Heart failure (HF) represents a complex clinical syndrome typically characterised by breathlessness, lower limb oedema and fatigue.1 When decompensated, HF is usually accompanied by observable signs, such as elevated jugular venous pressure, pulmonary crackles and peripheral oedema.1 HF stems from structural and/or functional abnormalities of the heart, leading to increased intracardiac pressure and/or insufficient cardiac output, both at rest and during physical exertion.1 Globally, approximately 64.3 million individuals are currently affected by HF.1–3 In developed countries, the prevalence of diagnosed HF typically ranges from 1% to 2% of the adult population.1–3
Following the initial diagnosis, individuals with HF typically require hospitalisation approximately once per year.4 In the Olmsted County cohort from 2000 to 2010, the average hospitalisation rate was 1.3 per patient per year.5 Research from various European countries and the US indicates that HF hospitalisation rates peaked in the 1990s before declining.6–8 However, recent findings from many European countries demonstrate an increase in both HF-related and non-cardiovascular admissions in the HF patient population.9–11 The rise in HF hospitalisations is particularly notable among women, possibly due to higher rates of comorbidities.12 Predictors of HF hospitalisations include AF, elevated BMI, increased HbA1c levels and reduced glomerular filtration rate.4,12 Of note, individuals with diabetes face a 1.5-fold higher risk of HF-related hospitalisation.4
Projections indicate a substantial increase in HF-related hospital admissions in the future, potentially rising by up to 50% over the next 25 years due to population growth, ageing and the increasing prevalence of comorbid conditions.4,12 Furthermore, it has been demonstrated that around 65% of HF patients experienced their first hospitalisation during a median follow-up of 3 years, and 55% died.13
Congestion is one of the hallmarks of HF, present in over 97% of patients with acute decompensated HF, whereas patients with de novo HF present mostly with cardiogenic shock and acute pulmonary oedema.14 Typical symptoms of congestive HF include breathlessness, orthopnoea, paroxysmal nocturnal dyspnoea, reduced exercise tolerance, fatigue, tiredness and increased time to recover after exercise.15 Typical signs include elevated jugular venous pressure, hepatojugular reflux, third heart sound (gallop rhythm) and laterally displaced apical impulse. Less specific signs include weight gain (>2 kg/week), cardiac murmur, peripheral oedema (ankle, sacral, scrotal), pulmonary crepitations, pleural effusion, tachycardia and tachypnoea, hepatomegaly, ascites, cold extremities and oliguria.15 Symptoms of congestion are strongly associated with HF hospitalisation, including pulmonary oedema (OR 3.08), shortness of breath (OR 2.94) and peripheral oedema (OR 2.16).13
Residual congestion, defined as a high left ventricular diastolic pressure associated with signs and symptoms of HF, such as dyspnoea, rales and oedema persisting despite guideline-directed medical treatment, is the major risk factor for HF readmission and mortality.16 Therefore, prompt recognition of congestion and rapid optimisation of medical and device therapy are crucial to induce remission in this malignant process. In this paper, we discuss the definitions, prevalence and prognosis of HF decompensation; the clinical profile of patients with acute HF; the significance of assessing residual congestion; the definition, identification and estimation of residual congestion; observational studies and randomised clinical trials in residual congestion; the impact of residual congestion on the prevalence of acute HF; and current guidelines to prevent recurrent HF decompensation.
Definitions, Prevalence and Prognosis of Heart Failure Decompensation
Acute HF is defined as both de novo (new-onset) acute HF or acute decompensated HF (in patients with previously diagnosed and/or treated HF).4 Acute HF arises from a rapid elevation of intracardiac filling pressures and/or acute myocardial dysfunction, potentially resulting in diminished peripheral perfusion and pulmonary oedema.17 In contrast to de novo acute HF, individuals experiencing acute decompensated HF typically manifest signs and symptoms of congestion and fluid retention (e.g. weight gain, exertional dyspnoea, orthopnoea and refractory oedema), rather than the pulmonary oedema or cardiogenic shock associated with acute left ventricular systolic dysfunction.17 This difference arises from chronic dysregulated neurohormonal compensatory mechanisms, which act to sustain a haemodynamic balance despite deteriorating left ventricular function.17
It has been reported that the incidence of HF in European countries ranges from 1.99 to 6.55 per 1,000 person-years, whereas the prevalence of HF in Europe varies from 10.0 to 39.0 per 1,000 persons.18 The number of HF-related hospitalisations in Europe ranges between 931 and 6,116 per million and the average length of hospital stay for HF varies from 6 to 11.5 days.18 The overall in-hospital mortality rate is 5.5%, with significant variations between each clinical profile or the congestion/hypoperfusion classification at admission.18–20 The highest mortality rate was seen among patients with cardiogenic shock, with 50% of deaths occurring within the first 24 hours of admission.18 This highlights the importance of the early detection of hypoperfusion and the need for appropriate initial treatment in these cases. Regarding outcomes within 1 year, 27% of patients with acute HF died, 26% were readmitted due to HF and 44% either died or were readmitted within 1 year after discharge.18 The rate of all-cause hospitalisations within 1 year was notably high (45%) and remained unchanged after 4 years.18
Clinical Profile of Patients with Acute Heart Failure
It is important to distinguish between the most common aetiologies of new-onset acute HF and acute decompensated HF because they have distinct pathophysiological mechanisms. The primary cause of new-onset acute HF is cardiac ischaemia, characterised by (sub)total coronary stenosis resulting in diminished contractility within the myocardium supplied by the affected coronary artery. In such cases, all therapeutic interventions prioritise addressing haemodynamic instability and implementing reperfusion strategies aimed at restoring myocardial contractile function and, therefore, increasing stroke volume.17
Decompensated acute HF may manifest in various clinical scenarios. Population-based investigations of individuals with decompensated HF have revealed a significant occurrence of concurrent conditions such as AF/atrial flutter (30–46%), valvular heart disease (44%) and dilated cardiomyopathy (25%).17,21 Moreover, patients admitted to hospital with acute decompensated HF typically exhibit advanced age (71–75 years) and a prior diagnosis of HF (65–87%), with the majority also presenting with coronary artery disease (50–68%) and hypertension (53–72%).22
It is advised that patients be evaluated using an approach that focuses on assessing the severity of their condition rather than its underlying cause.4 This involves an initial clinical evaluation considering signs and symptoms of congestion (e.g. orthopnoea, refractory oedema and elevated jugular venous pressure) and peripheral perfusion (e.g. cold extremities, oliguria and narrow pulse pressure).4,23 Patients are categorised as either ‘wet’ or ‘dry’ based on their fluid status and as ‘cold’ or ‘warm’ depending on the assessment of their perfusion status. This comprehensive clinical evaluation classifies patients into four groups (warm and wet; warm and dry; cold and dry; cold and wet), not only enabling initial guidance for best HF therapy but also providing prognostic insights.4,23
Strategies to detect and address residual congestion are crucial to stopping readmission after an acute HF hospitalisation and improving long-term prognosis.
Definition, Identification and Estimation of Residual Congestion
In a recent post hoc analysis of the DOSE-AHF and CARESS-HF trials, only half the patients with acute HF were devoid of congestion symptoms (as assessed using the ‘orthodema’ congestion score to evaluate orthopnoea and peripheral oedema) upon hospital discharge, and these patients had reduced mortality and rehospitalisation rates over a 60-day observation period by 26% (absolute risk reduction of 18%).24 Similarly, using the Composite Congestion Score (CCS; evaluating orthopnoea, jugular venous distension and peripheral oedema), Ambrosy et al. found that residual congestion persisted in 46% of patients by day 7 or discharge, and these individuals faced elevated risks of HF-related readmission and mortality by 13% per each CCS point.25 In both studies, a relatively simple points-based system was used to calculate features of congestion, increased pressures in cardiac chambers and peripheral oedema.24,25 These signs or symptoms are easily observable in patients, even from a distance, and, most importantly, do not require invasive or costly diagnostic measurements. Each of the features was scored, and the overall score translated into patients at high or low risk of residual congestion. However, the assessment of residual congestion based only on clinical finding may be not sensitive enough to determine the full potential of decongestion and, therefore, observe clinically meaningful outcomes.
Furthermore, findings from the Esc-EORP-HFA Heart Failure Long-Term Registry demonstrated that among discharged HF patients, 30.9% exhibited residual congestion, which was correlated with elevated 1-year mortality compared with patients without congestion (28.0 versus 18.5%).26 Non-cardiac comorbidities that were independently associated with an increased risk of congestion at discharge included tricuspid regurgitation, diabetes, anaemia and elevated New York Heart Association (NYHA) class.26 Conversely, β-blocker use upon admission, a diagnosis of new-onset acute HF or any cardiovascular intervention during hospitalisation were associated with a reduced likelihood of residual congestion.26
Moreover, it has been postulated that incorporating objective indicators of decongestion, such as N-terminal pro B-type natriuretic peptide (NT-proBNP) levels, haematocrit or response to diuretics, alongside clinical evaluation could enhance the detection of residual congestion.27 This approach may result in more precise and safe treatment and monitoring protocols.27 Rubio-Gracia et al. also found that residual congestion at day 7 or discharge is frequently observed in patients hospitalised for acute HF decompensation.27 The strongest predictors of significant residual congestion by day 7 were greater baseline congestion, reduced diuretic responsiveness and elevated blood urea nitrogen. Patients with significant congestion at day 7 had worse outcomes, including increased mortality and a higher risk of rehospitalisation for acute HF decompensation.27
As mentioned previously, achieving euvolaemia following acute HF decompensation can be challenging because residual congestion may persist even in patients with minimal signs of fluid overload.28 In addition, the resolution of dyspnoea and normalisation of body weight after treatment have been proven to be unreliable indicators of successful decongestion.24,29 Hence, simple clinical approaches for evaluating euvolaemia continue to represent an unmet need in acute HF management.28 Lung ultrasound (LUS) is gaining prominence for assessing pulmonary congestion, surpassing X-ray in detecting interstitial oedema or pleural effusions.
The Heart Failure Association of the European Society of Cardiology (ESC) recommends a multiparameter-based assessment for euvolaemia/residual congestion before discharge.30 The use of LUS along with other parameters indicative of congestion (natriuretic peptides, E/e′ ratio, inferior vena cava width) may aid in assessing residual congestion.31 Studies have demonstrated that residual congestion at discharge, determined by a B-line count (≥30), serves as a robust prognostic factor. For example, a systematic review demonstrated that in patients with acute HF, the presence of ≥15 B-lines on 28-zone LUS at discharge identified individuals at a greater than fivefold increased risk for HF readmission or death.32 Similarly, among ambulatory patients with chronic HF, the presence of three or more B-lines on five- or eight-zone LUS indicated patients at a nearly fourfold increased risk of 6-month HF hospitalisation or death.32 Therefore, LUS could serve as a valuable tool for detecting and tracking congestion and optimising therapy during and/or after HF hospitalisation, a proposition that warrants further validation through multicentre studies.31,32 Table 1 provides a summary of the methods for congestion assessment, including details on their implementation and interpretation, as well as our personal perspectives on the advantages and disadvantages of each method.
Observational Studies in Residual Congestion
The assessment of parameters indicative of residual congestion in observational studies has been extensively investigated. There is robust scientific evidence from observational studies confirming the association of natriuretic peptides, changes in echocardiographic parameters and LUS with residual congestion.33
The clinical course, predictors and prognostic value of congestion were studied in 1,572 patients admitted for acute decompensated HF by including different indirect markers of congestion (clinical congestion, natriuretic peptide dynamics, haemoconcentration and diuretic response).27 After 7 days of hospitalisation or at discharge (whichever came first), nearly 30% of patients were still significantly congested (CCS ≥3), approximately 50% were mildly congested (CCS 1 or 2) and only 24% had no signs of residual congestion (CCS 0). The presence of significant residual congestion at day 7 or discharge was associated with an 88% higher risk of readmissions for HF by day 60 and a 54% increase in all-cause mortality by day 180.27 Importantly, the diuretic response provided only modest additional prognostic value on top of residual clinical congestion.27 That study clearly demonstrated that most patients with acute decompensated HF still have signs of residual congestion 7 days after admission, which is associated with a higher likelihood of worse outcomes. Decongestion surrogates, such as diuretic response or natriuretic peptides, are significant predictors of outcomes, but they do not provide additive prognostic information, making clinical congestion still the most crucial part of patient evaluation at discharge.
In the OPTIMIZE-HF registry linked to Medicare claims in the US, 24,724 patients aged ≥65 years hospitalised for acute decompensated HF were analysed.34 Congestion severity was measured using a 15-point score that included clinical variables (dyspnoea, orthopnoea, fatigue, jugular venous pressure, rales and oedema). Patients with the highest congestion score had the highest rates of recent HF hospitalisations, an ejection fraction ≤40% and comorbidities (arrhythmias, diabetes and renal insufficiency). After adjustment for patient characteristics at baseline, it was found that an increase in congestion score by 3 points was associated with a 6% increase in mortality and a 9% increase in HF rehospitalisation at 1 year.34 The results of that study confirm the significance of easily accessible clinical parameters of congestion for risk stratification in patients with acute decompensated HF.
In the Japanese KCHF Registry, clinical congestion parameters (peripheral oedema, jugular venous pressure and orthopnoea) used to calculate the CCS were evaluated in 3,787 patients hospitalised due to decompensated HF.35 At discharge, as many as 85% of patients had a CCS of 0 (complete decongestion) and only 15% had a CCS of ≥1 (residual congestion), which is much less than the European data.27 In the Japanese KCHF Registry study, residual congestion increased the risk of all-cause death or recurrent HF hospitalisation by 30% at 1 year after discharge.35 However, the baseline CCS was also correlated with the risk of death or HF hospitalisation after discharge, even in the group with complete decongestion.35 These differences may be due to the fact that the length of hospitalisation in Japan is generally longer (median hospitalisation length 16 days), which facilitates more extensive decongestion, than in Europe, where hospitalisation length ranges between 8 and 10 days. Altogether, the severity of congestion at both admission and discharge are crucial factors to consider when planning a patient’s follow-up visits after hospital discharge because both are predictive of adverse outcomes.
In the retrospective population-based Japanese KUNIUMI Registry, the combined effect of residual congestion at discharge and worsening renal failure (WRF) during the hospital stay, defined as an increase of ≥0.3 mg/dl in serum creatinine levels, was analysed.36 Among 966 acute decompensated HR patients, rates of cardiovascular death or HF rehospitalisation over a mean follow-up period of 2.0 years were higher among those with either in-hospital WRF or residual congestion at discharge than in those without WRF and residual congestion (HR 1.61 and HR 2.34, respectively).36 Patients with both these two characteristics had the least favourable outcomes. However, residual congestion at discharge, but not WRF, was an independent predictor of outcomes in patients aged ≥80 years.36
These findings underline the importance of aggressive diuretic therapy to alleviate congestion, especially in older patients, who are generally more likely to be treated less intensively and discharged at lower diuretic doses (e.g. due to orthostatic hypotension concerns). Although WRF is a well-established negative prognostic factor in patients with acute HF, it seems less clinically important than the residual congestion.
The Congestion and Renal Index (CRI; ml/min/kΩ) is calculated as the ratio of the thoracic fluid content index divided by the estimated glomerular filtration rate before discharge. Patients with a higher CRI had a 56% higher risk of cardiovascular death or HF hospitalisation at 1 year compared with those with a lower CRI.37 CRI improved the risk stratification of the established risk models based on clinical variables and NT-proBNP.37 However, impedance cardiography, which is a non-invasive technology measuring total electrical conductivity of the thorax, is not routinely available in most cardiology departments, limiting the clinical utility of the CRI.
Another observational study investigated the association between residual congestion and renal dysfunction in 944 acute HF patients.37 The authors of that study proposed an interesting single indicator, the CRI, to reflect the combined effects of these two diseases on outcomes. Altogether, based on the results of observational studies, abnormalities in the natriuretic peptides, echocardiographic parameters and LUS are all associated with residual congestion and adverse outcomes, but the most robust evidence exists for the clinical congestion parameters routinely evaluated during physical examination of the patient. In addition to residual congestion, particular attention should be paid to the presence of in-hospital WRF, which aggravates the risk of post-discharge mortality and rehospitalisation in patients with acute decompensated HF.
In data from the STRONG-HF trial, biologically active adrenomedullin (bio-ADM) was found to be strongly associated with most indicators of congestion, with the exception of rales.38 Variations in bio-ADM levels were closely linked to changes in congestion status between baseline and day-90 follow-up.38 Individuals with highest baseline bio-ADM levels had significantly elevated risks compared with those in the lowest concentration of that biomarker, including a twofold greater likelihood of 180-day all-cause mortality or HF rehospitalisation and a more than twofold increased risk of 180-day HF rehospitalisation.38
Randomised Clinical Trials to Detect and Treat Residual Congestion
In this section, we present data from randomised control trials (RCT) regarding the assessment of residual congestion in the context of managing patients with acute HF and medications potentially resulting in successfully achieving euvolaemia.
In the single-blind LUS-HF RCT, 123 patients admitted for HF were randomly assigned to receive either standard follow-up (n=62; control group) or LUS-guided follow-up (n=61).39 The primary endpoint consisted of a composite of urgent visits, hospitalisations for worsening HF and death during the follow-up period. Visits were scheduled at 14, 30, 90, and 180 days after discharge. Physicians treating the patients were encouraged to adjust diuretic therapy based on the number of B-lines detected by LUS.39 The HR for the primary outcome in the LUS-guided group was 0.52 (p=0.049), primarily driven by a reduction in the number of urgent visits for worsening HF.39 The number of patients needed to treat to prevent an event was five. In addition, patients in the LUS-guided follow-up group received more loop diuretics and demonstrated an improvement in the 6-minute walking test.39
Previous studies have indicated that lung impedance-guided therapy can decrease hospitalisations for acute HF. In a single-blinded RCT from two medical centres involving 256 patients with HF and left ventricular ejection fraction ≤35% in NYHA Class II–IV who were admitted for acute HF within 12 months before recruitment, patients were randomly assigned to a control group managed by clinical assessment and a monitored group, whose therapy was supplemented by lung impedance guidance, and followed for at least 12 months.40 Patients, unaware of their group assignment, attended monthly outpatient clinic visits. The primary endpoint was acute HF hospitalisations, with secondary endpoints including all-cause hospitalisations and mortality.40 Significant reductions in HF hospitalisations were observed in the lung impedance-guidance group in both the first and second years of follow-up (58% and 45% decrease, respectively). In addition, the authors reported a substantial decrease in mortality (HR 0.52; 95% CI [0.35–0.78]).40 These data highlight the potential of new technologies, such as lung impedance-guided HF therapy, to alleviate the burden of HF-related hospitalisations.
Some RCT studies are currently under way. The CAVAL US-AHF (NCT04549701) study is designed as a single-centre RCT to compare the effectiveness of an ultrasound-guided healthcare approach targeting the inferior vena cava (IVC) and LUS against standard HF care in reducing residual congestion upon discharge.41 Acute HF patients will be randomly assigned to either the intervention group (LUS- and ultrasound-guided decongestion therapy) or the control group (standard of care: clinically guided decongestion therapy).41 Patients will be categorised based on the severity of congestion (none or mild, moderate or severe). Treating physicians will be aware of the test results and will adjust treatment accordingly based on a tailored therapeutic protocol. The primary endpoint is defined as the presence of more than five B-lines and/or changes in IVC diameter, with or without collapsibility.41 Secondary outcomes include a composite of HF rehospitalisation, unplanned visits due to worsening HF or death within 90 days.
Another trial (NCT05035459) has been proposed to assess whether intensive patient management using LUS-guided diuretic therapy will aid in identifying patients with residual congestion, thereby helping to lower the risk of HF hospitalisation and mortality through intensified decongestion.42 After receiving guideline-directed medical therapy for HF, patients with three or more B-lines detected by LUS within 48 hours before discharge will be randomly assigned in a 1:1 ratio to either the conventional HF management group (standard of care) or the LUS-guided intensive HF management group.42 The primary endpoint comprises a composite of HF readmission and all-cause mortality during the 1-year follow-up.42
Another interesting example of precise HF monitoring after discharge in patients with recent acute HF is the use of specialised, non-invasive, wearable cardiopulmonary management devices to assess residual congestion. A study on this topic is currently under way (NCT05026034), but no detailed design or rationale of the study has been published.
Similarly, evaluating patients during the pre-discharge period using bioimpedance scales may help identify patients with excess total body water, which may have prognostic significance. A study concerning this issue is also currently ongoing (NCT02394470), but no study design details have been published. While ongoing studies are investigating advanced methods for monitoring residual congestion and total body water in HF patients, such as wearable cardiopulmonary management devices and bioimpedance scales, existing large RCTs provide some insights into how additional therapeutic strategies can improve patient outcomes.
In the available large RCTs that evaluated additional diuretic therapy using acetazolamide,43,44 hydrochlorothiazide,45 sodium–glucose cotransporter 2 (SGLT2) inhibitors46–49 and intensive pharmacological management with frequent follow-up visits,50 achieving better euvolaemia (addressing residual congestion) demonstrated increased diuretic effect or clinical benefits for these patients, including a reduction in HF hospitalisations.51 Hence, large RCTs have shown that additional therapies using acetazolamide, hydrochlorothiazide or SGLT2 inhibitors and intensive management improve euvolaemia, leading to enhanced diuretic effects and reduced HF hospitalisations.
Impact of Residual Congestion on Acute Heart Failure Prevalence
The results of observational studies and RCTs have clearly demonstrated that the residual congestion not only represents a cardinal sign of HF but is also the primary cause of recurrent hospital admissions due to acute HF. Acute HF affects >10 million patients annually and, among patients who survive to discharge, early readmission rates are >30% regardless of the healthcare system.52 One reason for such a high readmission rate is the fact that most hospitalised patients do not achieve the optimal decongestion. Therefore, acute HF has been termed ‘a malignant condition’ and residual congestion ‘a pandemic’.16 The situation is complicated by the fact that up to 40% of patients considered ‘dry’ according to pulmonary auscultation present with subclinical congestion at hospital discharge, which can be detected by LUS and is associated with a worse prognosis.53
Every subsequent HF readmission is associated increased mortality. Compared with patients who are not hospitalised, those hospitalised for acute HF have a more than sixfold increased risk of death in the first month after discharge.53 This increase may be up to 12- to 16-fold in patients with multiple hospitalisations.54 In fact, the recent ESC guidelines define more than one unplanned visit or hospitalisation per year as one of the four criteria of advanced HF.4 Hence, residual congestion is one of the strongest predictors of acute HF, cardiovascular events and poor outcomes. Therefore, identification of residual congestion in HF patients at hospital discharge and optimal management in the early post-discharge phase are critical to prevent devastating consequences after acute decompensated HF. The current guidelines to detect clinical and subclinical congestion and prevent recurrent HF decompensation are discussed below.
Current Guidelines to Prevent Recurrent HF Decompensation
Patients admitted for HF should undergo a thorough assessment to ensure the absence of lingering signs of congestion prior to discharge, and pharmacotherapy should be optimised accordingly.4 It is also recommended to schedule an early follow-up visit within 1–2 weeks after discharge to evaluate the signs of congestion and medication tolerance, as well as to initiate and/or adjust evidence-based therapy as needed.4 Similarly, based on the favourable clinical outcomes of the STRONG-HF trial, which demonstrated the effectiveness of closely monitoring patients with acute HF after discharge, along with prompt adjustment of guideline-directed medical therapy (including increasing diuretic dosage), it is advisable to closely monitor signs of congestion in these patients.50 Consequently, the current 2023 ESC guidelines for managing chronic and acute HF advocate for an intensive approach to initiate and promptly uptitrate evidence-based medications before discharge and during frequent follow-up. This should be achieved through regular follow-up visits within the initial 6 weeks after HF hospitalisation to decrease the risk of rehospitalisation or mortality.55
Although two RCTs demonstrated an increase in diuresis, including natriuresis, using acetazolamide43,44 and hydrochlorothiazide45 added to standard loop diuretic therapy in patients with acute HF, current 2023 ESC recommendations do not advocate for the routine use of these drugs in patients with acute HF.55 However, these studies have shown the safety of such approach.43–45 The potential additional benefits of wearable technologies for monitoring heart rate, rhythm or lung congestion (such as bioimpedance or lung radar) compared with traditional HF management have been noted for identifying patients with an inadequate response to current HF therapy.40,56,57 However, the clinical outcomes of these procedures remain uncertain, and they are not currently recommended.4 An increase in diastolic pulmonary artery pressure could indicate early signs of congestion. A preliminary trial indicated a decrease in the likelihood of recurrent HF hospitalisations.58 However, in the GUIDE-HF trial, a management approach guided by haemodynamics for heart failure did not lead to a lower combined endpoint rate of mortality and overall heart failure events compared with the control group.59 Nevertheless, the 2021 ESC HF guidelines recommend considering a wireless haemodynamic monitoring system in symptomatic patients with HF to improve their clinical outcomes (Class IIb recommendation).4
Residual Congestions: Outlook
Gaps in Knowledge
Optimising the pre- and post-discharge management after an acute HF hospitalisation to improve patient care is a major unmet need. Although residual congestion should be rapidly identified and addressed using full doses of all available effective medications, it remains a challenge for at least three reasons.
First, it remains unclear how fluid status should be measured in acute HF patients at admission and/or discharge. Depending on the study, the criteria for congestion assessment take into account non-invasive clinical parameters (NYHA class, orthopnoea, peripheral oedema, jugular venous pressure), invasive pressure measurements (central venous pressure, pulmonary capillary wedge pressure), biomarkers (NT-proBNP, haemoglobin) and imaging (IVC ultrasound, B-lines LUS). Although the use of objective parameters to assess decongestion and guide diuretic therapy is appealing, their incremental value remains to be demonstrated in RCTs.60
Second, there are currently no standardised definitions regarding the decongestion endpoint. RCTs where optimal decongestion was used as an endpoint proposed to define it as the complete resolution of signs and symptoms of congestion, jugular venous pressure ≤8 cmH2O, central venous pressure ≤5 mmHg, pulmonary capillary wedge pressure ≤12 mmHg, NT-proBNP ≤400 pg/ml, IVC diameter <21 mm and ≥50% collapsibility.60 However, these absolute cut-offs do not take into account the dynamics of decongestion in an individual patient, depending on how severe the baseline condition was. To address this problem, the term ‘satisfactory decongestion’ has been introduced, covering the optimal decongestion and significant decongestion. Whereas optimal decongestion is the ultimate therapeutic target and refers to a ‘dry’ state, significant decongestion is defined as a substantial improvement in congestion parameters without achieving euvolaemia (mild residual congestion). Notably, a caveat of clinical parameters to assess decongestion is that they rely on subjective assessment, which may introduce potential bias in an RCT. Conversely, the objective parameters have hitherto failed to demonstrate superiority over clinical assessment.60
Third, ineffective decongestive therapy leads to serious adverse events ranging from prolonged length of stay through to the need to initiate vasoactive agents or renal replacement therapy, to the escalation to mechanical circulatory support and to death. However, the decongestive treatment failure endpoints also need to be objectively defined to increase the quality of RCTs aimed at assessing pharmacological- and device-related decongestion therapy.
Diagnostic Algorithm
The recent scientific statement by the ESC Heart Failure Association has addressed the challenges described above, providing guidance on the vulnerable period of predischarge and early post-discharge (<3 months) management of patients hospitalised for acute HF.61 During the predischarge phase, a multiparametric evaluation including clinical assessment, biomarkers and imaging is mandatory to exclude or minimise residual congestion, optimise pharmacotherapy and plan post-discharge management, including pharmacotherapy uptitration and specific post-discharge programmes (e.g. rehabilitation). The risk factors for residual congestion and/or HF readmission at discharge are summarised in Figure 1.
To evaluate residual congestion, the guideline-recommended CCS or simplified CCS should be used. The latter includes only three parameters (orthopnoea, oedema and jugular venous distension), easily obtained upon physical examination. Blood pressure and heart rate are also crucial for both risk stratification and treatment optimisation. For example, patients with hypotension are less likely to tolerate angiotensin-converting enzyme inhibitors or angiotensin receptor–neprilysin inhibitor, whereas those with a low heart rate are likely to be unable to tolerate β-blockers.
Among laboratory measurements, natriuretic peptides are important markers of congestion, beyond their major role in the diagnosis and risk stratification of patients with acute HF. Therefore, measurement of natriuretic peptides is recommended at both admission and hospital discharge. The prognostic value of additional parameters, including haemoconcentration, urinary sodium and/or urine volume measurements, carbohydrate antigen 125 and adrenomedullin level, remains to be further investigated.
Finally, comprehensive imaging has a major role for the detection of residual congestion at discharge. The guideline-recommended methods include chest X-ray to detect pulmonary congestion and/or pleural effusion, echocardiography to identify increased left ventricular filling pressures, mitral and tricuspid regurgitation severity, IVC size and collapsibility and LUS to assess the B-lines.
Treatment Optimisation
The recently proposed road map for the management of HF patients during the vulnerable phase after hospitalisation proposed to divide this period into three phases after hospital discharge: the very early vulnerable phase (0–30 days), the early vulnerable phase (30–60 days) and the late vulnerable phase (60–90 days).62 Treatment optimisation during this period is summarised in Figure 2.
The very early phase is associated with the highest risk of rehospitalisation and death because most patients are discharged with residual congestion and without the adequate uptitration of life-saving medications. According to the ESC guidelines, patients are eligible for discharge only if they are euvolaemic and haemodynamically stable on evidence-based oral medication for at least 24–48 hours, with stable renal function for at least 24 hours.4 The clinical and laboratory parameters that should be measured at discharge include orthopnoea, hypotension, tachycardia, impaired renal function, low sodium level, low albumin and increased levels of natriuretic peptides.62 In addition, the predischarge iron status (transferrin, ferritin) should be evaluated and addressed due to the association between IV iron administration and reduced risk of recurrent hospitalisation, especially in patients with heart failure with reduced ejection fraction.63
Guideline-recommended pharmacotherapy, including β-blockers, angiotensin-converting enzyme inhibitors, mineralocorticoid receptor antagonists and SGLT2 inhibitors, may be discontinued or reduced during hospitalisation due to haemodynamic instability and/or renal impairment, and not fully re-started before discharge; therefore, it is crucial to educate patients and/or their caregivers to promptly identify symptoms or signs of recurrent congestion and clearly define the plan for uptitration of pharmacotherapy in the early post-discharge period. An early follow-up visit (clinic, home visit, telephone call or telemonitoring) is recommended at 1–2 weeks after discharge.4
In the next phase (early vulnerable period), patients should be euvolaemic and receive guideline-recommended oral medications. Based on results of the STRONG-HF trial, rapid uptitration of oral HF therapies in the first 6 weeks after an acute HF hospitalisation is recommended to reduce HF readmissions and all-cause death.50 Although the STRONG-HF trial included triple therapy without SGLT2 inhibitors, this recommendation also includes empagliflozin or dapagliflozin, which can be initiated either in hospital or (at latest) in the early discharge period.55 If possible, patients should be enrolled in a cardiac rehabilitation programme to improve symptoms, functional capacity and quality of life, as well as to prevent rehospitalisation. The follow-up visit should be scheduled to assess congestion, drug tolerance and treatment adherence. Particular attention should be paid to blood pressure, heart rate, NT-proBNP levels, potassium levels and renal function. In patients with hyperkalaemia, potassium binders may enable the uptitration of renin–angiotensin–aldosterone system inhibitors.
Finally, in the late vulnerable phase, oral medication treatment should be uptitrated to the maximum tolerated doses and device therapy should be implemented, if necessary. In heart failure with reduced ejection fraction patients who are still symptomatic and have raised natriuretic peptide levels despite guideline-directed medical therapy, angiotensin receptor–neprilysin inhibitor should be given as a replacement for angiotensin-converting enzyme inhibitors or angiotensin receptor blockers. Preferably, patients should be participating in dedicated cardiac rehabilitation programmes, which can be delivered onsite or through telemedicine. Non-invasive home telemonitoring options to track congestion are growing exponentially and may be considered as well.4,64
Conclusion
The vulnerable post-discharge phase after an acute HF hospitalisation is characterised by an extremely high risk of death and rehospitalisation in patients with versus those without decompensation. Residual and recurrent congestion in the predischarge and early post-discharge phases is the major risk factor for adverse outcomes. Strategies to detect and address residual congestion are crucial to stopping readmissions after an acute HF hospitalisation and improving long-term prognosis.