (n = 163) beta-blockers (all p < 0.01). Peak S’ showed the highest accuracy in predicting peak VO2 < 10 mL/kg/min (cut-point ≤ 7.5 cm/s; AUC = 0.92, p < 0.0001) and peak VO2 > 20 mL/kg/min (cut-point > 12.5 cm/s; AUC = 0.84, p < 0.0001) in comparison to the other cardiac variables of the model (p < 0.05). Conclusions. A model incorporating different measures of cardiac mechanics is strongly related to peak aerobic capacity and may help in identifying different causes of effort intolerance from HF Stage A to C. Table 1 Variable Overall population (n = 387) Controls (n = 30) Stage A-B (n = 113) Stage C-HFpEF (n = 101) Stage c-HFrEF (n = 143) p-value Age, years 68.9 ± 11.1 67.1 ± 10.6 67.7 ± 10.4 70.5 ± 10.1 68.5 ± 11 0.1 Male, n (%) 247 (64) 18 (60) 70 (62) 57 (56) 102 (71) 0.1 VO2 @peak, mL/min/kg 15.7 (12.1-19.6) 23 (21.7- 29.7) 18 (15.4- 20.7)* 13.6 (11.8- 16.8)*† 14.2 (10.7- 17.5)*† <0.0001 Workload @peak, W 90 (65-120) 130 (115-195) 110 (84-130)* 70 (55-100)*† 80 (60-110)*† <0.0001 Heart rate @peak, bpm 123 ± 22 142 ± 12 130 ± 20* 115 ± 17*† 119 ± 23*† <0.0001 Stroke volume @peak, mL 83 (71-99) 98 (85-114) 86 (76-107) 83 (74-97)* 75 (63-95)*† <0.0001 Average S" @peak, cm/s 11.2 ± 3.8 17.1 ± 3.9 13.3 ± 2.9* 10.6 ± 2.5*† 8.7 ± 2.7*†‡ <0.0001 TAPSE/PAPs @peak, mm/mmHg 0.75 (0.46-0.97) 1.05 (0.93- 1.16) 0.81 (0.52- 0.91)* 0.52 (0.38- 0.83)*† 0.58 (0.41- 0.89)*† <0.0001 Left atrial reservoir strain/E/e" @low-load, % 2.25 (1.17-5.04) 6.23 (4.45-6.77) 4.34 (3.89- 5.58)* 2.23 (1.31- 2.86)*† 1.91 (1.07-2.44)*†‡ <0.0001 * p < 0.01 vs Controls; † p < 0.01 vs Stage A-B; ‡ p < 0.01 vs Stage C-HFpEF. PAPs systolic pulmonary artery pressure; TAPSE: tricuspid annular plane systolic excursion; VO2: oxygen consumption. Abstract Figure 1

Abstract Funding Acknowledgements Type of funding sources: None. Aims. We tested the prognostic role of a risk score including bio-humoral evaluation, cardiopulmonary-echocardiographic stress (CPET-ESE) and lung ultrasound, in patients with heart failure (HF) with reduced and preserved ejection fraction (HFrEF and HFpEF), and subjects at risk of developing HF (American College of Cardiology/American Heart Association Stages A and B). Methods and results. We evaluated 318 subjects: 94 in Stages A-B, 194 in Stage C (85 HFpEF and 109 HFrEF), and 30 age and sex-matched controls (Stage 0). During a median follow-up of 18.5 months, we reported 40 urgent HF visits, 31 HF hospitalisations and 10 cardiovascular deaths. Cox proportional-hazards regression for predicting adverse events identified five independent predictors and each was assigned a number of points proportional to its regression coefficient: Δstress-rest B-lines >10 (3 points), peak oxygen consumption <16 mL/kg/min (2 points), minute ventilation/carbon dioxide production slope ≥36 (2 points), peak systolic pulmonary artery pressure ≥50 mmHg (1 point) and resting N-terminal pro-brain natriuretic peptide (NT-proBNP) >900 pg/mL (1 point). We defined three risk categories: low-risk (<3 points), intermediate-risk (3-6 points), and high-risk (>6 points). The event-free survival probability for these three groups were 93%, 52% and 20%, respectively. Hazard Ratio was 4.55 for each risk category upgrade (95% confidence interval [CI], 3.44-5.93). The area-under-curve for the scoring system to predict events was 0.92 (95% CI 0.88-0.96). Conclusion. A multiparametric risk score including indices of exercise-induced pulmonary congestion, markers of cardiopulmonary dysfunction and NT-proBNP identifies patients at increased risk for HF events across the HF spectrum. Table 1 Variable EPYC score EPYC score <3 (low risk) n = 217 EPYC score 3-6 (intermediate risk) n = 70 EPYC score >6 (high risk) n = 31 p-value (between risk categories) Event-free (n = 244) 0 (0 - 2) 210 (97) 32 (46) 2 (6) <0.0001 With events (n = 74) 6 (4 - 9) 7 (3) 38 (54) 29 (94) <0.0001 p-value (event-free vs with events) <0.0001 <0.0001 <0.0001 <0.0001 Stage 0-Controls (n = 30) 0 (0 - 1) 30 0 0 <0.0001 Stages A-B (n = 94) 1 (0 - 2) 85 (45) 6 (9) 3 (10) <0.0001 Stage C-HFpEF (n = 85) 3 (1 - 6)*† 46 (25) 29 (41) 10 (32) <0.0001 Stage C-HFrEF (n = 109) 4 (2 - 7)*† 56 (30) 35 (50) 18 (58) <0.0001 p-value (between HF Stages) <0.0001 <0.0001 <0.0001 <0.0001 Values are mean ± standard deviation, n (%), or median [25th quartile, 75th quartile]. * p < 0.01 vs Stage 0-Controls; † p < 0.01 vs Stages A-B. Abstract Figure 1