Assessment of fluid status remains one of the main challenges in patients undergoing hemodialysis (HD), as both volume overload and excessive ultrafiltration are associated with increased morbidity and mortality.1,2 However, the assessment of fluid status is complex, and currently there is no single reference method that allows accurate evaluation in routine clinical practice.1–3

It is important to recognize that the different tools used to assess fluid status explore distinct pathophysiological domains.2–5 Bioelectrical impedance spectroscopy (BIS) provides an estimation of extracellular volume, whereas point-of-care ultrasound (PoCUS) primarily identifies ultrasound findings associated with elevated filling pressures and venous congestion. Clinical evaluation, including the concept of dry weight, represents a dynamic construct that is symptom-dependent and has limited sensitivity for detecting early abnormalities.2–8

As a result, discrepancies between these methods are frequently observed in clinical practice.2,4 Rather than reflecting measurement error, such discordance may represent the complex and multidimensional relationship between volume overload and congestion in HD patients, particularly in the presence of cardiovascular comorbidity, diastolic dysfunction, or altered ventricular compliance.2,5

In this context, the concept of congestion has gained increasing relevance, not only as a manifestation of extracellular volume overload but also as a consequence of elevated cardiac filling pressures and systemic venous congestion. These hemodynamic alterations may impair organ perfusion and contribute to adverse cardiorenal interactions, even in the absence of overt clinical symptoms.6–8 In HD patients, cardiovascular remodeling, diastolic dysfunction, and altered ventricular compliance may further dissociate extracellular volume expansion from clinically apparent congestion.2,5 Consequently, PoCUS may provide clinically relevant information beyond simple volume estimation by identifying ultrasound findings associated with elevated filling pressures and systemic venous abnormalities.6–8 Identifying these patterns may have important implications for the individualization of ultrafiltration strategies in HD patients. However, the relationship between volume-derived parameters and indicators of hemodynamic congestion remains incompletely understood in this population.

Despite the growing interest in multiparametric approaches,2,4 few studies have comprehensively evaluated the relationship and potential complementarity between the different tools available for assessing fluid status and congestion in this setting.

Therefore, the primary objective of this pilot study was to explore the concordance and discordance between clinical assessment, BIS-derived extracellular volume estimation, and PoCUS-derived hemodynamic congestion parameters in HD patients, recognizing that these tools assess distinct but potentially complementary pathophysiological domains. Secondary exploratory objectives included the evaluation of associations between biomarkers and parameters of volume overload and hemodynamic congestion.

A total of 15 patients undergoing chronic HD were included, with a mean age of 71.8±17.6 years and a predominance of males (73.3%). The cohort had a high comorbidity burden, including hypertension (80%), cardiovascular disease (46.7%), and type 2 diabetes mellitus (33.3%).

More than half of the patients had a central venous catheter as vascular access, and the mean dialysis vintage was 33.5±27.1 months. All patients were treated with online hemodiafiltration (HDF-OL), with adequate hemodynamic tolerance.

Most patients were anuric; only five (33.3%) had residual urine output greater than 500mL/day.

Baseline clinical and laboratory characteristics are summarized in Table 1.

Exploratory concordance and discordance between methods

Agreement was moderate between clinical assessment and PoCUS (κ=0.59), whereas it was low between clinical assessment and BIS (κ=0.19) and very low between BIS and PoCUS (κ=0.06) (Table 2).

In the ordinal analysis, BIS classified a greater proportion of patients into moderate and severe categories compared with clinical assessment and PoCUS.

In contrast, PoCUS identified a smaller subset of patients with ultrasound findings suggestive of hemodynamic congestion, with a distribution more closely aligned with clinical assessment.

Detailed distributions across methods are shown in Fig. 1.

Fig. 1.

Distribution of congestion and volume overload across assessment methods. Simplified binary matrices showing the distribution of patients classified according to clinical assessment, BIS-derived volume overload, and PoCUS-defined congestion. Rows correspond to the first method compared and columns to the second.

BIS: bioimpedance spectroscopy; PoCUS: point-of-care ultrasonography.

The main finding of this exploratory pilot study is the marked heterogeneity observed among the different tools used to assess extracellular volume overload and hemodynamic congestion in patients undergoing HD. These findings suggest that the different tools capture complementary aspects of fluid status and congestion, rather than simply representing methodological differences between techniques. Therefore, the low agreement observed between methods should not necessarily be interpreted as poor diagnostic performance, but rather as evidence that these techniques evaluate different but potentially complementary pathophysiological constructs.

BIS estimates extracellular volume, whereas PoCUS identifies ultrasound findings associated with elevated filling pressures and venous congestion. Clinical evaluation integrates later and symptom-dependent manifestations, which may explain its closer alignment with ultrasound findings in this cohort.2,4,5

This concept is particularly relevant in HD patients, in whom extracellular volume expansion and hemodynamic congestion may become dissociated. The relationship between volume and congestion is therefore not necessarily linear.6–8 The presence of diastolic dysfunction, which is common in this population, may lead to elevated filling pressures in the absence of overt volume overload, resulting in congestion despite relatively normal volume status. Conversely, increases in extracellular volume may occur without clear clinical or ultrasound manifestations, which may explain the high proportion of patients classified as volume overloaded by BIS compared with PoCUS.

In this context, our findings are consistent with previous studies2,4 showing limited agreement between volume-based techniques and ultrasound tools for assessing congestion, reinforcing the need for a multiparametric approach to fluid status evaluation in HD. PoCUS identified a distinct subset of patients with integrated ultrasound findings suggestive of hemodynamic and systemic venous congestion, which did not necessarily overlap with extracellular volume expansion detected by BIS.10,11

Regarding biomarkers, NT-proBNP levels were associated with both clinical congestion and BIS-estimated volume overload, suggesting that this marker primarily reflects increased filling pressures and myocardial stress3,12 rather than volume per se. In contrast, CA-125 showed an increasing trend with greater volume overload, consistent with its role as a marker of chronic tissue congestion,13,14 although its utility in detecting acute changes in HD patients appears to be limited.

It should be noted that the low proportion of patients with ultrasound-detected congestion in our cohort, all of mild severity, may have influenced the magnitude of the observed discordance and limits the ability to establish more robust comparisons between methods.

From a clinical perspective, these findings have relevant implications. The use of tools that exclusively assess volume may lead to an overestimation of fluid overload, whereas the integration of techniques targeting hemodynamic congestion, such as PoCUS, may allow a more individualized approach to ultrafiltration adjustment,1,15 particularly in patients with cardiovascular comorbidity. In this sense, combining parameters of volume and congestion may contribute to more precise decision-making, enabling more conservative or progressive ultrafiltration strategies according to the individual patient profile.

This study has several limitations. First, the absence of echocardiographic data precludes a comprehensive characterization of cardiac function and limits the ability to fully explore congestion phenotypes. However, this study was designed as a pilot study focusing on readily available bedside tools in routine HD practice, where access to echocardiography is often limited. Future studies incorporating larger cohorts, comprehensive cardiac phenotyping, and standardized congestion assessment are required to validate these exploratory findings and better define congestion phenotypes and their clinical implications.

Second, the small sample size and exploratory, hypothesis-generating pilot design limit the generalizability of the findings and preclude definitive conclusions regarding congestion phenotypes or method performance. Third, the lack of standardization using validated clinical congestion scores and the absence of blinding between evaluators may have introduced bias. Finally, no adjustment was made for potential confounders, such as cardiac function, dialysis vintage or the presence of atrial fibrillation, which may influence the interpretation of biomarkers and ultrasound findings. In addition, no formal intraobserver or interobserver reproducibility assessment was performed for ultrasound parameters due to the pilot nature of the study.

Overall, our findings suggest that fluid status assessment in HD patients requires the integration of tools exploring different and potentially complementary pathophysiological aspects.

Different bedside tools identified divergent patterns of extracellular volume overload and hemodynamic congestion in HD patients. BIS, PoCUS and clinical assessment should not be interpreted as equivalent or interchangeable tools, but rather as complementary methods providing distinct physiological and hemodynamic information relevant to fluid status evaluation. Biomarkers provided additional information regarding different volume overload and congestion profiles, although these findings should be considered exploratory.

These findings support the use of a multiparametric approach to fluid status assessment and treatment individualization. The observed discordance between methods likely reflects the multidimensional nature of congestion in HD patients, including the dissociation between extracellular volume expansion and hemodynamic/systemic venous congestion, rather than simple methodological disagreement. Given the pilot and hypothesis-generating nature of this study, larger investigations incorporating comprehensive cardiac phenotyping and standardized congestion assessment are needed to confirm these findings.

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

The authors declare no conflicts of interest related to this work.