Abstract
Introduction
Fluid overload is a major determinant of morbidity and mortality in maintenance hemodialysis (MHD) patients. This exploratory study describes the integration of lung ultrasound (LUS), modified Venous Excess Ultrasound Score (mVExUS), and bioimpedance analysis (BIA) for non-invasive fluid status assessment in MHD. Additionally, this study aimed to correlate congestion with malnutrition using echographic parameters (nutritional ultrasound [NUS]) in congestive and non-congestive patients.
Methods
In this single-center retrospective observational cohort study, 47 adult patients on MHD underwent pre-dialysis evaluation with LUS, mVExUS, NUS, and BIA. Patients were classified as congestive if they had an mVExUS score ≥2, LUS with ≥3 B-lines in ≥3 thoracic zones and BIA with the ratio of extracellular water to total body water ECW/TBW ≥0.39. NUS was used to assess the quadriceps rectus femoris (QRF) and preperitoneal visceral fat (PPVF), measuring Y-axis, Y-axis/height, cross-sectional muscle area rectus femoris (CS-MARF), and supramuscular fat (SMF). Demographic, biochemical, functional, and dialysis-related parameters were collected. Frailty, sarcopenia and nutritional status were evaluated. Congestive patients were reassessed after 5 weeks.
Results
Eight patients (17%) met criteria for congestion. As expected by the predefined congestion criteria, patients classified as congestive showed higher mVExUS grades (p < 0.001), greater B-line burden (11.9 vs. 2.9), and higher ECW/TBW ratios (0.42 vs. 0.40; p = 0.004). After 5 weeks, congestive patients exhibited improvements in N-terminal pro-B-type natriuretic peptide (NT-proBNP), portal vein pulsatility index, and pulmonary congestion, without adverse hemodynamic events. The CS-MARF was 1.81 ± 0.11 cm2 in congestive versus 2.91 ± 0.78 cm2 in non-congestive (p = 0.0004).
Conclusion
The integration of LUS, mVExUS, BIA and NUS provide a descriptive framework for multimodal assessment of fluid and nutritional status in MHD patients. These findings should be considered hypothesis-generating and require validation in prospective studies.
Keywords: Fluid overload, Lung ultrasound, Venous excess ultrasound score, Nutritional ultrasound, Maintenance hemodialysis
Introduction
Fluid overload is a major contributor to morbidity and mortality among patients with advanced chronic kidney disease (CKD) receiving renal replacement therapy (RRT), particularly those undergoing maintenance hemodialysis (MHD) [1–3]. The concept of “dry weight” – defined as the lowest weight a patient can tolerate after optimal ultrafiltration without symptoms of hypovolemia – remains a cornerstone of nephrological management. However, its determination remains a clinical challenge through physical examination with poor reproducibility [4, 5]. Adjunctive tools such as bioelectrical impedance analysis (BIA) have been introduced, and some studies have linked its use to improved cardiovascular risk stratification and survival [6]. Nonetheless, its diagnostic accuracy is hampered by serial confounding factors [7, 8]. Consequently, its use as a standalone tool for volume management in HD patients has been questioned, prompting calls for more integrative and dynamic strategies that incorporate complementary techniques [9].
In this context, point-of-care ultrasound (POCUS) has emerged as a valuable diagnostic modality due to its non-invasiveness, reproducibility, real-time applicability, and accessibility. Quantitative lung ultrasound (LUS) is a tool for assessing the extent of extravascular lung water that outperforms physical exam and plain chest radiography. In fact, B-lines have been shown to decrease dynamically during the hemodialysis treatment in proportion to ultrafiltration volume. Lung congestion is associated with a high death risk in dialysis patients and, therefore, represents a potential treatment target [10–12].
Concurrently, the Venous Excess Ultrasound Score (VExUS) has been proposed as a structured tool to evaluate systemic venous congestion as it helps overcome the limitations of conventional physical examination in detecting hemodynamic congestion and monitoring treatment efficacy [13, 14]. However, a modified version of the Venous Excess Ultrasound Score (mVExUS) protocol is required for use in hemodialysis patients as the assessment of intrarenal veins is often difficult to obtain in individuals with end-stage renal disease [13]. Most existing studies have focused primarily on tissue congestion, without simultaneously assessing venous congestion, which may result in an incomplete characterization of the patient’s true hemodynamic status [15].
Recent studies have highlighted that the coexistence of malnutrition and overhydration in hemodialysis patients represents a major risk factor for increased morbidity and mortality [16]. Although bioimpedance analysis (BIA) is widely used to evaluate hydration and body composition, its accuracy may be influenced by technical and population-related factors. The combination of LUS, mVExUS, and NUS offers a comprehensive approach to fluid and nutritional assessment providing complementary information to the physical examination or BIA alone in MHD patients. Therefore, the objective of the present study was to conduct an integrated, multimodal assessment of congestion and nutritional status by combining findings from LUS, mVExUS, NUS, and BIA in a cohort of prevalent MHD patients. Additionally, this study aimed to correlate congestion with malnutrition in congestive and non-congestive patients.
Methods
A single-center, retrospective observational cohort study of 47 patients with prognostic analysis was conducted at the Hemodialysis Unit of the Nephrology Department at the Central Defense Hospital Gómez Ulla, between January 2024 and January 2025. Adult patients (≥18 years) those undergoing MHD 3 times per week, for at least 3 months were included if they had complete evaluations using LUS, mVExUS, BIA, and NUS. Exclusion criteria included recent hospitalization due to heart failure (≤3 weeks), acute intercurrent conditions that could alter volume status (e.g., active infections, febrile syndromes, or clinical decompensation), or incomplete data. Patients were followed retrospectively up to 5 weeks after the initial assessment.
Follow-up and decongestion management (5-week period). “After the baseline pre-dialysis integrated assessment (LUS, mVExUS, BIA, and NUS), congestive patients underwent a pragmatic dry-weight optimization strategy as part of routine clinical care. UF targets were individualized by the treating nephrology team based on interdialytic weight gain, symptoms, and objective congestion markers. Dry weight was adjusted gradually over consecutive sessions, prioritizing hemodynamic tolerance. Dialysis prescription was modified when needed to achieve UF goals safely (e.g., extending treatment time and/or adjusting UF rate). Hemodynamic monitoring consisted of routine pre-, intra-, and post-dialysis blood pressure measurements and systematic recording of intradialytic intolerance (symptomatic hypotension, cramps, nausea/vomiting, headache). UF goals were reduced or dry-weight targets re-evaluated in the presence of intolerance.”
Clinical and demographic data were collected, including age, sex, comorbidities, etiology of CKD, body mass index (BMI), type of vascular access, dialysis modality, and time on dialysis. The analytical panel included complete blood count (including hemoglobin [Hb]), iron studies (serum iron [Fe], ferritin, transferrin [TRF], and transferrin saturation index [TSI]), total proteins (TP), prealbumin (PreAlb), albumin (Alb), and C-reactive protein (CRP). Parameters of phospho-calcium metabolism measured were total calcium (Ca), phosphorus (P), parathyroid hormone (intact PTH), and vitamin D. Urea, serum creatinine (Cr) and creatinine indexed to body surface area (Cr/SC, mg/dL/m2); normalized protein catabolic rate (nPCR) was also calculated. A lipid profile (total cholesterol [TC], high-density lipoprotein [HDL], low-density lipoprotein [LDL], non–HDL cholesterol, and triglycerides [TG]) was obtained. Electrolytes and N-terminal pro–B-type natriuretic peptide (pro-BNP) were measured. In addition, variables and assessment tools related to frailty, sarcopenia (SARC-F scale, handgrip), and nutritional status (MIS scale) were also included. Frailty is a medical condition characterized by the loss of physiological reserve across multiple domains or an increased vulnerability to stress. Frailty is not isolated to a single organ system or specific medical disease but rather is considered to be a multisystem syndrome. Frailty reflects an individual’s physiologic, as opposed to chronologic, age; is common among people with chronic kidney disease; and is associated with poor health outcomes [17, 18]. In non-dialysis CKD the 5-item FRAIL scale (fatigue, resistance, ambulation, illnesses, and loss of weight) showed consistent discriminative validity for mortality [17] and is also a valid and reliable tool in HD patients and has been shown to be a better frailty metric when compared to other tools [19, 20].
The FRAIL scale includes five questions that were applied by trained evaluators: Do you feel fatigued? Can you climb a flight of stairs? Can you walk a block without stopping? Do you follow up or treat more than 5 diseases? In the last 6 months, have you unintentionally lost 5% or more of your body weight? Patients were approached during the dialysis session and when they did not present acceptable cognitive conditions, their caregiver was consulted. The score ranges from 0 to 5 points, and each component of the assessment is worth one point. Individuals were classified as “non-frail” (score 0), “pre-frail” (score 1–2), and “frail” (score 3–5) [19].
Hemodynamic intolerance events during dialysis sessions (symptomatic hypotension, cramps, nausea, vomiting, headache) were also recorded. The following BIA parameters were recorded: total body water (TBW, L), extracellular water (ECW, L), intracellular water (ICW, L), ECW/TBW ratio, and Appendicular Skeletal Muscle Mass Index (ASMI, kg/m2).
Ultrasonographic parameters included (i) lung ultrasound (number of B-lines per field); (ii) inferior vena cava (IVC) expiratory diameters (cm); (iii) portal vein pulsatility index (PI), defined as (Vmax−Vmin)/VMax ×100(Vmax−V)/VMax ×100, and categorized as <30% (grade 0, normal), 30–49% (grade 1, mild), ≥50% (grade 2, severe); (iv) hepatic vein (HV) Doppler pattern, categorized as D>S (grade 0, normal), S≈D or S<D without systolic reversal (grade 1, mild), and systolic flow reversal (grade 2, severe); and (v) mVExUS score, computed from the IVC “gate” plus organ Doppler severity as follows: mVExUS 0: non-dilated IVC (physiological collapsibility), regardless of organ Doppler; mVExUS 1: dilated IVC (>2 cm) with no abnormal organ Doppler (HV grade 0 and PI grade 0); mVExUS 2: dilated IVC with severe congestion in one territory (HV grade 2 or PI grade 2); mVExUS 3: dilated IVC with severe congestion in ≥2 territories (HV grade 2 and PI grade 2). Left ventricular ejection fraction (LVEF, %) estimated by biplane Simpson was also recorded. In addition, NUS parameters were assessed, including the transverse rectus femoris muscle area (CS-MARF, cm2), the transverse MRF major (X) axis diameter (cm), the transverse MRF minor (Y) axis diameter (cm), the supra-muscular fat (SMF) thickness (mm), and the preperitoneal visceral fat (PPVF) thickness (cm).
BIA measurements were obtained using the InBody® S10 device. All ultrasound examinations were conducted by a nephrologist with formal training in clinical ultrasonography. The equipment used was Mindray Z50 with three probes: linear (75L38EA) for pulmonary ultrasound and nutritional ultrasound, convex (35C50EA) for modified VExUS score insonation, and micro-convex (65C15EA) for cardiac ultrasound. Both BIA and ultrasound assessments were performed prior to dialysis, always on the second session day of the weekly hemodialysis schedule, with the patient in the supine position. The inclusion of BIA was intended to quantify overall fluid overload rather than to compare assessment techniques. This approach aimed to describe a comprehensive congestive phenotype integrating intravascular, interstitial and extracellular components. Patients were then classified into two groups based on the presence or absence of significant echographic congestion, defined as the combination of an mVExUS score ≥2, lung ultrasound (LUS) evidence of ≥3 B-lines in more than three of the eight thoracic zones examined, and a BIA with ECW/TBW ≥0.39 [21]. Those fulfilling all criteria were defined as congestive, while the others were considered non-congestive. Given the absence of validated dialysis-specific NUS thresholds, muscle and fat ultrasound variables were analyzed as continuous measures; NUS was included as an exploratory morphometric tool and was not used to define sarcopenia or to drive standardized therapeutic decisions. Several variables known to influence volume and nutritional status were not systematically available in this retrospective cohort and were, therefore, not included in the analysis; among the variables analyzed, missing data were infrequent and occurred sporadically, and analyses were performed on a per-variable complete-case basis without imputation.
Statistical analysis included a descriptive evaluation of the entire cohort and comparisons between patients with and without significant congestion. Quantitative variables were expressed as mean ± standard deviation or as median and interquartile range, depending on distribution, while categorical variables were reported as absolute and relative frequencies. Group comparisons were performed using the Mann-Whitney U test for continuous variables and chi-square or Fisher’s exact test for categorical variables. Correlations between ultrasound parameters and BIA values were assessed using Spearman’s correlation coefficient. Analyses were performed using SPSS® version 25.0, with a p value <0.05 considered statistically significant.
Results
A total of 47 prevalent patients undergoing MHD were included in the study. Of these, 39 patients (82.9%) were classified as non-congestive and 8 patients (17.1%) as congestive. The mean age of the cohort was 71.9 years (SD 16.1), with a predominance of male patients (66%). Most individuals had hypertension (89.4%), type 2 diabetes mellitus (T2DM) (45.7%), and a mean Charlson comorbidity index of 7.2 (SD 2.8). A history of heart failure was present in 21.3% of patients, and 10.9% had a prior diagnosis of cancer. The mean BMI was 25.2 kg/m2 (SD 4.2). Hypertensive nephropathy was the most frequent cause (36.2%), followed by glomerular disease (14.9%). No significant differences were observed between the congestive and non-congestive groups in terms of CKD etiology, except for T2DM, which was more prevalent in congestive patients compared to non-congestive patients. The mean dialysis vintage was 41.3 months (SD 33.9), with 53.2% of patients dialyzing through a central venous catheter. The mean interdialytic weight gain was 2.3 kg (SD 0.8), with a mean ultrafiltration rate of 8.6 mL/kg/h (SD 3.7).
The congestive group exhibited greater comorbidity, with 100% having T2DM compared to 34.2% in the non-congestive group, and a higher Charlson index (8.5 vs. 7.0, respectively). Regarding laboratory parameters, no statistically significant differences were observed between the two groups, although the congestive group tended to exhibit higher values of pro-BNP, Cr/BSA and CRP. Of note, median NT-proBNP levels were higher in the congestive group (13.237 pg/mL [IQR: 6.193–21.482] vs. 3.568 pg/mL [IQR: 1.364–12.986]) (shown in Table 1).
Table 1.
Comparison of laboratory biomarkers between congestive and non-congestive patients
| Analytical variables | Total | Non-congestive | Congestive | p value |
|---|---|---|---|---|
| Patients, N | 47 | 39 | 8 | |
| Hb, mean (SD), g/dL | 11.2 (1.9) | 11.1 (1.7) | 11.6 (2.6) | 0.32 |
| Fe, mean (SD), mg/dL | 74.3 (32.6) | 71.9 (31.9) | 93.4 (35.5) | 0.19 |
| TSI, mean (SD), % | 38.5 (29.5) | 38.8 (31.0) | 37.3 (22.3) | 0.92 |
| TRF, mean (SD), mg/dL | 168.2 (29.8) | 168.4 (30.0) | 166.4 (31.5) | 0.82 |
| Ferritin, mean (SD), mg/dL | 735.7 (448.3) | 707.2 (454.7) | 874.5 (414.8) | 0.19 |
| Ca, mean (SD), mg/dL | 8.1 (0.5) | 8.2 (0.7) | 8.8 (0.6) | 0.65 |
| P, mean (SD), mg/dL | 4.2 (1.7) | 4.2 (1.7) | 4.2 (1.5) | 0.86 |
| Vit D, mean (SD), ng/mL | 29.1 (17.5) | 29.9 (18.8) | 24.7 (5.5) | 0.44 |
| PTH, mean (SD), pg/mL | 282.8 (254.1) | 295.3 (269.4) | 221.7 (157.9) | 0.81 |
| TC, mean (SD), mg/dL | 129.4 (36.3) | 130.7 (38.6) | 123.1 (22.3) | 0.92 |
| HDL, mean (SD), mg/dL | 48.3 (22.7) | 47.7 (23.3) | 50.8 (20.9) | 0.68 |
| LDL, mean (SD), mg/dL | 64.7 (29.9) | 66 (31.1) | 58.5 (24.1) | 0.62 |
| Non-HDL, mean (SD), mg/dL | 83.6 (31.6) | 85.8 (33.4) | 72.4 (18.4) | 0.32 |
| TG, mean (SD), mg/dL | 126.9 (77.4) | 130.2 (81.6) | 110.6 (53.4) | 0.53 |
| CRP, mean (SD) | 1.5 (2.8) | 1.6 (2.9) | 1.3 (2.2) | 0.06 |
| Lymphocytes, cell/µl 103, mean (SD) | 1 (0.5) | 1 (0.5) | 1 (0.5) | 0.42 |
| Na, mean (SD), mEq/L | 137.6 (3.1) | 137.8 (3.2) | 136.9 (2.6) | 0.42 |
| K, mean (SD), mEq/L | 4.5 (0.9) | 4.5 (1.0) | 4.7 (0.6) | 0.61 |
| Cl, mean (SD), mEq/L | 101.2 (3.4) | 101.5 (3.2) | 99.6 (3.8) | 0.2 |
| Urea, mean (SD), mg/dL | 108.6 (51.3) | 106.4 (51.5) | 119.2 (52.1) | 0.41 |
| Cr, mean (SD), mg/dL | 13.1 (27.7) | 12.6 (28.3) | 15.5 (26.6) | 0.92 |
| TP, mean (SD), g/dL | 6.5 (0.9) | 6.5 (1.0) | 6.3 (0.6) | 0.71 |
| PreAlb, mean (SD), g/dL | 25 (7.6) | 25.9 (7.5) | 20.8 (7.1) | 0.09 |
| Alb, mean (SD), g/dL | 4.3 (5.2) | 4.5 (5.7) | 2.9 (0.5) | 0.12 |
| Cr/SC, mean (SD), mg/dL/m2 | 2.8 (2.5) | 2.6 (2.6) | 4 (1.4) | <0.01 |
| nPCR, mean (SD), g/kg/day | 1.1 (0.4) | 1.1 (0.3) | 1.4 (0.4) | 0.12 |
| NT-proBNP, mean (SD), pg/mL | 3,568 [1,364–1,2986] | 2,971 [940–8,994] | 13,237 [6,193–21,482] | 0.02 |
Hb, hemoglobin; Fe, iron; TSI, transferrin saturation index; TRF, transferrin; Ca, calcium; P, phosphorus; Vit D, vitamin D; PTH, parathyroid hormone; TC, total cholesterol; HDL, high-density lipoprotein; LDL, low-density lipoprotein; Non-HDL, non-high-density lipoprotein cholesterol; TG, triglycerides; CRP, C-reactive protein; Na, sodium; K, potassium; Cl, chloride; urea, blood urea nitrogen; Cr, creatinine; TP, total proteins; PreAlb, prealbumin; Alb, albumin; Cr/SC, creatinine to body surface area ratio (mg/dL/m2); nPCR, normalized protein catabolic rate (g/kg/day); NT-ProBNP, N-terminal pro-B-type natriuretic peptide.
When analyzing the individual components of the mVExUS score and LUS, as determined by the predefined classification criteria, the congestive group presented higher mVExUS grades, larger IVC diameters, and more abnormal venous Doppler profiles, reflecting internal consistency of the multimodal congestion construct (Table 2). LUS findings were consistent with the classification criteria, with a higher B-line burden observed in patients categorized as congestive (shown in online suppl. Fig. S1; for all online suppl. material, see https://doi.org/10.1159/000552121), consistent with pulmonary interstitial congestion. Conversely, left ventricular ejection fraction was lower in the congestive group (42.4% vs. 56.2%; p < 0.001), indicating an associated impairment in systolic function. Table 2 depicts a coherent hemodynamic picture in which systemic venous congestion (mVExUS), pulmonary congestion (LUS), and cardiac function (LVEF) point in the same direction for the congestive phenotype. These NUS findings are reported descriptively and were not used to define malnutrition/sarcopenia categories or to guide clinical decisions.
Table 2.
Integrated ultrasound evaluation: echocardiographic, lung, mVExUS, and NUS (rectus femoris and abdominal fat) parameters by congestion status
| Ultrasound parameters | Non-congestive | Congestive | p value |
|---|---|---|---|
| Patients, N | 39 | 8 | |
| mVExUS score, n (%) | |||
| Score 0 | 39 (100) | 0 (0) | <0.01 |
| Score 1 | 0 (0) | 0 (0) | |
| Score 2 | 0 (0) | 7 (87.5) | <0.01 |
| Score 3 | 0 (0) | 1 (12.5) | <0.01 |
| IVC expiration diameter, mean (SD), cm | 1.3 (0.4) | 2.1 (0.1) | <0.01 |
| HV Doppler pattern, n (%) | |||
| D > S | 39 (100) | 0 (0) | <0.01 |
| S = D | 0 (0) | 7 (87.5) | <0.01 |
| S > D | 0 (0) | 1 (12.5) | 0.08 |
| PI, n (%) | 26.5 (15.5) | 46.5 (8.5) | <0.01 |
| PI (<30%) | 39 (100) | 0 (0) | |
| PI (30–49%) | 0 (0) | 7 (87.5) | |
| PI (≥50%) | 0 (0) | 1 (12.5) | |
| B-lines per lung field, mean (SD) | 2.9 (1.8) | 11.9 (2.8) | |
| Left ventricular ejection fraction (LVEF), n (%) | 56.2 (9.0) | 42.4 (5.0) | <0.01 |
| Nutritional ultrasound | |||
| CS-MARF, mean (SD) cm2 | 2.91 (0.7) | 1.81 (0.1) | <0.01 |
| X-axis, mean (SD), cm | 3.30 (0.5) | 2.95 (0.4) | 0.01 |
| Y-axis, mean (SD), cm | 0.90 (0.3) | 0.56 (0.1) | <0.01 |
| Transverse SMF, mean (SD), mm | 6.90 (1.9) | 6.00 (2.3) | 0.53 |
| Transverse PPVF, mean (SD), cm | 0.66 (0.3) | 0.75 (0.2) | 0.36 |
No p values are shown for B-lines, IVC diameter and portal PI as these variables were used to define the congestion phenotype. Continuous variables are presented as mean ± standard deviation or median (IQR), and categorical variables are expressed as percentages. Missing data were excluded per variable. p values come from two-sided Mann-Whitney U tests (non-parametric). No correction for multiple comparisons was applied.
mVExUS, Modified Venous Excess Ultrasound Score; IVC, inferior vena cava; HV, hepatic vein; D, diastole; S, systole; PI, Portal Vein Pulsatility Index; SD, standard deviation; CS-MARF, cross-section muscle area of the rectus femoris; X, major axis; Y, minor axis; SMF, Supramuscular fat; PPVF, preperitoneal visceral fat; IQR. interquartile range.
Regarding frailty and sarcopenia status, there were no significant differences between groups. The prevalence of FRAIL ≥3 was 50.0% in congestive patients (4/8) versus 46.2% in non-congestive (18/39; p = 1.0), and SARC-F ≥4 was 37.5% (3/8) versus 38.5% (15/39; p = 1.0). Likewise, the proportion meeting EWGSOP2 sarcopenia criteria did not differ significantly (50.0% [4/8] vs. 35.9% [14/39]; p = 0.6918). A greater share of congestive patients had low appendicular IMME (62.5% [5/8] vs. 38.5% [15/39]), but this did not reach statistical significance (p = 0.2581). Handgrip-defined dynapenia (75.0% vs. 58.9%; p = 0.356). Fisher’s exact test was used due to small cell counts (see Table 3).
Table 3.
Frailty, sarcopenia, and nutritional risk by congestion status
| Variable | Non-congestive | Congestive | p value |
|---|---|---|---|
| Patients, N | 39 | 8 | |
| FRAIL scale (score ≥3 points), n (%) | 18 (46.2) | 4 (50) | 1.00 |
| SARC-F Scale ≥4, n (%) | 15 (38.5) | 3 (37.5) | 1.00 |
| Sarcopenia (2019-EWGSOP2), n (%) | 14 (35.9) | 4 (50) | 0.70 |
| Low appendicular IMME (<5.5 kg/m2 in women; <7 kg/m2 in men, measured by BIA), n (%) | 15 (38.5) | 5 (62.5) | 0.26 |
| Dynapenia (<16 kg in women; <27 kg in men, measured by HGS), n (%) | 23 (58.9) | 6 (75) | 0.36 |
| PEW ≤1 (severe), n (%) | 16 (41) | 4 (50) | 0.71 |
| PEW >1, n (%) | 23 (59) | 4 (50) | |
| MIS score 2 (3–5), n (%) | 11 (28.2) | 3 (37.5) | 1.0 |
| MIS score 3 (6–8), n (%) | 20 (51.3) | 3 (37.5) | 0.70 |
| MIS score 4 (>8), n (%) | 8 (20.5) | 2 (25) | 1.00 |
| SGA, n (%) | |||
| SGA A | 19 (48.7) | 2 (20) | 0.12 |
| SGA B | 16 (41) | 3 (37.5) | 1.00 |
| SGA C | 4 (10.3) | 3 (37.5) | 0.06 |
p values are two-sided (Fisher’s exact test when any expected count <5; otherwise, Pearson’s χ2). For SGA A/B/C, the reported p value refers to the overall distribution across categories, not to a single row.
FRAIL Scale ≥3: criterion indicating frailty; SARC-F Scale ≥4: indicative of sarcopenia. Sarcopenia per EWGSOP2; low appendicular IMME: low appendicular skeletal muscle mass index (<5.5 kg/m2 in women; <7 kg/m2 in men, measured by BIA). HGS: handgrip strength. Dynapenia: reduced muscle strength (<16 kg in women; <27 kg in men, measured by handgrip). PEW Score (2014): Protein-Energy Wasting classification (0–1 = severe wasting; 2 = moderate wasting; 3 = mild wasting; 4 = normal nutritional status). MIS score: Malnutrition Inflammation Score (1 (normal) = <3; 2 (mild) = 3–5; 3 (moderate) = 6–8; 4 (severe) >8]. SGA (Subjective Global Assessment): A = well-nourished, B = moderate malnutrition or at-risk, C = severe malnutrition.
Nutritional risk scores showed broadly similar distributions by congestion status. Using dichotomized thresholds, PEW ≤1 (“severe”) did not differ between groups (50.0% (4/8) vs. 41.0% (16/39) (p = 0.7073). Considering full categories, MIS (1–4) distributions were comparable (overall p = 0.6896): MIS 2 (3–5) 37.5% vs. 28.2%, MIS 3 (6–8) 37.5% vs. 51.3%, and MIS 4 (>8) 25% vs. 20.5% (MIS 1: 0% in both). For SGA, there was a non-significant tendency toward worse status among congestive patients, with more SGA C (37.5% vs. 10.3%; p = 0.0576), while SGA B (37.5% vs. 41.0%; p = 1.0) and SGA A (20.0% vs. 48.7%; p = 0.1247) were similar (see Table 3).
In the morphometric NUS analysis of the rectus femoris (transverse view), congestive patients showed smaller muscle size. The CS-MARF was 1.81 ± 0.11 cm2 in congestive versus 2.91 ± 0.78 cm2 in non-congestive (p = 0.0004), with both the major (X) axis (2.95 ± 0.35 vs. 3.30 ± 0.47 cm; p = 0.0121) and minor (Y) axis (0.56 ± 0.07 vs. 0.90 ± 0.25 cm; p = 0.0001) being reduced. SMF thickness was similar (6.00 ± 2.33 vs. 6.90 ± 1.96 mm; p = 0.5326), while preperitoneal/visceral fat (transverse) tended to be higher in congestive patients without reaching significance (0.75 ± 0.19 vs. 0.66 ± 0.29; p = 0.3570) (Table 2).
Regarding body composition parameters, the data obtained between congestive and non-congestive patients, the ECW/TBW ratio was significantly higher in the congestive group (0.42 vs. 0.40; p = 0.004). Although absolute values of ECW, ICW, and TBW were greater in the congestive group (shown in online suppl. Fig. S2), these differences did not reach statistical significance compared to the non-congestive group and no relevant differences were found in other components including the phase angle, suggesting that the ultra-sound-detected congestion was not associated with greater cellular or inflammatory deterioration in this cohort.
During the 5-week follow-up, congestive patients showed improvements in several congestion-related parameters in parallel with routine dry-weight optimization, while intradialytic blood pressure remained stable and intolerance events were not increased. The mean number of B-lines per lung field on LUS significantly decreased, and similarly, the expiratory diameters of the IVC as well as the portal vein PI showed a slight downward trend (see online suppl. Table S1). Systolic and diastolic blood pressure values before and after dialysis remained stable, with no abrupt drops suggestive of hemodynamic intolerance during ultrafiltration, supporting the tolerability of the gradual dry weight adjustment.
Discussion
This study provides a descriptive integration of lung ultrasound (LUS), modified Venous Excess Ultrasound Score (mVExUS), nutritional ultrasound (NUS), and bioimpedance analysis (BIA) to characterize fluid and nutritional status in maintenance hemodialysis (MHD) patients. Rather than testing a predefined hypothesis, this approach explores the internal coherence of a multimodal congestion framework across vascular, interstitial, and body composition domains in a real-world clinical setting [22, 23].
An important methodological consideration is the presence of incorporation bias as the definition of congestion includes variables that are also described across groups. This limits internal validity and precludes independent inferential comparisons. Accordingly, these findings should be interpreted as descriptive assessments of internal coherence within a multimodal framework rather than as evidence of independent associations.
Within this context, the observed alignment between mVExUS, pulmonary findings (LUS), and extracellular volume markers (BIA) suggests internal consistency across different physiological compartments. This multimodal convergence may reflect the complex interplay between intravascular and interstitial fluid distribution in hemodialysis patients, rather than independent or causal relationships.
Lung ultrasound has been widely validated as a sensitive marker of pulmonary congestion, correlating with extravascular lung water, cardiac filling pressures, and adverse outcomes [24]. Previous studies have shown that LUS-guided strategies may reduce congestion and improve cardiac parameters, although their impact on hard clinical outcomes remains uncertain [25]. In our study, the observed reduction in B-line burden following dry-weight adjustment is consistent with these observations but should be interpreted as a dynamic description of congestion changes rather than as evidence of treatment effect.
The inclusion of VExUS extends congestion assessment to the systemic venous domain. Recent prospective data have demonstrated that VExUS scores decrease following hemodialysis, confirming their ability to capture dynamic changes in intravascular volume. Importantly, the lack of a linear relationship between ultrafiltration volume and VExUS variation suggests that venous congestion cannot be fully explained by net fluid removal alone but rather reflects complex hemodynamic interactions [26]. In line with these findings, the parallel evolution of mVExUS and LUS parameters in our cohort supports the concept that different ultrasound markers capture complementary dimensions of congestion.
Congestion in patients undergoing hemodialysis should, therefore, not be interpreted as a purely volumetric phenomenon but as a multidimensional cardiorenal interaction. In this context, natriuretic peptides such as NT-proBNP provide an integrative signal reflecting both fluid-related hemodynamic burden and underlying cardiac dysfunction. Elevated NT-proBNP levels have consistently been associated with increased all-cause and cardiovascular mortality in dialysis populations [27, 28], although their interpretation is influenced by cardiac structure, function, and reduced renal clearance. Together, these observations reinforce the concept that congestion cannot be adequately captured by a single parameter, supporting the use of multimodal assessment strategies.
The integration of nutritional ultrasound introduces an additional dimension by capturing body composition and potential interactions between fluid status, muscle mass, and inflammation. While emerging evidence suggests that fluid overload, malnutrition, and inflammation frequently coexist and may jointly influence outcomes in dialysis populations [29], the role of NUS in congestion phenotyping remains exploratory. In this study, NUS primarily provides objective morphometric characterization and highlights heterogeneity within conventional nutritional classifications, but its incremental clinical utility requires further validation.
These observations should also be interpreted in the context of known limitations of conventional volume assessment. The concept of dry weight remains inherently imprecise and dynamically defined [30] and single-parameter approaches – whether clinical, bioimpedance-based, or imaging-based – are insufficient to fully characterize congestion status [31]. The multimodal approach proposed here may help better describe this complexity, although its clinical applicability cannot be inferred from the present data.
Several limitations should be acknowledged. The study has a retrospective design and a relatively small sample size, particularly in the congestive subgroup, precluding multivariable adjustment. The absence of comprehensive cardiac phenotyping – including right ventricular function, diastolic parameters, and valvular disease – limits the ability to disentangle the relative contributions of cardiac dysfunction and volume overload. Additionally, the lack of formal interobserver variability assessment restricts external validity, particularly given the multimodal ultrasound approach. Follow-up management was not delivered under a standardized interventional protocol, and therefore observed changes in congestion parameters should be interpreted as hypothesis-generating rather than causal. Furthermore, NUS cutoffs are not validated in hemodialysis populations and may be influenced by comorbid conditions, limiting immediate clinical applicability.
In this context, the present study should be viewed as a proof-of-concept descriptive framework for multimodal congestion assessment in hemodialysis patients. Future prospective studies using non-circular definitions, standardized acquisition protocols, comprehensive cardiac characterization, and larger cohorts are required to determine its clinical relevance and potential applicability.
Conclusions
Point-of-care ultrasound using the combination of LUS and mVExUS allowed for improved identification of congestion phenotypes and enabled more individualized treatment planning. Congestive patients showed a greater tendency toward malnutrition by NUS, highlighting the need for a comprehensive, multimodal diagnostic strategy to accurately detect and manage this condition. In this context, bedside ultrasound is positioned as a key tool in the routine management of fluid overload in dialysis, with the potential to enhance both patient safety and clinical outcomes.
Statement of Ethics
The research was conducted ethically in accordance with the World Medical Association Declaration of Helsinki. Data collection was conducted in the context of routine clinical practice in compliance, since both ultrasound (vascular, tissue, nutritional) and bioimpedance measurement are performed periodically for the evaluation and monitoring of dialysis treatment of patients, with applicable regulations (Organic Law 3/2018 and EU Regulation 2016/679). This study protocol was reviewed and approved by the Ethics Committee and Institutional Review Board (IRB) of Hospital Central de la Defensa Gomez Ulla (version 4; protocol code 64/24, approved on 9 September 2024). Signed informed consent was obtained from all study participants.
Conflict of Interest Statement
The authors have no conflicts of interest to declare. Gregorio Romero-González was a member of the journal’s Editorial Board at the time of submission.
Funding Sources
The authors declare no financial support for the project.
Author Contributions
José Carlos de la Flor Merino, Prof. MD (primary author), conceived the ideas of the study; performed data collection, data analysis, and interpretation; provided revisions to scientific content of the manuscript. Avinash Chandu Nanwani MD, conceived the ideas of the study; performed data collection, data analysis, and interpretation. Celia Rodriguez Tudero, MD, provided revisions to scientific content of the manuscript and access to crucial research components. Elena Jimenez Mayor, MD, provided revisions to scientific content of the manuscript and access to crucial research components. Irwing Benites, MD, provided revisions to scientific content of the manuscript and access to crucial research components. Juan Lluncor, MD, provided revisions to scientific content of the manuscript and access to crucial research components. Susan Alcalde, MD, performed data analysis, interpretation, and data collection. Hugo Espinoza, MD, performed data analysis, interpretation, and data collection. Katia Hernández, MD, provided revisions to scientific content of the manuscript and performed data collection. Patricia Muñoz-Ramos, MD, provided revisions to scientific content of the manuscript. Gregorio Romero, MD (correspondence author) provided revisions to scientific content of the manuscript. Jesús Hernández, MD, provided revisions to scientific content of the manuscript and grammatical revisions to the manuscript.
Funding Statement
The authors declare no financial support for the project.
Data Availability Statement
The data that support the findings of this study are not publicly available due to institutional and privacy restrictions. However, they are available from the corresponding author upon request (contact J.C.D.L.F., josedelaflor81@yahoo.com).
Supplementary Material.
References
- 1. El BG, Dami F, Hanin H, Kabbali N, Arrayhani M, Sqalli Houssaini T. Bedside lung ultrasound in the assessment of volume status in chronic hemodialysis patients. J Nephrol Adv. 2016;1(1):48–57. [Google Scholar]
- 2. Loutradis C, Sarafidis PA, Ferro CJ, Zoccali C. Volume overload in hemodialysis: diagnosis, cardiovascular consequences, and management. Nephrol Dial Transpl. 2021;36(12):2182–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Pinter J, Canaud B, Mayne KJ, Stuard S, Moissl U, Kooman J, et al. Associations of abnormal fluid status, plasma sodium disorders, and low dialysate sodium with mortality in patients on hemodialysis. Clin J Am Soc Nephrol. 2024;19(11):1444–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Dekker MJE, Marcelli D, Canaud BJ, Carioni P, Wang Y, Grassmann A, et al. Impact of fluid status and inflammation and their interaction on survival: a study in an international hemodialysis patient cohort. Kidney Int. 2017;91(5):1214–23. [DOI] [PubMed] [Google Scholar]
- 5. Turrini F, Galassi M, Sacchi A, Riccò B, Chester J, Famiglietti E, et al. Intrarenal venous Doppler as a novel marker for optimal decongestion, patient management, and prognosis in acute decompensated heart failure. Eur Heart J Acute Cardiovasc Care. 2023;12(10):673–81. [DOI] [PubMed] [Google Scholar]
- 6. De la Espriella R, Núñez-Marín G, Cobo M, de Castro Campos D, Llácer P, Manzano L, et al. Intrarenal venous flow pattern changes Do relate with renal function alterations in acute heart failure. JACC Heart Fail. 2024;12(2):304–18. [DOI] [PubMed] [Google Scholar]
- 7. Koratala A, Teodorescu V, Niyyar VD. The nephrologist as an ultrasonographer. Adv Chronic Kidney Dis. 2020;27(3):243–52. [DOI] [PubMed] [Google Scholar]
- 8. Suarez J, Niyyar VD. Lung ultrasound: a “biomarker” for fluid overload? Adv Chronic Kidney Dis. 2021;28(3):200–7. [DOI] [PubMed] [Google Scholar]
- 9. El Shamy O, Ikizler TA. Fluid management protocols in hemodialysis patients using bioimpedance validation: time to look elsewhere? Kidney Int. 2023;104(3):425–7. [DOI] [PubMed] [Google Scholar]
- 10. Royse AG, Lui E, Gai D, Cid X, Canty D, Wang A, et al. Three zone scanning protocol for lung ultrasound: an anatomical basis. Heart Lung Circ. 2023;32(2):247–51. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Koratala A, Ross DW. Lung ultrasound in hemodialysis patients: is it practical to scan 28 zones? Am J Kidney Dis. 2020;75(5):815–7. [DOI] [PubMed] [Google Scholar]
- 12. Bhardwaj V, Vikneswaran G, Rola P, Raju S, Bhat RS, Jayakumar A, et al. Combination of inferior vena cava diameter, hepatic venous flow, and portal vein pulsatility index: venous excess ultrasound score (VEXUS score) in predicting acute kidney injury in patients with cardiorenal syndrome: a prospective cohort study. Indian J Crit Care Med. 2020;24(9):783–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Rodríguez-Espinosa D, Guzmán-Bofarull J, De la Fuente-Mancera JC, Maduell F, Broseta JJ, Farrero M. Multimodal strategies for the diagnosis and management of refractory congestion: an integrated cardiorenal approach. Front Physiol. 2022;13:913580. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Romero-González G, Manrique J, Slon-Roblero MF, Husain-Syed F, De la Espriella R, Ferrari F, et al. PoCUS in nephrology: a new tool to improve our diagnostic skills. Clin Kidney J. 2023;16(2):218–29. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Prager R, Argaiz E, Pratte M, Rola P, Arntfield R, Beaubien-Souligny W, et al. Doppler-identified venous congestion in septic shock: protocol for an international, multicentre prospective cohort study (Andromeda-VEXUS). BMJ Open. 2023;13(7):e074843. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Tian R, Chang L, Cheng L, Yang R, Zhang H. Malnutrition-inflammation-fluid overload complex syndrome and all-cause mortality in patients undergoing hemodialysis. Ren Fail. 2025;47(1):2512405. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Puri A, Lloyd AM, Bello AK, Tonelli M, Campbell SM, Tennankore K, et al. Frailty assessment tools in chronic kidney disease: a systematic review and meta-analysis. Kidney Med. 2025;7(3):100960. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Nair D, Liu CK, Raslan R, McAdams-DeMarco M, Hall RK. Frailty in kidney disease: a comprehensive review to advance its clinical and research applications. Am J Kidney Dis. 2025;85(1):89–103. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Barbosa EMS, Pereira AG, Mori V, da Silva Neves R, Vieira NM, Silva MZC, et al. Comparison between FRAIL Scale and Clinical Frailty Scale in predicting hospitalization in hemodialysis patients. J Nephrol. 2023;36(3):687–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Chao CT, Hsu YH, Chang PY, He YT, Ueng RS, Lai CF, et al. Simple self-report FRAIL scale might be more closely associated with dialysis complications than other frailty screening instruments in rural chronic dialysis patients. Nephrology. 2015;20(5):321–8. [DOI] [PubMed] [Google Scholar]
- 21. Shin J, Kim H, Jeon HE, Kwon S, Hwang JH, Son H-E, et al. Optimizing hemodialysis management using bioelectrical impedance analysis: emphasizing the significance of the phase angle. Ren Fail. 2025;47(1):2567037. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Flythe JE, Chang TI, Gallagher MP, Lindley E, Madero M, Sarafidis PA, et al. Blood pressure and volume management in dialysis: conclusions from a Kidney Disease: improving Global Outcomes (KDIGO) Controversies Conference. Kidney Int. 2020;97(5):861–76. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Loutradis C, Sarafidis PA, Ekart R, Papadopoulos C, Sachpekidis V, Alexandrou ME, et al. The effect of dry-weight reduction guided by lung ultrasound on ambulatory blood pressure in hemodialysis patients: a randomized controlled trial. Kidney Int. 2019;95(6):1505–13. [DOI] [PubMed] [Google Scholar]
- 24. Siriopol D, Hogas S, Voroneanu L, Onofriescu M, Apetrii M, Oleniuc M, et al. Predicting mortality in haemodialysis patients: a comparison between lung ultrasonography, bioimpedance data and echocardiography parameters. Nephrol Dial Transpl. 2013;28(11):2851–9. [DOI] [PubMed] [Google Scholar]
- 25. Zoccali C, Torino C, Mallamaci F, Sarafidis P, Papagianni A, Ekart R, et al. A randomized multicenter trial on a lung ultrasound-guided treatment strategy in patients on chronic hemodialysis with high cardiovascular risk. Kidney Int. 2021;100(6):1325–33. [DOI] [PubMed] [Google Scholar]
- 26. Leyba K, Longino A, Ormesher R, Krienke M, Van Ochten N, Zimmerman K, et al. VExUS point-of-care ultrasound tool to detect changes in volume status: a prospective observational study. JACC Adv. 2026;5(3):102538. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Madsen LH, Ladefoged S, Corell P, Schou M, Hildebrandt PR, Atar D. N-terminal pro brain natriuretic peptide predicts mortality in patients with end-stage renal disease in hemodialysis. Kidney Int. 2007;71(6):548–54. [DOI] [PubMed] [Google Scholar]
- 28. Wang AY-M, Lam CW-K, Yu C-M, Wang M, Chan IH-S, Zhang Y, et al. N-terminal pro-brain natriuretic peptide: an independent risk predictor of cardiovascular congestion, mortality, and adverse cardiovascular outcomes in chronic peritoneal dialysis patients: an independent risk predictor of cardiovascular congestion, mortality, and adverse cardiovascular outcomes in chronic peritoneal dialysis patients. J Am Soc Nephrol. 2007;18(1):321–30. [DOI] [PubMed] [Google Scholar]
- 29. Sabatino A, Maggiore U, Regolisti G, Rossi GM, Di Mario F, Gentile M, et al. Ultrasound for non-invasive assessment and monitoring of quadriceps muscle thickness in critically ill patients with acute kidney injury. Front Nutr. 2021;8:622823. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Agarwal R, Weir MR. Dry-weight: a concept revisited in an effort to avoid medication-directed approaches for blood pressure control in hemodialysis patients: a concept revisited in an effort to avoid medication-directed approaches for blood pressure control in hemodialysis patients. Clin J Am Soc Nephrol. 2010;5(7):1255–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. La Porta E, Faragli A, Herrmann A, Lo Muzio FP, Estienne L, Nigra SG, et al. Bioimpedance analysis in CKD and HF patients: a critical review of benefits, limitations, and future directions. J Clin Med. 2024;13(21):6502. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data that support the findings of this study are not publicly available due to institutional and privacy restrictions. However, they are available from the corresponding author upon request (contact J.C.D.L.F., josedelaflor81@yahoo.com).
