Rapid ultrafiltration (UF) of large volume excess during hemodialysis (HD) results in intradialytic hypotension.1,2 This is perceived by the body as any other blood water loss, and a series of defensive mechanisms are activated. Unless the control of blood volume is highly efficient,3,4 excessive UF may result in a hypotensive crisis, accompanied by symptoms, such as cramps and nausea, often requiring the premature interruption of the HD session,1,2 and increased mortality risk.5
The protein concentration in blood increases during UF, inducing oncotic recall of water from the interstitium. This phenomenon, called plasma refilling, is one of the mechanisms by which the body attempts to restore the imbalance caused by UF. Refilling is clinically important because of its role in preventing hypovolemia and hypotensive crises. The problem of adequate refilling is often addressed in clinical studies.3,6–14
Continuous on-line blood volume monitoring was applied by some authors to obtain qualitative characteristics of the process of refilling and its relation to hypotensive episodes.3,15–21 The quantitative analysis of the refilling process may be based on a two-compartment model to describe the rate of plasma refilling through the application of the Starling equation to the kinetics of plasma volume, Vp, during dialysis session22:
where on the right-hand side are transcapillary plasma refilling rate, lymphatic refilling rate (L) and UF rate. The transcapillary refilling rate is proportional to the balance of the Starling transcapillary forces (hydrostatic blood pressure P and oncotic pressure Π, subscripts c and i stand for capillary and interstitium, respectively) by the whole-body filtration coefficient Lp, which is usually calculated by fitting the model to the data. Lp may be used for the assessment of the functionality of the refilling process by virtue of its possible role as a marker of the vasodilation of the capillaries; particularly interesting is the correlation between the values of Lp and concentration of atrial natriuretic peptide (ANP).19,23
An index to quantify the refilling process of the system based on the filtration coefficient Lp, called the refilling coefficient (Kr), may be calculated from clinical data.21,24 Kr corresponds to the filtration coefficient in the Starling equation, but because additional assumptions must be introduced for its calculation, it may be considered only as an approximation of Lp.23,24 Kr is expressed as the ratio of plasma refilling rate (RF) to the increase in plasma oncotic pressure
over the initial oncotic pressure
(see Appendix A, Supplemental Digital Content, http://links.lww.com/ASAIO/A62):
Although the filtration coefficient in the Starling equation was expected to be constant, it was found that Kr decreases remarkably during UF and its values correlate well with vasodilatory peptides.19,23 It was shown that the end-of-session value of a time-dependent Lp might be a useful indicator in assessment of the fluid status of HD patients.25 The purpose of our study was to investigate the kinetics of the refilling process by assessing Kr for two standard HD sessions with: 1) shorter predialytic interval before the dialysis session and shorter session duration (SH), and 2) longer predialytic interval before the session and longer session duration (LH), in each of nine clinically stable HD patients. According to the design of the study, we investigated if and how the difference in fluid overload (FO) between sessions was reflected in the magnitude and kinetic behavior of the refilling coefficient.
The study was performed in nine stable end-stage renal disease patients (three males, median age 69, range 32–85 years) on maintenance HD with Gambro AK200 ULTRA S dialysis machine (Gambro AB, Lund, Sweden). The study was approved by the Regional Ethical Review Board in Stockholm, Sweden. Informed consent was obtained from all participants in the study. Patients were supine in beds during dialysis, and the position was maintained during the whole treatment, except in a few cases. The extracorporeal circuit was primed with approximately 300 ml saline that was not discarded. The patients were instructed to refrain if possible from intake of fluid and foods; and, only in a few cases food or drinks were given to the patients. Each patient underwent two sessions of different duration: 1) SH of 3.5 h, and 2) LH of 4.5 h. The predialytic interval was 44 h for SH and 68 h for LH, except for one patient with predialytic interval of 44 h for both sessions. No UF profiling was applied during the assessed dialysis sessions. Volume of body water compartments was measured through bioimpedance spectroscopy (Body Composition Monitor, Fresenius, Bad Homburg, Germany). Predialysis FO was calculated as the difference between total body water (TBW) content before and after HD, assuming that the final TBW value reached corresponded to FO = 0. Relative FO was derived scaling FO over its pre-HD value, and was assumed that it linearly decreased during the treatment.
Relative blood volume change was measured online (by hemoglobin light-absorption with Gambro BVS system, sampling period: 2.2 seconds) as a percentage of the initial blood volume. The initial blood volume was estimated using a commonly adopted anthropometric formula (see Appendix B, Supplemental Digital Content, http://links.lww.com/ASAIO/A62). To be able to proceed with the calculations, the volume decrease was approximated with an exponential function, c.f..24
Total protein concentration in serum (Cp) was measured from blood samples taken before and after HD and every hour during the session. Plasma oncotic pressure, Π, was calculated from total protein concentration through the Landis–Pappenheimer formula.26 An exponential function was then used to interpolate the oncotic pressure data (see Appendix B, Supplemental Digital Content, http://links.lww.com/ASAIO/A62). The refilling coefficient Kr(t) at time t was calculated using Equation 2. Assuming the mass of hemoglobin and the total volume of red blood cells constant during the HD session, we considered the rate of plasma volume change to be equal to the rate of blood volume change and used it to estimate the plasma refilling rate (see Appendix A, Supplemental Digital Content, http://links.lww.com/ASAIO/A62). Because of possible errors and disturbances in the body fluids distribution induced by the start of dialysis, the data were evaluated inside an interval between 60 and 180 or 240 minutes, for SH and LH sessions, respectively, as in the previous studies on Kr.23,24
Statistical comparisons between the sessions were performed with Wilcoxon matched pairs test. Differences were considered significant at p < 0.05. All values are expressed as mean ± standard deviation. Spearman’s rank correlation coefficient ρ was applied for the analysis of correlations; analysis of covariance was used to test for the effect of the grouping variable on pooling the observations from the SH and LH sessions.
As expected from the design of the study, initial body mass and TBW volume were higher before LH sessions than before SH sessions (Table 1). Ultrafiltration volume (but not UF rate), body mass drop, TBW drop, and extracellular volume drop (but not intracellular water drop) during the session were also higher after LH sessions than after SH sessions (Table 1). The volume of refilled fluid was higher by 0.75 L for LH compared with SH sessions (Table 1).
Mean arterial pressure (MAP), systolic and diastolic arterial pressures, and heart rate profiles remained on average constant during the treatment. No statistically significant difference was observed in blood pressure for different duration of the session: the average value of MAP was 90.5 ± 16.6 mm Hg for SH and 82.2 ± 16.7 mm Hg for LH (p = 0.14). The amplitude of fluctuations in pressure profiles during the session, i.e., the difference between the highest and the lowest value in pressure profile was 24.2 ± 9.3 mm Hg for SH and 24.2 ± 12.7 mm Hg for LH (p = 0.6).The average heart rate was 79.7 ± 16.3 bpm for SH and 78.7 ± 16.1 bpm for LH (p = 0.6). The amplitude of fluctuations in heart rate profile was 10 ± 7.4 bpm for SH and 16.9 ± 22.0 for LH (p = 0.5).
The relative blood volume change, calculated as a fraction of the pre-HD value, was not different between the two sessions (Figure 1 and Table 1). As expected, plasma oncotic pressure was increasing during the treatment (Figure 2); no significant difference was found between the two sessions. Ultrafiltration and refilling rates were similar during both sessions, but ultrafiltered and refilled volumes were higher for LH. However, there was a statistical tendency to higher refilling rate at 1 h for LH (392 ± 152 ml/h and 488 ± 164 ml/h for SH and LH, respectively; p = 0.07), but later on the values were not statistically different (p = 0.13 at 2 h; Figure 3). The refilling coefficient Kr was found, for both sessions, to decrease exponentially like toward a value reached at the end of the evaluation interval of 137 ± 56 ml/mm Hg/h for SH and 151 ± 74 ml/mm Hg/h for LH (p = 0.45; Figure 4A). The difference in Kr at 1 h between SH and LH (267 ± 127 ml/mm Hg/h and 392 ± 212 ml/mm Hg/h, respectively) was not statistically significant but showed a tendency (p = 0.07, effect size = 0.4). Furthermore, the median values of Kr were closer for the time interval 60–180 minutes, as seen in Figure 4B. The relative drop in Kr during the session, calculated as a percentage of the initial value, was similar for both sessions (p = 0.11, Table 1). Kr decreased along with FO between 1 h and the end of the evaluation period (Figure 5).
Analysis of covariance applied to the pairs of variables examined for correlations showed that interaction with the grouping variable was not significant; thus the observations from sessions SH and LH were pooled when calculating the correlation coefficients. Kr values at 1 h and at the end of the evaluation interval correlated with relative blood volume drop at the end of the evaluation interval, although the correlation was exponential (linear on the log(Kr) scale; Figure 6). The final value of Kr correlated with the drop in body mass and UF volume, as reported in Figure 7; similar significant correlations of final Kr where also found with extracellular and TBW drop (data not shown).
The complex mechanisms and the role of specific factors influencing the refilling capacity during a HD session are still not clear. Obviously, this makes it hard to reliably assess the efficiency of the refilling process, and how it affects the performance of the cardiovascular system during treatment sessions. The transcapillary permeability coefficient (Lp) in the context of the refilling process during HD was investigated for the first time by Schneditz et al.21Lp was estimated from a two-pool model of intradialytic fluid shifts and, separately, calculated from assumptions similar to those used later by Tabei et al.24; however, in the first case, calculations were restricted to an experimental period of 20 minutes of high-rate UF plus 20 minutes of plasma refilling, during which Lp was considered constant. In a more recent series of papers,23,24 Tabei and Imura attempted to calculate this parameter from volumetric data on the whole HD session using Equation 2 and found that its value changed during the course of the session; they proposed to use it as an index of the refilling process efficiency, and called it refilling coefficient, Kr. Their results showed a correlation between the drop in Kr and the vasodilation status of the patients.23 Yashiro et al.18–20 studied Lp applying a model derived from the one used in  but recalculating Lp at 20 second intervals; they also observed that the Lp values decreased during dialysis. In this case, the attention was focused on the final value of the permeability coefficient.
In this study, the two sessions in each patient differed as regards both the duration of session and the fluid status before the treatment, because of the longer predialytic interval before LH sessions. During LH, patients had higher initial values of body mass and TBW content. Because each patient was dialyzed with a similar UF rate but with different duration of dialysis, the LH sessions resulted in higher final UF volumes and therefore the patients undergoing LH sessions showed also a larger drop in body mass, TBW, and extracellular water volume (Table 1).
The decrease in blood volume was similar during SH and LH sessions even though the amount of water removed by UF was higher during LH. This was probably because, according to the Starling law, a higher removal of water is followed by a higher amount of volume recalled from the interstitium, as shown by the higher refilled volume calculated for the longer sessions (Table 1). However, we found this behavior to be nonlinear, as the ratio between volume refilled and volume removed by UF, which can be considered as an indicator of the efficacy of the refilling process, was higher among the initially more volume expanded patients in the LH sessions (refilled/filtrated: 74 ± 9% for SH vs. 81 ± 4% for LH; p < 0.05, Table 1).
The average Kr of 323 ml/mm Hg/h after 1 h of dialysis and 144 ml/mm Hg/h at the end of the evaluation interval obtained in this study were similar to those reported by Tabei et al.,24 405 ml/mm Hg/h after 1 h, and 94 ml/mm Hg/h after 4 h. After scaling to 50 kg of lean body mass (LBM),21 our results (361 and 159 ml/mm Hg/h, respectively) were similar to the values of Lp obtained by Schneditz et al.21 during the second hour of dialysis session, 336 ml/mm Hg/h. Yashiro et al.20 reported Lp values (scaled to 50 kg LBM) estimated at the end of a 4 h dialysis session for patients grouped according to the value of the ratio UF volume/body mass (postdialysis), UFV/BW, as: 154 ml/mm Hg/h for UFV/BW < 3%, 93 ml/mm Hg/h for UFB/BW in the range from 3% to 5 %, and 61 ml/mm Hg/h for UFV/BW > 5%. If, for comparison, we group dialysis sessions in our study according to UFV/BW into two groups (only one dialysis session had UFV/BW > 5%), then final Kr scaled to 50 kg of LBM was 196 ml/mm Hg/h for UFV/BW < 3%, and 111 ml/mm Hg/h for UFB > 3% (p < 0.05), in good agreement with Yashiro et al.20
The higher predialysis FO in patients before LH sessions seems not to have significantly affected Kr. However, a p value at 1 h close to the significance limit and a medium effect size suggest that Kr might be initially higher for LH, but even when extrapolating Kr to t = 0 the difference remains outside of the significance interval (p = 0.07). Further studies with higher sample size might solve this issue, but, on the other hand, the lack of homogeneity in the response in Kr to FO (see below) may preclude any clear interpretation of the presented statistical evaluation. Anyway this difference decreased with session time and, as shown in Figure 4A by the extrapolated values, Kr stabilized on similar values (p = 0.31). The stabilization occurred around 3 h for both SH and LH, and later on the changes were less than 5%. The relative stability of final as compared with initial Kr values and the negligible difference in final Kr between sessions of different duration suggest that after the removal of a sufficient amount of fluid, the patients tend to reach a similar value of refilling coefficient, regardless of initial fluid status.
Although, on average, patients had similar Kr after short and long interdialytic interval, and also the median of Kr was practically the same for both LH and SH dialysis sessions, the distribution of initial values of Kr was skewed toward high values for LH sessions because of high Kr measured in two patients (Figure 4B). Thus, the group of our patients seems to be inhomogenous in their response to incremental FO and this observation needs to be taken into account in new studies on the refilling process.
High values of Kr may protect against a high blood volume loss, as reflected by the negative correlation between Kr and percent blood volume decrease at the end of the evaluation period (Figure 6). The correlation reached at the end of the evaluation interval was especially strong (ρ = -0.7; p < 0.01), suggesting that the stable final value of Kr might be a more appropriate index of the ability of the refilling system to oppose the blood volume drop.
Iimura et al.23 have suggested that one reason why the initial values of Kr could be higher than the later values is that the predialysis levels of vasodilator hormones, especially of ANP and cyclic guanosine monophosphate, are elevated and then fall during the HD session; Kr was well correlated to the levels of these two hormones.23 In another study, B-type natriuretic hormone was shown to a have similar pattern in HD.27 Thus, the initial high degree of vasodilatation may be the main reason for the high initial values of the capillary hydraulic conductivity times capillary surface area, and therefore also of Kr, with a subsequent slow decrease in both vasodilatation and total hydraulic conductivity of the capillary bed thereafter, until the steady state of the capillary bed is reached. This decrease in the capillary surface area is accompanied by the increase in the peripheral vascular resistance and better control of the blood volume, especially of the central blood volume, and blood pressure.10 This finding may suggest that the vasodilatation did not change much after some threshold in FO was reached; the similar final values of Kr in both SH and LH might be interpreted that the patients reached a similar volume status after sessions of different length.
There may also be other factors that could impact on the estimated values of Kr. One of them is the change in interstitial hydrostatic pressure when the FO of the patient decreases during UF and vascular refilling.24 Actually, most of the removed excess fluid appeared to originate from the extracellular space (Table 1). Attempts to induce water flow from the intracellular to extracellular space by sodium profiling, aiming at the maintenance of the interstitial volume, are able to improve the plasma refilling.28,29
Another factor of potential importance might be the redistribution of plasma and red blood cells between the micro- and macrocirculation caused by UF during dialysis.30 The intravascular shift of blood with a low hematocrit from the microcirculation to the macrocirculation has been observed to lead to an underestimation of blood volume changes calculated from central hematocrit and hemoglobin measurements in the magnitude of 50%.31 We used the “worst-case” assumption to estimate a correction for our relative blood volume data, and calculated corrected Kr values (see Appendix C and Figure C1, Supplemental Digital Content, http://links.lww.com/ASAIO/A62). As expected, the corrected refilling coefficients were much smaller but Kr behavior did not change, and the difference between corrected and uncorrected values of Kr was small at the end of UF (see Appendix C, Supplemental Digital Content, http://links.lww.com/ASAIO/A62).
The observation that in some patients on HD the blood volume remains almost constant despite substantial UF (although not found in our patients) suggests that these and possibly other factors, beside oncotic pressure, may be responsible for refilling in some conditions.32A validated method of whole-body multifrequency bioimpedance analysis was applied in our study to obtain information about the volume status of the patients before and after HD sessions. This method may be replaced by a more elaborate method of segmental bioimpedance that may provide more exact data especially during the volume perturbation caused by UF.33–36 Nevertheless, whole-body bioimpedance seems to be sufficient for studying the relations between Kr values and volume of body compartments in the clinical setting. However, more precise information about the change of fluid volume in different body segments would be interesting for further understanding of the refilling kinetics.
Our study demonstrates, following the previous reports,23,37 that the FO during interdialytic intervals results in an increase in refilling coefficients probably because of vasodilatation and increased capillary surface area.23 However, the additional FO in the long compared with the short interdialytic interval is not accompanied by a substantial increase in Kr, except for some patients (Figure 4B). The subgroup of two patients sensitive to the difference in FO after short and long interdialytic intervals deserves further investigation. During UF with standard UF rates, body water and blood volumes decrease and concomitantly Kr values approach an equilibrium value in 2 to 3 h. Refilling coefficients stabilize despite continued UF, and this observation suggests that the capillary exchange system approaches equilibrium before the patient dry weight is reached. The assessment of Kr after short and long interdialytic intervals may help to identify patients with high sensitivity to the increment in FO during long interdialytic intervals.
We conclude that, irrespective of possible differences in the initial value, the refilling coefficient Kr, which may be considered as an approximation of the capillary hydraulic conductivity times capillary surface area, tends to reach a similar final value, about 144 ml/mm Hg/h, following both shorter and longer HD sessions. The low variability in the final, stable value of Kr and the fact that it was reached after a similar time interval during both dialysis sessions with different degree of fluid excess suggests that the final value of refilling coefficient, is largely independent on the initial fluid status of the patient.
The authors thank the patients who participated in the study and the nurses involved in the study for their excellent job in collecting the required samples. The bioimpedance spectroscopy monitor BCM was kindly provided by Fresenius Medical Care Sverige AB for the purpose of this study. The study was partly supported by the project PO KL: “Information technologies: research and their interdisciplinary applications”.
1. Daugirdas JT. Dialysis hypotension: A hemodynamic analysis. Kidney Int. 1991;39:233–246
2. Daugirdas JT. Pathophysiology of dialysis hypotension: An update. Am J Kidney Dis. 2001;38(4 Suppl 4):S11–S17
3. Agarwal R. How can we prevent intradialytic hypotension? Curr Opin Nephrol Hypertens. 2012;21:593–599
4. Nesrallah GE, Suri RS, Guyatt G, et al. Biofeedback dialysis for hypotension and hypervolemia: A systematic review and meta-analysis. Nephrol Dial Transplant. 2013;28:182–191
5. Henderson LW. Symptomatic intradialytic hypotension and mortality: An opinionated review. Semin Dial. 2012;25:320–325
6. Brummelhuis WJ, van Geest RJ, van Schelven LJ, Boer WH. Sodium profiling, but not cool dialysate, increases the absolute plasma refill rate during hemodialysis. ASAIO J. 2009;55:575–580
7. Harzallah K, Daiki M, Baffoun A, et al. Comparative study of perdialytic volemia variability in chronic dialysis patients between acetate-free biofiltration and bicarbonate hemodialysis. Saudi J Kidney Dis Transpl. 2011;22:982–989
8. Jabara AE, Mehta RL. Determination of fluid shifts during chronic hemodialysis using bioimpedance spectroscopy and an in-line hematocrit monitor. ASAIO J. 1995;41:M682–M687
9. Minutolo R, De Nicola L, Bellizzi V, et al. Intra- and post-dialytic changes of haemoglobin concentrations in non-anaemic haemodialysis patients. Nephrol Dial Transplant. 2003;18:2606–2612
10. Prakash S, Reddan D, Heidenheim AP, Kianfar C, Lindsay RM. Central, peripheral, and other blood volume changes during hemodialysis. ASAIO J. 2002;48:379–382
11. Seibert E, Zhu F, Kuhlmann MK, et al. Slope analysis of blood volume and calf bioimpedance monitoring in hemodialysis patients. Nephrol Dial Transplant. 2012;27:4430–4436
12. Nette RW, Krepel HP, van den Meiracker AH, Weimar W, Zietse R. Specific effect of the infusion of glucose on blood volume during haemodialysis. Nephrol Dial Transplant. 2002;17:1275–1280
13. Katzarski KS, Charra B, Luik AJ, et al. Fluid state and blood pressure control in patients treated with long and short haemodialysis. Nephrol Dial Transplant. 1999;14:369–375
14. Santoro A, Mancini E, Basile C, et al. Blood volume controlled hemodialysis in hypotension-prone patients: A randomized, multicenter controlled trial. Kidney Int. 2002;62:1034–1045
15. Beige J, Sone J, Sharma AM, et al. Computational analysis of blood volume curves and risk of intradialytic morbid events in hemodialysis. Kidney Int. 2000;58:1805–1809
16. Casagrande G, Teatini U, Romei Longhena G, Miglietta F, Fumero R, Costantino ML. A new method to evaluate patient characteristic response to ultrafiltration during hemodialysis. Int J Artif Organs. 2007;30:377–384
17. Arrigo G, Bucci R, Colasanti G, et al. Blood volume modeling and refilling rate estimation in hemodialysis by continuous hemoglobin monitoring. Int J Artif Organs. 1995;18:509–512
18. Yashiro M, Hamada Y, Matsushima H, Muso E. Estimation of filtration coefficients and circulating plasma volume by continuously monitoring hematocrit during hemodialysis. Blood Purif. 2002;20:569–576
19. Yashiro M, Watanabe H, Tomita M, Yamadori N, Muso E. Relationship between filtration coefficients of microvasculature and levels of atrial natriuretic peptide or echocardiographic measurements. Blood Purif. 2005;23:431–439
20. Yashiro M, Kamata T, Segawa H, Murakami T, Kadoya Y, Muso E. How does higher ultrafiltration within the conventional clinical range impact the volume status of hemodialysis patients? Blood Purif. 2009;27:253–260
21. Schneditz D, Roob J, Oswald M, et al. Nature and rate of vascular refilling during hemodialysis and ultrafiltration. Kidney Int. 1992;42:1425–1433
22. Chamney PW, Johner C, Aldridge C, et al. Fluid balance modelling in patients with kidney failure. J Med Eng Technol. 1999;23:45–52
23. Iimura O, Tabei K, Nagashima H, Asano Y. A study on regulating factors of plasma refilling during hemodialysis. Nephron. 1996;74:19–25
24. Tabei K, Nagashima H, Imura O, Sakurai T, Asano Y. An index of plasma refilling in hemodialysis patients. Nephron. 1996;74:266–274
25. Yashiro M, Ochiai M, Fujisawa N, Kadoya Y, Kamata T. The evaluation of filtration coefficients of microvasculature for the assessment of fluid status in hemodialysis patients. Int J Artif Organs. 2013;36:7–16
26. Landis E, Pappenheimer J.Hamilton WF. Exchange of substances through the capillary walls. Handbook of Physiology Circulation. 1963 Washington American Physiological Society:961–1034
27. De Mauri A, Bellomo G, Navino C, et al. Plasma B-type natriuretic peptide in dialyzed patients: Marker of cardiovascular disease or link to plasma refilling. J Nephrol. 2011;24:507–514
28. Fleming SJ, Wilkinson JS, Greenwood RN, Aldridge C, Baker LR, Cattell WR. Effect of dialysate composition on intercompartmental fluid shift. Kidney Int. 1987;32:267–273
29. Stiller S, Bonnie-Schorn E, Grassmann A, Uhlenbusch-Körwer I, Mann H. A critical review of sodium profiling for hemodialysis. Semin Dial. 2001;14:337–347
30. Mitra S, Chamney P, Greenwood R, Farrington K. The relationship between systemic and whole-body hematocrit is not constant during ultrafiltration on hemodialysis. J Am Soc Nephrol. 2004;15:463–469
31. Dasselaar JJ, Lub-de Hooge MN, Pruim J, et al. Relative blood volume changes underestimate total blood volume changes during hemodialysis. Clin J Am Soc Nephrol. 2007;2:669–674
32. Rodriguez HJ, Domenici R, Diroll A, Goykhman I. Assessment of dry weight by monitoring changes in blood volume during hemodialysis using Crit-Line. Kidney Int. 2005;68:854–861
33. Zhu F, Schneditz D, Wang E, Levin NW. Dynamics of segmental extracellular volumes during changes in body position by bioimpedance analysis. J Appl Physiol (1985). 1998;85:497–504
34. Zhu F, Schneditz D, Wang E, Martin K, Morris AT, Levin NW. Validation of changes in extracellular volume measured during hemodialysis using a segmental bioimpedance technique. ASAIO J. 1998;44:M541–M545
35. Zaluska WT, Schneditz D, Kaufman AM, Morris AT, Levin NW. Relative underestimation of fluid removal during hemodialysis hypotension measured by whole body bioimpedance. ASAIO J. 1998;44:823–827
36. Basile C, Libutti P, Di Turo AL, et al. Bioimpedance and the duration of the hemodialysis session. ASAIO J. 2011;57:310–313
37. Yashiro M, Kamata T, Yamadori N, Tomita M, Muso E. Evaluation of markers to estimate volume status in hemodialysis patients: Atrial natriuretic peptide, inferior vena cava diameter, blood volume changes and filtration coefficients of microvasculature. Ther Apher Dial. 2007;11:131–137
hypotension; ultrafiltration; blood pressure; extracellular volume; blood volume; plasma oncotic pressure
Supplemental Digital Content
Copyright © 2015 by the American Society for Artificial Internal Organs