Use of a Structured Interview to Assess the Association of Inadequate Energy and Protein Intake with Subclinical Congestion in Heart Failure Patients

Congestion is a leading cause of hospitalization in patients with heart failure (HF), and its recurrence is strongly associated with worse prognosis¹. While pulmonary and peripheral oedema are classical clinical signs, subclinical congestion, characterized by elevated cardiac filling pressures without overt symptoms, can develop weeks before decompensation and often remains undetected with conventional assessment methods². This early-stage congestion may promote gastrointestinal dysfunction through mechanisms such as intestinal wall oedema, increased mucosal permeability, bacterial translocation, and systemic inflammation, all of which can contribute to symptoms like nausea, early satiety, and cardiac cachexia³. However, the presence of nonspecific symptoms such as dyspnea or fatigue does not necessarily imply congestion, as they may result from other prevalent conditions in HF patients, such as poor physical conditioning, anaemia, or depressive symptoms⁴. Therefore, objective tools are essential for accurately identifying true fluid overload and differentiating it from other causes of symptom burden in this complex population⁵. This is especially relevant in cases of subclinical congestion, which can be identified through whole-body bioimpedance analysis and is much more common than previously thought, affecting nearly two-thirds of ambulatory patients with heart failure with reduced ejection fraction (HFrEF). Notably, this early, often unnoticed fluid accumulation is associated with increased rates of worsening heart failure events and worse clinical outcomes, highlighting the importance of using sensitive, noninvasive assessments in routine care⁶. Malnutrition, often intertwined with frailty and other comorbidities, has been consistently recognized as an independent predictor of adverse outcomes in HF, reinforcing the need for early detection and targeted interventions. Despite this, routine assessment of dietary intake is rarely incorporated into follow-up visits for patients with HF, even though evaluating nutrition is essential for identifying inadequate intake, recognizing early signs of catabolism, and guiding individualized dietary strategies that may prevent further functional decline^3,4.

Previous studies have suggested that congestion may influence nutritional status. For example, a study conducted in 2020 reported a significant association between higher B-type natriuretic peptide (BNP) levels at discharge and inadequate dietary intake, based on hospital meal consumption analysis⁷. As BNP reflects elevated cardiac filling pressures, this association may indicate persistent subclinical congestion impacting nutritional intake at discharge. Despite these insights, evidence on the relationship between subclinical congestion, dietary intake, and clinical outcomes, especially in stable ambulatory patients, remains limited. Therefore, further research is warranted to clarify these interactions using validated diagnostic tools for assessing congestion. This study aimed to determine whether subclinical congestion is associated with poor nutrient intake in patients with HF, hypothesizing that early detection and nutritional evaluation may play a critical role in patient management and prognosis.

This study was conducted according to the guidelines of the Instituto Nacional de Ciencias Médicas y Nutrición Salvador Zubirán (INCMNSZ) and approved by the Biomedical Research Ethics Committee (reference number 4504). Written and verbal informed consent was obtained from all participants prior to enrolment.

Study design and population

This was a cross-sectional study. Patients aged ≥ 18 years with a confirmed diagnosis of heart failure for at least six months, clinically stable, with no medication adjustments in the preceding three months, and no hospital admissions or acute decompensation within the last month who attended the HF Clinic at INCMNSZ between April 2023 and January 2025 were included. The exclusion criteria comprised current use of nutritional supplements, pregnancy, active cancer diagnosis, and gastrointestinal diseases associated with malabsorption. A total of 122 patients were recruited, and after applying the selection criteria, 117 were included in this study. The recruitment and selection of participants are depicted in the study flow diagram (Figure 1).

Assessment of subclinical congestion

Subclinical congestion was defined as the presence of objective signs of fluid overload detected by both the Venous Excess Ultrasound Score (VExUS) and Bioelectrical Impedance Vector Analysis (BIVA), in the absence of overt clinical signs of congestion (such as peripheral oedema, pulmonary rales, jugular venous distension, or orthopnea). Patients presenting only with nonspecific symptoms, such as fatigue or exertional dyspnea, were not classified as clinically congested. The clinicians who evaluated the patients were not blinded to the BIVA or VExUS results.

Venous Excess Ultrasound Score (VExUS) procedure

The VExUS assessment was performed by a trained cardiologist using a hybrid portable ultrasound device that offers pulsed and continuous Doppler with a low-frequency (2.5-5 MHz) sectorial transducer. Four main veins were measured: the inferior vena cava, portal vein, hepatic vein, and suprarenal vein. According to these measurements, the degree of congestion can be obtained, ranging from 0 to 3, where a degree equal to or greater than 1 is considered congestion. A more detailed description of the VExUS protocol has been described elsewhere⁸.

Bioelectrical Impedance Vector Analysis (BIVA) procedure

The bioelectrical impedance we used analyzes the whole body, measuring fat mass, lean, and body water across the entire body; it allows for detecting imbalances and tracking localized changes. BIVA was performed using a monofrequency (50 kHz), tetrapolar bioelectrical impedance device. Resistance and reactance measurements were obtained and standardized to the subject's height. According to the resistance and reactance calculations, we obtained the phase angle and vector. If the vector is below the 75^th percentile, it indicates that congestion is present. The procedure followed the standardized protocols described elsewhere⁹.

Dietary intake assessment

Dietary intake assessment was performed on the same day as the clinical visit and congestion evaluation. The questionnaire was evaluated using three non-consecutive 24 h dietary recalls (Supplementary 1) administered in person by trained nutrition professionals.

The multiple-pass method was used to enhance the accuracy of dietary recall. Patients were first asked to provide a free and uninterrupted list of all foods and beverages consumed during the previous 24 h. Subsequently, interviewers probed for commonly forgotten items, including snacks, beverages, sauces, and condiments, to minimize omissions. Patients then reported the time of consumption and the eating occasion for each item recalled. Detailed information regarding food preparation methods, portion sizes, ingredients, and, when applicable, product types was collected to allow for precise nutrient estimation. Finally, the entire recall was reviewed with each patient to verify completeness and accuracy.

The recalls were performed on random days, ensuring that one recall included a weekend day to account for variability in dietary patterns. The average intake from the three recalls was used for analysis.

Dietary adequacy was defined as an intake of at least 60% of the estimated energy requirements (25-30 kcal/kg/day) and a protein intake of 1.2 g/kg/day, based on clinical nutrition guidelines for patients with chronic diseases¹⁰. Lower intake was classified as inadequate, considering its association with increased malnutrition risk.

Nutritional analysis

Dietary data collected from the 24 h recalls were entered into the Food Processor Nutrition Analysis Software (version 7) for nutrient composition analysis. This software was previously used in another study to evaluate dietary phosphorus intake in patients treated with peritoneal dialysis, with good results¹¹. The software database includes comprehensive information on macronutrients, micronutrients, and fiber content. Inputting detailed food descriptions, preparation methods, and portion sizes enabled the precise calculation of total energy intake, macronutrient distribution, and fiber intake for each participant.

The dietary assessment and analysis were performed using ESHA's Food Processor software following a standardized workflow to ensure reproducibility and consistency. As part of the data entry process, an individual patient profile was created by selecting person > New, where demographic and anthropometric variables, including age, sex, weight, height, and relevant clinical information, were recorded. Measurement units were standardized to grams, milliliters, and kilograms, and individualized nutritional goals were defined when required, such as energy requirements of 25-30 kcal/kg/day and protein intake of 1.2 g/kg/day.

The nutrient output configuration was established prior to data analysis by selecting preferences > Nutrients to View or by customizing columns within reports. This configuration allowed for the evaluation of total energy intake, macronutrient distribution expressed as percentages, protein intake normalized to body weight (g/kg), total dietary fiber, soluble and insoluble fiber, and sodium intake. This setup ensured uniformity across all dietary analyses.

Dietary data collection was conducted independently of the software using the multiple-pass 24 h recall method. In Step 1 of the recall procedure, patients were asked to freely recall all foods and beverages consumed during the previous 24 h. In Step 2, interviewers systematically probed for commonly forgotten items, such as snacks, beverages, sauces, and condiments. In Step 4, detailed information regarding food preparation methods, portion sizes, ingredients, and product types was collected. This structured approach was used to improve recall precision before dietary data were entered into Food Processor.

All reported foods and beverages were entered into the software using the search function, and appropriate items were selected from generic, branded, or restaurant databases when applicable. Portion sizes were specified using standardized units or household measures that reflected actual consumption. To estimate usual dietary intake, recalls were completed on two to three nonconsecutive days within the same patient profile, including at least one weekend day, with each recall assigned to its corresponding date.

Daily and averaged nutrient intakes were generated by selecting reports > Spreadsheet, Nutrient Totals, or Intakes vs. Goals, allowing for evaluation of both single-day intake and multi-day averages across selected recalls.

All dietary data were subsequently exported using reports > Export in CSV, Excel, or PDF format for statistical analysis and inclusion in manuscript materials.

Dietary intake was therefore assessed using three nonconsecutive 24 h multi-pass recalls and analyzed with Food Processor, with the nutrient database updated to the corresponding version and multi-day average intakes used for analysis.

In this study, we selected a dietary intake questionnaire instead of commonly used nutritional risk tools because these tools rely primarily on biochemical markers (e.g., albumin and lymphocyte count) and anthropometric measurements to estimate nutritional risk, but they do not capture information on actual food intake or eating patterns. In contrast, the three non-consecutive 24 h dietary recalls employed in our methodology provided a direct and detailed evaluation of patients' consumption, making it possible to identify both qualitative and quantitative insufficiencies in macro and micronutrient intake. This approach was better suited to the study objective of examining dietary intake in relation to subclinical congestion.

Regarding grip strength, it is important to mention that a dynamometry test was used, considering the values for the mexican population, where in men, a dynamometry below 27 kg and in women, a dynamometry below 16 kg is considered low grip strength¹².

Statistical analysis

Categorical variables were summarized using frequencies and percentages. Continuous variables were expressed as means with standard deviations or medians with interquartile ranges, depending on data distribution. The Kolmogorov-Smirnov test with Lilliefors correction assessed normality¹³. Comparisons between groups (adequate vs. inadequate intake) were performed using Pearson's chi-square or Fisher's exact test for categorical variables, and Student's t-test or Mann-Whitney U test for continuous variables as appropriate. The significance threshold was p < 0.05. For the multivariate analysis, multiple linear regression was performed, in which subclinical congestion, dyspnea, and lack of appetite were included as covariates (potential confounding factors) to determine whether poor dietary intake was independently associated with any of these variables in a statistically significant manner. Jamovi software version 2.7.5 was used for statistical analysis.

Study population

A total of 117 patients met the inclusion criteria and were enrolled in the study. Among them, 54% were women, and the median age was 66.5 years. The patient selection process is detailed in the flowchart presented in Figure 1. The baseline characteristics of the study population, stratified by the presence or absence of subclinical congestion, are summarized in Table 1. No significant differences were observed between groups regarding the prevalence of dyspnea, fatigue, or gastrointestinal symptoms. The only statistical differences were in the New York Heart Association (NYHA) classification, phase angle, resistance, reactance Z-score, and serum biomarker levels.

Subclinical congestion assessment

Assessment using VExUS and BIVA identified distinct differences between patients with and without subclinical congestion. Notably, the group with subclinical congestion demonstrated lower diastolic blood pressure, higher heart rate, and elevated biomarkers like BNP and cancer antigen 125 (Ca125), suggesting a hemodynamic impact associated with fluid overload despite the absence of overt clinical signs of congestion14. It is important to note that these biomarkers were taken 1 week before the clinical visit. These findings support the hypothesis that subclinical congestion can be objectively identified, even in the absence of traditional clinical indicators.

Dietary intake evaluation

Table 2 presents the dietary intake characteristics of the patients categorized by subclinical congestion status. Patients with subclinical congestion exhibited significantly lower total energy intakes than those without congestion (p = 0.001). Protein intake was also notably lower in the congested group, both as a percentage of total dietary intake (p = 0.001) and when adjusted per kilogram of body weight. In addition, these patients had a significantly higher carbohydrate intake and reduced fat intake. Fiber intake was also diminished in the subclinical congestion group, with significant differences observed for total fiber (p = 0.008) and soluble (p < 0.001) and insoluble (p = 0.046) fractions.

Dietary intake and subclinical congestion association

Figure 2 illustrates the relationship between subclinical congestion and inadequate dietary intake. A significantly higher proportion of patients with subclinical congestion reported inadequate energy intake (78%) compared to those without congestion (22%) (p < 0.001; RR= 2.64, IC95% = 1.29-5.50). Furthermore, the protein intake rate relative to weight of participants with subclinical congestion is lower compared to those without congestion (0.84 vs. 1.44; p = 0.001), as shown in Table 2.

The multivariate linear logistic regression model (Table 3) demonstrated that subclinical congestion was significantly associated with an increased likelihood of inadequate dietary intake (odds ratio [OR] 10.04; 95% confidence interval [CI]: 1.03-97.75; p = 0.047). These findings persisted after adjusting for potential confounders, supporting the hypothesis that subclinical congestion negatively impacts dietary intake in heart failure patients.

This study demonstrated a significant association between subclinical congestion and inadequate dietary intake in ambulatory patients with heart failure. Patients with subclinical congestion exhibited lower energy and protein intakes, higher carbohydrate consumption, and reduced fiber intake than noncongested patients. These findings suggest that subclinical congestion may play a critical role in influencing nutritional status, emphasizing the importance of early detection and targeted nutritional interventions as part of comprehensive heart failure management.

Data availability

The database is provided as Supplementary materials.

Figure 1: Flowchart of patient selection for this study. Please click here to view a larger version of this figure.

Figure 2: Association between dietary intake and the presence of subclinical congestion in the studied population. Please click here to view a larger version of this figure.

Table 1: Baseline characteristics of the study population.

Table 2: Dietary characteristics of subjects with congestion vs. those without congestion.

Table 3: Multivariate linear logistic regression model for inadequate dietary energy intake.

Supplemental material 1: Multi-step 24 h recall format. Please click here to download this file.

Supplemental material 2: Protocol database. Please click here to download this file.

This study proposes that subclinical congestion, even in the absence of overt clinical signs, is significantly associated with inadequate dietary intake in patients with heart failure. Our results revealed that individuals with subclinical congestion consumed fewer calories and less protein, along with higher carbohydrate intake and reduced fiber intake, than those without congestion. These findings raise the possibility that nutritional status is intertwined with the mechanisms driving subclinical congestion; however, the directionality of this relationship cannot be determined from our data, as the study was not longitudinal. It may be more biologically plausible that advancing heart failure contributes to malnutrition rather than the reverse, through processes such as intestinal malabsorption, catabolism, neurohormonal dysregulation, and inflammatory cytokine activation, which can lead to anorexia and diminished appetite¹⁵. While causality cannot be established, our results highlight an important clinical intersection between nutrition and congestion, leading to future research aimed at clarifying the temporal sequence and identifying whether targeted nutritional interventions could modify this trajectory.

The findings also raise the possibility that inadequate nutrition could mediate the association between congestion and sarcopenia, as elevated BNP levels (indicative of increased cardiac filling pressure) have been linked to muscle wasting^16,17,18,19. Over the past decades, the understanding of body composition abnormalities in HF has evolved substantially. More recent consensus statements have emphasized the persistent gaps in assessing nutritional status and muscle health in this population and highlighted the need for integrated approaches that address both metabolic and functional decline. Our findings contribute to this ongoing discussion by suggesting that inadequate dietary intake may partly explain the association between congestion and reduced function. It is biologically plausible that congestion-driven neurohormonal and inflammatory pathways could accelerate sarcopenia, particularly in individuals who are unable to meet their nutritional requirements.

From a clinical standpoint, incorporating nutritional assessment alongside objective measures, such as handgrip strength, may improve the early identification of patients at risk for progressive muscle loss and help differentiate sarcopenia from more advanced cachexia. This distinction is crucial, as sarcopenia may still be reversible with timely nutritional and rehabilitative interventions, whereas cachexia often represents a more advanced systemic catabolic state²⁰.

From a scientific perspective, this research advances the field by highlighting a novel intersection between haemodynamic status and nutrition in heart failure. While prior studies have addressed malnutrition in advanced disease, our approach integrates objective congestion assessment tools with comprehensive dietary evaluation, thereby providing a reproducible method for identifying early nutritional deficits. The lower energy and protein intakes observed in the congested group align with previous studies associating poor nutritional status with disease progression^16,17, whereas the increased carbohydrate consumption may reflect metabolic and gastrointestinal alterations characteristic of heart failure, such as insulin resistance and oxidative stress, or abdominal distention or early satiety, which contribute to endothelial dysfunction and fluid retention²¹. Similarly, reduced fiber intake, both soluble and insoluble, may worsen systemic inflammation, intestinal permeability, and gut dysbiosis, all of which are implicated in heart failure pathophysiology via the gut-heart axis^22,23,24. Therefore, incorporating systematic dietary assessments into clinical practice could improve risk stratification and support timelier, patient-centered interventions.

Our findings should also be interpreted in the context of the "obesity paradox," a well-described phenomenon in heart failure in which patients with overweight or mild obesity exhibit better survival despite excess adiposity. Although higher BMI may confer greater metabolic and nutritional reserve, this measure does not adequately capture body composition nor distinguish between fat and lean mass. Emerging evidence highlights the concept of "malnourished obesity" or sarcopenic obesity, in which individuals classified as overweight may simultaneously present with inadequate dietary intake, low muscle mass, and functional impairment. This framework is particularly relevant to our cohort, where a proportion of patients had elevated BMI yet demonstrated suboptimal caloric and protein intake. Such findings suggest that excess body weight may mask underlying nutritional vulnerability, especially in the presence of subclinical congestion, which can exacerbate anorexia, promote catabolic pathways, and impair nutrient absorption. Therefore, assessments that incorporate dietary intake, muscle function (e.g., handgrip strength), and body composition may offer a more accurate characterization of nutritional risk than BMI alone. This perspective aligns with recent literature and reinforces the importance of integrating nutritional evaluation into routine heart failure care^25,26.

The limitations of this study should be acknowledged. The wide confidence intervals in the logistic regression analyses may be attributed to the sample size, suggesting the need for larger studies to confirm these associations. Moreover, while VExUS and BIVA provide sensitive, non-invasive detection of congestion, they do not directly measure splanchnic or myocardial tissue fluid accumulation, and their interpretation may vary according to operator expertise. The 24 h dietary recall, despite being conducted using a validated multiple-pass method, is subject to recall bias and may underestimate or overestimate true intake. Additional research could integrate objective dietary biomarkers, continuous haemodynamic monitoring, and imaging modalities such as cardiac magnetic resonance imaging or advanced echocardiography to validate and expand these findings.

The methodology used in this study, combining VExUS, BIVA, and detailed dietary intake analysis, holds potential applications in clinical and research settings. It could be employed to develop risk-stratification tools for heart failure patients, identifying those most likely to benefit from early dietary interventions before clinical decompensation, such as improving the qualitative composition of dietary intake^27,28. Furthermore, this approach may be valuable in other populations at risk of volume overload and malnutrition, such as those with chronic kidney disease or pulmonary hypertension, where subclinical congestion and nutritional deficits often coexist.

Future studies should explore whether correcting nutritional deficiencies can modify the course of subclinical congestion and improve clinical outcomes. This includes interventional trials assessing targeted strategies such as individualized caloric and protein optimization, dietary fiber enrichment, controlled carbohydrate distribution, and modulation of the gut microbiota with prebiotics or probiotics. Integration of these interventions into multidisciplinary heart failure management could reduce hospitalizations, improve quality of life, and mitigate the burden of heart failure on healthcare systems. Longitudinal studies are also warranted to determine causal pathways and to validate whether early nutritional correction can prevent the progression from subclinical to overt congestion.