Bioelectrical impedance is a measurable property of electrical ionic conduction of soft tissue, particularly lean tissue, as fat and bone are poor conductors.1–4 Impedance vector Z is composed of vector resistance (R), i.e., the opposition to flow of an alternating current through intra- and extracellular ionic solutions, representing the real part of Z and tissue reactance (Xc) vector, i.e., the capacitance produced by interfaces, including cell membranes and organelles, representing the imaginary part of Z.1–4 Thus, the rectangular coordinate representation of Z with R at the abscissa and Xc at the ordinate of the xy axis gives:
Main clinical applications of bioelectrical impedance analysis (BIA) include the estimation of body composition, the estimation of body fluids and of their compartments, and the estimation of clinical outcomes. As far as the specific issue of the estimation of hydration status in dialysis patients, several clinical methods of BIA (single frequency BIA, multifrequency BIA spectroscopy, BIA spectroscopy, and calf BIA spectroscopy) have been used for dry weights in end-stage renal disease patients.5–8 It must be underlined that, recently, an equivalence of information in fluid monitoring during hemodialysis (HD), using measurements at 50 kHz, conventional series circuit, vs. BIA spectroscopy with Cole and Hanai model has been demonstrated.9 Furthermore, as far as the specific issue of the estimation of clinical outcomes in dialysis patients by means of BIA is concerned, Pillon et al.3 have shown a significant association between shorter lengths of height-normalized Z vectors [√(R/H)2 + (Xc/H)2] and increased mortality; however, they could not demonstrate an association between longer vector lengths and decreased mortality. Now, we know that in the human body, more than 90% of the measured impedance Z is composed of R. Thus, it is obvious that the length of the BIA vector depends mainly on the R values: in other words, the higher the R value, the longer the vector. Recently, we were able to demonstrate what Pillon et al.3 were unable to demonstrate, i.e., an association between higher R values (equivalent to longer vector lengths) and decreased mortality.10
Thus, this study was designed to investigate into a research area in which published data are scarcely available: may BIA (measured as R and Xc) be influenced by the duration of the HD sessions?
SUBJECTS AND METHODS
Design of the Study
This crossover study was approved by our institutional review board and was conducted in accordance with good clinical practice guidelines and ethical principles of the Helsinki Declaration. The inclusion and exclusion criteria of the study are described elsewhere.11 A group of 11 stable white prevalent anuric uremic patients (nine men and two women, mean age: 54.1 ± 17.8 standard deviation years, dialysis vintage: 78.0 ± 60.2 months) was enrolled. After obtaining informed consent, they underwent one bicarbonate HD of 4 hours and one slow-flow bicarbonate HD session lasting 8 hours in a random sequence, always at the same interdialytic interval, at least 1 week apart. The experimental HD sessions used the GENIUS single-pass batch dialysis system.11 GENIUS provides 90 L of bicarbonate dialysate per dialysis session. It uses a double-sided roller pump that generates equal blood and dialysate flows up to 350 ml/min, as in the case of 4-hour HD sessions. As a consequence, dialysis sessions despite of markedly different duration still will apply an identical blood and dialysate volume, hence offering the opportunity to evaluate the impact of time as the sole variable. The system consists of a closed dialysate tank of 90 L, and although fresh and spent dialysate are stored together, dialysis may last up to 8 hours when using a blood and dialysate flow of 187 ml/min, without mixing of fresh and spent dialysate. The excess body water that is ultrafiltered out of the patient plasma is collected in an ultrafiltrate recipient. The sessions were pair matched as far as the dialysate and blood volume processed (90 L), and the volume of ultrafiltration (VUF) are concerned. Thus, the duration treatment was dictated by the achievement of the target dialysate and blood volume processed (i.e., 90 L): for this reason, it was slightly different from 4 and 8 hours (Table 1). Worth noting is the fact that the same dialysis machine and high-flux FX80 dialyzers (Fresenius Medical Care, Bad Homburg, Germany) were used in all sessions. The characteristics of the dialyzer are described elsewhere.12 Dialysate composition was as follows: total calcium 1.50 mmol/L; magnesium 0.5 mmol/L; potassium (K+) 2 mmol/L; sodium (Na+) 140 mmol/L; bicarbonate 35 mmol/L; chloride 113 mmol/L; glucose 5.55 mmol/L; and citrate 0.10 mmol/L.
Blood and Dialysate Sampling
Blood samples were taken from the inlet blood lines immediately before the onset and at the end of dialysis during the 4-hour and 8-hour sessions. Dialysate was collected from the inlet dialysate lines at the onset of dialysis sessions. Furthermore, at the end of dialysis, two samples were taken, respectively, from the ultrafiltrate recipient and from the dialysate tank, after thorough mixing, to quantify solute concentration in total spent dialysate. Particular attention was paid to the thorough mixing of the spent dialysate, strictly adhering to the ad hoc instructions present in the GENIUS operator's manual.
Plasma Na+, K+, ionized calcium (Ca++) concentrations, blood bicarbonate, and pH levels were measured immediately after blood sampling by means of an ion-selective electrode (Radiometer ABL 800, Kobenhavn, Denmark); Na+, K+, and Ca++ concentrations were measured in the fresh and in the spent dialysate and in the ultrafiltrate recipient (Radiometer ABL 800, Kobenhavn, Denmark). Details about the accuracy and precision of the measurements of Na+, K+, and Ca++ concentrations are reported elsewhere.11
Na+, K+, and Ca++ mass balances (Na+MB, K+MB, and Ca++MB, respectively) were calculated as follows (Na+MB is taken as an example):
Na+MB = (Na+ concentration in the fresh dialysate × 90) − ([Na+ concentration in the spent dialysate × 90] + [Na+ concentration in the ultrafiltrate recipient × VUF])
where 90 is the volume (L) of the fresh dialysate.
Solute removal during dialysis is expressed as a negative number, whereas solute gain is expressed as a positive number.
BIA measurements were determined at the start and the end of each experimental HD session, injecting 800 μA at 50 kHz alternating sinusoidal current with a standard tetrapolar technique (BIA 101 Impedance Analyzer; Akern, Florence, Italy). BIA was performed in standardized conditions: a quiet environment, ambient temperature of 22-24°C, and after being 20 minutes at rest in the supine position.13 The electrodes were left in situ during the session to avoid any possible change of their placement in the two measurements. The BIA variables measured were R, Xc, and the phase angle (arctan Xc/R).
Body weight was measured to the nearest 0.1 kg.
Data are reported as means ± SD. The Kolmogorov-Smirnov test was performed to assess the normality of the distribution of the data. The values of all the parameters were compared by means of the Student's t test for unpaired data. All the statistical inferences were made by means of the SPSS 11.0 software (SPSS Inc., Chicago, IL), and a p value <0.05 was considered for the statistical significance.
All the 22 dialysis sessions were uneventful; no session had to be interrupted because of clinical complications; no therapeutic interventions, such as saline infusions, were necessary. All the data collected in this study were normally distributed at the Kolmogorov-Smirnov test. Mean predialysis body weights were 72.2 ± 10.8 and 71.9 ± 10.7 kg (p = 0.205) in 4-hour and 8-hour treatments, respectively (Table 1). Mean postdialysis body weights were 69.2 ± 10.3 and 69.1 ± 10.1 kg (p = 0.458) in 4-hour and 8-hour treatments, respectively (Table 1). Mean VUF were 2.9 ± 0.8 and 2.9 ± 0.9 L (p = 0.851) in 4-hour and 8-hour treatments, respectively (Table 1).
Pre- and postdialysis plasma Na+, K+, Ca++ concentrations, blood pH, and bicarbonates levels and solute mass balances are listed in Table 1. No statistically significant difference was shown when comparing both (4 hours vs. 8 hours) pre- and postdialysis laboratory data (Table 1). Similarly, solute mass balances were not statistically significantly different between the two treatments (Table 1).
R values were statistically significantly higher at the end of the 8-hour dialysis sessions, when compared with the corresponding values at the end of the 4-hour dialysis sessions (p < 0.0001, Table 2). The same was true also for the Δ values (the difference between the post- and predialysis R values) and the percent increase of R values: both were statistically significantly higher in 8-hour dialysis sessions than in 4-hour dialysis sessions (for both, p < 0.02, Table 2). Xc values were statistically significantly higher at the end of the 8-hour dialysis sessions, when compared with the corresponding values at the end of the 4-hour dialysis sessions (p < 0.05, Table 2). However, the same was not true for the Δ values (the difference between the post- and predialysis Xc values) and the percent increase of Xc values: no statistically significant difference was observed in 8-hour dialysis sessions, when compared with the 4-hour dialysis sessions (Table 2). Phase angle values were not statistically significantly different at the end of the 8-hour dialysis sessions, when compared with the corresponding values at the end of the 4-hour dialysis sessions (p = 0.874, Table 2). The same was also true for the Δ values (the difference between the post- and predialysis phase angle values) and the percent increase of phase angle values: no statistically significant difference was observed in 8-hour dialysis sessions, when compared with the 4-hour dialysis sessions (Table 2).
Bioelectrical impedance is a measurable property of electrical ionic conduction of soft tissue, particularly lean tissue.1–4 In principle, at low frequencies (<5 kHz), current would pass through the extracellular fluid, whereas at higher frequencies (>100 kHz), it would penetrate all fluid compartments. In practice, a variable amount of current can cross muscle cells even at very low frequencies, particularly when the current path is parallel to the fiber.1 Physiologically, because of the fact that the current tends to follow the path of least resistance, measured R correlates most strongly with total body water (TBW), and correlations decrease for other body composition components, depending on the amount of water in these components. Whole-body Z is highly correlated with TBW measured by the isotope dilution techniques.14 In fact, Hoffer et al.,14 by injecting 100 μA at 100 kHz alternating sinusoidal current with a standard tetrapolar technique in 20 healthy volunteers and 34 patients with various diseases and degrees of fluid status (15 of them were affected by renal failure), found that the best relationship was TBW = H2/Z (where H is height, the correlation coefficients were 0.92 for the healthy volunteers and 0.93 for the patients). For this reason, most BIA applications use R, rather than impedance, to predict body composition.15,16
As already anticipated, we have shown that R was a significant independent predictor of long-term survival in a population of 328 incident HD patients.10 Specifically, the Cox regression analysis showed that higher R predicted a significantly better long-term survival of these patients.10 Furthermore, by stratifying the patients into the quartiles of R and evaluating them by means of the Kaplan-Meier survival analyses, we could show a significantly higher long-term survival in the groups of patients having R values above the 1st quartile (>467.8 Ohm).10 An indirect but interesting confirmation of the hypothesis generated by this prospective longitudinal study, i.e., that higher postdialysis R values predict a better long-term survival in HD patients, might come from the results of this study: 8-hour slow-flow bicarbonate HD sessions were associated with postdialysis R values, ΔR values, and percent increase of R values statistically significantly higher than the corresponding values of the 4-hour dialysis sessions. Thus, the hypothesis can be formulated, even though it must be proven in ad hoc studies, that patients undergoing long-hour nocturnal HD may have higher R values than those undergoing standard HD. Now, we know that nocturnal HD is associated with better clinical outcomes than standard HD.17 If this is mainly due to the maintenance of a correct dry body weight, of which higher R values may be a proxy, it remains a matter of future research. What is clearly known at the present time is that the hydration state is an important and independent predictor of mortality in chronic HD patients secondary only to the presence of diabetes.18 However, a crucial question remains: what is or what are the reasons why R values increase more at the end of the 8-hour sessions than at the end of the 4-hour session in this highly controlled crossover study, in which all the variables studied were not statistically significantly different? Different body fluid changes and/or a different body fluid distribution can be advocated: actually, VUF was perfectly pair matched, but the ultrafiltration rate was completely different between the two treatments, being in the 8-hour sessions exactly half compared with that of 4-hour sessions. Therefore, one can hypothesize that a slower ultrafiltration rate over 8 hours may be able to perform a more efficient vascular refilling by moving fluids from the hydrated interstitial gel into the vessels.5,11 Thus, it could be that during 8 hours of ultrafiltration at the same VUF with less hemodynamic stress, refilling from limbs is improved and that volume has sufficient time to equilibrate throughout the whole body with lower ultrafiltration rates. Actually, it has been shown that the different body regions are not equally contributing to so-called whole body BIA where almost 90% of the signal comes from the limbs.19 On the contrary, in the 4-hour sessions, part of the same VUF could be still bound to the interstitial gel, therefore yielding lower impedance values.
Furthermore, previous BIA studies done using real-time sampling have shown an increase in R during the dialysis procedure when no ultrafiltration is done, when only ultrafiltration is done, and also when ultrafiltration and dialysis are performed simultaneously.20 The rise in R during isolated ultrafiltration can be explained (as we have already postulated) by tissue dehydration. However, the increase in R during dialysis with no ultrafiltration also indicates that solute removal (urea and electrolytes) causes an independent increase in R.20 Small and middle molecule mass balances and kinetic modeling data are not presented in this study, but they have been published very recently in a companion article21: small and middle molecules are removed more adequately from the deeper compartments when performing a prolonged dialysis (8 hours), when compared with a standard dialysis, even if blood and dialysate volumes are kept constant.21 Thus, a new hypothesis alternative to fluid removal from different compartments as a likely explanation for the difference in R values can be formulated: being the amount of urea removed greater in the prolonged dialysis, a higher rate of intracellular, and interstitial volume removal could explain the higher R values found in this study.
A final question which our study cannot address is the following one: the postdialysis measures were taken at 4 or at 8 hours depending on the duration of treatment. This raises the question of what would be the R value in the shorter time scheme were it also measured at 8 hours? This is of interest as the predialysis values were not statistically significantly different.
In conclusion, the present highly controlled experiments using precise controls of solute mass balances show that 8-hour slow-flow bicarbonate HD sessions were associated with postdialysis R values, ΔR values, and percent increase of R values statistically significantly higher than the corresponding values of the 4-hour dialysis sessions. If higher R values may represent a proxy of a correct dry body weight, it remains a matter of future research.
1. Grimnes S, Martinsen ØG (eds): Bioimpedance and Bioelectricity Basics.
London, Academic Press, 2000.
2. Foster KF, Lukaski HC: Whole-body impedance—What does it measure? Am J Clin Nutr
64(suppl 3): S388–S396, 1996.
3. Pillon L, Piccoli A, Lowrie EG, et al
: Vector length as a proxy for the adequacy of ultrafiltration in hemodialysis. Kidney Int
66: 1266–1271, 2004.
4. Kushner RF: Bioelectrical impedance analysis: A review of principles and applications. J Am Coll Nutr
11: 199–209, 1992.
5. Piccoli A, for the Italian hemodialysis-bioelectrical impedance analysis (HD-BIA) study group: Identification of operational clues to dry weight prescription in hemodialysis using bioimpedance vector analysis. Kidney Int
53: 1036–1043, 1998.
6. Chamney PW, Krämer M, Rode C, et al
: A new technique for establishing dry weight in hemodialysis patients via whole body bioimpedance. Kidney Int
61: 2250–2258, 2002.
7. Zhu F, Kuhlmann MK, Sarkar S, et al
: Adjustment of dry weight in hemodialysis patients using intradialytic continuous multifrequency bioimpedance of the calf. Int J Artif Organs
27: 104–109, 2004.
8. Basile C, Vernaglione L, Di Iorio B, et al
: Development and validation of bioimpedance analysis prediction equations for dry weight in hemodialysis patients. Clin J Am Soc Nephrol
2: 675–680, 2007.
9. Piccoli A, Pastori G, Guizzo M, et al
: Equivalence of information from single versus multiple frequency bioimpedance vector analysis in hemodialysis. Kidney Int
67: 301–313, 2005.
10. Basile C, Vernaglione L, Lomonte C, et al
: Comparison of alternative methods for scaling dialysis dose. Nephrol Dial Transplant
25: 1232–1239, 2010.
11. Basile C, Libutti P, Di Turo AL, et al
: Haemodynamic stability in standard bicarbonate haemodialysis and long-hour slow-flow bicarbonate haemodialysis. Nephrol Dial Transplant
26: 252–258, 2011.
12. Eloot S, Van Biesen W, Dhondt A, et al
: Impact of hemodialysis duration on the removal of uremic retention solutes. Kidney Int
73: 765–770, 2008.
13. Bellizzi V, Scalfi L, Terracciano V, et al
: Early changes in bioelectrical estimates of body composition in chronic kidney disease. J Am Soc Nephrol
17: 1481–1487, 2006.
14. Hoffer EC, Meador CK, Simpson DC: Correlation of whole-body impedance with total body water. J Appl Physiol
27: 531–534, 1969.
15. Baumgartner RN: Electrical impedance and TOBEC, in Roche AF, Heymsfield SB, Lohman T (eds), Human Body Composition
. Champaign, Human Kinetics, 208–237: 1996.
16. Sun SS, Chumlea WC, Heymsfield SB, et al
: Development of bioelectrical impedance analysis prediction equations for body composition with the use of a multicomponent model for use in epidemiologic surveys. Am J Clin Nutr
77: 331–340, 2003.
17. Charra B, Calemard E, Ruffet M, et al
: Survival as an index of adequacy of dialysis. Kidney Int
41: 1286–1291, 1992.
18. Wizemann V, Wabel P, Chamney P, et al
: The mortality risk of overhydration in haemodialysis patients. Nephrol Dial Transplant
24: 1574–1579, 2009.
19. Zhu F, Schneditz D, Wang E, et al
: Validation of changes in extracellular volume measured during hemodialysis using a segmental bioimpedance technique. ASAIO J
44: M541–M545, 1998.
20. Dumler F, Schmidt R, Kilates C, et al
: Use of bioelectrical impedance for the nutritional assessment of chronic hemodialysis patients. Miner Electrolyte Metab
18: 284–287, 1992.
21. Basile C, Libutti P, Di Turo AL, et al
: Removal of uraemic retention solutes in standard bicarbonate haemodialysis and long-hour slow-flow bicarbonate haemodialysis. Nephrol Dial Transplant
26: 1296–1303, 2011.