Introduction
Intestinal failure is a complex clinical condition characterized by the reduction of gut function below the minimum necessary for the absorption of macronutrients, water, and electrolytes. In such cases, intravenous supplementation is required to maintain health, nutritional status, or growth.1,2 Intestinal failure may arise from a variety of underlying pathophysiological mechanisms, including short bowel syndrome, intestinal fistulas, mechanical obstruction, motility disorders, and mucosal disease. The degree of functional impairment depends not only on bowel length but also on the remaining bowel’s capacity for adaptation and absorption.3 In patients where the absorptive function of the digestive system is impaired, dehydration can be a major problem due to the reduced ability to absorb fluids.4,5 Dehydration has a significant impact on an individual’s well-being, quality of life, and clinical outcomes, making regular monitoring essential.6 Beyond hydration status, nutritional status is also profoundly compromised in intestinal failure patients, with malnutrition contributing to increased morbidity and mortality.
Among patients with intestinal failure, those with short bowel syndrome and receiving longterm home parenteral nutrition (HPN) represent a particularly vulnerable subgroup.7,8 These patients are at high risk for both dehydration and malnutrition, which significantly affect their quality of life and clinical outcomes. Dehydration, in particular, is a common yet often underrecognized complication. The clinical assessment of hydration status is challenging due to the lack of a definitive diagnostic standard and the limitations of traditional markers such as serum urea and creatinine. Consequently, regular monitoring using multiple parameters is essential in clinical practice.9 Monitoring of nutritional status should extend beyond BMI to include body composition assessment, as parameters such as muscle mass and fat-free mass provide more accurate indicators of nutritional status and functional outcomes.10
Bioelectrical impedance analysis (BIA) is a non-invasive method widely used for evaluating body composition and estimating hydration status. The diagnostic usage of BIA has increased because the equipment is portable and safe, the procedure is simple, non-invasive, and results are obtained rapidly.11,12 Although BIA is reliable in healthy individuals, its accuracy is reduced in patients with fluid imbalances, such as those with intestinal failure.12,13 Standard prediction equations used by BIA devices are typically developed for healthy populations, potentially leading to significant errors when applied to clinical populations with abnormal fluid distributions. Despite its limitations, BIA remains an accessible and practical tool in everyday clinical care, particularly when used for longitudinal monitoring within the same individual.
Given the increasing use of HPN and the burden of fluid imbalance in patients with intestinal failure, improving hydration assessment is essential. The present study aimed to evaluate the accuracy and clinical utility of two different BIA devices – one that offers only BIA results and the more advanced that offers a combination of BIA, bioelectrical impedance vector analysis (BIVA) and bioelectrical impedance spectroscopy (BIS) in monitoring hydration and nutritional status in patients receiving HPN. The BIA utilizes multifrequency bioelectrical impedance analysis primarily focused on assessing body composition parameters such as total body water, fat mass, and fat-free mass using four different frequencies.14 In contrast, the more advanced BIA/BIVA/BIS offers enhanced diagnostic precision by employing a broader spectrum of multifrequency impedance measurements (50 frequencies from 5 kHz to 1000 kHz), enabling segmental analysis and improved differentiation between intracellular and extracellular water compartments.15 Therefore, we hypothesized that the BIA/BIVA/BIS, which incorporates vector analysis and additional parameters, would show better agreement with clinical and laboratory hydration markers and would more accurately detect malnutrition.
Patients and methods
Study design and ethical approval
This cross-sectional observational study was conducted at the Department for Clinical Nutrition of the Institute of Oncology Ljubljana. Ethical approval was obtained from the National Medical Ethics Committee of the Republic of Slovenia (No. 0120-310/2019/9, dated December 10, 2019). All procedures were carried out in accordance with the ethical standards of the institutional and national research committees.
Participants
The study included patients with chronic intestinal failure treated with HPN who attend regular follow-up visits at the Department for Clinical Nutrition of the Institute of Oncology Ljubljana and provided written informed consent for participation in the study.
Measurement protocol
Measurements were conducted during routine clinical visits. Each patient underwent assessment with two BIA devices – only BIA QuadScan 4000 and more advanced BIA/BIVA/BIS Multiscan 5000 (Bodystat Ltd, Isle of Man, UK) – during the same visit. Body mass was measured using a calibrated medical scale (Seca 7997021099, UK), and height was retrieved from clinical records (measured annually using a wall-mounted stadiometer, Seca 206).
Both BIA assessments were performed in the supine position after a 10-minute rest period. Electrodes were placed on clean, dry skin at standardized anatomical landmarks: dorsally on the wrist and hand, and frontally on the ankle and foot, all on the right side of the body. The signal and measuring cables were connected according to manufacturer specifications.
The devices measured several parameters, including fat mass, fat-free mass, dry lean mass, total body water (TBW), intracellular water (ICW), extracellular water (ECW), phase angle, fat-free mass index (FFMI), and overhydration index (available only on the Multiscan 5000). The parameters were determined by the device’s internal algorithm. The phase angle was calculated from resistance and reactance using the manufacturer’s algorithm.
Laboratory data
Venous blood samples were drawn on the same day as the BIA measurements and analysed at the Institute of Oncology’s central laboratory. Serum urea and creatinine were recorded as biochemical markers of hydration status.
Malnutrition
The GLIM (Global Leadership Initiative on Malnutrition) criteria are a globally endorsed diagnostic framework for malnutrition, combining phenotypic criteria (unintentional weight loss, low BMI, reduced muscle mass) with etiologic criteria (reduced food intake or assimilation, and disease burden/inflammation).16 In our study malnutrition was diagnosed according to GLIM criteria based on FFMI thresholds (< 17 kg/m2 for males, < 15 kg/m2 for females).16
Statistical analysis
Descriptive statistics were calculated for all variables, including means, standard deviations, and distribution characteristics. Normality of data distribution was assessed using the Shapiro-Wilk test. For normally distributed variables, paired t-tests were used to compare the two BIA methods. For non-normally distributed data, non-parametric Mann-Whitney U tests were employed. Chi-square tests were used to evaluate categorical differences. Associations between BIA-derived parameters and biochemical markers were assessed using Spearman correlation coefficients. All analyses were conducted using IBM SPSS Statistics v23 and Microsoft Excel 2016.
Results
Participant characteristics
A total of 55 patients (33 males, 22 females) undergoing home parenteral nutrition (HPN) were included in the study. The mean age was 61.5 years (SD ± 13.1). Body weight ranged from 44.0 to 98.0 kg, with a mean of 68.1 kg (SD = 13.55). The distribution showed minimal skewness (0.18, SE = 0.32) and slightly negative kurtosis (–0.63, SE = 0.63).
Comparison of bioimpedance devices
Body composition was measured using both the BIA and the BIA/BIVA/BIS (Table 1, Table 2). The two devices showed similar values for fat mass, fat-free mass (FFM), and fat-free mass index (FFMI). However, the BIA/BIVA/BIS consistently provided higher estimates of dry lean mass and lower values for total body water (TBW), indicating greater sensitivity to hydration status.
TABLE 1.
Descriptive statistics of measured body composition parameters using the bioelectrical impedance analysis (BIA) device (Quadscan 4000)
| Min | Max | M | SD | Asm. | SEasm | Kurt. | SEspl | S-W | p | |
|---|---|---|---|---|---|---|---|---|---|---|
| Fat mass | ||||||||||
| kg | 7.30 | 42.70 | 19.22 | 7.05 | 0.85 | 0.32 | 1.34 | 0.63 | 0.96 | 0.04 |
| % | 11.30 | 49.80 | 28.24 | 8.40 | 0.32 | 0.32 | –0.05 | 0.63 | 0.98 | 0.61 |
| Fat free mass (FFM) | ||||||||||
| kg | 28.40 | 70.70 | 48.90 | 11.30 | 0.14 | 0.32 | –0.98 | 0.63 | 0.97 | 0.12 |
| % | 50.20 | 88.7 | 71.78 | 8.41 | –0.32 | 0.32 | –0.05 | 0.63 | 0.98 | 0.60 |
| Dry lean mass (kg) | ||||||||||
| 2.00 | 22.20 | 10.52 | 4.83 | 0.26 | 0.32 | –0.55 | 0.63 | 0.98 | 0.43 | |
| Fat free mass index (FFMI) (kg/m2) | ||||||||||
| 9.80 | 21.50 | 16.65 | 2.85 | 0.02 | 0.32 | –0.70 | 0.63 | 0.97 | 0.24 | |
| Phase angle (°) | ||||||||||
| 2.00 | 6.10 | 4.48 | 0.95 | –0.33 | 0.32 | –0.09 | 0.63 | 0.97 | 0.19 | |
| Total body water (TBW) | ||||||||||
| l | 26.80 | 54.2 | 38.53 | 7.28 | 0.26 | 0.32 | –1.00 | 0.63 | 0.96 | 0.07 |
| % | 39.50 | 71.50 | 57.16 | 6.92 | –0.15 | 0.32 | –0.52 | 0.63 | 0.98 | 0.61 |
| Extracellular water (ECW) | ||||||||||
| l | 12.60 | 23.20 | 17.13 | 2.59 | 0.34 | 0.32 | –0.75 | 0.63 | 0.97 | 0.14 |
| % | 19.40 | 31.30 | 25.50 | 2.70 | 0.22 | 0.32 | –0.40 | 0.63 | 0.98 | 0.51 |
| Intracellular water (ICW) | ||||||||||
| l | 12.80 | 28.20 | 20.56 | 4.52 | 0.06 | 0.32 | –1.12 | 0.63 | 0.96 | 0.06 |
| % | 22.60 | 36.50 | 30.25 | 3.57 | –0.14 | 0.32 | –0.98 | 0.63 | 0.97 | 0.15 |
| Prediction marker | ||||||||||
| 0.78 | 0.92 | 0.84 | 0.03 | 0.37 | 0.32 | –0.04 | 0.63 | 0.97 | 0.28 | |
TABLE 2.
Descriptive statistics of measured body composition parameters using the device of bioelectrical impedance analysis (BIA) / bioelectrical impedance vector analysis (BIVA) / bioelectrical impedance spectroscopy (BIS) (Multiscan 5000)
| Min | Max | M | SD | Asm. | SEasm | Kurt. | SEspl | S-W | p | |
|---|---|---|---|---|---|---|---|---|---|---|
| Fat mass | ||||||||||
| kg | 6.90 | 40.50 | 18.15 | 6.73 | 0.86 | 0.32 | 1.38 | 0.63 | 0.95 | 0.03 |
| % | 11.40 | 47.30 | 26.59 | 7.87 | 0.41 | 0.32 | 0.04 | 0.63 | 0.98 | 0.39 |
| Fat free mass (FFM) | ||||||||||
| kg | 30.90 | 73.00 | 50.00 | 11.19 | 0.24 | 0.32 | –0.88 | 0.63 | 0.97 | 0.18 |
| % | 52.70 | 88.60 | 73.41 | 7.87 | –0.41 | 0.32 | 0.04 | 0.63 | 0.98 | 0.39 |
| Dry lean mass | ||||||||||
| kg | 7.80 | 31.30 | 16.76 | 5.43 | 0.42 | 0.32 | –0.31 | 0.63 | 0.97 | 0.18 |
| Fat free mass index (FFMI) | ||||||||||
| kg/m2 | 10.70 | 22.10 | 16.99 | 2.71 | 0.08 | 0.32 | –0.64 | 0.63 | 0.98 | 0.36 |
| Phase angle | ||||||||||
| ° | 2.50 | 6.20 | 4.61 | 0.90 | –0.32 | 0.33 | –0.19 | 0.64 | 0.97 | 0.22 |
| Total body water (TBW) | ||||||||||
| l | 21.70 | 46.50 | 33.16 | 6.47 | 0.22 | 0.32 | –0.88 | 0.63 | 0.97 | 0.24 |
| % | 32.90 | 64.80 | 49.21 | 6.68 | 0.04 | 0.32 | –0.35 | 0.63 | 0.99 | 0.98 |
| Extracellular water (ECW) | ||||||||||
| l | 10.70 | 23.20 | 15.98 | 3.10 | 0.38 | 0.32 | –0.80 | 0.63 | 0.96 | 0.08 |
| % | 17.20 | 46.90 | 24.06 | 4.09 | 3.26 | 0.32 | 17.59 | 0.63 | 0.74 | 0.00 |
| Intracellular water (ICW) | ||||||||||
| l | 9.00 | 25.20 | 17.24 | 3.86 | 0.10 | 0.32 | –0.81 | 0.63 | 0.98 | 0.34 |
| % | 15.20 | 36.80 | 25.64 | 4.66 | 0.13 | 0.32 | –0.13 | 0.63 | 0.99 | 0.91 |
| Overhydration index | ||||||||||
| l | 0.50 | 7.40 | 2.89 | 1.59 | 0.97 | 0.32 | 0.83 | 0.63 | 0.92 | 0.00 |
| % | 0.60 | 11.50 | 4.34 | 2.30 | 0.78 | 0.32 | 0.75 | 0.63 | 0.96 | 0.05 |
| Prediction marker | ||||||||||
| 0.79 | 0.92 | 0.84 | 0.03 | 0.53 | 0.32 | 0.08 | 0.63 | 0.96 | 0.08 | |
Phase angle values measured by the BIA/BIVA/BIS (mean 4.61°) were slightly higher than those measured by the BIA (mean 4.48°), while extracellular water (ECW) and intracellular water (ICW) distributions differed more significantly between devices.
The BIA/BIVA/BIS also provided an additional parameter – overhydration (OHY) – with an average value of 4.34% body weight, not available on the BIA.
Laboratory markers of hydration
Serum urea and creatinine were used as laboratory markers of hydration status. The mean urea level was 7.54 mmol/L and creatinine was 85.2 μmol/L. Both markers showed non-normal distributions.
Correlation analysis revealed a statistically significant negative correlation between TBW (and ICW) measured by the BIA/BIVA/BIS and serum urea levels (Table 3). This relationship was stronger than the correlation observed with the BIA data, suggesting better concordance with clinical markers of dehydration. For creatinine, the associations were weaker and less consistent across devices.
TABLE 3.
The correlation between blood hydration markers and parameters measured by both devices
| Multiscan 5000 | Quadscan 4000 | |||
|---|---|---|---|---|
| Urea | Creatinine | Urea | Creatinine | |
| Total body water (TBW) | ||||
| l | –0.552** | 0.257 | –0.471** | 0.332* |
| % | –0.185 | –0.331* | –0.054 | –0.286* |
| Extracellular water (ECW) | ||||
| l | –0.453** | 0.283* | –0.465** | 0.297* |
| % | 0.007 | –0.297* | 0.145 | –0.419** |
| Intracellular water (ICW) | ||||
| l | –0.573** | 0.207 | –0.499** | 0.395** |
| % | –0.244 | –0.253 | –0.251 | –0.025 |
Identification of malnutrition
According to GLIM criteria BIA identified 16.4% of participants as malnourished, while the BIA/BIVA/BIS identified 24.4%. However, the difference was not statistically significant (p = 0.436).
Sensitivity in detecting dehydration
Based on relative TBW thresholds (< 55% in men, < 50% in women), the BIA/BIVA/BIS identified significantly more dehydrated patients (69.1%) compared to the BIA (29.1%) (χ2 = 17.61, p < 0.001) (Table 4). This finding suggests that the BIA/BIVA/BIS detects dehydration earlier or with greater sensitivity.
TABLE 4.
Results of Mann-Whitney U test in measured parameters according to the applied method
| Multiscan 5000 | Quadscan 4000 | Mann-Whitney U | p | |||
|---|---|---|---|---|---|---|
| M | SD | M | SD | |||
| Total body water (TBW) | ||||||
| l | 33.16 | 6.47 | 38.53 | 7.24 | 897.50 | < 0.001 |
| % | 49.21 | 6.68 | 57.16 | 6.92 | 626.50 | < 0.001 |
| Extracellular water (ECW) | ||||||
| l | 15.98 | 3.10 | 17.13 | 2.59 | 1148.00 | 0.03 |
| % | 24.06 | 4.09 | 25.50 | 2.70 | 985.00 | < 0.001 |
| Intracellular water (ICW) | ||||||
| l | 17.24 | 3.86 | 20.56 | 4.52 | 899.00 | < 0.001 |
| % | 25.64 | 4.66 | 30.25 | 3.57 | 672.00 | < 0.001 |
TABLE 5.
Comparison of the number of dehydrated patients based on blood measurements of urea and total body water
| Total body water (TBW) | |||||
|---|---|---|---|---|---|
| Dehydration | Normal hydration | Total | |||
| BIA | Urea | Dehydration | 4 | 16 | 20 |
| Urea | Normal hydration | 12 | 23 | 35 | |
| Urea | Total | 16 | 39 | 55 | |
| BIA/BIVA/BIS | Urea | Dehydration | 14 | 6 | 20 |
| Urea | Normal hydration | 24 | 11 | 35 | |
| Urea | Total | 38 | 17 | 55 |
TABLE 6.
Comparison of the number of dehydrated patients based on blood measurements of creatinine and total body water
| Total body water (TBW) | |||||
|---|---|---|---|---|---|
| Dehydration | Normal hydration | Total | |||
| BIA | Creatinine | Dehydration | 5 | 9 | 14 |
| Creatinine | Normal hydration | 11 | 30 | 41 | |
| Creatinine | Total | 16 | 39 | 55 | |
| BIA/BIVA/BIS | Creatinine | Dehydration | 13 | 1 | 14 |
| Creatinine | Normal hydration | 25 | 16 | 41 | |
| Creatinine | Total | 38 | 17 | 55 |
Correlation with phase angle
Using a median phase angle of 4.65° as the cut-off, 49% of participants were classified as dehydrated. The BIA/BIVA/BIS showed 38% concordance with phase angle-based classification, while the BIA showed only 13% agreement. Although not statistically significant, the BIA/BIVA/BIS showed better alignment with functional hydration markers such as phase angle.
Discussion
This study compared the performance of two BIA devices in assessing hydration and nutritional status among patients with intestinal failure on home parenteral nutrition. The results suggest that the newer and more advanced BIA/BIVA/BIS offers clinically meaningful advantages over the BIA, particularly in detecting dehydration and aligning with biochemical and functional markers.
One of the most important findings was the BIA/BIVA/BIS’s stronger correlation with serum urea, a widely accepted marker of hydration. To our knowledge no such study was yet published. BIA/BIVA/BIS measurements of TBW and ICW explained a greater proportion of variability in urea levels than those from the BIA. Although the associations with serum creatinine were more mixed, the BIA/BIVA/BIS still demonstrated a marginally better overall concordance.
In terms of clinical classification, the BIA/BIVA/BIS identified a substantially higher number of dehydrated patients using TBW thresholds, and its classifications more frequently matched with biochemical indicators. This supports the hypothesis that more advanced BIA technologies, particularly those incorporating vector analysis and spectroscopy, may offer enhanced sensitivity in detecting fluid imbalances in clinical populations with abnormal physiology.
Similar findings, but without comparison to biochemical indicators, were published by Fassini et al. This observational study found that BIVA identified significant differences in hydration and soft tissue status between short bowel syndrome patients and matched controls, even when traditional bioelectrical impedance measures did not differ.17 In patients with chronic kidney disease, Espinosa-Cuevas et al. demonstrated that BIVA showed good agreement with body composition and hydration parameters obtained using four different BIA technologies, particularly for the detection of fluid overload and post-dialysis changes in hydration status in hemodialysis patients.18 Moreover, a narrative review on body composition assessment in athletes published in 2021 highlighted that BIVA provides greater accuracy in evaluating hydration status and body fluid changes compared with conventional BIA, which is limited by the lack of athlete-specific predictive equations.19
Interestingly, although more patients were identified as malnourished using the BIA/BIVA/BIS, the difference did not reach statistical significance. This may be due to the small sample size or the limitations of FFMI alone as a diagnostic criterion. Nevertheless, the observed trend supports the need for further investigation into the use of advanced BIA tools for nutritional risk screening.
Phase angle, a parameter derived directly from resistance and reactance, is considered a robust indicator of cellular integrity and prognosis.20,21 In this study, phase angle-based classification of hydration showed greater agreement with BIA/BIVA/BIS. This is particularly relevant because phase angle is not reliant on predictive equations and is less affected by assumptions about body shape or hydration state20, making it valuable in populations with abnormal fluid distribution, such as IF patients.
Limitations of this study include the singletime-point measurement and a relatively small sample size. Nonetheless, results suggest that the BIA/BIVA/BIS offers added value in monitoring hydration and nutritional status. Further research with a larger cohort and longitudinal design could validate these findings and support their application in broader clinical practice.
Conclusions
Our findings have practical implications for the routine monitoring of HPN patients. Given the challenges of accurately assessing hydration and nutritional status in this population, incorporating more sophisticated BIA tools like the BIA/BIVA/BIS could aid in early detection of dehydration and malnutrition, guide individualized therapy, and ultimately improve patient outcomes.
Notes
[8] AI disclosure
During the preparation of this paper, the authors used Claude (Anthropic, Claude Opus 4.7) to improve the style and readability in some parts of the text. After using this tool/service, the authors have reviewed and edited the content as required and take full responsibility for the content of the publication.
[9] Contributed by Author contributions
TJ: Visualization; Validation; Writing – original draft; Writing – review & editing; JO: Formal analysis; Data curation; Investigation; Methodology; NRK: Conceptualization; Resources; Project administration; Supervision; Writing – review & editing.