Congestion is the leading cause of hospital admission in acute decompensated heart failure (ADHF), and residual congestion at discharge is a well-established indicator of poor prognosis.1 Congestion arises from multiple, interacting mechanisms. Cardiac dysfunction leads to increased filling pressures and a low cardiac output. To preserve circulatory volume and organ perfusion, compensatory activation of the neurohormonal and sympathetic nervous system promotes the retention of renal sodium and water, and blood volume redistribution that ultimately exacerbates congestion. Once considered a simple, uniform process of blood volume expansion, congestion is now understood to be a complex, heterogeneous and dynamic phenomenon that actively drives the progression of heart failure (HF) syndromes.2 Consequently, accurate and reliable tools for assessing congestion are essential when attempting to improve HF outcomes.
The European Society of Cardiology (ESC) provides comprehensive guidance for diagnosing and managing HF; however, no standardised framework exists for evaluating congestion.3 This review outlines the evidence base for current methods, including clinical examination, biomarkers, ultrasonography, invasive techniques and emerging novel tools, as well as their strengths and limitations (Supplementary Table 1).
Current Methods for Assessing Congestion
Clinical Signs and Symptoms
Clinical assessment remains the first step in evaluating congestion; however, its reliability has declined with increasing reliance on biochemical and imaging tests, while rising obesity rates further obscure signs of central and peripheral congestion (Figure 1 ).4
Jugular Venous Pressure and Near Infrared Spectroscopy
In ADHF, elevated central venous pressures (CVP) are detected as elevated jugular venous pressure (JVP), although the correlation is only modest. In a prospective study of 217 patients with ADHF, bedside JVP assessment demonstrated poor accuracy for detecting elevated CVP as measured by ultrasound, with a sensitivity and specificity of 57.3% and 43.6% for CVP ≥5 mmHg, and 67.8% and 30.4% for CVP ≥12 mmHg, respectively.5
To improve non-invasive estimates of filling pressures, newer techniques using near-infrared spectroscopy (NIRS) applied over the external jugular vein have proven benefits compared to clinical examination alone. In a prospective study, Pellicori et al. demonstrated that among patients with a clinically normal JVP, 36% had elevated right atrial pressure (RAP) ≥10 mmHg as measured by NIRS. Moreover, patients with elevated RAP ≥10 mmHg shown by NIRS had a higher risk of adverse outcomes (HR 2.38; 95% CI [1.19–4.75]; p=0.014). In addition, patients with low clinical JVP but elevated RAP shown by NIRS had a nearly twofold higher risk of death or HF hospitalisation (HR 1.93; 95% CI [0.85–4.37]; p=0.12), suggesting that NIRS may detect occult elevation in filling pressures missed by clinical examination of JVP and could enhance the assessment of congestion in HF.6
Rales, Pleural Effusion and Peripheral Oedema
Tissue congestion can be classified in two main types: pulmonary congestion and systemic congestion.
Pulmonary congestion can manifest as rales and pleural effusion; however, absence of these signs does not exclude the presence of congestion. Rales are not specific to congestion and may occur in various pulmonary conditions. Stevenson et al. found that paradoxically, pulmonary rales were present in only 19% of patients with a pulmonary capillary wedge pressure >22 mmHg.7
Peripheral oedema is a sensitive marker of systemic congestion but has low specificity as factors such as low oncotic pressure and increased vascular permeability can contribute to fluid shifting into the extravascular compartment and so the assessment of peripheral oedema in HF should always be interpreted within the overall clinical context.8
To standardise clinical assessment, the ADVOR trial developed a composite congestion score incorporating peripheral oedema, rales and other clinical markers to quantify congestion severity and guide diuretic therapy. Although the score supported early decongestion, its direct relationship with long-term clinical outcomes has not yet been established.9
Weight Monitoring
Weight loss is a commonly used marker to assess decongestion in patients admitted with ADHF. Urine output as a measure of fluid loss should theoretically correspond to equivalent weight reduction. Therefore, weight change is often used as a surrogate for net fluid loss. However, a pooled analysis of 708 patients from the DOSE-HF, ROSE-AHF and CARRESS-HF studies demonstrated concordance between weight loss and net fluid loss at 72–96 hours in only 55% of cases.10 Measurement errors related to patient factors, timing, scale variability and prolonged hospitalisation with muscle wasting may confound the relationship between weight and fluid loss.
Despite the modest correlation with fluid loss, weight change still has prognostic value. In the propensity-matched OPTIMIZE-HF registry (n=8,830), weight loss was linked to lower 30-day (HR 0.75; 95% CI [0.63–0.90]) and 60-day all-cause mortality (HR 0.80; 95% CI [0.70–0.92]) although this association was not sustained at 6 months. Overall, while observational data suggest greater in-hospital weight loss could predict improved short-term outcomes, evidence that targeted weight loss itself improves prognosis remains limited.
Clinical Congestion Scores
To improve the accuracy of standalone clinical signs, several composite congestion scores have been developed (Table 1). In a post hoc analysis of the placebo arm (n=2,061) of the EVEREST trial, Ambrosy et al. found that an EVEREST score ≥3 at discharge was associated with 35% re-hospitalisations for HF and 43% all-cause mortality risk at approximately 10 months, compared to 26% and 19% for those with a score of 0, respectively.11 The Stevenson classification, which stratifies patients by congestion and perfusion status, showed poor correlation with echocardiographic congestion markers in a multicentre study by Palazzuoli et al.12 While more accurate than individual signs, clinical congestion scores lack standardisation and robust validation and their role in guiding decongestive therapy remains to be defined.
Biomarkers
Natriuretic Peptides
Brain natriuretic peptide (BNP) and N-terminal pro B-type natriuretic peptide (NT-proBNP) are widely used biomarkers that are highly sensitive but non-specific markers of AHF. The biomarker can be considered a sum of two components: the ‘basal’ level and the ‘wet’ level. The basal NT-proBNP level in patients with chronic HF correlates with their New York Heart Association (NYHA) class and overall prognosis. The wet component, typically 25–50% above baseline, reflects acute increases in intracardiac pressure and volume overload during decompensation; this component rapidly declines with diuretic therapy.13 A relative reduction in NT-proBNP of approximately 30% is associated with good short-term prognosis, and the absolute natriuretic peptide (NP) level at discharge remains a stronger predictor of long-term outcomes.13
NT-proBNP has been validated as a prognostic marker in chronic HF, independent of ejection fraction and is incorporated in ESC guidelines for risk stratification.3,14 However, its role in guiding decongestion in AHF is controversial. A meta-analysis of nine randomised controlled trials (RCTs) comparing NP-guided decongestion with standard care in patients with AHF found no significant effect on all-cause mortality, cardiovascular death or the composite outcome of cardiovascular death and HF hospitalisation.15 The two largest trials included in this analysis were GUIDE-IT (n=894) and STRONG-HF (n=1,078). GUIDE-IT demonstrated no difference in outcomes between the NT-proBNP guided and usual care group, likely due to similar NT-proBNP target achievement (<1,000 pg/ml) and comparable use of guideline-directed medical therapy (GDMT). In contrast, STRONG-HF demonstrated that the study group that had intensive GDMT up-titration led to greater NT-proBNP reduction and improved outcomes compared to the usual care group. However, the heterogeneity in trial design and implementation among studies included in this meta-analysis makes it difficult to draw firm conclusions regarding NT-proBNP-guided therapy in the acute setting.
The use of NT-proBNP is also limited by significant intra-individual biological variations, genetic predisposition (such as polymorphisms in the NPPB gene and ethnicity), older age, and comorbidities such as AF, renal dysfunction, sepsis and obesity in a growing multimorbid geriatric population.16
Mitochondrial DNA
Mitochondrial dysfunction is a hallmark of HF as the myocardium shifts towards less efficient pathways of energy production to meet increased metabolic demands. This inefficiency leads to elevated oxidative stress and inflammation, driving adverse cardiac remodelling and progressive HF. Mitochondrial DNA (mtDNA) has been shown to be a good surrogate marker of mitochondrial function.17
Evidence regarding the use of cellular or circulating mtDNA as a biomarker for HF risk stratification has been inconsistent. A meta-analysis of three studies, each employing different methodologies to quantify cellular mtDNA, demonstrated an inverse relationship between cellular mtDNA content and the presence of HF (HR 1.30; 95% CI [1.07–1.56]).18 In contrast, an observational study by Mengozzi et al. found no association between cellular mtDNA and AHF but reported significantly higher levels of cell-free mtDNA, which independently predicted worsening congestion (defined as ≥2 ultrasound signs of congestion) (coefficient 1.203 or 3.329; 95% CI [1.017–10.902]).17 The authors proposed that this reflects a compensatory upregulation of mitochondrial biogenesis in stable HF to meet energy demands, which becomes exhausted during acute decompensation. This hypothesis is further supported by findings from Krychtiuk et al., who demonstrated that patients with AHF had markedly elevated circulating cell-free mtDNA levels compared to those with chronic HF and that higher levels correlated with a prognostic benefit for 30-day mortality, independent of renal function and NTproBNP (p=0.002).19
mtDNA is associated with systemic inflammation (measured by C-reactive protein levels) and reduced exercise tolerance (measured by cardiopulmonary exercise testing) indicating that mtDNA may serve as an integrative biomarker reflecting the interplay between mitochondrial dysfunction, inflammation and impaired cardiovascular reserve, extending its potential usefulness beyond traditional HF populations.17
Further research is needed to establish standardised measurement techniques and the clinical applicability of circulating mtDNA before it can be adopted.
Inflammatory Cytokines
Recent evidence highlights the interplay between persistent congestion, renal dysfunction and inflammation in determining outcomes in AHF. In a prognostic study by Pugliese et al., patients with concomitant chronic kidney disease (CKD) and congestion exhibited elevated levels of tumour necrosis factor-α (TNF-α) (p=0.037) and its soluble receptors, alongside reduced transforming growth factor-β1 (TGF-β1) (p=0.02) and a higher risk of unfavourable clinical outcomes.20 Both CKD (p=0.02) and congestion (p=0.01) independently predicted adverse events, but TNF-α mediated a substantial portion of their effect, explaining approximately 40% of the association with poor prognosis. These findings suggest inflammatory activation may act as a mechanistic link between haemodynamic congestion, renal impairment and disease progression in acute HF, underscoring the potential of cytokine profiling to refine risk stratification and guide targeted therapies.
Other Biomarkers
Various other biomarkers, evidence for their use and their strengths and limitations are detailed in Supplementary Table 2.
Urinary Sodium
Urinary sodium is currently the only urinary biomarker recommended to guide management in acute heart failure (AHF), as it helps early identification of patients at risk of poor diuretic response. The 2021 ESC guidelines recommend that spot urinary sodium <50–70 mEq/l within 2 hours of treatment indicates diuretic resistance and warrants early dose escalation.3
This approach has been evaluated in two RCTs: PUSH-HF (n=301) and ENACT-HF (n=401), which compared urinary sodium-guided diuresis to standard care.21,22 Both demonstrated significantly higher natriuresis in the guided arms (PUSH-HF: 409 versus 345 mmol, p=0.0061; ENACT-HF: 538 versus 365 mmol, p<0.001), although neither observed a difference in mortality or HF events. Notably, ENACT-HF reported shorter hospital stays in the interventional group (5.8 versus 7.0 days, p=0.036). The recently completed DECONGEST-HF study is expected to further clarify the role of serial urinary sodium monitoring in this context.23
Despite its promise, several limitations must be considered. Large-scale trials demonstrating its practicality in routine clinical settings or outcome measures are lacking. Additionally, urinary sodium levels are highly variable due to differences in diet, chronic diuretic use and individual renal sodium handling.24 Furthermore, conditions such as acute kidney injury (AKI) or CKD may confound results; intrinsic tubular injury in AKI can raise urinary sodium by impairing reabsorption while CKD may cause salt wasting, increasing sodium excretion up to fourfold.24 Finally, data on how diuretics such as tolvaptan or sodium-glucose transport 2 (SGLT2) inhibitors affect urinary sodium remain limited.25 There is also debate about urinary volume rather than arbitrary time measurements of urinary sodium, possibly offering a more valuable assessment of diuretic response.26
While early evidence is encouraging, further large-scale, well-designed studies are needed to confirm its role in improving clinical outcomes and to guide its integration into routine practice.
Ultrasound Assessment
Echocardiography
Echocardiography is a widely used, non-invasive tool for assessing cardiac structure and function. Assessment of left ventricular ejection fraction and valvular abnormalities using echocardiography in patients with AHF can identify the underlying substrate of HF and inform subsequent management strategies, whether medical or surgical that are essential for optimising therapy and improving long-term outcomes (Figure 2).
Echocardiography with tissue Doppler imaging can also be used to semi-quantitatively estimate elevated left ventricular filling pressures, which have been associated with adverse prognosis.27 The latest British Society of Echocardiography guidelines provide a structured approach for assessing filling pressures in patients with normal and impaired systolic function and patients with AF. Assessment includes left atrial (LA) area, LA strain, maximum tricuspid regurgitant velocity and the ratio of early mitral inflow velocity (E) to early diastolic mitral annular velocity (e’).28 A study by Dokainish et al. demonstrated that an E/e’ value ≥15, when combined with a BNP level of ≥250 pg/ml measured the day before discharge, provided incremental predictive power for readmission or cardiac death beyond traditional clinical risk factors.29
Assessment of right ventricular (RV) function using tricuspid annular plane systolic excursion (TAPSE) and RV–pulmonary artery (PA) coupling has also been shown to correlate with severe pulmonary congestion and adverse outcomes in a pooled analysis of four cohorts.30 RV–PA uncoupling reflects a mismatch between RV contractility and pulmonary arterial afterload, indicating limited RV reserve and higher risk of morbidity.
Echocardiography has its limitations. Visualisation may be challenging in patients with obesity and its reliability is reduced in patients with coexisting mitral valve disease.31 Serial echocardiographic assessments to monitor filling pressures or RV function are also impractical, as they require specialised expertise and they are time-consuming.
Lung Ultrasound
Lung ultrasound (LUS) is a non-invasive tool for assessing pulmonary congestion that requires minimal operator training. Several LUS scanning protocols have been proposed and the cumulative B-line count from 4–28 lung zones is commonly used to derive a congestion score for both diagnostic and prognostic assessment in ADHF.
In an observational study, Mazzola et al. demonstrated anterolateral B-lines at discharge were independent predictors of rehospitalisation for ADHF.32 The anterolateral scanning approach was also less affected by confounding from concomitant pneumonia compared to posterior scanning. Similarly, a pooled international analysis across diverse clinical settings has demonstrated that LUS provides incremental prognostic information beyond established risk scores such as MAGGIC and AHEAD.33 Moreover, LUS has demonstrated prognostic relevance across the spectrum of ejection fractions. In a prospective study of 1,021 patients with HF with reduced ejection fraction and HF with preserved ejection fraction (HFpEF), LUS was identified as an independent predictor of the composite outcome of death and HF rehospitalisation, with particularly strong significance in HFpEF.34
Studies also suggest that serial LUS can guide diuretic therapy, potentially shortening hospital stays and accelerating symptom relief.35,36 However, these benefits have not yet translated into improvements in post-discharge outcomes.
Inferior Vena Cava
The inferior vena cava (IVC) is a compliant vessel which is in direct continuation with the right atrium. Consequently, variations in RAP, as seen in ADHF, are transmitted back to the IVC. However, studies comparing IVC indices with invasively measured RAP using cardiac catheterisation have demonstrated only modest correlations.37 This is further limited in mechanically ventilated patients as artificial increases in intrathoracic pressure alter IVC dynamics.38
IVC distension carries prognostic significance in AHF, as persistent dilatation is associated with adverse outcomes.39 Moreover, IVC assessment can help detect subclinical congestion in patients with chronic HF. In a prospective study of 310 patients undergoing multimodal ultrasonographic evaluation of congestion, including LUS, IVC and renal Doppler flow, 42% of clinically euvolemic patients exhibited at least one ultrasonographic marker of congestion (p<0.0001).40
However, IVC ultrasound has limitations. Visualisation is challenging in patients with obesity, and IVC collapsibility may be reduced in patients with moderate to severe tricuspid regurgitation, resulting in an overestimation of the intravascular volume status.41 Although IVC size and collapsibility have been shown to correlate with diuretic response, evidence from RCTs on robust outcome data remains limited.42,43 The CAVA-ADHF-DZHK10 trial demonstrated that decongestion guided by IVC parameters did not result in greater reductions in NT-proBNP levels from baseline to discharge compared with standard care.44
Internal Jugular Vein
Clinical assessment of jugular venous distensibility (JVD) has limited sensitivity and specificity for detecting elevated CVP. This limitation can be mitigated by direct ultrasonographic visualisation of the internal jugular vein (IJV), which provides a more objective and reproducible estimate of right-sided filling pressures. A distended IJV on ultrasound has been shown to correlate with adverse outcomes in patients with HF.45
Under physiological conditions, the IJV has a resting diameter of about 0.10–0.15 cm which increases by up to 1 cm during the Valsalva manoeuvre. In AHF, elevated CVP results in an increased resting IJV diameter, while the maximum vessel diameter achieved during Valsalva remains unchanged. Consequently, the JVD ratio is reduced and is typically <4.46 A lower JVD ratio has been associated with more severe congestion, higher NT-proBNP levels and an increased risk of HF-related hospitalisation and mortality.47
Despite its potential usefulness, accurate IJV assessment requires appropriate technical skill as excessive transducer pressure can cause partial vein collapse. Furthermore, performing the Valsalva manoeuvre may not be feasible in acutely decompensated or dyspnoeic patients. Alternative dynamic assessments, such as the passive leg raise test, may be considered, although further studies are needed to standardise and validate these approaches.
Venous Excess Ultrasound Score
The Venous Excess Ultrasound (VExUS) score (Figure 3) is a tool used to assess systemic venous congestion in patients with ADHF, ranging from 0 (no congestion) to 3 (severe congestion), combining IVC diameter with Doppler flow patterns in the hepatic, portal and renal veins.
The VExUS score has demonstrated added value compared to IVC diameter alone. A study investigating the correlation between VExUS and right atrial pressure found a strong correlation, with VExUS scoring 99% for predicting RAP ≥12 mm Hg, outperforming IVC diameter, which had a 79% predictive value.48 Further, studies have shown that a high VExUS score at admission is a valuable prognostic indicator for patients admitted with ADHF, correlating with an increased risk of in-hospital death and HF rehospitalisation.49,50 An RCT of 140 patients with ADHF and AKI found that VExUS-guided management achieved decongestion twice as quickly as standard care (OR 2.6; 95% CI [1.9–3.0]; p=0.01).51 However, while promising, further research is needed to standardise and validate the VExUS protocol, which currently requires technical expertise and time, potentially limiting its clinical use.
Invasive Methods
Implantable Cardiac Devices
Implantable cardiac devices (ICDs), which provide continuous haemodynamic monitoring, have emerged as potential tools for HF management. These include multiparametric monitoring via cardiac implanted electronic devices, pulmonary artery pressure (PAP) monitoring devices, and newer IVC and left atrial pressure monitoring devices.
The SENSE-HF study was a prospective, multicentre, double-blind study that evaluated an impedance-based algorithm, OptiVol (Medtronic, Inc), in 501 NYHA class II and class III HF patients implanted with cardiac resynchronisation therapy with defibrillator (CRT-D) devices. The trial demonstrated a low sensitivity of 42% and low positive predictive value of only 38% for future HF events 6–24 months after implantation.52
Contemporary algorithms such as HeartLogic (Boston Scientific) and TriageHF (Medtronic), embedded within implanted cardiac devices, which integrate multiphysiological parameters including thoracic impedance, heart sounds, respiratory rate and heart rate, to generate early alerts for congestion and clinical deterioration, have recommendations for use in some health economies.53 However, these algorithms are not validated for use in an ADHF population.
Evidence regarding the benefit of PAP monitoring in HF remains heterogeneous. A meta-analysis of four major RCTs, COMPASS-HF, REDUCEhf, CHAMPION and GUIDE-HF, evaluating implantable PAP sensors (CardioMEMS, Chronicle, and Chronicle-ICD) in patients with prior HF hospitalisations, demonstrated a 25% reduction in total HF hospitalisations (2,224 patients; HR 0.75; 95% CI [0.58–0.96]; p=0.03), without a significant effect on all-cause mortality (RR 0.92; 95% CI [0.68–1.26]; p=0.48).54 GUIDE-HF showed a trend toward benefit before the COVID-19 pandemic; however, post-pandemic, PAP improved similarly in both groups, likely due to enhanced medical adherence, resulting in a neutral overall outcome. Notably, the absolute reduction in PAP across trials was modest, with only CHAMPION and GUIDE-HF employing structured algorithms for therapy titration based on target haemodynamic thresholds.54
The more recent MONITOR-HF trial (n=348) reported a greater mean PAP reduction (−8.4 mmHg) in the intervention group, accompanied by significant improvements in Kansas City Cardiomyopathy Questionnaire (KCCQ-12) scores and reduced HF hospitalisations, though without mortality benefit.55 A pooled meta-analysis combining CHAMPION, GUIDE-HF and MONITOR-HF reinforced these findings demonstrating consistent reductions in HF admissions but no effect on all-cause mortality.56
A novel invasive monitoring device is the FIRE-1 system, an implantable sensor positioned in the IVC that continuously monitors changes in vessel size as a surrogate for intravascular volume. First-in-human studies showed that the device could be safely implanted, provided reliable measurements correlating with imaging-based assessments and achieved good patient adherence. While primarily designed to detect early volume changes before pressure elevations occur, clinical outcome data are still limited, though it represents a promising tool for remote congestion monitoring.57
Adoption of these devices has been limited due to their high cost and the need for infrastructure to manage daily measurements. Further, these studies have evaluated patients in an ambulatory setting and not all patients who are admitted with ADHF have implanted cardiac devices. Therefore, the role of these devices in guiding decongestion in ADHF requires further investigation to establish their clinical usefulness.
Right Heart Catheterisation
Right heart catheterisation (RHC) provides invasive haemodynamic assessment in HF patients. There is a Class IC guideline recommendation for its use in patients with advanced HF who are being considered for mechanical support or transplant.3 A meta-analysis of six observational studies by Yoo et al. found that RHC-guided management in cardiogenic shock was associated with significantly lower mortality, likely due to earlier detection of haemodynamic decline.58 The ESCAPE trial examined a strategy of RHC-guided decongestion in 433 patients with ADHF. Although RHC-guided decongestion showed a trend towards greater functional improvement, there was no significant difference in the primary outcome of days alive out of hospital at 6 months, compared to the standard care arm.59
Newer Methods for Assessing Congestion
Artificial Intelligence and Machine Learning Models
Artificial intelligence (AI) and machine learning (ML) are increasingly being applied to enhance the accuracy, usability and integration of the existing methods to assess congestion described above.
ML methods have been used to increase the speed and accuracy of LUS and echocardiography. A secondary analysis of the BLUSHED-AHF trial incorporated an AI-based lung congestion score (LCS) derived from LUS video clips.60 Among 3,858 clips from 130 patients, the AI-LCS demonstrated high agreement with expert quantification of B lines, significantly outperforming operator scores. This suggests AI tools could enable broader and more accurate use of LUS by non-specialists.
Similarly, several AI-assisted echocardiographic platforms have been developed to increase the ease, speed and reproducibility of HF diagnosis. A recent systematic review of existing technologies in characterising HF through image analysis reported accuracy ranging from 57% to 99.3% compared to conventional methods.61 Most focus on automated detection of left ventricular ejection fraction, but their expansion into more advanced parameters, such as diastolic function, filling pressures and IVC dynamics, could further support their role in congestion assessment as well as being applicable to an ADHF population.
Remote Dielectric Sensing
Remote dielectric sensing (ReDS) is a non-invasive electromagnetic technology designed to quantify pulmonary congestion by measuring the dielectric properties of lung tissue. The device provides a congestion reading within minutes of application.
A meta-analysis by Sattar et al., which included seven prospective observational cohort studies, demonstrated that ReDS-guided management in patients with acute and chronic HF significantly reduced HF readmissions compared to standard care (OR 0.40; p<0.0001).62 Additionally, the ReDS-SAFE HF trial, a single-blinded proof-of-concept RCT, randomised patients hospitalised with ADHF to ReDS-guided decongestion or standard care.63 The primary composite outcome (unplanned HF visit, HF hospitalisation or all-cause mortality within 30 days of discharge) occurred in only 2% of patients in the ReDS-guided group versus 20% in the standard care group (p=0.005).
Despite promising results, ReDS adoption is limited because it is expensive and there are limited data linking its measurements to echocardiographic filling pressures. Further studies are needed to validate its role in guiding therapy and assess cost-effectiveness.
Bioimpedance Analysis
Bioimpedance analysis (BIA) is a quick, non-invasive and user-friendly tool that has been widely studied in CKD patients who are undergoing dialysis. More recently, its role in HF management has been explored. It provides rapid and reproducible estimates of total body water, intracellular water, extracellular water and derived parameters such as dry weight. These measurements enable clinicians to assess both intravascular and interstitial congestion.
Principles and Practicalities of Bioimpedance Analysis
BIA works by applying an imperceptible, high-frequency, low-amplitude alternating electrical current through the body via surface electrodes, typically placed on the chest, hands or feet. The resistance and reactance encountered by this current are influenced by the volume and distribution of fluid compartments in the body. These electrical properties are then used to generate estimates of body water content and fluid distribution.
Bioimpedance technology has been applied in a variety of devices, including standalone bioimpedance analysers, impedance cardiography (ICG) devices, and wearable monitoring systems.
Impedance Cardiography
Impedance cardiography uses bioimpedance technology for continuous haemodynamic measurements such as stroke volume, cardiac output, systolic time intervals and thoracic fluid content (TFC). A systematic meta-analysis found significant but modest correlations between NTproBNP and ICG-derived TFC (r 0.33; 95% CI [0.18–0.48]; p<0.001) and inverse relationships with stroke volume index (r −0.37; 95% CI [−0.66 to −0.08]; p=0.011), suggesting potential usefulness in assessing congestion and haemodynamic status.64
Conversely, in the BIG substudy of the ESCAPE trial, among 170 patients who underwent impedance cardiography, 82 also underwent cardiac catheterisation. ICG showed only modest correlation with invasively measured cardiac output (r = 0.4–0.6) and poor agreement with invasively derived haemodynamic profiles (κ≤0.1).65 Furthermore, no single ICG variable was independently prognostic. These findings suggest that while ICG may help track trends during management of HF, the accuracy and prognostic value of its absolute values remain limited.
Bioimpedance Analyser
Bioimpedance analysers are widely used for the rapid assessment of body composition and fluid status. Portable devices typically use electrodes attached to the hands and feet, similar to an ECG (Figure 4), to measure whole-body impedance. Bioimpedance scales, in contrast, allow individuals to stand directly on the device, providing both weight and detailed body composition metrics. While bioimpedance scales are commonly used in fitness settings to monitor weight and body composition trends, clinical-grade bioimpedance analysers have been employed to assess fluid status in conditions such as CKD and HF. A systematic review of 20 HF studies, including 6 reporting on CV outcomes, linked fluid overload markers, such as bioimpedance vector analysis, hydration index, and extracellular water ratio to increased cardiovascular risk.66
Beyond diagnosis and prognosis, BIA may help guide decongestion. Parameters such as phase angle, oedema index and hydration index have shown correlations with clinical improvement and weight loss.67 However, evidence on its clinical impact remains limited. A small RCT by Venegas-Rodriguez et al., involving 56 obese HFpEF patients found BIA-guided therapy (using estimated dry weight) was associated with a lower incidence of AKI, although the difference was not statistically significant, and cardiovascular outcomes were similar between groups.68 The ongoing BIO-HF trial (NCT07173426) is expected to provide high-quality evidence regarding the usefulness of BIA in guiding fluid management in ADHF and may help define its role in routine practice.69
While BIA is inexpensive, portable, rapid, and reproducible, it is not without limitations. Factors such as electrode placement, and hydration state may influence accuracy. Clinical adoption has also been slow due to limited awareness and variability among available devices.
Bioimpedance Wearable Technologies
Wearable bioimpedance devices extend the principles of BIA to continuous, non-invasive monitoring in ambulatory settings using a single lead. They enable real-time assessment of thoracic fluid and haemodynamic changes, offering potential for early detection of decompensation in HF. However, despite being portable and low-cost, their accuracy is limited by motion artefacts, electrode placement variability, and body composition.70 Current data remain largely exploratory and large-scale validation is required before these devices can be routinely integrated into clinical management.
Conclusion
No single method to assess congestion in ADHF offers a comprehensive, foolproof solution. Newer tools, such as BIA, are emerging as promising alternatives, capable of evaluating both intravascular and extravascular congestion in a non-invasive, easy-to-use and rapid manner, particularly in settings where more complex assessments are impractical. Overall, more research in this area would have a direct positive effect on patient outcomes and the considerable burden on healthcare cost and usage.