Original Paper Received: June 17, 2014 Accepted: September 2, 2014 Published online: December 16, 2014
Blood Purif 2014;38:180–187 DOI: 10.1159/000368157
A Simple and Feasible Method to Determine Absolute Blood Volume in Hemodialysis Patients in Clinical Practice Joachim Kron a Daniel Schneditz b Til Leimbach a Sabine Aign a Susanne Kron c
a
KfH Kidney Center Berlin-Köpenick, Berlin, Germany; b Institute of Physiology, Medical University of Graz, Graz, Austria; c Department of Nephrology, Charite Universitätsmedizin Berlin, Berlin, Germany
Abstract Background: We developed a simple method to determine the absolute blood volume (V) during hemodialysis in everyday clinical practice and examined its relationship with volume overload, clinical relevance, and accuracy. Methods: The increase in relative blood volume (RBVpost – RBVpre) measured before and after infusion of 240 ml of ultra-pure dialysate using the bolus function of a commercial online hemodiafiltration machine incorporating a relative blood volume monitor was applied to determine absolute blood volume. The specific blood volume (Vs, blood volume per kg body mass at dry weight, in ml/kg) was compared to volume status as assessed by bioimpedance analysis and clinical criteria. Results: The blood volume measured in 30 stable hemodialysis patients was 6.51 ± 1.70 l at the beginning, corresponding to a specific blood volume of 80.1 ± 12.8 ml/kg, and dropped to 5.84 ± 1.61 l or 72.0 ± 12.1 ml/kg at the end of the dialysis session, respectively. Specific blood volume correlated with volume status assessed both clinically and by bioimpedance analysis. Intradialytic morbid events oc-
© 2014 S. Karger AG, Basel 0253–5068/14/0384–0180$39.50/0 E-Mail [email protected] www.karger.com/bpu
curred only in treatments where specific blood volume fell below 65 ml/kg. The reproducibility of the technique was better than 4% and the in vitro accuracy corresponds to a resolution in Vs of better than 1 ml/kg. Conclusion: Absolute blood volume can be easily measured at the beginning of the dialysis session using the current dialysis technology. Information about V and Vs could be a promising tool to avoid intradialytic morbid events. This technique could be completely automated without altering the hardware of currently available online dialysis devices. Therefore, it is recommended that this technique be integrated into all hemodiafiltration machines. © 2014 S. Karger AG, Basel
Introduction
Volume management is a central issue of dialysis therapy with the aim to normalize the patient’s blood pressure and blood volume and to avoid excessive changes during inter- and intradialytic treatment phases. Both chronic expansion and acute depletion of intravascular volume lead to long-term end-organ damage [1, 2]. Information on absolute blood volume and a better fluid management in hemodialysis could substantially improve patient outcome [3]. A simple method to deterJoachim Kron, MD KfH Kidney Center Berlin-Köpenick Erwin-Bock-Strasse 5 DE–12559 Berlin (Germany) E-Mail joachim.kron @ kfh-dialyse.de
Downloaded by: University of Tokyo 157.82.153.40 - 5/17/2015 2:44:31 PM
Key Words Absolute blood volume · Blood volume monitoring · Hemodialysis · Intradialytic morbid events · Volume management
Materials and Methods Theory Relative blood volume monitoring is a standard feature of modern dialysis devices. Furthermore, defined volumes of ultrapure dialysate can automatically be injected into the extracorporeal circulation without direct manipulation of fluids and blood lines to treat symptomatic intradialytic hypotension as an emergency function. The dilution of blood and the calculated increase in relative blood volume (in %) caused by the infusion of ultra-pure dialysate can be used to determine the absolute blood volume (V, in l) at the time of infusion as: V=
Vbolus 1 , RBVpost RBVpre 10
(1)
where Vbolus is the volume of the dialysate bolus (in ml), and RBVpre and RBVpost refer to the relative blood volume (in %) before and after infusion. Blood volume at any other treatment time (Vt, in l) is given as:
Clinical Procedure The study was performed with the 5008 online hemodiafiltration machine (FMC, Bad Homburg, Germany). This device includes a blood volume monitor (BVM) to continuously record the RBV by ultrasonic means [14]. The 5008 dialysis machine offers the possibility to inject a defined volume of ultra-pure dialysate into the venous blood line as one continuous bolus. The magnitude of the bolus volume can be preset in multiples of 30 ml. For the purpose of this study and to maximize the dilution effect and the resulting RBV increase, the maximum infusion volume of 240 ml was chosen. Treatments were started without UF. After having attained a stable relative blood volume display, the infusion of 240 ml dialysate was started by pressing the so-called emergency button on the keypad of the dialysis machine. The infusate flow was automatically set at 200 ml/min; the blood flow was automatically reduced from 300 to 50 ml/min during the bolus administration. Relative blood volume data before and after bolus administration were manually recorded from the blood volume monitor and absolute blood volume was calculated using Eq. 1. There was no read-out of BVM data during dialysate infusion, which lasted 1.2 min as the display of rapid changes in RBV caused by the infusion was blocked by machine software. During that time, the indicator added to venous blood recirculated through the heart and lungs because of cardio-pulmonary recirculation as all patients were dialyzed using a peripheral arterio-venous access [15]. Any recirculation of indicator resulting from access complications would also have occurred during that phase. However, recirculation of indicator can be assumed to dissipate within 1 to 2 min after the end of infusion, comparable to the dissipation of urea gradients following an abrupt stop of extracorporeal clearance [16]. One to two minutes after the end of infusion, after dissipation of effects caused by recirculation and after having reached a stable relative blood volume, data display was un-blocked so that RBVpost could be recorded manually. After the test, UF was started to remove the prescribed UF volume as well as the volume infused for the blood volume test. Apart from this variation, patients underwent their regularly prescribed hemodialysis treatments. All treatments were performed in midweek sessions in hemodialysis modus. For comparison, blood volume was related to dry weight body mass (in kg) to obtain specific blood volume (Vs, in ml/kg). Reproducibility Intra-individual reproducibility was tested in ten patients. For this purpose, a 240 ml bolus was applied in hourly intervals three times during the same treatment session with a constant UF rate. Bolus administration was performed as described above. UF was automatically stopped during the infusion.
where RBVt is the RBV at that time (in %). To determine volume (V0) at treatment start t0 RBVt is 100% by definition. A more detailed description of the theory is described elsewhere [13].
Patients The study was done in maintenance hemodialysis patients of the institutional dialysis program and approved by the local ethics committee. All patients provided written informed consent to participate in this study. Thirty stable chronic hemodialysis patients (11 females) with a mean age of 75.7 ± 10.2 years and a dialysis vintage of 59 (range 32 to 161) months were studied during hemodialysis. Renal failure resulted from diabetic nephropathy (fourteen), nephrosclerosis (six), glomerulonephritis (six), polycystic kidney disease (three),
Absolute Blood Volume in Hemodialysis Patients
Blood Purif 2014;38:180–187 DOI: 10.1159/000368157
Vt = V
RBVt , RBVpre
(2)
181
Downloaded by: University of Tokyo 157.82.153.40 - 5/17/2015 2:44:31 PM
mine absolute blood volume in everyday practice has not been available so far. Online measurement of hemoconcentration and the derivation of relative blood volume (RBV) from such measurements is a standard feature of modern dialysis devices. However, changes in relative blood volume do not translate into absolute blood volume expected to vary in dialysis patients depending on their individual actual volume status [3, 4]. The same drop in relative blood volume may result in considerable different postdialysis absolute blood volumes in treatments, even when observed in the same patient [5]. About 20 years ago, several investigators tried to estimate the absolute blood volume analyzing the relative blood volume curve before and after an abrupt change in ultrafiltration (UF) rate [3, 6–9]. Suffering many limitations [10], this approach has not been transferred into clinical practice. Later, the controlled infusion of ultrapure dialysate during online hemodiafiltration has been suggested as an alternative [11–13]. The aim of this study was to evaluate this infusion technique with available dialysis equipment under routine conditions in clinical practice, and to analyze its clinical application.
105 104 103 102
Statistics Data are provided as mean ± standard deviation (SD). Significance of differences was assessed by Student t-test. A probability p < 0.05 was assumed as significant to reject the null-hypothesis. All statistical analyses were performed with SPSS statistics 18 (SPSS Inc., Chicago, Ill., USA).
101 100 99 –10
–5
0
5
10
Time (min)
Fig. 1. In-vitro analysis. Example of dialysate dilution in the presence of ultrafiltration. Dialysate was added to the test volume at time t = 0, leading to a sharp increase in RBV. The RBV before and after the addition of the indicator volume was used for the calculation as shown in Eq. 1.
Table 1. Specific blood volume (Vs), volume excess measured by bioimpedance analysis (Ve), and systolic blood pressure (BPsys) in relation to clinical judged volume status
Results
Clinical Results The absolute blood volume as measured with the bolus method at treatment start was 6.50 ± 1.70 l, corresponding to a specific blood volume of 80.1 ± 12.8 ml/kg, respectively. These volumes were significantly lower in female (5.25 ± 1.26 l, 75.9 ± 11.3 ml/kg, respectively) patients compared to male patients (7.23 ± 1.50 l, 82.6 ± 13.2 ml/kg, respectively) (p < 0.001). At treatment end, absolute blood volume dropped to 5.84 ± 1.61 l, and specific blood volume fell to 72.0 ± 12.1 ml/kg. Specific blood volume at the beginning of dialysis significantly (p < 0.05) correlated with the volume overload measured by bioimpedance immediately before treatment (r = 0.45). There was no correlation between specific blood volume and systolic blood pressure at the beginning and at the end of dialysis (r = 0.26 and r = 0.23, respectively). Specific blood volume as well as volume excess measured by bioimpedance in patients clinically assessed as volume overloaded, euvolaemic and ‘too dry’ was signifi182
106
Blood Purif 2014;38:180–187 DOI: 10.1159/000368157
Variable
Overloaded Euvolaemic ‘Too dry’ (n = 6) (n = 19) (n = 5)
Vs, ml/kg Beginning of dialysis 96.7±9.2* 78.6±9.3† End of dialysis 87.3±8.5* 70.7±8.9† Ve, l Pre-dialysis 3.97±1.13# 2.29±1.30 BPsys, mm Hg Pre-dialysis 140.7±23.5 136.2±15.9 Post-dialysis 140.6±16.1 131.6±20.8
66.2±5.0 58.6±5.7 1.30±0.92 125.4±10.6 118.0±12.0
* p < 0.001 compared with the other groups, † p < 0.01 compared with ‘to dry’ patients, # compared with clinical euvolaemic patients p < 0.01 and with ‘too dry’ patients p < 0.01.
cantly different between these three groups (table 1). Figure 2 shows the specific blood volumes of all 30 patients according to their clinically judged volume status. Only mild intradialytic and postdialytic morbid events were observed. But only clinically stable patients had been Kron/Schneditz/Leimbach/Aign/Kron
Downloaded by: University of Tokyo 157.82.153.40 - 5/17/2015 2:44:31 PM
107
RBV (%)
In vitro Studies Additionally, in vitro studies were done to check the accuracy of the technique with a 4008B dialysis machine (FMC, Bad Homburg, Germany) equipped with a BVM [14]. Dialysate temperature was 37 ° C. Blood and dialysate flows were 250 and 300 ml/min, respectively. Approximately, four to six liters of bovine blood anticoagulated with EDTA-K2 were placed in a thermostat bath and held at a temperature of 37 ° C. The volume of the blood was measured at the beginning of the test. The volumes of the primed blood lines (145 ml) and the dialyzer blood compartment (82 ml) were added to this volume to determine the initial test volume. The test volume was subsequently corrected for any added indicator volume and removed UF volume. Testing was done with (fig. 1) and without UF. Indicator volumes ranged between 100 and 300 ml.
Color version available online
and pyelonephritis (one). Patients with cardiac or fistula dysfunction or severe volume expansion were excluded. Based on clinical judgment, 11 patients appeared either volume overloaded (n = 6) or ‘too dry’ (n = 5) but had previously refused to have their ‘dry weight’ adjusted. Before treatment, the actual fluid volume overload was also assessed by bioimpedance analysis using the body composition monitor (FMC, Bad Homburg, Germany) [17].
Color version available online
110
100
*
80
Patient
Symptoms
Vs, ml/kg
1 2 3 4 5 6 7 8
None Nausea Mild cramps None Cramps, nausea Mild cramps None Loss of voice
64 63 63 60 59 59 53 49
70
Table 3. Reproducibility of RBV increase in three subsequent dilutions measured within the same treatment session
60
*
50
40
Start
End
Overloaded
Start
End
Euvolemic
Start
End
Too dry
Fig. 2. Specific blood volumes (Vs) at the beginning (Start, open circles) and at the end (End, full circles) of dialysis in 6 clinically volume overloaded, 19 euvolemic, and 5 ‘too dry’ patients. The bar indicates mean group values. Notice two euvolemic patients (*) differed from volume status based on clinical judgment and measured by bioimpedance.
included in this study. Morbid events occurred in 5 of 8 patients with blood volume below 65 ml/kg at the end of the treatment. Table 2 shows the symptoms in all patients with a blood volume below this value. Morbid events were absent when specific blood volume remained above 65 ml/kg. Reproducibility The response of relative blood volume to repeated bolus administration within the same treatment session was highly reproducible. The average increase in relative blood volume (RBVpost – RBVpre, in %) was 5.1 ± 0.19% (standard deviation range 0.06 to 0.51%), resulting in a coefficient of variation of 3.97% (table 3).
Patient
RBVpost – RBVpre, %
SD
CV, %
1 2 3 4 5 6 7 8 9 10 Average
5.60 5.90 5.13 3.13 3.76 5.47 4.03 5.63 5.90 5.80 5.04
0.20 0.17 0.51 0.15 0.25 0.06 0.25 0.06 0.10 0.10 0.19
3.6 2.9 9.9 4.9 6.7 1.1 6.2 1.0 1.7 1.7 3.97
CV = Coefficient of variation.
analysis between measured (Vdis) and experimental blood volumes (V) gave the following relationship: Vdis = 1.014*V – 0.004; r2 = 0.97 (fig. 3, left panel). The mean absolute difference between measured distribution volumes and experimental blood volumes was 0.066 ± 0.107 l (n = 49) and not different from zero (fig. 3, center panel). This corresponded to a mean relative difference of 1.29 ± 2.07%. The relative error of the measurement decreased as indicator volume increased (fig. 3, right panel).
Discussion
In vitro Studies A total of 49 dilutions were done in five experimental setups using indicator volumes of 100, 200, and 300 ml with and without concomitant UF. Linear regression
This study shows that absolute blood volume during hemodialysis can be easily determined by infusion of a well-defined volume of ultra-pure dialysate into the extracorporeal circulation and concomitant measurement
Absolute Blood Volume in Hemodialysis Patients
Blood Purif 2014;38:180–187 DOI: 10.1159/000368157
183
Downloaded by: University of Tokyo 157.82.153.40 - 5/17/2015 2:44:31 PM
Vs (ml/kg)
90
Table 2. Intradialytic morbid events in 8 patients with specific blood volume (Vs) below 65 ml/kg at treatment end
0.4
5.0
–0.2
4.0
–0.4 3.5
4.0
4.5
5.0
5.5
6.0
6.5
Mean
0
4.5
3.5
6.0 +2 SD
0.2 ©V (l)
Vdis (l)
5.5
8.0
–0.6 3.5
V (l)
–2 SD
4.0
4.5
5.0
5.5
Vmean (l)
sured distribution volume (Vdis, in l) compared to experimental blood volume (V, in l). The full line indicates the line of identity; the broken line indicates the linear regression line. Center panel: Bland-Altman analysis of measured and experimental blood volumes. The difference Vdis – V (ΔV, y-axis) is plotted as a function
of relative blood volume changes using the blood volume monitor. Normal saline has been used to measure the cardiac output and blood volume in experimental applications [18]. Some have labeled this saline bolus technique to determine absolute blood volume a ‘poor man’s Swan Ganz’ [19]. The additional fluid administration might have prevented a systematic investigation of this method in hemodialysis patients in the past. With modern online hemodiafiltration devices, ultrapure dialysate can now directly be infused into the patient. In comparison with saline, dialysate has the important advantage that it is a balanced solution and available at the proper temperature and osmotic concentration during dialysis, that it can be precisely delivered without the manipulation of blood lines and solutions using the intrinsic volume balance system [13] when activating the emergency function of the dialysis machines, and that does not infer additional costs. Data of absolute blood volume in hemodialysis patients are rare. There are significant variations depending on individual volume status, fluid distribution between body compartments, and the duration of the interdialytic period [20, 21]. Attempts to quantify absolute blood volume during hemodialysis have been limited by the lack of suitable methods. Estimates of blood volume based on anthropometric data are derived from databases including normal healthy individuals (about 70 ml/kg body Blood Purif 2014;38:180–187 DOI: 10.1159/000368157
2.0 0 –2.0
Fig. 3. In-vitro accuracy (n = 49). Left panel: Identity plot of mea-
184
4.0
6.0
6.5
–4.0
50
100
150 200 250 Vbolus (ml)
300
350
of average volumes (Vdis + V)/2 (Vmean, x-axis). The average difference (Mean) and the range covered by two standard differences (±2 SD) are shown by the broken lines. Right panel: The relative error (E, in %) between measured and experimental volumes decreased as the infusion volume (Vbolus) increased.
weight, or 1/13 body weight). These indirect calculations significantly underestimate directly measured blood volume in hemodialysis patients, especially at the upper end of the range [22]. Mitra et al. found about 20% higher blood volumes with a direct measurement using the indocyanine green in comparison with four different anthropometric estimates [22]. More recently, in a small study exploring a new online indocyanine measurement specific blood volume was only 60.9 ± 10.2 ml/kg dry body weight when measured half-away into dialysis [23]. In a parallel study done in five patients using dialysate dilution and analysis of continuously recorded BVM data, the initial specific blood volume was 77.9 ± 12.9 ml/kg [13]. Specific blood volumes of our patients are well comparable to data measured with radio-labeled markers [24, 25]. In 10 patients, Puri et al. measured specific blood volumes of 75 and 67 ml/kg before and after dialysis, respectively [24], while Dasselaar et al. measured 81 ml/kg pre and 67 ml/kg post dialysis in 7 patients [25] when their data are normalized for body mass. However, indocyanine green and radioisotope methods are too expensive and impractical for everyday clinical routine. In contrast, the dialysate bolus method presented here is easy and feasible to measure the absolute blood volume in everyday clinical practice as the delivery of a dialysate bolus is a standard feature of online hemodiafiltration Kron/Schneditz/Leimbach/Aign/Kron
Downloaded by: University of Tokyo 157.82.153.40 - 5/17/2015 2:44:31 PM
6.0
Color version available online
0.6
E (%)
6.5
The described method has some limitations. First, absolute blood volume is measured at the time of bolus infusion and therefore does not exactly correspond to the blood volume before dialysis. With the connection to the extracorporeal circulation, the distribution space for the infused bolus is expanded by the volume of the extracorporeal circulation. Second, the distribution of red blood cells is not constant throughout the vasculature. The hematocrit is lower in the microcirculation and blood with a low hematocrit translocates from the microcirculation to the macrocirculation to prevent central hypovolemia. Dasselaar et al., found a rise of the so-called F-cell ratio from 0.90 to 0.99 during hemodialysis [25]. The total blood volume decreased twice as much as that derived from hemoconcentration measured in extracorporeal blood lines and assuming in a single blood compartment. The discrepancy varies considerably between individuals [25]. The technique presented here underestimates the blood volume reduction in the peripheral compartment occurring during the late phase of UF and dialysis. This explains the discrepancy with results obtained by Dasselaar et al. that reflect the changes of absolute total blood volume in both compartments. However, for hemodynamic stability, it is more important to maintain central blood volume, while the peripheral volume is allowed to decrease to larger extent so that the importance of total absolute volume (i.e., the sum of the central and peripheral blood volume) reduction can be questioned [27, 28]. In the study of Dasselaar et al. [25] specific blood volume was 81 ml/kg (derived from absolute volume and body mass) before and 67 ml/kg after dialysis. While the pre-dialysis value almost exactly matches our data, our post-dialysis data at 72 ml/kg appear to exceed their value. The discrepancy can be explained by the systematic underestimation of relative blood volume changes using data on central hemoconcentration as discussed above. Our data are more reflective of the central absolute blood volume, but this is of clinical importance. Nevertheless, the advantage of our method is the applicability in everyday clinical routine with the potential to prevent inappropriate blood volume reductions and to avoid intradialytic morbid events. The described technique for measuring the absolute blood volume opens some important new perspectives. First of all, we suggest that this method be used to identify critical thresholds for specific blood volumes to avoid intradialytic morbid events. A specific blood volume of 65 ml/kg seemed to represent this critical threshold for mild morbid events in our study. But only clinically stable patients were included in this study. Further research es-
Absolute Blood Volume in Hemodialysis Patients
Blood Purif 2014;38:180–187 DOI: 10.1159/000368157
185
Downloaded by: University of Tokyo 157.82.153.40 - 5/17/2015 2:44:31 PM
machines. The procedure is safe, noninvasive, and inexpensive and absolute blood volumes are obtainable at the beginning of every dialysis session. Our clinical and experimental data show that bolus volume of 240 ml proved to be a reasonable compromise between accuracy, volume load, and technical scope of the dialysis machine. None of the patients had any complaints during or after the bolus infusion. In principle, a lower bolus volume is possible, but then the accuracy is also expected to be lower (fig. 3). The technique has the potential for complete automation without alteration of dialysis machine hardware. A software modification would be sufficient to automate the procedure and to provide a continuous display of actual blood volume data. Moreover, this feature could be implemented into feedback-controlled UF programs. The approach described here is not limited to the used equipment but can be applied to all online dialysis machines with an incorporated device for relative blood volume measurement. Therefore, the implementation by automated means into all these machines is recommended. Absolute blood volumes measured in our study are clinically plausible. They correlate with volume status assessed both by clinical judgment and bioimpedance analysis. Only stable patients were included in our study so that severe morbid events were not expected. Nevertheless, a specific blood volume of 65 ml/kg seemed to represent a critical threshold for mild events. This is close to the 64 ml/kg threshold measured many years ago [26]. The new method is highly reproducible with a coefficient of variation of better than 4%. This is comparable to the indocyanine green (4.1%) [22] and better than radioisotope dilution techniques (8.1%) [9]. With an actual variation of ±3 ml/kg blood volume, the accuracy of this method appears sufficient for practical purposes. In two patients, however, absolute blood volumes did not match clinical symptoms. In another patient, the coefficient of variation was almost 10%. Manual analysis of relative blood volume data presents a potential weakness of the method used in this study. However, more accurate measurements can be expected if the entire dilution curve is analyzed rather than relying on only two RBV values, as shown for in-vitro measurements (fig. 1) or in a more technical companion paper [13]. Transferring the accuracy of our in vitro data to the clinical setting, this corresponds to a resolution of less than ± 1 ml/kg of specific blood volume. Analysis of continuous data has the potential to provide both flexibility and accuracy for the period of measurement [13]. This could easily be installed by a software modification.
pecially in hypotension-prone patients is required to characterize such critical values. In connection with a blood volume-controlled UF feedback program [29], this may prevent intradialytic morbid events more efficiently than earlier times [30]. Second, the assessment of dry weight remains a difficult clinical judgment. But it is the expansion of intravascular volume that causes long-term cardiovascular damage. In some of our patients assessed as ‘euvolaemic,’ specific blood volume remained elevated even at the end of dialysis. Routine use of our method might initiate the concept of ‘dry blood volume.’ This could supplement or even substitute the current ‘dry weight’ concept [3, 9]. For example, a target blood volume could be prescribed at the beginning of a treatment. Feedback programs could prevent blood volume from falling below a critical threshold by continuously adjusting the UF rate to reach the target blood volume [29]. At the end of every dialysis session, blood volume would be in the optimum state of lowest cardiovascular strain for the patient. Third, the technique could also be used in intensive care. Especially volume management in septic shock is difficult because of considerable volume loss into the interstitial space. On the one hand, volume resuscitation is essential for maintaining circulation and organ perfusion. On the other hand, extravascular fluid accumulation is associated with an impaired outcome [31]. Volume management in renal replacement therapy therefore is always caught up in the conflict of aggravating hypovolaemia by undue UF and insufficient reduction in fluid overload, thus contributing to a poor prognosis. In
this setting, the relative blood volume is only of partial help because it provides information of about the deviation from baseline blood volume. But baseline blood volume is unknown as of now. Information on specific intravascular volume could substantially improve volume management in acute kidney injury, especially in septic shock. And last but not the least, the method could be used as an additional tool to study the distribution and the dynamics of extracellular fluid between intra- and extravascular compartments under different conditions, for example with different degrees of volume load, distribution in the short or long interdialytic period [20, 21], or in patients with heart failure. In conclusion, the described method is simple, safe, inexpensive, noninvasive, and feasible using current dialysis technology in everyday clinical practice. The technique has the potential for complete automation with appropriate modifications in system software. Additional changes in hardware are not required. The absolute blood volume measured at the beginning of every dialysis session offers the opportunity to significantly improve fluid management in hemodialysis and related applications. Therefore, we call upon manufacturers to implement this technique into machines used for treatment of both chronic and acute renal failure.
Disclosure Statement No conflicts of interest to declare.
References
186
7
8
9
10
Blood Purif 2014;38:180–187 DOI: 10.1159/000368157
measurement of blood volume during haemodialysis. Life Support Syst 1987; 5: 293– 305. Schneditz D, Roob J, Oswald M, Pogglitsch H, Moser M, Kenner T, Binswanger U: Nature and rate of vascular refilling during hemodialysis and ultrafiltration. Kidney Int 1992;42: 1425–1433. Leypoldt JK, Cheung AK, Steuer RR, Harris DH, Conis JM: Determination of circulating blood volume by continuously monitoring hematocrit during hemodialysis. J Am Soc Nephrol 1995;6:214–219. Johnson DW, McMahon M, Campbell S, Wilkinson J, Kime N, Shannon G, Fleming SJ: Non-invasive, optical measurement of absolute blood volume in hemodialysis patients. Kidney Int 1996;49:255–260. Schneditz D, Roob JM, Vaclavik M, Holzer H, Kenner T: Noninvasive measurement of
11
12
13
14
blood volume in hemodialysis patients. J Am Soc Nephrol 1996;7:1241–1244. Schneditz D, Wimmer J, Haditsch B, Holzer H: Measurement of systemic blood flow by slow infusion of ultra-pure dialysate during hemodialysis. J Am Soc Nephrol 2003; 14: 213A. Wimmer J, Haditsch B, Holzer H, Schneditz D: Measurement of central blood volume by slow infusion of ultra-pure dialysate during hemodialysis. J Am Soc Nephrol 2003;14:851A. Schneditz D, Schilcher G, Ribitsch W, Krisper P, Haditsch B, Kron J: On-line dialysate infusion to estimate absolute blood volume in dialysis patients. ASAIO J 2014;60:436–442. Schneditz D, Pogglitsch H, Horina J, Binswanger U: A blood protein monitor for the continuous measurement of blood volume changes during hemodialysis. Kidney Int 1990; 38:342–346.
Kron/Schneditz/Leimbach/Aign/Kron
Downloaded by: University of Tokyo 157.82.153.40 - 5/17/2015 2:44:31 PM
1 Agarwal R: Volume overload in dialysis: the elephant in the room, no one can see. Am J Nephrol 2013;38:75–77. 2 Daugirdas JT: Pathophysiology of dialysis hypotension: an update. Am J Kidney Dis 2001; 38(suppl 4):S11–S17. 3 Thijssen S, Kappel F, Kotanko P: Absolute blood volume in hemodialysis patients: why is it relevant, and how to measure it? Blood Purif 2013;35:63–71. 4 Booth J, Pinney J, Davenport A: Do changes in relative blood volume monitoring correlate to hemodialysis-associated hypotension? Nephron Clin Pract 2011;117:c179–c183. 5 Dasselaar JJ, van der Sande FM, Franssen CF: Critical evaluation of blood volume measurements during hemodialysis. Blood Purif 2012; 33:177–182. 6 Schallenberg U, Stiller S, Mann H: A new method of continuous haemoglobinometric
Absolute Blood Volume in Hemodialysis Patients
21 Dasselaar JJ, de Jong PE, Huisman RM, Franssen CF: Influence of ultrafiltration volume on blood volume changes during hemodialysis as observed in day-of-the-week analysis of hemodialysis sessions. ASAIO J 2007; 53: 479– 484. 22 Mitra S, Chamney P, Greenwood R, Farrington K: Serial determinations of absolute plasma volume with indocyanine green during hemodialysis. J Am Soc Nephrol 2003;14: 2345–2351. 23 Schneditz D, Haditsch B, Jantscher A, Ribitsch W, Krisper P: Absolute blood volume and hepato-splanchnic blood flow measured by indocyanine green kinetics during hemodialysis. ASAIO J 2014;60:452–458. 24 Puri S, Park JK, Modersitzki F, Goldfarb DS: Radioisotope blood volume measurement in hemodialysis patients. Hemodial Int 2014;18: 406–414. 25 Dasselaar JJ, Lub-de Hooge MN, Pruim J, Nijnuis H, Wiersum A, de Jong PE, Huisman RM, Franssen CF: Relative blood volume changes underestimate total blood volume changes during hemodialysis. Clin J Am Soc Nephrol 2007;2:669–674.
26 Kim KE, Neff M, Cohen B, Somerstein M, Chinitz J, Onesti G, Swartz C: Blood volume changes and hypotension during hemodialysis. Trans Am Soc Artif Intern Organs 1970; 16:508–514. 27 Secher NH, Van Lieshout JJ: Normovolaemia defined by central blood volume and venous oxygen saturation. Clin Exp Pharmacol Physiol 2005;32:901–910. 28 Shulman T, Heidenheim AP, Kianfar C, Shulman SM, Lindsay RM: Preserving central blood volume: changes in body fluid compartments during hemodialysis. ASAIO J 2001;47:615–618. 29 Gabrielli D, Krystal B, Katzarski K, Youssef M, Hachache T, Lopot F, Lasseur C, Gunne T, Draganov B, Wojke R, Gauly A: Improved intradialytic stability during haemodialysis with blood volume-controlled ultrafiltration. J Nephrol 2009;22:232–240. 30 Nesrallah GE, Suri RS, Guyatt G, Mustafa RA, Walter SD, Lindsay RM, Akl EA: Biofeedback dialysis for hypotension and hypervolemia: a systematic review and meta-analysis. Nephrol Dial Transplant 2013;28:182–191. 31 Bagshaw SM, Brophy PD, Cruz D, Ronco C: Fluid balance as a biomarker: impact of fluid overload on outcome in critically ill patients with acute kidney injury. Crit Care 2008; 12: 169.
Blood Purif 2014;38:180–187 DOI: 10.1159/000368157
187
Downloaded by: University of Tokyo 157.82.153.40 - 5/17/2015 2:44:31 PM
15 Schneditz D: Recirculation, a seemingly simple concept. Nephrol Dial Transplant 1998; 13:2191–2193. 16 Schneditz D, Kaufman AM, Polaschegg HD, Levin NW, Daugirdas JT: Cardiopulmonary recirculation during hemodialysis. Kidney Int 1992;42:1450–1456. 17 Moissl UM, Wabel P, Chamney PW, Bosaeus I, Levin NW, Bosy-Westphal A, Korth O, Müller MJ, Ellegard L, Malmros V, Kaitwatcharachai C, Kuhlmann MK, Zhu F, Fuller NJ: Body fluid volume determination via body composition spectroscopy in health and disease. Physiol Meas 2006; 27: 921–933. 18 Krivitski NM, Depner TA: Cardiac output and central blood volume during hemodialysis: methodology. Adv Ren Replace Ther 1999;6:225–232. 19 Steuer RR, Bell DA, Barrett LL: Optical measurement of hematocrit and other biological constituents in renal therapy. Adv Ren Replace Ther 1999;6:217–224. 20 Movilli E, Cancarini GC, Cassamali S, Camerini C, Brunori G, Maffei C, Maiorca R: Inter-dialytic variations in blood volume and total body water in uraemic patients treated by dialysis. Nephrol Dial Transplant 2004;19: 185–189.
Blood Purif 2014;38:180–187 DOI: 10.1159/000368157
A Simple and Feasible Method to Determine Absolute Blood Volume in Hemodialysis Patients in Clinical Practice Joachim Kron a Daniel Schneditz b Til Leimbach a Sabine Aign a Susanne Kron c
a
KfH Kidney Center Berlin-Köpenick, Berlin, Germany; b Institute of Physiology, Medical University of Graz, Graz, Austria; c Department of Nephrology, Charite Universitätsmedizin Berlin, Berlin, Germany
Abstract Background: We developed a simple method to determine the absolute blood volume (V) during hemodialysis in everyday clinical practice and examined its relationship with volume overload, clinical relevance, and accuracy. Methods: The increase in relative blood volume (RBVpost – RBVpre) measured before and after infusion of 240 ml of ultra-pure dialysate using the bolus function of a commercial online hemodiafiltration machine incorporating a relative blood volume monitor was applied to determine absolute blood volume. The specific blood volume (Vs, blood volume per kg body mass at dry weight, in ml/kg) was compared to volume status as assessed by bioimpedance analysis and clinical criteria. Results: The blood volume measured in 30 stable hemodialysis patients was 6.51 ± 1.70 l at the beginning, corresponding to a specific blood volume of 80.1 ± 12.8 ml/kg, and dropped to 5.84 ± 1.61 l or 72.0 ± 12.1 ml/kg at the end of the dialysis session, respectively. Specific blood volume correlated with volume status assessed both clinically and by bioimpedance analysis. Intradialytic morbid events oc-
© 2014 S. Karger AG, Basel 0253–5068/14/0384–0180$39.50/0 E-Mail [email protected] www.karger.com/bpu
curred only in treatments where specific blood volume fell below 65 ml/kg. The reproducibility of the technique was better than 4% and the in vitro accuracy corresponds to a resolution in Vs of better than 1 ml/kg. Conclusion: Absolute blood volume can be easily measured at the beginning of the dialysis session using the current dialysis technology. Information about V and Vs could be a promising tool to avoid intradialytic morbid events. This technique could be completely automated without altering the hardware of currently available online dialysis devices. Therefore, it is recommended that this technique be integrated into all hemodiafiltration machines. © 2014 S. Karger AG, Basel
Introduction
Volume management is a central issue of dialysis therapy with the aim to normalize the patient’s blood pressure and blood volume and to avoid excessive changes during inter- and intradialytic treatment phases. Both chronic expansion and acute depletion of intravascular volume lead to long-term end-organ damage [1, 2]. Information on absolute blood volume and a better fluid management in hemodialysis could substantially improve patient outcome [3]. A simple method to deterJoachim Kron, MD KfH Kidney Center Berlin-Köpenick Erwin-Bock-Strasse 5 DE–12559 Berlin (Germany) E-Mail joachim.kron @ kfh-dialyse.de
Downloaded by: University of Tokyo 157.82.153.40 - 5/17/2015 2:44:31 PM
Key Words Absolute blood volume · Blood volume monitoring · Hemodialysis · Intradialytic morbid events · Volume management
Materials and Methods Theory Relative blood volume monitoring is a standard feature of modern dialysis devices. Furthermore, defined volumes of ultrapure dialysate can automatically be injected into the extracorporeal circulation without direct manipulation of fluids and blood lines to treat symptomatic intradialytic hypotension as an emergency function. The dilution of blood and the calculated increase in relative blood volume (in %) caused by the infusion of ultra-pure dialysate can be used to determine the absolute blood volume (V, in l) at the time of infusion as: V=
Vbolus 1 , RBVpost RBVpre 10
(1)
where Vbolus is the volume of the dialysate bolus (in ml), and RBVpre and RBVpost refer to the relative blood volume (in %) before and after infusion. Blood volume at any other treatment time (Vt, in l) is given as:
Clinical Procedure The study was performed with the 5008 online hemodiafiltration machine (FMC, Bad Homburg, Germany). This device includes a blood volume monitor (BVM) to continuously record the RBV by ultrasonic means [14]. The 5008 dialysis machine offers the possibility to inject a defined volume of ultra-pure dialysate into the venous blood line as one continuous bolus. The magnitude of the bolus volume can be preset in multiples of 30 ml. For the purpose of this study and to maximize the dilution effect and the resulting RBV increase, the maximum infusion volume of 240 ml was chosen. Treatments were started without UF. After having attained a stable relative blood volume display, the infusion of 240 ml dialysate was started by pressing the so-called emergency button on the keypad of the dialysis machine. The infusate flow was automatically set at 200 ml/min; the blood flow was automatically reduced from 300 to 50 ml/min during the bolus administration. Relative blood volume data before and after bolus administration were manually recorded from the blood volume monitor and absolute blood volume was calculated using Eq. 1. There was no read-out of BVM data during dialysate infusion, which lasted 1.2 min as the display of rapid changes in RBV caused by the infusion was blocked by machine software. During that time, the indicator added to venous blood recirculated through the heart and lungs because of cardio-pulmonary recirculation as all patients were dialyzed using a peripheral arterio-venous access [15]. Any recirculation of indicator resulting from access complications would also have occurred during that phase. However, recirculation of indicator can be assumed to dissipate within 1 to 2 min after the end of infusion, comparable to the dissipation of urea gradients following an abrupt stop of extracorporeal clearance [16]. One to two minutes after the end of infusion, after dissipation of effects caused by recirculation and after having reached a stable relative blood volume, data display was un-blocked so that RBVpost could be recorded manually. After the test, UF was started to remove the prescribed UF volume as well as the volume infused for the blood volume test. Apart from this variation, patients underwent their regularly prescribed hemodialysis treatments. All treatments were performed in midweek sessions in hemodialysis modus. For comparison, blood volume was related to dry weight body mass (in kg) to obtain specific blood volume (Vs, in ml/kg). Reproducibility Intra-individual reproducibility was tested in ten patients. For this purpose, a 240 ml bolus was applied in hourly intervals three times during the same treatment session with a constant UF rate. Bolus administration was performed as described above. UF was automatically stopped during the infusion.
where RBVt is the RBV at that time (in %). To determine volume (V0) at treatment start t0 RBVt is 100% by definition. A more detailed description of the theory is described elsewhere [13].
Patients The study was done in maintenance hemodialysis patients of the institutional dialysis program and approved by the local ethics committee. All patients provided written informed consent to participate in this study. Thirty stable chronic hemodialysis patients (11 females) with a mean age of 75.7 ± 10.2 years and a dialysis vintage of 59 (range 32 to 161) months were studied during hemodialysis. Renal failure resulted from diabetic nephropathy (fourteen), nephrosclerosis (six), glomerulonephritis (six), polycystic kidney disease (three),
Absolute Blood Volume in Hemodialysis Patients
Blood Purif 2014;38:180–187 DOI: 10.1159/000368157
Vt = V
RBVt , RBVpre
(2)
181
Downloaded by: University of Tokyo 157.82.153.40 - 5/17/2015 2:44:31 PM
mine absolute blood volume in everyday practice has not been available so far. Online measurement of hemoconcentration and the derivation of relative blood volume (RBV) from such measurements is a standard feature of modern dialysis devices. However, changes in relative blood volume do not translate into absolute blood volume expected to vary in dialysis patients depending on their individual actual volume status [3, 4]. The same drop in relative blood volume may result in considerable different postdialysis absolute blood volumes in treatments, even when observed in the same patient [5]. About 20 years ago, several investigators tried to estimate the absolute blood volume analyzing the relative blood volume curve before and after an abrupt change in ultrafiltration (UF) rate [3, 6–9]. Suffering many limitations [10], this approach has not been transferred into clinical practice. Later, the controlled infusion of ultrapure dialysate during online hemodiafiltration has been suggested as an alternative [11–13]. The aim of this study was to evaluate this infusion technique with available dialysis equipment under routine conditions in clinical practice, and to analyze its clinical application.
105 104 103 102
Statistics Data are provided as mean ± standard deviation (SD). Significance of differences was assessed by Student t-test. A probability p < 0.05 was assumed as significant to reject the null-hypothesis. All statistical analyses were performed with SPSS statistics 18 (SPSS Inc., Chicago, Ill., USA).
101 100 99 –10
–5
0
5
10
Time (min)
Fig. 1. In-vitro analysis. Example of dialysate dilution in the presence of ultrafiltration. Dialysate was added to the test volume at time t = 0, leading to a sharp increase in RBV. The RBV before and after the addition of the indicator volume was used for the calculation as shown in Eq. 1.
Table 1. Specific blood volume (Vs), volume excess measured by bioimpedance analysis (Ve), and systolic blood pressure (BPsys) in relation to clinical judged volume status
Results
Clinical Results The absolute blood volume as measured with the bolus method at treatment start was 6.50 ± 1.70 l, corresponding to a specific blood volume of 80.1 ± 12.8 ml/kg, respectively. These volumes were significantly lower in female (5.25 ± 1.26 l, 75.9 ± 11.3 ml/kg, respectively) patients compared to male patients (7.23 ± 1.50 l, 82.6 ± 13.2 ml/kg, respectively) (p < 0.001). At treatment end, absolute blood volume dropped to 5.84 ± 1.61 l, and specific blood volume fell to 72.0 ± 12.1 ml/kg. Specific blood volume at the beginning of dialysis significantly (p < 0.05) correlated with the volume overload measured by bioimpedance immediately before treatment (r = 0.45). There was no correlation between specific blood volume and systolic blood pressure at the beginning and at the end of dialysis (r = 0.26 and r = 0.23, respectively). Specific blood volume as well as volume excess measured by bioimpedance in patients clinically assessed as volume overloaded, euvolaemic and ‘too dry’ was signifi182
106
Blood Purif 2014;38:180–187 DOI: 10.1159/000368157
Variable
Overloaded Euvolaemic ‘Too dry’ (n = 6) (n = 19) (n = 5)
Vs, ml/kg Beginning of dialysis 96.7±9.2* 78.6±9.3† End of dialysis 87.3±8.5* 70.7±8.9† Ve, l Pre-dialysis 3.97±1.13# 2.29±1.30 BPsys, mm Hg Pre-dialysis 140.7±23.5 136.2±15.9 Post-dialysis 140.6±16.1 131.6±20.8
66.2±5.0 58.6±5.7 1.30±0.92 125.4±10.6 118.0±12.0
* p < 0.001 compared with the other groups, † p < 0.01 compared with ‘to dry’ patients, # compared with clinical euvolaemic patients p < 0.01 and with ‘too dry’ patients p < 0.01.
cantly different between these three groups (table 1). Figure 2 shows the specific blood volumes of all 30 patients according to their clinically judged volume status. Only mild intradialytic and postdialytic morbid events were observed. But only clinically stable patients had been Kron/Schneditz/Leimbach/Aign/Kron
Downloaded by: University of Tokyo 157.82.153.40 - 5/17/2015 2:44:31 PM
107
RBV (%)
In vitro Studies Additionally, in vitro studies were done to check the accuracy of the technique with a 4008B dialysis machine (FMC, Bad Homburg, Germany) equipped with a BVM [14]. Dialysate temperature was 37 ° C. Blood and dialysate flows were 250 and 300 ml/min, respectively. Approximately, four to six liters of bovine blood anticoagulated with EDTA-K2 were placed in a thermostat bath and held at a temperature of 37 ° C. The volume of the blood was measured at the beginning of the test. The volumes of the primed blood lines (145 ml) and the dialyzer blood compartment (82 ml) were added to this volume to determine the initial test volume. The test volume was subsequently corrected for any added indicator volume and removed UF volume. Testing was done with (fig. 1) and without UF. Indicator volumes ranged between 100 and 300 ml.
Color version available online
and pyelonephritis (one). Patients with cardiac or fistula dysfunction or severe volume expansion were excluded. Based on clinical judgment, 11 patients appeared either volume overloaded (n = 6) or ‘too dry’ (n = 5) but had previously refused to have their ‘dry weight’ adjusted. Before treatment, the actual fluid volume overload was also assessed by bioimpedance analysis using the body composition monitor (FMC, Bad Homburg, Germany) [17].
Color version available online
110
100
*
80
Patient
Symptoms
Vs, ml/kg
1 2 3 4 5 6 7 8
None Nausea Mild cramps None Cramps, nausea Mild cramps None Loss of voice
64 63 63 60 59 59 53 49
70
Table 3. Reproducibility of RBV increase in three subsequent dilutions measured within the same treatment session
60
*
50
40
Start
End
Overloaded
Start
End
Euvolemic
Start
End
Too dry
Fig. 2. Specific blood volumes (Vs) at the beginning (Start, open circles) and at the end (End, full circles) of dialysis in 6 clinically volume overloaded, 19 euvolemic, and 5 ‘too dry’ patients. The bar indicates mean group values. Notice two euvolemic patients (*) differed from volume status based on clinical judgment and measured by bioimpedance.
included in this study. Morbid events occurred in 5 of 8 patients with blood volume below 65 ml/kg at the end of the treatment. Table 2 shows the symptoms in all patients with a blood volume below this value. Morbid events were absent when specific blood volume remained above 65 ml/kg. Reproducibility The response of relative blood volume to repeated bolus administration within the same treatment session was highly reproducible. The average increase in relative blood volume (RBVpost – RBVpre, in %) was 5.1 ± 0.19% (standard deviation range 0.06 to 0.51%), resulting in a coefficient of variation of 3.97% (table 3).
Patient
RBVpost – RBVpre, %
SD
CV, %
1 2 3 4 5 6 7 8 9 10 Average
5.60 5.90 5.13 3.13 3.76 5.47 4.03 5.63 5.90 5.80 5.04
0.20 0.17 0.51 0.15 0.25 0.06 0.25 0.06 0.10 0.10 0.19
3.6 2.9 9.9 4.9 6.7 1.1 6.2 1.0 1.7 1.7 3.97
CV = Coefficient of variation.
analysis between measured (Vdis) and experimental blood volumes (V) gave the following relationship: Vdis = 1.014*V – 0.004; r2 = 0.97 (fig. 3, left panel). The mean absolute difference between measured distribution volumes and experimental blood volumes was 0.066 ± 0.107 l (n = 49) and not different from zero (fig. 3, center panel). This corresponded to a mean relative difference of 1.29 ± 2.07%. The relative error of the measurement decreased as indicator volume increased (fig. 3, right panel).
Discussion
In vitro Studies A total of 49 dilutions were done in five experimental setups using indicator volumes of 100, 200, and 300 ml with and without concomitant UF. Linear regression
This study shows that absolute blood volume during hemodialysis can be easily determined by infusion of a well-defined volume of ultra-pure dialysate into the extracorporeal circulation and concomitant measurement
Absolute Blood Volume in Hemodialysis Patients
Blood Purif 2014;38:180–187 DOI: 10.1159/000368157
183
Downloaded by: University of Tokyo 157.82.153.40 - 5/17/2015 2:44:31 PM
Vs (ml/kg)
90
Table 2. Intradialytic morbid events in 8 patients with specific blood volume (Vs) below 65 ml/kg at treatment end
0.4
5.0
–0.2
4.0
–0.4 3.5
4.0
4.5
5.0
5.5
6.0
6.5
Mean
0
4.5
3.5
6.0 +2 SD
0.2 ©V (l)
Vdis (l)
5.5
8.0
–0.6 3.5
V (l)
–2 SD
4.0
4.5
5.0
5.5
Vmean (l)
sured distribution volume (Vdis, in l) compared to experimental blood volume (V, in l). The full line indicates the line of identity; the broken line indicates the linear regression line. Center panel: Bland-Altman analysis of measured and experimental blood volumes. The difference Vdis – V (ΔV, y-axis) is plotted as a function
of relative blood volume changes using the blood volume monitor. Normal saline has been used to measure the cardiac output and blood volume in experimental applications [18]. Some have labeled this saline bolus technique to determine absolute blood volume a ‘poor man’s Swan Ganz’ [19]. The additional fluid administration might have prevented a systematic investigation of this method in hemodialysis patients in the past. With modern online hemodiafiltration devices, ultrapure dialysate can now directly be infused into the patient. In comparison with saline, dialysate has the important advantage that it is a balanced solution and available at the proper temperature and osmotic concentration during dialysis, that it can be precisely delivered without the manipulation of blood lines and solutions using the intrinsic volume balance system [13] when activating the emergency function of the dialysis machines, and that does not infer additional costs. Data of absolute blood volume in hemodialysis patients are rare. There are significant variations depending on individual volume status, fluid distribution between body compartments, and the duration of the interdialytic period [20, 21]. Attempts to quantify absolute blood volume during hemodialysis have been limited by the lack of suitable methods. Estimates of blood volume based on anthropometric data are derived from databases including normal healthy individuals (about 70 ml/kg body Blood Purif 2014;38:180–187 DOI: 10.1159/000368157
2.0 0 –2.0
Fig. 3. In-vitro accuracy (n = 49). Left panel: Identity plot of mea-
184
4.0
6.0
6.5
–4.0
50
100
150 200 250 Vbolus (ml)
300
350
of average volumes (Vdis + V)/2 (Vmean, x-axis). The average difference (Mean) and the range covered by two standard differences (±2 SD) are shown by the broken lines. Right panel: The relative error (E, in %) between measured and experimental volumes decreased as the infusion volume (Vbolus) increased.
weight, or 1/13 body weight). These indirect calculations significantly underestimate directly measured blood volume in hemodialysis patients, especially at the upper end of the range [22]. Mitra et al. found about 20% higher blood volumes with a direct measurement using the indocyanine green in comparison with four different anthropometric estimates [22]. More recently, in a small study exploring a new online indocyanine measurement specific blood volume was only 60.9 ± 10.2 ml/kg dry body weight when measured half-away into dialysis [23]. In a parallel study done in five patients using dialysate dilution and analysis of continuously recorded BVM data, the initial specific blood volume was 77.9 ± 12.9 ml/kg [13]. Specific blood volumes of our patients are well comparable to data measured with radio-labeled markers [24, 25]. In 10 patients, Puri et al. measured specific blood volumes of 75 and 67 ml/kg before and after dialysis, respectively [24], while Dasselaar et al. measured 81 ml/kg pre and 67 ml/kg post dialysis in 7 patients [25] when their data are normalized for body mass. However, indocyanine green and radioisotope methods are too expensive and impractical for everyday clinical routine. In contrast, the dialysate bolus method presented here is easy and feasible to measure the absolute blood volume in everyday clinical practice as the delivery of a dialysate bolus is a standard feature of online hemodiafiltration Kron/Schneditz/Leimbach/Aign/Kron
Downloaded by: University of Tokyo 157.82.153.40 - 5/17/2015 2:44:31 PM
6.0
Color version available online
0.6
E (%)
6.5
The described method has some limitations. First, absolute blood volume is measured at the time of bolus infusion and therefore does not exactly correspond to the blood volume before dialysis. With the connection to the extracorporeal circulation, the distribution space for the infused bolus is expanded by the volume of the extracorporeal circulation. Second, the distribution of red blood cells is not constant throughout the vasculature. The hematocrit is lower in the microcirculation and blood with a low hematocrit translocates from the microcirculation to the macrocirculation to prevent central hypovolemia. Dasselaar et al., found a rise of the so-called F-cell ratio from 0.90 to 0.99 during hemodialysis [25]. The total blood volume decreased twice as much as that derived from hemoconcentration measured in extracorporeal blood lines and assuming in a single blood compartment. The discrepancy varies considerably between individuals [25]. The technique presented here underestimates the blood volume reduction in the peripheral compartment occurring during the late phase of UF and dialysis. This explains the discrepancy with results obtained by Dasselaar et al. that reflect the changes of absolute total blood volume in both compartments. However, for hemodynamic stability, it is more important to maintain central blood volume, while the peripheral volume is allowed to decrease to larger extent so that the importance of total absolute volume (i.e., the sum of the central and peripheral blood volume) reduction can be questioned [27, 28]. In the study of Dasselaar et al. [25] specific blood volume was 81 ml/kg (derived from absolute volume and body mass) before and 67 ml/kg after dialysis. While the pre-dialysis value almost exactly matches our data, our post-dialysis data at 72 ml/kg appear to exceed their value. The discrepancy can be explained by the systematic underestimation of relative blood volume changes using data on central hemoconcentration as discussed above. Our data are more reflective of the central absolute blood volume, but this is of clinical importance. Nevertheless, the advantage of our method is the applicability in everyday clinical routine with the potential to prevent inappropriate blood volume reductions and to avoid intradialytic morbid events. The described technique for measuring the absolute blood volume opens some important new perspectives. First of all, we suggest that this method be used to identify critical thresholds for specific blood volumes to avoid intradialytic morbid events. A specific blood volume of 65 ml/kg seemed to represent this critical threshold for mild morbid events in our study. But only clinically stable patients were included in this study. Further research es-
Absolute Blood Volume in Hemodialysis Patients
Blood Purif 2014;38:180–187 DOI: 10.1159/000368157
185
Downloaded by: University of Tokyo 157.82.153.40 - 5/17/2015 2:44:31 PM
machines. The procedure is safe, noninvasive, and inexpensive and absolute blood volumes are obtainable at the beginning of every dialysis session. Our clinical and experimental data show that bolus volume of 240 ml proved to be a reasonable compromise between accuracy, volume load, and technical scope of the dialysis machine. None of the patients had any complaints during or after the bolus infusion. In principle, a lower bolus volume is possible, but then the accuracy is also expected to be lower (fig. 3). The technique has the potential for complete automation without alteration of dialysis machine hardware. A software modification would be sufficient to automate the procedure and to provide a continuous display of actual blood volume data. Moreover, this feature could be implemented into feedback-controlled UF programs. The approach described here is not limited to the used equipment but can be applied to all online dialysis machines with an incorporated device for relative blood volume measurement. Therefore, the implementation by automated means into all these machines is recommended. Absolute blood volumes measured in our study are clinically plausible. They correlate with volume status assessed both by clinical judgment and bioimpedance analysis. Only stable patients were included in our study so that severe morbid events were not expected. Nevertheless, a specific blood volume of 65 ml/kg seemed to represent a critical threshold for mild events. This is close to the 64 ml/kg threshold measured many years ago [26]. The new method is highly reproducible with a coefficient of variation of better than 4%. This is comparable to the indocyanine green (4.1%) [22] and better than radioisotope dilution techniques (8.1%) [9]. With an actual variation of ±3 ml/kg blood volume, the accuracy of this method appears sufficient for practical purposes. In two patients, however, absolute blood volumes did not match clinical symptoms. In another patient, the coefficient of variation was almost 10%. Manual analysis of relative blood volume data presents a potential weakness of the method used in this study. However, more accurate measurements can be expected if the entire dilution curve is analyzed rather than relying on only two RBV values, as shown for in-vitro measurements (fig. 1) or in a more technical companion paper [13]. Transferring the accuracy of our in vitro data to the clinical setting, this corresponds to a resolution of less than ± 1 ml/kg of specific blood volume. Analysis of continuous data has the potential to provide both flexibility and accuracy for the period of measurement [13]. This could easily be installed by a software modification.
pecially in hypotension-prone patients is required to characterize such critical values. In connection with a blood volume-controlled UF feedback program [29], this may prevent intradialytic morbid events more efficiently than earlier times [30]. Second, the assessment of dry weight remains a difficult clinical judgment. But it is the expansion of intravascular volume that causes long-term cardiovascular damage. In some of our patients assessed as ‘euvolaemic,’ specific blood volume remained elevated even at the end of dialysis. Routine use of our method might initiate the concept of ‘dry blood volume.’ This could supplement or even substitute the current ‘dry weight’ concept [3, 9]. For example, a target blood volume could be prescribed at the beginning of a treatment. Feedback programs could prevent blood volume from falling below a critical threshold by continuously adjusting the UF rate to reach the target blood volume [29]. At the end of every dialysis session, blood volume would be in the optimum state of lowest cardiovascular strain for the patient. Third, the technique could also be used in intensive care. Especially volume management in septic shock is difficult because of considerable volume loss into the interstitial space. On the one hand, volume resuscitation is essential for maintaining circulation and organ perfusion. On the other hand, extravascular fluid accumulation is associated with an impaired outcome [31]. Volume management in renal replacement therapy therefore is always caught up in the conflict of aggravating hypovolaemia by undue UF and insufficient reduction in fluid overload, thus contributing to a poor prognosis. In
this setting, the relative blood volume is only of partial help because it provides information of about the deviation from baseline blood volume. But baseline blood volume is unknown as of now. Information on specific intravascular volume could substantially improve volume management in acute kidney injury, especially in septic shock. And last but not the least, the method could be used as an additional tool to study the distribution and the dynamics of extracellular fluid between intra- and extravascular compartments under different conditions, for example with different degrees of volume load, distribution in the short or long interdialytic period [20, 21], or in patients with heart failure. In conclusion, the described method is simple, safe, inexpensive, noninvasive, and feasible using current dialysis technology in everyday clinical practice. The technique has the potential for complete automation with appropriate modifications in system software. Additional changes in hardware are not required. The absolute blood volume measured at the beginning of every dialysis session offers the opportunity to significantly improve fluid management in hemodialysis and related applications. Therefore, we call upon manufacturers to implement this technique into machines used for treatment of both chronic and acute renal failure.
Disclosure Statement No conflicts of interest to declare.
References
186
7
8
9
10
Blood Purif 2014;38:180–187 DOI: 10.1159/000368157
measurement of blood volume during haemodialysis. Life Support Syst 1987; 5: 293– 305. Schneditz D, Roob J, Oswald M, Pogglitsch H, Moser M, Kenner T, Binswanger U: Nature and rate of vascular refilling during hemodialysis and ultrafiltration. Kidney Int 1992;42: 1425–1433. Leypoldt JK, Cheung AK, Steuer RR, Harris DH, Conis JM: Determination of circulating blood volume by continuously monitoring hematocrit during hemodialysis. J Am Soc Nephrol 1995;6:214–219. Johnson DW, McMahon M, Campbell S, Wilkinson J, Kime N, Shannon G, Fleming SJ: Non-invasive, optical measurement of absolute blood volume in hemodialysis patients. Kidney Int 1996;49:255–260. Schneditz D, Roob JM, Vaclavik M, Holzer H, Kenner T: Noninvasive measurement of
11
12
13
14
blood volume in hemodialysis patients. J Am Soc Nephrol 1996;7:1241–1244. Schneditz D, Wimmer J, Haditsch B, Holzer H: Measurement of systemic blood flow by slow infusion of ultra-pure dialysate during hemodialysis. J Am Soc Nephrol 2003; 14: 213A. Wimmer J, Haditsch B, Holzer H, Schneditz D: Measurement of central blood volume by slow infusion of ultra-pure dialysate during hemodialysis. J Am Soc Nephrol 2003;14:851A. Schneditz D, Schilcher G, Ribitsch W, Krisper P, Haditsch B, Kron J: On-line dialysate infusion to estimate absolute blood volume in dialysis patients. ASAIO J 2014;60:436–442. Schneditz D, Pogglitsch H, Horina J, Binswanger U: A blood protein monitor for the continuous measurement of blood volume changes during hemodialysis. Kidney Int 1990; 38:342–346.
Kron/Schneditz/Leimbach/Aign/Kron
Downloaded by: University of Tokyo 157.82.153.40 - 5/17/2015 2:44:31 PM
1 Agarwal R: Volume overload in dialysis: the elephant in the room, no one can see. Am J Nephrol 2013;38:75–77. 2 Daugirdas JT: Pathophysiology of dialysis hypotension: an update. Am J Kidney Dis 2001; 38(suppl 4):S11–S17. 3 Thijssen S, Kappel F, Kotanko P: Absolute blood volume in hemodialysis patients: why is it relevant, and how to measure it? Blood Purif 2013;35:63–71. 4 Booth J, Pinney J, Davenport A: Do changes in relative blood volume monitoring correlate to hemodialysis-associated hypotension? Nephron Clin Pract 2011;117:c179–c183. 5 Dasselaar JJ, van der Sande FM, Franssen CF: Critical evaluation of blood volume measurements during hemodialysis. Blood Purif 2012; 33:177–182. 6 Schallenberg U, Stiller S, Mann H: A new method of continuous haemoglobinometric
Absolute Blood Volume in Hemodialysis Patients
21 Dasselaar JJ, de Jong PE, Huisman RM, Franssen CF: Influence of ultrafiltration volume on blood volume changes during hemodialysis as observed in day-of-the-week analysis of hemodialysis sessions. ASAIO J 2007; 53: 479– 484. 22 Mitra S, Chamney P, Greenwood R, Farrington K: Serial determinations of absolute plasma volume with indocyanine green during hemodialysis. J Am Soc Nephrol 2003;14: 2345–2351. 23 Schneditz D, Haditsch B, Jantscher A, Ribitsch W, Krisper P: Absolute blood volume and hepato-splanchnic blood flow measured by indocyanine green kinetics during hemodialysis. ASAIO J 2014;60:452–458. 24 Puri S, Park JK, Modersitzki F, Goldfarb DS: Radioisotope blood volume measurement in hemodialysis patients. Hemodial Int 2014;18: 406–414. 25 Dasselaar JJ, Lub-de Hooge MN, Pruim J, Nijnuis H, Wiersum A, de Jong PE, Huisman RM, Franssen CF: Relative blood volume changes underestimate total blood volume changes during hemodialysis. Clin J Am Soc Nephrol 2007;2:669–674.
26 Kim KE, Neff M, Cohen B, Somerstein M, Chinitz J, Onesti G, Swartz C: Blood volume changes and hypotension during hemodialysis. Trans Am Soc Artif Intern Organs 1970; 16:508–514. 27 Secher NH, Van Lieshout JJ: Normovolaemia defined by central blood volume and venous oxygen saturation. Clin Exp Pharmacol Physiol 2005;32:901–910. 28 Shulman T, Heidenheim AP, Kianfar C, Shulman SM, Lindsay RM: Preserving central blood volume: changes in body fluid compartments during hemodialysis. ASAIO J 2001;47:615–618. 29 Gabrielli D, Krystal B, Katzarski K, Youssef M, Hachache T, Lopot F, Lasseur C, Gunne T, Draganov B, Wojke R, Gauly A: Improved intradialytic stability during haemodialysis with blood volume-controlled ultrafiltration. J Nephrol 2009;22:232–240. 30 Nesrallah GE, Suri RS, Guyatt G, Mustafa RA, Walter SD, Lindsay RM, Akl EA: Biofeedback dialysis for hypotension and hypervolemia: a systematic review and meta-analysis. Nephrol Dial Transplant 2013;28:182–191. 31 Bagshaw SM, Brophy PD, Cruz D, Ronco C: Fluid balance as a biomarker: impact of fluid overload on outcome in critically ill patients with acute kidney injury. Crit Care 2008; 12: 169.
Blood Purif 2014;38:180–187 DOI: 10.1159/000368157
187
Downloaded by: University of Tokyo 157.82.153.40 - 5/17/2015 2:44:31 PM
15 Schneditz D: Recirculation, a seemingly simple concept. Nephrol Dial Transplant 1998; 13:2191–2193. 16 Schneditz D, Kaufman AM, Polaschegg HD, Levin NW, Daugirdas JT: Cardiopulmonary recirculation during hemodialysis. Kidney Int 1992;42:1450–1456. 17 Moissl UM, Wabel P, Chamney PW, Bosaeus I, Levin NW, Bosy-Westphal A, Korth O, Müller MJ, Ellegard L, Malmros V, Kaitwatcharachai C, Kuhlmann MK, Zhu F, Fuller NJ: Body fluid volume determination via body composition spectroscopy in health and disease. Physiol Meas 2006; 27: 921–933. 18 Krivitski NM, Depner TA: Cardiac output and central blood volume during hemodialysis: methodology. Adv Ren Replace Ther 1999;6:225–232. 19 Steuer RR, Bell DA, Barrett LL: Optical measurement of hematocrit and other biological constituents in renal therapy. Adv Ren Replace Ther 1999;6:217–224. 20 Movilli E, Cancarini GC, Cassamali S, Camerini C, Brunori G, Maffei C, Maiorca R: Inter-dialytic variations in blood volume and total body water in uraemic patients treated by dialysis. Nephrol Dial Transplant 2004;19: 185–189.
