Validating the use of bioimpedance spectroscopy for assessment of fluid status in children

Validating the use of bioimpedance spectroscopy for assessment of fluid status in children Background Bioimpedance spectroscopy (BIS) with a whole-body model to distinguish excess fluid from major body tissue hydration can provide objective assessment of fluid status. BIS is integrated into the Body Composition Monitor (BCM) and is validated in adults, but not children. This study aimed to (1) assess agreement between BCM-measured total body water (TBW) and a gold standard technique in healthy children, (2) compare TBW_BCM with TBW from Urea Kinetic Modelling (UKM) in haemodialysis children and (3) investigate systematic deviation from zero in measured excess fluid in healthy children across paediatric age range. Methods TBW_BCM and excess fluid was determined from standard wrist-to-ankle BCM measurement. TBW_D2O was determined from deuterium concentration decline in serial urine samples over 5 days in healthy children. UKM was used to measure body water in children receiving haemodialysis. Agreement between methods was analysed using paired t test and Bland-Altman method comparison. Results In 61 healthy children (6–14 years, 32 male), mean TBW_BCM and TBW_D2O were 21.1 ± 5.6 and 20.5 ± 5.8 L respectively. There was good agreement between TBW_BCM and TBW_D2O (R = 0.97). In six haemodialysis children (4–13 years, 4 male), 45 concomitant measurements over 8 months showed good TBW_BCM and TBW_UKM agree- ment (mean difference − 0.4 L, 2SD = ± 3.0 L). In 634 healthy children (2–17 years, 300 male), BCM-measured overhydration was − 0.1 ± 0.7 L (10–90th percentile − 0.8 to + 0.6 L). There was no correlation between age and OH (p =0.28). Conclusions These results suggest BCM can be used in children as young as 2 years to measure normally hydrated weight and assess fluid status. . . . . . . Keywords Fluid volume Bioimpedance Chronic kidney disease Overhydration Total body water Children Haemodialysis Electronic supplementary material The online version of this article (https://doi.org/10.1007/s00467-018-3971-x) contains supplementary material, which is available to authorized users. * Indranil Dasgupta Pediatric Nephrology Division, Center for Pediatrics and Adolescent Indranil.Dasgupta@heartofengland.nhs.uk Medicine, University of Heidelberg, Heidelberg, Germany Institute for Human Nutrition and Food Science, Christian-Albrecht University, Kiel, Germany Heartlands Hospital and University of Birmingham, Birmingham, UK Universität Hohenheim, Hohenheim, Germany Departments of Renal Medicine and Medical Physics, Leeds Department of Pediatrics, Institute of Clinical Sciences, Sahlgrenska Teaching Hospitals NHS Trust, Leeds, UK Academy at University of Gothenburg, Gothenburg, Sweden Department of Children’s Nephrology, Leeds Teaching Hospitals Global R&D, Fresenius Medical Care, Bad Homburg, Germany NHS Trust, Leeds, UK 1602 Pediatr Nephrol (2018) 33:1601–1607 Introduction measured OH from zero in healthy children across the paedi- atric age range. Fluid management in haemodialysis impacts patient experi- ence, morbidity and mortality [1, 2]. While inadequate fluid removal can lead to oedema and may precipitate heart failure, Methods volume depletion can cause hypotension, dizziness, cramps, abdominal symptoms, prolonged recovery time following Study participants haemodialysis and accelerated loss of residual renal function. An increased risk of morbidity and mortality has been associ- For the method comparisons, healthy children aged 6 to ated both with chronic fluid overload [3–5]and with 14 years related to hospital staff and children receiving intradialytic hypotension and loss of residual function [6–9]. haemodialysis at the Paediatric Nephrology Department of Determination of optimal fluid status in a dialysis patient is Leeds Teaching Hospitals NHS Trust UK were recruited. For challenging [10]. Conventionally, the assessment of fluid status the investigation of age-related changes in BCM-measured is based on clinical symptoms and signs. Progressive reduction OH, a large cohort of healthy children aged 2 to 17 years from of target weight until the patient becomes symptomatic is de- Germany and Sweden was studied. Children with limb ampu- scribed as ‘probing for dry weight’. Assessment of fluid status is tations, cardiac or other chronic diseases were excluded. even more challenging in children who often have rapid changes in flesh weight. A number of technologies are now available to Measurement of TBW_BCM and OH aid fluid management in renal patients, of which bioimpedance spectroscopy (BIS) is one of the most widely studied. BCM measurements to obtain TBW (TBW_BCM) and OH Whole-body (wrist-to-ankle) BIS measurements can be were made using standard whole-body electrode configura- used to determine extracellular, intracellular and total body tion after the child had been lying down for at least 5 min. water (ECW, ICW and TBW) volumes [11]. The addition of TBW_BCM in healthy children was measured once on the a 3-compartment body model to distinguish excess extracel- dominant side. In children on haemodialysis, measurements lular fluid from lean and adipose tissue allows an objective were performed on the dominant or non-fistula side before the assessment of excess fluid or ‘overhydration’ (OH) [12]. The start of dialysis once a month for up to 8 months. BIS fluid volume model and the body model are integrated The BCM measures whole-body impedance over 50 fre- into the Body Composition Monitor (BCM, Fresenius AG, quencies (from 5 kHz to 1 MHz) and determines extracellular Bad Homburg), which has been well validated [13] and shown and total body resistance by Cole modelling [16]inorder to to be associated with better survival in adult dialysis patients estimate ECW and ICW using the fluid volume model [11]. [3, 5]. The BCM is increasingly being used to assist the man- The 3-compartment body composition model uses these vol- agement of adult dialysis patients [14, 15]. For paediatric pa- umes to separate the body weight into normally hydrated lean tients to benefit from this technology, evidence of the applica- tissue mass (LTM), normally hydrated adipose tissue mass bility of the underlying models in children is required. (ATM) and excess fluid (or overhydration, OH) [12]. As OH The extensive validation of the BCM models against gold is simply the discrepancy between the actual body weight and standards that has been carried out in adults is impractical in the normally hydrated weight (LTM + ATM), it can be posi- children. However, it is possible to measure TBW in both tive or negative. OH is typically within a range of − 1.1 to + healthy subjects and haemodialysis patients using child- 1.1 L (10th to 90th percentile) in healthy adults without car- friendly techniques, against which the TBW estimated by the diac or renal complications [14]. However, in both healthy BCM fluid volume model can be assessed. While there is no subjects and dialysis patients, the deviation from normal hy- gold standard for quantifying excess extracellular fluid, it dration represented by a BCM-measured OH of 1.1L scales should be absent in healthy subjects. BCM OH measurements with the size of the individual. To compare individuals of that are close to zero, and without systematic variation with different size, OH is normalised to ECW as excess fluid pri- age, in healthy children would provide further confirmation of marily accumulates in the extracellular space. the validity of the fluid volume model and indicate that the 3- compartment body model developed for adults can be applied Measurement of TBW_D2O to help manage fluid status in children on dialysis. In this study, we investigated the agreement of BIS-derived Deuterium oxide (or ‘heavy water’,D O) dilution can provide TBW (i) with fluid volume obtained by deuterium dilution in an accurate measure of TBW [17]. Deuterium is a naturally healthy children and (ii) with the urea distribution volume occurring, stable isotope of hydrogen that is safe for use in (equivalent to TBW) derived from urea kinetic modelling children. To avoid the need for infusions or blood samples, the (UKM) in children receiving haemodialysis. A further objec- D O was taken orally in a drink containing 1 mL of 7% D O 2 2 tive was to check for systematic deviation of the BCM- per/kg/body weight (after emptying the bladder) and the Pediatr Nephrol (2018) 33:1601–1607 1603 children were asked to provide a small (7 mL) urine sample x/y graph. For overhydration, the 10th and 90th percentiles every evening for 6 days starting from the day before taking were also reported. Paired t tests were used, and a p value of the D O drink (the baseline). The D O concentration in the < 0.05 was considered to indicate significance. 2 2 urine samples was analysed by isotope ratio mass spectrome- try (IRMS) [18] in the Medical Research Council Human Nutrition Research Laboratory in Cambridge, UK. D Odis- Results tributes throughout the total body water and the initial distri- bution volume (TBW_D2O) was determined from the mass of Table 1 details age, gender, height, weight and BMI with D O administered and the decline in the concentration of deu- standard deviation scores of the different groups of healthy terium in the urine samples [19]. TBW_D2O was calculated children participating in this analysis. using both the multi-point back-extrapolation method and the two-point plateau method for quality control. The volume Agreement between D O- and BCM-derived total obtained using the multi-point method was used for analysis. body water Measurement of TBW_UKM Sixty-one healthy children received TBW assessments by D O dilution. Sixty children (28 female, median age 10.3) In the haemodialysis cohort, TBW was determined using a were able to provide sufficient urine samples for analysis. modified version of the formal UKM procedure developed TBW_BCM was calculated using unadjusted model as pro- by Sargent and Gotch [20]. For each monitored dialysis ses- vided by BCM Version 3.2 and above. The mean TBW_BCM sion, pre- and post-dialysis serum urea samples were taken. (±SD) was 21.1 ± 5.6 L and that by deuterium dilution was The average urea clearance rate for the session was calculated 20.5 ± 5.8 L. There was good agreement between for the dialyser, flow rates and current haematocrit. Urea gen- TBW_BCM and TBW_D2O (R = 0.97, Fig. 1) with a bias eration was assumed to be constant and residual renal function of + 0.6 L and 95% limits of − 2.0 to +3.2 L (Fig. S1). was neglected. The urea distribution volume, the ‘kinetic’ V, required to give the measured change in serum urea level Agreement between BCM- and UKM-derived total (after correction for rebound) with the recorded treatment time body water in children receiving haemodialysis and calculated clearance was found by iteration. There were no problems with dialysis delivery such as access recirculation Six children between the ages of 4 and 13 years (2 female, or clotting that would have led to an exaggerated kinetic V in median age 10.5 years) were receiving regular haemodialysis any of the monitored sessions. Like D O, urea distributes at the time of the study. Table 2 details age, gender, primary throughout the total body water (giving TBW_UKM). renal diagnosis, months on dialysis, height, BMI, requirement for ultrafiltration and months to transplantation for children on Funding and ethical approval haemodialysis participating in the study. Forty-five concomi- tant measurements of TBW were taken by BCM and UKM The method comparison was funded by a grant from the over an 8-month period. There was good correlation between British Renal Society and the work of one of the main inves- TBW_BCM and TBW_UKM (Fig. 2). The mean difference tigator was supported by the National Institute of Health between the methods was − 0.4 L, 2SD = ± 3.0 L (Fig. S2). Research (NIHR) Devices for Dignity Healthcare The mean difference between individually averaged data (N = Technology Consortium, UK. The Leeds East Local 6) was 0.2 L, range (min to max) = − 1.2 to 1.9 L (Figs. S3 and Research Ethics Committee approved the study protocol and S4). TBW_BCM showed better precision of the individual healthy participants were recruited through the hospital’son- monthly measurements with a mean coefficient of variation line bulletin board. Parents provided informed consent for the (CV=SD/mean) of 3.5% compared to 7.7% for TBW_UKM. study and children ‘assented’ to take part in line with local recommendations. Data collection for the extended cohort in Germany and Sweden was supported by Fresenius Medical Investigation of age-related trends in BCM-measured Care, Bad Homburg, by providing the BCM device. Local OH in healthy children ethics approval was sought at each site. For all healthy children (n = 634) who took part in the method Data analysis comparison described above, the average BCM-measured OH was − 0.1 ± 0.7 L (10th to 90th percentile − 0.8 to + 0.6 L). Agreement between methods was assessed using mean and The OH normalised to extracellular water (OH/ECW, mean ± SD (adjusted for multiple measurements when appropriate) SD) was − 1.0 ± 6.3% (10th to 90th percentile −8.5to+ of paired differences (Bland-Altman analysis), and R in an 6.4%). There was no correlation between age and OH (p = 1604 Pediatr Nephrol (2018) 33:1601–1607 Table 1 Subject characteristics Patient cohort N Age [years] Height [cm] BMI [kg/m ] BMI_SDS stratified by gender and origin [mean ± SD]. BMI is provided in Heidelberg 0.1 ± 1.1 both kg/m and standard Female 180 9.1 ± 3.8 134.4 ± 21.6 17.7 ± 3.7 deviation score Male 168 9.9 ± 4.5 141.4 ± 28.2 18.4 ± 3.9 Kiel 0.0 ± 1.0 Female 65 10.5 ± 2.3 144.7 ± 13.4 18.2 ± 3.3 Male 65 10.5 ± 2.8 146.5 ± 18.1 17.6 ± 3.4 Gothenburg 0.0 ± 1.0 Female 23 12.0 ± 0.8 158.1 ± 8.0 19.4 ± 4.5 Male 36 11.8 ± 0.8 155.1 ± 10.5 17.9 ± 2.4 Fresenius −0.1 ± 0.2 Female 6 12.8 ± 6.6 146.3 ± 29.8 18.7 ± 2.9 Male 1 5.0 123 15.3 Leeds 0.0 ± 0.9 Female 29 10.1 ± 2.5 140.3 ± 14.4 17.8 ± 2.7 Male 31 10.7 ± 2.3 143.4 ± 15.3 17.1 ± 2.2 BMI body mass index, SD standard deviation 0.28), (Fig. 3). The age distribution of all children under in- to − 1.4 ± 6.2% relative to ECW (10th to 90th percentile − 8.6 vestigation is shown in Fig. S5. to 5.9%). Again, there was no systematic variation in OH For the subset of healthy children from Leeds (n =60), the measured by BCM with age (p = 0.28) (Fig. S6). BCM-measured OH (mean ± SD) was 0.2 ± 0.5 L (10th to 90th percentile −0.4to+0.9L). TheOH/ECW (mean± SD) was + 2.3% ± 5.8 (10th to 90th percentile − 5.0 to Discussion 9.9%). There was no correlation between age and OH in this subset (p =0.25). Bioimpedance measurements are very sensitive to the amount The cohort from Germany and Sweden included 574 and distribution of fluid in tissue, allowing objective assess- healthy children (349 from Heidelberg, 130 from Kiel, 34 ment of fluid status. Several bioimpedance-based techniques from Bad Homburg and 61 from Gothenburg) with a median have been used in studies of children on haemodialysis. age of 11 years. Two hundred eighty-six (49.8%) were female. Changes in whole-body impedance at 50 kHz reflect changes The OH (mean ± SD) was − 0.1 ± 0.7 L which corresponded in body water volume during dialysis [21]. The reactance component of this measurement (Xc) can help identify dry weight [22] and the change in the resistance component (R) with fluid removal correlates with intradialytic hypotension and left ventricular mass index [23]. The variability in blood pressure and heart rate during dialysis can be partially ex- plained by vector analysis [24]. Despite its obvious application potential in dialysis, the use of bioimpedance has been largely restricted to research studies due to difficulties in interpreting the data and problems apply- ing predictive equations derived in healthy subjects in patients with abnormal fluid status. The combination of BIS with the 3-compartment body model in the BCM can provide users with a straight-forward assessment of how far a patient is from their normally hydrated weight. The fluid volume model used to derive ECW and ICW from BIS data includes parameters related to the resistivity of intracellular and extracellular water and body shape, while Fig. 1 TBW BCM vs. TBW D20 in 60 healthy children aged between 6 the 3-compartment model relies on ‘hydration parameters’ and 14 years. Corresponding Bland-Altman plot is shown in (i.e. the proportion of ICW and ECW per kg of lean and 2 2 Supplementary Fig. S1. All subjects R =0.97 (p <0.001), Girls R = adipose tissue). For the BCM to be used in subjects that differ 0.97 (p < 0.001) and Boys: R =0.98 (p <0.001). BCM Body Composition Monitor, D O deuterium oxide, TBW total body water significantly from the reference population, the assumption 2 Pediatr Nephrol (2018) 33:1601–1607 1605 Table 2 Characteristics of haemodialysis cohort Patient Gender Primary renal diagnosis Age Months Height BMI [kg/m ] UF required Months [years] on HD [cm] on HD to tx 1 M Congenital renal dysplasia 4 38 89 18.9 Yes 8 2 M Congenital renal dysplasia 6 6 107 20.5 No 14 3 M Atypical haemolytic uraemic syndrome 10 3 137 14.6 Yes 6 4 F Congenital renal dysplasia 11 1 139 20.5 No 4 5 M Nephronophthisis 12 18 140 16.7 Yes 15 6 F Unknown cause 13 9 154 20.3 Yes 6 HD haemodialysis, BMI body mass index, UF ultrafiltration, Tx kidney transplantation Please note the age, height, BMI and time on HD are at the start of the study that the hydration parameters do not require modification sampling. Urea kinetic modelling was used to calculate the needs to be tested [25]. Dialysed children are a particularly urea distribution volume as an equivalent to TBW in a cohort critical population in this regard; differences in model param- of children undergoing haemodialysis. Again, we saw good eters in growing and maturing children would be expected to agreement between the BCM- and UKM-derived TBW esti- impact the agreement between BCM-measured parameters mates without systematic differences across an age range of 4 and measurements using other techniques. to 13 years. The repeated measurements showed superior re- To assess the validity of the fluid model, we compared the producibility of BCM-derived TBW estimates as compared to TBW estimates by BCM with D O dilution-derived measure- UKM-derived values, which is not surprising considering the ments in a cohort of healthy children whose BMI distribution simplicity of bioimpedance measurements as compared to the was consistent with that of a large German cohort reported substantial biological and technical variation associated with a previously [26]. There was good agreement between BCM- methodology requiring repeated blood sampling and labora- and D O-derived TBW readings without any systematic dif- tory measurements. The advantage of simplicity was also ferences related to age, suggesting these findings are likely to emphasised in a recent study comparing fluid status in chil- be generalisable. dren with nephrotic syndrome by bioimpedance and echocar- The deuterium dilution method used was not suitable for diography [27]. children on haemodialysis for it required serial urine The validity of the 3-compartment model for fluid overload was assessed in healthy children with an assumed normal state of hydration, i.e. only small differences between actual body weight and the normally hydrated weight (the sum of the normally hydrated lean and adipose tissue) reported by BCM. Such differences will be reported by BCM as ‘overhydration’ since the device is primarily designed to de- tect and quantify excess fluid in the extracellular compart- ment. The apparent OH reported by BCM in healthy children was small, without any systematic age-related variation from age 2 to 17 years. Furthermore, the relative overhydration observed in the cohort of healthy children (OH/ECW, mean ± SD) was − 1.0 ± 6.3% (10th to 90th percentile − 8.5 to + 6.4%), showing similar variation to adult healthy controls (10th to 90th percentile −8to 8%) [28]. While this study does not provide a rigorous validation of the models used by the BCM to determine body water volumes and normally hydrated tissue mass in children, Fig. 2 TBW_BCM vs. TBW_UKM in 6 children on haemodialysis. our findings indicate that the models developed for use Measurements were made over 8 months. Corresponding Bland-Altman in adults can also be applied in children. Further justifica- plot is shown in Supplementary Fig. S2. The same data but individually tion for implementing the BCM to help the management averaged is shown in Supplementary Figs. S3 and S4. Dashed line indi- of fluid status in paediatric dialysis patients will come cates line of identity. TBW total body water, BCM Body Composition Monitor, UKM Urea Kinetic Modelling with practical experience. 1606 Pediatr Nephrol (2018) 33:1601–1607 Fig. 3 BCM-measured OH (a) and OH/ ECW (b) in all healthy children whiskers are 1.5*interquartile range (covering 99.3% of data assuming (n = 634). Boxes indicate interquartile range from 25th to 75th percentile, normal distribution), crosses = outliers. BCM Body Composition line is the median, notches indicate 95% confidence interval for median, Monitor, OH over hydration, ECW extra cellular water It is important to note that the BCM provides an estimate of measurements are highly reproducible, allowing changes in apatient’s normally hydrated weight, which is not necessarily lean body mass to be distinguished from changes in fluid identical with the actual target weight due to several con- status and adjustments in target weight to be made before founding factors, including variations in the length of limbs symptoms occur. Further studies are required to demonstrate relative to the trunk, body temperature, electrolyte levels and the benefits of using BCM in guiding fluid management and the fluid and electrolyte content of the gut, that vary between improving clinical outcomes in this vulnerable population. measurements. Whereas the use of BCM cannot replace good Acknowledgements We are grateful to Antony Wright and Les Bluck clinical judgement, it can help prevent potentially harmful (who sadly died in 2014) for their advice on measuring total body water adjustments in target weight based on deceptive signs and in children and their independent analysis of the data, which was made symptoms. In addition, serial BCM readings allow changes possible by MRC Programme Funding (Physiological Modelling and in fluid status to be monitored accurately, provided a consis- Metabolic Risk: MC_UP_A090_1005). We are also grateful to Fresenius Medical Care for providing the tent measurement protocol is used [29]. equipment, consumables and technical support for the study. The added value of BCM to blood pressure monitoring was impressively documented in a study of 463 dialysis sessions in Compliance with ethical standards 23 haemodialysed children [30]. Hypertension was present in 39% of dialysis sessions, of which only 31% were associated Conflict of interest EL received honoraria from Fresenius Medical Care with moderate to severe BCM-measured OH. The authors for providing training in the use of bioimpedance spectroscopy in renal concluded that hypertension is not always related to care. PC, PW and UM are employees of Fresenius Medical Care. overhydration, and that use of BCM could avoid inappropriate ‘dry weight probing’ in patients with volume-independent hy- pertension. Furthermore, a recent study in children attempted Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http:// to determine the clinical utility of BCM-measured fluid status creativecommons.org/licenses/by/4.0/), which permits unrestricted use, by comparing it to clinical assessment and cardiovascular in- distribution, and reproduction in any medium, provided you give dicators. Assessment of fluid status based on clinical assess- appropriate credit to the original author(s) and the source, provide a link ment was shown to be misleading; BCM measurements cor- to the Creative Commons license, and indicate if changes were made. related with established biomarkers and cardiovascular mea- The results presented in this paper have not been published previously in sures [31]. whole or part, except in abstract format. Conclusion References Our study suggests that the BCM can be used in children as 1. Hecking M, Karaboyas A, Antlanger M, Saran R, Wizemann V, young as 2 years, at least for measurements of normally hy- Chazot C, Rayner H, Hörl WH, Pisoni RL, Robinson BM, Sunder- drated weight and fluid overload. The measurements are non- Plassmann G, Moissl U (2013) Significance of interdialytic weight invasive, well tolerated, inexpensive and easy to perform. If gain versus chronic volume overload: consensus opinion. Am J carried out according to a standardised protocol [29], BCM Nephrol 38:78–90 Pediatr Nephrol (2018) 33:1601–1607 1607 2. Weiner DE, Brunelli SM, Hunt A, Schiller B, Glassock R, Maddux 17. van Marken Lichtenbelt WD, Westerterp KR, Wouters L (1994) Deuterium dilution as a method for determining total body water: FW, Johnson D, Parker T, Nissenson A (2014) Improving clinical outcomes among hemodialysis patients: a proposal for a Bvolume effect of test protocol and sampling time. Br J Nutr 72:491–497 first^ approach from the chief medical officers of US dialysis pro- 18. Hilkert AW, Douthitt CB, Schlüter HJ, Brand WA (1999) Isotope viders. Am J Kidney Dis 64:685–695 ratio monitoring gas chromatography/mass spectrometry of D/H by 3. Wizemann V, Wabel P, Chamney P, Zaluska W, Moissl U, Rode C, high temperature conversion isotope ratio mass spectrometry. Rapid Malecka-Masalska T, Marcelli D (2009) The mortality risk of Commun Mass Spectrom 13:1226–1230 overhydration in haemodialysis patients. Nephrol Dial Transplant 19. Coward WA. Calculation of pool sizes and flux rates. In: Prentice 24:1574–1579 AM, eds. The doubly-labelled water method for measuring energy 4. Agarwal R (2010) Hypervolemia is associated with increased mor- expenditure. Technical recommendations for use in humans. tality among hemodialysis patients. Hypertension 56:512–517 Vienna: International Dietary Energy (http://archive.unu.edu/ 5. Chazot C, Wabel P, Chamney P, Moissl U, Wieskotten S, unupress/food2/UID05E/UID05E00.HTM, Table of Contents - Wizemann V (2012) Importance of normohydration for the long- Chapter 4. Accessed 27 Mar 2018) term survival of haemodialysis patients. Nephrol Dial Transplant 20. Sargent JA, Gotch FA (1980) Mathematic modeling of dialysis 27:2404–2410 therapy. Kidney Int 18(Suppl. 10):S2–S10 6. Stefánsson BV, Brunelli SM, Cabrera C, Rosenbaum D, Anum E, 21. Bradbury MG, Smye SW, Brocklebank JT (1995) Assessment of Ramakrishnan K, Jensen DE, Stålhammar NO (2014) Intradialytic the sensitivity of bioimpedance to volume changes in body water. hypotension and risk of cardiovascular disease. Clin J Am Soc Pediatr Nephrol 9:337–340 Nephrol 9:2124–2132 22. Paglialonga F, Ardissino G, Galli MA, Scarfia RV, Testa S, Edefonti 7. Flythe JE, Xue H, Lynch KE, Curhan GC, Brunelli SM (2015) A (2012) Bioimpedance analysis and cardiovascular status in pedi- Association of mortality risk with various definitions of atric patients on chronic hemodialysis. Hemodial Int 16(Suppl 1): intradialytic hypotension. J Am Soc Nephrol 26:724–734 S20–S25 8. Termorshuizen F, Dekker FW, van Manen JG, Korevaar JC, 23. Oh G, Wong C, Begin B, Salsbery K, Sutherland S, Chaudhuri A Boeschoten EW, Krediet RT, NECOSAD Study Group (2004) (2014) Whole-body single-frequency bioimpedance analysis in pe- Relative contribution of residual renal function and different mea- diatric hemodialysis patients. Pediatr Nephrol 29:1417–1423 sures of adequacy to survival in hemodialysis patients: an analysis 24. Brooks ER, Fatallah-Shaykh SA, Langman CB, Wolf KM, Price of the Netherlands Cooperative Study on the Adequacy of Dialysis HE (2008) Bioelectric impedance predicts total body water, blood (NECOSAD)-2. J Am Soc Nephrol 15:1061–1070 pressure, and heart rate during hemodialysis in children and adoles- 9. Vilar E, Farrington K (2011) Emerging importance of residual renal cents. J Ren Nutr 18:304–311 function in end-stage renal failure. Semin Dial 24:487–494 25. Lindley EJ, Lopot F (2015) The use of bioimpedance to aid volume 10. Agarwal R, Weir MR (2010) Dry-weight: a concept revisited in an assessment in dialysis patients. Kidney Int 87:240 effort to avoid medication-directed approaches for blood pressure con- 26. Kromeyer-Hauschild K, Wabitsch M, Kunze D, Geller F, Geiß HC, trol in hemodialysis patients. Clin J Am Soc Nephrol 5:1255–1260 Hesse V, von Hippel A, Jaeger U, Johnsen D, Korte W, Menner K, 11. Moissl UM, Wabel P, Chamney PW Bosaeus I, Levin NW, Bosy- Müller G, Müller JM, Niemann-Pilatus A, Remer T, Schaefer F, Westphal A, Korth O, Müller MJ, Ellegård L, Malmros V, Wittchen HU, Zabransky S, Zellner K, Ziegler A, Hebebrand J Kaitwatcharachai C, Kuhlmann MK, Zhu F, Fuller NJ (2006) (2001) [Percentile for the body mass index for children and adoles- Body fluid volume determination via body composition spectros- cents using various German samples]. Monatsschr Kinderheilkd copy in health and disease. Physiol Meas 27:921–933 149:807–818 12. Chamney PW, Wabel P, Moissl UM Müller MJ, Bosy-Westphal A, 27. Özdemir K, Mir MS, Dinçel N, Bozabali S, Kaplan Bulut İ, Yilmaz Korth O, Fuller NJ (2007) A whole-body model to distinguish ex- E, Sözeri B (2015) Bioimpedance for assessing volume status in cess fluid from the hydration of major body tissues. Am J Clin Nutr children with nephrotic syndrome. Turk J Med Sci 45:339–344 85:80–89 28. http://www.bcm-fresenius.de/files/information_on_reference_ 13. Wabel P, Chamney P, Moissl U, Jirka T (2009) Importance of ranges.pdf (Accessed 18 Feb 2018) whole-body bioimpedance spectroscopy for the management of 29. Lindley EJ, Keane DF, Schneditz D (2016) Comparison of fluid balance. Blood Purif 27:75–80 intradialytic changes in weight and fluid status. Nephrology 21:632 14. Passauer J, Petrov H, Schleser A, Leicht J, Pucalka K (2010) 30. Zaloszyc A, Schaefer B, Schaefer F, Krid S, Salomon R, Niaudet P, Evaluation of clinical dry weight assessment in haemodialysis pa- Schmitt CP, Fischbach M (2013) Hydration measurement by tients using bioimpedance spectroscopy: a cross-sectional study. bioimpedance spectroscopy and blood pressure management in Nephrol Dial Transplant 25:545–551 children on hemodialysis. Pediatr Nephrol 28:2169–2177 15. Davies SJ, Davenport A (2014) The role of bioimpedance and bio- 31. Eng CSY, Bhowruth D, Mayes M, Stronach L, Blaauw M, Barber markers in helping to aid clinical decision-making of volume as- A, Rees L, Shroff RC (2017) Assessing the hydration status of sessments in dialysis patients. Kidney Int 86:489–496 children with chronic kidney disease and on dialysis: a comparison 16. Cole KS, Cole RH (1941) Dispersion and absorption in dielectrics: of techniques. Nephrol Dial Transplant. https://doi.org/10.1093/ndt/ 1. Alternating current characteristics. J Chem Phys 9:341–351 gfx287 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Pediatric Nephrology Springer Journals
Loading next page...
 
/lp/springer_journal/validating-the-use-of-bioimpedance-spectroscopy-for-assessment-of-nRKjsDG2rH
Publisher
Springer Berlin Heidelberg
Copyright
Copyright © 2018 by The Author(s)
Subject
Medicine & Public Health; Pediatrics; Nephrology; Urology
ISSN
0931-041X
eISSN
1432-198X
D.O.I.
10.1007/s00467-018-3971-x
Publisher site
See Article on Publisher Site

Abstract

Background Bioimpedance spectroscopy (BIS) with a whole-body model to distinguish excess fluid from major body tissue hydration can provide objective assessment of fluid status. BIS is integrated into the Body Composition Monitor (BCM) and is validated in adults, but not children. This study aimed to (1) assess agreement between BCM-measured total body water (TBW) and a gold standard technique in healthy children, (2) compare TBW_BCM with TBW from Urea Kinetic Modelling (UKM) in haemodialysis children and (3) investigate systematic deviation from zero in measured excess fluid in healthy children across paediatric age range. Methods TBW_BCM and excess fluid was determined from standard wrist-to-ankle BCM measurement. TBW_D2O was determined from deuterium concentration decline in serial urine samples over 5 days in healthy children. UKM was used to measure body water in children receiving haemodialysis. Agreement between methods was analysed using paired t test and Bland-Altman method comparison. Results In 61 healthy children (6–14 years, 32 male), mean TBW_BCM and TBW_D2O were 21.1 ± 5.6 and 20.5 ± 5.8 L respectively. There was good agreement between TBW_BCM and TBW_D2O (R = 0.97). In six haemodialysis children (4–13 years, 4 male), 45 concomitant measurements over 8 months showed good TBW_BCM and TBW_UKM agree- ment (mean difference − 0.4 L, 2SD = ± 3.0 L). In 634 healthy children (2–17 years, 300 male), BCM-measured overhydration was − 0.1 ± 0.7 L (10–90th percentile − 0.8 to + 0.6 L). There was no correlation between age and OH (p =0.28). Conclusions These results suggest BCM can be used in children as young as 2 years to measure normally hydrated weight and assess fluid status. . . . . . . Keywords Fluid volume Bioimpedance Chronic kidney disease Overhydration Total body water Children Haemodialysis Electronic supplementary material The online version of this article (https://doi.org/10.1007/s00467-018-3971-x) contains supplementary material, which is available to authorized users. * Indranil Dasgupta Pediatric Nephrology Division, Center for Pediatrics and Adolescent Indranil.Dasgupta@heartofengland.nhs.uk Medicine, University of Heidelberg, Heidelberg, Germany Institute for Human Nutrition and Food Science, Christian-Albrecht University, Kiel, Germany Heartlands Hospital and University of Birmingham, Birmingham, UK Universität Hohenheim, Hohenheim, Germany Departments of Renal Medicine and Medical Physics, Leeds Department of Pediatrics, Institute of Clinical Sciences, Sahlgrenska Teaching Hospitals NHS Trust, Leeds, UK Academy at University of Gothenburg, Gothenburg, Sweden Department of Children’s Nephrology, Leeds Teaching Hospitals Global R&D, Fresenius Medical Care, Bad Homburg, Germany NHS Trust, Leeds, UK 1602 Pediatr Nephrol (2018) 33:1601–1607 Introduction measured OH from zero in healthy children across the paedi- atric age range. Fluid management in haemodialysis impacts patient experi- ence, morbidity and mortality [1, 2]. While inadequate fluid removal can lead to oedema and may precipitate heart failure, Methods volume depletion can cause hypotension, dizziness, cramps, abdominal symptoms, prolonged recovery time following Study participants haemodialysis and accelerated loss of residual renal function. An increased risk of morbidity and mortality has been associ- For the method comparisons, healthy children aged 6 to ated both with chronic fluid overload [3–5]and with 14 years related to hospital staff and children receiving intradialytic hypotension and loss of residual function [6–9]. haemodialysis at the Paediatric Nephrology Department of Determination of optimal fluid status in a dialysis patient is Leeds Teaching Hospitals NHS Trust UK were recruited. For challenging [10]. Conventionally, the assessment of fluid status the investigation of age-related changes in BCM-measured is based on clinical symptoms and signs. Progressive reduction OH, a large cohort of healthy children aged 2 to 17 years from of target weight until the patient becomes symptomatic is de- Germany and Sweden was studied. Children with limb ampu- scribed as ‘probing for dry weight’. Assessment of fluid status is tations, cardiac or other chronic diseases were excluded. even more challenging in children who often have rapid changes in flesh weight. A number of technologies are now available to Measurement of TBW_BCM and OH aid fluid management in renal patients, of which bioimpedance spectroscopy (BIS) is one of the most widely studied. BCM measurements to obtain TBW (TBW_BCM) and OH Whole-body (wrist-to-ankle) BIS measurements can be were made using standard whole-body electrode configura- used to determine extracellular, intracellular and total body tion after the child had been lying down for at least 5 min. water (ECW, ICW and TBW) volumes [11]. The addition of TBW_BCM in healthy children was measured once on the a 3-compartment body model to distinguish excess extracel- dominant side. In children on haemodialysis, measurements lular fluid from lean and adipose tissue allows an objective were performed on the dominant or non-fistula side before the assessment of excess fluid or ‘overhydration’ (OH) [12]. The start of dialysis once a month for up to 8 months. BIS fluid volume model and the body model are integrated The BCM measures whole-body impedance over 50 fre- into the Body Composition Monitor (BCM, Fresenius AG, quencies (from 5 kHz to 1 MHz) and determines extracellular Bad Homburg), which has been well validated [13] and shown and total body resistance by Cole modelling [16]inorder to to be associated with better survival in adult dialysis patients estimate ECW and ICW using the fluid volume model [11]. [3, 5]. The BCM is increasingly being used to assist the man- The 3-compartment body composition model uses these vol- agement of adult dialysis patients [14, 15]. For paediatric pa- umes to separate the body weight into normally hydrated lean tients to benefit from this technology, evidence of the applica- tissue mass (LTM), normally hydrated adipose tissue mass bility of the underlying models in children is required. (ATM) and excess fluid (or overhydration, OH) [12]. As OH The extensive validation of the BCM models against gold is simply the discrepancy between the actual body weight and standards that has been carried out in adults is impractical in the normally hydrated weight (LTM + ATM), it can be posi- children. However, it is possible to measure TBW in both tive or negative. OH is typically within a range of − 1.1 to + healthy subjects and haemodialysis patients using child- 1.1 L (10th to 90th percentile) in healthy adults without car- friendly techniques, against which the TBW estimated by the diac or renal complications [14]. However, in both healthy BCM fluid volume model can be assessed. While there is no subjects and dialysis patients, the deviation from normal hy- gold standard for quantifying excess extracellular fluid, it dration represented by a BCM-measured OH of 1.1L scales should be absent in healthy subjects. BCM OH measurements with the size of the individual. To compare individuals of that are close to zero, and without systematic variation with different size, OH is normalised to ECW as excess fluid pri- age, in healthy children would provide further confirmation of marily accumulates in the extracellular space. the validity of the fluid volume model and indicate that the 3- compartment body model developed for adults can be applied Measurement of TBW_D2O to help manage fluid status in children on dialysis. In this study, we investigated the agreement of BIS-derived Deuterium oxide (or ‘heavy water’,D O) dilution can provide TBW (i) with fluid volume obtained by deuterium dilution in an accurate measure of TBW [17]. Deuterium is a naturally healthy children and (ii) with the urea distribution volume occurring, stable isotope of hydrogen that is safe for use in (equivalent to TBW) derived from urea kinetic modelling children. To avoid the need for infusions or blood samples, the (UKM) in children receiving haemodialysis. A further objec- D O was taken orally in a drink containing 1 mL of 7% D O 2 2 tive was to check for systematic deviation of the BCM- per/kg/body weight (after emptying the bladder) and the Pediatr Nephrol (2018) 33:1601–1607 1603 children were asked to provide a small (7 mL) urine sample x/y graph. For overhydration, the 10th and 90th percentiles every evening for 6 days starting from the day before taking were also reported. Paired t tests were used, and a p value of the D O drink (the baseline). The D O concentration in the < 0.05 was considered to indicate significance. 2 2 urine samples was analysed by isotope ratio mass spectrome- try (IRMS) [18] in the Medical Research Council Human Nutrition Research Laboratory in Cambridge, UK. D Odis- Results tributes throughout the total body water and the initial distri- bution volume (TBW_D2O) was determined from the mass of Table 1 details age, gender, height, weight and BMI with D O administered and the decline in the concentration of deu- standard deviation scores of the different groups of healthy terium in the urine samples [19]. TBW_D2O was calculated children participating in this analysis. using both the multi-point back-extrapolation method and the two-point plateau method for quality control. The volume Agreement between D O- and BCM-derived total obtained using the multi-point method was used for analysis. body water Measurement of TBW_UKM Sixty-one healthy children received TBW assessments by D O dilution. Sixty children (28 female, median age 10.3) In the haemodialysis cohort, TBW was determined using a were able to provide sufficient urine samples for analysis. modified version of the formal UKM procedure developed TBW_BCM was calculated using unadjusted model as pro- by Sargent and Gotch [20]. For each monitored dialysis ses- vided by BCM Version 3.2 and above. The mean TBW_BCM sion, pre- and post-dialysis serum urea samples were taken. (±SD) was 21.1 ± 5.6 L and that by deuterium dilution was The average urea clearance rate for the session was calculated 20.5 ± 5.8 L. There was good agreement between for the dialyser, flow rates and current haematocrit. Urea gen- TBW_BCM and TBW_D2O (R = 0.97, Fig. 1) with a bias eration was assumed to be constant and residual renal function of + 0.6 L and 95% limits of − 2.0 to +3.2 L (Fig. S1). was neglected. The urea distribution volume, the ‘kinetic’ V, required to give the measured change in serum urea level Agreement between BCM- and UKM-derived total (after correction for rebound) with the recorded treatment time body water in children receiving haemodialysis and calculated clearance was found by iteration. There were no problems with dialysis delivery such as access recirculation Six children between the ages of 4 and 13 years (2 female, or clotting that would have led to an exaggerated kinetic V in median age 10.5 years) were receiving regular haemodialysis any of the monitored sessions. Like D O, urea distributes at the time of the study. Table 2 details age, gender, primary throughout the total body water (giving TBW_UKM). renal diagnosis, months on dialysis, height, BMI, requirement for ultrafiltration and months to transplantation for children on Funding and ethical approval haemodialysis participating in the study. Forty-five concomi- tant measurements of TBW were taken by BCM and UKM The method comparison was funded by a grant from the over an 8-month period. There was good correlation between British Renal Society and the work of one of the main inves- TBW_BCM and TBW_UKM (Fig. 2). The mean difference tigator was supported by the National Institute of Health between the methods was − 0.4 L, 2SD = ± 3.0 L (Fig. S2). Research (NIHR) Devices for Dignity Healthcare The mean difference between individually averaged data (N = Technology Consortium, UK. The Leeds East Local 6) was 0.2 L, range (min to max) = − 1.2 to 1.9 L (Figs. S3 and Research Ethics Committee approved the study protocol and S4). TBW_BCM showed better precision of the individual healthy participants were recruited through the hospital’son- monthly measurements with a mean coefficient of variation line bulletin board. Parents provided informed consent for the (CV=SD/mean) of 3.5% compared to 7.7% for TBW_UKM. study and children ‘assented’ to take part in line with local recommendations. Data collection for the extended cohort in Germany and Sweden was supported by Fresenius Medical Investigation of age-related trends in BCM-measured Care, Bad Homburg, by providing the BCM device. Local OH in healthy children ethics approval was sought at each site. For all healthy children (n = 634) who took part in the method Data analysis comparison described above, the average BCM-measured OH was − 0.1 ± 0.7 L (10th to 90th percentile − 0.8 to + 0.6 L). Agreement between methods was assessed using mean and The OH normalised to extracellular water (OH/ECW, mean ± SD (adjusted for multiple measurements when appropriate) SD) was − 1.0 ± 6.3% (10th to 90th percentile −8.5to+ of paired differences (Bland-Altman analysis), and R in an 6.4%). There was no correlation between age and OH (p = 1604 Pediatr Nephrol (2018) 33:1601–1607 Table 1 Subject characteristics Patient cohort N Age [years] Height [cm] BMI [kg/m ] BMI_SDS stratified by gender and origin [mean ± SD]. BMI is provided in Heidelberg 0.1 ± 1.1 both kg/m and standard Female 180 9.1 ± 3.8 134.4 ± 21.6 17.7 ± 3.7 deviation score Male 168 9.9 ± 4.5 141.4 ± 28.2 18.4 ± 3.9 Kiel 0.0 ± 1.0 Female 65 10.5 ± 2.3 144.7 ± 13.4 18.2 ± 3.3 Male 65 10.5 ± 2.8 146.5 ± 18.1 17.6 ± 3.4 Gothenburg 0.0 ± 1.0 Female 23 12.0 ± 0.8 158.1 ± 8.0 19.4 ± 4.5 Male 36 11.8 ± 0.8 155.1 ± 10.5 17.9 ± 2.4 Fresenius −0.1 ± 0.2 Female 6 12.8 ± 6.6 146.3 ± 29.8 18.7 ± 2.9 Male 1 5.0 123 15.3 Leeds 0.0 ± 0.9 Female 29 10.1 ± 2.5 140.3 ± 14.4 17.8 ± 2.7 Male 31 10.7 ± 2.3 143.4 ± 15.3 17.1 ± 2.2 BMI body mass index, SD standard deviation 0.28), (Fig. 3). The age distribution of all children under in- to − 1.4 ± 6.2% relative to ECW (10th to 90th percentile − 8.6 vestigation is shown in Fig. S5. to 5.9%). Again, there was no systematic variation in OH For the subset of healthy children from Leeds (n =60), the measured by BCM with age (p = 0.28) (Fig. S6). BCM-measured OH (mean ± SD) was 0.2 ± 0.5 L (10th to 90th percentile −0.4to+0.9L). TheOH/ECW (mean± SD) was + 2.3% ± 5.8 (10th to 90th percentile − 5.0 to Discussion 9.9%). There was no correlation between age and OH in this subset (p =0.25). Bioimpedance measurements are very sensitive to the amount The cohort from Germany and Sweden included 574 and distribution of fluid in tissue, allowing objective assess- healthy children (349 from Heidelberg, 130 from Kiel, 34 ment of fluid status. Several bioimpedance-based techniques from Bad Homburg and 61 from Gothenburg) with a median have been used in studies of children on haemodialysis. age of 11 years. Two hundred eighty-six (49.8%) were female. Changes in whole-body impedance at 50 kHz reflect changes The OH (mean ± SD) was − 0.1 ± 0.7 L which corresponded in body water volume during dialysis [21]. The reactance component of this measurement (Xc) can help identify dry weight [22] and the change in the resistance component (R) with fluid removal correlates with intradialytic hypotension and left ventricular mass index [23]. The variability in blood pressure and heart rate during dialysis can be partially ex- plained by vector analysis [24]. Despite its obvious application potential in dialysis, the use of bioimpedance has been largely restricted to research studies due to difficulties in interpreting the data and problems apply- ing predictive equations derived in healthy subjects in patients with abnormal fluid status. The combination of BIS with the 3-compartment body model in the BCM can provide users with a straight-forward assessment of how far a patient is from their normally hydrated weight. The fluid volume model used to derive ECW and ICW from BIS data includes parameters related to the resistivity of intracellular and extracellular water and body shape, while Fig. 1 TBW BCM vs. TBW D20 in 60 healthy children aged between 6 the 3-compartment model relies on ‘hydration parameters’ and 14 years. Corresponding Bland-Altman plot is shown in (i.e. the proportion of ICW and ECW per kg of lean and 2 2 Supplementary Fig. S1. All subjects R =0.97 (p <0.001), Girls R = adipose tissue). For the BCM to be used in subjects that differ 0.97 (p < 0.001) and Boys: R =0.98 (p <0.001). BCM Body Composition Monitor, D O deuterium oxide, TBW total body water significantly from the reference population, the assumption 2 Pediatr Nephrol (2018) 33:1601–1607 1605 Table 2 Characteristics of haemodialysis cohort Patient Gender Primary renal diagnosis Age Months Height BMI [kg/m ] UF required Months [years] on HD [cm] on HD to tx 1 M Congenital renal dysplasia 4 38 89 18.9 Yes 8 2 M Congenital renal dysplasia 6 6 107 20.5 No 14 3 M Atypical haemolytic uraemic syndrome 10 3 137 14.6 Yes 6 4 F Congenital renal dysplasia 11 1 139 20.5 No 4 5 M Nephronophthisis 12 18 140 16.7 Yes 15 6 F Unknown cause 13 9 154 20.3 Yes 6 HD haemodialysis, BMI body mass index, UF ultrafiltration, Tx kidney transplantation Please note the age, height, BMI and time on HD are at the start of the study that the hydration parameters do not require modification sampling. Urea kinetic modelling was used to calculate the needs to be tested [25]. Dialysed children are a particularly urea distribution volume as an equivalent to TBW in a cohort critical population in this regard; differences in model param- of children undergoing haemodialysis. Again, we saw good eters in growing and maturing children would be expected to agreement between the BCM- and UKM-derived TBW esti- impact the agreement between BCM-measured parameters mates without systematic differences across an age range of 4 and measurements using other techniques. to 13 years. The repeated measurements showed superior re- To assess the validity of the fluid model, we compared the producibility of BCM-derived TBW estimates as compared to TBW estimates by BCM with D O dilution-derived measure- UKM-derived values, which is not surprising considering the ments in a cohort of healthy children whose BMI distribution simplicity of bioimpedance measurements as compared to the was consistent with that of a large German cohort reported substantial biological and technical variation associated with a previously [26]. There was good agreement between BCM- methodology requiring repeated blood sampling and labora- and D O-derived TBW readings without any systematic dif- tory measurements. The advantage of simplicity was also ferences related to age, suggesting these findings are likely to emphasised in a recent study comparing fluid status in chil- be generalisable. dren with nephrotic syndrome by bioimpedance and echocar- The deuterium dilution method used was not suitable for diography [27]. children on haemodialysis for it required serial urine The validity of the 3-compartment model for fluid overload was assessed in healthy children with an assumed normal state of hydration, i.e. only small differences between actual body weight and the normally hydrated weight (the sum of the normally hydrated lean and adipose tissue) reported by BCM. Such differences will be reported by BCM as ‘overhydration’ since the device is primarily designed to de- tect and quantify excess fluid in the extracellular compart- ment. The apparent OH reported by BCM in healthy children was small, without any systematic age-related variation from age 2 to 17 years. Furthermore, the relative overhydration observed in the cohort of healthy children (OH/ECW, mean ± SD) was − 1.0 ± 6.3% (10th to 90th percentile − 8.5 to + 6.4%), showing similar variation to adult healthy controls (10th to 90th percentile −8to 8%) [28]. While this study does not provide a rigorous validation of the models used by the BCM to determine body water volumes and normally hydrated tissue mass in children, Fig. 2 TBW_BCM vs. TBW_UKM in 6 children on haemodialysis. our findings indicate that the models developed for use Measurements were made over 8 months. Corresponding Bland-Altman in adults can also be applied in children. Further justifica- plot is shown in Supplementary Fig. S2. The same data but individually tion for implementing the BCM to help the management averaged is shown in Supplementary Figs. S3 and S4. Dashed line indi- of fluid status in paediatric dialysis patients will come cates line of identity. TBW total body water, BCM Body Composition Monitor, UKM Urea Kinetic Modelling with practical experience. 1606 Pediatr Nephrol (2018) 33:1601–1607 Fig. 3 BCM-measured OH (a) and OH/ ECW (b) in all healthy children whiskers are 1.5*interquartile range (covering 99.3% of data assuming (n = 634). Boxes indicate interquartile range from 25th to 75th percentile, normal distribution), crosses = outliers. BCM Body Composition line is the median, notches indicate 95% confidence interval for median, Monitor, OH over hydration, ECW extra cellular water It is important to note that the BCM provides an estimate of measurements are highly reproducible, allowing changes in apatient’s normally hydrated weight, which is not necessarily lean body mass to be distinguished from changes in fluid identical with the actual target weight due to several con- status and adjustments in target weight to be made before founding factors, including variations in the length of limbs symptoms occur. Further studies are required to demonstrate relative to the trunk, body temperature, electrolyte levels and the benefits of using BCM in guiding fluid management and the fluid and electrolyte content of the gut, that vary between improving clinical outcomes in this vulnerable population. measurements. Whereas the use of BCM cannot replace good Acknowledgements We are grateful to Antony Wright and Les Bluck clinical judgement, it can help prevent potentially harmful (who sadly died in 2014) for their advice on measuring total body water adjustments in target weight based on deceptive signs and in children and their independent analysis of the data, which was made symptoms. In addition, serial BCM readings allow changes possible by MRC Programme Funding (Physiological Modelling and in fluid status to be monitored accurately, provided a consis- Metabolic Risk: MC_UP_A090_1005). We are also grateful to Fresenius Medical Care for providing the tent measurement protocol is used [29]. equipment, consumables and technical support for the study. The added value of BCM to blood pressure monitoring was impressively documented in a study of 463 dialysis sessions in Compliance with ethical standards 23 haemodialysed children [30]. Hypertension was present in 39% of dialysis sessions, of which only 31% were associated Conflict of interest EL received honoraria from Fresenius Medical Care with moderate to severe BCM-measured OH. The authors for providing training in the use of bioimpedance spectroscopy in renal concluded that hypertension is not always related to care. PC, PW and UM are employees of Fresenius Medical Care. overhydration, and that use of BCM could avoid inappropriate ‘dry weight probing’ in patients with volume-independent hy- pertension. Furthermore, a recent study in children attempted Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http:// to determine the clinical utility of BCM-measured fluid status creativecommons.org/licenses/by/4.0/), which permits unrestricted use, by comparing it to clinical assessment and cardiovascular in- distribution, and reproduction in any medium, provided you give dicators. Assessment of fluid status based on clinical assess- appropriate credit to the original author(s) and the source, provide a link ment was shown to be misleading; BCM measurements cor- to the Creative Commons license, and indicate if changes were made. related with established biomarkers and cardiovascular mea- The results presented in this paper have not been published previously in sures [31]. whole or part, except in abstract format. Conclusion References Our study suggests that the BCM can be used in children as 1. Hecking M, Karaboyas A, Antlanger M, Saran R, Wizemann V, young as 2 years, at least for measurements of normally hy- Chazot C, Rayner H, Hörl WH, Pisoni RL, Robinson BM, Sunder- drated weight and fluid overload. The measurements are non- Plassmann G, Moissl U (2013) Significance of interdialytic weight invasive, well tolerated, inexpensive and easy to perform. If gain versus chronic volume overload: consensus opinion. Am J carried out according to a standardised protocol [29], BCM Nephrol 38:78–90 Pediatr Nephrol (2018) 33:1601–1607 1607 2. Weiner DE, Brunelli SM, Hunt A, Schiller B, Glassock R, Maddux 17. van Marken Lichtenbelt WD, Westerterp KR, Wouters L (1994) Deuterium dilution as a method for determining total body water: FW, Johnson D, Parker T, Nissenson A (2014) Improving clinical outcomes among hemodialysis patients: a proposal for a Bvolume effect of test protocol and sampling time. Br J Nutr 72:491–497 first^ approach from the chief medical officers of US dialysis pro- 18. Hilkert AW, Douthitt CB, Schlüter HJ, Brand WA (1999) Isotope viders. Am J Kidney Dis 64:685–695 ratio monitoring gas chromatography/mass spectrometry of D/H by 3. Wizemann V, Wabel P, Chamney P, Zaluska W, Moissl U, Rode C, high temperature conversion isotope ratio mass spectrometry. Rapid Malecka-Masalska T, Marcelli D (2009) The mortality risk of Commun Mass Spectrom 13:1226–1230 overhydration in haemodialysis patients. Nephrol Dial Transplant 19. Coward WA. Calculation of pool sizes and flux rates. In: Prentice 24:1574–1579 AM, eds. The doubly-labelled water method for measuring energy 4. Agarwal R (2010) Hypervolemia is associated with increased mor- expenditure. Technical recommendations for use in humans. tality among hemodialysis patients. Hypertension 56:512–517 Vienna: International Dietary Energy (http://archive.unu.edu/ 5. Chazot C, Wabel P, Chamney P, Moissl U, Wieskotten S, unupress/food2/UID05E/UID05E00.HTM, Table of Contents - Wizemann V (2012) Importance of normohydration for the long- Chapter 4. Accessed 27 Mar 2018) term survival of haemodialysis patients. Nephrol Dial Transplant 20. Sargent JA, Gotch FA (1980) Mathematic modeling of dialysis 27:2404–2410 therapy. Kidney Int 18(Suppl. 10):S2–S10 6. Stefánsson BV, Brunelli SM, Cabrera C, Rosenbaum D, Anum E, 21. Bradbury MG, Smye SW, Brocklebank JT (1995) Assessment of Ramakrishnan K, Jensen DE, Stålhammar NO (2014) Intradialytic the sensitivity of bioimpedance to volume changes in body water. hypotension and risk of cardiovascular disease. Clin J Am Soc Pediatr Nephrol 9:337–340 Nephrol 9:2124–2132 22. Paglialonga F, Ardissino G, Galli MA, Scarfia RV, Testa S, Edefonti 7. Flythe JE, Xue H, Lynch KE, Curhan GC, Brunelli SM (2015) A (2012) Bioimpedance analysis and cardiovascular status in pedi- Association of mortality risk with various definitions of atric patients on chronic hemodialysis. Hemodial Int 16(Suppl 1): intradialytic hypotension. J Am Soc Nephrol 26:724–734 S20–S25 8. Termorshuizen F, Dekker FW, van Manen JG, Korevaar JC, 23. Oh G, Wong C, Begin B, Salsbery K, Sutherland S, Chaudhuri A Boeschoten EW, Krediet RT, NECOSAD Study Group (2004) (2014) Whole-body single-frequency bioimpedance analysis in pe- Relative contribution of residual renal function and different mea- diatric hemodialysis patients. Pediatr Nephrol 29:1417–1423 sures of adequacy to survival in hemodialysis patients: an analysis 24. Brooks ER, Fatallah-Shaykh SA, Langman CB, Wolf KM, Price of the Netherlands Cooperative Study on the Adequacy of Dialysis HE (2008) Bioelectric impedance predicts total body water, blood (NECOSAD)-2. J Am Soc Nephrol 15:1061–1070 pressure, and heart rate during hemodialysis in children and adoles- 9. Vilar E, Farrington K (2011) Emerging importance of residual renal cents. J Ren Nutr 18:304–311 function in end-stage renal failure. Semin Dial 24:487–494 25. Lindley EJ, Lopot F (2015) The use of bioimpedance to aid volume 10. Agarwal R, Weir MR (2010) Dry-weight: a concept revisited in an assessment in dialysis patients. Kidney Int 87:240 effort to avoid medication-directed approaches for blood pressure con- 26. Kromeyer-Hauschild K, Wabitsch M, Kunze D, Geller F, Geiß HC, trol in hemodialysis patients. Clin J Am Soc Nephrol 5:1255–1260 Hesse V, von Hippel A, Jaeger U, Johnsen D, Korte W, Menner K, 11. Moissl UM, Wabel P, Chamney PW Bosaeus I, Levin NW, Bosy- Müller G, Müller JM, Niemann-Pilatus A, Remer T, Schaefer F, Westphal A, Korth O, Müller MJ, Ellegård L, Malmros V, Wittchen HU, Zabransky S, Zellner K, Ziegler A, Hebebrand J Kaitwatcharachai C, Kuhlmann MK, Zhu F, Fuller NJ (2006) (2001) [Percentile for the body mass index for children and adoles- Body fluid volume determination via body composition spectros- cents using various German samples]. Monatsschr Kinderheilkd copy in health and disease. Physiol Meas 27:921–933 149:807–818 12. Chamney PW, Wabel P, Moissl UM Müller MJ, Bosy-Westphal A, 27. Özdemir K, Mir MS, Dinçel N, Bozabali S, Kaplan Bulut İ, Yilmaz Korth O, Fuller NJ (2007) A whole-body model to distinguish ex- E, Sözeri B (2015) Bioimpedance for assessing volume status in cess fluid from the hydration of major body tissues. Am J Clin Nutr children with nephrotic syndrome. Turk J Med Sci 45:339–344 85:80–89 28. http://www.bcm-fresenius.de/files/information_on_reference_ 13. Wabel P, Chamney P, Moissl U, Jirka T (2009) Importance of ranges.pdf (Accessed 18 Feb 2018) whole-body bioimpedance spectroscopy for the management of 29. Lindley EJ, Keane DF, Schneditz D (2016) Comparison of fluid balance. Blood Purif 27:75–80 intradialytic changes in weight and fluid status. Nephrology 21:632 14. Passauer J, Petrov H, Schleser A, Leicht J, Pucalka K (2010) 30. Zaloszyc A, Schaefer B, Schaefer F, Krid S, Salomon R, Niaudet P, Evaluation of clinical dry weight assessment in haemodialysis pa- Schmitt CP, Fischbach M (2013) Hydration measurement by tients using bioimpedance spectroscopy: a cross-sectional study. bioimpedance spectroscopy and blood pressure management in Nephrol Dial Transplant 25:545–551 children on hemodialysis. Pediatr Nephrol 28:2169–2177 15. Davies SJ, Davenport A (2014) The role of bioimpedance and bio- 31. Eng CSY, Bhowruth D, Mayes M, Stronach L, Blaauw M, Barber markers in helping to aid clinical decision-making of volume as- A, Rees L, Shroff RC (2017) Assessing the hydration status of sessments in dialysis patients. Kidney Int 86:489–496 children with chronic kidney disease and on dialysis: a comparison 16. Cole KS, Cole RH (1941) Dispersion and absorption in dielectrics: of techniques. Nephrol Dial Transplant. https://doi.org/10.1093/ndt/ 1. Alternating current characteristics. J Chem Phys 9:341–351 gfx287

Journal

Pediatric NephrologySpringer Journals

Published: Jun 4, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off