Detecting subclinical congestion in stage A/B pre‐heart failure: a glimpse into the future?

This article refers to ‘The prevalence and clinical associations of ultrasound measures of congestione in patients at risk of developing heart failure’ by J.J. Cuthbert et al., published in this issue on pages 1831–1840. Signs and symptoms related to congestion are the predominant cardiac reason why patients with heart failure are being admitted to the hospital. Congestion is a setting in which a culmination of a wide range of cardiac structural or functional abnormalities together with changes in intravascular volume or venous capacitance function lead to symptoms of breathlessness and fatigue or signs of extravascular volume accumulation.1-3 Data from implantable haemodynamic sensors have taught us that slight elevations in filling pressures can occur weeks before an acute decompensation in heart failure and in the early stages often occur without symptoms or signs of heart failure, being termed subclinical congestion.4, 5 Furthermore, subclinical congestion has long been recognized to herald an ominous prognosis in heart failure patients awaiting cardiac transplantation or in patients being discharged after an acute heart failure episode.6 Targeted ultrasound evaluation of extra-cardiac structures including lung tissue, venous structures and the kidneys allows to determine the presence of systemic congestion and is becoming increasingly integrated in the ultrasound evaluation of patients with established heart failure.7 This is further driven by the appreciation that congestion itself is a perpetuating mechanism driving disease progression in heart failure. Indeed, the recent universal definition of heart failure recognizes this important role of congestion, defining heart failure as a clinical syndrome with current or prior symptoms and/or signs caused by structural and/or functional cardiac abnormalities being corroborated by either an elevated natriuretic peptide (NP) level or objective evidence of cardiogenic pulmonary or systemic congestion by diverse diagnostic modalities (however not specifically naming lung ultrasound, jugular vein diameter analysis or renal vein Doppler analysis).8 This definition of heart failure requires current or past symptoms of heart failure. However, population-based registries indicated that up to 40–50% of the adult population can be categorized as having stage A (risk factors for) or stage B (pre) heart failure.9 This being related to the high prevalence of disease conditions such as diabetes, hypertension and coronary artery disease. However, this also begs the question if all those patients need to be defined as having heart failure and which patient end up progressing to clinical overt heart failure. Clearly choosing the right preventive strategies in selected stage A patients such a diuretic-based antihypertensive therapies or angiotensin-converting enzyme inhibitors or angiotensin receptor blockers in hypertensive patients or sodium–glucose co-transporter 2 (SGLT2) inhibitors in diabetic patients, can reduce the occurrence of symptomatic heart failure. Additionally, the St Vincent's Screening to Prevent Heart Failure (STOP-HF) illustrated that using a NP biomarker approach in patients with stage A heart failure allows to better diagnose and manage the transitioning towards more clinically overt heart failure than classic preventive strategies for stage A (at-risk heart failure).10 In the STOP-HF trial, 41.6% of patients had an elevated NP in the intervention group and these patients underwent more cardiac investigations and were more often treated with renin–angiotensin–aldosterone system-based therapies at follow-up than patients undergoing classic preventive strategies (standard of care approach in stage A patients). Such approach ultimately leads to a lower occurrence of left ventricular (LV) dysfunction with or without heart failure, asymptomatic LV dysfunction, or emergency hospitalization for major cardiovascular events. Not surprisingly, the recent universal definition of heart failure recognizes this finding and more broadly defines stage B heart failure (Figure 1) incorporating the role of elevated NPs.8 However, the potential role of dedicated ultrasound congestion evaluation remains unstudied in this population of at-risk (stage A) and pre- (stage B) heart failure. In this issue of the Journal, Cuthbert and colleagues report on an interesting prospective sub-study of the Heart OMics in AGEing (HOMAGE).11 This prospective trial included patients at risk for heart failure (≥1 of: diabetes, hypertension, a previous myocardial infarction or coronary artery revascularization) and excluded patients with clinical suspected heart failure, atrial fibrillation or grossly elevated NPs [B-type natriuretic peptide (BNP) >280 ng/L or N-terminal pro-B-type natriuretic peptide (NT-proBNP) >1000 ng/L] or presence of isolated systolic dysfunction defined as a LV ejection fraction <45%. Patients underwent ultrasound evaluation of congestion at two study sites using a same approach of lung ultrasound, jugular vein analysis at rest and during Valsalva manoeuvre and vena cava analysis. However, Doppler analysis of the renal veins was not performed. Additionally, echocardiographic assessment of LV systolic and diastolic function was performed and NPs were measured, with cutoffs for abnormality being aligned with the European Society of Cardiology guidelines or the recent universal definition for pre- (stage B) heart failure (NT-proBNP >125 ng/L or BNP >35 ng/L). A total of 238 patients with stage A heart failure were prospectively included and enrolled in this analysis. Of these patients, a total of 73% patients had an elevated NP and, according to the universal definition of heart failure, can be deemed to have stage B or pre-heart failure. This percentage is higher in comparison to the screening of stage A patients in the STOP-HF trial (41.6% vs. 73%), but the STOP-HF trial used a higher BNP cutoff (>50 ng/L).10 Interestingly, of the total patient population with all ultrasound evaluations present (n = 214), a total of 30% (n = 64) had ≥1 ultrasound congestion marker and these patients more often had a dilated left atrium (92%) or elevated NP (75%). If ultrasound congestion imaging is a reliable indicator of elevated filling pressures, one-fourth of patients were missed to be reclassified from at-risk (stage A) to pre- (stage B) heart failure if NPs were only used. Indeed, building on the question of the incremental value of ultrasound congestion imaging on top of NPs, authors show only a moderate correlation of ultrasound congestion imaging with NPs. Furthermore, 19% of patients (n = 12 of 64) only had congestion on ultrasound imaging and no abnormal NPs, no structural abnormalities (e.g. left atrial volume index <34 mL/m2) or functional abnormalities (E/e′ <13). Taken together potentially suggesting incremental information being gained by ultrasound congestion imaging. In a purely hypothesis generating (underpowered) analysis, the addition of presence of congestion on ultrasound imaging on top of NPs in patients at risk for heart failure better identified patients developing a post-hoc defined combined endpoint (unplanned cardiovascular admission, prescription of a loop diuretic, diagnosis of atrial fibrillation, diagnosis of heart failure, or death). This interesting analysis raises several questions. Can ultrasound congestion imaging help us better define those with pre- (stage B) heart failure at risk (Figure 1)? Can identification of a different mechanism (ultrasound congestion imaging) help us identify patients in need of potentially different treatment strategies (e.g. more emphasis on diuretic-based hypertensive instead of calcium antagonists, better follow-up and diagnosis like in the STOP-HF trial, SGLT2 inhibitors even in the absence of diabetes, and so on)? Can ultrasound congestion imaging help us better restage patients with at-risk (stage A) heart failure in settings where NP testing is not available? Next to the questions raised, several limitations need to be taken into account: (i) these ultrasound evaluations have never been validated in at-risk patients to reflect increased filling pressures (arguably one of the most important perpetuation mechanism in the transition between stages of heart failure) and therefore need invasive haemodynamic validation; (ii) this study is not powered to determine if adding ultrasound congestion imaging can better detect the transitioning to clinical overt (stage C) heart failure; (iii) absence of other biomarkers (soluble CD146 or carbohydrate antigen-125) potentially also reflecting a more vascular/systemic phenotype of congestion could have been informative as one-fourth of patients with congestion on ultrasound imaging had a normal NT-proBNP. Clearly, additional studies are necessary to further define the place of ultrasound congestion imaging in patients with at-risk (stage A) and pre- (stage B) heart failure, but Cuthbert and colleagues should be congratulated on their effort in paving the way. Pieter Martens is supported by a grant from the Belgian American Educational Foundation (BAEF) and from the Frans Van de Werf Fund. Conflict of interest: none declared.